Matsuo Bashô's Poetic Spaces: Exploring Haikai Intersections 9781403972583, 1403972583

This collection of essays explores certain neglected aspects of this haikai master's literary and philosophical con

303 50 1MB

English Pages 288 [281] Year 2006

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Matsuo Bashô's Poetic Spaces: Exploring Haikai Intersections
 9781403972583, 1403972583

Table of contents :
Cover......Page 1
Part II: The Artist as Poet......Page 116
5 Double Voices and Bashō's Haikai......Page 118
6 Loosening the Links: Considering Intention in Linked Verse and its Consequences......Page 140
7 Exploring Bashō's World of Poetic Expression: Soundscape Haiku......Page 172
Part III: The Poet as Painter......Page 212
9 Bashō and the Haiga......Page 214
10 Interactions of Text and Image in Haiga......Page 230
11 Buson's Bashō: The Embrace of Influence......Page 256
Contributors......Page 270

Citation preview

‒ P S M B ’

This page intentionally left blank

‒ ’  P     M B   S   Ex plor ing Ha ika i Int er s ect ion s

Edited by

E l e anor K erkham

– ’S POETIC SPACES MATSUO BASH O © Eleanor Kerkham, 2006. All rights reserved. No part of this book may be used or reproduced in any manner whatsoever without written permission except in the case of brief quotations embodied in critical articles or reviews. First published in 2006 by PALGRAVE MACMILLAN™ 175 Fifth Avenue, New York, N.Y. 10010 and Houndmills, Basingstoke, Hampshire, England RG21 6XS Companies and representatives throughout the world. PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN-13: 978–1–4039–7258–3 ISBN-10: 1–4039–7258–3 Library of Congress Cataloging-in-Publication Data Matsuo Bash–o’s poetic spaces : exploring haikai intersections / edited by Eleanor Kerkham. p. cm. Two articles translated from the Japanese. Includes bibliographical references and index. ISBN 1–4039–7258–3 1. Matsuo, Bash–o, 1644–1694—Criticism and interpretations. 2. Haikai—History and criticism. I. Kerkham, Eleanor. PL794.4.Z5M366 2006 895.6132—dc22

2006041585

A catalogue record for this book is available from the British Library. Design by Newgen Imaging Systems (P) Ltd., Chennai, India. First edition: October 2006 10 9 8 7 6 5 4 3 2 1 Printed in the United States of America.

To the fond memory of our late friends, colleagues, and haikai scholar-teachers, Dr. Earl Miner And Dr. Leon Zolbrod

This page intentionally left blank

C

Illustrations

ix

Acknowledgments

xi

Introduction Haikai Intersections Eleanor Kerkham

1

Part I

The Artist as Thinker

1. Bashp at the Center of Creation Hori Nobuo, translated by Cheryl Crowley

23

2. Zpka: The Creative in Bashp’s View of Nature and Art David Landis Barnhill 3. Reinventing the Landscape: The Zhuangzi and the Geographical Imagination of Bashp Peipei Qiu

33

4. Skeletons on the Path: Bashp Looks Forward William R. LaFleur

Part II

61 79

The Artist as Poet

5. Double Voices and Bashp’s Haikai Haruo Shirane 6. Loosening the Links: Considering Intention in Linked Verse and its Consequences I. Leopold Hanami 7. Exploring Bashp’s World of Poetic Expression: Soundscape Haiku Horikiri Minoru, translated by Cheryl Crowley

105

127

159

C      

viii

8. And Us Too Enclosed in Mori Atsushi’s Ware mo mata, Oku no hosomichi Eleanor Kerkham

Part III

173

The Poet as Painter

9. Bashp and the Haiga Joan O’Mara

201

10. Interactions of Text and Image in Haiga Stephen Addiss

217

11. Buson’s Bashp: The Embrace of Influence Eri F. Yasuhara

243

Contributors

257

Index

261

I 

Book Cover Yosa Buson, Portrait of Master Bashp (Bashp-p-zp) Ink and light gray, browns, and green on paper, 98.8  32.1 cm. Courtesy of Itsup Art Museum, Itami. The most appealing and perhaps the best of Buson’s twelve surviving portraits of Bashp; the text, in Buson’s hand, includes a maegaki (introduction) and hokku (date uncertain) by Bashp and is signed, Midnight Cottage, Buson: hito no tan o iu koto nakare

Do not speak of other’s faults

onore ga chp o toku koto nakare Do not mention one’s own fine points mono ieba kuchibiru samushi aki no kaze

1.1 9.1 9.2

9.3 9.4

9.5 9.6 9.7 10.1

When one speaks the lips are cold, autumn winds

“Wind” experiment Kyoriku, Portrait of Bashp (as the “new” Saigyp), Kakimori Bunko, Itami Bashp (verse) with Kyoriku (illustration), Crow on a Withered Branch (“Kare eda ni”), Idemitsu Museum of Arts, Tokyo Bashp, Gourd-beating (Hachi-tataki), Wataya Bunko, Tenri Central Library, Tenri Bashp (verse) with Itchp (illustration), Bagworm (“Minomushi no”), Wataya Bunko, Tenri Central Library, Tenri Bashp, Solitary Traveler in the Rain, Kakimori Bunko, Itami Bashp, Yellow Rose (Yamabuki), Kakimori Bunko, Itami Bashp, Banana Tree by the Gate to the Bashp-an (“Minomushi no”), Idemitsu Museum of Arts, Tokyo Enomoto Kikaku (1661–1707), Melon Skin, Shpka collection Midlothian, VA

29 204

205 208

209 210 211 212 219

x

10.2 10.3 10.4 10.5 10.6 10.7 10.8 10.9 10.10 10.11 10.12 10.13 10.14 11.1

I         

Nakagawa Otsuyu, (Bakurin, 1675–1739), Deer, Beckett collection Denver Colorado Kakujp (1664–1747), Sailing on the Blue Sea, Beckett collection Kaga no Chiyo (1703–1775), Late Spring, Yabumoto collection, Tokyo Ki Baitei (1734–1811), Portrait of Bashp, private collection Yosa Buson (1716–1784), Portrait of Chigetsu, private collection Inoue Shirp (1742–1812), Moon, private collection Takebe Ayatari (1719–1794), Moon, Shpka collection Hakuin Ekaku (1685–1768), Bag of Hotei, private collection Takebe Spchp (1761–1814), Discussion Under a Mosquito Net, Masuda collection, Tokyo Takebe Spchp (1761–1814), Strange Figure, Masuda collection, Tokyo Inoue Shirp (1742–1812), Self-Portrait, private collection Kobayashi Issa (1763–1827), Garden Butterfly, Shpka collection Fujimori Spbaku (1758–1821), Dawn of the Thirteenth Night, private collection Yosa Buson (1716–1784), Pilgrim’s Willow Haiga (Yugyp yanagi jigasan), Itsup Arts Museum, Itami

220 222 223 227 229 230 232 233 234 235 237 238 240 250

A

I

would like to thank Professor Matsuda Yoshiyuki and Yamabushi Priest Hoshino Fumihiro for suggesting this project, and Professors William LaFleur, Marlene Mayo, and Eri Yasuhara for encouraging its completion. Professor Thomas Rimer and the Inter-College Committee on East Asian Studies, University of Maryland; the Japan Information and Culture Center, Embassy of Japan; the Japan Foundation Center for Global Partnership; and the Washington Southeast Regional Seminar on Japan generously supported several efforts to bring Japanese and American scholars together for symposia and performances celebrating Bashp’s life and poetry. I would also like to thank Professor S. Robert Ramsey and the Department of Asian and East European Languages, Literatures, and Cultures, University of Maryland; the University of Maryland Japanese Studies Workshop; Tenri University Library; the Kakimori Bunko; and Professor Akiko Okada, Itsup Fines Arts Museum, for their support of our use of haiga illustrations. I am especially grateful to Kenneth Tanaka, East Asia Collection, University of Maryland Libraries, and Janel BrennanTillmann, School of Languages, Literatures, and Cultures, for providing technical expertise and support. Finally, on behalf of the contributors to this volume, I would like to express my gratitude to Professor Ogata Tsutomu, who sent us our Japanese scholars, and who has nurtured the quests of so many readers, scholars, and Bashp lovers for greater understanding of the depth and variety in those poetic spaces Bashp created for all of us to step into.

This page intentionally left blank

I       

H    I  Eleanor Kerkham

T

he power of Matsuo Bashp (1644–1694) to maintain a central spot in Japan’s postwar popular consciousness must be seen as remarkable, even in a nation that listens regularly to its literary figures. While Japan’s and the world’s poetic tastes are in constant flux, it is still safe to say that in the early twenty-first century Matsuo Bashp remains the most beloved poet in Japan, and that he is the most well-known of all Japanese poets outside of Japan. His American audience hardly blinks when a major poet and future poet laureate, Robert Hass, suggests that Bashp transformed the haiku/hokku “into one of the great lyric forms in human culture and himself into one of the world’s great lyric poets.”1 It is clearly Bashp’s identification as founder and supreme practitioner of the world’s “most widely recognizable poetic form,” the haiku, that accounts for his international popularity.2 The corpus of English language material on Bashp reflects this overwhelming interest in the seventeen-syllable haiku, even though Bashp seems to have valued more highly his creative work with linked verse (haikai no renga or renku, comic linked verse). If we divide studies in English into three overlapping groups, the most numerous are presentations of Bashp as haiku poet, including a great many translations and imaginative recreations, particularly of one hundred or so of his most wellknown hokku. Closely related to these are the works in English on Bashp as Buddhist poet, student of Zen Buddhism, or exemplar of the Zen spirit. These studies too turn primarily to Bashp’s hokku and to a lesser extent to his travel journals as texts for illustration and discussion. A third category is the relatively small body of scholarly monographs and articles that have focused to date on biographical information and on introductions to and interpretations/translations of the most well-known texts in three of the major forms in which he worked, the hokku, the comic linked verse, and haibun (haiku-like

E     K     

2

prose).3 The essays collected in this volume reflect the desire to add to this third body of works in English by exploring in more detail some of the varied intersections among this haikai master’s literary, philosophical, and painterly contributions.

T O KU

NO HOSOMICHI /B¯ 

B

The seed for this desire was planted with a series of “productions” begun in Japan in 1988 centering on Bashp’s prose masterpiece, the Oku no hosomichi (Narrow Road Through the Deep North). The story of these happenings represents a remarkable cultural/political phenomenon, dubbed by the Japanese media the “Oku no hosomichi Boom,” which thrust a favorite poet into the media spotlight, giving him the new role of ushering Japan into the twenty-first century. It is worth recounting here because of what it says about Japan and about the poet. As early as the mid-1980s plans had begun for a set of celebrations of the 300th anniversary of Bashp’s most famous poetic journey. Bashp and his disciple Kawai Sora (1649–1710) traveled through northern Japan from spring to late autumn of 1689. Celebratory events (ibento) were staged in all major prefectures through which they had traveled.4 Bashp had himself commemorated the six-month journey in the poetic work canonized as one of the masterpieces of the Japanese language and the finest of Bashp’s poetic prose works.5 Completed not long before his death in 1694, the Oku no hosomichi narrates a journey taken by two travelers from Edo up the Pacific coast to Matsushima and Hiraizumi, across Japan to Kisakata, down the rough Japan Sea coast and inland toward Lake Biwa and the Ise Shrine.6 The 1988–1989 Oku no hosomichi celebrants seem to have made no conscious distinction between Bashp’s significant life experiences (1689) and his literary text (1694). They were commemorating both. The beauty of Bashp’s masterpiece was celebrated in music, in dramatic performances, television documentaries, professional radio readings, and in popular and scholarly studies. At the same time, a whole nation commemorated the fact that two historical persons had taken a journey at a specific time, that both had written about it, and that their own responses and those of later generations of writers and poet–travelers to the areas through which they traveled had transformed these spots into some of Japan’s premier historic and tourist sites. At appropriate moments during the two-year period (1988–1989), corresponding seasonally whenever possible to the Oku no hosomichi’s narrative frame, special commemorative exhibits were held all along its route at museums, temples, shrines, libraries, galleries, and tourist

I       

3

areas. National and international haiku contests were held. There were public “haiku radio excursions” (rajio haikuingu) along those portions of the route that retain their natural beauty. Well-known TV comedians or radio announcers often joined in to entertain, direct haiku readings, and present prizes and presents to participants from all over the country who had responded to the plea to spend their leisure time with Bashp and Sora. Commenting on some of the Oku no hosomichi celebrations in the year 1988, Ogata Tsutomu describes events such as the Mitsukoshi Department store’s memorial celebration in the city of Sendai; an “Oku no hosomichi Summit,” held in Pgaki, Bashp’s final stop on the journey; and the publication of news items in dedicated sections of daily newspapers, special editions of weekly and monthly magazines, and various haiku journals.7 City and prefectural officials, backed by help from large corporations and the national government, produced symposia and conferences (several of which were televised by local broadcasting subsidiaries). They arranged for televised linked verse and haiku composition sessions and for dance, theatrical, and musical performances featuring native arts and artists. There was a traveling European haiga (haiku paintings) exhibit sponsored by the Belgian government and the European Community, which aimed at strengthening European– Japanese relations, as well as other international symposia in the United States and Japan. Ogata suggests that the primary audience for such productions in Japan was neither the academic community nor the many amateur haiku societies scattered across Japan, but rather ordinary people pursued by local and central government bodies carrying out the official “Revival of Villages” or “Revival of the Rural Areas” (mura-okoshi or chihp-okoshi) policy of the national government.8 Taking advantage of government support available to them during a period of national affluence, even seemingly obscure areas mentioned in the Oku no hosomichi worked at staging symposia, essay contests, and public lectures. The City of Spka (the travelers’ first stop on their journey), for instance, presented in 1988 and 1989 a series of four “international symposia.” The city had constructed in a public park area a splendid new community auditorium and a replica of a Tokugawa bridge of the type Bashp may have crossed upon entering the town. The public symposia, featuring Bashp scholars as well as other prominent writers, poets, academics, and critics, staged for the citizens of the city and prefecture in celebration of their new auditorium and, of course, of their relationship to Bashp and his journey.9 Yamagata Prefecture (representing the area in which Bashp spent more time than any other on his journey) outdid all others in this

4

E     K     

regard. Officials of its various city tourist bureaus first conceived of a five-year plan of events. This plan was later elaborated and extended to at least eight years and included the construction, among other things, of two Oku-no-hosomichi-related tourist theme centers/museums, one at the Ryushakuji Temple or Yamadera and one in the village of Haguro, the religious center of what was known in Bashp’s time as the Three Sacred Mountains of Dewa. Funding also went into the production of a variety of public events, reproduction of numerous Bashp-related primary manuscripts, and into the preservation of other key Yamagata Oku-no-hosomichi-related spots, such as the home of the merchant Seifu in Obanazawa or the so-called hpjin no ie or borderguard hut in which the travelers spend three days waiting for a storm to pass. The Oku no hosomichi Boom tastefully combined its educational and commercial sides. There was the publication of new and reissued monographs on Bashp and numerous “special editions” of both scholarly and commercial journals.10 Reproductions of Bashp “poem cards” (rectangular tanzaku and square kaishi), haikupaintings (haiga), and of two Yamagata-Prefecture-related Yosa Buson (1716–1783) depictions of Oku no hosomichi, in handscroll and miniature screen format, appeared on the tourist market. There were various other expensive-to-modest souvenir gifts: Bashp and Sora dolls and miniature sculpture, door and wall hangings, cups, goodluck charms, T-shirts, towels, coasters, key chains, and such, most inscribed with selected Bashp hokku composed on the journey or included in the Oku no hosomichi. In addition to the many new scholarly monographs, illustrated texts, and personal reflections on Oku no hosomichi, major publishers and bookstores produced and promoted popular Oku no hosomichi manga, film companies produced special Oku no hosomichi anime, and educational companies produced audio readings, as well as imaginative video dramatizations of the journey and readings of the text. Important social, cultural, and political issues were of course involved in the general Oku no hosomichi fervor in Japan. An official “cultural politics” (bunka gypsei) was at play and its goals were complex. National, prefectural, city, and village governments worked together but sought to accomplish different ends. One might ask why they elected to use Matsuo Bashp and not other master Japanese poets or why not other journeys and their literary depictions? Was it just luck that a memorial opportunity coincided with the monetary means to celebrate in such high style? Why is a journey celebrated? Why and how does a seventeenth-century poet capture the sustained attention

I       

5

of Japanese students and teachers, bureaucrats, officials, salary men and women, housewives, scholars, and amateur and professional poets? And what is the nature of the Bashp attraction? We try to suggest answers here to some of these questions, based in part on the intrinsic artistic interest, strength, and variety of Bashp’s accomplishments as an artist. A more important question, however, might be how and why a writer who elected to live on the margins of his society and who chose to work with nonconventional literary genres became Japan’s supreme canonical poet, and how was his canonization used by the political and cultural establishments of the late 1980s–early 1990s as part of a larger movement to encourage what might be seen as a selective if ironic return to “traditional” values? As suggested earlier, the complex of Bashp events was designed in part to attract new visitors to less well-trod areas and to rekindle native interest in local history. Oku no hosomichi ’s 1988–1989 anniversary years coincided with the height of the economic “bubble” of the 1980s and a period of conspicuous affluence. However, national and local governments were concerned with other related problems as well. One of these was environmental—preserving historically and scenically important spots and sacred areas while attracting the populace to them. Another was cultural and economic—the protection and promotion of traditional arts and artists. Yet another was demographic, political, and social—the desire to decentralize Japan and to encourage citizens to remain in or return to native areas rather than moving into over-crowded cities. A related goal was the desire to elevate certain “globalization” issues: selling an artistically and economically successful Japan and making a political issue of Japan’s responsibility for assuming an international leadership role in the twenty-first century. Cultural diplomacy must be part of this important task. The producers and promoters of the Oku no hosomichi events were clearly concerned with the presentation of Japanese arts, culture, and what they saw as Japanese values to the rest of the world. Haiku in particular had proven to be an accessible, near-universal poetic form, one that was already serving as a vehicle of international understanding. As a poetic genre based on a keen awareness of the workings of interrelationships of all sorts, the haiku might play a role in taking Japan’s remarkable postwar reconstruction as a nation of peace and cultural achievement to a new level. Japan was also struggling with the problem of how to encourage its own citizens to spend their leisure time and extra cash in more creative, restorative ways. Bashp and Sora could be seen as model travelers. Sora’s diary of the journey reveals that the two carefully

6

E     K     

performed their pretravel homework. They knew how to enrich themselves and their journeys with older travel books, earlier prose narratives and poetry, local history, and a lively narrative and dramatic folk repertory. Once they arrived at their destinations, they were able to meet new and old friends and fellow poets and to learn from temple and shrine histories, folk legends, stories told by people of all classes and professions, and from mountains, rivers, trees, insects, and rocks. They were informed travelers, aesthetically and emotionally involved with the new spaces they entered. The 1988–1989 event planners seem to have reasoned that Japanese group or individual excursions to their own “back roads and far towns,” as well as to the churches or art museums of Europe, organized spiritual pilgrimages, such as the week-long yamabushi (mountain priest) “rebirth” or rejuvenation workshops in Haguro, or visits to newly constructed theme parks and museums could take the place of shopping sprees in Hawaii and LA, sex tours to Thailand, or showtime and gambling excursions to New York City and Las Vegas.11 It is an irony familiar in the Western literary tradition that a writer, employing a mixture of traditional and unconventional language, poetic forms, and subject matter, might manage posthumously to position his or her art and fictional images of unconventional lifestyles within the mainstream of his or her own cultural traditions. It is also not unusual to see such an icon used by the cultural establishment to spearhead certain political agendas. While we cannot address all of the questions raised by these issues, we can note two points that link these Oku no hosomichi celebrations to the present collection of essays. First is that fact that the myths of Bashp and of the fictional images his art has created represent something more than the accomplishments of a single poet. Second is the observation that in spite of, or perhaps along with, the desire to harness the commercial and political potential of the carefully created “Bashp boom,” there was also a genuine intellectual and spiritual thrust behind the many Oku no hosomichi productions. My own participation in several of these consistently revealed what must be described as straightforward, earnest attempts, both on the part of local officials and of individual citizens, to glean hints, from Bashp’s literary texts, his life experiences and lifestyle, and his way of thought, on how to structure their own life sojourns, how to learn from people whose paths crossed their own, and how to spend travel time (leisure time or any time) more creatively. The success of the 1988–1989 Oku no hosomichi events, in any case, fueled the desire to cultivate and prolong the community spirit and local enthusiasm by planning a new series of Bashp commemorations.

I       

7

In July 1993 two almost unrelated “events” were celebrated together at an international symposium in the then newly opened EdoTokyo Museum in Tokyo. The two were the 1400th year anniversary of the founding of the Dewa Sanzan Shugendp (Buddhist/Daoist derived secret religions discipline) tradition and Matsuo Bashp’s relationship to Yamagata Prefecture.12 Professor Matsuda Yoshiyuki, a specialist in the field of leisure, travel, and recreation, and producer– coordinator of many of the Bashp events mentioned earlier, reminded his audience that it would be possible, in 1994, to celebrate not just the completion of an important literary text (Oku no hosomichi), but also the 300th and 350th anniversaries of its author’s birth (1644) and death (1694). Professor Matsuda challenged his Japanese and international guests to continue to think about how Bashp’s philosophy of pilgrimage, his extraordinary devotion to his art and to his disciples, and his artistic accomplishments might become part of Japan’s broader global legacy, contributing to a new international cultural consciousness. These and other similar challenges were accepted by a number of different groups, and although the 1994 commemorations of Bashp’s birth and death did not matched the 1988–1989 Oku no hosomichi excitement (Japan’s financial bubble was deflating and less cash flowed for spiritual and cultural events), a far broader area in Japan was able to become involved, ranging from Bashp’s birthplace in Iga Ueno to his death site in Osaka and grave site at the modest Buddhist temple Yoshinaka-dera, overlooking Lake Biwa. This second set of celebrations took on a more international cast, and haiku and Bashp enthusiasts from Europe to the United States joined in the celebrations. Once again, there were haiku and linked verse gatherings world-wide, musical events, scholarly meetings, essay contests, and important commemorative exhibits, most notably at Tenri University Central Library, home of the most extensive collection of haikai manuscripts; at the Matsuo Bashp Museum in the poet’s birthplace, Iga Ueno; and at the Historical Museum of the City of Ptsu. Important research tools—catalogues and monographs—and opportunities to view primary materials again resulted. Focus on the life and on all of the areas frequented by Bashp and funds for the publication of newly uncovered documents helped scholars and Bashp lovers ferret out invaluable information on lesser poets, painters, and friends who had sought out the master’s instruction and companionship and who had preserved and passed on his teachings, letters, calligraphic productions, and haiga. A more scholarly approach may have been present in these events but as in the first set of “happenings,” the Japanese people

8

E     K     

were championing a man and his thought, haiku as an international poetic form, travel to out-of-the-way spots, Bashp and Sora as ideal travelers, Bashp’s poetic texts, and—to borrow from chapter 4 by William LaFleur in this volume—a way of life that celebrated rather than debased nature and oneself.

R B¯  Horikiri Minoru divides contemporary scholarship on Bashp in Japan into two broad categories: studies illustrating an “exterior” approach, or the work of scholars who examine the “social and cultural history of Bashp’s literary career,” and “interior” approaches, which focus on the linguistic, expressive aspects of language and specific literary texts.13 While the categories are not exact, and many scholars necessarily combine aspects of both when approaching an individual literary text or topic, the exterior approach, described by Horikiri as “flourishing in Japan today,” would include haikai scholars who have found their critical center within the haikai za. The za can be seen as a creative space in which complex professional, intellectual, personal, and literary “intersections” constantly alter the flow of literary creation.14 These include points of interaction among haikai masters and their disciples; among different but fluid haikai “schools” (poets all over Japan gathered around various haikai masters or their representatives); between the two genres, renga and haikai; and between these two relatively late poetic forms and the Japanese waka/monogatari tradition. As an art that parodies and often subverts its linguistic, generic, and individual poet–predecessors, haikai’s intersections also include imaginative links to the rest of Japanese literature and culture, to Chinese prose, poetry, and its philosophical traditions, and to the social, intellectual, and everyday realities of seventeenth-century Tokugawa life. The essays in this collection seek to draw our attention to the nature of some of these intersections and to the creative spaces they generate. Part I, The Artist as Thinker brings together three essays that examine Daoist, Buddhist, and Neo-Confucian resonances in Bashp’s thought and creative life. Hori Nobuo (“Bashp at the Center of Creation,” translated by Cheryl Crowley) opens with the question, “what is haikai?” and by extension, “what is a haikai no hito (a haikai person)?” He answers this question by presenting the probing answer of Bashp’s disciple Mukai Kyorai to a similar query, “what is the foundation of haikai?” Kyorai’s response, “If you write a prose text suffused with haikai, it becomes haikaibun (haikai prose). If you compose a waka in the haikai spirit, it becomes haikai waka. If you

I       

9

make haikai your way of life, you are a haikai person,” suggests that haikai was not simply a particular art form but was, as Hori suggests, “a way of thinking, speaking, and behaving that is suffused with certain tendencies.” To define these certain tendencies Hori examines the earliest intersections between the Chinese notion of kokkei (comical, humorous) and Japanese definitions of the term haikai. He then examines later Chinese-influenced characterizations that suggest that haikai was always seen in Japan as a literary form using humor to challenge “correct,” “proper,” or “right” views on literature and on human experience. Hori thus grounds haikai and Bashp as haikai no hito around the artists’ iconoclastic attitude toward himself and his subject. Haikai or kokkei was, in Hori’s words, “an innovative perspective or way of thinking that calls into question viewpoints, values, and aesthetics that have become habitual, fixed, and formalized, yet expresses deeper principles.” Hori next outlines briefly the process by which earlier Danrin Haikai School theorists as well as Bashp School haikai poets, in their attempts to defend and legitimate their nonorthodox literary form, discovered, particularly in the Chinese Daoist classic, the Zhuangzi, both philosophical and stylistic foundations for their poetic path. Hori argues that Bashp’s personal encounter with and “close connection to the profound philosophical tenets of Zhuangzi,” helped situate haikai and himself as haikai master “at the center of creation,” slightly out of tune with the proper way, but in harmony with zpka (“the workings of the universe, the activity of the universe”). David Barnhill’s essay (“Zpka: The Creative in Bashp’s View of Nature and Art”) presents an examination of the meaning and effect of Bashp’s encounter with this key notion of zpka and with other related concepts embedded in “the multiplex Chinese religio-aesthetic tradition” upon which Bashp drew. Centering on the meaning of the term zpka as used in the well-known introductory passage to the travel journal Oi no kobumi (“Knapsack Notebook”), Barnhill argues that the concept is basic to an understanding of the haikai master’s poetic theories and practice. Asking how Bashp’s “texts mean, particularly in relationship to other texts,” and what the contexts are “in which we should place Bashp’s texts as we interpret them,” Barnhill explores certain Western critical approaches that help us answer the first question, and certain aesthetic ideals in Chinese poetry and painting and a variety of Chinese Daoist, Confucian, and Neo-Confucian sources that help situate us in the second question. His discussion of the reasons for his own choice of “the creative” as a translation for zpka goes beyond the problem of accurate rendering

10

E     K     

into English and includes a contextual examination of the meanings and philosophical nuances of all of Bashp’s uses of this and other related religio-aesthetic terms—terms that reflect the poet’s understanding of nature’s creativity and of its relationship to artistic creativity. By explicating the complex, multifaceted thought-structure behind important prose passages drawn from Bashp’s haibun, Barnhill places Bashp’s work in a broad cultural context, which intersects not only with Chinese aesthetic and philosophical thought, but also with Japanese Buddhist, Shintp, and folk traditions, and with the actual vitality and creativity of nature. Peipei Qiu (“Reinventing the Landscape: The Zhuangzi and the Geographical Imagination of Bashp”) focuses on the literary and intellectual intersections between Bashp’s work and the Daoist classic, the Zhuangzi. Bashp’s encounter with this text resulted, she argues, in the reinvention of “the poetic significance of landscape in his travel journals.” She examines his creation of the image of a traveler who has creatively adapted the Zhuanzi’s shpypyu (carefree wandering) aesthetic and spirit to his own, haikai poet’s poetic and life goals. Qiu’s analysis of Bashp’s “geographical imagination,” formed not simply as a result of material “viewing” of the physical spaces his traveler enters, but as a poetic response to “broader cultural, aesthetic, and philosophical frameworks,” reveals a traveler/poet as emotionally and intellectually involved with earlier traditions and in particular with the Zhuangzi, as with the landscape itself. Seeing the Zhuangzi not just as a literary source of new poetic phrases but also as a source for a radical break from medieval Japanese attitudes toward the traveler and his/her interaction with the landscape, Qiu, like Hori and Barnhill, is drawn to the importance for Bashp of the philosophical/aesthetic implications behind, and the rich legacy of, the Chinese term, zaohua (zpka, “create and transform”), which she too seeks to redefine and expand upon as a path into Bashp’s creative approach. Qiu’s presentation of the Song scholar Lin Xiyi’s commentary on some of those passages in the Zhuangzi that seem to have helped shape Bashp’s thinking about larger questions of space, time, and travel are especially helpful in revealing the nature of the haikai poet’s intersection with this unique Chinese literary/philosophical text. William LaFleur (“Skeletons on the Path: Bashp Looks Forward”) begins his essay with the interesting observation that during the 1960s, just as Bashp was being discovered by a select American audience as a prime exemplar of an attitude and way of life that eschewed rootedness in material accumulation and in the abuse of nature and the self, he was being dismissed by a “politically radicalized younger

I       

11

group” of Japanese scholars as “insufficiently critical of the political structure and powers of his own time.” LaFleur continues with a fascinating critical tour de force that compares and contrasts two ideologically opposed approaches to Bashp’s world. He presents first a critique of a form of cultural criticism that, in its 1960s’ Japanese Marxist guise, tended as a matter of ideology to devalue Bashp as a poet who “provides no social criticism.” LaFleur does this by showing how select elements in Bashp’s work (in this case passages from the Oku no hosomichi) and recent historical research on the relationships between Tokugawa political and cultural centers and its border areas in the northeast can be marshaled as evidence that Bashp was not only insufficiently critical of the political structure and powers of his own time and a poet“ ‘who provides no social criticism’ could be located,” but was also a kind of cultural emissary of the Tokugawa shogunate. In this approach LaFleur argues that as one prong of an attempt to “colonize” outlying areas not yet under the full cultural sway of the Kyoto/Edo political and cultural center, Bashp succeeded, through his travel and his fictional accounts of these sojourns, in drawing local officials, merchants, peasants, and poets into “an aesthetic–personal ‘network,’ ” which could be used as a tool for political and economic control. LaFleur next employs similar careful reasoning combined with a close reading of texts to make the case that there is much in Bashp’s work that, in Robert Alter’s words, runs “counter to the dominant ideology of the culture” and turns “the tables on any agenda that would envision colonization by enculturation.” Finally, drawing on echoes of the Zhuangzi and on Buddhist attitudes toward death and the disappearance of the body, LaFleur illustrates the limits of both approaches by moving more deeply into an examination of Bashp’s creation of an “aesthetic and etiquette of death”—one that dares to “look forward” to his own self as a “skeleton on the path.” Part II, The Artist as Poet, focuses on an examination of reading practices and approaches to individual genres and texts. Haruo Shirane’s essay (“Double Voices and Bashp’s Haikai”) opens this section with an examination of the manner in which Tokugawa haikai generally and Bashp’s haikai in particular “emerged out of the intersection of contrastive languages/cultures,” resulting in what Shirane describes as the “double voiced,” “dialogic,” or “double-movement” literary genres identified as haikai. Placing haikai within the broader cultural and linguistic transformations that marked seventeenthcentury Japan, Shirane pays special attention to the interactions between contrastive languages found not only in individual hokku, but also within haikai links: “haikai thrived in the intersection of these

12

E     K     

languages, especially between classical diction (gago), highly encoded, self-enclosed, almost disembodied language, which bore the voices of a refined, aristocratic past, and various types of vernacular, popular language (zokugo), which had a strong sense of immediacy, physicality, and referentiality, closely tied to everyday commoner life and material culture.” In relating a variety of topics key to an understanding of haikai—such as the nature of the creative process, of poetic and of “socially and ideologically inscribed” language, of “intertextual double-voicedness,” of parody, or of “tradition”—Shirane illustrates ways in which we might draw from a rich body of Western critical theory (Bakhtin, Althusser, Riffaterre, Gates, Benjamin, and Hutcheon, for instance) in thinking about the hokku and haikai as literary genres centrally involved in an open, ever-evolving dialogue with the immediate present, the past, and the future. Leopold Hanami (“Loosening the Links: Considering Intention in Linked Verse and its Consequences”) presents a theory and illustration of an innovative new approach to the reading and interpretation of orthodox linked verse and of haikai. He examines first the traditional approach to the presentation and reading of linked verse found in modern Japanese and in English studies and translations of the poetic form. Arguing that analyses of renga should address features peculiar to the form—“a change in direction, the mutability of meaning, a destabilization of perspective”—Hanami demonstrates the possibilities of a new approach by presenting a rereading of the first ten verses of the well-known Minase sangin (1488) by the renga poets Spgi, Shphaku, and Spchp and of the first twelve verses of the Bashp School kasen (thirty-six verse comic linked verse), Ichinaka wa (In the City), from the Bashp School anthology Sarumino (Monkey’s Raincoat, 1691). His rereadings illustrate the limitations of the traditional focus on the two-verse sequence and reveal the possibilities of moving beyond the associations between two consecutive stanzas “by examining one stanza in relation to any stanza within the poem, and, ultimately, without it.” Drawing on a wide range of Western critics and critical theorists in considering the role of interpreting “intent” and “meaning” in poetic composition, Hanami uses insights gained from these critics to clarify how, in the composition process, renga and haikai poets had to remain aware of far more than simply the previous verse. Hanami’s readings suggest that the traditional approach to reading renga and haikai, in its demand for immediate intelligibility between any two consecutive verses, tends to obscure meanings hovering at the boundaries of words and phrases and to hide from the reader some of the unexpected productive effects of language and the

I       

13

imagination. His application of what I believe is a unique and innovative approach to the reading process and to linked verse composition helps free the reader’s imagination from the dominant and, it seems, ultimately limiting two-stanza reading. Horikiri Minoru (“Exploring Bashp’s World of Poetic Expression: Soundscape Haiku,” translated by Cheryl Crowley) opens with an assessment of recent scholarship on Bashp in Japan. As mentioned earlier, he divides this scholarship into two broad categories, studies having an exterior and those with an interior approach. Horikiri focuses his attention on interior approaches that center on linguistic analysis, meaning, intention, interpretation, and poetic context. Studies employing interior approaches address such issues as “how should we approach Bashp’s works as literature, what are the main characteristics of his poetry, and most importantly, why does Bashp’s literary style hold such fascination for us?” He divides the interior category further into two types: studies that examine the “linguistic and cultural circumstances of the age in which these works were created in order to isolate the distinctive features of these expressions within the context of the Genroku period (1688–1704)” and a more comparative type, which “analyzes the unique structure of Bashp’s expression within the general context of the art of poetry.” Horikiri illustrates this more general, comparative approach with a discussion of the three elements he sees as intrinsic to Bashp’s haiku expression. The first is the “haikai” element or the “quality of playfulness, humor, or wit.” The second is “a conceptual method based on dialogue, metaphorical hyperbole, or conjecture,” in which “the poet decisively grasps the poetic object with strong, subjective discernment and projects his emotions onto that object.” The third element and the one on which Horikiri centers his attention is described by the term keiki (landscape), “a poetic device that depicts an image, especially a visual image, while communicating the poet’s emotions.” Providing a discussion of both the Chinese and Japanese theoretical backgrounds for the phrase that defines this approach—keisen jpgo (scene first, emotion next)—as well as Western theories on the poetics of sound, Horikiri calls attention to the importance in the Bashp repertoire of a relatively large number of “soundscape” verses (borrowing a term coined by the Canadian musician and theoretician R. Murray Schafer): hokku, which employ aural imagery to create “a poetic world of landscapes of sound.” In order to provide a more extended look at Professor Horikiri’s readings and provocative analyses of Bashp’s soundscape hokku, Cheryl Crowley presents, with special permission granted by Professor Horikiri, a translation and summary of his discussion

14

E     K     

(printed elsewhere) of ten of Bashp’s soundscape hokku (see chapter 7, note 11). Finally, as a way of bringing more of the sounds of Bashp’s poetic travel piece, the Oku no hosomichi, into this volume, I provide an analysis of an unusual reading and translation into modern Japanese of Bashp’s masterpiece by the award-winning novelist Mori Atsushi (1912–1992) (Kerkham, “And Us Too Enclosed in Mori Atsushi’s Ware mo mata, Oku no hosomichi ” [And Me Too, Once Again, into Oku no hosomichi]).15 In his lyrical study of Oku no hosomichi, Mori attempts to pull his readers and himself as writer into what he sees as the ever-expanding circles of meaning that surround and reside in the depths (oku) of Bashp’s last prose piece. Mori’s monograph was written for the audience celebrating the 1988 Oku no hosomichi Boom described earlier. In it he first articulates what can be seen as a structuralist/intertextual critical approach to the literary work. He then develops a theory on the shape and meaning of its overall structure and describes what he sees as Bashp’s most important literary technique, that of “taip ” (correspondences or intersections). Defining taip as temporarily balanced intersections between images, words and phrases, individual hokku or waka, earlier tales, human beings, seasons, geographical settings or seasonal movements, Mori reads his audience through his own modern Japanese version of the text, progressively uncovering these intersections and analyzing how they function within the work to create an interconnecting series of echoes and counter-echoes that keep the reader moving forward and backward among the text’s individual, semi-independent haibun. A closer look at Mori’s study suggests that his rereading of the Oku no hosomichi aims at drawing himself and his readers into a journey and a way of thought that looks openly at death and that believes in the power of literature and of memories of the past to provide both wisdom and aesthetic pleasure. The last section, The Poet as Painter, introduces a topic that has not been attended to sufficiently in Western scholarship on Bashp, the haiga (haikai painting). Each of the three essays here calls our attention to a complex of questions surrounding the intersections of poetry and painting. Joan O’Mara (“Bashp and the Haiga”) presents an introductory essay on Bashp’s haiga, outlining their nature, scope, and possible line of development (many of Bashp’s haiga are not dated). After a discussion of several of his more interesting haiga, O’Mara assesses his overall contribution to the genre, concluding that “haiga became a major form of artistic expression with Matsuo Bashp.” While his haiga seldom “add secondary meanings to the poems,” his

I       

15

approaches, O’Mara argues, “set the tone for most haiga that was to follow in the next half century.” Dividing the body of Bashp’s paintings into three categories—“simple sketches,” “illustrations based on journeying,” and haiga (combination of painting and at least one hokku)—O’Mara compares and contrasts the haiga that she sees as being “simply illustrative” and those employing a more imaginative “associative linking” of poem and visual image. It is the “associative” haiga that O’Mara see as bringing the genre “to its fullest expressive potential, as a type of mixed media variant on the haikai no renga process.” O’Mara provides insight into Morikawa Kyoriku’s (1656–1715) influence on Bashp’s haiga and discusses the importance to Bashp and to contemporary and future haikai poet–painters of collaborative works ( gassaku) and of haiga that were products of communal haikai sittings. Stephen Addiss (“Interactions of Text and Image in Haiga”) broadens O’Mara’s history of haiga by taking the story from its beginnings to the early nineteenth century. His primary focus is on the vital intersections in haiga among visual image, calligraphic style, poetic meaning, color, and overall design. Addiss confirms O’Mara’s assessment of Bashp’s contribution to the genre, suggesting that while his “modest, subtle haiga don’t seek to impress viewers with technical prowess, bright colors, or dynamic brushwork . . . his simple but flavorful paintings contribute gently evocative images to his poems” and “set the tone for most haiga that was to follow in the next half-century.” To illustrate what he sees as common features of works by Bashp, his followers, and poet–painters working during the first fifty years after Bashp’s death, Addiss looks at the work of four direct and indirect Bashp followers. Common features of the group include informal and modest brushwork, little or only soft, unobtrusive color, visual imagery that seldom goes beyond the text, prominence of calligraphy and thus of the verse, and use of simple imagery and empty space. Addiss concludes his historical survey with a discussion of changes in the haiga tradition brought first by Yosa Buson, who was influenced by the literati painting tradition, and continued in the early nineteenth century by professional painters working in a more naturalistic style associated with Matsumura Goshun (1752–1811). While the common characteristics listed earlier continue to apply generally to later haiga, Addiss argues that regardless of painting style, the ways in which text and imagery intersect gradually became more complex or, in O’Mara’s words, more subtle and less direct in their associative techniques. Eri Yasuhara (“Buson’s Bashp: The Embrace of Influence”) completes the collection of essays with an examination of ways to think about

16

E     K     

the forward and backward moving intersections between earlier and later haikai poet–painters and their literary and painterly texts. Yasuhara defines the traditional view of the Bashp–Buson relationship—including Buson’s own statements about it—as one in which the latter poet “revered Bashp as poetic ideal or inspiration” and lamented his own ability to live up to his poet master. She next questions Buson’s literary protestations of his inability to stand beside his chosen predecessor by examining other passages from his prose works that reveal a more complex and shifting relationship between the two poets—a relationship, she suggests, that demands a different sort of characterization. This she supplies by applying to the Buson–Bashp nexus the theories of influence put forth by Harold Bloom in his The Anxiety of Influence (1973). Yasuhara interweaves Bloom’s insights into relationships between two strong poets with her own close readings of key hokku, haiga, and prose works by Buson to characterize a painter–poet who commemorates, incorporates, embraces, and then swerves away from his chosen predecessor. Focusing her attention on Buson’s haiga Yugyp yanagi jigasan (Pilgrim’s Willow Haiga) or Ishi jigasan (Rock Haiga; figure 11.1), Yasuhara reads into this piece complex textual and visual evidence of Buson’s intricate gestures toward two of Bashp’s own precursors, Saigyp and the Chinese poet Su Dongpo, toward a np play featuring Saigyp and alluded to by Bashp, and of course toward Bashp himself. She concludes that “in this complex and sophisticated work, Buson pays homage to his poetic predecessors, places himself in their lineage, then contradicts them in a complex gesture that brings his own poetry and painting to the fore.” Her description of the Buson/Bashp relationship and her application of the theories of Bloom to it reveal that though Buson may be acting out, in certain commemorative life activities and in multiple literary/painterly texts, psychological and creative needs somewhat different from those described by Bloom, he still appears to be caught up in a conscious and unconscious drama that involves the desire to incorporate as well as to repress, outdo, or overcome established traditions and his major precursor. The unifying subject of these essays is, of course, Bashp as haikai master, and our primary goal is to provide greater understanding of the varied intersections alive in all of the creative spaces he entered. Our hope is that, armed with the wealth of information now available about the poet and his life and times and with the insights of both Japanese and Western haikai scholars and of contemporary literary theorists, we might all evolve toward a sufficiently dialectical critique— one that is unafraid of attesting to the literary value of Bashp’s works

I       

17

(their insights, complexity, and beauty) and that seeks to integrate his accomplishments as haikai master with the social and cultural realities of Tokugawa Japan.

N For transliteration of Chinese names, terms, and places into English, we have used the pinyin system and for those in Japanese the modified Hepburn system. We have rendered names of Japanese and Chinese authors and scholars in the normal East Asian name order, surname followed by the given name. Pen names are used for poets and painters known primarily by those names, such as Bashp or Saigyp. 1. Robert Hass, The Essential Haiku: Versions of Basho, Buson, & Issa (Hopewell, NJ: The Ecco Press, 1994), 4. Hass served two terms as United States’ poet laureate, 1995–1997. Bashp and his contemporaries used the term hokku (“first verse”) or haikai, rather than haiku, to designate compositions in the 5/7/5 syllable poetic form. The hokku served as the first verse of a linked verse or were seen as potential first verses. 2. In The Sound of Water, Haiku by Basho, Buson, Issa, and Other Poets (Boston: Shambhala Centaur Editions, 2000) Sam Hamill states: “Haiku may be the most widely recognizable poetic form in the world,” xiii. 3. The term haikai originally distinguished humorous or light-hearted verse from more serious verse (for a discussion of the origins and meanings of the term, see chapter 1 by Hori Nobuo in this volume); haikai is used both as an abbreviation of haikai no renga (comic linked verse) and as a general term to designate one or all of the poetic forms (hokku, renku, haibun, and haiga [haiku-style painting]) or the aesthetic attitudes associated with haikai no renga. 4. Counting calendar years (using the Japanese kazoedoshi method), 1988 is the 300th year while 1989 marked the actual 300 years. A debate ensued in Japan on which year should be chosen for anniversary celebrations, and in the end Oku no hosomichi events were staged both years. Ogata Tsutomu discusses the arguments made for when best to stage the 300th year anniversary in his article, “Oku no hosomichi sanbyakunen” (Oku no hosomichi Three Hundred Years), Bungaku, 5, Vol. 56 (Tokyo: Iwanami Shoten, May, 1988), 2. Ogata is a major figure in Bashp studies, and there are few areas of haikai or Bashp research that do not bear his imprint. He figured prominently in many of the Bashp events described here. 5. There are several very good translations of the Oku no hosomichi. For the convenience of those wishing to study, teach, or simply enjoy Bashp in English translation, we have tried in this volume to cross-reference citations to translations of Bashp’s travel works and longer haibun with the recent renderings by David Barnhill in Bashp’s Journey: The Literary Prose of Matsuo Bashp (Albany: State University of New York Press, 2005). An attractive illustrated translation of Oku no hosomichi in a single edition is

18

6.

7.

8. 9.

10.

11.

12.

E    K      that by Cid Corman with Kamaike Susumu, Back Roads to Far Towns, Bashp’s Oku no hosomichi (New York: Grossman, 1968). Sora is named and identified in the narrative, and some of the hokku he composed en route are placed in the work under the name Sora. The work’s I-narrator, however, is not given a name and his verses (that is to say, verses by Bashp) are not identified by name. Ogata, “Oku no hosomichi sanbyakunen,” 1–9. In this article, Ogata describes several of the events and assesses their cultural/political significance. Much of the information here on the “happenings” presented by local governments and participated in too by the most important scholarly haikai association, the Haibun Gakkai, is anecdotal, based in large part on my own personal experience and observations at six of these Oku no hosomichi events. Ibid., 2. One of the four, for instance, was participated in by the well-known poet and critic, Poka Makoto; the distinguished Bashp scholar Professor Ogata Tsutomu; the prominent scholar of German literature Professor Koiso Masashi; the late Buson and haikai scholar Professor Leon Zolbrod, and myself. As was the usual practice with such events, all talks and proceedings of the symposium were published by Spka City, Ibaraki Prefecture: Dai nikai, Oku no hosoomichi kokusai shimpojiumu, kpenroku (The International Symposium on Oku no hosomichi, Proceedings) (Tokyo: Gypsei, 1990). Ogata, “Oku no hosomichi sanbyakunen,” 1–2, compares the Oku no hosomichi Boom with two earlier commercial/cultural booms triggered by popular, long running television specials, the “Oshin” and the “Date Masamune” Booms. The comparison is interesting in that in each case a whole nation seems to have had its attention captured by one cultural event. Neither the Oshin nor the Masamune craze was quite as persistent or varied, however, and neither had the political clout associated with the Oku no hosomichi Boom. For reporting on the original Oshin run and its popularity in Korea, China, South East Asia, and the Middle East, see The East, August, 1983; or Japan Times, March 24, 1984. The scholar who supplied the theoretical foundation (based on his academic research on “leisure time”), a personal connection (having been born in the Yamagata area), and considerable energy and imagination to the planning and production of many of these events was Matsuda Yoshiyuki, a professor at Tsukuba University and a research consultant to the government-funded Yoka Kaihatsu Senta (Leisure Time Development Center). For Matsuda’s theories on leisure and its connection to Bashp and the Japanese people, see his essays in Rejâsangyp o kangaeru (Thoughts on the Leisure Industry) (Tokyo: Jikkyp Books, 1993), 1–90 and especially 36–66. The symposium was the second of a series of four symposia presented in four different locations throughout the year by Yamagata Prefecture, the City of Tsuruoka, the town of Haguro, and the Ideha Museum in

I       

19

Haguro. The overall theme of the series was “The Three Sacred Mountains of Dewa and Japanese Spiritual Culture, Past, Present, and Future.” For an introduction to the celebrations and the proceedings of the four symposia see Matsuda Yoshiyuki, ed., Dewa Sanzan to Nihonjin no seishin bunka: kako, genzai, soshite mirai (Dewa Sanzan and Japanese Spiritual Culture, Past, Present, and Future) (Tokyo: Perikan, 1994). 13. See chapter 7 by Professor Horikiri in this volume. 14. There has been an extended intellectual debate on and opposition to the notion, stated in the extreme, that because of the collaborative nature of linked verse production, the renga or haikai poet should not be evaluated in the context of individual achievement. For an early, carefully considered, and extended argument made in opposition to Ogata Tsutomu’s views on collaborative literature (literature of the za ), see, for instance, Murata Boku, “Bashp, Hitotsu no hansetsu: za no bungaku no kokufuku” (Bashp, an Alternate View, Moving Beyond “Literature Produced through Companionship”), Kokugo Kokubungaku, Vol. 26, July 1977, 1–19. 15. Tokyo: Nihon Hpsp Shuppan Kypkai, 1988; Mori received the Akutagawa Prize for literature in 1974 for his novel Gassan (Mount Gassan, 1973); for more on the novel see note 1, chapter 8.

This page intentionally left blank

P   I

T A   T  

This page intentionally left blank



B¯   C  C Hori Nobuo Translated by Cheryl Crowley

W  H? As many people know, Matsuo Bashp was the greatest figure active in Japanese haikai during the latter half of the seventeenth century. Not only did he compose haikai poetry, but he created masterpieces in a variety of genres, including renku (linked verse), haibun (haikai-style prose), and haiga (haikai paintings). Moreover, his way of life was itself a work of art. If we were to sum him up in one phrase, we might call him a haikai no hito, a haikai person. In Kyoraishp (Kyorai’s extracts, 1704), a haikai treatise written by one of Bashp’s disciples Mukai Kyorai (1561–1704), the author writes that Bashp responded to a question about “the foundations of haikai” (haikai no motoi) in these words, “If you write a prose text with haikai in it, it becomes haikaibun [haikai prose]. If you compose a waka with haikai in it, it is haikai waka. If you make haikai part of your daily life, you are a haikai person.”1 As this passage shows, haikai is an idea that goes beyond material forms of art. That is to say, haikai is nothing else but a mode of thinking, a mode of speaking and acting, and a mode of human behavior that is influenced by a certain aesthetic. But what kind of aesthetic does haikai have? In the Chinese classic Han shu (Annals of the Han, first century CE) there is the definition, “haikai is the comic (kokkei).” That is to say, if Bashp is a haikai no hito, then he is also a kokkei no hito, a comic person. However, kokkei is not simply frivolous humor. On this point, at first glance, the section “Biographies of Comic Performers” (in Japanese, Kokkeiden) in the Chinese classic Shiki (Records of the Grand Historian, first century BCE) seems to imply that kokkei is silliness; however, it actually means that clowns and other performers who use kokkei to cleverly

24

H    N  

overcome obstacles are kokkei no hito. This is why both the words haiku and haiyu (actor) have the same element, hai (wazaogi, odoke: buffoonery, farce). The first time in Japanese literary history that the word haikai comes to be used is at the beginning of the tenth century, when fiftyeight “haikai no waka” were included in the poetry anthology Kokinshu (Anthology of Ancient and Modern Japanese Poems, ca. 905). The twelfth-century poet and scholar Fujiwara no Kiyosuke (1104–1177) wrote about these haikai no waka in Pgishp (Treatise on Arcane Matters) in the following terms: The word “haikai” is interpreted as “comic.” For this reason, everyone thinks it only means “jokes.” But is that necessarily so? Now it is thought that those who are kokkei are not on the proper path, and yet they are on the Path. Moreover, haikai is not the path of correctness, yet it holds to the highest principles.

The phrase michi ni arazushite shikamo michi o nasu mono ([those who are kokkei] are not on the proper path, yet they are on the Path) is clearly a logical contradiction, but to enlarge on its meaning for clarity’s sake, the sense of it is: “judged on the basis of accepted ideas about common sense, order, and values, it can’t be called correctness, but looked at in a broader perspective, now it is clear that the new viewpoints and values that kokkei offers are correct.” To put it a different way, kokkei is an innovative perspective or way of thinking that calls into question viewpoints, values, and aesthetics that have become habitual, fixed, and formalized, and uncovers new relationships between things.

T C B H  Z Given the stress placed on this kind of radically innovative viewpoint, naturally it might be expected that there would be attacks from the supporters of the established order and older aesthetic theories. And indeed there were. What kinds of response did haikai poets have against such attacks? Most of them formulated their counterarguments citing as their authority Kiyosuke’s phrase, “[those who are kokkei] are not on the proper path, and yet they are on the Path.” Among these counterarguments, one that merits particular attention was made by the founder of a haikai style slightly antedating Bashp, the Danrin haikai poet Nishiyama Spin (1605–1682). Spin was a

B    ¯     C      C   

25

distinguished intellectual of his day, a master of both renga and haikai. Refuting the charge that the haikai of his school was too wild and undisciplined, he writes: Above all the way of haikai is to put Falsity (kyo) first and Truth (jitsu) last. It is waka’s allegory (gugen); it is renga’s mad words (kypgen). “Using renga as our foundation, but forgetting renga” is the basic precept of our worthy predecessors. I have been traveling on this path for years. It is said, a certain Chinese person said, “to live fifty years is to know forty-nine of ‘not right’ (hi).” How much more so should one who has reached the age of seventy know one’s own “not right” even better than others? I know there is a principle (ri) in favoring the “not right.” However, in the world one finds wisdom and foolishness, righteousness and wickedness, greater families and lesser ones, samurai who act like townspeople and priests who fight, red eboshi hats, zukin hoods with trailing flaps, and flimsy garments made in Date2—all have their own capacities. You cannot understand people unless you share their thinking. There is old style haikai and contemporary style haikai, skillful and unskillful haikai, and there is no distinction between them. It is best to follow your own tastes. [Haikai] is the fictional language of dreams and illusions.3

There are two major points here: one, Spin emphasizes the fact that favoring the “not right” has a classical precedent; and two, in saying that the unconventional behavior of outlaws is something that we will not comprehend without sharing their way of thinking, he shows an attitude of sympathy for marginal people. But why does he say that it is good to favor the not right? According to Spin, it is because the Chinese philosopher Zhuangzi articulated this in the doctrine of “making all things equal.” A basic tenet of the Zhuangzi, the doctrine of making all things equal, suggests that because all things in the world arise from the same source, the Dao, there is absolutely no way to make distinctions between them. In light of the Dao, distinguishing “this” and “that” is meaningless; so everything depends on the individual capacities of the person who tries to choose between them.4 Moreover, the Zhuangzi states that because the person who is at one with the fundamental Dao becomes non-discriminating and accepts all things with equanimity, he or she is able to live life with authentic freedom. Accordingly, this doctrine attempts to completely overturn conventional values, rejecting human endeavors as harmful to nature, and seeing true usefulness in the things that people regard as useless. It is likely that the Zhuangzi’s ideas had a strong influence on Kiyosuke’s theory of

26

H    N  

kokkei being “not on the proper path, yet on the Path.” In this situation, “haikai” and “kokkei” inherently come close to the Zhuangzi. Furthermore, the profound philosophical principles of the Zhuangzi become the mechanism directly connecting haikai with the universe as a whole.

H P’ D At the beginning of the seventeenth century, haikai enjoyed great popularity as a new form of literature and of culture; at the same time, however, fierce debates arose among its practitioners and factions. Style, of course, underwent dramatic changes. Bashp commented on the conditions of the time in his epilogue to Tokiwaya no kuawase (Tokiwaya Verse Contest, 1680) by his disciple Sugiyama Sampu (1647–1732): It is said that during the more than four centuries between the Han and the Wei dynasties, poets, people of genius, and literary styles changed three times. Waka has renewed itself in every age. Haikai transforms itself every year; it becomes new again every month.5

There were lively controversies over what was innovative and what was outdated. In the preface to the haikai collection Azuma nikki (Eastern Diary, 1681), Bashp’s friend Tani Saimaro (1656–1738) advised that the best thing to do was to leave things to the Infinite: Lamenting for the dead in front of a blowfish; with a water rail, wishing it would peck at a melon. Imagining the sea in early summer rain (samidare); thinking of colored leaves in a winter shower (shigure); not knowing nature for oneself. We do not see texts that distance themselves from other texts. Do words say something, or do they say nothing? Are they different from the peeping of chicks, or are they the same? When one speaks of old and new, one should attend to the righteous and leave things up to the Infinite.6

The phrase “peeping of chicks” is a reference to the Zhuangzi, which has the theory that, when considered by universal standards, human speech and the peeping of chicks are the same thing. However, the declaration that in terms of clearly expressing the truth, the signs of human language do not have the same capacity as does the peeping of chicks is a truly astounding statement. What led haikai poets to this astounding statement was the Zhuangzi. In the same vein, Shuchikudp,7 another contemporary of Bashp, clearly links the

B    ¯     C      C   

27

Zhuangzi to his argument: The parables ( gugen) of the Zhuangzi say that if we declare the thousand differences and the ten thousand distinctions that arise in the mind are without one hair of individual direction, the “That” is the same as “This” and “This” is the same as “That,” and words in which falsehood and truth are mixed naturally return to the Way of Non-action.8

B¯    Zhuangzi Bashp’s strong admiration for the Zhuangzi began in his late thirties and lasted until the end of his life at the age of fifty. In a haibun that he wrote at age forty, Kasen no san (Comment On A Thirty-Six-Link Verse Sequence), he mentioned that “a poet who listens to the piping of Heaven is ideal.” The “piping of Heaven” (Chinese, tianlai; Japanese, tenrai) is a term that appears in the Zhuangzi, and it means the sounds of heaven or of nature. In the Zhuangzi, “listening to the piping of heaven” refers to taking in the diverse sounds of heaven and earth just as they occur. That is to say, listening to the piping of heaven means to listen with a non-discriminating frame of mind to the sound a human being makes while blowing a flute or that made by the wind blowing through trees. This acceptance of everything with the same standards and values is the total opposite of making distinctions between higher and lower, superior and inferior. It is accepting all things as natural and coming into harmony with them. It is, for example, letting go of the fear of death and treating both life and death as natural. In other words, it is an all-affirming point of view, which regards birth, maturation, aging, and dying, all as good things. The aforementioned haikai theory, not on the proper path, yet on the Path, was at odds with the established order and value system and its absolute authority, but it could easily be tied in with Zhuangzi’s doctrine of making all things equal. Of course, the true “haikai person” strove to keep a certain distance from order imposed by authorities. As is well-known, the fact that Bashp boldly chose an unconventional lifestyle of reclusion and travel was a result of precisely his insistence on freeing himself from conventionalized and routine points of view. Also, something else that Bashp wholeheartedly believed in was “creation” (zpka), what the teachings of the Zhuangzi’s philosophy called the work of the universe. We might also call creation the work of the universe’s movements. In the opening passage of Oi no kobumi (Rucksack Notebook, ca. 1690), which has been called Bashp’s most elegant expression of his ideas, there are the phrases “following Creation, make the four seasons your friends,” and

28

H    N  

“following Creation, return to Creation.” We might say that Bashp endeavored to always place himself in the very midst of creation.9

I   Zhuangzi  C  B¯ ’s W Bashp was neither purely a philosopher nor a religious thinker. He underwent many changes, but until the end of his life he ultimately remained an artist influenced by the haikai aesthetic. Thus, while Bashp viewed the possession of consistency of thought as important, the nature of the art he produced was probably much more significant to him. As mentioned, above all, his life itself was art. In any case, Bashp, who resisted social customs and conventional values and was determined to be in harmony with the dynamic force behind creation, strove always for an awareness of things as they fit into the larger framework of the universe. The remainder of this essay will focus on the question of what kinds of distinctive characteristics this brought to his work. The first of these characteristics is the strong effect that the permutations of time had on his work. The reason for this is that the universe’s dynamism is clearly made manifest in the shift from day to night, the waxing and waning of the moon, the transformations of the four seasons, and the myriad vicissitudes of time. This is why Bashp paid such careful attention to the seasons. It was not just because he thought this showed a highly refined sense of taste. The second characteristic is that Bashp found rich potential in “empty space” (yohaku, as in a painting). Bashp, who thought of things in the larger framework of the universe, was very well-aware that there are limits to the power that words have to express truth and perceptions of the world. The fact that he used few words and that haiku itself is ascetic in verbal expression stem from the fact that he knew that the world he could not describe in words would be all the wider for it. Accordingly, his use of empty space was neither insufficiency nor a disenchantment with words. Tosa Mitsuoki (1617–1691), a painter in Bashp’s own era, explained empty space in these terms: Do not fill up the whole picture with lines; also apply colors with a light touch. Some imperfection in design is desirable. You should not fill in more than a third of the background. Just as you would if you were writing poetry, take care to hold something back. The viewer, too, must bring something to it. If one includes some empty space along with an image, then the mind will fill it in.10

If we replace Mitsuoki’s “lines” and “colors” with Bashp’s “words,” this statement would express Bashp’s argument very well. The empty

B    ¯     C      C   

29

space in Bashp’s work must be filled in by the mind that senses what words cannot express. Let us try a psychological experiment. What is written in figure 1.1a? And in figure 1.1b? And, finally, in figure 1.1c? Each of the illustrations has wind written on it, of course. Now, a different question. Did you get the same impression from all of them? Or perhaps not? It is likely that each was different. The reason that the same word conveys differing impressions is, of course, the difference in the quality of the space beyond the words. As many people know, this is an experiment often conducted in Gestalt psychology. The site of meaning beyond the “figure” of letters is called the “ground.” Even though we focus our consciousness on the figure of a written word, there is also a ground surrounding it similar to empty space in a painting, and it exerts an influence on our apprehension of meaning. The influence of the ground is of great significance in art. In Bashp’s case, the empty space activates an important conduit that directs the mind toward a universe that is inexpressible in words. (a)

(b)

(c)

Figure 1.1 Gestalt experiment: (a) “wind” with white background (b) “wind” with red background (c) “wind” with blue background

H    N  

30

One of Bashp’s verses is: yamaji kite nani yara yukashi sumiregusa

coming along a mountain path what sweetness! violets11

To understand this verse, let us imagine the vast limitless space of the universe. Next, picture just a patch of violets right in the middle of a mountain path. Nothing else needs to be said. In an instant, the vitality of the violets harmonizes with the dynamism of the universe and the rhythms of life and shines vividly. But it doesn’t stop here. I always think that in order to better understand Bashp’s writings, the best thing to do is to read them as works whose imagery arises in the context of the whole universe. Finally, I would like to add something in order to forestall misunderstanding. Do not think that Bashp mastered perfection or consistency of form in his thought or his ways of expression. Rather, he was someone who spent his whole life without reaching that level and was plagued with uncertainty. Indeed, it was precisely because his doubts ran so deep that his works were able to achieve such great profundity. In the writings of his disciples, we see that this doubt haunted him even on his deathbed. His last verse was: tabi ni yande yume wa kareno o kakemeguru

taken ill on a journey— in withered fields dreams wander12

He was troubled with thoughts of how to revise the verse even then. And it is said that with his last painful breath, he whispered to his followers something like: “To be or not to be / That is the question.”

N This essay is based on a talk, with some revisions, presented by Professor Hori at the University of Maryland, November 5, 1994. 1. Kphon Bashp zenshu, Imoto Npichi, Miyamoto Saburp, Kon Eizp, and Puchi Hatsuo, eds., Vol. 7 (Tokyo: Fujimi shobp; reprint, 1989), 127–128; for more detailed commentary, see Okamoto Akira, ed., Kyoraishp hypshaku (Tokyo: Meichp Henkpkai, 1970), 256–258. 2. The hat, hood, and robe are garments associated with eccentrics or peculiar men-about-town. 3. “Afterword” to a solo one-hundred-verse haikai sequence, published in the anthology Oranda maru nibansen (1690); see Inui Hiroyuki, Iida

B    ¯     C      C   

31

Masakazu, Esaka Hiranao, eds., Danrin haikai shu, II Koten haibungaku taikei, Vol. 4 (Tokyo: Shueisha, 1972), 439–440. 4. For English translation of passages from Zhuangzi relevant to the doctrine see, for instance, Burton Watson, The Complete Works of Chuang Tzu, “Discussion on Making All Things Equal,” chapter 2 (New York: Columbia University Press, 1968), 39–40: Words are not just wind. Words have something to say. But if what they have to say is not fixed, then do they really say something? Or do they say nothing? People suppose that words are different from the peeps of baby birds, but is there any difference, or isn’t there? What does the Way rely upon, that we have true and false? What do words rely upon, that we have right and wrong? . . . Everything has its “that,” everything has its “this.” From the point of view of “that” you cannot see it, but through understanding you can know it. So I say, “that” comes out of “this” and “this” depends on “that”—which is to say “this” and “that” give birth to each other. But where there is birth there must be death; where there is death there must be birth. Where there is acceptability there must be unacceptability; where there is unacceptability there must be acceptability. Where there is recognition of right there must be recognition of wrong; where there is recognition of wrong there must be recognition of right. 5. Kphon Bashp zenshu, 394. 6. Iida Masukazu, Esaka Hiranao, and Inui Hiroyuki eds., Danrin haikaishu, I, Koten haibungaku taisei, Vol. 3 (Tokyo: Shueisha, 1971), 572. 7. Shuchikudp is probably a pseudonym for Okanishi Ichu (1639–1711), although it is also possible that Shuchikudp is a haikai poet influenced by Ichu. 8. Haikai wakubun (Haikai Hypothetical Questions, 1678). 9. For more on the term, “creation” (zpka), see chapters 2 and 3 by David Barnhill and Peipei Qui, respectively, in this volume. For translation of Bashp’s Oi no kobumi see David Barnhill, “Knapsack Notebook, Oi no kobumi,” in Bashp ’s Journey : The Literary Prose of Matsuo Bashp (Albany: State University of New York Press, 2005), 29–43. 10. Honchp gahp daiden (An Authoritative Summary of the Rules of Japanese Painting); for discussion, translation, and Japanese sources, see Makoto Ueda, Literary and Art Theories in Japan (Cleveland: Western Reserve University Press, 1967), 138–139; 262. 11. Kphon Bashp zenshu, Imoto Npichi, Yayoshi Kan’ichi, Yokozawa Saburp, and Ogata Tsutomu, eds., Vol. 6, Nozarashi kikp (Record of a WeatherBeaten Skeleton, 1685–1687), 60; Barnhill, Bashp ’s Journey, 20. 12. Kphon Bashp zenshu, Vol. 2, 109.

This page intentionally left blank



Z ¯ : T C  B ¯’  V   N   A David Landis Barnhill

N

ear the beginning of his travel journal “Knapsack Notebook” (Oi no kobumi), Bashp makes an important and complex statement about nature and art. Saigyp’s waka, Spgi’s renga, Sesshu’s painting, Rikyu’s tea ceremony— one thread runs through the artistic Ways. And this artistic spirit is to follow zpka, to be a companion to the turning of the four seasons. Nothing one sees is not a flower, nothing one imagines is not the moon. If what is seen is not a flower, one is like a barbarian; if what is imagined is not a flower, one is like a beast. Depart from the barbarian, break away from the beast, follow zpka, return to zpka.1

I have left untranslated the key term zpka, a rather uncommon word derived from Chinese sources. The conventional translation of this term is “nature,” found in the standard translation of his journals by Nobuyuki Yuasa, as well as in renderings by Makoto Ueda, Eiichi Hayashi, and Eleanor Kerkham.2 Thus, the final line has been read as “follow nature, return to nature.” But the translation as “nature” does not solve the issue of what zpka means. The notion of nature is a cultural construction, not a universal given.3 And it is a personal construction, a distinctive element in any individual worldview. What, then, is nature in Bashp’s literary writings? This is a huge issue with many facets, only some of which I will be able to consider here, in particular his notion of nature’s creativity and its relationship to artistic creativity. In doing so I will consider other related issues, for there is more at stake in interpreting this passage

34

D    L      B    

than just Bashp’s understanding of nature. One such issue is the nature of the meaning of his texts: how do his texts (or any others) mean, particularly in relationship to other texts? Tied directly to that issue is another: what is the context(s) in which we should place Bashp’s texts as we interpret them? I think the common, implicit answer in the West to these questions has been that Bashp’s texts are relatively self-sufficient and can be interpreted without significant attention to texts of other writers. For many Western commentators, if his texts are placed in a context, it generally has been the medieval Japanese literary tradition and Zen Buddhism. Yet, his world is more complex and multifaceted. It certainly does partake of the Japanese literary and religious traditions. But it also participates in the Chinese religio-aesthetic tradition, which includes Daoism and Confucianism, as well as aesthetic ideas and ideals in Chinese poetry and painting.4 The relative neglect by Western commentators of these aspects of his broader cultural context has led not only to a narrow interpretation of his worldview, but also to an inability to recognize (and thus to translate) important texts.5 In response to this situation, we need to maintain an exploratory perspective on Bashp, looking for ideas and attitudes that do not conform to our conventional view, watchful for aspects of a worldview richer than we have imagined, examining texts that have been overlooked. And we need to place his works in a broader context. Such an approach is necessary because the meaning of his texts is dependent on our understanding and use of this more comprehensive intertextual framework. Meaning is not simply textual (confined to the text at hand) but cultural and intertextual, and we need to consider those texts that shed light on the meaning of Bashp’s writing. The choice and range of the cultural context in which we place his works are crucial. A brief comment should be made here about an intertextual approach. There are different ways of conceiving of intertextuality. In a broad sense of the term, the influence of one text or writer on another and allusions to older texts are kinds of intertextuality. However, as Haruo Shirane has pointed out, contemporary approaches to intertextuality do not depend on the discovery of direct influence or reference.6 Rather, they involve the articulation of the cultural context within which a writer or text should be read, a context that helps to define the meanings of the text. Bashp’s texts make clear that the traditions of Chinese aesthetics and religious thought are contexts within which he developed his ideas and attitudes.

T   C       B   ’ ¯  V   N      A 

35

Z¯   B¯ ’ W The importance of the term zpka in the passage cited earlier is seen in the fact that it is used three times, twice in the emphatic conclusion. This term also appears three other times in his writings. Later in “Knapsack Notebook” is the following line: sanya kaihin no bikei ni zpka no kp o mi . . . 7 The passage has been translated by Yuasa as “it was a great pleasure to see the marvelous beauties of nature, rare scenes in the mountains or along the coast . . .”8 His translation presents zpka as beautiful landscape, the object of aesthetic perception. But the passage actually says “I gazed at the accomplished skill of zpka in the beautiful scenes of mountains and fields, sea and shore.” Yuasa has ignored the term kp (skill) and made beautiful scenes the object of attention rather than the skillfulness of zpka. He has, in effect, conventionalized Bashp, transmuting his particular vision into a generic Western concept of nature as landscape. Significantly, Yuasa also has ignored the very Chinese structure of polar opposites, the yin-yang of mountains and fields, the sea and the shore.9 The term zpka also occurs in Bashp’s last journal, “Narrow Road to the Deep North” (Oku no hosomichi). The passage concerns Matsushima, “Pine Islands,” a place famous for its beauty. At the conclusion of a description of the islands, we find the sentence: zpka no tenkp, izure no hito ka hitsu o furui, kotoba o tsukusamu.10 Here, Yuasa’s translation is closer to the original: “My pen strove in vain to equal this superb creation of divine artifice.”11 However, he again makes the scene (“creation”) the object, rather than the “heavenly skill” of zpka. The passage reads: “the heavenly skill of zpka: who could fully capture it with [painting] brush or [poetic] words?”12 There is another version of this passage on Matsushima, found in one of his many haibun.13 After stating that there are “islands upon islands of various shapes and sizes,” Bashp adds, kikyoku tenkp no myp o kizaminaseru ga gotoshi: “as if heaven’s extraordinary artistry had carved a wondrous mystery”.14 The term zpka is not used here, but we do find again the notion of divine skill (tenkp) as an active and creative force. It is worth noting that a number of Chinese thinkers associated the term zpka (Chinese, zaohua) with the term myp (wonder, mystery, subtlety, excellence: miao in Chinese).15 There is one other use of zpka in Bashp’s writings, in a haibun written during the journey that gave rise to “Narrow Road to the Deep North.” “The sound of water, the call of birds, the green of pine and sugi too in fine detail and deep color: scenic beauty of consummate

36

D    L      B    

skill. The greatness of zpka—does it not bring joy?” (bikei takumi o tsukusu. Zpka no kp no pinaru koto, mata tanoshikarazu ya).16 Again, emphasis is placed on the skillful workings of zpka. Perhaps the most conspicuous part of all these passages is the distinction made between the beautiful scenes of nature and zpka. Zpka is not nature as landscape but that which fashions such scenes and their transformations,17 and it is characterized by artistic creativity. While beauty is the principal quality of the scene, skill is the primary quality of zpka. But such a close reading of these passages is not sufficient to understand the term zpka. To uncover what meaning this term has in Bashp’s writings, we need to do two things: consider the meanings of zpka in earlier uses, and look to other passages in Bashp’s writings that are related to such “divine creativity.”

B  ¯   C L T One of the characteristic features of Bashp’s sense of his tradition is his presentation of Japanese and Chinese poetry as a single, multiplex tradition. While the famous passage from “Knapsack Notebook” quoted at the beginning of this chapter delineates a purely Japanese tradition (Saigyp, Sesshu, Spgi, and Rikyu), in numerous other writings, Bashp closely associates Japanese poets (particularly Saigyp) with Chinese poets. Perhaps the clearest example is his introduction to the haikai no renga collection Minashiguri (“Shriveled Chestnuts”) where he outlines four poetic tastes, three linked to Chinese poets, the fourth to Saigyp. This one book called “chestnuts” has four tastes. The first is like drinking the wine of the poetic heart of Li Bo and Du Fu. Another is like eating the rice gruel of Han Shan’s dharma. These poems have the subtlety of something seen far away or heard from a great distance. Then there is the rare flavor of lonely poverty and elegance: it is found at the mountain home of Saigyp in the wormy chestnuts that people ignore.18

The fourth taste “encompasses all the feelings of love” and is associated with the Chinese poet Bo Juyi (772–846). In addition to bringing together the Japanese and Chinese poetic traditions, Bashp saw poetry as one part of a larger aesthetic complex that included painting and calligraphy.19 On two occasions, for instance, he wondered about the ability of art to capture nature and spoke of poetry and painting as if they were two modes of a single artistic pursuit.20 In one case, as we have already seen, he speaks of the

T   C       B   ’ ¯  V   N      A 

37

splendor of the Matsushima area in “Narrow Road to the Deep North” and wonders: “the heavenly skill of zpka: who could fully capture it with (painting) brush or (poetic) words?” In an early haibun he speaks of the beauty of Mt. Fuji, and says: Even poets can’t exhaust this scene in verse; those with great talent and men of letters give up their words; painters too abandon their brushes and flee. If the demigods of faraway Gushe mountain were to appear, I wonder if even they could succeed in putting this scene into a poem or a painting.21

But, perhaps, Bashp’s clearest expression of the mutuality of poetry and painting is found in a later haibun about his disciple Morikawa Kyoriku (1656–1715). Richly talented, he loves both painting and poetry, and I thought I’d assay his interests. “Why are you fond of painting?” I asked. “Because of poetry,” he replied. “Why do you love poetry?” “Because of painting.” Thus these two arts have one use. Yes, “It is shameful for a gentleman to have many accomplishments,” so turning two arts to one purpose is praiseworthy indeed. In painting he is my teacher; in poetry I instruct him and he is my disciple. My teacher has penetrated the spiritual depths of painting and his brush moves with wondrous subtlety. I cannot take in all the mysterious profundity of his works.22

Bashp’s concern is not simply with this aesthetic complex, however. References to Chinese religious thinkers abound in his prose. The most common figure is the Daoist Zhuangzi,23 but Confucius, NeoConfucians, and Confucian values such as filial piety are also integrated into his world. Like Chinese literati, Bashp considered the artistic and religious traditions as one multiplex religio-aesthetic tradition. One of many examples is found in an important haibun called “Postscript to ‘Comments on the Bagworm,’ ”24 which provides an introduction to a Chinese poem and essay written by Bashp’s friend Yamaguchi Sodp. Bashp compares Yamaguchi’s work to that of several Chinese poets, including Su Shi (1037–1101), and then notes that Sodp’s piece speaks of Chinese examples of filial piety and of the need to “look again to the heart of Zhuangzi.” Bashp considers this combination of poetry, Confucian values, and Daoist vision to be an ideal. Another example of associating poetry and religious thought is found in a late haibun lamenting the death of his disciple Matsukura Ranran (1647–1693). There, he praises his disciple for having mastered Laozi and Zhuangzi and having “frolicked” with haikai poetry—an exemplary combination for Bashp.25

38

D    L      B    

The point I want to make is that Bashp did not have the propensity to categorize and separate cultural traditions as we do. He saw as one tradition what we see as poetry and painting; the arts and religious thought; the Chinese and Japanese traditions; and Buddhism, Daoism, and Confucianism. And he explicitly placed himself in that multiplex tradition. His early haibun, for instance, emphasize the poverty and wabi (aesthetically refined poverty) of his condition and its relation to that of Chinese poets. In “Sleeping Alone in a Grass Hut,” he feels such an identification with Du Fu and Su Shi that he feels he is listening to the same rain that poured on them. The Elder Du wrote a poem of a thatched hut tearing in the wind. Then old man Su Shi, moved by the lonely poverty in that poem, wrote a verse about a leaking cottage. Now I listen to their rain on the banana leaves, lying alone in my grass hut. banana plant in a windstorm: a night of listening to rain in a tub bashp nowaki shite / tarai ni ame o / kiku yo kana26

Bashp thus continues the tradition of poems on huts in a storm, and in doing so, places himself in a lineage with Du Fu and Su Shi. In a much later haibun (1692) that is highly Chinese in character, Bashp states that In moments of leisure, I lean my elbow on it, and forgetting myself and breathing deeply, I cultivate my qi. In quiet times I open a book and search for the spirit in the sage’s mind and the wise man’s genius. In moments of tranquility I take out my brush, entering the inner heart of Wang Xizi and Huai Su.27

The first sentence is taken from the opening of the second chapter of the Zhuangzi. Bashp then sets his ideal as communing with—and entering into—both the wisdom of the sages and the creative heart of calligraphers. Bashp’s identification with the Chinese religio-aesthetic world can be seen in another aspect of his prose writing: his conception of space. His views of both space and time are highly complex and I can touch on just a few aspects that are relevant to our subject here.28 Bashp frequently refers to Chinese places, particularly mountains, often tying together Japanese locations with Chinese. In some cases he simply associates Chinese and Japanese locales. In others he presents a Japanese site as having the similar nature and power as certain special

T   C       B   ’ ¯  V   N      A 

39

sites in China. In the beginning of the haibun “On Mount Fuji” cited earlier, Bashp associates Mt. Fuji with mountains of Daoist legend. “Mount Kunlun is said to be far away, and in Mount Penglai and Mount Fangzhang dwell Daoist immortals. But right here before my eyes: Mount Fuji’s great peak rises from the earth.”29 In still other works he identifies a mountain in Japan with sacred spaces in China. In “Journal of Bleached Bones in a Field” (Nozarashi kikp), for instance, he presents Yoshino as equivalent to a mountain in China. The passage has a Chinese structure, with strong parallelism. I wandered alone into the heart of Yoshino. The mountains were so deep. White clouds lay piled on the peaks, and misty rain filled the valley. The woodsmen’s tiny huts were scattered all around, and the sound of wood cut to the west echoed on the east. Temple bells struck to the base of my heart. From of old many who abandoned the world and entered these mountains fled into Chinese poetry, took refuge in Japanese verse. Surely one can call this Mount Lu, like the mountain in Cathay.30

Again, note the identification of the Chinese and Japanese poetic traditions. In other passages, Bashp suggests that he is able to make a “spirit journey” to Chinese places. Stephen Owen has discussed the nature and importance of the spirit journey in Chinese poetic theory. “On the spirit journey, the poet can instantly encompass all time (in the Chinese sense conceived as past and present) and space.” “. . . Imagination is a spirit journey, a movement through space. . . .” Concerning the “spirit thought” described by Liu Xie (c. 465–c. 520), Owen notes that “in its ‘finest’ or ‘most subtle’ form (miao, presumably the fine points and subtle changes, that which is essential but invisible to the ordinary senses), ‘spirit wanders with things.’ ”31 Similarly, in his most famous haibun, “An Account of the Unreal Dwelling,” Bashp speaks of his spirit wandering with things: “My spirit rushes off southeast to Wu and Chu; my body stands at the Xiao and Xiang Rivers and Lake Dongting.”32 In such writings, he is (meta)physically locating himself in China. The way Bashp views the landscape is also presented as Chinese in character. In the haibun “An Account of Eighteen View Tower,” he relates the estate of a friend to two different sets of famous views in China: “The eight views of the Xiao River and the ten sites of the Xiang River are experienced together in the one flavor of the cool wind.”33 The motif of the Eight Views and similar combinations of aesthetic locales began in the Northern Song Dynasty (960–1126) by

40

D    L     B     

Song Di (fl. eleventh century) when he painted eight views in Hunan Province. These became known as the Eight Views of Xiao and Xiang Rivers and is a common theme in painting and poetry in China and eventually in Japan. They were also incorporated into the famous Edo period garden at Shisendp. In this haibun, Bashp similarly incorporates this Chinese way of seeing the landscape into his own writing while suggesting that in the landscape he beholds he can experience eighteen famous views in China.34 Yet another aspect of Bashp’s “Chinese” way of viewing nature is his use of the motif of climbing a tower to view the landscape. This theme was very common in Chinese poetry. Hiroyuki Suzuki has noted that two-story structures were uncommon in Japan until the Edo period, when they became popular among bunjin (Japanese literati), with the Tower for Whistling at the Moon at Shisendp being a prime example.35 In several haibun, as well as in the Matsushima section of “Narrow Road to the Deep North,” Bashp climbs towers to view a scene, a physical location and visual perspective on the landscape that situates him in the Chinese aesthetic tradition.36 In his literary prose, then, Bashp presents himself as a Sino-Japanese literati. He explicitly locates himself and the Japanese poetic tradition within the Chinese multiplex of the religio-aesthetic tradition. To understand his view of nature, and particularly his notion of the Chinese-derived term zpka, we need to place his texts within the context of that tradition. We must ask, then: what meanings has the term zpka had in the Chinese tradition?

T C U  Zaohua  R T The term zpka comes from the Chinese zaohua, which might literally be translated as “make-transform.” The first significant use of the term is found in several humorous passages in chapter six of the Zhuangzi where it appears along with a cognate term zao wu zhe (“that which makes things”).37 Both terms are translated by Burton Watson as “Creator.” Suddenly Master Lai grew ill. Gasping and wheezing, he lay at the point of death. His wife and children gathered round in a circle and began to cry. Master Li, who had come to ask how he was, said, “Shoo! Get back! Don’t disturb the process of change!” Then he leaned against the doorway and talked to Master Lai. “How marvelous the Creator is! What is he going to make of you next?

T   C       B   ’ ¯  V   N      A 

41

Where is he going to send you? Will he make you into a rat’s liver? Will he make you into a bug’s arm?” Master Lai said, “The Great Clod burdens me with form, labors me with life, eases me in old age, and rests me in death. . . . Now, having had the audacity to take on human form once, if I should say, “I don’t want to be anything but a man! Nothing but a man!” the Creator would surely regard me as a most inauspicious sort of person. So now I think of heaven and earth as a great furnace, and the Creator as a skilled smith. Where could he send me that would not be all right?”38

The notion of the creator in this passage is complex and ambiguous. Several points are worth noting here. First, life involves a process of transformation that is unpredictable, sometimes wild, and potentially destructive. Second, this process is the movement of life itself and thus something we should accept, value, and follow, even if it means our death. Third, this creation refers not to the coming-into-being of the universe at the beginning of time, but rather to the ongoing transformations of life.39 Fourth, the process involves some kind of agency and skill that can’t simply be reduced to the natural world. It is not a simple “naturalism” in which things act “by themselves.” On the other hand, this skilled agency is not a transcendent deity separate from the natural world, and thus “the Creator” is misleading as long as Western connotations are involved. Zhuangzi’s creator, then, refers to a quite vague sense that the transformations of life involve a kind of skillful agency. This “sense” remains undefined because, I think, any attempt to concoct a specific metaphysical belief would distort the nature of the creative process that is in and of this world but that involves some kind of proficiency and control. Norman Girardot has pointed out that the Zhuangzi is generally critical of attempts at such metaphysical specificity. He notes “the repugnance Zhuangzi has for contrived intellectual ‘disputations’ [bian]. For the Zhuangzi all such controversies are only hollow linguistic games. In the first seven chapters of the Zhuangzi there is a refusal to talk theoretically about what can only be experienced . . .”40 The Creator should be seen as a metaphor for an experience of a distinctive quality of the natural world, rather than as the name of a defined metaphysical being. Centuries after Zhuangzi, Chinese poets and painters frequently referred to this notion of the zaohua, and its complexity and ambiguity remained in their works. For the poet–painter Su Shi, for instance, “the clear breeze over the river, or the bright moon between the hills, which our ears hear as music, our eyes see beauty in—these we may take without prohibition, these we may make free with and they will

42

D    L    B    

never be used up. These are the endless treasures of the Creator, here for you and me to enjoy together!”41 That the creation of zaohua concerns transformation rather than origin is seen in the following passage by the Yuan dynasty painter Tang Hou: “Landscape is a thing incorporating all the excellences of creation [zaohua], and it is inexhaustibly protean in that it takes on different appearances in light or darkness, shade or shadows, fair or rainy weather, cold or heat, morning or evening, noon or night.”42 One aspect of the functioning of the creator is that it works spontaneously. Su Shi’s brother Su Che (1039–1112) praised a painting of bamboo by Wen Tong saying that “Truly they are things produced spontaneously, which only natural creation [zaohua] could make. . . . In its being twisted and straight, crosswise and slanting, thick and thin, low and high, you have seized the secret thoughts of the Creator [zao wu] and produced a sense of life in a single morning. Can you indeed be a man of Tao?”43 Important for Chinese artists, and for Bashp, is the close relationship between the creativity of zaohua and that of art. Writing about the skill and spontaneity of the painter Zhang Dunjian, Bo Juyi stated that “I realized that learning what is in one’s bones and marrow is achieved by mental art, and skill matching creation [zaohua] comes from natural harmony. Chang merely received from his mind and transmitted to his hand, and it was so without his being conscious of its being so.”44 At times, artists were so impressed with the creativity of art that it seemed superior to the creativity of nature. The poet Mei Yaochen (1002–1060), commenting on a painting of plants and insects, stated that “And so I learn that the Creator’s power / can’t match the agility of the artist’s brush.”45 Stephen Owen has made an important comparative statement concerning zaohua and artistic creativity. Because it is particularly relevant to our understanding of Bashp’s notion of zpka, it is worth quoting at length here. Speaking of a passage in the Wen fu (RhymeProse on Literature) by Lu Ji (261–303), he states that the passage and a few other roughly contemporary statements constitute the beginning of a tradition in Chinese literary thought in which the poet becomes equivalent to the forces of Ongoing Creation [zaohua] . . . [T]his notion of the poet as Zaohua is much closer to the Western notion of literary “creation.” The similarity can, however, be deceptive. The stress here is not on voluntary creation ex nihilo, but on a comprehensive and animate whole in which the operations of a poet’s mind reenact the animate totality of Nature. . . . Lu Ji does exult

T   C       B   ’ ¯  V   N      A 

43

in the writer’s power, but it is the power to be Nature, not to make that “second nature” whose identity comes from its distinction from the first Nature.46

The notion of zaohua was referred to in the Neo-Confucian tradition as well. In fact, it was through Neo-Confucian commentaries on the Zhuangzi that Bashp became familiar with the ideas of the early Daoist thinker. Ogata Tsutomu has discussed the role of zaohua in NeoConfucian philosophy, noting that it is not nature per se but that which governs its movements. In Neo-Confucian thought, the activity of zaohua is qi (Japanese: ki), which is always moving and bringing things into existence. This power of creation is controlled by li (Japanese: ri), an unchanging principle of movement, which provides order to life in the form of patterns of growth and flourishing.47 Bashp’s use of zpka and his conception of nature must be understood in light of this rich and complex tradition, though we cannot, of course, assume his understanding of the term is the same as that of any particular Chinese thinker. We need now to consider his use of the term and other related ideas in his writings.

B ¯ ’ U  Z¯ O ka  L   C T We can now understand more fully Bashp’s call to “follow zpka, return to zpka” in the beginning of “Knapsack Notebook.” For Bashp, zpka does not mean nature as a specific collection of phenomena (e.g., a forest), a location (e.g., the country or wilderness), a specific scene, or nature as a whole.48 Nor is it, as Okazaki has noted, some metaphysical reality separate from the natural world.49 While Bashp, like Chinese thinkers before him, was not precise concerning its nature, I think we can suggest two related ways of conceiving of it. First, we can speak of it as a creative force in nature; it is an energy that transforms the world in a way that makes the world flourishing and beautiful. Second, we can think of it in terms of disposition, that is, the tendency and ability of nature to produce skillfully creative transformations. The notion of force parallels the idea of cosmic qi, while disposition is closer to the Neo-Confucian idea of li, the harmonious pattern of growth and flourishing in the world.50 The notion of force suggests the creative dynamism of the notion of zpka, while disposition suggests the brilliant pattern of the transformations.51 The notion of disposition is a useful concept because it can suggest the ability to act skillfully without conscious, intentional, rational control

44

D    L      B    

over the actions.52 Athletics can provide a model of this mode of acting. The split-second reactions of a great basketball player adjusting to several defenders on the way to the basket exhibit a spontaneous skill that allows him or her to avoid the defenders and make the shot. Rational, intentional decision making based on plans or principles would be unproductive in such a situation. Yet, there is skill and a kind of control involved. Zpka is, I suggest, a controlling agency not in the sense of a being that consciously creates or directs nature. Rather, it is the creative force of nature that has the spontaneous tendency and ability to exhibit transformations that are beautiful. These transformations occur at different levels, from the four seasons (as seen in the “Knapsack Notebook” passage) to the changes in a scene that occur from moment to moment. With these ideas in mind we can reconsider the anthropomorphic component of the notion of zaohua/zpka. I take the anthropomorphic language as metaphoric, a technique of suggesting primarily two things: (i) the awesome and wonder-full skill of nature at creating a vitalistic beauty, displaying a proficiency so marvelous that it seems to transcend the individual thing or moment; and (ii) the participation of each thing and each scene in a single cosmic energy and pattern, which manifests itself in ongoing creative transformations. The skill and “control” of this cosmic energy is the spontaneous disposition of nature rather than the imposed direction of a separate metaphysical being.53 The skill and beauty involved in the metamorphoses of nature suggested to Bashp, as well as to Chinese thinkers before him, that there is a parallel between art and the creativity of nature. In both cases, there is a “controlling agency” only in the sense of an ingrained disposition to create works of art in a skillful way without willfully trying to.54 Artistic creativity is essentially the same mode of activity as nature’s ongoing creative work. As Owen notes, the true artist has the power to be zpka, to perform as nature performs, and the search for artistic creativity is the search to be nature in this sense, to return to and follow its mode of creativity. In order to suggest these meanings in the term zpka, I translate the term as “the Creative.”55 Such a term (i) distinguishes between zpka and the natural world per se; (ii) suggests the creative force and disposition of nature; and (iii) avoids the connotations found in the term “Creator” of a separate deity who originated and controls the world.56 What of the anthropomorphic aspect of this term in the case of Bashp? One passage, the Matsushima section of “Narrow Road to the

T   C       B   ’ ¯  V   N      A 

45

Deep North,” does seem to suggest this dimension of the term. Before referring to the creative, it mentions a Shintp deity: “Is this the work of Pyamazumi from the age of the gods? The heavenly skill of the Creative: who could fully capture it in brush or in words?”57 Pyamazumi, the child of Izanagi and Izanami (creators of the Japanese islands), was associated with mountains. This passage is significant in part as an example of the association of Shintp with Chinese thought. The Shintp–Buddhist syncretism in Japan is well-known, but the blending of Shintp with Chinese (i.e., Daoist and Confucian) thought is often overlooked, although certain Neo-Confucian thinkers in the Tokugawa period show such a blending, and Shintp was, in fact, influenced by both Daoism and Confucianism from the early period.58 For our purposes here, however, the passage is important for the issue of anthropomorphism in the creative. Does the Shintp reference imply that the creative is some kind of “creation deity?” I think the answer is no. The other references to the creative in Bashp’s writings lack any clear sense of the anthropomorphic, and particularly the passage at the beginning of “Knapsack Notebook” would make little sense if interpreted in terms of a creation deity. His principal focus is on the ongoing transformations from moment to moment and season to season. I would argue that in this particular passage Bashp has complemented his fundamentally Chinese view of creation as transformation with a suggestion of the Shintp sacrality of the site. The importance of the role of Shintp and folk religion in Bashp’s religiosity has, I have argued, been underestimated.59 This passage is evidence of that element in what we might call his religious amalgam, for it is no systematic syncretism. His religious world included Buddhism, Daoism, Confucianism, Shintp, and folk religion, but there was no attempt to create some kind of coherent philosophical system. While it is important to recognize the sophistication of the metaphysical elements in his worldview, we should not impose our Western tendency to metaphysical specificity or logical consistency on the complexity of Bashp’s experience or the multifaceted expressiveness of his language. Given this view of the creative, to “return to zpka” is not to dwell in the wilderness, return to country living, or go sightseeing in nature. It is a call for people to recognize and follow this natural mode of creativity. Great art is based on this creativity, and thus the true artist is a companion of nature’s transformations. However, the passage is not directed simply to artists but to all who would truly discern

46

D    L      B    

beauty. All of life is a manifestation of the creative and is therefore beautiful. The moon and flower mentioned in the passage suggest nature’s beauty. Those who fail to see the moon and flower of all of life lack the religio-aesthetic vision of the creative in and of nature.60 This association of nature’s creativity with human creativity has an interesting implication. In the West we are used to the dichotomy of nature versus culture. The natural is that which has not been formed by culture, and conventionally the barbarian and the beast have been considered more natural than (though “beneath”) those in cultured human society. But for Bashp, this is not the case. The height of human culture, artistic creativity, is precisely the natural: it follows and embodies the creativity of nature. To be cultured is to have the capability to be natural. To lack culture—to be a barbarian or a beast—is to be unable to discern nature’s beauty or follow its creativity. Only those highly cultured in human art can truly perceive and act according to nature. And only those who intimately follow nature can be considered the great bearers of culture. Because Bashp has this vision of the creative, he can truly see nature’s beauty and participate in its mode of creativity.61 Despite the connection between nature’s creativity and human art, at times Bashp, like many Chinese poets and artists before him, questioned the artist’s ability to capture nature’s beauty. As we have seen, in the Matsushima section of “Narrow Road to the Deep North,” he wonders: “the heavenly skill of zpka: who could fully capture it with brush or words?” In the haibun “On Mount Fuji” translated earlier, he raises the same question in the context of Daoist legend. While this passage is in part exuberant hyperbole, the issue of the inability of words to capture reality is an important one. Why should this be a disturbing issue? Bashp is not explicit about the source of his concern. One possible reason might be the very transformations of nature, in which there are a hundred scenes in a brief moment. Stephen Owen, speaking of Lu Ji’s Wen fu, notes that The outer world is always in transformation, coming-to-be. Given in written language, the things of the world are torn out of transformation and fixed: in a text the words “autumn leaf” will never fall to the ground and decay, to be replaced by a spring leaf six months later. . . . That very power of transmutation in which the poet exults also fails to recreate the outer world in its most important attribute— the capacity for change.62

Related to this issue is a term Bashp uses in the Mt. Fuji haibun and in many other places, a usage that is another indication of the importance

T   C       B   ’ ¯  V   N      A 

47

of the Chinese context for his writings. The term is tsukusu (“exhaust”), which appears twice in this passage (translated in the hokku as “fulfill”). It is a term frequently used in Chinese aesthetics for bringing a creative work to culmination.63 What is it for a poet to exhaust a scene—and is it possible? The implication of this particular passage is that art cannot exhaust the scene, cannot artistically fulfill or capture a scene of rich transformations in poems or paintings.64 And yet the concluding hokku suggests that nature is its own fulfillment— in and through each moment of transformation. Here, at least, Bashp seems to be suggesting that artistic creativity is inferior to that of the creative in failing to represent nature’s flourishing transformations.

C: N’  H The creative is not Bashp’s only reference to what we can call his Chinese-based metaphysics of creativity. An examination of these related passages can give us a fuller sense of his notion of nature’s creativity. Bashp wrote a haibun praising a kasen (a haikai no renga sequence) by the poet Matsuyama Seikai. In it he makes sustained use of Daoist imagery, and in particular refers to an image from the Zhuangzi for the vitality of life. A windstorm from Pine Mountain in the province of Iyo has blown upon the withered leaves at Bashp Cave, and its voice sings a linked verse. Ah, the distant sound of wind, the rustling of leaves: like the ringing of a jewel, the echoes of gold and iron—sometimes blowing forcefully, sometimes soft. They arouse such deep feelings, bringing forth tears. Through a thousand holes comes a fierce roar, the echoes ever shifting, the meanings of each verse distinct. This is Heaven’s spontaneous piping, a wafting wind that tears the banana leaves.65

The reference to “a thousand holes” comes from the second chapter of the Zhuangzi on the topic of the “piping of Heaven.” “The Great Clod belches out breath and its name is wind. So long as it doesn’t come forth, nothing happens. But when it does, then ten thousand hollows begin crying wildly.”66 The Great Clod is a term that suggests the undifferentiated reality from which the phenomenal world springs.67 It is also associated with the notion of creator.68 Each thing in the world is presented as a hole through which the breath (qi) of the Great Clod blows. Humans are included in this: we too are holes through which heaven’s wind blows (note that his first reference to “bashp ” [banana] is put in terms of a cave). The central function of

48

D    L      B    

the passage is to present an image for the “anima” (literally “breath” of life) of things. More specifically, the passage suggests metaphorically three things about the natural world. First, each thing’s vital energy or qi (which means both breath and vitality) is given to it; it is not one’s own creation. Since it is not one’s own creation and is vital, it seems as if it comes from some agent, but the nature of that agency cannot be specified. Second, the passage suggests that the breath of vitality is shared by all things: everything in life is the piping of heaven coming from the single breath of the Great Clod. Third, in addition to a fundamental cosmic unity, the passage highlights the distinctness of each thing: while all things share the same breath, the Great Clod blows on the ten thousand things in a different way, so that each can be itself.69 The distinctness of all things within an essential unity is important for Bashp in part because he is writing about a haikai no renga sequence, in which the kasen is a single sequence and yet “the meanings of each verse [are] distinct.” In a renga sequence, each verse must be connected to the verses both preceding and following, but must also be independent so that each verse can be “detached” from the preceding one when it is read in terms of the following verse.70 The emphasis on distinctness also seems part of Bashp’s vision of nature. In a number of writings he emphasizes the individuality of things and the detail of perception. One example is found in a passage noted earlier: “The sound of water, the call of birds, the green of pine and sugi too in fine detail and deep color: scenic beauty of consummate skill. The great accomplishments of the Creative—bringing such joy.”71 The detail and thus the distinctness of everything is, I would argue, related to his “Zhuangzi-ian” view of nature. Distinctness is seen also in the haibun “Matsushima,” which states that “Each single pine is flourishing, all so lovely, gorgeous, beyond words.”72 Not only are all things distinct, they are always in flux. Again, this notion is highly appropriate to the alterations of image and mood that characterize a renga sequence. In another haibun (quoted earlier concerning “the four tastes”), which also concerns a haikai no renga, Bashp repeats this theme when he praises the fact that in the sequence “the echoes ever shifting, the meanings of each verse distinct.” The distinctness of each hole is put in terms of producing a different “sound,” just as a flute produces different sounds from the single breath of the flutist because the holes are different. This image of the flute suggests that the creativity of nature is like human creativity in music. As in the case of the image of the creative, what nature does is parallel to what humans do in art, the implication being that art

T   C       B   ’ ¯  V   N      A 

49

should imitate the “spontaneous piping” of nature. In this haibun, Bashp does present a theme not found in the Zhuangzi passage: this creativity of nature and of humans “gives rise to deep feelings” (hito ni kokoro o tsuku). This phrase is a reference to a Saigyp poem in the Shinkokinshu (#299): oshinabete mono o omowanu hito ni sae kokoro o tsukuru aki no hatsukaze

Even in a person most times indifferent to things around him they waken feelings— the first winds of autumn.73

Bashp is adding poetic feeling to Zhuangzi’s notion of nature’s creativity: the creative wind of life also creates feelings so deep it tears at one, just as wind rips the easily torn leaves of the bashp tree.74 Two later haibun repeat this reference to the nature’s creative wind and its relation to art. “An Account of Pure Washed Temple” begins with very Chinese imagery of mountains and water, tranquility and activity: Mountains are still and cultivate the nature; water moves and consoles the feelings. There is a person who makes his dwelling between tranquility and movement. His name is Hamada Chinseki. With his eyes he takes in all [tsukusu, “exhaust”] of the scenery. With his mouth he gives voice to aesthetic creativity [ f u ga] . . . At Omono Bay, Seta and Karasaki extend to the right and left like two sleeves which embrace Lake Biwa and face Mikamiyama. Because the lake resembles the shape of a biwa, the echoing pine wind performs music with the waves.75

As in the Zhuangzi passage, nature’s creative wind is like music. The same theme is repeated in another haibun, which begins: There is a pine here reaching to nine feet. The lower branches extend over ten feet, growing on top of each other with a thick growth of needles. The wind plucks the koto pine, calling forth rain and giving rise to waves. The sound resembles a lute, a flute, and a drum, and the waves blend with the echoes of heaven.76

The wind is heaven’s, which brings forth rain and waves and creates music in the pine. And the undulating movement of waves merges with the echoes of heaven’s creative movement. As in the case of the creative, these Zhuangzi-based passages suggest that nature is vital and artistic, and human art corresponds to the creativity of nature.77

50

D    L     B     

C: B¯  N’ C Bashp’s literary prose shows a sophisticated and complex view of nature’s creativity. Nature is more than the beautiful scenes we see about us. There is a creative force in nature that fashions beauty with skillful artistry. The creative is Bashp’s term for the spontaneous creativity of nature, which parallels the creativity of great art. The creative animates all things, and in doing so gives to them the beauty of flower and moon. Life is animated by a divine breath, which unifies all things in a single cosmic vitality, yet makes all things distinct. Nature is evershifting, and these transformations—of each moment and through the four seasons—are the flourishing of life. They give rise to deep feeling and to outstanding art. The artist, and every cultured person, should return to this cosmic creativity, recognize its beauty, and follow its movements. Such a view is principally Chinese. There is, of course, much more to Bsthp’s view of nature, and much of it is also Chinese in concept and flavor. It is important for us to expand the intertextual range of Bashp’s text beyond medieval Japanese literary tradition and Zen Buddhism. The countless references to Chinese religious and aesthetic thought require that we place his texts in the context of Daoism and Confucianism as well as Buddhism, and in the context of the Chinese aesthetic tradition (both poetry and painting) as well as Japanese literature. If we do that, we will find a wonderfully rich worldview.

N Research for this chapter has been supported by the Shinshu Spgp Kenkyujo in Kyoto, where the author was a visiting researcher in 1993, and by a Guilford College faculty research grant. I am indebted to Professors – Yamamoto Yuiitsu of Otani University, Yoshida Sanroku of Kansai Gaigoku Daigaku, William R. LaFleur of the University of Pennsylvania, Gary Ebersole of the University of Chicago, and Jonathan Chaves of George Washington University for stimulating discussions regarding this material. 1. The edition of Bashp’s writings used in this chapter is by Imoto Npichi, Hori Nobuo, and Muramatsu Tomotsugu, eds., Matsuo Bashp shu (Nihon koten bungaku zenshu) (Tokyo: Shpgakukan, 1972), 41, abbreviated as NKBZ. This quotation is found at NKBZ, 311–312. I also give the page number to my translation of his prose, Bashp’s Journey: The Literary Prose of Matsuo Bashp (Albany: State University of New York Press, 2005). This passage is on page 29.

T   C       B   ’ ¯  V   N      A 

51

2. Nobuyuki Yuasa, trans., The Narrow Road to the Deep North and Other Travel Sketches (Harmondsworth, England: Penguin, 1966), 72; Makoto Ueda, Matsuo Bashp (New York: Twayne, 1970), 132; Eiichi Hayashi, “Oi no kobumi,” The Reeds 8 (1962): 77; Eleanor Kerkham, “Notes from the Traveler’s Satchel,” The Tea Leaves 2 (1965): 29. 3. For two discussions of the notion of nature as cultural construction, see George Robertson, Melinda Mash, Lisa Tickner, Jon Bird, Barry Curtis, and Tim Putnam, eds., Future Natural: Nature, Culture, Science (London: Routledge, 1996), and Michael Soule and Gary Lease, eds., Reinventing Nature? Responses to Postmodern Deconstruction (Washington, D.C.: Island Press, 1995). 4. For an examination of Bashp’s participation in another aspect of East Asian culture, the “popular” religiosity of Shintp and the folk traditions, see “Folk Religion and Shintp in the Ecosystem of Bashp’s Religious World;” paper delivered at the Annual Meeting of the American Academy of Religion, Chicago, Illinois, November 23, 1994. It is also important to examine his texts not simply in terms of his East Asian tradition, but also in terms of cross-cultural categories. For two examples of this approach, see David Barnhill, “Impermanence, Fate, and the Journey: Bashp and the Problem of Meaning,” Religion 16 (1986): 323–341 (which applies Clifford Geertz’s notions of religion and the problem of meaning), and his “Bashp as Bat: Wayfaring and Anti-Structure in the Journals of Matsuo Bashp (1644–1694),” Journal of Asian Studies 49 (1990): 274–290 (which applies Victor Turner’s notion of anti-structure). 5. A significant number of his haibun show strong Chinese flavor, perhaps related to the fact that the haibun is a Japanese adaptation of the fu, the Chinese prose poem. There have been relatively few translations of his haibun until recently, despite the fact that some are minor masterpieces. (For a translation of seventy-nine haibun, see my Bashp ’s Journey.) 6. Haruo Shirane, “Lyricism and Intertextuality: An Approach to Shunzei’s Poetics,” Harvard Journal of Asiatic Studies 50.1 (1990): 76. In discussing Shunzei, Shirane notes that in many cases it is unclear and indeterminable whether or not he had in mind specific poems from the Kokinshu, yet “these words and phrases, with their associative clusters, form as integral a part of the intertext as does the obvious allusion to the Fukakusa poems” (76–77). See the earlier mentioned article for a discussion of some of the theoretical bases for intertextuality. For a helpful volume of essays on this approach, see Jay Clayton and Eric Rothstein, eds., Influence and Intertextuality in Literary History (Madison: University of Wisconsin Press, 1991). For an important use of intertextual theory that provides a new perspective on Japanese religion and its relation to the Chinese context, see Steven Heine, Dpgen and the Kpan Tradition: A Tale of Two Shpbpgenzp Texts (Albany, NY: State University of New York Press, 1994). 7. NKBZ, 324. 8. Yuasa, The Narrow Road, 85.

52

D    L    B    

9. The passage has been translated by Eiichi Hayashi as “Seeing the beauties of mountain and field, sea and beach, I simply marvel at the works of nature,” “Oku no hosomichi,” The Reeds 7 (1961): 91; and by Kerkham as “Among the beauties of the mountains, the fields and the sea, I contemplated the works of nature,” “Traveler’s Satchel,” 37. The phrase used by both translators, “works of nature,” begins to suggest the notion of creative activity that is absent in Yuasa. But without explanation, the term “nature” is misleading in this context, as it implies our conventional understanding of nature. 10. NKBZ, 361–362. 11. Yuasa, The Narrow Road, 116. 12. Bashp’s Journey, 61. Of the many translators of Oku no hosomichi, two have used the term “creator” (the common translation for the equivalent Chinese term, zaohua) in rendering this passage. Helen Craig McCullough’s translation is “what author can describe the wonder of the creator’s divine handiwork?” Classical Japanese Prose: An Anthology (Stanford: Stanford University Press, 1990), 535. Dorothy Britton’s translation is “Ah! who could possibly do justice with his brush to this wondrous divine work of the Creator of the Universe or presume to describe it adequately in words!” A Haiku Journey: Bashp’s Narrow Road to a Far Province (Tokyo: Kodansha, 1980), 53. 13. Haibun are short pieces of poetic prose, usually including a hokku poem. They range from a few sentences to a lengthy paragraph, with one (Genjuan no ki, An Account of the Unreal Dwelling) a few pages in length. Editors of his works disagree on the number of his haibun, some including more prose introductions to his hokku than others. The NKBZ edition lists 109 haibun, almost all written between 1680 and 1693. 14. “Matsushima,” NKBZ, 482; Bashp’s Journey, 117. Existing translations of this passage by Hiroaki Sato and Makoto Ueda do suggest this notion of divine skill: “as if wonderfully carved by a heavenly artist,” Hiroaki Sato and Burton Watson, From the Country of Eight Islands: An Anthology of Japanese Poetry (New York: Anchor Press, 1981), 291; “the wonders of a supreme artistry possessed only by nature,” Ueda, Matsuo Bashp, 113. 15. For example, the painter Zhu Yunming (1460–1526) stated that “everything in the universe has some kind of life and that the mystery [miao] of creation [zaohua], changing and unsettled, cannot be described in forms,” Susan Bush, trans., The Chinese Literati on Painting: Su Shi (1037–1101) to Dong Qichang (1555–1636) (Harvard-Yenching Institute Studies XXVII. Cambridge, MA: Harvard University Press, 1971), 135. See also Osvald Siren, trans., The Chinese on the Art of Painting (New York: Schocken Books, 1963), 48, 199–200. 16. “Concerning the Beautiful Scenery At the Home of Master Shua,” NKBZ, 470; Bashp’s Journey, 113. What I have translated as “scenic beauty of consummate skill” literally reads “beautiful scenes exhaust skill.”

T   C       B   ’ ¯  V   N      A 

53

17. Peipei Qiu makes this same point in “Daoist Concepts in Bashp’s Critical Thought,” in (eds.) Steven Tötösy de Zepetnek and Jennifer W. Jay, East Asian Cultural and Historical Perspectives (Edmonton: Research Institute for Comparative Literature and Cross-Cultural Studies, University of Alberta, 1997), 325. 18. NKBZ, 413–414; Bashp’s Journey, 97. Ogata also has pointed to this haibun as one of Bashp’s statements of his conception of his own literary tradition and as an indication of the importance of such an inheritance to Bashp. Ogata Tsutomu, Bashp no sekai (Tpkyp: Kodansha, 1988), 296. 19. In China, poetry, painting, and calligraphy were thought of together as the “three perfections.” See Michael Sullivan, The Three Perfections: Chinese Poetry, Painting, and Calligraphy (New York: G. Braziller, 1980); and Alfreda Murck and Wen Fong, eds., Words and Images: Chinese Poetry, Calligraphy, and Painting (New York: Metropolitan Museum of Art and Princeton University Press, 1991). For a discussion of Su Shi’s views on poetry, painting, and calligraphy, see Ronald C. Egan, Word, Image, and Deed in the Life of Su Shi (Harvard-Yenching Institute Monograph Series 39, Cambridge, MA: Harvard University Press, 1994), 261–309. One reason these arts were considered together was that they all used the same implement: the brush. 20. In this, Bashp follows the Chinese literati tradition seen in Yu-Shi Chen’s comment on Su Shi: “what is crucial in Su Shi’s association of the poet with the painter is not so much a comparison as a convergence of the painter’s and the poet’s minds on the conceptual level with regard to an understanding of the working of nature and an appraisal of the role of artistic expression in the total process of creating a work of art, whether verbal or pictorial,” Yu-Shih Chen, Images and Ideas in Chinese Classical Prose: Studies of Four Masters (Stanford: Stanford University Press, 1988), 139. 21. NKBZ, 416; Bashp’s Journey, 98. 22. NKBZ, 541–542; Bashp’s Journey, 138. 23. Zhuangzi lived around 300 BCE, but the text attributed to him was the work of many hands over an extended period of time. For a discussion of the text, see A.C. Graham, “How Much of ‘Chuang-tzu’ did Chuang-tzu write?” in (eds.) H. Rosemont, Jr. and B. Schwartz, Studies in Classical Chinese Thought (Journal of the American Academy of Religion, Thematic Issue 47 [1979]), 459–502. 24. NKBZ, 436–438; Bashp’s Journey, 105–106. 25. NKBZ, 547–549. See NKBZ, 430, for a haibun with references to the religious thinkers Confucius and Zhuangzi and the poets Du Fu and Li Bo. For a discussion of Bashp’s writings in the context of Chinese religious and aesthetic thought, see David Pollack, The Fracture of Meaning: Japan’s Synthesis of China from the Eighth through the Eighteenth Centuries (Princeton: Princeton University Press, 1986), 209–213. 26. NKBZ, 409; Bashp’s Journey, 94–95.

54

D    L    B    

27. NKBZ, 537–538; Bashp’s Journey, 137. Egan notes how Ouyang Xiu began a tradition of believing that the viewer of calligraphy could commune with the spirits of calligraphers of the past. Egan, Word, Image, and Deed, 265. 28. The complexity of Bashp’s view of space affects his use of nominal reference. For a brief discussion of this, see my review of Earl Miner’s important book Naming Properties: Nominal Reference in Travel Writings by Bashp and Sora, Johnson and Boswell (Ann Arbor, MI: The University of Michigan Press, 1996) in Monumenta Nipponica 53 (1998): 105–108. 29. NKBZ, 416; Bashp’s Journey, 98. 30. NKBZ, 292; Bashp’s Journey, 17. Most translators have refused to take Bashp at his word and conventionalized the text by inserting the word “compare.” For example, “It was only natural for them to compare these mountains with Mount Rozan in China,” Yuasa, The Narrow Road, 56; “Indeed, would it not be fitting to compare the environs to Mount Lu in China?” McCullough, Classical Japanese Prose, 517. Bashp, I would argue, is identifying the two mountains. His haibun on the grass hut quoted earlier is another example of spatial (and temporal) identification: Du Fu, Su Shi, and Bashp experience the same rain. Okazaki Yoshie sees in this passage a depiction of the secret workings of zpka that go beyond surface beauty and draw Bashp’s spirit into its metaphysical world. Okazaki, Bashp no geijutsu (Tpkyp: Hpbunkan, 1959), 110. 31. Stephen Owen, Readings in Chinese Literary Thought (Council of East Asian Studies. Cambridge, MA: Harvard University Press, 1992), 104, 203. 32. NKBZ, 502; Bashp’s Journey, 124. 33. NKBZ, 457; Bashp’s Journey, 109. 34. For a discussion of this motif and its place in the Shisendp garden, see Hiroyuki Suzuki’s discussion in J. Thomas Rimer, Stephen Addiss, and Jonathan Chaves, eds., Shisendp: Hall of the Poetry Immortals (New York: Weatherhill, 1991), 102ff. I am grateful to Jonathan Chaves for pointing out this reference to me. 35. Ibid., 100. 36. Okazaki, Bashp no geijutsu, 116, argues that when Bashp climbs to the second flower of his lodging to “sleep within the wind and clouds, overcome by a feeling of mystery and wonder” (NKBZ, 362), he is experiencing a blessing of being a companion to zpka. 37. For discussions of the importance of Zhuangzi in Bashp’s conception of zpka, see Kon Eizp, Bashp: sono shpgai to geijutsu (Tokyo: Nihon hpsp shuppan kypkai, 1989); Nonomura Katsuhide, Bashp (Tokyo: Yuseidp, 1969); and Tamaki Tpru, Bashp no kyp (Tokyo: Kadokawa shoten, 1989). 38. Burton Watson, trans., The Complete Works of Chuang Tzu (New York: Columbia University Press, 1968), 84–85. A.C. Graham avoids the term creator. He is more literal and distinguishes between zaohua, which he

T   C       B   ’ ¯  V   N      A 

39.

40. 41. 42. 43. 44. 45.

46. 47.

48.

49.

55

translates as “the process which fashions and transforms us,” “he that fashions and transforms us,” and “the fashioner and transformer,” and zao wu zhe, which he translates as “the maker of things,” Chuang-tzu: The Inner Chapters (London: Unwin, 1981), 88–89. There is, of course, a sense of cosmogony in the Daoist tradition. It is an important theme in the Daodejing, but as Donald Munro has pointed out, the emphasis shifts in the Zhuangzi from “production” (sheng) to “transformation” (hua), The Concept of Man in Early China (Stanford: Stanford University Press, 1969), 120. Norman Girardot has described this as an “endless incarnational process,” Myth and Meaning in Early Daoism: The Theme of Chaos (hun-tun) (Berkeley: University of California Press, 1983), 80. There is, however, a sense of the coming-to-be of things in the Zhuangzi, usually from a primordial and undifferentiated reality that goes by various names, such as the Great Clod and Chaos (huntun). For an in depth analysis of the theme of Chaos, see Girardot, Myth and Meaning. However, the term zaohua generally refers to creation as the ongoing process of transformation. Girardot, Myth and Meaning, 79. Burton Watson, trans., Su Tung-p’o; Selections from a Sung Dynasty Poet (New York: Columbia University Press, 1965), 90. Bush, Chinese Literati, 128. Ibid., 38. Ibid., 66, n. 109. “On Seeing a Painting of Plants and Insects by Chu-ning.” Translation by Jonathan Chaves in both Chaves, Mei Yao-ch’en and the Development of Early Sung Poetry (New York: Columbia University Press, 1976), 202, and in Wu-chi Liu and Irving Yucheng Yo, eds., Sunflower Splendor: Three Thousand Years of Chinese Poetry (Garden City, NY: Anchor Books, 1975), 319. Recall Bashp’s haibun “On Mount Fuji” translated earlier for a different view—that art fails where nature’s creativity succeeds. Owen, Chinese Literary Thought, 111. Ogata, Bashp no sekai, 284–292. Ogata sees a parallel between qi and li (the changing and unchanging aspects of zaohua) and Bashp’s distinction between ryukp and fueki (the changing and unchanging aspects of haikai poetry) (289). For another discussion of the Neo-Confucian context of Bashp’s thought, see Nonomura Katsuhide’s article, “Bashp to Spshi to Spgaku” (Bashp, Zhuangzi, and Song Learning) reprinted in Bashp, Society for the Publication of Scholarly Essays on Japanese Literature, eds., (Yuseidp, 1969), 234–240. In an important discussion of zpka, Kuriyama Riichi begins by stating that the term “clearly does not mean nature in the sense of things.” Instead, it refers to the dynamic and spontaneous creativity of nature. Kuriyama Riichi, Spgp Bashp jiten (Tokyo: Yusankaku, 1982), 70. Ogata, Bashp no sekai, 100.

56

D    L    B    

50. The conventional translation of the term li has been “principle,” an abstract term that is more appropriate to Greek philosophy than to NeoConfucianism. Recently, scholars have recognized that the meaning of li is closer to “the inherent natural disposition of things,” and the term “pattern” has been used to translate the term. See, for instance, Donald Munro, Images of Human Nature: A Sung Portrait (Princeton: Princeton University Press, 1988) and Egan, Word, Image, and Deed. 51. Thus, my interpretation of zpka is similar to Ogata’s summary of the Neo-Confucian view, though it is not necessary to think of zaohua in terms of qi and li. As in the Neo-Confucian view, however, I think it is important to think in terms of both force and disposition. 52. One of the complexities of zpka is that it combines the notion of skill and spontaneity. As passages from the previous section make clear, zpka is not characterized by an intentional, consciously controlled skill (what we might call “craftsmanship”) but rather by spontaneous transformation that is supremely skillful. 53. Yamamoto Yuiitsu has argued that there is also in Bashp a parallel to the idea of an undifferentiated reality in the Zhuangzi and Daoism. See Yamamoto Yuiitsu, Haibungaku no keifu (Kypto: Hpzpkan, 1978), 84ff and 103ff. But except for the single passage in Oku no hosomichi concerning the Shintp deity Pyamazumi (NKBZ, 361–362; Bashp’s Journey, 60–61), Bashp shows little interest in cosmogony. Like Zhuangzi, he is more interested in the creative cosmology of the natural world. 54. Bashp’s valorization of spontaneity can be seen in his distinction between a poem that naturally “grows” and a poem artificially “made” by the will of the poet. For a discussion of his poetics in English, see Ueda, Matsuo Bashp, 147–169, and Ueda, Literary and Art Theories in Japan (Cleveland: Western Reserve University Press, 1967), 145–172. Kuriyama argues that zpka is related to Bashp’s notion of spontaneity in artistic creativity, in particular to the famous statement in the Sanzpshi: “The Master said, ‘Learn of the pine from the pine, learn of the bamboo from the bamboo.’ In other words, one must become detached from the self.” To create art by design or intention and with a sense of self is contrary to Bashp’s ideal. Similarly, Kuriyama states that zpka is the opposite of human-made (Kuriyama, Spgp Bashp jiten, 70). The use of the term “skill” in this context thus does not imply the intention-bound skill of a craftsman but a creativity that is wonderful in two senses: stunningly beautiful and full of wonder because it transcends human design. 55. Shirane also uses the term “Creative” in his discussions of zpka; see Traces of Dreams: Landscape, Cultural Memory, and the Poetry of Bashp, (Stanford: Stanford University Press, 1998). 56. Peipei Qiu has used the terms “Nature’s Workings” and “the Natural” in translating the term. See “Adaptation and Transformation: A Study of Daoist Influence on Early Seventeenth-Century Haikai,” in (ed.) Amy V. Heinnick, Currents in Japanese Culture: Translations and Transformations (New York: Columbia University Press, 1997), 185–203, and “Daoist

T   C       B   ’ ¯  V   N      A 

57. 58.

59. 60.

61.

62. 63.

57

Concepts in Bashp’s Critical Thought.” Each of these translations, as well as my own, emphasize different aspects of this complex term. With the term “the Creative,” I am highlighting the creative dynamics of zpka, although they are, in fact, nature’s workings and fully natural and spontaneous. NKBZ, 361–362; Bashp ’s Journey, 60–61. For an examination of Shinto/Neo-Confucian syncretism in the Tokugawa period, see Peter Nosco, “Masuho Zankp (1655–1742): A Shinto Popularizer between Nativism and National Learning,” in (ed.) Peter Nosco, Confucianism and Tokugawa Culture (Princeton: Princeton University Press, 1984), 166–187. For a study of the Chinese influence on early “Shintp” including a critique of our conventional view of Shintp as a purely native Japanese religion, see Toshio Kuroda, “Shintp in the History of Japanese Religion,” Journal of Japanese Studies 7 (1981): 1–21. In “Folk Religion and Shintp in the Ecosystem of Bashp’s Religious World” (paper delivered at the Annual Meeting of the American Academy of Religion, Chicago, Illinois, November 23, 1994), which focuses on Bashp’s cultural” which focuses on Bashp’s cultural and intertextual connection with popular religiosity, I suggest an ecological model for understanding the nature of religions as interdependent rather than discrete. That model is also appropriate to understanding the relationship between Bashp and the Chinese religio-aesthetic tradition. “Folk Religion and Shintp in the Ecosystem of Bashp’s Religious World.” Ogata argues that the notion that those who are cut off from the heart of the creative are barbarians and beasts is a Confucian idea, while the notion of a heart that unifies with the manifestation of the creative in the beauty of moon and blossoms is Bashp’s own aesthetic standpoint. Ogata, Bashp no sekai, 289. However, note Su Shi’s “Prose Poem on the Red Cliff” cited earlier for a Chinese example of the tie between the Creative and beauty, though not specifically in terms of “moon and flowers.” Okazaki makes a similar point. He considers an interpretation in which humans are considered the masterpiece of the creative, but calls this a Western idea. For Bashp, the moon and flowers (what we would call the natural world) are the masterpieces of the creative. Humans are incomplete unless they achieve a unity with the moon and flowers and become a companion to the four seasons. Okazaki, Bashp no geijutsu, 102. Owen, Chinese Literary Thought, 112. The Chinese term jin is used with different connotations in different contexts, but in its general meaning it has the implication of “to do something fully.” The painter Dung Yu stated, e.g., “At this time the artist suddenly forgets his four limbs and body; then what he sees with his insight is all mountains, so he is able to achieve [jin] Dao. Later men who try to understand his method through his paintings do not know that they were painted unconsciously,” Bush, Chinese Literati, 153. Guo Xi made the following statement about a painting: “Wonderfully lofty and divinely beautiful are these [painted] mountains. In order to exhaust

58

64.

65. 66. 67.

68. 69.

70.

71. 72. 73. 74.

D    L     B     their marvels and grasp the work of the Creator, one must love their spirit, study their essential features, wander about them widely, satiate the eyes and store up the impressions in the heart. Then, even if the eye does not see the silk and one realizes that the hand does not govern brush and ink, marvelous, mysterious, boundless becomes that picture of mine!” Osvald Siren, The Chinese on the Art of Painting (New York: Schocken Books, 1963), 48. Stephen Owen points out that the way literary theorist Lu Ji “conceives his procedures is worth our attention: the central terms are chin, ‘exhaust’ (one of the main terms used in the question of adequacy) and miao, the ‘subtleties,’ the ‘fine points,’ ” Owen, Chinese Literary Thought, 85. For a related discussion of Bashp’s view of the inability of words to capture reality, see Qiu, “Daoist Concepts in Bashp’s Critical Thought,” 333–334. NKBZ, 414–415; Bashp’s Journey, 97–98. Watson, The Complete Works of Chuang Tzu, 36–37. Such an idea is not limited to Daoism. Chen notes that Su Shi was interested in the doctrine of the Neo-Confucian Zhou Dunyi in which “countless differentiated phenomena of existence derive[s] from an original source which is itself . . . undifferentiated.” Yu-Shih Chen, Images and Ideas in Chinese Classical Prose: Studies of Four Masters (Stanford: Stanford University Press, 1988), 135. See, for instance, Watson, The Complete Works of Chuang Tzu, 85. In his haibun “Postscript to ‘Comments on the Bagworm,’ ” Bashp states that “if one looks with tranquility, one sees that all things are selfrealized” (NKBZ, 437; Bashp’s Journey, 106), a phrase drawn from the Song poet Cheng Mingdao but resonant with Zhuangzi’s portrait of individual distinctness within cosmic unity. For other discussions of this haibun, see Kon, Bashp: sono shpgai to geijutsu, 63 and Nonomura, Bashp, 235ff. For a discussion of Su Shi’s emphasis on the distinctness of each individual thing in a continually transforming natural world, see Egan, Word, Image, and Deed, 294. For a discussion of the Buddhist significance of mujp (impermanence) in the creation and reading of a renga sequence, see Gary Ebersole, “The Buddhist Ritual Use of Linked Poetry in Medieval Japan,” Eastern Buddhist, 16.2 (1983): 50–67. NKBZ, 470; Bashp’s Journey, 113–114. NKBZ, 482; Bashp’s Journey, 117. Burton Watson, trans., Saigyp: Poems of a Mountain Home (New York: Columbia University Press, 1991), 67. The leaves of the bashp (plantain) tree were known to tear easily, and in several passages Bashp refers to this aspect of his namesake and other images of vulnerability. Note also that in this passage Bashp brings together the Chinese and the Japanese, as well as the philosophical and

T   C       B   ’ ¯  V   N      A 

59

poetic traditions (the vision of nature’s creativity in the Zhuangzi and the “deep feeling” of Saigyp). 75. NKBZ, 497—498; Bashp’s Journey, 122. 76. NKBZ, 523; Bashp ’s Journey, 131. 77. This haibun concludes with another Daoist reference, the notion of controlling one’s breath in order to prolong life. “Bo Juyi said ‘the pine exhales the old qi so it is preserved for a thousand years.’ ”

This page intentionally left blank



R     L: T Z H UA N G Z I   G   I  B¯ Peipei Qiu

L

andscape is an essential component of haikai (comic linked verse) imagination. The pivotal significance of the famous places (meisho) in haikai verses and the emphasis on geographical imagination in haikai prose (haibun) all demonstrate the importance of landscape in haikai composition. This chapter explores how the renowned haikai master Matsuo Bashp (1644–1694) uses the Daoist classic Zhuangzi to reinvent the poetic significance of landscape in his travel journals. It shows that Bashp’s geographical imagination is shaped not simply by the material qualities of space, but more importantly by conceptions based on broader cultural, aesthetic, and philosophical frameworks, particularly the aesthetic of shpypyu (C. xiaoyaoyou, carefree wandering), which is highlighted by the Zhuangzi and reiterated in Chinese poetic tradition.

SHO ¯ YO ¯ Y U¯   H AIKAI T RAVELER In haikai history, Bashp is as famous for his eccentric way of life as for his poetry. He abandoned his house in Edo city and moved to a thatched hut on the bank of the Fukagawa River in the middle of a successful career as a haikai teacher. Four years after his move to Fukagawa, he went on perpetual wandering. He took a major journey in 1684, which was recorded in his first travel journal, Nozarashi kikp (A Weather-Beaten Journey). In 1687, he took a trip to Kashima, which resulted in another travel account, Kashima no ki (A Journey to Kashima). Bashp spent most of the following year on journey again,

62

P     Q 

traveling to places famous in classical poetry (utamakura) and visiting poets in different provinces. He wrote two pieces of prose based on these journeys, Oi no kobumi (Essay in My Pannier) and Sarashina kikp (A Journey to Sarashina, 1688). In the spring of 1689, he undertook another major journey to the northern areas of Honshu; his experiences during this five-month wandering became the material for his best-known travel account, Oku no hosomichi (The Narrow Road to the Depths). Between these major journeys, Bashp traveled constantly to meet his disciples and friends and to visit poetic and historical monuments, devoting a great amount of creative energy to a prose genre he named haibun, particularly the haikai kikpbun (haikai travel journal). Stemming from the travel poems (kiryoka) of Manypshu (compiled 759) and being closely associated with utamakura and poetic diaries (utanikki), the Japanese travel journal as a literary genre is inseparable from poetry. From the earliest extant travel diary Tosa nikki (Tosa Diary, ca. 935) by Ki no Tsurayuki, the Japanese literary travel journals had followed a tradition of weaving poems and the introductory narratives in a sequential order. The travel journals that existed before Bashp were typically written in a first-person voice, with the traveler’s itinerary revolving around the utamakura and the narrative centering on poems composed on them. Given that Japanese poetry always has accentuated the significance of landscape, the marriage of kikpbun and poetry is not a coincidence; in fact, the kikpbun provides an ideal form for poetic expression. The fusion with poetry, however, simultaneously enriched and limited the aesthetic representation of the landscape of the kikpbun: when centering on classical poetic diction, the geographical imagination of the travel journal often was defined by conceptions and conventions that had been molded by classical poetry. In classical Japanese poetry, each poetic toponym (utamakura or meisho), or seasonal word (kigo), has its established essence (hon’i), which determines not only what but also how landscape should be portrayed. In addition, the canonical literary travel journals bear a predominately melancholy tone inherited from the classical poetry. For example, in the works of the famous travel poets, such as the waka (Japanese song) poet Saigyp (1118–1190) and the renga (linked verse) master Spgi (1421–1502), the lament over the passing of seasons, the impermanence of life, and the chaos of the age are recurrent themes. When Bashp aspired to develop a new type of travel journal in haikai style, he met the same challenge that faced haikai when it re-flourished in the early seventeenth century: in the shadows of tradition, Bashp had to

R             L      

63

re-present a classically defined landscape through a popular haikai vision and by using haikai language—the vernacular Japanese that did not have the refined hon’i of classical diction, and Chinese words that were not associated with classical poetic toponyms. In order to reinvent the kikpbun as well as the poetic landscape, Bashp widely referred to the Daoist classics, especially the Zhuangzi, to generate a new poetic essence. The Zhuangzi, though not a native text, had been known in Japan since the Nara period and was popular among the educated people. Haikai poets had envisioned the Zhuangzi as a source of poetic essence before the Bashp School arose. The Teimon School borrowed the parabolic expression of the Zhuangzi as a model for their allegorical poetry. The Danrin School, on the contrary, took the bold laughter, the deliberate reversal of conventional meaning, and the unrestrained imagination of the Zhuangzi as a congenial frame of reference for its characteristic approach to haikai.1 More significantly, the Zhuangzi appeals to haikai poets because it asserts an aesthetic attitude that sees beauty in ordinary and even “low” things/beings, making it possible to discover profound meaning in the down-to-earth topics and vernacular language of haikai and to regenerate poetic essence. In other words, the Daoist classic can help transform a newly invented haikai word (haigon) into a mediating sign, which translates the superficial meaning of a verse or text into the intended meaning and provides the necessary context for poetic dialogue. The last feature of the Zhuangzi, in particular, explains Bashp’s adaptation of the shpypyu spirit to the recluse/traveler theme of his kikpbun: the spirit of carefree wandering and its association with Chinese recluse/travel literature provide a new meaning, giving context to his travel poems and journals. In reinventing the poetic landscape, Bashp carefully reconfigured the traveler—the viewer of the land. From his first kikpbun, the traveler appears as a carefree wanderer. Nozarashi kikp opens with the following paragraph. An ancient priest said: “Traveling a thousand li, I gather no provisions. Under the midnight moon, I entered the land of Not-Even-Anything.” Following his trek, I left my shattered hut on the bank of Fukagawa in the autumn of the first year of Jpkyp. The whistling winds were exceptionally cold. Ready to become a skeleton in the fields— the winds piercing my heart. nozarashi o / kokoro ni kaze no / shimu mi kana2

64

P     Q 

The first line of the poem has often been cited by critics to illustrate the tragic and solemn nature of Bashp’s journey and his travel account. An intertextual reading, however, reveals other implications. The ancient priest to whom Bashp refers is Guangwen (Japanese, Kpmon), a Chinese Chan (J. Zen) Buddhist priest of the Southern Song Dynasty (1127–1279). His verse paraphrased by Bashp was collected in Jianghu fengyue ji (J. Gpko fugetsu shu, Winds and the Moon At Rivers and Lakes, An Anthology), which was widely read by the Five Mountains Buddhist priests in medieval Japan and by the intellectuals and poets of Bashp’s time. The first couplet of Guangwen’s poem reads: “On my journey, I gather no provisions, but laugh and sing. / Under the midnight moon, I entered the land of Not-EvenAnything.”3 Guangwen’s lines, in turn, allude to the Zhuangzi. The first line is based on the following words in the “Xiaoyaoyou” chapter: “If you go off to the green woods nearby, you can take along food for three meals and come back with your stomach as full as ever. If you are going a hundred li, you must grind your grain the night before; and if you are going a thousand li, you must start getting the provisions together three months in advance.”4 The second line of Guangwen’s poem alludes to the famous Not-Even-Anything village (C. Wu he you zhi xiang), a fictional place that signifies in the Zhuangzi a world free of conventional values and institutions. Bashp rephrased the first line of Guangwen’s verse and made the line closer to the wording in the Zhuangzi. This change must not have been random, for the statement in the Zhuangzi is not advice for travel provisions, but a metaphor for people’s differing aspirations in life and understanding of the world. In the Zhuangzi, this statement appears right after the famous yuyan (J. gugen, allegory, metaphor) about the cicada and the little dove laughing at the Peng bird.5 Immediately after the statement, the Zhuangzi concludes: “What do these two creatures understand? Little understanding cannot come up to great understanding; the shortlived cannot come up to the long-lived.”6 By rephrasing Guangwen’s verse, Bashp associates the significance of his travel closely with the Zhuangzi, creating an image of an eccentric traveler who has attained the great understanding of the Dao and who is determined to enter the world of carefree wandering. In light of its association with the Zhuangzi, the beginning of Nozarashi kikp creates a transcendental but also humorous tone that is necessary to distinguish Bashp’s haikai kikpbun from previous travel journals. As Ogata Tsutomu has observed, even the skeleton, an image commonly seen as a symbol for death and considered as Bashp’s serious resolution for his dangerous journey, finds its humorous

R             L      

65

reference in the Zhuangzi.7 In the chapter “Perfect Happiness” in the Zhuangzi, the Daoist master saw a skeleton on his way to Chu. He poked it with his carriage whip and asked questions about how it had come to this. After he had finished speaking, he dragged the skull over and using it for a pillow, lay down to sleep. In the middle of the night, the skull came to him in a dream and said, “You chatter like a rhetorician and all your words betray the entanglements of a living man. The dead know nothing of these! Would you like to hear a lecture on the dead?” “Indeed,” said Zhuangzi. The skull said, “Among the dead there are no rulers above, no subjects below, and no chores of the four seasons. With nothing to do, our springs and autumns are as endless as heaven and earth. A king facing south on his throne could have no more happiness than this!” Zhuangzi couldn’t believe this and said, “If I got the Arbiter of Fate to give you a body again, make you some bones and flesh, return you to your parents and family and your old home and friends, you would want that, wouldn’t you?” The skull frowned severely, wrinkling up its brow. “Why would I throw away more happiness than that of a king on a throne and take on the troubles of a human being again?” it said.8

Clearly, this yuyan states the Daoist belief that perfect happiness exists in limitless freedom, even if this freedom is found in death. With this yuyan/gugen as a mediating text, the skeleton becomes the signifier not only of tragic death, but also of the perfect happiness as defined by Daoist discourse. Bashp’s “ready to become a skeleton,” therefore, suggests not simply the speaker’s resolution to face death on the road, but more importantly the Daoist vision through which he sees life and death as well as the world. As we have seen in this example, in Bashp’s kikpbun the landscape is perceived through the eyes of a traveler in the shpypyu spirit, or, to be more precise, a carefully designed poetic vision that is tinted with Daoist color. The Daoist influence in this poetic vision can be seen more clearly in Oi no kobumi,9 which begins with a peculiar self-portrait of the traveler: “In my body, which has one hundred bones and nine openings, exists something I have called Furabp. I must have meant that my body resembles spun silk that is easily torn in the wind.”10 As has been pointed out repeatedly, Bashp’s peculiar description here is inspired by the following passage in the Zhuangzi: “The hundred joints, the nine openings, the six organs, all come together and exist here (as my body) . . . It would seem as though there must be some

66

P     Q 

True Lord among them. But whether I succeed in discovering his identity or not, it neither adds nor detracts from his Truth.”11 “The hundred bones and nine openings,” though not the exact wording of the Zhuangzi, serves as a mediating text associating Bashp’s haibun and the Daoist classic. This mediating text reveals that Furabp is an indicator of the true identity of the poet, just as is “True Lord” in the Zhuangzi. Furabp is a term coined by Bashp with three characters: fu, wind; ra, thin silk; and bp, priest or boy.12 It is a creature that is broken easily, that is devoid of worldly values, and that willingly submits to nature’s force, the wind. This image shares the qualities of the plantain tree (bashp), a metaphorical image Bashp uses to identify himself. Both are beautiful and fragile. It needs to be noted that Bashp’s kikpbun inherit not only Daoist ideas; names of great Japanese travel poets, particularly Saigyp and Spgi, are mentioned in his travel records frequently. While clearly aware of the kikpbun tradition that framed his haikai kikpbun, Bashp worked hard to break established conventions to create a new style. He writes in Oi no kobumi: Speaking of the travel journal, great writers such as Lord Ki no Tsurayuki, Chpmei, and the nun Abutsu brought this genre to its apogee.13 Later travel journals are by and large little more than imitations of the great masters, and none is able to change the conventions. Shallow-brained and talentless, much less could I make new contributions. It would be easy to write, for example, that such and such a day was rainy in the morning but turned sunny in the afternoon, that there was a pine tree at a certain place, or that there was a river called such and such at a certain place. Records like this, of course, are not worth mentioning unless they present the uniqueness of Huang Tingjian and the novelty of Su Dongpo. Yet, views of the landscape at different places remained in my mind, and the touching impression of places, such as a house in the mountains, or an inn at a remote province, provided the seeds of words. I decided to jot down randomly the unforgettable places, with a hope that they might record traces of messages from the winds and clouds. My words are like the reckless words (mpgo) of the intoxicated, and therefore the audience should take them as no more than the rambling talk of the dreaming and should listen to them recklessly (bpchp).14

Facing the challenge that haikai must demonstrate its newness against the existing kikpbun tradition, Bashp approached the established genre from a haikai-esque perspective. Bpchp (listen recklessly), though including a character different from the word in the Zhuangzi,

R             L      

67

shares the same Japanese pronunciation with the word used there. Mpgo (reckless words) is a standard word neither in Japanese nor in Chinese, but its unusual use of character and the context in which it is used indicate that the word is derived from the same paragraph of the Zhuangzi. As Imoto Npichi and Yayoshi Kan’ichi suggest, “Reckless words” (mpgo) and “listen to them recklessly” (bpchp), the expressions with which Bashp describes his writing and how it must be appreciated, are both drawn from the second chapter of the Zhuangzi.15 These Chinese words not only serve as haigon to transform the existing travel journal that had relied primarily on the classical poetic diction into a new haikai narrative, but also introduce, through their intertexual association with the Daoist classic, an iconoclastic attitude that distinguishes haikai kikpbun from the melancholy tone of earlier travel journals.

Z¯OKA   L R As seen in Bashp’s discussion of kikpbun in Oi no kobumi, commemoration of beautiful landscape views and intimacy with nature have been central themes of Japanese travel journals since the earliest travel diaries. Bashp’s travel journals continued this thematic emphasis, but his perception and depiction of the landscape are deeply rooted in Daoist concepts, particularly, zaohua (J. zpka, create and transform). A term widely used in traditional Chinese texts, it designates in the Zhuangzi both the working of the Dao—the natural way in which all phenomena come into being and transform—and the accomplishment of the Dao (the existence of all things and beings). The notion is used in Chinese literary theory to imply the natural and sponteneous creative process or the unsullied outcome of such a process. This usage is also found in Bashp’s travel journals.16 In examining Bashp’s portrayal of the landscape, the influence of Saigyp cannot be neglected. Saigyp’s journeys and his poetic depictions of the landscape are no doubt among the major inspirations for Bashp’s kikpbun, but Bashp’s presentation of it is distinctively different from Saigyp’s. As represented by Saigyp, the aesthete-travel tradition in Japan before Bashp had been associated with the religious tradition: the major travel poets were mostly Buddhist priests, the themes and images were imparted with Buddhist significance, and retreating to nature was often presented as a way toward religious salvation or enlightenment. Saigyp’s nature poems, for example, embody an unavoidable conflict between his religious commitment to

68

P     Q 

renounce the phenomenal world and his love of scenic beauty. Such a conflict is seen clearly in the following poem: Why is a heart attracted to cherry blossoms still in this body which, I thought, had forsaken the world? hana ni somu / kokoro no ikade / nokoriken / sutehateteki to / omou waga mi ni17

This kind of conflict is not seen in Bashp’s poetry. Bashp joyfully sees the beauty of the landscape as the marvelous work of zpka and assumes that following zpka—constantly immersing oneself in the embrace of nature, appreciating its beauty, and following its course— is the essential way to maintain aesthetic sensibilities. Oi no kobumi contains the following statement: In the waka of Saigyp, the renga of Spgi, the paintings of Sesshu and the tea ceremony of Rikyu, one fundamental principle runs through all arts: those who pursue art follow zpka, and have the four seasons as their companion. Everything they see is like a flower and everything they imagine is like the moon. If one sees no flower, he is the same as a barbarian; if one has no moon in mind, he is no different from the birds and the beasts. Go beyond the barbarians and depart from animals; follow zpka and return to zpka.18

By saying that those who follow zpka see nothing but flowers and think of nothing but the moon, Bashp has made zpka the precondition of artistic perception. His call for “returning to zpka” suggests that zpka is not only where artistic creativity begins, but also its ultimate attainment. Based on this aesthetic belief, Bashp’s travel journals praise the beauty of the landscape as the splendid accomplishment of zpka: “The singular peaks and grotesque mountains vie with each other in their shapes, forming a hair-like dark line and a glimpse of faint green as in a painting. The sounds of the water, the singing of birds and the green of pines and cedars are extremely exquisite—the beautiful scene demonstrates the perfection of artistry. How could one not feel joyous for the great accomplishment of zpka!”19 The same usage of the term also is found in Oi no kobumi: “I have seen the masterwork of zpka in the beautiful scenes of mountains, fields and the coast, and following the footprints of devoted travelers who are free of worldly concerns, I have come to know the heart of true poets.”20 Here, zpka is not the beautiful landscape per se but what has brought it into being. Like the Daoist sages who have joined with the

R             L      

69

course of zaohua, the traveler Furabp wanders on a haikai journey beyond worldly concerns, submits himself to the force of the winds, and harmonizes with the four seasons. To the Furabp, the poetic self of Bashp, “following zpka” in the physical sense is to discover and appreciate the masterwork of zpka. From the time of the journey that he commemorates in Oi no kobumi,21 the exploration and depiction of the immeasurable power of zpka and the magnificent beauty of its creation are central themes of Bashp’s travel journals. In Oku no hosomichi, the best-known of Bashp’s travel accounts, exploration of the wonder of zpka is a keynote. Many poems in the kikpbun contain a place name or focus on a scenic site, creating an impression that the salutation to the landscape is one of the major preoccupations of the traveler/poet. Yet, Bashp often has avoided writing a poem at the most famous sites in his travel accounts. Oi no kobumi tells us that at Yoshino, known for the most beautiful views of cherry blossoms in Japan, the poet “was unable to find proper language to compose a poem, and hence, kept silent.”22 In Oku no hosomichi, Bashp avoids writing a poem on Matsushima, though he tells us that it is “the most beautiful place in Japan.” He writes: “Matsushima must have been made by the Mountain God in the distant past when the deities created the world. Who could capture this heavenly work of zpka with his brush and words?”23 Evidently, Bashp’s silence before such a magnificent landscape is intended to demonstrate the inadequacy of language in comparison with the creation of zpka. It also implies an aesthetic belief that overtones of silence are more powerful than insufficient language—a belief that is repeatedly stressed by the Zhuangzi: “The Great Way is not named; Great Discriminations are not spoken.”24 “Heaven and earth have their great beauties but do not speak of them; the four seasons have their clear-marked regularity but do not discuss it; the ten-thousand things have their principles of growth but do not expound them. The sage seeks out the beauties of Heaven and earth and masters the principles of the ten-thousand things.”25 Bashp’s treatment of Mount Fuji reveals this aesthetic belief most clearly. His best-known poem on Mount Fuji describes how he was unable to see the celebrated mountain: It was raining the day when I crossed the Barrier, and all the mountains were concealed in clouds. The misty rain— the day unable to see Fuji leaves great charm. kirishigure / fuji o minu hi zo / omoshiroki26

70

P     Q 

This poem reminds us of Yoshida Kenkp’s classical comments on aesthetic sensibility: “Are we to look at cherry blossoms only in full bloom, the moon only when it is cloudless? To long for the moon while looking on the rain, to lower the blinds and be unaware of the passing of the spring—these are even more deeply moving.”27 Echoing the classical aesthetic tradition, Bashp’s verse on Fuji is no doubt a superb poem. However, a eulogy on Mount Fuji by Bashp, probably written around the same time, tells us a different reason for his not providing a direct portrait on the grandeur of Mount Fuji: On Mount Fuji I heard that Mount Kunlun is in a distant country, and Penglai and Fangzhang are places where immortals live. Here, before my eyes, Mount Fuji stands steeply above the earth, towering into the vast sky. As if the mountain is opening its cloud gates to welcome the sun and the moon, wherever I turn, I see its magnificent face, and its beautiful appearances change in myriad ways. [Facing Fuji,] poets could not produce a fitting verse, talented writers would be out of words, and artists could only put down their painting brushes. People say that there is a Holy Man living on faraway Gushe Mountain. I wonder if he could compose a poem or draw a picture of Mount Fuji. Amid cloudy mists, in a second its view has changed one hundred times. kumokiri no / zanji hyakkei o / tsukushikeri28

In the waka tradition, the poetic hon’i of Mount Fuji lies in its lofty peak covered by white snow. The following waka from Ogura hyakunin isshu (The Ogura Sequence Of One Hundred Poems By One Hundred Poets)29 is a typical example: At Tago Bay I came out, and looked afar— to see the hemp-white of Mount Fuji’s lofty peak under a flurry of snow. tago no ura / uchiidete mireba / shirotae no / Fuji no takane ni / yuki wa furitsutsu30

Bashp’s poem, however, gives no attention to the classical essence of Mount Fuji defined by the waka tradition, although he does mention the peak in the short haibun that precedes the poem. Even in his haibun, the depiction of the geographical features of Fuji is minimized, mystified, and projected through the poet’s imagination of famous Daoist sites—Kunlun, Penglai, Fangzhang, and Gushe. Mount Kunlun is at the center of the Kunlun mountain range, which is located at the border

R             L      

71

of today’s Xizang and Xinjiang provinces in China. Chinese mythology has it that Kunlun is the pivot of the universe, where the Queen Mother of the West, an immortal figure who appears in many Daoist legends, lives. In the Zhuangzi, Kunlun is described as a place where those who have attained the Dao and who have achieved immortality live.31 Penglai and Fangzhang are also holy sites in Daoist mythology and believed to be places where immortals live. The phrase, “a Holy Man living on Gushe Mountain” appears in the “Xiaoyaoyou” chapter of the Zhuangzi. The original passage describes him as an immortal who “doesn’t eat the five grains, but sucks the wind, drinks the dew, climbs up on the clouds and mist, rides a flying dragon, and wanders beyond the four seas.”32 The juxtaposition of Mount Fuji and the Daoist toponyms presents a symbolic landscape carefully designed: the appearances and definitions of the geographical space have changed as the visitor reimagines it through a Daoist context. Mount Fuji is no longer simply a physical place of grandeur; it has become an aesthetic landscape whose wonder manifests the power of zpka. Bashp’s avoidance of portraying Mount Fuji in his poem demonstrates clearly the aesthetic belief that the creation of zpka is so magnificent that no language can properly describe it.

T A L  B O  Z¯oka As evident in the poems cited earlier, in Bashp’s portrayal of the aesthetic landscape, the identity of the speaker is usually omitted, while the speaker’s voice is unmistakably present, evoking debates on whether Bashp’s nature poems are impersonal. This controversial quality, however, can be explained by his aesthetic belief in following zpka and returning to zpka: when portraying the landscape, the identity of the poet is not absent, but has merged into the entity of nature; the poet’s voice is evident, yet conveyed through harmony with the work of zpka. The following poem from Oku no hosomichi is a good example of the amalgamation of the speaker’s poetic self with the creation of zpka: How tranquil it is! Penetrating into the rocks the sound of cicadas. shizukasa ya / iwa ni shimiiru / semi no koe33

This poem appears to be strictly on nature: no trace of man is found in the tableau to disturb the profound tranquility of the universe. Yet, there is obviously a beholder through whose senses the eternal

72

P     Q 

tranquility is apprehended, internalized, and expressed. The opening phrase, shizukasa ya, is derived from shizuka, an adjective that means quiet, still, or tranquil. The word is commonly written with a character whose Chinese-origin reading is “sei,” but in this verse Bashp uses a different character whose Chinese-origin reading is “kan,” meaning “leisure” or “idle.” The implications of the latter character are highly valued by the Daoist thinkers, and Bashp’s choice is not a coincidence. Textual studies show that this poem has gone through careful revisions. The original draft recorded in Sora zuikp nikki (Travel With the Master—Sora’s Journal) is as follows: A mountain temple— seeping into the stones, the sound of cicadas. yamadera ya / ishi ni shimitsuku / semi no koe34

This version seems to be more “impersonal.” Comparing this verse with the one cited earlier, it is clear that the revision intends to convey more the poet’s perception of the stillness of the landscape, as Bashp describes in the prose preceding the poem: “Wandering along the cliffs, climbed the rocky mountains, and visited Buddhist temples— the profound tranquility of the beautiful landscape penetrated deeply into my heart.”35 By carefully choosing a character whose connotations are celebrated in the Daoist texts to transcribe the word “shizukasa,” Bashp expresses simultaneously the external world and the speaker’s internal feelings, presenting not only a picture of the landscape, but also an aesthetic evaluation of it, an evaluation that is informed by the Daoist discourse. This aesthetic landscape embodies the beholder’s attitude toward the world, and it is in this landscape that the poetic self merges into zpka. Indeed, in Bashp’s poetry shizukasa (tranquility) often designates an existential state in which one’s inner serenity and the fundamental silence of the external world merge, a state close to the ontological solitude and silence defined in the Zhuangzi: “Emptiness, stillness, limpidity, silence, inaction—these are the level of Heaven and earth, the substance of the way and its Virtue.”36 In Bashp’s later works, this state often is described as kanjaku, a compound word that is written with the character “kan” (leisure or idle) and “jaku” (lonely or still), and Bashp declares that kanjaku is the state in which his “mind is.”37 But kanjaku as a basic tone became evident in Bashp’s poetry much earlier. From around the time of his first poetic journey, his poems already show the tone of kanjaku. The following poem from Nozarashi kikp typically captures the atmosphere of profound solitude and silence.

R             L      

73

The sea darkens: the voices of wild ducks are faintly white. umi kurete / kamo no koe / honoka ni shiroshi38

This poem has been widely praised for its extraordinary depiction of the seascape: as darkness begins to permeate the sea, the faint cry of wild ducks fades into the infinite silence; it deepens the stillness, like a piece of whiteness heightens the darkness of the sea. In this poem, the auditory image, “the voice of wild ducks,” is described with a visual term, “white.” This “transference of the senses” has invited numerous comments.39 Some scholars believe that with “whiteness” Bashp really is not describing sound but something else—the waves, the sea, the vapor over the sea, the color of wild ducks, and so forth. Others maintain that whiteness does depict “the voice,” and this deliberate confusion of senses is a superb rhetorical technique.40 The transference of the senses as a rhetorical device is a familiar element in Japanese poetry. Konishi traces it to “synaesthesia” in classical Japanese poetry and the method of “conceit” derived from the Zhuangzi-style gugen in earlier haikai.41 As convincingly argued by Konishi and other scholars, Bashp would not have had to take the trouble to break the 5/7/5 syllabic rule if he had intended to use whiteness as a modifier of the sea, the wild ducks, or any visual images. He simply could have placed the second line after the third, and the poem would be perfectly regular; the word white would grammatically describe the sea or the color of the wild ducks. As seen in the verse presented earlier, when the transference of the senses occurs, the original significances of relevant images are transferred, transformed, and greatly expanded at the same time. Apparently, by using the transference of the senses Bashp deliberately dispensed with the regular hokku format as well as with conventional reader’s expectations in order to convey a certain significance. The intended significance, as Konishi points out, is the tone of kanjaku, the profound stillness perceived through a mental state of emptiness.42 Interestingly, in the Zhuangzi the fundamental emptiness and stillness are also often associated and described with whiteness. The Zhuangzi says: “Look into that closed room, the empty chamber where brightness [literally, whiteness] is born! Fortune and blessing gather where there is stillness.”43 About this passage, the Song annotator Lin Xiyi explains44: Zhan means to “look.” The Zhuangzi uses the “empty chamber” as an emblem of “mind.” “That closed room” is a metaphor of the emptiness of the self. Be empty, then brightness is naturally born in the emptiness.

74

P    Q  The phrase, “Whiteness is born,” means “brightness is born.” But, instead of saying “brightness,” the Zhuangzi says “whiteness”—this is a distinctive characteristic of Zhuangzi’s writing. Where there are emptiness and brightness is the origin of ten thousand things. Fortune and blessing gather where there is stillness. “Stillness” is also the “emptiness.” Only when it is still, can it be empty; only when it is empty, can it be bright.45

The Zhuangzi uses whiteness to represent the attainment of Dao, a state that is born in emptiness and stillness, in which fortune and blessing gather. Lin further elaborates that the whiteness is “brightness,” and brightness is born in a mental state of stillness and emptiness. Lin emphasizes particularly that the transference of the image whiteness to brightness is a distinctive feature of Zhuangzi’s writing. Although it may be arbitrary to conclude that the transference of senses in the wild duck poem is inspired directly by the passages in the Zhuangzi and Lin’s interpretation of it, it is elucidating to read these passages as comparable references. Read in conjunction with the Zhuangzi, the image white in Bashp’s poem conveys a similar state; it is the poetic figuration of the primal emptiness and stillness the poet perceives when hearing the faint voice of wild ducks fading into the boundless seascape. Through the transferred image, whiteness, the poem powerfully creates a tone of kanjaku. As seen in the discussion above, the Daoist classic Zhuangzi played an important part in the haikai landscape created by Matsuo Bashp. Bashp’s use of the Zhuangzi, however, was not simply stylistic imitation or textual reference, but was based on the larger concern of reinventing existing literary conventions. By adapting the Zhuangzi into the context of the haikai travel journals, Bashp successfully recreated the landscape defined by waka and renga poetry and presented a new horizon on the world of haikai. His incorporation of the shpypyu aesthetic into the haikai imagination also helped to transform the popular genre into a profound art.

N 1. For a fuller discussion of the use of the Zhuangzi by earlier haikai schools, see Hirota Jirp, Bashp no geijutsu—Sono tenkai to haikei (Tokyo: Yuseidp, 1968), 192–237; and Peipei Qiu, Bashp and the Dao: The Zhuangzi and the Transformation of Haikai (Honolulu: University of Hawai’i Press, 2005), 13–40. 2. Komiya Toyotaka, comp., Kphon Bashp zenshu (hereafter KBZ) (Tokyo: Kadokawa Shoten, 1962–1969), VI, 53. Complete translations of this

R             L      

3.

4.

5.

6.

7.

8. 9.

75

kikpbun can be found in Donald Keene, “Bashp’s Journey of 1684,” Asia Major no. 7 (November 1959); Nobuyuki Yuasa, “The Record of a Weather-Exposed Skeleton,” in Bashp: The Narrow Road to the Deep North and Other Travel Sketches (Harmondsworth: Penguin Press, 1966), 51–64; Dorothy Britton, A Haiku Journey, Bashp’s Narrow Road to a Far Province (Tokyo and New York: Kpdansha International, 1974); and Helen McCullough, “The Narrow Road of the Interior,” Classical Japanese Prose, An Anthology (Stanford: Stanford University Press, 1990), 513–522. My translation owes much to earlier translations. (Editor’s note: we have, for convenience of students and teachers, cross-referenced allusions to Bashp’s kikpbun (and haibun when possible) with translations in David Landis Barnhill’s Bashp’s Journey: The Literary Prose of Matsuo Bashp [Albany: State University of New York Press, 2005], 13–22; abbreviated as Bashp’s Journey.) Quoted in KBZ, VI, 53, annotation to Nozarashi kikp. The intertextual relationship between Bashp’s passage and Guangwen’s poem and the Zhuangzi has been pointed out by Ogata Tsutomu in Bashp no sekai (Tpkyp: Kpdansha, 1963), 52–54. My discussion here owes much to his insight. Burton Watson, “Free and Easy Wandering,” in The Complete Works of Chuang Tzu (hereafter CWC), chapter 1 (New York: Columbia University Press, 1968), 30. Yuyan, or gugen in Japanese, is an important rhetorical method of the Zhuangzi. Burton Watson in his translation of CWC rendered the meaning of the word into “imputed words.” It refers to words spoken through the mouth of historical or fictional figures to make them more compelling. The Peng bird appears in the first chapter of the Zhuangzi. “In the Northern Depth there is a fish called Kun. The fish is so huge that no one knows how many thousand li it measures. The fish changes and becomes a bird called Peng. No one knows how many thousand li the back of the bird measures. When the sea begins to move, Peng sets off for the Southern Depth, which is the Lake of Heaven.” Cf. Watson, CWC, 29. Lin Xiyi, Zhuangzi Juanzhai kouyi (hereafter ZJK); (1629 Kyoto edition), reproduced in Nagasawa Kikuya, comp., Wakokubon shoshi taisei, XI, XII (Tokyo: Kyuko Shoin, 1976), 1/4a and b/p. 413; Watson, CWC, 30. Ogata, Bashp no sekai, 59. While noting the possible intertextual relationship between Bashp’s “nozarashi” and the “dulou” (old skull), Ogata sees a paradox between Bashp’s conceptual allusion to the Zhuangzi and his physical experience at the moment, and he interprets the phrase nozarashi o kokoro ni in the hokku, as the poet’s determination to die on his trek. Lin, ZJK, 6/23a and b/p. 14. Watson, CWC, 193–194. Complete translations of the work can be found in Nobuyuki Yuasa, “The Record of a Travel-worn Satchel,” in Bashp: The Narrow Road to the Deep North and Other Travel Sketches (Harmondsworth: Penguin Books,

76

10. 11. 12.

13.

14. 15. 16.

17.

18. 19.

20. 21.

22. 23. 24.

P    Q  1979), 71–90; Eleanor Kerkham, “Notes from the Traveler’s Satchel,” The Tea Leaves 2 (autumn 1965): 26–46; and Barnhill, Bashp ’s Journey, 29–43. Oi no kobumi, KBZ, VI, 75; cf. Barnhill, Bashp ’s Journey, 29. Watson, CWC, 38. Nieda Tadashi suggests that Furabp might be a comic twist of the word furai, which means “being blown here by winds.” See Nieda, Bashp ni eikypshita kanshibun (Tokyo: Kypkiku shuppan sentâ, 1976), 3–7. Abutsu (?–ca. 1283) was the second wife of Fujiwara no Tameie (1198–1275) and the author of Izayoi nikki, a journal written on her journey from Kyoto to Kamakura. She took Buddhist tonsure after her husband’s death. KBZ, VI, 76—77; cf. Barnhill, Bashp ’s Journey, 30. “Additional Annotations” to Oi no kobumi, KBZ, VI, 167. Regarding the meaning and derivation of Bashp’s zpka, Japanese scholars have different opinions. While some note its roots in Chinese sources, especially in the Daoist classics, others interpret it as an autonomous expression that refers to “nature” in general. See Nose Asaji, Bashp kpza, VI (Tokyo: Sanseidp, 1943), 34; Nonomura Katsuhide, “Bashp to Spji to Spgaku,” Renga haikai kenkyu 15.11 (1957): 33–39; Konishi Jin’ichi, “Bashp to gugensetsu,” Nihon gakushiin kiyp 18. 3 (November 1960), 151–158; Hirota, Bashp no geijitsu, 372–444; and Imoto Npichi, Bashp kpza, I (Tokyo: Spgensha, 1953), 204. For a discussion in English on this subject, see Peipei Qiu, Bashp and the Dao: The Zhuangzi and the Transformation of Haikai (Honolulu: University of Hawai’i Press, 2005), 81–82. Watanabe Tamotsu, annot., Saigyp Sankashu zen chushaku (Tokyo: Kazama Shobp, 1971), 48; cf. translations by William R. LaFleur, Mirror for the Moon: A Selection of Poems by Saigyp (1118–1190) (New York: New Directions Books, 1978), 6, and Burton Watson, trans., Saigyp: Poems of a Mountain Home (New York: Columbia University Press, 1991), 39. KBZ, VI, 75; cf. Donald Keene’s translation in World Within the Walls, 92–93 and Barnhill, Basho’s Journey, 29. KBZ, VI, 390. The passage is found in a haibun (“Viewing the Lovely Scenery of our Host, Shua”) composed by Bashp in 1688 on his Oku no hosomichi journey. KBZ, VI, 85; cf. Barnhill, Bashp’s Journey, 39. The date of the completion of Oi no kobumi is uncertain. Ogata Tsutomu suggests that the work was completed around 1690, about three years after the journey upon which the work was based. See “Haiku hairon,” in Nihon koten kanshp kpza (1959) XIX, 217. But evidence shows that the formation of the concept under discussion took place much earlier. See Hirota, Bashp no geijutsu, 372–424. KBZ, VI, 84. Ibid., 118. Lin, ZJK, 1/41a, b/p. 432; translation is from Watson, CWC, 44.

R             L      

77

25. Lin, ZJK, 7/19a, b/p. 41; translation is from Watson, CWC, 236. 26. KBZ, VI, 53. 27. Donald Keene, trans, Essays in Idleness: The Tsurezuregusa of Kenkp (New York: Columbia University Press, 1967), 115. 28. KBZ, VI, 305. 29. I have borrowed Steven Carter’s translation of the title of the sequence in his Traditional Japanese Poetry, An Anthology (Stanford: Stanford University Press, 1991), 203–204. 30. Carter, Traditional Japanese Poetry, 207. 31. See Watson, CWC, 81–82. 32. See ibid., 33. 33. KBZ, VI, 123; Barnhill, Bashp ’s Journey, 65. 34. KBZ, VI, 255. 35. Ibid., 123; cf. Barnhill, Bashp ’s Journey, 65. 36. Lin, ZJK, 5/2a, b/p. 525; translation of the passage is from Watson, CWC, 142. 37. “Rakushisha no ki” (On The Hut of Fallen Persimmons, 1691), KBZ, VI, 482; Barnhill, Bashp ’s Journey, 129. 38. KBZ, VI, p. 59; the translation of the poem is from Keene, World Within the Walls, 86. 39. Keene, World Within the Walls, 87. 40. For a brief summary and translation of comments on the poem by Japanese scholars, see Makoto Ueda, Bashp and His Interpreters, Selected Hokku with Commentary (Standford, CA: Standford University Press, 1991), 123–124. 41. Konishi Jin’ichi, “Kamo no koe honokani shiroshi—Bashp ku bunseki hihyp no kokoromi,” Bungaku 31, no. 7–12 (1963): 845–848. 42. Ibid., 842. 43. Lin, ZJK, 2/21a, b/p. 448; translation is from Watson, CWC, 58. 44. Lin Xiyi (ca. 1200–1273) was a scholar–official of the Song dynasty. His vernacular explanations of the Daoist classics were widely read in Japan and it has been proven that Bashp and his fellow haikai poets read the Zhuangzi through Lin’s annotations. 45. Lin, ZJK, 2/21a, b/p. 448.

This page intentionally left blank



S    P: B ¯ L  F William R. LaFleur

I saw myself a ring of bone in the clear stream of all of it and vowed always to be open to it that all of it might flow through and then heard “ring of bone” where ring is what a bell does —Lew Welch, “Ring of Bone”

C

ross-cultural borrowing can at times resemble a garage sale. Items that the original owners have come to dislike and wish not to have to see again in their own home can become items that seem like discoveries of great value to the exploring passerby. During the 1960s the poetry of Bashp had become an item somewhat like that. In Japan a politically radicalized younger group of scholars had begun to dismiss or at least downgrade Japan’s premier poet. New criteria employed for estimating what is truly valuable in literature had, when applied to the famous writer of haikai and poetic travel diaries, found him insufficiently critical of the political structure and powers of his own time. This animus against Bashp had gotten so strong that in 1970 Konishi

80

W      R. L F   

Jin’ichi, quickly gaining a reputation as the most important scholar of Japanese literature in the world, appraised and lamented the literary climate of the 1960s as follows: Among a younger generation . . . today there is a tendency for the haikai of Bashp to be valued much less than they once were. Today one frequently runs up against the opinion that, since there is no ideology in Bashp’s work and since in him there is absolutely no criticism of the society of his time, it becomes impossible [for these younger scholars] to see Bashp as having written literature of real value.1

Konishi implied that a radical cleaning within the house of Japanese literature had, at least in the minds of some engaged in the purge, found Bashp to be a rather dispensable item. On the other side of the Pacific, however, poets were just then in the process of finding something radically important—and importantly radical!—in Bashp. It is helpful to recognize that early in that decade, on January 17, 1961, a former general and republican, Dwight Eisenhower, took the occasion of his retirement from the presidency of the United States to issue a warning (without impact at the time!) concerning the drastic and fearful changes afoot in American society—namely, the subversion of its traditional values by what he called “the military–industrial complex.” At the same time some of America’s poets, making use of their own sensitive antennae, had already begun to envision an alternative to the kind of society America was rapidly becoming. And Kenneth Rexroth, Gary Snyder, Philip Whalen, and others were positioning Bashp high among the sources from the past proving that humans can choose not to debase nature and debase themselves in that process. Snyder would refer repeatedly to Bashp’s phrase about “going to the pine tree to learn of the pine and to the bamboo to learn of bamboo.”2 And there can be little doubt that the corpus of Bashp’s travel diaries, especially the off-the-beaten-track ambience of Oku no hosomichi, was a major source of inspiration to American poets seeing a far better future in being on the road rather than in the sedentary, arming-to-the-teeth culture of a misnamed “Cold” War that would, by the latter half of the 1960s decade, ignite into a tragically hot mode in Southeast Asia. Here, then, was the irony of Bashp in two societies during the 1960s. Peripheralized by younger Japanese scholars as a poet “in whom no criticism of society” could be located, Bashp to a new wave of younger America writers was celebrated as having incomparable authenticity and the courage to break from a suffocating society in

S           P

81

order to see a range of optional personal and societal possibilities. The late Cid Corman, whose translation of Oku no hosomichi as “Back Roads to Far Towns” came to be widely read, wrote the following about him in 1968: We too move with him to and through the backwater regions of North Central Honshu. His words are our provision, breath, rhythm. And they can never not be our time. The end of his journey is the end of ours.3

The contrast between Japanese ready to knock Bashp off his pedestal and Americans eager to put him up on one was sharp. The project I undertake in the following pages is to examine this difference in the evaluation of Bashp—first by reconstructing the viewpoint that finds him culpable and then by introducing other considerations in an attempt to find a more balanced perspective.

I C? In 1970 Konishi summarized his contemporaneous critics of the seventeenth-century poet as charging that “there is no ideology in Bashp’s work” and “absolutely no criticism of the society of his time.” It is worth noting that during the 1960s American literary critics, in the thrall of New Criticism, were not yet prepared to engage in the detection of ideology or Ideologiekritik. The serious application of Marx and the approach of the “Frankfurt School” to the analysis of literary texts would not appear here until the 1980s and would not show up in the writings of students of Japanese literature until the 1990s. And even today I know of no American study of Bashp that could be described as carried out in the mode of Ideologiekritik. In view of this there would be, I suggest, at least heuristic value in attempting to sketch out the basic shape of the approach to which Konishi objected. What follows is put together out of the basics of Ideologiekritik, what was being said about Bashp in Japan during the 1960s, and additional information now at our disposal—especially from historians exploring the relationship between northeastern Honshu and the shoguns’ efforts to exert more control over it. The critics of ideology insist that the dissemination or extension of “culture” is never an innocent project. Therefore, in traveling into the “far” or “interior” north and bringing his mode and standard of poetry to that area in 1689 Bashp had agreed to be the “cultural front” in what was, in fact, a colonization of that territory by the political powers based in Edo and Kyoto. Such an interpretation, of course,

W      R. L F   

82

is dependent on the historical scholarship suggesting that the politically integrated unit we today call “Japan” was very different in the seventeenth century. The substantive claim of Amino Yoshihiko and others is that, during the medieval and much of the early modern periods, what we call “Tphoku” had not only a spoken language but also a culture distinctively different from that of Western Japan.4 It was a region that, at that time inclusive of people called ezo or emishi at the extreme point of this difference, was in its entirety a place that might in some significant sense be thought of as “a different country.” Linguistically, culturally, and even in some sense politically, it was a place that could be meaningfully differentiated from Yamato, its counterpart to the west and south. Building on work by Japanese historians, Alexander Bay calls this frontier zone “Kita pu” and claims that “it did not necessarily embrace, and sometimes even rejected, both Sinocentric social and political institutions prevalent in Yamato Japan and the transmarine Northeast Asian culture to its north.”5 According to this hypothesis, Hideyoshi’s military sallies into the north were not so much to reunify what had once been a single nation as it was to make a nation for the first time. Moreover, since the rice culture of Yamato had never been ecologically or culturally appropriate to the climate and ways of the northeast, the wider extension of rice cultivation patterns into that area during subsequent centuries was part of what was creating the fiction of one nation. The “internal colonization” of the northeast was a many-pronged process. On this hypothesis Bashp may have been a participant in this replacement of reality with a fiction. If so, then it would not be a mere coincidence that it was in connection with his passage through the barrier at Shirakawa—that is, having finally entered into the northeast— that Bashp writes as if rice cultivation had been an ancient practice there. He wrote:6 furyu no hajime ya oku no taue uta

That might be translated as: Refinement rose up from the rice-planting songs of these country-folk

Although Shirakawa would have been far to the south of Kita Pu, and I will return to this poem later to provide a different interpretation,

S           P

83

there can be value in seeing how such a poem could be fitted into an attempt to see Oku no hosomichi as ideologically tainted. Moreover, if the northeast were a country apart from Yamato, it would likely be that from the perspective of people there the emperor or tennp would have seemed to be the monarch of another nation, a person who lived in a city far from them in space and alien to them in its cultural fashions. Some residents of today’s Tphoku have absorbed this into their own view of things. It was so stated to me by a well-read Tphoku resident I met in 1990. During the spring of that year, while making my own tour of the most northerly territory covered in Bashp’s Oku no hosomichi and while in the city of Sakata, I stopped to see the miira or self-mummified Buddha in Kaikp-ji, a temple there. That evening I had dinner with a gentleman I had fortuitously met along the way. When in our conversation I expressed curiosity about the high percentage of miira that had been produced in that part of Japan and asked why miira were so concentrated in the Tphoku area, my conversation partner provided the following explanation. It was, he said, that the people of that area, with neither the opportunity nor the inclination to revere the sacred emperor of that other culture/ country to the west of their own, had a deep need for some kind of extraordinary, deeply charismatic type of person to revere within their own cultural ambit. This need, he said, gave rise to the cult of the miira, spiritual athletes and heroes of the “Tphoku nation.” Bashp too, he said, may have been responding to something of that when he gave expression to an unusual sense of “sacred awe,” calling attention to the “severe austerities” of religious practitioners in the areas of Gassan and Yudono.7 In other words, self-embalmed miira and living ascetics were, at least to that nameless, unnameable nation of people inhabiting the most northerly portion of the island of Honshu, roughly equivalent in function to the tennp in Kyoto. I neither want to nor can judge the accuracy of such a claim about miira, but I do surmise that the new research on the history of the “northeast” is fascinating, provocative, and, if I can judge by the opinions of the man I met in Sakata, something with its own local appeal. Let us, hypothetically, push what might be some implications of such historical research a bit further. We know that at least during the Heian period portions of whatever peoples were living in northeastern area had been resistant to political control by the hegemonic powers of Heian-kyp—so resistant, in fact, that it became necessary for the capital authorities to establish garrisons in Tphoku and set in place the castle known as Taga-jp, a fort whose ruins Bashp went to see.

84

W      R. L F   

Let us assume, furthermore, that the northeast remained, even after Taga-jp was no longer extant and even as late as the seventeenth century, a territory whose people remained, at least to some degree, resistant both to the political control and to the cultural hegemonizing of the tennp’s capital to the far south and of the somewhat closer upstart shogunal capital called Edo. Let us go on to assume there may be cogency in Muramatsu Tomotsugu’s marshalling of evidence suggesting that Kawai Sora was quite likely paid by the Tokugawa shogunate to engage either in systematic or part-time “intelligence work” while traveling along as Bashp’s companion on journeys such as the one into the interior north.8 Let us accept at least the possibility that at certain points Sora—who wore a monk’s robes—feigned poverty and engaged in mendicancy in order to disguise the fact that he had income from the shogunate for services rendered. Let us also note that most of the time Bashp seems to have had not much actual difficulty finding places to spend the night—or a sequence of nights—and that most of the time these stayovers were at the homes of local dignitaries and persons whose wealth was connected to their power. Based on considerations such as these, the scholar doing a critique of ideology could surely go on to say that Bashp was engaged in an enterprise that at its core represented and reinforced the goals and values of central authorities in Edo and/or in Kyoto. On this reading and in terminology that tends now to sound jargonistic, Bashp traveling into and through northern hinterlands was the representative and purveyor of the “hegemonic discourse” and values of the culture of the political center. Bashp, after all, commenced and concluded his cultural forays into the hinterlands from one or the other of the capitals, either that associated with the court or that of the shogun. Although he portrays the ellipse of his journey north as in the manner of the “aim-less,” the fact of the matter is that he went from, and returned to, places that served as bases. Tabi as travel, then, in this interpretation is seen as a form of cultural colonization, a drawing of outlying areas into the dominant and increasingly extended political realm. By meeting and poetizing with local haikai poets, almost all of whom were fairly wealthy local persons eager to possess and show off their own acquaintance with the culture of the distant capitals, Bashp is assumed to have fused these persons into an aesthetic–personal “network.” And, of course, this network of the arts placed the roof of aesthetics over a power-hungry political– economic system. Bashp in this was nothing more, then, than a very

S           P

85

sophisticated tool in a process whereby outlying areas were colonized and brought under the control of a centralizing authority.

A  B The assumption of Ideologiekritik is that literary texts, once “unmasked”—that is, stripped of their aesthetic disguise—will be forced to stand naked as word deployments in fairly hard-nosed efforts to hold or extend political and social power. And, it is assumed, this will work well only if a given writer or poet in the service of some ruler or would-be ruler skillfully conceals the fact of that service but then invents narratives or poetic structures that in one way or another rationalize his or her hidden sponsor’s possession of power. One way a writer might do this is by importing into the text references to either philosophical or religious concepts that have the effect of shifting the blame for deplorable social conditions away from those who rule. The notion of karma in India or the West’s concept of “God’s Will” are viewed, at least by persons skeptical of them, as blame-deflecting constructs. Bashp, however, lived in an era when intensive Japanese studies of Chinese philosophy would certainly have given him knowledge of those Chinese thinkers who interpreted deplorable social conditions as indications that a given emperor or dynasty was itself unjust and deserving of replacement. If so, what are we to make of the following? A text that has proven irksome to readers and scholars fond of Bashp and his perspective on things is the following from Nozarashi kikp (“Journal of Bleached Bones in a Field”, 1684). I render it as: While walking along the Fuji River, I caught sight of a child, barely three years old, crying its heart out on the riverbank. It was obviously abandoned by parents who must have concluded that this child would never be able to survive the ride over life’s river and the rough rapids it includes. The child seemed as fragile as bush-clover buffeted by autumn winds, about to be torn off and scattered. I gave it some food from my satchel. saru o kiku hito sutego ni aki no kaze ika ni

Poets pitied monkeys, but a discarded child, stung by Fall winds— what’s the response to that?

How did this happen? Are you, child, hated by your father? Neglected by your mother? No, it’s not a matter of either a father’s

86

W      R. L F    hatred or a mother’s failure to care. There is only one thing for you to do, child. That is, raise a complaint about your bad lot to “heaven.”9

This passage and the moral dilemmas it poses have, to be sure, not escaped the attention of critics in modern Japan. Kuwabara Takeo, for instance, criticized Bashp severely: He compares the child to bush-clover in autumn winds, gives it something to eat, and goes on his way. Maybe this dealing with another human being as merely a thing in nature is what can be expected from a poet of nature. But this is not the behavior of someone we think of as humane. Why could he not at least have brought the child along with him to the nearest inn?10

Yamamoto Kenkichi sharply disagreed, preferring to pass a much less harsh judgment on the poet. He wrote: There are lots of people who criticize Bashp’s coolness in turning away from the abandoned child and continuing on with his journey. But to criticize him is easy for moderns like ourselves who can easily bring a lost child to the nearest police-station. Those who would criticize Bashp, however, should do so only on the basis of having themselves given up all their own material possessions to benefit orphans, impoverished folk, and the wounded veterans of war.11

Yamamoto backs up this perspective by materials showing that during the Edo period there were places in Japan where the poverty was so severe that people faced with pregnancy and another “mouth to feed” routinely aborted such fetuses or denied them breath when they emerged from the womb. Yamamoto, correctly I think, insisted that what would be moral responsibility in the twentieth century may not be automatically imposed upon a time, place, and societal situation vastly different from our own. It is highly unlikely that an innkeeper in the next town—or any town in that time and place—would have accepted for adoption a three-year-old waif brought in by Bashp. Kuwabara’s “solution” to the moral dilemma does appear, in that sense, simplistic.12 Although, as we will see, there are ambiguities in this passage, there is one thing free of all equivocation. What Bashp does make very clear in this short text is that parents forced to rid themselves of children were not to be judged as at fault. The poet’s words, of course, are addressed beyond the three-year-old child to the reader. This kind of terrible situation, he insists, does not come about because parents hate

S           P

87

their children or are insouciant about abandoning them. Bashp is very clear about this point; he explicitly rejects the notion that the plight of the waif was due to parental failure.13 Where Bashp leaves things in tantalizing ambiguity is in what comes next, where he writes of himself as having said: “There is only one thing for you to do, child. That is, raise a complaint about your bad lot to ‘heaven.’ ” My reason for putting scare quotes around “heaven” is, of course, that how we interpret the simple term ten in the text is of crucial import. Bashp, I suggest, presents us here with a word that has been very strategically placed and meant to be interpreted in and through the poet’s own rich knowledge of Chinese texts. We can begin with accepting the view, probably first suggested by Hirota Jirp but endorsed by Tanaka Yoshinobu, that the textual background to this whole episode is found at the end of the sixth book of the inner chapters of the Zhuangzi.14 In Victor Mair’s felicitous translation of that work Sir Chariot visits his impoverished friend Sir Mulberry, and standing outside his door, finds him inside strumming on a lute and articulating a plaint that sounded as though it were between singing and crying: “Was it father? Was it mother? Heaven? Earth?” The voice could hardly sustain itself and the verses were uttered in haste. Sir Chariot entered and asked, “Why is your song like this?” “I am thinking about who might have brought me to this extremity, but can’t come up with an answer. Surely my father and mother would not wish for me to be so poor. And heaven shows no preference in whom it covers nor earth in whom it supports. How could they show preference in making me poor? I seek to find who might have done it, but can’t succeed. Well, perhaps it was simply destiny that brought me to this extremity!”15

Surely this passage informs Bashp’s insistence that the abandoned child’s parents were not to blame. But if not they, then who? Bashp had surely raised the ethical ante. Sir Mulberry complained of poverty, but Bashp confronts a wailing child who will likely die. And, if poets such as Du Fu, Li Bo, and their Japanese imitators had seen fit to lament the plight of monkeys, surely that of an abandoned

88

W      R. L F   

human child, Bashp claims, was a fortiori pitiable. I think we cannot be satisfied with the notion that his only recourse was to place the blame on something beyond the human realm such as “destiny” or “fate.” This is why I find inadequate Nobuyuki Yuasa’s translation of Bashp as having said here: “this child’s suffering has been caused by something far greater and more massive—by what one might call the irresistible will of heaven.”16 One problem with such a translation is that it does not take with sufficient seriousness the tense of the words that Bashp uses to address the child, namely, “There is only one thing for you to do, child. That is, raise your cry of complaint about your bad lot to ‘heaven.’ ” Bashp writes nake, an imperative form of naku, to cry. I suggest, however, that in this context “to cry” seems to mean something other than an internally directed and self-enclosed act of self-pity. The crying here is a crying out. It is directed outward and, in fact, really suggests something such as “Do, indeed, make an audible complaint!” We are getting, I suggest, closer to be able to say what ten means in this passage. Furthermore, we are benefited, I suggest, by accepting the gloss on ten as tenmei in the Sugiura text.17 And precisely because tenmei is the Japanese rendering of the Chinese concept of tianming, we are immediately put into a context where politics, rulers, the ethics of ruling, and the quality of a given reign cannot be kept out of the picture. Tianming, usually translated as “heaven’s mandate,” may have one pole in the invisible empyrean but its other is in the visible empire. The two are interconnected. Responsibility, effectively, devolves upon those within the human world, however much persons there may have gotten deified. Benjamin Schwartz notes: “At its deepest level, the idea of Heaven’s Mandate presents us with a clear apprehension of the gap between the human order as it ought to be and as it actually is.”18 And, as is well-known, Mencius counseled those within a given realm to pay close attention to its conditions. The presence of inequality, injustice, and so on could be indicating that a given ruler or dynasty had lost heaven’s mandate and that a change of dynasty, however achieved, may be in order. Of course, Japanese philosophers and political thinkers were very skittish about seeing the Mencian concept of a withdrawable tenmei as in any way applicable to the “eternal” institution of their own tennp and Japan’s single, unchanging dynasty.19 However, exclusion from the Mencian idea did not, we may infer, apply to the pseudo dynasty of the Tokugawa shogunate. Edo-era people with some historical sense knew that the shogunate had had a beginning; and, if so, it

S           P

89

possibly might have an end as well. Unlike the imperial institution in Japan, the shogunate could potentially lose its “mandate.” Tianming/ tenmei could to that degree have political, even potential regime-change relevance in Japan. It is noteworthy that Bashp finds this deplorable social situation somewhere near Hakone and in the larger vicinity of Mount Fuji— that is, West of Edo and well within an area under the domain of the shogunate. He is not in the far northeast. Thus, we can find, I suggest, an implication not too far from the surface of Bashp’s advice to the child to raise a cry of complaint to ten. To complain to heaven would at the same time be to express dissatisfaction about how the realm was being ruled and managed. To write about heaven as needing to hear a complaint about one’s lot in life was necessarily also to implicate those who administer the realm as being persons who, in concrete and specific terms, were at fault. And, of course, what matters here is not the child’s lament, one certainly not going to be heard in Edo, but rather the poet’s inclusion of this in a work intended to have a readership. In Bashp’s case this locating of responsibility would have pointed to whatever persons or agencies were responsible for conditions of poverty, infanticide, child abandonment, and the like that the poet came upon while on his travels. If Ideologiekritik were correct about Bashp being clandestinely engaged in doing the business of the shogunate, we would not expect him to have made a reference to abandoned children. To do so was to indicate that all was not well in Japan—that is, Japan under the shogunate. Clearly said to be neither the fault of parents nor something to be blamed on an impersonal fate, the responsibility for conditions found by Bashp, he implies, must be laid at the doorstep of another human institution. Other than the shogunate there was no obvious choice. In fact, in Oku no hosomichi, when deep in the interior, Bashp registered his sense of seeing deplorable social conditions. When referring to how struck he was by the highly unusual and pathos-eliciting names of towns there, he says that they are “fitting names such as Not-Knowing-Parents, Not-Knowing-Children, Rejected-Dog, Sentback Horse, and the like.”20 In noting that such names are fitting the poet passes along his judgment that they reflect real conditions, states of poverty leading to various kinds of social dislocation and a jumble of broken relationships. This passage, significantly, immediately precedes his depiction of the pathetic lives of two traveling prostitutes met by him at an inn at the Ichiburi barrier. In that case too we have Bashp pausing to describe people in distress, people whose lives were

90

W      R. L F   

made miserable by the conditions of the society in which they lived. These also are things that a poet on assignment to whitewash society’s problems—and the regime that had responsibility for them—could simply have left out of his text. Bashp clearly chose not to do so. And that is significant.

P   P I hold that any reduction of literature to politics—and to nothing but politics—is misdirected and mistaken. This is not, however, to deny that literature will often have a political dimension or to suggest that we need not be aware of such. Of political treatises we may reasonably say that they are, when all is said and done, about politics. The same may be said about religious tracts—although not about multilayered works such as the Bible. To the degree that a given text will qualify as a literary one, we should never be able with justice to apply to it the when-all-is-said-and-done kind of wrap-up finalization. What counts as literature will escape our attempts to reduce it to one kind of thing. So, although we need to recognize a political dimension in a text, it is very different from assuming that, since all other dimensions are merely disguises put on to dissimilate, the real meaning of a text comes clear only when we have stripped off its masks and see it as nothing other than a statement about power and a ploy in a cultural–political contest. In The Pleasures of Reading in an Ideological Age (1989), Robert Alter astutely observes that “the literary canon, for all its supposed attachment to ideological values, often incorporates texts that run counter to the dominant ideology of the culture.”21 This is instructive. In the earlier section I argued that even if hypothetically Bashp had been given some material support from Edo authorities and/or had been expected to extend the capital’s ways and cultural norms into a still only loosely incorporated part of the main island of Japan, a close look at a text such as Oku no hosomichi suggests that he was not acting like an ideological patsy. He could, with subtlety, suggest that all was not well in the shogun’s larger “realm” and that the shogunate deserved the blame for that. Alter is right about things in a literary text that “run counter to the dominant ideology.” In a truly rich text we can sometimes see the poet expected to be peddling ideology actually subverting it. If, theoretically, Bashp was expected to be bringing the “hegemonic” culture to a backward area, we can see him resisting what is simplistic in that assumption as well. He did not assume that persons

S           P

91

or communities relatively impoverished were necessarily incapable of having or knowing refinement. No Sei Shpnagon, Bashp actually subverted the assumptions of her Heian era urban snobbery. If expected to be merely celebrating the culture of the privileged and sophisticated while traveling among rustics and the north’s poor, Bashp was, in fact, a poet insisting on the discovery and disclosure of beauty in the life contexts of rural, agricultural people. One of Bashp’s concerns, it seems to me, was to locate and celebrate a form of beauty that had absolutely nothing to do with the kind of beauty associated with the collectible objects and commodifiable skills prized by both merchants and samurai—persons for whom the possession of “art” was taken as an indicator of status and worth. It was, after all, when he was moving away from the capitals and into what, in the eyes of people in the capital, was the lowly esteemed interior, that he redescribed furyu—or “stylishness” in the arts—in a strikingly new way. At Shirakawa he wrote:22 furyu no hajime ya oku no taue uta

That might be translated as: Refinement rose up from the rice-planting songs of these country-folk

However, because in English the word culture has a fortuitous equivocality—as both the ambit of human creativity and as an automatic growth process in the botanical realm—we could also render this poem into a more universalized idiom as: Culture: that which grows from the songs of country peasants planting

The poem’s deft image is that of peasants bent over, surely uncomfortably, even excruciatingly, to insert young rice seedlings into flooded paddies. The vocalizations normally expected from them would be those of pain—moans, groans, or deep sighs. But this is precisely where Bashp captures something that independent, empirical observation of such scenes has fully verified, namely, that such farmers have long done this kind of work often while singing rhythmically. That is, Bashp cites such peoples’ choice to react in a way other than in what their physical discomfort and pain might dictate. Rather than groan, they sing. And it is in the making of precisely such a choice that Bashp

92

W      R. L F   

sees the origin of the human being’s cultivation of its humanity, even the genesis of what others would see as the formation of “refinement.” He turns the tables on any agenda that would envision colonization by enculturation. Rejecting any assumption that poor peasants have empty lives and live without the refinement of true—that is, urban— culture, he ascribes the very origins of cultural refinement to something that arose spontaneously within the work patterns and practices of dirt-poor farmers. And the presence of these songs in the paddies was, we may assume from confirmations by other sources, not a romanticized projection but a description of what, in fact, was there.

S E One other problem with the critique of ideology is that it too breezily dismisses—or reduces to mere politics in another guise—what sometimes is a very valuable existential dimension within a literary text. Ideologiekritik focuses exclusively on the living and their livelihood and obscures the fact that all humans have an additional concern about dying and what it might mean to be dead. Bashp’s work, I hold, is seriously short-changed if this concern of his gets lost in the processes of modern interpretations. I approach this in a somewhat roundabout fashion and begin by noting that people in medieval and even modern Japan puzzled, with considerable intellectual profit, over a question that the modern West ignores or largely assumes to have been “solved” by the creation of legal fictions. That question has to do with ownership both of “fresh” corpses and of whatever turns out to be the physical residue—hair, ash, bones—of a body no longer alive. Medieval folk recognized that the question of who exactly owns such items is fraught with religious and philosophical significance; the modern West, by contrast, dismisses such medievals’ concerns as superstition attached to “relics” and goes on to assume that the whole problem can be solved simply by having the not-yet-deceased person prepare something called a “will”—that is, what can, at least legally, determine in advance all matters having to do with postmortem ownership. And it has come to be assumed that even the corpse of the person contemplating death is his or her property and is something that can be included in this directive issued in advance. Via such a will and testament the deceased’s capacity both to own and to designate disposition of his or her corpse survives that body’s capacity to be empirically alive. Our own law has hidden within it an unstated philosophical, maybe even a religious, assumption: the “person” can outlive

S           P

93

his or her biological body and can own it even though unable to affect it in any other way. This, for us, has become a normalized procedure and one we assume should also be normative. Yet, even for many Japanese who are contemporaneous with ourselves, such matters are allowed to remain more complex—both in concept and in praxis. We can see that the ownership question remains subtly nuanced among them in their continuing practice of giving bodily residue, usually in the form of ash and/or very fine bones, to immediate family members for safe-keeping in one place or another. Residue of the deceased sits on a home or temple altar. And we see the whole question of disposition remaining interestingly mooted when, for instance, Japan’s Organ Transplant Law of 1997 gives living relatives of the deceased the right to veto the removal of organs for transplantation even if the deceased had expressly willed their reuse by way of an organ donation card. Something willed by someone no longer alive does not, in such a case, trump the expressed will of the living relatives. Clearly, to be able claim the right to possess something, it is beneficial to be alive. Even though efforts may be made to revise Japan’s organ donation law so as to bring it more in line with the laws of Europe and America, there can be value in recognizing what is, in fact, an item of common sense in the resistance to such change. It is based on the obvious fact that although it may be legally possible for someone now deceased to go on “owning” his or her body, physical ownership of that body by the deceased has, by virtue of that death, become a patent impossibility. A kind of common sense about things also appears to move Biblical interpreters, for instance, to refrain from seeing Jesus as having suggested a physical action when in Matthew 8:22 he issued the injunction: “let the dead bury the dead.” Clearly it would have to be someone not dead who does the burying or the cremating or whatever. And, thus, in the most primary and empirical way both the body of the recently deceased and its later residues will become the actual possession of persons other than the deceased. Medieval Japan’s preoccupation with mujp or the Buddhist teaching of radical impermanence fostered an ongoing conversation about the necessary limitations of ownership. One especially powerful way to do this was through pictorial representations of the human being’s physiological deterioration after dying. Not only the skeleton but stages in the material decomposition of the corpse were put under review. And not infrequently the corpse was portrayed as already largely owned by the worms, insects, and dogs that were in the process of consuming it.

W      R. L F   

94

The most arresting example of this in the visual arts of the early medieval period is the Kuspshi emaki, a scroll that graphically depicts nine stages in the natural deconstruction of the body of a dead woman—in that case most likely that of the celebrated beauty Ono no Komachi. The scroll moves the viewer all the way from Komachi alive and in gorgeous court robes through the following nine phases: just dead; bloated-belly; blood-smeared; dispersed flesh; greenish liquifying; being consumed by beasts; stripped skeletal; scattered bones; and part of a mound. In the end what was once a person is no more than sun and windswept bones scattered on an open field.23 This final image, although less grotesquely presented, was in many ways to be the connective theme and title of Bashp’s first travel diary, Nozarashi kikp. In this connection it can be helpful to recall that Saigyp, selected by Bashp as one of his exemplars in the arts, probably also was known by him to have contemplated dying while traveling. And for Saigyp, whose sensibility was that of a medieval person, katami, the word we now translate as “keepsake,” often encompassed some physical remnant of the mi, the body of the deceased. On the first of his two journeys to the northeast, the one he undertook in 1147, Saigyp stopped—as Bashp was to do more than five hundred years later—at the grave of Sanekata, a poet who had gotten himself exiled from the capital. Saigyp wrote:24 While in the [far northern] province of Mutsu I came across an unusual looking grave-mound. I asked whose it was and was told that it belonged to a middle-captain of the palace guard. When I persisted in inquiring exactly who this person might have been, I was informed that it was Fujiwara Sanekata, and I was deeply saddened. Even before learning the details, I had sensed the pathos in this scene of frost-shriveled pampas grass—so fragile it was almost invisible. Later, in trying to express my feelings, adequate words were almost unavailable: kuchi mo senu sono na bakari o todomeokite kareno no susuki katami mi zo miru

The verb kuchiru and its cognates come in waka as a bit of jolt. Meaning to “rot,” “decompose,” or “decay,” kuchiru signaled a topic less than comfortable to courtier sensibilities. It is, though, the kind of word we should not be surprised to see Saigyp use. If we insist on

S           P

95

the physicality and materiality of Saigyp’s diction here, we would need to translate this verse somewhat as follows: One part of him escaped decay—his name, still around here like this field’s withered grass: my view of the relic he left.

Pampas or miscanthus may very well be the brownest, driest flower in nature, and Saigyp understandably saw the linkage between it and what he had hoped might be at least a bone, however dry and bleached, remaining of what had been the body, the katami, of a dead friend named Sanekata. Bashp, no doubt, had this poem in mind when, also in the north, he wrote the following after being moved by seeing the ruined castle where the twelfth-century warrior Yoshitsune and his retinue had met deaths:25 natsugusa ya tsuwamonodomo ga yume no ato summer grasses: all that’s left of all the warriors’ dreams

As a warrior who had become a monk himself, Saigyp had included in his verse and prose clear criticisms of the folly of militarist ambitions. And Bashp appears to have underscored that point. After Saigyp and on the way to Bashp we find the fascinatingly ludic work by the radically eccentric Zen poet Ikkyu (1394–1481) entitled Gaikotsu (Skeletons).26 This work’s text and especially its illustrations mock the notion of death as repugnant or tragic. In it skeletons make merry, dance, and copulate. The point would seem to be that people’s wariness about corpses —as if constellations of bone somehow still have life, personality, and the capacity to harm the living—is really something laughable. And in finding these superstitions and fears risible, Ikkyu surely showed his rootedness in the trajectory taken by the Chan/Zen tradition in which he stood. In that subgenre of poetry called the jisei, a poem composed as one’s “last” verse before dying, one can trace throughout Japan’s medieval period a progressive playfulness, even jocularity, expressed in the face of death. A comic spirit drives its way even into this genre in

W      R. L F   

96

which some kind of utter seriousness would ordinarily have been expected. This change, I surmise, is also due to a growing impact that the Zhuangzi was having on Japanese literature. Statements such as the following in that work enabled Buddhists to think of the impermanence found in dying as able to be greeted with playfulness rather than lugubriousness: “The true man of old knew neither fondness for life nor aversion to death, was neither elated by going forth nor reluctant to return. Casually he went and casually he came.”27 Having much earlier written about how Bashp “looked back,”28 I want here to note a few things about how he looked forward. By that I mean his ability to anticipate his own death and bodily remains—and to incorporate such into the particularity and special genius of his poetry. Bashp picked up and further refined what I would call the etiquette of how one should regard death. He lived, I suggest, at a time when social convention held that however free we might be to mourn the death of others, we ought to envision our own dying and death as no great loss to either the world or to ourselves. I do not mean to suggest either that this is “only” a social convention—and, therefore, dismissible as insincere—or that this was the only social context where one might find it. I only mean that as an etiquette—and indeed with strong Buddhist ethical teaching behind it—it is brilliantly present in Bashp. The contrast between feelings of heaviness concerning the death of others and lightness in view of one’s own demise is sharp in Nozarashi kikp. Stopping during the fall of 1984 at his own ancestral home for the first time since his mother’s death the previous year, he writes of having been shown, as a katami, strands of her whitened hair and of then writing:29 te ni toraba kien namida zo atsuki aki no shimo If held in my hand, my hot tears would melt these autumn-frosted items

In the same work, his first travel journal, Bashp was already looking ahead to his own death. It was one he optimally viewed as occurring while journeying and sufficiently alone so that his remains would become—as in the earlier medieval texts and pictures—merely scattered bones lying on a moor. This was not just an aesthetic but also an ethic of dying. A deep concern to collect and care for what might be “left”—that is, the

S           P

97

katami—of others who were deceased was not, according to the operative notion of “the good” here, to be transferred to one’s own remains. Concerning them a virtual nonchalance, one in the mode of the Zhuangzi fused with Buddhism, was deemed appropriate. A clear-eyed, unworried, and even light-hearted vision of one’s own remains as skeleton fit well into such an approach. It was exactly this that Bashp adopted in the opening lines of Oi no kobumi, his “Knapsack Notebook” of 1687. Just prior to his short and personal list of artistic heroes—Saigyp, Spgi, Sesshu, and Rikyu—he describes his own “self” as:30 A hundred bones, nine orifices, and something inside. This provisional thing is called “In-the-Wind-Flapping-Priest.” With the slightest hint of wind it gets moved and makes sounds. This something in me started mouthing haikai a long time ago and that has turned out to be the preoccupation of a lifetime.

Commentaries that read “something inside” (uchi ni mono) as meaning tamashii—that is, “soul” or “ghostly spirit”—are, I suggest, mistaken. The direction of this passage lies away from any substantializing of a death-transcending self-stuff, even that of the intangible variety. Provisionality and contingency are what are important here. The image of himself as made up of bones and orifices is not gratuitous; this text analogizes the way vocalized—that is, sounded—poems passed through him as poet to the way exposed human bones might make audible sounds when a breeze passes through them. This image, strange perhaps to modern sensibilities, is one through which Bashp, ironically, naturalizes his own talent. His lifelong voicing of verse is, he suggests, like an ordinary event in nature: automatic, unstoppable, present there because of the nature of things and not because he is something unique. Although we make think of it as a poetic conceit, the “journeying” image had a similar naturalizing role. Being on a journey, by definition, puts a person at a distance from the urban contexts of gathered human beings and their social projects. No amount of late-twentieth-century insistence that “everything is a social construction” can wipe away the Zhuangzi’s sense that “the way” is not a cultural artifice but, instead, its opposite. For Bashp, just as for Saigyp and many Chinese poets, actualizing the way by literally going on a journey was to enhance the seriousness of life and to reduce radically the impression that dying constitutes irrecoverable loss. In the balanced structure of contrasts that comprise that core image, one side is comprised of being alive/being awake and daytime;

W      R. L F   

98

its opposite is made up of being dead/being asleep and darkness. But Bashp can play somewhere between those poles of contrast in a poem such as the following:31 shini mo senu tabine no hate yo aki no kure not even dead yet at a journey nap’s end: autumn dusk

The surprise of being still alive does not vitiate the sense that there would be nothing tragic or regrettable in being dead. Medieval Japanese lore had accounts of people coming across windswept skeletons in fields and hearing eerie whistling sounds—as if tongueless skulls still had some kind of voice. For the whole of his life, Bashp here suggests, a poetic something had been inside the mix of bones and orifices that was his body and it, just like a skeleton in a field whistling in the wind, made noises that came in the form of iterated poems. But it was as a skeleton on an open moor that became Bashp’s most striking way of imaging himself as dead. Nozarashi kikp opens with the poet announcing that during the early autumn of the first year of the Jpkyp era (1684) he, in imitation of an earlier Chinese poet–monk who took along no provisions but traveled ecstatically looking at the moon, left his dilapidated house by Edo’s Sumida River. But immediately he was smitten by a very chill wind and wrote:32 nozarashi o kokoro ni kaze no shimu mi kana

One way of attempting a translation would be: moor bleached bones: right to the heart anticipating this comes a body-slicing wind

Stepping outside the door of his Edo home, Bashp’s mind and heart are full of the image of being, somewhere far down the road, a skeleton through which the wind might whistle. But just then a wind so cold that it seems to go right into his body hits him. His body, although not yet a skeleton, feels penetrated as the wind slices into it and seems to reach the very interior mind/heart that had been imagining its future.

S           P

99

The genius of this poem lies in how it collapses an imagined future under the weight of an experienced present but at the same time insists that the content of the present experience is fundamentally one with what had been imagined, a mere moment ago, only as a far-off eventuality. The whole of the future is thrown right into the present. The skeleton he will someday be, we are asked to recognize, is none other than the bones that are the enfleshed core of what he now is. And the wind, cold but in no way an antagonist, goes right into and through him. And like the sounds made by winds passing through skeletal bones in a field, the ensuing paragraphs and poems of Journal of Bleached Bones in a Field, having been so intriguingly introduced, are, so the image suggests, words and lyrics that have made their way through the mind and body of Bashp. A kind of natural event is unfolding. And the wind-carried poems are now free to roam the world.

N Some paragraphs and one section of this chapter were published earlier in an essay, “Poverty in Extemis: Reduction and Bashp,” in (eds.) Hans-Georg Möller and Günter Wohlfart, Philosophieren über den Tod (Köln: Editions Chpra, 2004), 113–151. 1. “Taidan: Bashp to tetsugaku,” in Konishi Jin’ichi, ed. Fuga no makoto (Tokyo: Kadokawa shoten, 1970), 338. The recorded conversation was between Konishi and Donald Keene. Konishi adds that there were, of course, scholars who resisted this dismissal of Bashp (338). 2. For translation of Hattori Dohp’s (1657–1730) quote of and commentary on Bashp’s words, see Makoto Ueda “Basho on the Art of the Haiku: Impersonality in Poetry,” reprinted in (ed.) Nancy Hume, Japanese Aesthetics and Culture: A Reader (Albany: State University of New York, 1995), 161; for Dohp’s Sanzpshi passage, see Minami Shin’ichi, Sanzpshi spshaku, “Akazpshi” (Tokyo: Kazama Shobp; reissued 1980), 114–115. 3. “Introduction,” in (trans.) Cid Corman and Kamaike Susumu, Back Roads to Far Towns: Bashp ’s Oku-no-Hosomichi (New York: Grossman Publishers, 1968), 8. 4. See, for instance, Amino Yoshihiko, Higashi to nishi no kataru Nihon no rekishi (Tokyo: Soshiete, 1982); and Uma fune jpmin: Tpzai Kpryu no Nihon rettpshi (Tokyo: Kawai Shuppan, 1992). 5. Alexander Bay, “The Swift Horses of Nukanobu: Bridging the Frontiers of Medieval Japan,” in (eds.) Gregory M. Pflugfelder and Brett L. Walker, JAPANimals: History and Culture in Japan’s Animal Life (Ann Arbor, MI: Center for Japanese Studies, The University of Michigan,

100

6.

7.

8. 9.

10.

11. 12.

13.

14.

W      R. L  F    2005), 92. Other scholars in the West stressing the difference are Mark J. Hudson, Ruins of Identity: Ethnogenesis in the Japanese Islands (Honolulu: University of Hawai’i Press, 1999); and Mimi Hall Yiengpruksawan, Hiraizumi: Buddhist Art and Regional Politics in Twelfth-Century Japan (Cambridge, MA: Harvard University Press, 1998). “Kita pu” names the northern (kita) portions of the provinces of Mutsu (or Michinoku, p) and Dewa (u). Sugiura Shpichirp, Miyamoto Saburp, and Ogino Kiyoshi, eds., Bashp bunshu (Tokyo: Iwanami shoten, 1959), 76. References to this work in subsequent paragraphs will simply refer to “Sugiura” and provide the relevant page reference. All translations here, unless otherwise indicated, are my own. Sugiura, 88; cf. David Landis Barnhill, Bashp ’s Journey: The Literary Prose of Matsuo Bashp (Albany: State University of New York Press, 2005), 66. Interestingly, mummified bodies of three Northern-branch Fujiwara leaders—Kiyohara, Motohira, and Hidehira—remain in the Hiraizumi temple. Muramatsu Tomotsugu, Bashp no tegami (Tokyo: Daishukan, 1985), 225–234. Sugiura, 36–37. The rendering into English of the title of Nozarashi kikp, as well as some others later, are ones I find especially felicitous and are borrowed from David Landis Barnhill’s Bashp ’s Haiku (Albany: State University of New York, 2004) and his Basho’s Journey; for translation of this section of the journal, see Basho’s Journey, 12. From Kawabara Takeo, Daini geijutsuron as quoted in Uwagawa Shpsuke, Nozarashi kikp no kaishaku to hypron (Tokyo: Oofusha, 1968), 25. Since the Japanese loanword hyumanisuto differs considerably from what our word “humanist” now signifies, I have translated it as “humane.” Whether this was an actual or merely fictive episode hardly matters in evaluating the moral stance embraced within the text. Ibid., 29. The effect of practices in roughly contemporaneous Europe was not much different. Although there were religious institutions to which abandoned infants—euphemistically called “foundlings”—were brought, the mortality rate among these infants sometimes approached 90 percent. See John Boswell, The Kindness of Strangers: The Abandonment of Children in Western Europe from Late Antiquity to the Renaissance (New York: Pantheon Books, 1988). This contrasts sharply with the judgment by some Confucian and National Learning (kokugaku) writers in the early nineteenth century who focused exclusively on parents as responsible for such things. See my Liquid Life: Abortion and Buddhism in Japan (Princeton: Princeton University Press, 1992), 105–111. Tanaka Yoshinobu, Bashp: Futatsu no kao, (Tokyo: Kodansha, 1998), 225. Also exploring the profound impact of the Zhuangzi on Bashp’s

S           P

15. 16. 17. 18.

19.

20. 21.

22. 23.

24.

25. 26. 27. 28.

29. 30. 31. 32.

101

sense of the meaning of travel is Peipei Qiu’s essay in this volume (chapter 3). Victor H. Mair, Wandering on the Way: Early Taoist Tales and Parables of Chuang Tzu (Honolulu: University of Hawaii Press, 1998), 64–65. Nobuyuki Yuasa, trans., The Narrow Road to the Deep North and Other Travel Sketches (Baltimore: Penguin Books, 1966), 52. Sugiura, 37, n. 22. Benjamin I. Schwartz, The World of Thought in Ancient China (Cambridge, MA: Belnap Press, 1985), 53. See also Robert Eno, The Confucian Creation of Heaven: Philosophy and the Defense of Ritual Mastery (Albany: State University of New York Press, 1990), 23ff. See William R. LaFleur, “Heart Purity and Utilitarianism: Ethics in Mencius, Wang Yang-ming, and Nishida Kitarp,” in (eds.) Marthe Chandler and Ronnie Littlejohn, Polishing the Chinese Mirror: Essays in Honor of Henry Rosemont Jr. (Chicago & La Salle, IL. Open Court Publishing Co., 2006) (in press). Sugiura, 91; cf. Barnhill, Basho’s Journey, 70. Robert Alter, The Pleasures of Reading in an Ideological Age (New York: Simon and Schuster, 1989), 31. See also Michael Rosen, On Voluntary Servitude: False Consciousness and the Theory of Ideology (Cambridge, MA: Harvard University Press, 1996), for an analysis of the blind spots within Ideologiekritik. Sugiura, 76; cf. Barnhill, Bashp ’s Journey, 55. A fine, easily accessible reproduction of the “Kuspshi emaki” is in Komatsu Shigemi, ed., Nihon no emaki, Vol. 7 (Tokyo: Chupkpronsha, 1987), 109–119. Kubota Jun, ed., Saigyp zenshu (Tokyo: Nihon koten bungakkai, 1982), 85–86. William R. LaFleur, Awesome Nightfall: The Life, Times, and Poetry of Saigyp (Boston: Wisdom Publications, 2003), 24. Sugiura 84; cf. Barnhill, Bashp ’s Journey, 62. James H. Sanford has translated this illustrated text in his Zen-Man Ikkyu (Chico, CA: Scholars Press, 1981), 201–216. Mair, Wandering on the Way, 52. “The Poet as Seer: Bashp Looks Back,” in William R. LaFleur, The Karma of Words: Buddhism and the Literary Arts in Medieval Japan (Berkeley: University of California Press, 1983), 149–164. Sugiura, 38; cf. Barnhill, Bashp ’s Journey, 16. Sugiura, 52; cf. Barnhill, Bashp ’s Journey, 29. Sugiura, 40; cf. Barnhill, Bashp ’s Journey, 18. Sugiura, 36; cf. Barnhill, Bashp’s Journey, 13.

This page intentionally left blank

P  II

T A A  P 

This page intentionally left blank



D V  B¯ ’  H  Haruo Shirane

O

ne of the most dramatic transformations in the social, political, and cultural history of Japan occurred in the seventeenth century, in the transition from the medieval period to the early modern era, a century that witnessed the emergence of urban culture, centered on the chpnin (urban commoners), the spread of mass education, and the advent of printing, which transformed literature into a commodity for huge markets. Until then, written literature was almost entirely the possession of a small elite group of aristocrats, priests, and high-ranking samurai. In the medieval period biwa hpshi or traveling minstrels had chanted such military epics as the Heike monogatari (Tale of Heike) and the Taiheiki (Chronicle of Medieval Japan) to a populace that could neither read nor write. The average samurai was illiterate, as were farmers and craftsmen. With the emergence of a new socioeconomic structure, the government encouragement of education, and the spread of print capitalism, however, this situation changed drastically. By the middle of the seventeenth century, almost all samurai, now the bureaucratic elite, were able to read, as were the middle to upper levels of the farmer and chpnin classes. This newly literate populace transformed haikai, heterodox linked verse, into the first truly popular literature of Japan in the sense of being widely practiced and read by commoners. In the second half of the seventeenth century, haikai books (over 650 separate titles) were second in popularity only to Buddhist texts among Kyoto publishers, who published an estimated 300,000 volumes in first editions alone. Sequel to a Mountain Well (Zoku yama no i), a haikai anthology published in 1667, includes the note: “967 authors, 48 provinces, 5035 verses.”

106

H  S 

In the seventeenth century, literary interest turned toward both the new culture and lives of Tokugawa commoners and toward the Japanese and Chinese “classics,” which were printed and widely distributed for the first time. In the medieval period, there had been a pressing need for a frame of reference that could transcend the constant warfare and social turmoil, that could provide an explanation for the constant uncertainty and seeming lack of control over individual fate. This had led to a cultural gravitation toward Buddhism and a nostalgic reverence for the Heian classical tradition, which presented an elegant, self-contained, seemingly timeless “other world” in which one could escape from the ugly and disturbing realities of contemporary life. These two movements resulted in a stream of medieval literature—produced mainly by a small circle of poet–priests and scholars deeply rooted in the traditional culture of Kyoto—that was spiritually informed and neoclassical in focus and that turned to Buddhism and the Heian literary canon for authority. As Japan entered an extended period of peace and political stability and a new society emerged in the seventeenth century, the cultural perspective turned toward “this world,” and literature began to separate from Buddhism. The need to seek authority and cultural models in the Heian classical tradition, however, continued to exist, this time among a newly literate commoner audience. At the urging of scholar/poets such as Matsunaga Teitoku, haikai became a means of understanding and absorbing the Japanese “classics,” which were rapidly reconstructed (to include, e.g., Noh drama) and reformulated to meet the tastes and needs of the new readership. Haikai masters such as Matsuo Bashp gave wealthy commoners and high-ranking samurai a sense of cultural identity (which included China), of continuity to a hallowed past, which they had never possessed but were eager to acquire. Significantly, this acquisition of literary and cultural “tradition” occurred through the medium of or with the aid of the new vernacular languages. A salient characteristic of Tokugawa culture was the intense interaction of diverse and frequently binary languages—urban and provincial, Kamigata (Kansai region) and Edo (Kantp region), samurai and chpnin, aristocratic and commoner, classical ( gago) and vernacular (zokugo), Chinese and Japanese, to mention only the most prominent. In previous periods, regional and social differences had also generated contrastive languages, but the cultural center had always been in Kyoto or with those who had been closely associated with its cultural milieu. The shift of the political center to the east, to Edo, which became a major metropolis, and the emergence of two distinctive

D  V  B ¯ ’  H 

107

cultural regions (Kansai and Kantp) created a new and highly contested cultural field, and increased the awareness of linguistic and sociocultural differences. Haikai thrived in the intersection of these languages, especially between classical diction ( gago), a highly encoded, self-enclosed, almost disembodied language, which bore the voices of a refined, aristocratic past, and various types of vernacular, popular language (zokugo), which had a strong sense of immediacy, physicality, and referentiality, closely tied to everyday commoner life and material culture. Mikhail Bakhtin has argued that language is not a monolithic, neutral, and closed system but instead is pluralistic and socially and ideologically inscribed. Bakhtin sees language as a constant struggle among systems or groups, between a centripetal, unifying force imposed by those in power, which attempts to impose a unitary order on language, and centrifugal, stratifying forces created by various social groups, which fracture and divide language into many different, socially accented tongues—a phenomenon that he calls heteroglossia. In Bakhtin’s view, every utterance serves as a point where these centripetal and centrifugal forces converge, often in tension or conflict. Commenting on European literature, Bakhtin contrasts poetry, particularly the lyric, which in his view attempts to maintain the unitary order of the established literary language, with what he calls the polyphonic novel, which reflects and encourages the dynamics of heteroglossia.1 A similar kind of sociolinguistic interaction between centripetal and centrifugal forces, unitary and polyphonic genres, had occurred in Japan within the Japanese poetic tradition, particularly from the sixteenth century, an age of great social and political upheaval, through the seventeenth century, when the chpnin emerged to create the base for a new cultural order. Waka, the thirty-one syllable quintessential classical form, was based on the notion of a unitary language; it was a high genre that banned all forms of language not found in the highly circumscribed, aristocratic diction of the Heian classics, consisting primarily of the Kokinshu (Collection of Japanese Poems Old and New, ca. 908), Ise monogatari (Tales of Ise) and Genji monogatari (Tale of Genji). The subject matter was likewise confined to a small cluster of highly conventionalized, elegant topics, focused primarily on love and the four seasons. The same restrictions applied to renga, orthodox linked verse, which carried on the classical tradition into the late medieval period. By contrast, haikai, which drew freely on colloquial Japanese, regional idioms, Chinese phrases, Buddhist vocabulary, and other centrifugal languages, was based on the notion of challenging, inverting, and otherwise subverting the

108

H  S 

unitary language of waka and the Heian classics. Dog Tsukuba Collection (Inu tsukuba shu, 1532), one of the earliest anthologies of haikai, opens with the following sequence. kasumi no koromo suso wa nurekeri

the robe of mist dampened at the hem

The opening verse metaphorically describes a mist that surrounds a mountain, making it look dark and damp at the bottom. The tsukeku (added verse) composed by Yamazaki Spkan (d.ca. 1539–1540), one of the pioneers of haikai and presumed editors of Dog Tsukuba Collection, is: Sahohime no haru tachinagara shito o shite

Princess Saho, with the coming of spring, stands, pissing

In classical waka, Sahohime, the goddess and harbinger of spring, usually wears a robe of mist. The added verse humorously inverts the tone of the previous verse and the elegant associations of Sahohime by using the startling colloquial phrase shito o shite (pissing) and having her urinate while standing (tachinagara) as commoner women did in those days. In a manner that echoes the social upheavals of the time, of gekokujp (overcoming the high by the low), Muromachi haikai poets parodied and debased the metaphors, language, and conventions of the high tradition. In the course of its long history, haikai gravitated toward popular culture, toward bold and often ribald humor, toward those elements banned from classical literature and that initially led to the emergence of haikai. At the same time, haikai also tended to move toward aesthetically refined, spiritually informed literature. These opposing tendencies led to the split in the late medieval period between renga (orthodox linked verse)—referred to by its practitioners as ushin renga—which expunged zoku from both diction and content, and haikai, or heterodox linked verse. In the seventeenth century, with the unprecedented popularization of haikai and the rapid decline of orthodox renga, which disappeared by the end of the century, a similar split occurred again within haikai. Matsunaga Teitoku (d. 1653), the founder of the Teimon school and the leader of early seventeenthcentury haikai, sought to bring haikai closer to high literature by gentrifying the subject matter and tempering the use of colloquial Japanese. By contrast, what came to be called Danrin haikai, which came to the fore in the 1670s and which looked back to Muromachi haikai, was highly parodic and aggressively used the new languages of the time. The latter type of haikai, which gravitated toward vulgar

D  V  B ¯ ’  H 

109

topics, continued to flourish in the form of maeku-zuke (single verse capping) and then eventually as satirical and humorous senryu, which jettisoned the seasonal base and diction of classical waka. The former stream of haikai, which reached from Teitoku through Matsuo Bashp (1644–1694) and Yosa Buson (1716–1783) and then to Masaoka Shiki (1867–1902) and modern haiku, maintained the seasonal base of classical poetry and sought to recreate and widen the tradition through haikai and its heterodox languages. The following hokku, which Bashp composed in 1688, is typical of the manner in which his haikai emerged out of the intersection of contrastive languages/ cultures, takotsubo ya hakanaki yume o natsu no tsuki

octopus trapshaving ephemeral dreams beneath a summer moon

Here, the octopus, which were lowered in the afternoon and raised the next morning, are asleep in the traps, having “ephemeral dreams” (hakanaki yume). The reader’s mind jumps not only across the “cutting word” (kireji)—marked by a hyphen in the English translation— but moves between two kinds of language, between a refined classical phrase, the “summer moon” (natsu no tsuki), associated with the brevity of summer nights, and “octopus traps” (takotsubo), an everyday, colloquial word rooted in commoner culture. The “low,” potentially vulgar word humorously injects new life into a classical cliché—summer moon and the familiar theme of impermanence—while the classical imagery endows the vernacular word with complex connotations and poetic depth. One result of this intense interaction of different languages was the emergence in the seventeenth century of a vernacular literature or culture of mitate (literally, “seeing by comparison”), a literature of double vision that moved back and forth between two starkly different worlds, between that of Japanese and Chinese classics and that of the new material culture of seventeenth-century Japan, each providing a lens or filter to view the other. The audience simultaneously “saw” the world of the classics alongside or through a redone contemporary version. Haikai, which may be considered the ultimate literature of double vision, flourished in this new literary and linguistic space, which required both a familiarity with the classics as well as a ironic distance from it. The kind of parody found in haikai, especially the mitate, resembles honkadori, or allusive variation, a key poetic technique in medieval waka in which the poet must “borrow” sufficiently from the “foundation poem” (honka) so that it can be readily recognized while altering the subject or approach enough to avoid plagiarism.

110

H  S 

The result in the hands of the best medieval poets was the creation of a thirty-one syllable poem with considerable depth and complexity. The same paradigm prevailed in the medieval arts where the function of the artist was not to be as individualistic or as unique as possible, but rather to master the various kata (patterns or artistic conventions) and add cautiously to the established repertoire. In contrast to honkadori, which assumed a common base of diction, tone, and subject matter with the foundation poem and the classical tradition, the mitate was enjoyed primarily for the startling, dramatic, and often witty changes that it imparted to the target text. Ukiyo-e artists such as Hishikawa Moronobu (1618–1694), the so-called father of ukiyo-e and a contemporary of Matsuo Bashp, used the technique of mitate in the visual arts, both alluding to and radically transforming the topics and imagery of the classical source. In reference to the mitate-e (painting) by Suzuki Harunobu (1725?–1770), a successor of Moronobu, the art historian Kobayashi Tadashi notes two types: In one, a modern interpretation is superimposed over the original subject; figures and animals familiar to everyone as characters from famous events, narratives, and paintings from the classical past are converted into stylish young men and women of the present. In the other, . . . (t)hey attempt to amuse by abbreviating, inflecting, or adapting the tone and meaning of the classical poem to current customs and manners.2

In these mitate-e, as in much of haikai, the world of ga (classical refinement and elegance) is transformed into or juxtaposed to that of zoku (popular, contemporary culture) so as to create a striking double vision. Bashp’s mitate-e differs from these mitate-e and from the haikai of Ihara Saikaku in the nature of the zoku, which is not the stylish men and women of the floating world, of the great urban centers, but rather the zoku of the mundane, of the everyday, especially as found in farms and fishing villages of the provinces. The new popular literature and drama, including haikai, emerged out of urban, chpnin culture, especially the three urban centers of Kyoto, Osaka, and Edo— Matsunaga Teitoku was based in Kyoto, and Nishiyama Spin, and Ihara Saikaku, the leaders of Danrin haikai, came from Osaka—but Bashp was a socially marginal figure born in Iga, a small province to the southeast of Kyoto. His family, which came from the lowest rung of the samurai class, had, by Bashp’s time, fallen to the level of a farmer, and ultimately belonged to no particular class or social group. Liminal or marginal figures, the beggar, the old man, the outcast, and

D  V  B ¯ ’  H 

111

the traveler pervade Bashp’s poetry and prose. The natural world of Bashp’s poetry likewise tends to be off-center, dominated by subdued, withered images: shigure (winter rains), susuki (miscanthus), obana (miscanthus in ear), karasu (crow), kpmori (bat). It was not the classical images of Ariwara no Narihira, Ono no Komachi, and the shining Genji (the famous lovers so popular in Tokugawa mitate-e and the frequent source of parody for Saikaku) that Bashp looked back to but rather Npin, Saigyp, Spgi, and Chinese poets such as Du Fu and Li Bo (all perceived as recluse poets). The difference is perhaps most evident in Bashp’s treatment of Noh, or more precisely, the utaibon, the librettos for Noh, which, by the late seventeenth century, had became extremely popular, serving as a kind of new classic for Tokugawa commoners. In contrast to the typical Tokugawa mitate-e, which alluded to the more erotic and elegant aspects of the Noh, especially the “women” plays (kazura-mono), Bashp focused on the muted image of the waki, or the traveling priest, and on the fate of tragic heroes of the “warrior” plays (shura-mono). Bashp’s poetry and travel literature changed the make up of the “ancients” (kojin)—the major poetic predecessors—for subsequent generations: it was not the noted Heian poets, but rather Saigyp, Npin, and Spgi—as well as their Chinese recluse counterparts—who became the “great” poetic precursors. Bashp likewise drew heavily on the language and imagery of medieval poetry rather than on Heian aristocratic literature. He, in turn, was to be canonized, placed in a self-created literary genealogy that carried the medieval poet/traveler into the Tokugawa period. In the seventeenth century a sharp distinction was maintained between the forms of traditional genres—such as waka, renga, Chinese poetry, Noh drama—and the new, popular genres—such as kypka (comic waka), haikai, kanazpshi (vernacular fiction), jpruri (recitation to music), and kabuki. Comic genres such as kypka, kypshi (comic Chinese poetry), and kypbun (playful prose), which became extremely popular in the Tokugawa period, parodied their high counterparts by borrowing the elegant, aristocratic forms of ga literature— of waka, kanshi (Chinese poetry), and classical prose—and giving them zoku, especially erotic, content. The Kokin wakashujo (Preface to Young Men of Old and New), for example, composed in late sixteenth century, depicted the world of pederasty while parodying, almost word for word, the famous kana preface to the tenth century Kokin wakashu (Collection of Japanese Poems Old and New). Nise monogatari (Fake Tale), published 1704–1711, similarly parodies another classic, The Tales of Ise. Sometimes, the reverse phenomenon occurs: what initially appears to be low turns out to be “high.” Ga

112

H  S 

takes the form of zoku, a process referred to as yatsushi (literally, the “dressing down” of a person of high station), a phenomenon that extends beyond literature to the clothing and lifestyle of the chpnin (such as wearing colorful silk lining under plain cotton). Ishikawa Jun, a modern novelist and scholar, regards the popular folk tale of Otake, the lowly maid who turns out to be a contemporary incarnation of Dainichi Nyprai, as an archetypal example that finds its historical precedent in the Noh play Eguchi, in which the prostitute from Eguchi turns out to be the Fugen Bodhisattva. Not only does the profane turn out to be sacred, the high (in this case, a Buddhist deity that had lost some of its power in the Tokugawa period) takes on new life and meaning through the low form. This literature of reversal or double vision often inverted—no doubt to the delight of commoner audiences—the strict social hierarchy in which the samurai were the elite and the chpnin occupied the lowest rung. Perhaps the most famous example is the kabuki play Sukeroku, first performed in 1716, in which the dashing (otoko-date) young chpnin protagonist turns out to be Soga Gorp, the younger of the famous samurai Soga brothers (who died in 1193), in a contemporary Edo setting. In contrast to Yonosuke, the lascivious hero of Ihara Saikaku’s Kpshoku ichidai otoko (Life of a Man of Love, 1682), who, in Danrin haikai fashion, becomes a vulgarized, contemporary version of the refined, aristocratic hero of the Tale of Genji, the fictional “I” of Oku no hosomichi, the “beggar old man” and homeless traveler, is a yatsushi, dressed down embodiment of Zpga, Npin, Saigyp, and other high spiritual figures. Most of all, Bashp sought out the high in the commonplace, in seventeenth-century popular culture, endowing popular language and topics, particularly those drawn from provincial life, with the kind of poetic and spiritual depth hitherto found only in classical diction and poetry. As the following hokku in Bashp’s Oku no hosomichi suggests, the traveler finds furyu (translated here as “poetry” but also meaning art or culture in the widest sense) in the farm life and country villages of Michinoku (Deep North). Furyu no hajime ya oku no taueuta

The beginnings of poetry! the rice-planting songs of the Deep North

Bashp’s haikai frequently transformed and inverted the mainstream literary and linguistic conventions through what Henry Louis Gates (in reference to African American literary aesthetics) calls “repetition with a signal difference” or “intertextual double-voicedness.”3 The

D  V  B ¯ ’  H 

113

famous frog poem, which was written by Bashp in 1686, exemplifies this kind of ironic echo. furu ike ya kawazu tobikomu mizu no oto

an ancient pond— a frog leaps in, the sound of water.

An earlier version of the poem was kawazu tondaru, a colloquial phrase, which would have drawn attention to the zokugo or haikai language. The final version, however, contains neither haigon (haikai words) nor zoku content, and instead twists the traditional hon’i, or poetic associations, of the frog. From as early as the Man’ypshu (Collection of Ten Thousand Leaves, 759) the kawazu was admired for its singing, its beautiful voice, as in the following waka, which appears in a section on frogs in the Man’ypshu (vol. 10, no. 2165). kami tsu se ni kawazu tsuma yobu yu sareba koromode samumi tsuma makamu toka

On the upper rapids a frog calls for his love. Is it because, his sleeves chilled by the evening, he wants to share his pillow?

In the Heian period, the kawazu also became associated with the blossoms of the yamabuki (kerria), the bright yellow rose, and with limpid mountain streams, as in the following anonymous poem from Kokinshu (spring, no. 125). kawazu naku Ide no yamabuki chirinikeri hana no sakari ni awamashi mono o

At Ide, where the frogs cry, the yellow rose has already scattered. If only I had come when they were in full bloom!

According to one source, Kikaku, one of Bashp’s disciples, suggested that Bashp use yamabuki ya (“Bright yellow rose—”) in the opening ku (first five syllables), which would have left the poem inside the circle of traditional associations. Instead, Bashp worked in and against the traditional hon’i of the kawazu. In place of the plaintive voice of the frog singing in the rapids or calling out for his lover, Bashp gives the sound of the frog jumping into the water. And instead of the elegant image of a frog in a fresh mountain stream beneath the yamabuki, the hokku presents a stagnant pond. Haikai-ness in Bashp’s texts implied not only ironic echo but the discovery of new and startling perspectives (e.g., the jumping frog), which from the established, conventional point of view (the singing

114

H  S 

frog) appeared deviant, grotesque, or incongruous, and thus comic or humorous. This approach recalls the theory of defamiliarization put forward by the Russian Formalists, who argued that literary texts, rather than reflecting reality, tend to “make it strange,” to “defamiliarize,” to dislocate our habitual perceptions, to challenge dominant conceptions.4 Bashp’s haikai, while not political in the sense of attempting to alter the political or social structure, strove to make strange the system of literary and social commonplaces. Bashp achieved this in part through the creation of poetic personae, especially the recluse and the traveler, who bring an “outside” or liminal perspective on the world. Bashp also engaged in refamiliarization, which gave new life to poetic cliches through new contexts, languages, and perspectives. As a kigo (seasonal word) for spring, kawazu implied the emergence of the frog from its winter hibernation. Bashp’s hokku places this association in a new perspective: the sudden movement of the frog, which suggests the awakening of life in spring, stands in contrast to the implicit winter stillness of the old pond. In Semiotics of Poetry (1978) Michael Riffaterre argues that the poetic text is marked by indirections—such as displacement (metaphor, metonymy), distortion (ambiguity, hyperbole, contradiction, etc.), and the creation of textual space (symmetry, rhyme, etc.)—that threaten the first level of reading, that of the words as “literary representation of reality.” These incompatibilities cause the reader to move to the second level of reading, where he or she makes various “transcodings” from the first level and reads the words and phrases as parts of other networks or systems of signs.5 The lack of obvious signs of indirection combined with conventions of reading modern haiku as “snapshots” of a scene often cause readers of Bashp’s haikai to stop at the first level. In contrast to Danrin haikai, which delighted in mimetic distortions, in the fantastic and the absurd (referred to as gugen haikai), the keiki style, which characterized Bashp’s poetry from the late 1680s and which dominated all of Genroku haikai, stressed verisimilitude. As a consequence, there has been a strong tendency to read Bashp’s haikai on the mimetic level. Significantly, even after the informed reader moves to the second level, the first level is not denied or erased, as in Riffaterre’s model. Instead, the first level continues to coexist with the second, creating a double voice and allowing the reader to take pleasure in the disjunction or resonance between the two. Indeed, one of the innovations of Bashp’s frog poem was the creation of an external scene or landscape with subtle, internal overtones: the faint sound of water rising from an old pond implies a larger meditative silence, a loneliness, if not the attentive eye

D  V  B ¯ ’  H 

115

and ear of a recluse. The reader of the frog poem thus simultaneously enjoys this hokku as a descriptive, image-oriented poem about a frog and as a haikai, parodic challenge to a classical topos. A number of Bashp’s poems in his mature period recall the ancient rhetorical figure of syllepsis (Gr. “taking together”), which has been redefined by Jacques Derrida and Michael Riffaterre as a kind of pun, as a trope that consists in understanding the same word in different ways at the same time, one meaning being literal and primary, the other figurative. The second meaning is not just different from and incompatible with the first: it is tied to the first as its polar opposite the way the reverse of a coin is bound to its obverse.6

The haikai-ness of Bashp’s poetry and prose often derives from two levels of meaning—referential and encoded, literal and figurative, contextual and intertextual—which move in opposite directions at the same time. As Ogata Tsutomu has shown, to be a “traveler” in Bashp’s poetics meant literally to journey alone on distance and difficult roads and figuratively to join and enjoy the company of other noted poet–travelers such as Npin, Saigyp, and Spgi. In contrast to medieval writers, Bashp treats the poetic personae of the recluse and the traveler ironically, with a haikai-esque doubleness in which the “outsider” became an “insider,” the “low” the “high”. In his early Chinese-style hokku, Bashp parodies Chinese texts, lowering Du Fu to the level of a beggar at Fukagawa, moving temporally from the past to the present, and linguistically from Chinese to vernacular Japanese—all of which is implicitly a downward, popularizing movement. At the same time, however, the poem raises the low life at Fukagawa to the level of the renowned Chinese poet, enabling Bashp to “join” the company of Du Fu and other great Chinese poets. To be a “recluse” meant to live alone in a hermitage, as a self-imposed social outcast, and to be surrounded by other medieval and Chinese recluse poets. Travel and the grass hut were tropes—two-faced signs mediating between text and intertext—through which the poet joined other poets across time, in which haikai, the low popular genre, became a part of high literature. In a sylleptic twist, being outside meant being “inside,” being “poor” (materially) meant being “rich” (spiritually or poetically). Wabi (physical and psychological suffering), paradoxically, implied literary or aesthetic pleasure, as did the medieval aesthetics of the somber, the lean (yase), and the cold (hie). In medieval poetry, shigure, cold intermittent winter showers, was a dark image associated with impermanence and the

116

H  S 

uncertainty of life. In Bashp’s literature, however, the haikai poet deliberately seeks out shigure, which gives him intertextual pleasure and a sense of communion with those ancient poets who wrote about shigure. The poet of fukyp (“madness in poetry”), a poetic ideal developed by Bashp in the mid-1680s, lives and acts in the world of poetic signs, on the level of the intertext, thereby appearing insane (kyp) to the ordinary person, who sees those actions only on the level of everyday, object-oriented, referential discourse and for whom winter rain is only to be avoided. The haikai-ness of Bashp’s shigure, which became a symbol of the Bashp style, lay in this double voice, in the implied pleasure in the unpleasant. Haikai humor originally derived from working against established values, conventions, commonplaces, from undercutting images of power and authority. Muromachi haikai parodied Buddhism and its sacred icons as well as those aesthetic values associated with the classical tradition and aristocratic society. At the same time, as the new literature of the urban commoners and samurai, haikai increasingly sought to assume the functions of more high literature, becoming a transmitter of moral and spiritual values (such as makoto) and in some senses replacing the role of religion, especially Buddhism, which declined in the Tokugawa period. One consequence was that, particularly from the Genroku period (1688–1704) onward, one stream of haikai tended to gravitate toward those very elements—aesthetic and religious values—that it had earlier mocked and playfully degraded, while another stream (which became senryu) continued to move in the direction of comic, satirical verse. Bashp, who stood at this crossroads in haikai history, took on the difficult task of creating poetry that had profound spiritual and aesthetic implications even as it retained its anticonventional character and embraced popular language and culture. The result was a literature that was double-voiced and profoundly paradoxical in nature—iconoclastic yet traditional, humorous yet sorrowful, sacred yet profane, momentary yet lasting—a tension embodied in Bashp’s notion of fueki ryukp, the “unchanging and the ever-changing.” Bashp eventually went beyond the ga/zoku dialectic to achieve a new kind of haikai-ness that did not depend on the inclusion of haigon or popular content, as earlier haikai had, but instead arose out of a haikai-esque attitude that worked against established values, conventions, and perceptions, including those created by haikai itself. Parody has traditionally been defined as the imitation of a distinctive style and thought of a literary text, author, or tradition for comic effect. The success of this kind of parody—exemplified by Spkan’s haikai sequence in the opening of Inu Tsukuba shu—depends on a

D  V  B ¯ ’  H 

117

recognition of the target text and the negative reversal of the values embedded in it. Bashp’s mature haikai, by contrast, is closer to a type of resonant irony in which the echoing of the target text also functions as a form of homage, as a ritual and sometimes filial greeting (aisatsu) to the poetic ancestor. Linda Hutcheon has redefined parody along these lines as a form of “imitation characterized by ironic inversion, not always at the expense of the parodied text.” this irony can be playful as well as belittling; it can be critically constructive as well as destructive. The pleasure of parody’s irony comes not from humor in particular but from the degree of engagement of the reader in the intertextual “bouncing” . . . 7

Hutcheon gives the example of the relationship of James Joyce’s Ulysses to Homer’s Odyssey or T.S. Eliot’s “The Wasteland” to Western classical texts.8 Ulysses follows the Homeric model on the level of character and plot, but it does so with an ironic difference. Instead of being mocked, the classical model becomes the norm from which the text departs. Parody redefined in this fashion can range from the disdainful laugh (found in satire) to the knowing smile (of irony) to the respectful nod or oblique homage, which is often found in allusion and is frequently a characteristic of Bashp’s literature. Hutcheon has also noted that “parody is normative in its identification with the Other, but it is contesting in its Oedipal need to distinguish itself from the prior Other.”9 Npin, Saigyp, and others became Bashp’s poetic fathers, to whom he paid homage, but he also broke from them, deliberately inverting and even erasing the traces of his poetic predecessors, revealing what Harold Bloom has called “the anxiety of influence.”10 Not to have done so would in fact have meant ceasing to be a haikai poet. This double movement, of homage and displacement, of appropriation and transformation, took the form, especially in his travel literature, of refiguring the landscape, particularly utamakura, the classical toponymies. For Bashp, as for most linked verse poets in the premodern period, the tradition or poetic canon was conceived not so much as a specified body of texts—specifically Kokinshu, Ise monogatari, and Genji monogatari, which had become required reading for all waka poets from the time of Fujiwara Shunzei and Fujiwara Teika (1162–1241)—but rather as a highly encoded body of hon’i, or poetic associations, which existed for almost all aspects of nature and for such topics as love and travel. In the medieval period, Kokinshu, Ise monogatari, and the Genji monogatari became the canon, not because they were considered

118

H  S 

great books of philosophy, history, religion, or narrative literature, but because they were fountainheads of the hon’i that the waka poet must compose on and were considered indispensable for the cultivation of poetic sensibility. The same became even more true of orthodox renga poets, who depended on a knowledge of the hon’i to link verses. Haikai poets also inherited the classical hon’i, but they used it for different purposes. Muromachi poets such as Spkan inverted or vulgarized the traditional hon’i, while Genroku poets such as Bashp broadened and reconstructed the canon, approaching the classical hon’i from new perspectives or with new languages. Since the topics of poetry concentrated on the phases of the four seasons, the poetic canon could be said to reside in “nature,” and any reconstruction of tradition meant no less than the re-vision or rewriting of nature. Each seasonal word (kigo), which became a requirement of the hokku, or opening verse, had a hon’i that tied it to the larger poetic tradition. (In the Muromachi and Edo periods, these associations were made readily available in renga and haikai handbooks, which were categorized by the seasons.) The hokku, now called haiku, is often considered the shortest form of poetry in the world, but each hokku, via the kigo, is actually part of a larger, complex text, a gigantic seasonal wheel of poetic associations, which existed in the collective memory of both the poet and his or her audience. For Bashp, this tradition was not only alive but highly changeable, subject to haikai twists and transformations, especially as a result of one’s own experience or close observation of nature. In “On Some Motifs in Baudelaire” (1939), Walter Benjamin (1892–1940) makes a distinction between two types of experience, integrated (Erfahrungen) and atomistic (Erlebnisse), that resonate with Bashp’s notion of tradition. Erfahrungen integrates tradition into experience, fusing collective and personal experience, particularly through ritual. Experience is indeed a matter of tradition, in collective experience as well as private life. It is less the product of facts firmly anchored in memory than of a convergence in memory of accumulated and frequently unconscious data.11

Erlebnisse, by contrast, was experience as a series of atomized, fragmented, unarticulated passing moments merely lived through. Both modern haiku and Bashp’s haikai gravitate toward the passing moment, the atomized Erlebnisse, the “here and now,” but Bashp’s haikai was simultaneously deeply rooted in collective experience, in Erfahrungen. The collective poetic memory of the frog—its association

D  V  B ¯ ’  H 

119

with spring, new life, singing—was as critical as the “now,” the frog leaping into the pond. Bashp’s haikai embodies both collective experience, in which past contexts became part of the context for subsequent reception, and immediate, fragmentary experience focused on the present moment. The fueki (“the unchanging”) in Bashp’s notion of fueki ryukp implied a sense of poetic and cultural history, which was concentrated in the classical hon’i, while ryukp (“the ever-changing”) implied a constant need for “newness” (atarashimi)—new languages, approaches, styles, and hon’i. The “unchanging” was a living entity that required constant revision and revitalization. Indeed, Bashp preferred the word honjp, which suggests a living essence or spirit, to hon’i, which implies a fixed, preestablished essence. Bashp urges his disciples to study the ancients not, as in classical renga, in order to follow the traditional hon’i, the classical associations of a topic, but to give the hon’i new life, to find new hon’i, to expand the borders of the hon’i, to express the old hon’i in new ways. Since the tradition did not exist so much in the form of texts and canonical works (though it may have included them) as in the collective memory of the hon’i, the objects of nature in Bashp’s worldview came to bear what Benjamin referred to as “aura,” which results when an object exists in the fabric or web of history. In Benjamin’s words, such objects “look back” at the viewer. The experience of the aura thus rests on the transposition of a response common in human relationships between the inanimate or natural object and man. The person we look at, or who feels he is being looked at, looks at us in turn. To perceive the aura of an object we look at means to invest it with the ability to look at us in return. This experience corresponds to the date of the “memoire involontaire”. . . 12

The traces of the past inscribed on the object or the landscape transform the relation between the poet and nature into one similar to that between human individuals, a dialogic relationship in which the landscape returns the viewer’s gaze, responds with its own voice. Indeed, in Bashp’s Oku no hosomichi (Journey to the Deep North) travel becomes a means of engaging in dialogue with the spirits of the ancients via the medium of place. The meisho (famous places) and the utamakura (poetic toponymies) in this travel literature possess Benjamin’s aura. The “I” reexperiences and renews the collective memory of the object or place even as he experiences the physical place anew, as a present moment. Bashp also endowed places that had no aura, that had not appeared in classical poetry, with a “face” or a “voice” that subsequent poets could gaze upon and listen to. For him

120

H  S

the journey became a means of recreating the landscape, of giving it new life and new voices, as he did with other objects of nature. The double voice in Bashp’s haikai as well as his poetics of double vision, in which the poet simultaneously sees, often in mitate (“to see one thing as another”) or shakkei (“borrowed landscape”) form of two landscapes, is directly related to the nature of haikai linked verse, in which each verse is read as double, in relationship to the text that both precedes and follows it. Linked verse opens with a hokku, a seventeensyllable (5/7/5) verse, to which is added a wakiku, or fourteen-syllable (7/7) verse, which is capped in turn by a seventeen-syllable (5/7/5) daisanku, and so forth, until a sequence of thirty-six, fifty, or one hundred links was completed. Each new verse (tsukeku) is linked to the previous verse (maeku) to form a new poetic world while breaking away from the verse prior to the previous one (uchikoshi). With the exception of the hokku, or the seventeen-syllable opening verse, which is read independently, each new verse is bound to the previous verse while standing in a negative relationship to the uchikoshi from which it pushes off. The manner in which the tsukeku (added verse) inverts, displaces, or otherwise twists, usually in humorous fashion, the world of the maeku (previous verse) and uchikoshi (verse prior to the maeku) has a direct parallel to the manner in which haikai works against classical texts and conventions. If parody is fundamentally intertextual, the ironic bouncing between the text and the intertext, the relationship of the tsukeku and the maeku to that of the maeku and the uchikoshi may be called intratextual: both involve a negative movement, a pushing off, an ironic recontextualization, so as to create new meanings and perspectives. A good example is the following threeverse sequence (no. 22, 23, 24) in the Ko no moto (Beneath the Cherry Trees) kasen (thirty-six verse sequence) in Hisago (Gourd, 1690), a Bashp-school haikai anthology edited by Shadp (d. 1737). Kumano mitaki to nakitamaikeri

“I want to see Kumamo,” she wept. Bashp

tatsukayumi Ki no sekimori ga katakuna ni

bow in hand the barrier guard at Ki stood unyielding Chinseki

sake de hagetaru atama naruran

the bald head no doubt from too much drinking Kyokusui

D  V  B ¯ ’  H 

121

In the first verse, in refined classical Japanese, the speaker (presumably an aristocratic lady traveler) is weeping because she is anxious to visit Kumano (in Ki Province), a popular site for religious pilgrimage in the Heian and medieval periods. The next verse combines with the previous verse to create a scene in which the traveler is weeping because the fierce-looking guard at the barrier is refusing to let her pass into Ki Province. The third verse, which combines with the second verse to create a new scene, humorously transforms the barrier guard into a tippler, whose head has grown bald, or so it appears, from excessive drinking. Orthodox linked verse likewise recontextualized or “misread” the previous verse, but in contrast to orthodox renga, which maintained a harmonious tone and preserved classical content and diction, haikai moved radically from one social world or language to another. In this case, the aristocratic, classical world and language of the first two verses is unexpectedly transformed into an everyday, commoner scene. The pleasure of haikai derived as much from the new combinations, the creation of new worlds, as from the process of decontextualization, the negative movement, the pushing off from the intratext or the intertext. As we can see here, haikai can only be understood in ironic relationship to other texts, codes, or systems, and in this regard it is always double voiced. The relationship between the added verse and the previous verse is simultaneously one of fusion and juxtaposition, of combination and equivalence, which allows haikai to both bridge and maintain the difference between classical diction and other languages. Sigmund Freud notes in Jokes and Their Relationship to the Unconscious that humor can derive from the discovery of “similarity between dissimilar things.” Jean Paul (Richter) has expressed this thought itself in joking form: “Joking is the disguised priest who weds every couple.” Vischer carries this further: “He likes best to wed couples whose union their relatives frown upon.”13

In contrast to Danrin haikai, which tended to emphasize the difference between ga and zoku, to bring the two into violent collision, Bashp sought to find continuity in discontinuity, to fuse ga and zoku, contemporary and classical, to find hidden similarities. A good example is the following hokku, which Bashp composed in 1690. kogarashi ya hohobare itamu hito no kao

wintry gusts— a person’s face pained by swollen cheeks

122

H  S 

The two parts of the hokku—separated by the cutting word ya—can be read both together, as one continuous, combined scene, or separately, discontinuously, as two parts that reverberate against each other. In the former, a person suffering from “swollen cheeks” or mumps (hohobare) stands outside buffeted by winter gusts (kogarashi), the piercing winter winds that blow the leaves off the trees. In the latter, the image of a person’s face hurting from mumps echoes that of kogarashi, of biting cold and desolation. The hokku has the kind of peripeteia found in linked verse, in which the expectation or context established by the previous verse is reversed or radically altered by the added verse. Here, the expectation generated by kogarashi, a classical seasonal word, is humorously upset by the low haigon phrase hohobare itamu—illness is never mentioned in classical poetry—which then leads to a double reversal: after the initial shock caused by the collision of ga and zoku, we discover, in the fashion of Bashp’s “link by scent” (nioi-zuke), a connotative equivalence or harmony between the winter gusts and the swollen cheeks, the marriage of Jean Paul Richter’s “odd couple.” Sometimes, the relationship between the two parts of the hokku is not that of similarity but of antithesis or contrast. In the famous frog poem, the two parts create a scene in which the frog leaps into an old pond followed by the sound of water. At the same time, however, cutting word ya sends the reader moving back and forth between the two parts in a circular motion. The result is a montage effect, an implicit contrast between the quiet world of the old pond (furuike), with its connotations of winter, hibernation, stillness, and the sudden movement of the frog, with its implications of spring, vitality, rebirth, motion. The poem, like much of haikai, occupies a double zone between a scenic or narrative representation on the one hand and a fragmentary collage on the other. The process of representing an imagined scene is not cut off but coexists with a spatial (almost cubistic) arrangement. Haikai is profoundly dialogic, not only in the manner in which different languages and voices interact and the reader bounces between texts, but in the way in which it emerges out of an intimate dialogue with other contemporary poets, in the nature of its “addressivity.” Mikhail Bakhtin has stressed three dimensions or coordinates of “dialogue”: the writing subject, the addressee, and exterior texts. On a horizontal axis, the word is oriented toward a specific addressee in a subject/addressee relationship; and on a vertical axis, the same word or text is oriented toward an anterior or synchronic literary corpus in an intertextual (text/context) relationship. Haikai, as both a communal activity and as a parodic or ironic echoing, constantly

D  V  B ¯ ’  H 

123

works at the intersection of these two axes. In contrast to classical renga poets such as Spgi or earlier haikai poets such as Teitoku, Spin, and Saikaku, who frequently composed dokugin, or solo linked verse, Bashp always composed haikai in the form of the za, as communal literature in which the participants (renju) gathered together to link verses, usually at the invitation of a host and under the guidance of a haikai master. The hokku, the opening verse, which was usually composed by the main guest, served as a greeting, a compliment, or expression of gratitude to the host, who in turn composed the wakiku, or second verse. More than half of Bashp’s hokku are in fact aisatsu (greetings) that have either a social or religious function. shiragiku no me ni tatete miru chiri mo nashi

white chrysanthemums right before my eyes: not a speck of dust

This hokku, which Bashp composed at the residence of Sonome, one of his disciples, is typical of his mature style in that the poem is in the keiki style, describing the scene before the poet’s eyes while at the same time functioning on the connotative level as an aisatsu, implicitly praising the character of the hostess. (The honi of shiragiku, or white chrysanthemums, was elegance and refinement.) The art of the address here lies in the double voice, the ability of the poem to exist simultaneously on the literal and figural, referential and symbolic levels, as both monologue and dialogue. Bashp’s hokku are not only aisatsu, or greetings, they implicitly ask for a response to the same topic. Yamamoto Kenkichi, who was a modern pioneer in studying the dialogic aspect of haikai, has even argued that each hokku implicitly ends with the phrase, “Isn’t it?” As Ogata Tsutomu points out, the famous frog poem is not only a poem about a frog, it is an invitation to Bashp’s haikai partners, suggesting something like: “The frog has always been regarded as a creature that sings, especially in fresh streams, in the middle of spring, but I want to look at the frog differently. Wouldn’t you be interested in doing this together?” The implicit quietness and loneliness of Bashp’s frog poem are, in a haikai twist, to be shared with others. Kikaku responded to Bashp’s hokku with the following wakiku: ashi no wakaba ni kakaru kumo no i

on the young reed leaves hangs a spider’s nest14

In a manner typical of Bashp-style haikai, the added verse is linked to the previous verse both by scenic extension (the edge of the water) and by implied equivalence: the connotations of the wakiku—the

124

H  S 

sense of spring implied in the new leaves (wakaba) on the swamp reeds (ashi) and the aftermath of winter suggested by the spider’s nest (kumo no i)—directly echo the contrast between the spring (frog) and winter (old pond) in Bashp’s hokku. Other disciples “answered” Bashp’s hokku by composing their own frog poems, which were collected, matched, and judged in “Kawazu awase” (“Frog Competition,” published 1686, edited by Senka), which opens with: First round: Left side furuike ya kawazu tobikomu mizu no oto

an ancient pond— a frog leaps in, the sound of water

Right side itaike ni kawazu tsukubau ukiha kana

tenderly, a frog crouches on a floating leaf! Senka

Like Kikaku’s wakiku, Senka’s hokku extends the scene of Bashp’s hokku: the frog, which in Bashp’s poem had jumped in the old pond, has swum to a floating leaf on which it now crouches. The notion of addressivity here extends beyond the horizontal za relationship between the subject and the immediate addressee toward a vertical axis in which the text speaks across time to the ancients. Yamamoto Kenkichi has argued that the hokku is both a statement and a question, monologue and dialogue, that Bashp, of all poets, developed this double voice to the fullest and that the pleasure of reading often lies in listening to both voices intermingle. This process, in which composition of a poem is directly answered or parodied by another poem even as it stands on its own, derives directly from the process of linked verse in which the participants constantly alternate between being poet and audience, addresser and addressee. Bashp’s haikai, in short, is acutely informed by what Bakhtin calls the “already spoken,” by previous utterances about the topic, as well as by the “not-yet-spoken.” In Problems of Dostoyevsky’s Poetics, Bakhtin describes three kinds of creative processes: the polyphonic creative process in which the polyphonic writer sets in motion a multivoiced, open-ended dialogue marked by surprising and “unfinalizable encounters”; the “Romantic” (inspirational) model, which sees the creative process as a sudden burst of imagination in the mind of the individual; and the “classical” (formalist) model, in which the author follows a plan and completes

D  V  B ¯ ’  H 

125

the work like a mathematician solving a problem. Bashp engaged in the polyphonic process: the journey, with its unexpected encounters and engagement with other voices, past and present, with other languages and subcultures, became the overriding metaphor for his poetic process. In the last decade-and-a-half of his life, after initially establishing himself in Edo, Bashp traveled from group to group, place to place, mainly in the provinces, constantly seeking out new partners. He was like a jazz musician who gathers new musicians around him or joins other groups as a guest performer in order to create new music. Bashp’s disciples were not just students who followed their master but poets whose voices had a profound influence on his poetry. Bashp himself had a tendency to move toward the past, toward classical and medieval literature, to join the company of the ancients, but his socially and geographically diverse haikai partners—ranging from high samurai, merchants, artisans, priests, to rich farmers— simultaneously anchored him in the material culture and variegated languages of the late seventeenth century. Inversely, Bashp’s clients and students, rooted as they were in everyday Edo society, absorbed the high tradition through the conduit of haikai and their learned haikai master. Whenever his poetry began to stagnate, Bashp departed on a journey, engaging in poetic dialogue with other poets, often unknown or still young, in different parts of the country. For example, the publication of Fuyu no hi (Winter Days), his first major haikai anthology in the Bashp style, was a direct result of the journey, commemorated in Nozarashi kikp, to Owari Province and Nagoya. Sarumino (Monkey’s Straw Coat, published in 1691), considered to be the Kokinshu of haikai, grew out of his journey through the Deep North (Michinoku) and his subsequent encounter with poets in Kyoto and the Konan area (south of Lake Biwa). His return to Edo resulted in the karumi (light) style found in Sumidawara (Charcoal Sack, 1694), the last of the major Bashp-style haikai anthologies. By working with different partners, editors, and social groups—the machishu or affluent merchants in Nagoya, the samurai in Konan, the shop merchants in Edo—in diverse places (Nagoya for Fuyu no hi, Kyoto for Sarumino, and Edo for Sumidawara), Bashp developed strikingly different types of poetry in a relatively short span of time. In a relentless drive to seek new poetic ground, Bashp either abandoned or lost, one after another, disciples who had played a major role at one stage but were unable to contribute or participate in his latest movement. Bashp traveled simultaneously in two fundamental directions: on one hand, he journeyed vertically into the past, seeking out the traces of the ancients,

126

H  S 

revitalizing and recasting the tradition. At the same time, he moved horizontally from disciple to disciple, style to style, in constant pursuit of new languages and perspectives, without which the life of the tradition would be lost. Bashp once described haikai as “thirty-six steps forward, no steps backward.” The added verse must push off the penultimate verse to create a new world; it can not return to earlier worlds. The same was true of Bashp’s poetic career as it was of his haikai.

N 1. Mikhail Bakhtin, trans., The Dialogic Imagination, Caryl Emerson and Michael Holquist (Austin: University of Texas, 1981); and Rabelais and His World, Helene Iswolsky, trans. (Bloomington, IN: Indiana University Press; reprint, 1984). 2. In Donald Jenkins. The Floating World Revisited (Portland, OR: Portland Art Museum, 1993), 87. 3. Henry Louis Gates, The Signifying Monkey: A Theory of African-American Literary Criticism (New York: Oxford University Press, 1988). 4. Tony Bennett, Formalism and Marxism (London and New York: Methuen, 1979), 42. 5. Michael Riffaterre, Semiotics of Poetry (Bloomington, IN: Indiana University Press, 1978), 2. 6. Michael Riffaterre, “Syllepsis,” Critical Inquiry 6, no. 4 (summer 1980), 629. 7. Linda Hutcheon, A Theory of Parody (New York and London: Methuen, 1985), 6. 8. Ibid., 32. 9. Ibid., 77. 10. Ibid. Harold Bloom, The Anxiety of Influence (London and New York: Oxford University Press, 1973). 11. Walter Benjamin, “On Some Motifs in Baudelaire,” in (ed.) Walter Benjamin; (trans.) Harry Zohn, Illuminations (New York: Schocken Books, 1968), 157. 12. Benjamin, Illuminations, 188. 13. Sigmund Freud, Jokes and Their Relationship to the Unconscious, translated and edited by James Strachey (New York: Norton, 1960), 11. 14. Ashi no wakaba (young leaves of the reed) is a kigo for spring.



L   L : C I  L V   C I. Leopold Hanami

narenu sumai zo sabishisa mo uki

uki ware o sabishigarase yo kankodori

in an unfamiliar dwelling even loneliness is miserable —Minase sangin, Spgi, 1488.1.22 I who am miserable— make me feel lonely, cuckoo Saga nikki, Bashp, 1691.4.22

For the contemporary reader familiar with both Minase sangin and

Saga nikki, Spgi’s verse brings to mind Bashp’s entry in his brief Saga diary. Both express the inexorable association between loneliness and misery. They also manifest the value of loneliness in Japanese aesthetics: Spgi hopes to find tranquility in solitude, but in an unfamiliar place “even loneliness is miserable.” Bashp, for his part, is miserable but hopes to find serenity in the lonely but aesthetically pleasing “kanko” cry of the cuckoo. However, under conventional circumstances, an association between these two verses—separated by more than two hundred years—might not be acceptable, for Spgi’s verse is only a single line from a renga, a one-hundred verse linked poem. A traditional approach requires the reader to consider any meaning the verse may offer only in the context within which it finds itself, forever linking it to its neighboring verse and limiting its possibilities in the greater context of Japanese literature.

I . L  H

128

This is a paradox. Renga, even in its linear diachronic format, compels the reader to recognize the reality that any one verse can have different meanings in a synchronic moment. An initial perusal of a renga sequence reveals a stream of individual stanzas, each linked to its previous or subsequent stanza through an association seemingly unrelated to other stanzas. This association between individual stanzas is regarded conventionally as a single poetic unit similar in structure to the classical poetic form of waka. Indeed, much of the conventional analysis of renga and haikai (the linked verse popularized by Bashp) focuses on this unity. Howard Hibbett insists that “no more than two links may be bound together.”1 Robert Brower and Earl Miner see the composition of renga as “units of ever-shifting sets of two stanzas.”2 In his own book on renga, Miner consistently refers to the two associated stanzas in renga as “units.”3 In fact, his translations of renga and haikai sequences show his strong inclination to see the successive stanzas as units: one stanza A is united with the next one B to create one poetic unit AB; the second stanza B is then paired with the next one in the sequence C creating another new unit BC. This continues throughout the sequence until the translation reaches the last stanza. These are represented as a poetic unit of five lines, the traditional transliteration of waka. Verse A First poetic unit

Verse B Verse C

Second poetic unit

This, however, does not seem to address a characteristic peculiar to linked verse—a change in direction, a shift in perspective, a destabilization of meaning. An analysis of renga and haikai should also address these features, not by looking at the association between two consecutive stanzas, but rather by examining one stanza in relation to any stanza within the poem and, ultimately, beyond it. Such an analysis will allow greater flexibility in “reading” relationships beyond the poetic unit of individual linked verses. Focusing on the first ten sequences of Minase sangin,4 I will make observations on the role of interpreting intent—demonstrating the impracticality of ignoring common sense when strictly following renga rules—and reveal a continuity beyond two stanzas that is usually denied linked verse. I will then consider the first twelve verses in a haikai sequence by Bashp, Bonchp, and Kyorai—Ichinaka wa from Sarumino—which will not

L   L 

129

only illustrate the continuity of meaning beyond two adjoining verses, but suggest the poets’ disregard for the prescriptive rules that modern scholars still apply.

T R A conventional approach to renga begins with a discussion of its sequential composition and the way it moves in different linear directions. The first of usually three poets composes the hokku (first stanza) in the 5/7/5 syllable format resembling a kami no ku (upper verse of a waka), the next poet composes a stanza in the 7/7 format of the shimo no ku (lower verse), thereby completing a poetic unit that resembles a waka poem. Alternating between 5/7/5 and 7/7 syllable stanzas, each poet composes in turn until they have composed one hundred stanzas altogether. The object of each succeeding sequence, however, is not to continue in a fashion that lends coherence over more than two stanzas, but rather to transform the meaning of the one stanza immediately before it, thereby causing a shift in direction. The result is a poetic form that rejects most traditional ideas of stable images and meanings. The first three stanzas in Minase sangin illustrate this. 1. yuki nagara yamamoto kasumu yube kana

as the snow remains as is the foot of the hills grow hazy— such an evening! Spgi (1421–1502)

2. yuku mizu tpku ume niou sato

the flowing water is distant and a village fragrant with plum blossoms Shphaku (1443–1527)

3. kawakaze ni hito mura yanagi haru miete

in a river breeze a cluster of willows spring appears Spchp (1448–1532)

A cursory look at these verses illustrates a subtle shift in the images. The first stanza describes the cold first days of spring; according to convention, the term kasumu (to become hazy) automatically identifies the season as spring, while the snow indicates that spring has begun in calendric terms only. The second stanza introduces flowing water and plum blossoms, indications of melting snow and early spring, respectively. Since the stanza should relate with the previous one, the image of the beginning of spring in the first stanza is

130

I . L  H

transformed to one of early spring. The third stanza introduces the idea that spring is easily visible in the green leaves rustling in the wind—an image that suggests that spring is well on its way. This three-verse sequence represents a change, a shift, a move in direction that disrupts the stability of images. Of course, a logical temporal movement is discernable and furnishes it with a kind of unity, and this is in step with the rules of renga: a sequence on spring— as well as autumn and love—must continue for a minimum of three stanzas. However, the poet is free to associate any aspect of spring with the previous stanza, thus fulfilling the requirements of change. What is of interest is the destabilization of an image due to this change. The plum blossoms in the second stanza are a traditional image of early spring, and so link neatly with the first verse. But in conjunction with the third verse, the plums are in bloom unseasonably late, for nearby the willows blow in the breeze, an indication that mid-spring has arrived. The destabilization of the image prevents in a traditional sense a cohesive unity beyond two successive links. This is perhaps the most logical reason why conventional analysis of renga focuses on one pair of stanzas, the “poetic unit,” at a time. Historically, renga commentary approached each poetic unit as a waka. One likely reason for this is its resemblance to classical court poetry. Besides the 5/7/5/7/7 syllabic structure, words and phrases were borrowed directly from waka diction. Nijp Yoshimoto (1320–1388) articulated this quite clearly in Chirenshp: “Originally, renga relied on [waka] poetry; one must understand earlier waka, distinguish its diction, and apply it in practice.”5 In the Renri hishp, he is even blunter: “Renga is a miscellaneous style of [waka] poetry” (ibid). According to Saitp Yoshimitsu, the idea that the poetic units of renga are essentially the same as waka was probably born from these statements (p. 42). While Saitp goes on to refute this approach as somewhat outdated—referring to the writings of Yoshimoto and Spgi explaining the importance of diction unacceptable in waka and other distinguishing points (pp. 42–46)—other scholars, among them Katp Shuichi, persist in viewing renga as a string of waka. “In renga different poets took turns composing the first and second halves of a single waka poem (emphasis added).”6 Most renga analyses focuses on the sequential linear movement in a renga sequence, the technique of association and progression suggested by Jin’ichi Konishi.7 In reference to Yoshimoto’s words on the manifestations of impermanence and the fluctuations of life in renga, Makoto Ueda states that the renga sequence “is usually thought of as one special type of mood poem, and [the reader] enjoys

L   L 

131

the transition of the continually changing mood at each stanza.”8 He implies that the enjoyment of renga is based on the “changing moods” in a succession of transitions between two stanzas, a progression moving in a sequentially forward direction. While not referring to the poetic unit as waka, his comments would lead to an analysis that viewed any two consecutive stanzas as a single poetic unit. This approach of examining association and progression at each coupling finds its roots in the classics, as illustrated in the classical commentary accompanying the Minase sangin.9 4. fune sasu oto mo shiruki akegata

the splashes of the boat’s pole are distinct at the break of dawn Spgi

5. tsuki ya nao kiri wataru yo ni nokoruran

does the moon linger on a night of drifting fog? Shphaku

[Verse 5 in conjunction with 4] means that, although day breaks, the moon lingers as if it was still night due to the darkness caused by the mist. This stanza depicts a scene of a moon on an autumn evening. 6. shimo oku nohara aki wa kurekeri

over fields covered with frost autumn has come to a close Spchp

Although the moon is something that always lingers, in this instance, [the poet] has taken it as a late autumn moon and has attached [his verse] accordingly. It means that, the moon lingers in night where the fog drifts, and the frost-covered field is the end of autumn. The mist and frost contrast. 7. naku mushi no kokoro to mo naku kusa karete

without a thought for the feelings of the crying insect, the grass withers Spgi

It means that, the grass withers and frost covers it, but because the insect hates these, it is as if they give no thought to [the insect’s] feelings.

This commentary treats two linked verses as a poetic unit if not an actual waka poem, but implicitly recognizes a shift with each successive stanza as demonstrated by the phrase “it means that.”10 This indicates that the commentator was attempting to distinguish a different meaning

132

I . L  H

with the linking of a subsequent sequence, that is, “it meant one thing in the previous link but now it means something else.” The fact that the commentaries of stanza 1 and 2 do not include this phrase further indicates the recognition of shifting meanings, for a shift in meaning does not take place in the first two stanzas; hence, the commentator does not have to make this distinction. Miner uses this approach as well. His technique is more sophisticated, stressing the characteristics of response and sequence: the poet is inspired by the previous stanza and then creates his own. However, much like the classical commentary, a sequence is associated only with its subsequent element, a process linear and diachronic. While it seems to allow for the displacement of meaning, it denies the simultaneous coexistence of different meanings. And yet, it is significant, for it reveals a crucial aspect in the composition of the renga sequence: intention.

I: A I  R To enable me to estimate a beauty of nature, as such, I do not need to be previously possessed of a concept of what sort of a thing the object is intended to be . . . the mere form pleases on its own account. If, however, the object is presented as a product of art, and is as such to be declared beautiful, then, seeing that art always presupposes an end in the cause (and its causality), a concept of what the thing is intended to be must first of all be laid at its basis.11

If Immanuel Kant were alive today, he may hold this opinion a little closer to the vest. In a post-structuralist, postmodern environment, an understanding of the intentions of a writer/composer certainly holds less value than the interpretation of a reader. Indeed, only a handful would argue against the flux of meanings and the merits of intertextual readings. And yet, renga composition suggests that understanding a poem’s intentions is crucial when composing one’s own verse. Of course, this is not to suggest that our analysis of renga should seek the intentions of a poet, thereby deeming to interpret, explicate, and elucidate his intentions. Rather, as Kant suggests in the above quote, in examining a poem, we should realize that intention is inherent in any composition; that in creating a poetic effect, there must be certain elements in the poem that manifest what its ends might be, that is, its intent. Note that there is a difference between the intentions of a poet and the intent of a poem. The distinction can be confusing. E.D. Hirsch, in Validity in Interpretation (1967) endorses an understanding of authorial intent. He bases this position on his belief that the meaning

L   L 

133

of a text cannot change. “[I]f any theory of semantic mutability were true, it would legitimately banish the author’s meaning as a normative principle in interpretation, for if textual meaning could change in any respect there could be no principle for distinguishing a valid interpretation from a false one.”12 A few pages later, Hirsch himself banishes the author and makes a statement that easily could be construed as a contradiction in terms: “Meaning is that which is represented by the text; it is what the author meant by his use of a particular sign sequence; it is what the signs represent.” (p. 8). If meaning is represented by the text, it stands to reason that if the context within which a text is situated changes, then the meaning would also change—as it does in renga—regardless of the author’s particular sign sequence. W.K. Wimsatt and Monroe Beardsley, fathers of the “intentional fallacy” concept, find authorial intent completely irrelevant.13 As modern critics, they indicate that the poem is an entity unto itself, free from the subjective influences of man: “The poem is not the critic’s own and not the author’s. . . . The poem belongs to the public. It is embodied in language, the peculiar possession of the public, and it is about the human being, an object of public knowledge” (p. 5). For them, grasping author intent is futile; it is a design in the author’s mind, and an attempt to understand intent is to believe that it is something that can be “transferred” from one person to another through the medium of words. Unfortunately, by divorcing the text from the author, they objectify it as something sterile and uninfluenced by outside sources, which would suggest that the role of the reader is diminished as well. This does not bode well for renga, for we cannot ignore the fact that a poet also played the role of reader when composing. He had to try—successfully or not—to interpret the intent of the previous poem, for if he did not, then he might compose a poem that would not alter the direction of the sequence. In other words, in trying to create a verse independent in meaning, a poet is ironically dependent on the poetic intentions of his predecessor, despite the following comment by Konishi. [T]here is no consultation between poets on the meaning or intention of their verses. When a poet links a verse to the verse before, he has nothing to go on but his own immediate understanding [emphasis added] of the preceding verse. It is thus inevitable that difference of interpretation should arise and unpredictable shifts of direction occur.14

Konishi contends that shifts in direction are caused by different interpretations of a particular stanza, and this is because the poet cannot

134

I . L  H

recognize the intention of the previous poet. The fallacy of this statement is immediately clear. There is no consultation between poets, perhaps, but that does not mean that he is incapable of grasping the meaning or intention of the previous stanza. On a physical level, a renga sequence, as H.M. Horton describes in detail, is an oral and performative process.15 Each poet is close enough to the designated scribe so that each recited verse can be heard and jotted down. As a result, each poet heard everyone else’s composition. Further, a renga master (as well as the scribe) had to have an excellent memory, for the success of a sequence depended on his ability to recognize immediately the overall consistency of tone and development of an entire sequence at a moment’s notice during the composition process. “The renga master might at any time decide that a submitted verse . . . lacked the appropriate tone for the current moment in the hundred-verse sequence” (p. 454). With this ability to recall all poems already recited, how could he not be influenced by the others’ stanzas? On a practical level, a poet’s goal was to provide a shift in meaning in the sequence with his stanza. But how could he be sure that he was actually causing a shift in meaning without taking into consideration the intention of the previous stanza? Indeed, how could he not take into account the previous two stanzas? The poet must out of necessity interpret the meaning of the previous stanza in order for his stanza to cause a shift in the sequence, in compliance with renga requirements. Additionally, in interpreting the meaning of the previous stanza, he must view it in relation to the stanza prior to it, for without doing so he would have difficulty interpreting the intentions of the previous poet. Then and only then can he compose his own stanza without fear of breaking the rules of renga. Stanzas 7, 8, and 9 of the Minase sangin will serve as an illustration of this process. 7. naku mushi no kokoro to mo naku kusa karete

without a thought for the feelings of the insect, the grass withers Spgi

8. kakine o toeba arawanaru michi

as I visit along the fence its barren dirt path Shphaku

9. yama fukaki sato ya arashi ni okururan

deep in the mountain does the village spend its days in a raging storm? Spchp

L   L 

135

Before Spchp can compose his own verse (verse 9), he must out of necessity interpret the meaning of Shphaku’s. However, in order to do so, he must consider Shphaku’s stanza in relation with Spgi’s stanza to “interpret” its intent. Shphaku describes a person visiting someone whose residence, surrounded by a fence, is beside a dirt path. The naked dirt path suggests that no one lives there. In conjunction with Shphaku’s, Spgi’s autumn verse can be reparced and reinterpreted as a metaphor for emotion. naku mushi no kokoro tomo naku kusa karete

The heart of the crying insect, without a friend the grass withers.

In the play of sounds, tomo (friend) suggests that the former resident was a friend. The withered grass not only strengthens the suggestion that the friend no longer lives there, but that his friend has died (naku), withered away as the grass. The absence of the friend saddens the visitor who laments as the crying insect. As a result, in conjunction with the previous stanza, Shphaku’s verse illustrates a visitor who is lonely and sad. After having come to his own conclusion concerning the meaning of Shphaku’s stanza, Spchp must now compose his own altering stanza. I should with all due prudence add that the above interpretation is mine, and I do not imply that Spchp’s interpretation would have been the same. The point is that Spchp must have gone through a similar process of interpretation in order to submit a stanza that would be associated with Shphaku’s stanza but significantly different from his intention with regard to the visitor. Spchp’s visitor may be calling upon a friend, or since he is in a village deep in the mountain, suggesting a place he rarely visits, simply asking for information. In either case, the village has few if any inhabitants as seen in the naked dirt path. However, the visitor seems to be more contemplative than sad; he concludes, with a rhetorical question, that the naked, empty dirt path was probably caused by constant stormy weather. Ultimately, Spchp’s stanza has changed the attitude of the visitor—from one who has lost a friend to one who wonders about the weather conditions of the area. The key is that he undoubtedly accomplished this by hearing, remembering, and interpreting previous poems successfully. In effect, the poet is first a reader—in that he attempts to interpret— and then a composer. Konishi may argue that the scribe’s other role as rule enforcer allowed the poet to be oblivious to the intentions of the previous poets. Should a rule be broken, the scribe would disallow the stanza and require the poet to create a new stanza. This situation is

136

I . L  H

conceivable with regard to the rule of intermission—the requirement that limits the appearance of particular terms over a designated number of stanzas—particularly if a word requires a great number of stanzas between appearances and if the poet was not a “master.” However, it is difficult to imagine a situation in which a poet such as Spgi, Spchp, or Shphaku would forget any of the stanzas, let alone those just recited. Such an implication would be, to say the least, unlikely. For renga analysis, a significant consequence of this discussion of intention is the fact that a single stanza is not composed with just the immediately previous stanza in mind. In order to create the next verse in a sequence, the poet must interpret the previous stanza only in the context of the stanza that preceded the previous one. Any new approach should recognize the poet’s understanding of verses beyond the previous one and consider ways of treating renga in ways that are not limited to the conventional two-stanza unit. If a poet obviously considered more than the previous verse to create his own, then why should we as readers be bound to such restrictive regulations? This is in contradistinction to most approaches. Miner states that “no stanza makes integral and continuous sense with any other stanza except that immediately before it or immediately after it.”16 While this approach seems to recognize the displacement of meaning in a single verse, it nonetheless clings to some sense of ontological purity by treating these as separate, independent paired units, unrelated to each other. Further, while thematic integrity for more than two stanzas is a requirement for certain topics, such as spring or love, temporality is not a consideration. If this is so, how does one explain the temporal qualities of the first three stanzas in Minase sangin alluded to earlier? To reiterate: the first stanza suggests the beginning of spring with the image of snow in early spring as evoked by the use of “haze”; the second, early spring with plum blossoms and the flowing water from melting snow; the third verse, spring nearing its peak as reflected in the willows. There is clearly a temporal association that makes integral and continuous sense beyond the immediate links from verse 1 to 3. If we were to take Miner’s statement at face value, should we presume that the temporal progression of spring is an accident of composition? Let us consider numbers 7 through 9 once again, this time from a temporal point of view. Stanza 7 and 8 describe the subject who calls upon a friend who has died, as discussed earlier. The addition of 9 locates the friend’s former residence in a village deep in the mountains, while the subject remains a visitor. If we add to this the tenth verse, however, we see that the

L   L 

137

subject is an inhabitant of the village 10. narenu sumai zo sabishisa mo uki

in an unfamiliar dwelling ven loneliness is miserable Spgi

The suffering of the village through a storm is transferred to the subject as suggested when Spgi confesses that even sabi is miserable. The subject has transformed from visitor to resident. A scenario can be constructed within a temporal framework. The subject goes to a village to see a friend, only to discover that he no longer lives there and has, in fact, passed away. The visitor notices the absence (or at least a low number) of people in the village, and this, as well as the demise of his friend, was no doubt caused by the storm. This situation provides the visitor with a place where he can practice an ascetic’s life of solitude, which may have been his intention in the first place when he visited his friend. (Indeed, we can imagine that the friend was also an ascetic living deep in the mountain.) Having taken up residence at this new unfamiliar dwelling, he spends his days through the raging storms. This eventually leads to regret as he realizes that this life of solitude is miserable. The continuity is further affirmed when we consider that both verses 7 and 10 express the desolation of being alone and away from civilization: an insect in the withered grass and a man experiencing a miserable loneliness. Since the same poet, Spgi, composed these two verses, it would be difficult to claim coincidence. A poet may feign ignorance to the meaning of verses of other poets, but how could he be oblivious to his own? In addition, if any kind of unity was disallowed by the rules, should we consider the scribe as unqualified since he failed to disallow such unity? Hardly. Since a poet needed to compose with consideration for the two previous stanzas—and likely even his own verse three stanzas removed—it would not have been surprising for his composition to manifest a connection with a stanza beyond the previous one. The scribe understood this. The poet couldn’t escape it. And we should not deny it. Renga study should resist the conventional poetic unit approach and explore the possibilities. So where to start? I have come to the earlier conclusion by considering the role of intention in a renga composition. However, I believe intention is useful only as a means to make plain an ignored characteristic of renga composition: the fact that the poet takes into consideration more than the previous verse. I do not suggest that we, like the poets did and like Hirsch would suggest, try to grasp the intentions of the poet, for we have already seen that meaning is mutable. And yet,

138

I . L  H

the role of the outsider—for us the reader—should not be discounted, a la Wimsatt and Beardsley. Instead, we might take to heart the words of Paul de Man. As he points out, trying to interpret a poem in any one way would simply be an attempt to explicate what was already there. For the interpreter of a poetic text, this foreknowledge [Heidegger’s knowledge within the circle of understanding, i.e. truth] is the text itself. Once he understands the text, the implicit knowledge becomes explicit and discloses what was already there in full light. Far from being something added to the text, the elucidating commentary simply tries to reach the text itself, whose full richness is there at the start. Ultimately, the ideal commentary would indeed become superfluous and merely allow the text to stand fully revealed. But it goes without saying that this ideal commentary can never exist as such . . . because it goes against the temporal structure of the hermeneutic process. The implicit foreknowledge is always temporally ahead of the explicit interpretive statement that tries to catch up to it.17

This is the crux of renga reading and interpreting. As discussed earlier, the poet interprets the intention of the previous poet, and then composes his own stanza that, while having altered the meaning of the previous stanza, creates its own particular meaning in association with the previous one. Although the subsequent stanza has its own intention—its own truth as it were—it is ultimately altered by the next poet. And the scholar who attempts to analyze the association between two renga sequences falls into the same trap: focusing analysis on the single-paired poetic unit. But in spite of all the good intentions of the critic/reader, the analysis will be rejected by the next verse. Every analysis of a stanza is immediately outdated by the subsequent stanza. Therefore, as de Man suggests, “implicit foreknowledge” cannot be grasped in literary works—in our case renga—for it is always one step ahead of the analysis. And this, perhaps more than anything else, reveals the foremost characteristic of renga.

: T P  A A [I]n language there is only difference. Even more important: a difference generally implies positive terms between which the difference is set up; but in language there are only differences without positive terms. Whether we take up the signified or the signifier, language has neither ideas nor sounds that existed before the linguistic system, but only

L   L 

139

conceptual and phonic differences that have issued from the system. The idea or phonic substance that a sign contains is of less importance than the other signs that surround it.18

As Ferdinand de Saussure points out, words only have meaning because they are different from other terms. And this difference must be seen in negative terms; that is, if we were to accept a term as positive, it would imply that the term was absolute, that it had its own identity independent of other terms. But this is not possible. Any word, particularly because it was arbitrarily designated to represent a given concept, can only be distinguished by its difference from other words. Jacques Derrida considered Saussure’s comments and came to an enlightening conclusion. The first consequence to be drawn from this is that the signified concept is never present in itself, in an adequate presence that would refer only to itself. Every concept is necessarily and essentially inscribed in a chain or a system, within which it refers to another and to other concepts, by the systematic play of differences. Such a play, then—difference—is thus no longer simply a concept, but the possibility of conceptuality, of the conceptual system and process in general.19

Derrida conceives concepts in and of themselves as not positive. That is, they do not have an innate, unchangeable, absolute meaning; any significance they may manifest rests only in their relationship with other concepts within the same system. In essence, they have no essence. They are merely possibilities; potentialities one step ahead of understanding because their significance is contingent upon any number of associations with other concepts in a ‘systematic play of differences.” They are, as Derrida coined it, differance. And so it is with renga. Each stanza in a sequence may or may not be composed with a particular meaning in mind. However, for the reader, whatever meaning that may be present always hinges upon the previous stanza. The particular meaning intended by the poet when composing can only be extracted in relation to the previous stanza. But this meaning is displaced and another applied by the subsequent stanza. For the reader, the value/significance of a stanza results solely from the simultaneous presence of the other stanzas around it. Each independent stanza is merely a possibility. And this is where an analysis of renga should eventually begin—the single, individual stanza and the possibilities from different perspectives.

I . L  H

140

The three poets participating at Minase were involved in the creation of a sequence of individual poems. And each verse can offer different meanings in relation to the verses around it. Verse 9 of the Minase sangin in relation to 8 and 10 will serve as an example. 8. kakine o toeba arawanaru michi

as I visit along the fence the naked dirt path Shphaku

9. yama fukaki sato ya arashi ni okururan

deep in the mountain does a village spend its days in a raging storm? Spchp

10. narenu sumai zo sabishisa mo uki

in an unfamiliar dwelling even loneliness is miserable Spgi

In Spchp’s mind, the location described in his stanza is intended to be a village that was visited by the subject as can be seen in association with Shphaku’s stanza. As mentioned earlier, Spchp’s stanza takes a contemplative stance, coming to the conclusion that the naked dirt path, the absence of people, was caused by a storm; and expressing his sadness for the people who experienced it. He seems to say that life is fragile; that in this world, man, as represented by the village, is impermanent, an attitude reflected by many poet–priests of medieval Japan.20 Spgi’s stanza offers a different perspective. He changes the village from a place the subject visited to his residence (sumai). The subject is relatively new to the mountain village as the word unfamiliar (narenu) suggests. And the storm has created in the subject such an overwhelming attitude of lament that even sabi, the aesthetic ideal of loneliness, has become miserable. He emphasizes his miserableness by changing Spchp’s stanza into a question: Should I spend my days deep in the mountains suffering through these storms? Spgi’s stanza has transformed the storm from an observation into a personal experience. And in this transformation, we see the possibilities in Spchp’s stanza to express two distinct and different perspectives.

A C  I  W Renga followed traditional rules of composition and its content manifested a sense of decorum focusing on themes closely related to court poetry. Indeed, established renga poets such as Nijp Yoshimoto (1320–1388) and Spgi were traditional waka poets as well, and their

L   L 

141

linked verses reflected this tradition. Diction was generally limited to the approved lexicon on Japanese words used in court poetry. The important images in a renga sequence—flowers, seasons, the moon—were conventional topics in waka composition. However, by the Edo period, linked verse had turned away from traditional poetry. Early in his career, Matsuo Bashp was a member of the Danrin School, which practiced a style that exhibited a relaxed attitude toward tradition. He no longer restricted himself to the limited diction of waka. Everyday words deemed unpoetic by traditional standards such as door (tobira) or yam gelatin (kon’nyaku) sprinkle his compositions. Buddhist and Chinese terms pronounced in Chinese also find their way into his verses, a practice unthinkable in waka (lit. Japanese poetry) composition. The topics of his poetry also reveal his attempts to distance himself from traditional poetry. Until around 1680, he composed much of his haikai by incorporating the Danrin’s sense of irreverence and humor into his style, often using the time honored technique of classical allusion. Bashp would allude to a classical piece of literature and twist it in an amusing way, as his neko no tsuma (the cat’s mate) hokku, which he parodies the courtly love affair of Ariwara Narihira of the Heian period by comparing it to cats in heat.21 However, Bashp’s inclination to distinguish himself from the classical poetic tradition is perhaps best exemplified by the style he developed and practiced toward the end of his life: karumi. According to Ogata Tsutomu, it was established as a poetic principle upon his return to the capital from the journey to northern Japan that resulted in his masterpiece, Oku no hosomichi, and was a consistent technique throughout his final years, as reflected in the writings of his disciples, in which was related in one way or another to karumi.22 Developed— as discerned from Bashp’s writings—from 1689 to 1694, karumi was a principle that accepted life as it is; that practiced non-contrivance, reflecting the rhythm of the mind, as is, in verse; and that attempted to discover poetry in the mundane everyday things in life through common words.23 This approach to poetry is not only reflected in his hokku but also in his linked-verse compositions. Sarumino—a collection of haikai by Bashp and his disciples edited by Mukai Kyorai (1651–1704) and Nozawa Bonchp (d. 1714) in 1691, under Bashp’s supervision—represented a peak in Bashp’s haikai composition. The different haikai sequences in this collection were composed soon after his Okunohosomichi journey and as such, it is not surprising to see his principle of karumi reflected in these poems.24 Within this compilation, the thirty-six verse sequence “In the City” (Ichinaka wa) is perhaps the best representative of Bashp’s

142

I . L  H

final legacy, the principles of karumi, and his separation from classical poetic traditions. The sequence in general and Bashp’s contributions specifically, all suggest an attitude that is uncontrived, natural, and, to a greater or lesser degree, mundane. The sequence has been previously addressed by well-known scholars such as Miner and Ueda, but their approach has seemingly been influenced by the traditional approach of renga. Predictably, they combine two verses at a time, discussing the merits of each thirty-one-syllable unit much as most scholars have approached traditional renga. While this approach is understandable for renga itself, it seems inconsistent for haikai, for according to Bashp, haikai was “free.” With waka poetry, given its elegance, they had to go to such lengths to work out a design and compose, but with haikai, given its freedom, there is no achievement in simply composing an ordinary scene.25

While Bashp was speaking of haikai diction, it still suggests his attitude to haikai in general. Waka—and renga, which shared much of its practices and traditions—was bound by “elegance” whereas haikai provided a freedom to choose any word or image to compose simple scenes. And this is key. For Bashp, the freedom in haikai was a way to distinguish it from traditional poetry. Indeed, Kyorai echoed these very sentiments. “People in society think as if haikai is a servant of renga. The master’s judgment was completely different.”26 In other words, Bashp did not see haikai—and by extension, those who composed it—as being subservient to renga. Indeed, it was its own style of poetry, one that was equal to but different from renga. And modern readers should view them as such. Yet, modern scholars treat haikai as if it were the equivalent of renga. In virtually every analysis of haikai, verses are taken two at a time.27 They are arranged, read, and interpreted in units of two consecutive sequences, providing for a limited, restricted reading. The following is a consideration of the first twelve verses of Ichinaka wa, an attempt to illustrate the pleasures of a freer, unbound approach. It will not only consider the effects of neighboring verses, but will also allow for the intertextual realities that affect all reading. 1. ichinaka wa mono no nioi ya natsu no tsuki

In the city the smells of things— a summer moon. Bonchp

L   L 

143

Most students of Japanese literature will recognize the hokku, the first verse of this well-known haikai sequence. One such student, Dazai Osamu, considers it a fine verse. When it’s hot, what I unexpectedly recall is “A summer moon” that is in Sarumino. . . . It is a fine verse. The expressions of senses are precise. I call to mind a fishing village. Depending on the person, one will imagine the area around Jinbpchp of Kanda, or recall the night store in Hatchpbori. They are likely diverse, but anything one imagines is fine. It is mystifying because a single summer night from one’s own past is brought back to life vividly.28

Dazai’s comments regarding Bonchp’s opening verse is suggestive of the freedom each individual has to read a poem, an opinion that reflects the ideas of poets as far back as Fujiwara Shunzei.29 It provides us with an opening to consider poems beyond traditional readings, to contemplate other meanings based on the reader’s experience, a particularly welcome attitude when reading linked verse, a poetic form whose very concept promotes the fluidity of meaning. As the hokku, this opening verse is self-contained. The poem depicts a summer evening in the city, both hot and somehow cool. Bonchp accomplishes this by presenting two images separately but combined in a synesthetic fashion. The smell of things in the city— food, alcohol, garbage—suggests a hot and humid evening among the citizenry. However, the heat gradually cools as the smell lingers beneath a full summer moon; but it also seems to combine into a single sensation: the dynamic aroma of a city moon. This sensation is all the more striking in that the elements at the core of this sensation are opposites—the smells of man and the moon of nature, the activity that generates the smells and the tranquility commanded by the moon, the ephemeral qualities of a whiff of smells and the timeless view of the moon. The ultimate effect is one of pleasant activity in the city, a sensation made real and immediate through the union of sight and smell, an effect that Dazai likely felt as well. Unfortunately, Dazai’s favorable opinion dies with the opening verse. He continues his critique of the first eleven sequences of Ichinaka wa in his essay “Tengu,” indicating that most of what follows is mediocre. The title of this piece is a reflection of the author’s attitude at the moment he wrote it. “I am going mad from the summer heat and have become a Tengu. Forgive me” (p. 234). A supernatural being in human form with a long nose, Tengu is used to describe a conceited and arrogant person. And, indeed, Dazai’s

144

I . L  H

criticisms are arrogant, if not tongue in cheek. But his comments reflect the thoughts of the unconventional reader, one who is not particularly concerned with the sanctity of the two-verse poetic unit, and I will refer to his criticisms in conjunction with my own. Dazai considers the subsequent poem by Bashp as superfluous and therefore tedious. His blunt opinion notwithstanding, he suggests that the second verse attached by Bashp attempts to explain Bonchp’s poem and so it is tedious. 2. atsushi atsushi to kado kado no koe

“it’s hot, it’s hot” voices from gate to gate Bashp

At first glance, Dazai’s comment seems convincing. Bashp’s 7/7 verse, seemingly simplistic, repeats two of the words—atsushi (hot) and kado (gate)—leaving the reader with a total of only three words and two particles. However, Dazai may have overlooked the cool quality of the moon in the opening verse, an aspect that Bashp, well-versed in the classics despite his frequent parody of them, was well aware of. The moon, even a summer one, traditionally projects a sense of coldness through its whiteness, as indicated by the following poems by Saigyp, a poet whose poems Bashp was very familiar with. “When he composed poems on the summer moon.” natsu no yo mo Even on a summer night ozasa ga hara ni frost settles over a field shimo zo oku of small bamboo tsuki no hikari no because the light of the moon sae shi watareba crosses chillingly clear yamagawa no iwa ni sekarete chiru nami o arare to mi suru natsu no yo no tsuki

The mountain river is blocked by boulders and the splashing waves appear as bits of hail: a summer night’s moon30 (Sankashu 245 and 246)

These two waka were composed on the topic of “summer moon” but convey a cold sensation. In the first poem, dew settles over a field of small bamboo, but it turns into a chilly frost due to the clear moonlight. Similarly, the second poem suggests a chill through hail, an image created by splashing beads of water reflecting the moonlight. Both of these poems lend a sense of coolness to Bonchp’s hokku, thereby suggesting a pleasant summer evening.

L   L 

145

However, Bashp’s verse underscores the oppressive heat by effecting a tension between the coolness suggested in Bonchp’s summer moon and the unbearable heat expressed by the townspeople. Juxtaposed to the previous poem, Bashp conveys an unrelenting heat on a hot, humid Japanese summer night despite the moonlight. Further, the simplicity and repetition of words urges us to focus less on its meaning and more on its sounds. The almost panting effect of the unvoiced consonants of “atsushi” is joined by the harsher, unvoiced alliteration of the last line, “kado kado” and “koe.” It is far from the “redundant annotation” that Dazai would dismiss it as. When associated with the previous poem, the sounds of people complement the sights and smells of the city. However, there is a sudden shift in direction when considered with Kyorai’s poem. 3. nibangusa tori mo hatasazu ho ni idete

without completing the Second Weeding stalks sprout ears of rice Kyorai

This verse evokes a new environment for the heat expressed in Bashp’s poem. Nibangusa is the second weeding of rice fields, a job undertaken to prevent weeds and other unwanted plants from sharing the soil with the rice plants. It is generally carried out in the sixth month of the lunar calendar and is associated with the hottest time of summer, which relates it with the previous verse by Bashp. As the verse indicates, the second weeding has yet to be completed, but the stalks have sprouted ears of rice already, suggesting that the entire growing season has been hotter than usual, as the rice plants mature ahead of schedule. The effect of this poem, of course, is to shift the heat from the odor-filled city to a rural setting and to change the mood from one evoked by a pleasant evening in the city to one reflecting active labor in the fields. As prescribed in the rules of traditional linked verse, Kyorai has shifted the poem in another direction. More interestingly, his poem provides a different context for Basho’s poem, thereby changing the meaning and images from town people to farm folk working their fields. While we cannot deny the instability of meaning, there is a sense of continuity present. The verses sandwiching Bashp’s poem— urban and rural—provide a degree of poetic tension, underscoring the difference between the tranquility brought on by the moon and the agitation of farming, a tension unappreciated if a reader does not consider all three poems together. The different settings also suggest the anxiety of influence brought about by the rules of renga. In his

146

I . L  H

attempt to change the direction of the sequence, Kyorai sent it on a rather predictable path of opposites—urban to rural. This predictability mirrors Dazai’s rather condescending comments on this verse. He judges it as “uninteresting,” typical of Kyorai whom he characterizes as “unsophisticated.” Indeed, it evokes from him a “wry smile” (nigawarai) that slowly—after a few more readings—arouses a sense of embarrassment. But he does offer a “not bad” for the next link by Bonchp. 4. hai uchi-tataku urume ichimai

brushing off the ashes a single sardine Bonchp

The verse portrays someone preparing a simple meal. Teruoka and Nakamura envision a busy farmer eating his meal indoors by an ashfilled hearth. However, the word ichimai (one flat object) indicates dried fish, something that can be left outside without fear of spoiling while working in the fields. Brushing off the ashes, then, might suggest that the fish was covered by the ash when the farmer fanned the fire, a motion suggestive of eating outdoors. Given the flow of the sequence, it is not difficult to imagine this person as a farmer out in the fields, perhaps too busy from weeding to go home for lunch. The verse seems to complement the previous poem although Dazai, for his part, sees it as affected (kiza) and overly stylish (haikara sugiru). Still, it is better than his opinion of the next poem by Bashp. 5. kono suji wa gin mo mishirazu fujiyusayo

Around here they don’t even recognize silver. What a predicament! Bashp

Conventional readings of this poem suggest that a person from the city has entered into the scene and observes the inconveniences of a rural area. The visitor suggests that these farm people have never even seen money, and that he finds this to be inconvenient when he cannot purchase food, perhaps the fish in the previous poem, muttering, “What a predicament.” However, Dazai believes that this reading renders the verse as trite: this observation would be too obvious and the words kono suji sound overly sarcastic. Instead, he suggests that the words are uttered by the farmer brushing the ashes off his fish, complaining about his life and how little money he has. He reads the poem as, “We all don’t even recognize money. What a predicament,” and pronounces it a “verse that is not good.” If we view this poem as he

L   L 

147

does, then it is perhaps trite. But more importantly, Dazai makes it clear that he will read the poem differently despite his knowledge of other conventional readings. He recognizes and indeed asserts his privilege to read a poem his own way. 6. tada tohypshi ni nagaki wakizashi

simply out of synch a long sidearm Kyorai

This verse by Kyorai is suggestive of an awkward-looking man, one who wears a sword that is obviously too long for him. Poisoned, as it were, by the two different readings of the previous poem, the image becomes quite complex. Perhaps we have a rpnin, a wandering masterless samurai, who is perturbed by the fact that his money means nothing to the farmers visualized in earlier links. Indeed, we can envision him leaving an urban center into a rural one where he is inconvenienced by those who put little stock in money. On the other hand, Dazai’s reading of the previous poem might suggest a farmer who straps on a sword, as awkward and comical as the image of a meek farmer—Yphei, who dons armor and learns to fight in Kurosawa Akira’s Seven Samurai, certainly comes to mind—complaining about his lot in life. As such, Dazai’s comments of the previous poem actually make this verse quite complex and more interesting, even though he himself derides this verse as “messed up” (mecha kucha) and “beyond incredulity” (kisp-tengai) as to be a “cow from the void,” going as far as to assess blame on Bashp for having composed the “vague” verse that preceded it. 7. kusamura ni kawazu kowagaru yumagure

in a cluster of grass fearing a frog in the evening dusk Bonchp

The awkward-looking, sword-wielding man is, in this verse, startled by a frog lurking in a cluster of grass. Dazai considers this a tedious composition as well. And he is not really surprised at this for, according to him, linked verses are difficult to compose, and “in one-hundred flawed verses, eventually there will be one successful one.” Still, it is not necessarily a substandard (gehin) poem, writes Dazai, but rather one that attempts to “make up” (ma ni awase) for the previous poem by Kyorai. More importantly for our discussion, however, is the fact that this verse continues a single stream of consciousness; a consistent and by

148

I . L  H

now rather lengthy thread has run through more than the normal two or sometimes three verses allowed by the prescriptive rules of renga. The sequence begins with a summer moon above the city. While the smell of the streets suggests the heat, the moon provides a sense of coolness, although this only serves to underscore the heat that everyone breathlessly complains about at the entrance of their dwellings. The heat extends predictably from an urban setting to a rural one where farmers indicate that the rice is sprouting even before the second weeding, a sign that the summer is hotter than usual. As these farmers work, pulling the unnecessary weeds around the rice, they eat a modest lunch of one dried sardine, brushing off the ashes that have collected on it. A wandering samurai is upset that he cannot buy any food here as the farmers will not accept his money. Or, as Dazai suggests, a tired farmer complains about his lot in life: that he has no money. And this man, samurai or farmer, is an awkward one at that, wearing a long sword on his side that is completely disproportionate perhaps to his height or build, but more likely his station in life, for the disproportion between man and sword is highlighted when he reveals that he is easily frightened by a frog croaking or jumping in the tall grass in the evening dusk. These seven links exhibit a logical continuity that is interesting but would otherwise be unacceptable if judged by the prescriptions of renga. Did Bashp, Bonchp, and Kyorai not realize this? Or were they perhaps more relaxed in their adherence to the conventions of linked verse? Certainly, words such as ‘sardine” and “silver” suggest their reluctance to apply traditional diction. So why not a reluctance to follow the other rules? Or did Bashp realize this and think that enough was enough, for in the following link, we have the first radical shift in perspective. 8. fuki no me tori ni ando yurikesu

a lantern’s light extinguished with a jolt as she hunts for butterburs Bashp

While Dazai considers this verse as one that—following Bonchp’s lead—finally and quietly erases the image of “a long sidearm,” it is much more than that. From the beginning, we have had a man observing a town, leaving it and traveling through a rural area, observing the life around him. But in this verse, we clearly have a female: gathering butterburs ( fuki; Petasites japonicus) for their edible sprouts is an activity that indicates a woman. As this women gathers butterburs at dusk, she is startled by something rustling in the grass, and with a jolt, the light in the lantern is extinguished. The sound,

L   L 

149

when juxtaposed with the previous verse, is caused by the frog. As are most images of women gathering herbs, it is one of spring, a departure from the summer heat that seemed to underlie the previous verses. 9. dpshin no awakening to aspirations okori wa hana no to the path to enlightenment tsubomu toki

at a time when blossoms bud Kyorai

In another significant shift, what has been a secular image now shifts to a religious one. This verse has Buddhist overtones as indicated by the word dpshin, which suggests aspirations to enlightenment. The image suggests that the woman of the previous verse is now a nun. Taken together, the verse depicts a woman gathering butterburs, who realizes the ephemeralilty of life when the light in her lantern extinguishes. It is this realization, which occurred in her youth—the time when blossoms bud—that has induced her to take Buddhist vows and become a nun. Dazai finally has a positive comment for a verse by Kyorai: “outstanding.” But he has nothing else to say. “For an outstanding verse, one can only be awestruck.” 10. Noto no Nanao no fuyu wa sumiuki

at Nanao on the Noto peninsula winter brings a harsh life Bonchp

Bonchp does not go along with Kyorai, as Dazai sees it, and “slams the door to his heart.” This “stiff” verse is “simply prosaic without a melody.” Dazai seems to be referring to the severity of the imagery. Nanao on the Noto Peninsula is a seaside village off the Japan Sea. During the winter, it is assailed by the cold Siberian winds from the continent and represents a harsh winter. And leading a harsh life during the winter is the nun. While Dazai may characterize Bonchp’s verse as prosaic, it provides a stark counterpoint to the beginning of this short sequence that spans the last four verses. We can envision a young lady who is plucking herbs startled by a frog, causing her lantern to extinguish. Upon which she realizes the ephemerality of life and determines that she is to become a nun, all of which she is now recalling as she spends a harsh winter in Nanao, a far cry from the spring of her life. Winter and spring are different, suggesting a drastic change in direction, but like the urban–rural opposites at the beginning of the sequence, there is an obvious parallel between spring and winter. And this leads us to another obvious parallel: winter and old age. 11. uo no hone

aging to the point

I . L  H

150

shiwaburu made no where one must gum at fish bones— oi o mite reaching old age. Bashp

As is typical of Japanese, the subject is often dropped and there are no pronouns to indicate gender, but in this particular verse, English-language commentators assign a male gender.31 This is likely due to the imagery of a wretchedly toothless old man sucking on fish bones. Few have questioned this assignment because scholars and students alike readily accept the two-sequence unit, thereby closing off the previous image of the lady–nun. But if we allow ourselves to break these boundaries, to go beyond this limitation as we have in the previous sequences, then perhaps we can imagine an old woman, one who is in the winter of her life, living out her remaining years on the harsh Noto Peninsula. Indeed, how pitiable a figure would she present? How effective the pathetic image when juxtaposed to the young lady picking herbs? But this begs the question, why do most scholars presume the subject is a man? Perhaps those who readily take this view may, in fact, be influenced because they are familiar with the subsequent poem. 12. machibito ireshi komikado no kagi

having let in one who waits— the keys to a side gate Kyorai

Kyorai himself comments that this verse was composed to incorporate a tale (monogatari), indicating it is the “old man in charge of the gate,” a reference to a scene in the Suetsumuhana chapter in the Tale of Genji, in which Genji leaves Prince Hitachi’s daughter (Suetsumuhana) on a snowy morning.32 As Genji prepares to leave through the middle gate, he realizes that it is closed and an old man comes with keys to unlock it. As such, the verse is a kind of mirror image of the tale: instead of a person being let out, he is being let in. Any reader familiar with Kyorai’s comment would then be inclined to view the elder gnawing at fish bones in the previous poem as an old man, because the gatekeeper in verse 12 has already been assigned this gender by the poet himself. In other words, while a simple sequential reading would not automatically indicate a change in gender, familiarity of the verse to come could induce the reader to presume that the subject is a male. This is indicative of how easily readers can be influenced by a variety of stimuli. One’s reading can be affected by poems anywhere in the sequence, before and after a given verse. Is this wrong? I would dare say no. Indeed, it is precisely this possibility that reveals the many aspects of linked verse. By loosening the links of the traditional rules

L   L 

151

of renga, we can appreciate and enjoy the exquisite depths and complexities therein.

B  S An obvious consequence of liberating a renga or haikai verse from its traditional link is to treat it as an independent verse. If we view a verse as able to relate freely with either or any of its neighboring stanzas, then perhaps we can view each link in a context that goes beyond the sequence itself. This is not, of course, a novel idea. The classical commentator mentioned earlier states that the first verse composed by Spgi relates to a waka composed many years earlier by the retired emperor Gotoba (1180–1239): 1. yuki nagara yamamoto kasumu yube kana

as the snow remains as is the foot of the hills grow hazy— such an evening!

It is the essence of the source poem: Miwataseba yamamoto kasumu Minasegawa yube wa aki to tare ka iiken

As I gaze across the foot of the hills grow hazy at Minase river— who could have possibly said “for evenings, it is autumn?”

It states that, even an evening when the foothills simply grow hazy is beautiful and moreover a scene where the snow remains is rare and profound.

Given the circumstances of the Minase sangin, the commentary is likely correct in attributing the inspiration of the opening poem to Gotoba. Tradition also holds that the sequence was to be offered up in homage to the retired emperor at the shrine of his namesake.33 The sequence was created in Minase of Yamashiro Province (modern Osaka Prefecture) where he maintained an imperial villa. Further, 1488 is the 250th anniversary of Gotoba’s death. There is no reason to doubt the commentary. Spgi’s stanza describes the location; by alluding to Gotoba’s poem, he lets us know that the hazy foothills refer to those at Minase River. Gotoba’s poem is also considered an early spring poem, as reflected in its placement as the 36th poem in Shinkokinshu.34 Spgi vividly describes the early spring theme, if not the beginning, by incorporating yuki nagara (lit. while snow remains as is) into his stanza. Further, the allusion suggests the beauty of the occasion and setting, for the retired emperor implies that a spring’s evening is just as beautiful and evocative as an autumn’s eve. However, there may be more to the poem than just its aesthetic qualities. The comments

152

I . L  H

say that Spgi’s verse is the essence of Gotoba’s poem. However, Gotoba’s poem is more than a statement of the beauty of an evening where the foothills grow hazy. It is indeed one of the strongest statements of the changing aesthetics sensibilities of the time. In this poem, Gotoba critiques the Heian aesthetics that ruled court poetry for centuries by questioning the words of one of the de facto arbiters of taste for the Heian period, Sei Shpnagon. In her Makura no spshi, a collection of entries that clearly reflected the aesthetic sensibilities of her peers, she writes that the most beautiful time of day in autumn is the early evening. Of autumn poems in imperial anthologies that refer to the time of day, the overwhelming number mention twilight or evening. However, Gotoba stands up to this tradition, asking “who could have possibly said / ‘for evenings, it is the autumn.’ ” Was Spgi trying to mention that this particular sequence too was breaking from tradition? Does Spgi’s appointment in the same year as an official to the Kitano Renga Kaidokoro—the highest honor in the world of renga at the time—have anything to do with his image of himself?35 Are we compelled to consider that Spgi did have a reason for alluding to the essence of Gotoba’s poem due to the fact that Minase sangin is traditionally considered the best model for a one-hundred-verse sequence composed by three poets?36 Indeed, even the unknown commentator seems to manifest a similar thought through his own error. He has obviously confused two different poems from the Shinkokinshu. The poem to which he refers is Gotoba’s except for the last line, which should read: nani omoiken (what was I thinking?) (Shinkokinshu, 36). The last seven syllables mistakenly inserted are from another poem by Fujiwara Kiyosuke (d. 1177), one to which Gotoba himself likely alludes. usugiri no magaki no hana no asa shimeri aki wa yube to tare ka iiken

In a light mist flowers by the bamboo fence loll from the morning dew— who could have possibly said “in autumn it is the evening?” (Shinkokinshu 240)

Kiyosuke’s poem is similar to Gotoba’s in tone. He too questions the aesthetic values of Sei Shpnagon by asking “who could have possibly said / in autumn it is the evening?” Kiyosuke admires the flowers hanging heavily from the morning dew and concludes that autumn mornings are just as moving as the evening. If the commentator believed that Spgi’s verse manifested a sense of breaking with tradition, then the commentator could have thought of both of these poems and easily mixed them up.

L   L 

153

Given the reality that a renga verse can relate to a poem either before or after it, we should accept then that it not only refers to past poems but to future ones as well. This is not to suggest Spgi saw into the future. But as the poet could only compose diachronically, looking to the previous poem, the reader is privileged to read synchronously in either direction, which brings us back to the beginning of this chapter. 10. narenu sumai zo sabishisa mo uki

In an unfamiliar dwelling even loneliness is miserable.

Sogi’s verse suggests finding tranquility in solitude, but even loneliness is miserable in a place unfamiliar. Indeed, it suggests he has to endure storms in a mountain village, if we read it in conjunction with its preceding verse—as we are privileged to do—his loneliness is heightened and his sense of misery augmented. The verse is given greater significance if we consider that loneliness is an aesthetic quality that transcends centuries. As contemporary readers familiar with both Minase sangin and Saga nikki, we can allow ourselves the luxury of letting Spgi’s verse bring to mind Bashp’s. Nothing is as pleasing as living alone. Chpshp Inji says, “When a guest obtains a half day of tranquility, the host loses a half day of tranquility.” Sodp was always moved by these words. As for myself, I recited this verse when I was alone at a certain temple. uki ware sabishigarase yo kankodori

I who am miserable— make me feel lonely, cuckoo.37

In 1691, Bashp traveled to Saga, then a rustic area on the outskirts of the capital, and stayed there from May 15 to June 1. He enjoyed the surrounding solitude, but had frequent visitors, much to his dismay. According to the diary, there were only four or five days when he was able to be alone, including the twenty-second day of the fourth month, the date of the excerpt mentioned earlier. He quotes the poet Kinoshita Chpshpshi (1569–1649), who explains the incompatibility of entertaining guests and seeking tranquility, then states that a poetic colleague Yamaguchi Sodp (1642–1716) was always moved by these words. Bashp also seeks the pleasures of solitude but suggests he needs a hand to reimmerse himself into a state of tranquility. He is miserable because of all the visitations and now asks the cuckoo to cry its lonely kanko so he may feel lonely. With Bashp’s verse and its own circumstances, Spgi’s poem indeed evokes a great sense of sorrow and misery. Not only is the poet/subject

154

I . L  H

desolated by the storm, he is lonely because he lives in an unfamiliar place where no one visits him. Unlike Bashp, who has the luxury of wanting to discourage visitors, Spgi’s subject—who may want visitors to help him through a trying time—is left to deal with his situation alone, a loneliness that is truly miserable. This, consequently, suggests that the aesthetic pleasure of loneliness is not so much an aspect of real loneliness as a separate artificial construct cultivated and maintained by the literati of premodern Japan. But, that is a topic for another time and place. This chapter has dwelled on the possibilities of renga and renku, attempting to suggest to the reader that there are options in reading and studying these literary forms. The conventional approach is valid when analysis centers on the actual diachronic act of composing a linked verse. But I might suggest that intention should play a greater role, for it will give the researcher the freedom to go beyond the paired poetic unit. I also propose a more synchronic approach, one that will allow the reader to let the single verse exist on its own accord, free to pair with any verse it wants, prior, later, in, or out of the sequence itself, and consequently free to mean whatever it wants. This approach will provide the opportunity to explore the limitless and inherent possibilities of traditional and comic linked verse waiting to be found.

N 1. Howard Hibbett, “The Japanese Comic Linked-Verse Tradition,” Harvard Journal of Asiatic Studies 23 (1960–1961): 79. 2. Robert Brower and Earl Miner, Japanese Court Poetry (Stanford: Stanford University Press, 1961), 417. 3. Earl Miner, Japanese Linked Poetry (Princeton: Princeton University Press, 1979). 4. I have consulted the following published text for this study: Ijichi Tetsuo, Rengashu, in Nihon koten bungaku taikei, 39 (Tokyo: Iwanami Shoten, 1960), 345–347. 5. Saitp Yoshimitsu, ed., Chusei renga no kenkyu (Tokyo: Yuseido, 1979), 42. 6. Shuichi Katp, A History of Japanese Literature, David Chibbett, trans. (New York, Tokyo: Kodansha International, 1979), 294. 7. Jin’ichi Konishi, “Association and Progression: Principles of Integration in Anthologies and Sequences of Japanese Court Poetry, A.D. 900–1350,” in (trans.) R.H. Brower and E. Miner, Harvard Journal of Asian Studies 21 (1958): 122–125. 8. Makoto Ueda, Nihon no bungaku riron: kaigai no shiten kara (Tokyo: Meiji Shoin, 1975), 131. 9. Ijichi, Rengashu, 346. There is only one copy of the Minase sangin with premodern commentary, and it is used as the source for the printed

L   L 

10.

11.

12. 13. 14. 15.

16. 17. 18. 19.

20.

21.

22. 23.

155

version in the Nihon koten bungaku taikei series. It is owned by Konishi Jin’ichi and is indeed considered a rare item (27). Unfortunately, the transcription date is missing and the commentator is apparently unknown. The Japanese commentary ends the interpretation with “to nari” (lit. it is that; mod. J. to iu koto de aru). The copula form is unmistakably distinguished by the use of the Chinese character employed to indicate the declarative (dantei) auxiliary verb nari. Immanuel Kant, “Critique of Judgement,” in (ed.) Immanuel Kant; (trans.) James Creed Meredith, Critique of Aesthetic Judgement (Oxford, England: Clarendon Press, 1911), 172–173. E.D. Hirsch, Validity in Interpretation (New Haven, CT: Yale University Press, 1967), 6. W.K. Wimsatt, The Verbal Icon (Lexington: University of Kentucky Press, 1954), 3–18. Jin’ichi Konishi, “The Art of Renga,” K. Brazell and L. Cook, trans., Journal of Japanese Studies 2 (1975): 49. H. Mack Horton, “Renga Unbound: Performative Aspects of Japanese Linked Verse,” Harvard Journal of Asiatic Studies 53, no. 2 (December 1993), 443–512. Miner, Japanese Linked Poetry, 58–59. Paul de Man, Blindness and Insight (Minneapolis: University of Minnesota Press, 1983), 30. Ferdinand de Saussure, Course in General Linguistics, Wade Baskin, trans. (New York: McGraw-Hill, 1966), 120. Jacques Derrida, “Différance,” in (trans.) David Allison, Speech and Phenomena: And Other Essays on Husserl’s Theory of Signs (Evanston, IL: Northwestern University Press, 1973), 140. It should be noted that Derrida accepted Saussure’s linguistic premise of the sign but could not accept his conclusions. In his book Of Grammatology, Derrida criticized Saussure for privileging speech over the written word, which indicated that speech was “closer” to the signified. To Derrida, this conclusion marginalized text. Perhaps the best example would be found in Hpjpki by Kamo no Chpmei (1155?–1216). Chpmei contemplated the apparent unchanging qualities of families, great and small, over generations in Kyoto, concluding that “should we inquire whether this is true, [we find] the houses that had existed long ago are now rare.” Nishio Minoru, Hpjpki, Tsurezuregusa, in Nihon koten bungaku taikei: 30 (Tokyo: Iwanami Shoten, 1957), 23. Neko no tsuma / hetsui no kuzure yori / kayoikeri. (The cat’s mate / commutes through the crumblings / of the earthen stove). This poem alludes to the fifth section of Ise monogatari. See Kon Eizp (ed.) Bashp kushu, Shinchp Nihon koten shusei, 51 (Tokyo: Shinchpsha, 1982), 32. Ogata Tsutomu, Za no bungaku (Tokyo: Kadokawa shoten, 1973), 153. See ibid., 152–153. Also see, Hori Nobuo, “Karumi” in Nihon koten bungaku daijiten, Vol. 2 (Tokyo: Iwanami Shoten, 1984), 38; and Makoto Ueda, Matsuo Bashp (New York: Kodansha International, 1982), 64–66, 158–160.

156

I . L  H

24. Not everyone sees it this way. Earl Miner views Bashp’s late style as “more a matter of hokku than of sequences,” suggesting that karumi was more readily applied to hokku—the opening verse of a haikai that could also be seen as a haiku—and that verses attached in sequences reflected Bashp’s earlier style of okashimi, a style that is “peculiar, odd, and perhaps a bit humorous.” See Miner, Japanese Linked Poetry, 116–123. 25. Bashp’s disciple Kyorai is quoting his master here. See Kyoraishp in Ijichi Tetsuo, Omote Akira, and Kuriyama Riichi, eds., Rengaron shu, npgakuron shu, hairon shu, Nihon koten bungaku zenshu, 51 (Tokyo: Shpgakukan, 1973), 435–436. 26. Kyoraishp., 476. 27. See, e.g., Ptani Tokuzp and Nakamura Shunjp, Bashp kushu, Nihon koten bungaku taikei, 45 (Tokyo: Iwanami Shoten, 1962); Teruoka Yasutaka, Nakamura Shunjp, Kaneko Kinjirp, Renga haikai shu, Nihon koten bungaku zenshu, 32 (Tokyo: Shpgakukan, 1974); and Miner, Japanese Linked Poetry. 28. Dazai Osamu, “Tengu,” Dazai Osamu zenshu, Vol. 10 (Tokyo: Chikuma Shobp, 1976), 230. This essay was originally published in a journal called Mitsukoshi, 9 (1942). 29. At the Sumiyoshi Shrine poetry match (Sumiyoshi taisha utaawase), Shunzei stated the following in his closing comments. “Needless to say, when it comes to the hearts of men pulled by the nets set in the distant sea lanes, it is rare for the floating fishermen to come to terms as one single mind. Although [a particular plant] is named ‘beach reed’ (hamaogi) at the famous Kamikaze Isejima, [the same plant] is simply referred to as a rush (ashi) at the Naniwa ferry-crossing, and a reed (yoshi) in the eastern provinces; although concerning the same poem, the sentiments of men are also different.” Shinpen kokka taikan, Vol. 5 (Tokyo: Kadokawa Shoten, 1983), 210. 30. Saigyp, Sankashu, Shinchp Nihon koten shusei: 49 (Tokyo: Shinchpsha, 1982), 72. 31. See Ueda, Matsuo Bashp, 95 and Miner, Japanese Linked Poetry, 305. 32. Ijichi, Omote, and Kuriyama, Rengaronshu, npgakuronshu, haironshu, 511. Also see, Teruoka, Nakamura, and Kaneko, Renga haikai shu, 466–467, n. 12. 33. See Miner, Japanese Linked Poetry, 172, and Ijichi, Rengashu, 26. 34. The poems of Shinkokinshu, as those in all imperial anthologies, are classified first by the four seasons, beginning with spring. The Shinkokinshu is particularly noted for its temporal progression of seasonal poems, as pointed out by Konishi, “Association and Progression,” 74. 35. See Shimazu Tadao’s entry, “Spgi • Spchp” in Kubota Jun and Kitagawa Tadahiko, eds., Chusei no bungaku, Nihon bungaku shi 3 (Yuikaku, 1976), p. 327; and Ijichi Tetsuo’s entry, “Spgi” in Nihon koten bungaku daijiten 4 (Iwanami, 1984), p. 13. 36. For typical remarks asserting that Minase sangin is the best example of renga, see Ariyoshi Tamotsu, Chusei nihon bungaku shi (Tokyo: Yuikaku,

L   L 

157

1978), 51 and Niho nihon bungaku daijiten, Vol. 5 (Tokyo: Iwanami Shoten, 1984), 609–610. 37. Imoto Npichi, Hori Nobuo, and Muramatsu Tomotsugu, eds., Matsuo Bashp shu, Nihon koten bungaku zenshu, 41 (Tokyo: Shpgakukan, 1976), 393; for translation of this passage in Saga nikki see Barnhill, Bashp’s Journey, 82–83. The poem is actually a revision of a poem composed 1689.10.19, at Daichiin Temple in Ise Province. uki ware o sabishigarase yo aki no tera

I who am miserable— make me feel lonely autumn temple.

This page intentionally left blank



E B ’ ¯  W   P E: S  H Horikiri Minoru Translated by Cheryl Crowley

Scholarship on Bashp in Japan can be broadly divided into two

categories. The first kind includes biographical inquiry and documentary scholarship focusing on the various Bashp-related primary sources, and research that looks into the state of the haidan (haikai community) of Bashp’s day in order to ascertain his position in it. This is what is known as research exterior to Bashp’s literary works, or sociocultural research. The exterior approach is flourishing in Japan today; one outstanding example is the publication of Ptani Tokuzp’s Bashp zenzufu (Reproduction and discussion of Bashp’s Complete Calligraphic Works and Paintings, 1993).1 The second kind can be termed the interior approach; it is research into Bashp’s poetic expression, which looks into the ways Bashp’s works are read and identifies the characteristics of his poetry; ultimately and above all, it investigates the reasons why Bashp’s texts hold such fascination for us. Research into Bashp’s poetic expression can be further divided into two types. The first of these is that which isolates the distinctive features of poetic expression in the context of Japan’s Genroku period (1688–1704) while closely comparing the linguistic and cultural circumstances in which the works were created. It is, in other words, research that examines the characteristics of expression found in Bashp’s haikai. For example, in haikai there were improvised verses (sokkypshi), humorous verses (kokkei no shi), verses that functioned as greetings (aisatsu), and works composed in the context of a “za” (communal setting)—for this reason, it is often said that it is necessary to study renku (haikai no renga, comic-linked

160

H  M

verse) and the like. Also, there is research into the works’ linguistic phases: studies of words that were used in the haikai of the time, or the ways in which honkadori (borrowing from a source poem)—or parody—was employed, and so on. The second type is analysis of the unique structures of expression in Bashp’s works in the general context of poetics as a whole. An international, comparativist viewpoint is needed to do this. Recently, books such as Kawamoto Kpji’s Nihon shika no dentp (Poetics of Japanese Verse, 2000) and my own Hypgen toshite no haikai (Haikai As Expression, 2002)2 have taken a related approach, but this kind of research methodology remains rare in Japan, and is something that deserves more attention in the future. I would argue that there are three elements intrinsic to expression in Bashp’s haiku. The first is the so-called most basic characteristic of haikai, the element of kokkei (playfulness), humor, or wit. This is something that Bashp inherited from the haikai of his predecessors Matsunaga Teitoku (1571–1653) and Nishiyama Spin (1605–1682)3; it is laughter elicited by word play centered around paronomasia (kakekotoba—overlapping meanings or pivot words), parody, and mitate (metaphor). We see this in verses such as the following: sakura yori matsu wa futaki o mitsukigoshi

after cherry blossom season I waited three months to see the two trunks of these pines (Oku no hosomichi [Narrow Road to the Interior], compiled 1694)

tako tsubo ya hakanaki yume o natsu no tsuki

octopus traps: a fleeting dream under the summer moon (Oi no kobumi [Rucksack Notebook], 1690–1691)

In “after cherry blossom season” (sakura yori) the word for “pine” (matsu) is homophonous with the word for “wait” (matsu); the number “three” is juxtaposed with “two” and also the word for three (mi) is homophonous with the word for “see” (mi). In “octopus traps” (tako tsubo ya) Bashp connects the faintly comic fate of an octopus sleeping at the bottom of the sea to an awareness of the impermanence of all things. The second element is an expressive mode, which, by means of Bashp’s strong subjective discernment, decisively grasps the poetic object and projects an emotion onto it. This is yobikake (address)—a way of thinking through dialogue, expressing meaning through metaphorical exaggeration, or conceptualizing through the hypothetical subjunctive.

B   ¯ ’ W   P E uki ware o sabishigarase yo kankodori

161

make my melancholy even lonelier, cuckoo (Saga nikki [Saga Diary], compiled 1691)

Kisagata ya ame ni Seishi ga nebu no hana

Kisagata: Xishi sleeps in the rain amid mimosa flowers (Oku no hosomichi)

imo arau onna Saigyp naraba uta yoman

woman washing potatoes: were I Saigyp I might compose a verse (Nozarashi kikp [Record Of A Weather-Beaten Skeleton], 1685–1687)

In “make my melancholy” (uki ware o) the speaker addresses the cuckoo, asking it to intensify the feeling of wistful sadness in his heart; in “Kisagata” (Kisagata ya) the speaker compares the sight of rainsoaked mimosa blossoms to the appearance of the tragic Chinese beauty Xishi; “woman washing potatoes” (imo arau onna) is subjunctive, alluding to the incident when Saigyp (1118–1190) composed a poem to ask a courtesan for shelter from the rain.4 The third element is one in which Bashp implies jp (emotion) while describing keiki (scenery or environment), that is to say an image, especially a visual image. This is a mode of expression that is described as “environment first, then feeling” (keisen jpgo) or “form first, then feeling” (shisen jpgo). It includes verses such as the following: natsukusa ya tsuwamonodomo ga yume no ato

summer grasses: the aftermath of warriors’ dreams (Oku no hosomichi)

ara umi ya sado ni yokotau amanogawa

rough sea: reaching across to Sado Island the Milky Way (Oku no hosomichi)

In “summer grasses” (natsukusa ya), by writing of the landscape of the ancient battleground Hiraizumi that is lush with verdant summer grasses, he depicts the ephemerality of human life in a world of constant change. “Rough sea” (ara umi ya) conveys the loneliness of

162

H  M

humanity in the vast landscape of the Japan Sea, Sado Island, where many criminals were exiled, and the celestial “River of Heaven” (the Milky Way). In my view, out of all these three expressive elements, Bashp ultimately placed the most importance on the third one, which expresses meaning through images. I came to this conclusion after recently completing an essay on the overall structure of the haikai theories that Bashp espoused.5 I found that Bashp’s terms, such as the “ultimate principle of poetry” (fuga no makoto), “following Creation” (zpka zuijun), “changing and unchanging” (fueki ryukp), “original essence” and “original emotion” (hon’i, honjp), “quiet austerity” (sabi), “delicacy” (shiori), and “lightness” (karumi), are all connected to making emotion manifest through images. Bashp established his true literary style—the so-called Shpfu (Bashp style)—after his Oku no hosomichi journey in 1689. This was his karumi period, from the time of Hisago (Gourd, 1690) and Sarumino (Monkey’s Straw Coat, 1691), until Sumidawara (Charcoal Sack, 1694). Bashp’s disciple Kyorai (1651–1704) states this clearly in Kyoraishp (Kyorai’s Extracts, compiled 1702–1704): The Bashp school and other schools, above all, are different in how they conceive things. The Bashp school composes verses on environment and emotion together, just as they are; other schools seem to mainly focus on artifice.6

What makes the “Bashp style” fundamentally different from those of other haikai schools is that it frankly expresses keijp (environment and emotion) just as it is. This certainly is karumi’s season. The kei of keijp means scenery or landscape, expressing the concrete image of a thing just as it is, without clever artifice. In Chinese poetics we also find the phrase “unification of environment and emotion” (J. keijp itchi); basically it means expression in which landscape is depicted, charged with emotional resonance. Therefore, it is not a mere copy or sketch of a scene. Like the “summer grasses” and “rough sea” verses cited previously, they are expressions of environments that come to possess emotion of their own accord. Also, of course, many different qualities exist in the emotions that derive from their environments, but the one that Bashp most emphasized in his last years was ordinary feeling, feelings that can casually emerge from everyday life. This is what is called “zokutan heiwa” (lit., commonplace stories in ordinary language)—writing verse about everyday things in plain, easy words.

B   ¯ ’ W   P E shiodai no haguki mo samushi uo no tana

163

salted bream, cold up to their gums: fishmonger’s shelf (Komojishishu [Straw Lion Collection], 1693)

This verse describes an impression of bitter cold from the gums of salted bream lined up in a fishmonger’s stall. This is indeed one of the pinnacles of haiku as the poetry of ordinary people. In first touching on it, this quality of ordinariness was one of the things that Bashp’s disciple Kagami Shikp (1665–1731) most strongly emphasized as a way of teaching commoners; another was respect for the power of images, telling students to make the content of their verse just like a picture. From among all of those Bashp-style verses that show this regard for images, the ones I would especially like to focus on are those that rely on sound and aural imagery. There is a neologism derived from putting together the word “landscape,” which refers to visual scenery and “sound” to form “soundscape.” Soundscape is a word used in the book The Tuning of the World (1977) by Canadian musician R. Murray Schafer, but it is entirely appropriate for discussing poetry formed out of an “aural landscape.”7 Moreover, the Japanese are a people who from ancient times have appreciated soundscapes. In the famous Makura no spshi (Pillow Book, ca. 1000), Sei Shpnagon writes of how wonderful it is to hear the sound of the wind and insects on an autumn evening.8 Japanese people do not call small sounds such as the songs of insects oto (sound), but ne (sound). Sounds even fainter than ne, those that the human ear cannot actually perceive, such as that of a lotus blossom opening, are thought to sound like “pon” or “bassu.” One example is the verse by haikai poet Pemaru (1722–1805): hitori kiku toki ya hachisu no hiraku oto

when listening alone: sound of a lotus opening into bloom

Also, there is an Edo-period ukiyo-e called “Listening to Insects at Dpkanyama.” Dpkanyama is known to have been famous as a place for listening to natural “concerts” of insect sounds. Furthermore, Edo people enjoyed hearing the sound of wind chimes in the evening. Japanese gardens were provided with shishi odoshi (deer scarers), a device made of bamboo, which periodically strikes a rock, making a pleasant knocking sound. Also, according to the theories of the scholars Tsunoda Tadanobu and Sano Kiyohiko, Japanese, unlike people of other countries, perceive sounds such as the tinkling of streams and

164

H  M

the chirping of birds not with the right brain, which processes music, but with the left one, which processes language.9 In any case, the verses in which Bashp creates a poetic world by means of soundscapes are extremely numerous: shizukasa ya iwa ni shimi iru semi no koe

silence: rocks penetrated by the cicadas’ buzz (Oku no hosomichi)

This famous verse uses sound to depict the landscape of quiet Risshakuji Temple and the mountains of Yamadera. In this verse, the world of animals and plants is at one with the human heart. Bashp’s mind is at one with the mind of nature. No doubt people in Western countries think that cicadas’ songs are noisy and oppressive, but in this verse the cicadas’ song is the sound of life itself penetrating into nature’s cliffs. In the Zen master Dpgen’s teachings there is the phrase “body and mind drop off ” (shinjin datsuraku), which means breaking through the shell of body, mind, and ego, and achieving a state of mind that accepts things just as they are. This is what the verse “silence” (shizukasa ya) is all about; it is not simply a depiction of silence per se. I believe that it is an expression of a state of mind in which nature and the mind become one. furu ike ya kawazu tobikomu mizu no oto

old pond: a frog jumps in sound of water (Kawazu awase [Frog Match], 1686)

This is also a famous verse; its focus is without question the “sound of water.” Generally “mizu no oto” (sound of water) means the sound of flowing or dripping water, but because in this verse it is the tiny sound made when a frog jumps into water, it expresses limitless silence all the more. Some years ago, when the present Japanese emperor was the crown prince, I was one of several scholars invited to a talk on haiku and haikai, and at that time, sure enough, the subject of whether or not a frog jumping into water really makes a sound came up. One of the participating scholars declared that, in his own experience, frogs have a streamlined shape, and in fact do not make a noise. One might say that the sound of water that Bashp heard was not one audible to the ear, but one that was audible to the mind. However, I myself think that the sound of water was indeed faintly audible. Were we to compare Bashp’s ear to the modern-day ear, physiologically or

B   ¯ ’ W   P E

165

biologically, we would not see any differences in sensory function; but in terms of social structure and environment, we might think, could it be that our ways of listening to that sound are very different? Bashp wrote many verses that refer to the songs of living beings in the natural world such as birds and insects, the sounds of natural phenomena such as wind, rain, rivers, and the like, the various sounds of daily life that arise as part of people’s lives, human voices, musical sounds, and so forth. According to my count, the number of Bashp verses that use aural imagery amount to 110 out of more than 980. Surprisingly, Bashp sometimes used things that make no sound as part of his soundscapes: minomushi no ne o kiki ni koyo kusa no io

come listen to the sound of a bagworm— my thatched hut (Zoku minashiguri [Sequel to “Empty Chestnuts”], 1687)

In this verse the speaker calls to an addressee to come to his hut and join him in savoring the autumnal atmosphere while listening to the sound of a bagworm; but, in fact, bagworms do not make a noise. However, it was considered amusing because of its tradition as a literary conceit, as in Makura no spshi (Pillow Book) there is the phrase “[the bagworm] crying plaintively ‘Milk! Milk.’ ”10 Also, kumo nani to ne o nani to naku aki no kaze

spider, with what sound and for what are you crying? autumn wind (Mukai no oka [Yonder Hill], 1680)

In an autumn wind, the speaker wishes that the voiceless spider would find a voice. The verse in which this kind of imaginary auditory sense appears most strongly is surely the following: yoru hisoka ni mushi wa getsuka no kuri o ugatsu

at night, in secret, beneath the moon, an insect burrows into a chestnut (Azuma nikki [Eastern Diary], 1681)

It describes hearing an insect assiduously digging a hole in a chestnut on a night of bright moonlight, but this too is certainly imaginary. We might well call it unusually sensitive nerves.

166

H  M

Modern culture has overwhelmingly developed a focus on visual images in preference to those of the other senses. However, hearing, no less than sight, offers rich imagery. There is a long history of sounds such as the quiet patter of the rain, the rustle of autumn leaves, and the lapping of waves on the shore being used to create magnificent poetic worlds. And perhaps we could say that within that history, Bashp was a poet who expressed the soundless world of quietude by means of the world of audible sound.11 So far I have focused on visual and aural images as important aspects of the special characteristics of Bashp’s poetic expression. In the future, I would like to do a more detailed stylistic analysis of the 5/7/5 syllable expressive structure of haikai and haiku—for example, the function performed by “ya” (an emotive particle) as a means of creating structure. Also, in the past, in an essay about the structure of Oku no hosomichi, I wrote that its organization was exactly that of a linked verse sequence, unfolding like an illustrated handscroll.12 I wrote that Oku no hosomichi is not just the narration of a freewheeling journey, rather, it has many shifts in “focal points of meaning,” from meetings and courtesies between people to visits to cultural sites made famous by tradition, and this is how it develops. However, recently I read Eleanor Kerkham’s essay, “The Construction and Climax of Oku no hosomichi” and was quite impressed. Professor Kerkham argues that Oku no hosomichi, like the design of a Buddhist mandala, is made of overlapping layers of multiple worlds.13 In Japanese scholarship, essays such as this, which consider the structure of works, are extremely few, but indeed a few years back, an essay written by Japanese and American comparative literature specialist Makoto Ueda, “Taxonomy of Textual Structure,” which was published in The Structure of Japanese Literature (Nihon bungaku no kpzo), addressed issues such as the “multilayered narrativity” characteristic of Japanese literature, and it is extremely useful.14 I myself would like to undertake research from this angle in the future. Also, I am interested in confirming that haikai is, more than anything else, a form of culture that prevailed in the context of the daily lives of ordinary people, and I am doing research on this accordingly. This is partly because today fine Japanese traditions continue to be destroyed by karaoke. However, above all, what I most want to do is to investigate the expressive world of the text. For this reason, it is essential that I pursue my research in cooperation with American and European scholars.

B   ¯ ’ W   P E

167

N This translation of a talk presented at the University of Maryland, November 5, 1994, has been slightly revised with permission of the author. Revision is based in part on subsequent publications of the talk in Japanese; see Japanese Education News no. 94 (1995): 7–10 (Tokyo: Kadokawa Shoten) and Horikiri Minoru’s Bashp no saudosukeepu (Tokyo: Perikan, 1998), 293–303. See also note 10 later. The review, in note 10, of Professor Horikiri’s presentation and discussion of several of Bashp’s “soundscape” hokku is presented by the translator with permission by the author. 1. Tokyo: Iwanami Shoten, 1993. Edited by the Society for the Publication of the Complete Calligraphic Works and Paintings of Bashp under the supervision of Otani Tokuzp and including the input of ten of the most renowned Bashp scholars, this extensive and authoritative work includes reproductions and commentary on Bashp’s tanzaku (poem cards) and kaishi (poem sheets), haiga (haikai paintings), and haisho (haikai talks). 2. Kawamoto, Nihon shika no dentp (Tokyo: Iwanami Shoten, 1991); English translation published as Poetics of Japanese Verse (Tokyo: University of Tokyo Press, 2000); Horikiri, Hypgen to shite no haikai (Tokyo: Iwanami Shoten, 2002). 3. Waka, renga, and haikai poet Matsunaga Teitoku, was founder of the Teimon School of haikai. Nishiyama Spin (1605–1682), renga and haikai poet, was the founder of the Danrin school (translator’s note). 4. This exchange is included in Saigyp’s Sankashu (Poems of A Mountain Home): “On the way to the temple called Tennp-ji, I got caught in the rain. In the area known as Eguchi I asked at one place for a night’s lodging. When refused, I replied as follows: yo no naka o itou made koso katakarame kari no yado o oshimu kimi kana

It is hard, perhaps, To hate and part with the world; But you are stingy Even with the night I ask of you, A place in your soon-left inn.

The response by a ‘woman-of-play’: ie o izuru hito to shi kikeba kari no yado ni kokoro tomuna to omou bakari zo

It is because I heard You’re no longer bound to life As a householder That I’m loath to let you get attached To this inn of brief, bought, stays.”

Translation in William Lafleur, The Karma of Words: Buddhism and the Literary Arts in Medieval Japan (Berkeley: University of California Press, 1983), 70–71 (translator’s note).

168

H  M

5. See my “Bashp hairon taikeika no kokoromi” (An attempt at systematizing Bashp’s poetic theories) in Ferisu Jogakuin Daigaku spritsu sanju shunen kinen bunshu (Essays commemorating the thirtieth anniversary of the founding of Ferris University), April 1995. 6. Kyoraishp hypshaku (Training). See Okamoto Akira, Kyoraishp hyphaku (Meichp Kankpkai, 1970), 284–285. 7. R. Murray Schafer, The Tuning of the World (New York: Alfred Knopf, 1977); for Schafer’s definitions of “soundscape” see 7–10 and 271–275. 8. See, for instance, Ivan Morris, trans., The Pillow Book of Sei Shpnagon (Harmondsworth, Middlesex: Penguin Books, 1967), 21: In autumn the evenings, when the glittering sun sinks close to the edge of the hills and the crows fly back to their nests in threes and fours and twos; more charming still is a file of wild geese, like specks in the distant sky. When the sun has set, one’s heart is moved by the sound of the wind and the hum of the insects. 9. Tsunoda Tadanobu, Nihonjin no np (Tokyo: Daishukan Shoten, 1980); Sano Kiyohiko, Oto no bunkashi: Tpzai hikaku bunkakp (Tokyo: Yuzankaku shuppan, 1991). 10. See Morris, The Pillow Book, 69–70: I feel very sorry for the basket worm. He was begotten by a demon, and his mother, fearing that he would grow up with his father’s frightening nature, abandoned the unsuspecting child, having first wrapped him in a dirty piece of clothing. “Wait for me,” she said as she left. “I shall return to you as soon as the autumn winds blow.” So, when autumn comes and the wind starts blowing, the wretched child hears it and desperately cries, “Milk! Milk!” 11. For an extended essay by Professor Horikiri on “Bashp’s soundscapes” in which these verses are discussed, see his Bashp no saundoskeepu, 7–91. “Bashp’s Soundscapes” is the lead article in a collection of essays on Bashp and Buson published previously in various scholarly journals between 1983 and 1996. Dawing on a broad range of Japanese and Western scholars who have thought about the activity of “hearing” and “listening” and who, like Schafer, have attempted to define spaces by the human and/or natural sounds that “mark” them, Horikiri examines all of the hokku by Bashp that he sees as “soundscape poetry.” He includes in his analysis not only a great many other, similar verses by Bashp’s disciples, but also many single two-verse links from key Bashp-school linked verse (editor’s note). The following are additional examples of soundscape hokku from Professor Horikiri’s article “Bashp’s soundscapes,” 31–40 (translator’s note). Drawing on the research of Tsunoda Tadanobu, Professor Horikiri argues that Japanese people process natural sounds with the same part of

B   ¯ ’ W   P E

169

their brain as they use to process language, i.e., the left brain, and for this reason Bashp was extremely receptive to the emotional content of sounds in his environment. This following verse is an example: arare kiku ya kono mi wa moto no furu kashiwa

listening to the hail: myself, the same as ever, an old oak

Its headnote is “On the construction of a new Bashp Hut.” The speaker’s experience of the sound of hail triggers the feeling of relief that despite the change of residence, his sensibilities remain exactly the same as they used to be. Likewise, in the following verse: ikameshiki oto ya arare no hinoki kasa

with a powerful noise, hail falls on my cypress travel hat

The sound of hail on the speaker’s hat of woven cypress is inextricably linked with renewed awareness of the melancholy circumstances of life on the road. The same is true in this verse, whose headnote is, “in memory of a certain person”: uzumibi mo kiyu ya namida no niyuru oto

the banked fire also goes out—tears boil away with a hiss

The startling sound of water striking coals in a fireplace is a poignant expression of the intensity of the speaker’s grief. Professor Horikiri writes that in these verses, the sounds are received by the same mental faculties that process language, and that the emotions described unconsciously are processed intellectually and logically. Other verses that rely on this kind of left-brain thinking are the following: fuyu niwa ya tsuki mo ito naru mushi no gin

garden in winter: the moon dwindled to a thread too insects’ chorus

This verse juxtaposes an image of the moon and the nocturnal sounds of insects in the setting of a barren winter garden. They are both metaphorically and rhetorically connected by “ito naru,” becoming thin as a thread. The speaker sees the wasted-looking moon at the same time as hearing the chirping of insects late in the year—formerly worthy of the word “chorus” but now reduced to a sound as thin as the silver of the moon itself. The ironic linkage of the short life of insects with their song is also the focus of the next two verses: yagate shinu keshiki wa miezu semi no koe

they cannot see that they soon will die— cicadas’ song

muzan ya

“What a tragedy!”

H  M

170 kabuto no shita ni kirigirisu

underneath the helmet, a katydid

The first acknowledges that the life of the cicada is brief, however lustily it sings. The second draws on a story from the Heike monogatari (Tale of the Heike), in which elderly Saitp Sanemori dyed his hair black in order to hide his age and went into battle against the Minamoto. When his helmet was removed after his death, his enemies’ triumph turned to pity when they realized they had slain an old man. In this verse, the tragic warrior is replaced by a katydid, but the katydid’s voice, like that of the cicadas in the previous verse, is a reminder of how soon life comes to an end. In the next two verses, the human voice is treated as something to be processed with the left brain, much as a Japanese person would hear the sound of insects or animals according to Tsunoda’s thesis: hitogoe ya kono michi kaeru aki no kure tsuke mo ugoke waga naku koe wa aki no kaze

people’s voices: coming home on this road autumn nightfall O grave—you too, move! my grieving voice on the autumn wind

In the first one, the sound of the human voice is not something that offers information or meaning, but is rather a feature of the autumn dusk that only serves to heighten the atmosphere of quiet and loneliness. In the second, the sound of “my grieving voice” blends with that of the “autumn wind,” but it is not the words of the speaker; it is, rather, the plaintive sound of his voice itself, which like the wind conveys intense and penetrating sorrow. In the following verse, one variety of human voice, snoring, engages both left and right sides of the brain when introduced in a scene whose origins derive from art and religion: tsuki ka hana ka toedo Shisui no ibiki kana

you may ask, “do you prefer the moon or the blossoms?” but the Four Sleepers snore on

The “Four Sleepers” are iconic figures of Chan (Zen) Buddhism: Han Shan, Shi De, Feng Gan, and Feng Gan’s tiger. The four of them huddled together in sleep were a popular subject among Chinese and Japanese painters. Their carefree pose conveys their achievement of spiritual realization; asking them to judge between the beauties of autumn and spring (a standard poetic conceit) will earn nothing but an indifferent snore in return. Professor Horikiri then takes up the theories of Yoshida Hidekazu comparing Western and Japanese ways of perceiving musical sounds. Yoshida argues that Japanese process music played on instruments such as the flute, koto, and shamisen with the left brain. This is because, he explains, Japanese music harmonizes with the sounds of nature. Professor Horikiri

B   ¯ ’ W   P E

171

explores how this might apply to Bashp’s poetry, and analyzes the following verses: Biwa Kp no yo ya shamisen no oto arare Sumadera ya fukanu fue kiku koshita yami

a night out of “Song of the Lute” the shamisen sounding like a hailstorm Sumadera temple: listening to a silent flute darkness under the trees

In “A night out of ‘Song of the Lute’ ” (Biwa Kp no), the speaker confuses the sound of the shamisen with that of falling hail, then it makes him think of the classic Bo Juyi poem “Song of the Lute” (Pipa xing) about a female entertainer sadly longing for her glorious past. In “Sumadera temple” (Sumadera ya), the silent flute of Heike monogatari hero Atsumori reminds the speaker that the vanity and splendor of the powerful inevitably fade into ruin. 12. See my essay “Oku no hosomichi o renkuteki ni yomu––kpseiron e teigen,” Edo bungaku 1, no. 3 (1990): 74–93. 13. Eleanor Kerkham, “Oku no hosomichi to sono kuraimakkusu,” in (ed.) Matsuda Yoshiyuki, Dewa Sanzan to Nihonjin no seishin bunka (The Sacred Mountains of Dewa and Japanese Spiritual Culture) (Tokyo: Perikan, 1994), 145. A scholar who presents an extended analysis of Oku no hosomichi juxtaposing a theory of the “mandalization” of the geography of Japan upon the course of Bashp’s historical journey and onto that of the literary work Oku no hosomichi itself is Takeshita Kazuma. See, for instance, his Oku no hosomichi no kpzp to shinjitsu: Bashp bungaku no nazo o kaku kagi wa nani ka (Tokyo: PHP Kenkyusho, 1986). Kerkham’s discussion of Oku no hosomichi ’s structural “overlapping layers of multiple worlds” as a metaphor for the Buddhist mandala is not comparable to Takeshita’s imaginative and detailed juxtapositions. It refers rather to Oku no hosomichi’s complex, three-dimensional movement back and forth through past and present time and space and the meditative structure of the narrator’s (the traveler’s) mind (editor’s note). 14. Konishi Jin’ichi and Nakanishi Susumu, eds., Nihon bungaku no kpzo (Tokyo: Spjusha, 1987).

This page intentionally left blank



A  U T E   M A’ W A R E M O M ATA , OKU NO HOSOMICHI Eleanor Kerkham

M

ori Atsushi (1912–1992), novelist and literary essayist, published in 1988 a lyrical study of Matsuo Bashp’s masterpiece, Oku no hosomichi, entitled Ware mo mata, Oku no hosomichi (And Me Too, Once Again, Into Oku no hosomichi).1 The monograph can be seen as part of the 300th year memorial celebration of Oku no hosomichi described in the introduction to this volume. As Mori suggests in his opening chapter, his study represents a personal reaction to the public expectation that he, the “guest traveler” and “commentator” in a 1984 five-part NHK docudrama on the Oku no hosomichi, must also be an expert on the literary work and could thus be asked to perform during the many 1988–1989 Oku no hosomichi memorial events. Mori tells us, however, that he had had no special relationship with Bashp or with the Oku no hosomichi. He allows, nevertheless, that while he is not a poet and had not spent his life in travel, his method of work might be seen as a possible link with the earlier poet. When at work, it may indeed seem that he, like Bashp, was “off on a journey.” Both were creative artists and both brought into being new literary worlds—worlds, as Mori sees them, into which they might lure their readers. Mori’s readers know too that he, like Bashp, adopted special relationships to the spaces that are a part of his literary worlds, and that a crucial one of these areas for both was Gassan (Moon Mountain) and the two smaller peaks it overlooks, Mount Yudono and Mount Haguro, or the “Three Sacred Mountains” in Yamagata Prefecture. Although these “links” to Bashp may account in part for

174

E     K     

why Mori was chosen to host the Oku no hosomichi documentary, it did not necessarily follow, he tells us, that he might thus be expected to have and to publicly present his own interpretation of Bashp’s famous travel journal. The public was expecting just this, however, so Mori went back to the Oku no hosomichi and to his memories of time spent in key spots along its route to work out the interpretation he was supposed to have. Mori assumed that if he set himself the task of returning to, carefully rereading, and then picking up his pen to write about the Oku no hosomichi, then the creative result would be the personal “understanding” that the city and prefectural officials and a popular audience were expecting. He trusted that the process of writing would reveal to him what he thought and felt, and that in the act of writing he would discover both his subject and its relationship to himself. This assumption helps explain some of the nuances in the title Mori chose for his book: Ware mo mata, Oku no hosomichi (And Me Too, Once Again, Into the Oku no hosomichi). The title identifies the work as autobiographical. A chapter, at least, of his own “story” would emerge in the act of reliving and writing about a chapter (or chapters—Bashp’s journey and his composition of the Oku no hosomichi) in the life of another artist. Having focused, in his well-known novel Gassan, on the Dewa Sanzan or sacred mountains area that represents a highlight of Bashp’s journey and of the Oku no hosomichi, and having lived in and written about Sakata, Fukuura, and other towns along the Japan Sea coast traversed by Bashp and Sora, when Mori set out with the NHK crew in 1984 from Tokyo’s Fukagawa or the site of the Bashpan, he was literally going back once again—mata. Likewise, when he sat down with his memories of this and other personal experiences, with the Oku no hosomichi text, with Sora’s record of the journey, and with modern commentaries on the texts, he was returning a third, fourth, or any number of times to the Oku no hosomichi and to his own past. As Mori and other scholars have pointed out, Bashp’s journey and his poetic journal might also be seen as a return to those sacred spaces that had “housed” earlier travelers, the tragic warrior, Yoshitsune (1159–1189), or Bashp’s favorite poet Saigyp (1118–1190), whose death five hundred years earlier corresponded roughly with the year of Bashp’s journey. My reasons for examining a work in Japanese that may only be available in East Asian libraries are twofold. One is to try to reveal the imaginative way in which the work, in its theory, structure, and methods of presentation, enlivens Bashp’s masterpiece and becomes part of its circular, concentric lines of respect for great wanderers/poets of the

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

175

past. Another is to make available for application and further thought an inventive set of ideas about a literary masterpiece for which we have multiple and excellent English translations, which is read and taught often, and which has the power to reward a rich variety of different interpretations and approaches. I examine first the critical contexts from which Mori operates and the organizing structures he postulates as he takes us on a lively excursion through the Oku no hosomichi, both as a physical journey and as a literary masterpiece. I discuss next the stylistic and structural “intersections” or “correspondences” (Mori uses the term, taip) that he sees as characterizing the work’s movement and source of energy. As sketched out in his brief opening section, Mori’s theory of literature resembles what might be seen as a classic structuralist approach. He assumes that the function and purpose of literature is not to mirror the world, but to provide an imaginative recreation of it. The writer creates an artistic or literary world and “charms” his or her readers into becoming a part of it.2 In order to do this, Mori says, an author must provide this imaginative realm with a coherent philosophical vision, a satisfying rationale or theme (he uses the term meidai, “proposition” or “thesis”)—a unifying theme that major elements of the work support.3 To provide a vision of a world wherein one’s readers will feel they have a place or one that will eventually have the power to “tightly enclose” (mippei) the reader, the artist must create a multilayered literary structure, one into which the reader might gradually be ever more deeply drawn. Bashp accomplishes this task, Mori explains, by creating within his work a complex series of taip. Mori’s reading of the Oku no hosomichi makes clear that he means by taip intersections between two entities within which correspondences or oppositions appear. He describes taip as something akin to fields of tension and balance between all manner of concrete and abstract things: between words, phrases, or images, between motifs or themes, between human beings, ideas, places, poems, and most importantly between Bashp’s text, the Oku no hosomichi itself, and other literary texts of the past—earlier waka, the np drama, narrative works such as Tales of the Heike, folk legends, Chinese prose and poetry, or earlier works by himself and his disciples. The constant inner movement caused by these correspondences contained within a carefully constructed, many-layered structure allows Bashp’s imagined construct to become, in Mori’s words, “a living world in which real time flows.”4 Theorizing further on Oku no hosomichi ’s structural and stylistic intersections and their complex relations, Mori argues that as haikai

176

E     K     

master and almost as an unconscious phenomenon, Bashp makes use in his prose of all of the “associative” strategies and techniques of the art of haikai. A linked verse moves because of lines of association reaching backward and forward from verse to verse. To create associations is to set up correspondences (taip). States of tension and balance are achieved and then broken against a background of newly formed balances. Mori stresses, however, that the balances Bashp sought in his world were not “balance as equilibrium” (in Mori’s words, matching yin to yin or yang to yang) but “balance as antithesis” (allowing yin to stand beside yang). Rereading Oku no hosomichi in much the same way Leopold Hanami suggests we might reread renga or haikai, Mori sees the lines of association moving not just between contiguous sections (individual haibun) in the Oku no hosomichi, but among disparate haibun in the work as a whole. One half of an opposition may be sounded and seem to disappear, only to reappear many passages later as it corresponds with its other half.5 Furthermore, in yoking opposites (yin and yang) a transformation inevitably occurs; a new something is always coming into being. The constant operation of this principle—the interplay of oppositions—inevitably produces appropriate transformations or brings about the “ten thousand changes” (again, to use the words of Chinese cosmology that Mori borrows to explain his theory).6 Change as the fundamental law of existence is thus established as the basic conceptual frame within Bashp’s literary world. Carefully constructed comparative and contrastive correspondences are always at work within a structure that lures the reader into deeper and deeper levels of experience. Mori suggests that this law of change proceeds in what he presents as a “natural” pattern: there is first an origin (ki, a “beginning to be” something), which is then matched with (taip suru) its “suitable response” (shp, “reaction” or a “receiving”); ki and shp are, in turn, matched by a third unit, the “transformation” (ten), a unit in which a change occurs or new elements appear; and finally all three—ki, shp, and ten—are matched by a conclusion or ending (ketsu), a wrapping up of the whole.7 In postulating this ki-shp-ten-ketsu pattern alive as the structure of the Oku no hosomichi, Mori again borrows from the Chinese. Ki-shp-ten-ketsu are the designations given for the “rules” of composition of the four- or the eight-line regulated Chinese verse. The four are normally described as representing, roughly: an introduction of theme, its elucidation or development, a change, and a wrapping together. The ki-shp-ten-ketsu structure is discussed most often in connection with the structure of the Chinese four-line verse, the jue-ju (zekku in Japanese, “cut off lines”), in which the function of

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

177

each of the four lines of the verse is described: a theme or vision is “established” in the first line; the second “receives” and develops this; the third brings a transformation or turn; and the fourth “binds” or resolves all into an ending or closure.8 In much the same way that medieval Japanese renga theorists argued that the traditional jo-ha-kyu (introduction/development/ swift ending) pattern was a natural structural principle at work out in the world and in artistic compositions (it was the way a flower grows, a bird sings its song, or a np drama and linked verse move), Mori seems to see this four-part structure as a natural principle at work in the Oku no hosomichi.9 It is Bashp’s four-part ordering of his imaginative world—an organic, living frame or “anatomy” in which the poet moves us from a formal introduction into his world of travel, establishing the “vision” that informs it and the themes it will explore (its ki) to an acceptance and further sounding of these themes (shp); into a transformation ten, in which the traveler reviews, broadens, and recreates his original vision; and on to a quickened “return and closure.” Although Mori does not try to provide a detailed discussion or “proof” of his four-part division or of its relationships to the work’s thematic development, he does provide an intuitive, poetic sort of logic of movement as he leads his readers through the entire text. His divisions are broad enough to allow us to accept at least provisionally that the work is indeed ordered around this or other similar principle. Furthermore, his skillful discussion of the workings of taip provides insight into the ways in which parts of the work sing and echo with one another like an intricate piece of music. Mori argues that the ki-shp-ten-ketsu principle shapes the work as follows. After the formal, philosophical opening passage (Mori, Ogata Tsutomu, and others call it the work’s “overall introduction,” jo, and Mori sees it as the author’s separate aisatsu or “greeting” for the whole), Bashp’s ki (the narrator’s establishment of his world and sounding of its main themes) extends from the poet–traveler’s leavetaking from friends and disciples in the second passage up to the Shirakawa Barrier crossing or the traveler’s visit to Saigyp’s willow and the village of Ashino. While Mori does not try to justify his divisions in terms of the movement of the mind of the traveler, one does not need argument here to read each of the sections up to the Shirakawa Barrier as being introductory and preparatory in nature. The narratorprotagonist and his companion are prepared for the journey physically, emotionally, and spiritually with each step toward the Barrier. In these first eleven or twelve passages the narrator “introduces” himself and his traveling companion and describes their preparations for

178

E     K     

potential hardships and for their poetic and spiritual journeys. At the same time he prefigures a major theme—the traveler’s search for the spirits of the poet Saigyp and the warrior Yoshitsune and for the spiritual purity and power of such persons as Kukai (presented as founder of temple and sacred mountain at Nikkp), En no Gypja (traditionally a founder of Shugendp, the yamabushi [mountain priest] tradition of travel and mountain worship) at the Shugen Kpmypji temple, or the traveler’s friend and teacher, the priest Butchp Oshp (1643–1715), both part of the Kurobane visit. Finally, the Shirakawa Barrier represents the entrance into Pu—the provinces of Michinoku and Dewa or deep into the north—the traveler’s stated goal. Mori links the fact that he selects this point in the text as the place to end Bashp’s ki section to his own interpretation of the hokku, which closes the Ashino passage: ta ichimai / uete tachisaru / yanagi kana (one field / planted and we leave / the willow). Different interpretations of the verse revolve around the identity of the subjects for the two verbs, uete (plant) and tachisaru (take one’s leave). Women planting rice seedlings plant and go away; the narrator and Sora watch the planting and then go away; the travelers plant a portion and go away; the travelers sit and watch the women plant, and they all go away. In each reading, of course, the willow doesn’t move. It was there when Saigyp visited five hundred years earlier and is still planted in the field. Mori does not make a choice among these, but he adds to the possible complexity by pointing out that the local official who invited the travelers to see the willow had, by the time Bashp composed Oku no hosomichi four or five years after the journey, died and gone away. Butchp, whose hut the travelers visited two passages earlier, had left the hut and the area; likewise, the spirit of the willow tree in the np play Yugyp Yanagi mysteriously appears, dances, and fades away. All of these elevate the travelers’ move away from the tree into an act symbolizing their own perpetual departures and eventual deaths. The reason the travelers come and go away from the willow in this manner is of course because Saigyp had once stopped beside the tree and had not, his verse would have us believe, easily gone away: michinobe no / shimizu nagaruru / yanagi kage / shibashi tote koso / tachidomaritsure (“Just a brief stop,” / I said when stepping off the road / into a willow’s shade / where a bubbling stream flows by, / as has time since my “brief stop” began).10 Mori thus argues that the feeling in this passage of a departure, of people and spirits leaving a specific area, is strong. As an equally definitive opening to the next portion of the journey, the traveler tells us that he moved through the Barrier with a new consciousness—the feeling that he is finally truly into his journey.

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

179

The shp or “appropriate response” section extends from the Shirakawa Barrier crossing through the visit with the painter Kaemon in Sendai on the Miyagi Plain. Mori points out that the first passage of the shp, the Shirakawa crossing, is, like the work’s opening passage, formal, highly stylized, and allusive (playing with colors, with the four seasons, and with a brocade of phrases from earlier prose and poetry). His explanation for the break on the Miyagi Plain seems to be that with the Sendai passage the traveler is once again (after, presumably, having reaffirmed his search for the spirits of Saigyp and Yoshitsune and other souls of the past and after having experienced, at Iizuka, doubts about, but then acceptance of his vulnerable body and his sometimes shaky spiritual goals) sent off by his new friend with a gift of walking sandals, the traveler’s home, decorated with the season’s ritual iris-purple. The important ten or transformation section begins, then, with the travelers’ passage along the Oku no Hosomichi, the designation given to a portion of the road from Sendai en route to the Tsubo no Ishibumi (the oldest monument the travelers encounter). It is here, in the words chiseled on the stone, that the traveler feels he sees into “the hearts of the kojin (great persons of the past).”11 This encounter gives the narrator the strength to redefine his goals and to proceed with a new consciousness of his role as poet and discoverer of the past. The ideas expressed by the narrator in this passage “correspond with” (taip suru) and then “transform” his earlier philosophical but more negative musings at Iizuka. The traveler has felt a glimmer of the eternal ( fueki) and this further stimulates his journey’s search.12 The ten section continues through what Ogata Tsutomu and other scholars have seen as the work’s high points: Shiogama, Matsushima, and Zuiganji; Hiraizumi and the crossing into Dewa; the visit with the merchant Seifu in Obanazawa; discovery of the Risshaku-ji and ride down the Mogami River; and finally the crucial pilgrimages to the Three Sacred Mountains and walk from Sakata to Kisakata and back. It is not difficult to read this significant portion (in terms of the number of hokku composed, amount of time spent, and even insights gleaned) of the Oku no hosomichi as transformative and recreative in nature. One can easily argue, with Mori and others, that something new and significant “happens” to the traveler during this portion of the trip, and that something “ends” when he leaves Sakata, the province of Dewa, and the sacred mountains. Mori’s ketsu or “ending” corresponds with what scholars who postulate a renku-like three-part jo-ha-kyu (introduction, development, ending) division for the Oku no hosomichi have describe as the

180

E     K     

work’s kyu or fast ending: the often hurried walk down the Echigo-Echizen road along the Japan Sea coast and into Pgaki. This portion of the journey sees a fainter sounding of themes, a gradual lightening of tone, and a movement toward temporary closure with the traveler’s return to old friends and final departure for the Ise Shrine. Although other equally coherent theories of Oku no hosomichi’s overall structure have been and will continue to be argued, Mori’s analysis is thought provoking. More discussion from Mori on the relationships between this structure and the development of recurring themes or between this structure and the movement of the mind of the narrator would have been welcome. Mori’s goal as a creative writer seems to be to lure his readers back into a layered, three-dimensional Oku no hosomichi. His ongoing discussion of the structurally and stylistically interrelated taip within this four-part division elicits an awareness of a complex structure and a high degree of formal coherence, which is indeed not obvious if we do not enter with him into the work. Mori not only adopts the ki-shp-ten-ketsu framework for his reading of Oku no hosomichi, but also imposes it on his own book as well. He divides Ware mo mata, Oku no hosomichi into these four parts and proceeds within them on two distinct levels. The first is the dramatization of his own experiences at those spots chosen by the NHK documentary director as most “representative,” or perhaps most accessible today, of Bashp’s Oku journey. The narrative of Mori’s journey through contemporary Japan is most interesting to the foreign reader for what it reveals about the continuities and changes that have occurred over this momentous three-hundred-year period. In a city like Sendai there would appear on the surface to be few spots worthy as monuments of the journey. Memories do remain a vital part of the life of the city, however. The NHK research staff managed to find an impressive variety of people in Sendai and elsewhere whose depth of concern for the values and traditions Bashp represents was genuine and moving. It is interesting to discover what is and is not preserved of the historical Bashp’s world and to see how the present keepers of tradition—the priestly families, city officials, and ordinary householders—have made it their responsibility (as Bashp had done) to preserve their own versions of the past. Mori’s descriptions of the people and the landscapes, temples, and shrines, which retain their mystery and beauty today, are masterful, and they too serve to enrich our own subsequent personal readings of the Oku no hosomichi. The second level on which Mori proceeds is the presentation of his inventive reading of the text. As mentioned, his goal is to follow the complex intersections that link different pieces of the work into what

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

181

Mori sees as a rhythmical, evolving three-dimensional whole. His primary critical task is to describe how the Oku no hosomichi moves, and his reading thus focuses on a progressive uncovering of the taip that he believes activate its movement. As sketched in his theoretical opening and worked out in his analysis of the workings of different sorts of intersections, Mori repeatedly points to the fact that the frame of reference for Bashp’s poetic language is not that empirical reality of the actual journey of 1689 but the world of poetry and legend—an imaginative literary realm that the Oku no hosomichi then becomes part of. Thus the critical context that grounds his reading and defines his basic orientation toward Bashp’s masterpiece is poetry—or the classical poetic, religious, and folk traditions—that is to say “literature” more broadly, as Bashp knew it. It is interesting to a non-Japanese reader that Mori decides not to present the original text of the Oku no hosomichi in his work. Rather, he presents and even bases his critical comments on his own translation into modern Japanese of the prose sections. He does allow the hokku to speak for themselves, however, not presenting modern renderings of the originals. This method assumes, perhaps, that most readers will have Bashp’s original in their laps or maybe even in their heads. It may also be designed to call attention to the fact that the book represents Mori’s own personal “return.” Bashp’s art is deeply embedded in his words and in the differences between his and the ordinary (much less the modern) language, so not having his words in the text was initially somewhat disconcerting. The problem is one every translator must face. Certain aspects of a literary work can, in the hands of a skillful translator, be recaptured—elements of structure or narrative method, certain images, metaphors, and symbols, for instance, or techniques of character and thematic development. Thus correspondences that function on these levels can be effectively discussed in the context of a translation into modern Japanese. Other linguistic and literary conventions, however, the use of irony, the compression of language, or the rhythm and tone of Bashp’s style, cannot always be captured. It is when Mori discusses correspondences on this level that not having Bashp’s original text becomes something of a liability. Mori weaves into his translation information that Bashp does not supply, including amplification of the narrator’s thoughts and feelings and facts about the actual historical journey—facts gleaned from Sora’s diary and from other contemporary and modern sources. This is the method used by many scholars presenting a modern Japanese rendering of the original classical Japanese, and it assumes that one’s audience wants to know not what Bashp said (they can see this in the

182

E     K     

original) but what Bashp meant, or what the scholar in question believes that Bashp meant, to say with his highly allusive, poetic language. If it were only this—a matter of interpreting Bashp’s text— then we could say that every translation is an interpretation, and that having Mori’s is very interesting. The danger with this approach, however, is that the translator might provide details of the real journey that have little to do with the Oku no hosomichi as literary text; and even that we all might lose sight of the critical object of study (is it Bashp’s life and the 1689 journey? or is it the Oku no hosomichi as literary text? or a text in its social/historical context?). When Mori amplifies his translation with names of specific real places, people, and events gleaned from letters, Sora’s diary, and notes, he is, of course, changing its nature. Part of Mori’s dilemma was probably that his objects of inquiry were two, the 1689 journey and the literary work written five years later. If Mori is taking the author’s life as his critical context—that is to say if his critical approach is based on the belief that the key to the meaning of the Oku no hosomichi lies in the meaning Bashp meant it to have—then he should have examined more closely Bashp’s life and thought after completing his journey and particularly after his return to Edo, 1692–1694. Although Mori’s translation may be more autobiographical than the original, when in his commentary he compares Sora’s accounts of the journey with Bashp’s—that is to say, when he asks himself whether or why there are “fictional” elements in Bashp’s text—he is not doing so to “prove” or “disprove” the veracity of Bashp’s “version.”13 Rather, he is using what he thinks we know of the actual journey or “reality”—represented by Sora’s records—as a sort of first version of the literary narrative of the journey. Bashp’s “changes”—the differences between Sora’s version of what “happened” and the narrative action of the Oku no hosomichi—might be seen as “revisions”, and we can glean from comparisons of the two some information about Bashp’s creative process. Mori assumes that Sora’s rendering represents a way Bashp first remembered his journey. When the poet chose to present events in a different manner, then he was revising, and in a poet’s revisions we can see revealing artistic choices. Mori always asks why, given the special literary world Bashp is constructing, certain changes were made. His answer to this critical (not biographical) question is invariably, “because Bashp was intent upon the creation of certain correspondences.”14 This approach always pulls Mori back into his primary concern, the Oku no hosomichi as a separate literary world. A final aspect of Mori’s approach to the rendering of the Oku no hosomichi into modern Japanese—one that may again be designed to

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

183

serve as aid to the modern reading public—is his addition of original texts of the waka or of translations into modern Japanese of relevant portions of other earlier literary pieces to which Bashp so often alludes. This characteristic too tends to keep Mori committed to his intertextual approach. In adopting this method Mori has to make many crucial interpretative choices. He cannot present all of the work’s to which Bashp may be alluding and thus must commit himself to one or two. His choices nevertheless allow us to see what sources the ongoing Oku no hosomichi scholarship has uncovered. More importantly, Mori’s antecedents are normally representative, and this is what is most important to the intertextual critic. As Hirota Jirp’s studies of Bashp’s “relationships” (kakawari) to the classics reveal, tracing specific “influences” on a work such as Oku no hosomichi is far less important than pointing to general traditions or to certain of the set of five- and seven-syllable poetic phrases to which Bashp seems to have been attracted.15 When we as critics try to move from the historical poet’s mind to the poetic work, then we are obliged to continue to seek for the specific “correct” source. When our context, however, is not the author and not even the literary text as unique, independent entity but the world of literature more broadly, then we can concentrate on becoming as aware as possible of the stylistic and structural conventions Bashp takes for granted, as well as of the networks of traditional phrases, images, symbols, and patterns of action that give Bashp’s imaginative world its richness. So whether Mori has made all of Bashp’s possible sources available is not as important as the fact that his method of translation keeps us constantly aware of the work’s extraordinary allusive richness. Bashp borrows constantly, and his text is a multilayered tapestry of stylistic and structural threads stretching out to many earlier texts. Mori’s decision to translate in this manner (and it had to be a self-conscious “literary” choice) reveals that he is highly aware of the intertextual nature of Bashp’s text, and that he is taking as his most constant critical context not Bashp’s mind at the time of writing, nor the reality of his journey, nor even the Oku no hosomichi as the masterpiece of a certain genre, but classical Japanese literature more broadly. The translation suggests that Mori believes that the words from earlier waka, Chinese poetry, Tales of the Heike, the np and other folk narrative and drama are an intricate part of the work’s meaning and that a key to understanding the Oku no hosomichi will somehow be found in its literary antecedents. There is a final aspect of Mori’s presentation that helps keep our attention on the fact that the Oku no hosomichi is a text woven of many

184

E     K     

threads from the rich tapestry of literature that lives on Bashp’s mind. This is Mori’s lively retelling—not in his translation but as part of his commentary—of stories and folk legends that Bashp’s traveler uncovers on his journey. Mori sees Bashp’s role as similar to that of his own— and that of his own narrator in the novel Gassan—to rediscover the power and mystery of an area’s living tales about itself—a role, as Ogata Tsutomu has pointed out, similar to that of the waki (priestly “side actor”) in the np drama. Examples of Mori’s retellings include his recreation of the many np drama settings to which the traveler alludes or his ongoing recitation of the Yoshitsune cycle, particularly, of course, the Minamoto warrior’s flight to the northeast, his last battle, and ritual death (with wife, son, and newly born daughter) at Takadachi in Hiraizumi. This narrative strain includes many other players whose fates Yoshitsune had touched: Nasu no Ypichi, the Satp brothers and their wives and parents; Izumi Saburp, his father, Fujiwara Hidehira, and treacherous brother Fujiwara Yasuhira; and Benkei, Kanefusa, and other loyal warriors. Mori weaves into his commentary moving scenes from Heike monogatari, from kpwakamai (narrative folk drama), np, old jpruri texts, and other court and Buddhist collections of folk stories. An interesting example of the way in which Mori interweaves his recreation of legendary material into his analysis of the workings of taip is his discussion of Bashp’s use of the Tamamo no Mae legend in the Kurobane passages. Tamamo no Mae was, the story goes, a beautiful court favorite of Emperor Go-Toba and a nine-tailed golden fox in disguise. When the court astrologer discovers that she is a fox and the possible cause of the emperor’s illness, warriors are called in who chase her from the capital to the Nasu Plains. The same warriors proceed to Nasu, practice their marksmanship in a dog-chasing arena, and continue to hound her. The warrior Miura no Suke finally sees her form reflected in Mirror Pond from the tree upon which she is hiding. He shoots her, and she turns into a frightening poisonous rock, the Sesshpseki or “life destroying stone.” It is the manner in which Bashp structures his experiences at Kurobane that intrigues Mori. According to Sora, the order of outings during their several-day stay at Kurobane was as follows: they went first to the Unganji and the Shugen Kpmypji temples, and then to the Hachiman Shrine. In the first of three separate Kurobane passages, Bashp’s narrator, however, describes being taken to the dog-chasing arena, to Tamamo no Mae’s old tomb on the Nasu plains, and on to the Hachiman Shrine, where the travelers are moved by memories of the story of Nasu no Yoichi’s prayer to the god Hachiman as he lets his arrow sail toward a small fan on a distant ship.

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

185

At nightfall they worship before the travel sandals of En no Gypja at the Shugendp Kpmypji temple. Mori suggests that although Bashp carefully sets up the Tamamo no Mae imagery and atmosphere in the first haibun, he chose not to complete references to her story here with a mention of their visit to the Sesshpseki. Rather, he orders his successive haibun in a way that will set up a correspondence not just between the dog-chasing arena and Tamamo no Mae’s old tomb but also between the “skillful,” even though unrelated, deeds of Miura no Suke and Nasu no Yoichi. Bashp next creates an independent and very different haibun for a visit to Butcho Oshp’s hut at the Unganji. This he does in order to express his special respect for an honored priest, poet, and friend. In a third passage (traveling on horseback to see the Sesshpseki) he finally brings back the Tamamo no Mae story and the more violent warrior imagery by creating a correspondence in the hokku that ends the passage (no o yoko ni / uma hikimuke yo / hototogisu [across the moor / lead my horse! / hototogisu]). It happens that this hokku’s image of a horse on the plains alludes to a waka to which the narrator had alluded in the first Kurobane passage describing the trek over the Nasu plains: mononofu no / yanami tsukurou / kote no ue ni / arare tabashiru / Nasu no shinohara (The brave warrior / adjusts his quiver / while over his wrist guard / hail beats down / on the Nasu bamboo plains). Mori points out that these words are alluded to in the np play Sesshpseki (with its story of a traveling priest attempting to calm Tamamo no Mae’s angry fox-spirit), which then become a part of the Oku no hosomichi context. Bashp thus completes a complex “correspondence” of the type that draws his readers into an ever more intricate literary world.16 Mori suggests further that in creating a separate Sesshpseki haibun to end the Kurobane section Bashp creates an “intersection” with the later Fukushima haibun in which the traveler mentions a brief peek made at the Black Mound Cavern (Kurozuka no Iwaya), an allusion to the frightening home of another legendary female demon and subject of the np play Kurozuka. In this work a traveling priest arouses an angry female spirit by peeking into a mysterious old woman’s forbidden and nightmarish room full of half-eaten men and bones.17 Mori’s discussion of intersections, his translation, and his lively narrative of earlier stories, all help dramatize the fact that the haikai poet–traveler was constantly dipping into his own classic storehouse, one that encompassed both the Chinese and Japanese literary and philosophic traditions. Mori’s instinctive, primary interest as a critic is clearly in literature “as an imitation of literature” or in “poems which imitate other poems and take their meaning from these literary

186

E     K     

relationships.”18 This is why his approach can best be described as intertextual and why his reading of the text, one important aspect of his translation, and much of his commentary present the Oku no hosomichi as an imaginative construct that might indeed “tightly enclose” its readers into a rich literary world. Mori’s method illustrates his belief that with great literary masterpieces, the further one enters in, the deeper the work becomes.

I As indicated earlier, the unique feature of Mori’s reading of Oku no hosomichi is his extended and ongoing discussion of taip. His method is such that the monograph must be read through to be fully appreciated, but illustrations of some of the different types of echoes and counter-echoes he hears sounding through the work, and a presentation of what he reads as the “deep correspondence” that enlivens its base will reveal something of Mori’s vision of how the Oku no hosomichi functions as a literary work. The most numerous correspondences are those between verses: between waka or Chinese verses, for instance, of the type discussed in the Kurobane passage or between two hokku in the same, in contiguous, or in widely separate haibun. An example of the latter is the correspondence seen between the content of two hokku that were “asked for and presented to young men”: the hokku presented to the young man leading the traveler’s horse at Sesshpseki—no o yoko ni / uma hikimuke yo / hototogisu (across the plain / lead my horse/ hototogisu)— stands in balance with the hokku given much later to young monks on the other side of Japan at the Zenshpji—niwa haite / ideba ya tera ni / chiru yanagi (garden swept / I leave the temple / willows scatter). Or Mori sees correspondences between two verses that are the result of having been urged by others to visit a certain place: the Ashino hokku—ta ichimai / uete tachisaru / yanagi kana— (one field / planted and we leave / the willow) and that composed at Risshakuji— shizukasa ya / iwa ni shimiiru / semi no koe (silence! / sinking into the rocks / cicadas’ songs). In this case, Mori suggests that only when the two scenes and the two hokku are brought together in the intersection of the two poems can the reader feel the depth of “silence” (shizukasa) gleaned from the image of the ancient rocks that reaches back to the willow left alone in the rice field (rice planters gone, the travelers gone, no murmuring Saigyp stream, and Saigyp is gone into the quiet of death). There are also multiple correspondences between a phrase in the prose and a verse. The traveler’s memory of an earlier poet’s

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

187

conversation with a late-blooming cherry tree, recalled upon seeing such a tree buried under the snow on Mount Yudono, for instance, corresponds back to an earlier hokku presented to the traveler by his disciple Kyohaku, a poem addressed to the late-blooming cherry at Takekuma. The two prostitutes’ description of themselves, heard by the travelers in the night—“we are children of the shore where white waves crash all around”—sounds back against the well-known hokku ending the previous haibun that narrates a tormented, dark walk down the rough Japan Sea coast and ends with this hokku—ara umi ya / Sado ni yokotau / ama no gawa (Wild seas! / falling in on Sado / River of Heaven). Mori postulates many geographic correspondences, such as that set up between the Kitagami River on the Pacific coast and the Mogami River running to the Japan Sea coast. He points out that the narrator uses his own actions in these spots to dramatize a sense of their length and power—looking down on the Kitagami from above or flowing past the banks of the swift Mogami in a boat. He also names the rivers’ exits to the sea (Ishi no Maki and Sakata) and their upper reaches (in far northern areas)—areas that the traveler, again in both cases, longs to enter. There are the more general geographic and seasonal correspondences between the traveler’s walk up the Pacific coast to the sound of rice planting in the spring and their trek down the Japan Sea coast to the scent of early rice in the autumn. Many have pointed to the contrasting geographical and female imagery used to describe a smiling Matsushima on the Pacific side and a gloomy Kisakata on the Japan Sea side. Mori also points to the contrasting bird and allusive links suggested in the two more extended and carefully composed haibun, on Matsushima and Kisakata. Mori describes correspondences combining personalities and places, such as the slightly comic figure of the artist Kaemon in Sendai whose map led to a michiyuki (travel song) to Shiogama Bay and the colorful Tpsai in Fukui who led the traveler along a moon viewing michiyuki to Tsuruga. Another such example is the kind help a farmer provides during their crossing of the Nasu Plain, ending with meetings with waiting friends at Kurobane as opposed to the warning to avoid farmers at Nago Bay followed by their being led on, again, to waiting friends at Kanazawa. Related to these are correspondences between types of human events, such as the stay at the Zenshpji and the leave-taking with Sora, which corresponds to the stay at the Tenryuji and leave-taking with Hokushi. At Unganji, early in their journey, the travelers seek remains of a retreat of an old friend and Buddhist priest Butchp Oshp, and again near the end Bashp seeks an old friend and Buddhist priest at Tenryuji.

188

E      K    

There are ongoing correspondences between ideas, such as that set up in the work’s third passage with the mention of “hair turning white under Wu skies,” that is, the notion that the traveler will grow old and may die in faraway, inhospitable lands. The color white is later echoed first at Shirakawa (the White River) Barrier with its white roses, snow, and white u flowers decorating Sora’s hair, which, we had learned at Mount Kurokami (Black Hair Mount), was black but is now shaved off. The white hair image sounds again in a hokku by Sora on Kanefusa’s white hair and his death scene at Hiraizumi, and is finally matched by the narrator’s verse on another famous death scene—that of the old warrior Sanemori whose dyed black hair was washed white by his enemies and whose helmet the traveler views at the Tada Shrine.

O KU ’s O P Mori argues, as do many other commentators, that the first passage of the Oku no hosomichi is, like the first verse of a linked verse, its most important section, determining the work’s characteristic tone, its movement, direction, and goals.19 It depicts a complex departure— the hermit–poet’s philosophical departure from a particular way of life and his actual physical departure from the hermitage, a symbol of the life he abandons. In the famous opening phrase of the Oku no hosomichi—tsukihi wa hakutai no kakaku ni shite (months and days are the passing guests of a hundred generations)—Bashp alludes to a popular and, in Tokugawa times, much-quoted poetic prose piece, a “preface,” by another famous wanderer, the Chinese poet Li Bo (701–762). As the traveler’s formal opening statement, this phrase becomes the work’s basic lifeline, prefiguring the meaning of its narrative action. Mori discusses Bashp’s complex allusion in detail, seeing it as the work’s foundational taip. Here is Li Bo’s preface: “Banqueting in a Peach and Plum Garden on a Spring Night” Heaven and earth are inns for the ten thousand things, day and night are the passing guests of a hundred generations, (emphasis mine) and this floating world is like a dream. What happiness, then, should we hope to find here? Poets of old held their lanterns to wander late into the night, and truly, they had reason to do so. The mild spring calls me with scenes of misty beauty, while nature lends me her designs. We meet together here, in a garden, fragrant with flowering peach and plum—

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

189

what could be more joyous! Though I am no match for young poets, skilled at verse, we shall quietly enjoy the view, talking together of things deeply felt. A feast is spread; let us pass the food and wine and get drunk with the flowers and moon. What will happen to the inept one who composes no poem? Three more cups of wine for the culprit.20

Mori is interested in what Bashp borrows and what he changes of Li Bo’s ideas. He notes that Bashp builds the first of his parallel ideas not with Li Bo’s first phrase (heaven and earth are inns for the ten thousand things), but rather with his second (day and night are passing guests of a hundred generations). It should be mentioned too that Bashp presents a near literal translation into Japanese of Li Bo’s phrase, but he also changes one Chinese term, using the compound tsukihi (month and days, moon and sun, or time) in place of Li Bo’s kpin (day and night, light and darkness, or time). The Japanese word tsukihi brings to the line the more concrete and vivid images of the moon and sun with all the connotations the two carry in the Japanese poetic tradition. The term can also be expanded as Bashp’s narrator is led in his second phrase into his own variation on Li Bo’s attempts to spatialize and humanize time: yukikp toshi mo mata tabibito nari (and the years that come and go are also travelers). Mori suggests that Bashp did not borrow both metaphors (nature, or the world out there, as an inn for all creation, and time as the eternal traveler) as his contemporaries Saikaku and Michikaze had done, and that he chose the broader of the two with which to create a correspondence.21 Bashp’s narrator chooses not to bring into his poetic world the spatial metaphor—heaven and earth as an inn for the ten thousand things. The literal meaning of the Chinese term ni lu (that which “greets the traveler” or “opposes the journey”) is not what the traveler envisions at this point in his account. It provides the image of a stable roof, something that interrupts the journey. The two opening parallel phrases in Oku no hosomichi emphasize the temporal, the movement of time: nights and days and the years, coming and going, are travelers. This metaphor highlights the relationship upon which the narrator wishes to focus: that between movement (the flow of time) and actual human life—human life as an eternal present, a movement that is time and space. Both of Bashp’s images strengthen the analogy between human life as the flow of a specific, limited time and the seasonal movement of creation. Mori points to the fact that Bashp continues in the rest of his opening passage (and in the second as well) to echo Li Bo’s words. In

190

E     K     

his second sentence he alludes to Li Bo more subtly, presenting his implicit “opposition” to the thought of playing all night in order not to “lose” any precious time: Fune no ue ni shpgai o ukabe, uma no kuchi o toraete oi o mukpru mono wa hibi tabi ni shite tabi o sumika to su. Those who float through life aboard a boat or greet old age leading a horse make each day a journey and the journey their home.

Here again an ordinary unit of time, each day, is a journey, while the individual’s movement with time through ever-changing space is home. The narrator is drawn to the examples of ordinary people who step out onto the road to greet time, their old age and death—people whose everyday lives bring them into harmony with the flow, who make it their home, and who do not linger in the garden. Bashp’s extended allusion to Li Bo continues in the third sentence. He first borrows one of the Chinese poet’s terms, kojin (the ancients or poets of old). To expand upon Mori’s discussion here, we might note that the narrator’s intricate use of the technique of parallelism in the first three balanced pairs sets up a rhythm that suggests we might find other parallels, other natural phenomena (besides the moon and sun) and other ordinary human beings (besides traveling merchants) who live in awareness of the fact of change, and who act out this truth in their daily lives. The rhythm and structure of the first two sentences lead the narrator, at any rate, to this type of expansion as he searches for further support for his ideas and continues to think about Li Bo and his words: kojin mo pku tabi ni shiseru ari (and many were the poets of old too who died on the road). The sentence can be read on one level as a “response” to Li Bo: yes, many of the venerable poets of old, aware of the illusory, dream-like nature of our selves and our floating world, amused themselves late into the night, unwilling to miss a moment of pleasure. But then many also died on the road, making the journey their home. Finally, the narrator’s search for parallels leads him, impels him, into the words yo mo, which begin the long last sentence moving through the hokku to the end of the passage: Kojin mo pku tabi ni shiseru ari. Yo mo . . .(Many too were the poets of old who died on the road. Just as I . . .).22 Bashp’s narrator links himself emotionally here with the conceptual truth of impermanence, which he has borrowed from Li Bo, and with the metaphor he has created by juxtaposing

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

191

human life and a journey. The tension the narrator feels in making and maintaining this identity is expressed in the constant repetition of the word “travel” or “traveler” in the three sentences leading up to his own yo mo: kakaku (traveler or passing guest), tabibito (traveler), and tabi (in the phrases hibi tabi ni shite, tabi o sumika to su, and tabi ni shiseru ari).23 If we go back to the ware mo in Mori’s own title (Ware mo mata, Oku no hosomichi), we can see that while it has the modest sense of “even I”—“I too, who had no special interpretation of the Oku no hosomichi,” –– it also contains a more complex suggestion, one that links Mori with Bashp’s yo mo, and with Li Bo, Du Fu, Npin, Saigyp, Spgi, and other kojin or “poets of old.” Mori’s phrase ware mo mata corresponds with Bashp’s text at another key point in the text. In the ketsu or final portion of the work, when Sora becomes ill at the Yamanaka hot springs and sets out alone for Ise, he leaves this parting verse: yuki yukite taorefusu to mo hagi no hara

On and on I go and should I fall, may it be in fields of hagi

Sharing the sorrow of parting, Bashp’s narrator presents his hokku response, prefaced with the phrase, “and me too, once again”(yo mo mata): kyp yori ya kakitsuke kesan

from today we erase those words!

kasa no tsuyu

dew in the traveler’s hat

The maegaki (preface), yo mo mata, responds directly to Sora’s verse, “and I too must travel on and on, and if I die on the road, I would like it to be in the flowers in a manner which links me with Saigyp and with you.”24 His own departure verse to Sora expresses both personal sadness and the deeper and more positive, if paradoxical, understanding that we all ultimately travel alone with the Buddha.25 With his use of the phrase ware mo Mori links himself and his reader to Saigyp, the haikai poet Bashp, his Oku no hosomichi narrator, and Sora—all who die on the road. Just as Bashp’s narrator was hurrying on, hurrying to include himself or to say I too am a poet and a traveler and thus live and die on the road, Mori too seeks to add himself to Bashp’s vision of death and of his fate as traveler. Bashp fashions a traveler who rejects or at least ignores Li Bo’s more hedonistic attitude: that we must enjoy our brief lives before time runs out. He does not deny the fact of change, however. His life philosophy, like Li Bo’s, is based on it. Nor does he deny the pleasures and sufferings that change brings. He is, rather, obsessed by

192

E     K     

images of time, space, and the traveler: the moon, the sun, days, months, years, ordinary humans as travelers, and earlier poets, including Li Bo, who were travelers. The particles mo also echo back to the opening sentence (toshi mo) and to Mori’s title suggesting that the narrators (Bashp’s and Mori’s) are concerned with constantly moving images: coming and going days, nights, and years, merchants, and poets—travelers who come and go, alone and without the illusion that their lives are any other than the brief flow from birth to death. Bashp continues to suggest the influence of Li Bo’s preface in this opening passage with the description of his own desire to roam. This desire is brought on not only by the thought of earlier poets but also by the coming of spring, by inspiration from the gods of wanderlust and travel, and by the prospect of the beauty of the flowers at Shirakawa and the moon over Matsushima. In the second passage, Bashp echoes Li Bo when he depicts the gathering of friends or when he adopts another Li Bo-like metaphor for human life—maboroshi no chimata (“crossroads of the unreal” or the illusory crossroad between birth and death).26 The narrator’s opening quotation of and variation on Li Bo’s Preface sounds out as a formal introduction to his poetic journal. His inclusion of a direct literary allusion as his opening line suggests strongly that the meaning of the work being presented will somehow be found in its relationships to words of the past. This and the continued echoing of Li Bo’s (and other poets’) thoughts and images allow the narrator to say, in effect, that not only have certain key actions of his past life sprung in part from the great literature of the past, but also his way of thought and his sense of the meaning of life is somehow related to this Chinese poet’s thoughts. His changes of Li Bo’s ideas suggest that while he recognizes the sameness between himself and the earlier poet, he also feels a difference, which he must try to express. The narrative of the journey that follows thus stands in part in opposition to Li Bo’s world. With a lingering moon overhead and tears in his eyes, Bashp’s poet departs from friends. Li Bo’s poet gathers with them to talk, eat, drink, compose poems, and play games in the flowers and moonlight. Although Bashp had not picked up on Li Bo’s spatial metaphor in his opening passage, Mori makes the interesting observation that he did do so much later in the journey. It is during the important pilgrimage to the three sacred mountains and climb to Gassan that Bashp’s narrator faces the Li Bo metaphor he earlier ignores: heaven and earth are the inn for all creation. Mori calls attention to the

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

193

following phrase from the Gassan passage: unmu sanki no naka ni hypsetsu o funde noboru koto hachi ri (the climb, treading ice and snow into clouds, mists, and cold mountain air, was eight li). This phrase, Mori suggests, “corresponds with” a verse presented much earlier. During the ki (preparatory) portion of the journey, when the traveler bowed before the traveling clogs of En no Gypja at the Kpmypji in Kurobane, he presented this hokku: natsu yama ni ashida o ogamu kadode kana

In summer mountains praying to his travel clogs— my leavetaking

En no Gypja was traditionally considered to be one of the founders of the Shugendp religious discipline and the yamabushi tradition of secret ascetic practices associated with sacred mountains. Here at the Kpmypji early in the journey, the traveler must have prayed for the strength and spiritual power to pursue his journey. Months later, after his travel sandals had carried him to this goal, the most sacred of yamabushi mountains, as he walks through ice and snow in a spot where heaven and earth seem to meet and swirl together, the poet brings back his earlier feelings for the yamabushi monk captured in this hokku.27 He then experiences a feeling of awe, as if he had entered into an unearthly realm: sara ni jitsugetsu gypdp no unkan ni iru ka to ayashimare (I felt as though we had stepped through the cloud-barrier onto the pathway of the sun and moon). When the sun sets and moon appears, the travelers make their bed on the mountain top.28 It is here—on an actual border between “heaven and earth,” where all of “nature,” or the sky and the earth, can be experienced as an open space that shelters and envelops him—that the traveler seems to accept Li Bo’s metaphor, not as an image of false permanence and security but as a moving, open center where the travelers sleep. Mori suggests it was probably on this night on top of Gassan that Bashp (Mori, here as elsewhere, tends to equate Bashp and his poet–narrator) was able for the first time to take Li Bo’s pair of space–time metaphors as a single truth: space, the universe, is the inn for all creation and our flow through it, time, is its passing guest.29 Bashp, perhaps, and his fictional narrator seem to have arrived at a new consciousness of space and time as a flowing present, all travelers’ home. If Bashp’s narrator draws closer to Li Bo here, he nevertheless continues in basic opposition to that other area in which he had created a foundational, structural correspondence—the image of friends judging poetic elegance ( f u ga) and drinking three more cups of wine. At

194

E     K     

several key points in his journey the traveler meets old friends and aspiring haikai poets; he is fed, cared for, and asked to participate in poetry gatherings at Kurobane, Sukagawa, Sendai, Obanazawa, Pishida on the Mogami River, Haguro, Kanazawa, Fukui, Iro no Hama, and Pgaki. As a traveler, however, he always departs from friends, guarding his original intuitive sense that he must continue to move on and on. Mori introduces the idea that Bashp’s next-to-last “party”—the entertainment arranged as a boat outing to Iro no Hama in Tsuruga in a lonely and somewhat bedraggled autumn setting—stands as a taip once again (as yin to yang), with Li Bo’s banquet on a spring night. Our poet leaves Tpsai at Tsuruga, of course, and moves on with Rptsu to Pgaki. Here he is met by friends who are stunned and overjoyed to see him, “as one returned from the dead.” One more party and he moves on again—departing by boat, as he had at the beginning of his journey, to Ise, and leaving this verse to the Pgaki contingent:30 hamaguri no Futami ni wakare yuku-aki zo

The clam’s body and shell parting, to view Futami once again disappearing fall

This final hokku “corresponds” with the traveler’s first departure verse left for friends in Senju: yuku-haru ya tori naki uo no me wa namida

Disappearing spring! Birds cry out, while fishes eyes fill with tears

In a similar way the term yuku-haru takes us back to the opening words of Oku no hosomichi—to the moon and sun forever passing and to yukikp toshi mo (the years coming and going) and the seasons that passed the pretraveler by when he lived in his hut and moved with the traveler on his journey. Mori says that for readers of the Oku no hosomichi the poet has, “as a white haired traveler to unknown lands and with his own body, completed his 100-verse renku and led us through ever deepening 3000 realms, carefully and finally enclosing us within an expanding theme.”31 To continue to act out the vision, the philosophy or meidai, he has based his literary illusion (the Oku no hosomichi) upon, the traveler must move on to other ever-changing worlds. This paradox, Bashp’s deconstruction of his carefully created world, is what he ultimately leaves us while his traveler goes on to Ise, seeking still Saigyo’s spirit and maybe too, says Mori, the two

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

195

entertainment women and pilgrims to Ise he chanced to meet at Ichiburi.

N 1. Mori Atsushi, Ware mo mata, Oku no hosomichi (And Me Too, Once Again, into Oku no hosomichi) (Tokyo: Nihon Hpsp Shuppan Kypkai, 1988); Mori received the Akutagawa Prize for literature in 1974, for the novel Gassan (1973). Gassan is a lyrical novel based on the author’s experiences during a period of semi-reclusion in the Churenji temple at the foot of Mount Gassan and Mount Yudono in the village of Asahi, Yamagata Prefecture. The work’s I-narrator, a writer, seems to have been drawn to the temple to renew his links to language and words and to death and life through death. The narrator describes his gradual move into a dream-like autumn and winter during which villagers lead him into the mysteries of Gassan, the sacred “mountain of death.” As guest in the temple and outsider in the village, the narrator becomes an empathetic eye describing bits of the lives of the temple inhabitants, their efforts to survive through the cold and isolation of winter, and their relationships to the sparse village community. Making his base an igloo-like tent constructed of plastic as protection from leaks in the ceiling of his attack room, he listens, observes, and records. The center of his attention is the life story of a former carpenter, now paralyzed from an accident, who has been taken in by the temple priest. To contribute to the care provided him, he whittles chopsticks and sells them to the villagers and other temple visitors. As this life unfolds, the narrator weaves snatches of other “stories” he hears at gatherings and in conversations with the temple family, mountain men, and villagers. These stories become the warm liquid that flows through a narrative of hardship, rough lives, and survival at the foot of the imposing snow covered Gassan rising over those who live in its shadow. 2. Mori, Ware mo mata, 6. 3. In stating that “in a unified work, every element works together toward a theme,” Steven Lynn (Texts and Context: Writing about Literature with Critical Theory [New York: Addison Wesley Longman, 1997], 2) is describing “New Criticism,” but as Donald Keesey and others point out, in attempting to link a major theme to a work’s structural and stylistic characteristics, both “intertextual” and “New Criticism” critics are functioning as structuralists; see Donald Keesey’s introduction to Chapter Five, “Intertextual Criticism: Literature as Context,” in his Contexts for Criticism (Palo Alto, CA: Mayfield Publishing, 1987), 258 and 262. 4. Mori, Ware mo mata, 7. 5. See examples later. 6. Mori, Ware mo mata, 8.

196

E     K     

7. Mori, Ware mo mata, 8. 8. Adapted from the definition in the standard Japanese dictionary Kpjien, and from the brief discussion by James Robert Hightower in Topics in Chinese Literature (Cambridge, MA: Harvard University Press, 1962), 69–70. 9. Mori, Ware mo mata, 8. 10. Translation is by William LaFleur, Awesome Nightfall, The Life, Times, and Poetry of Saigyp (Boston: Wisdom Publications, 2003), 143. 11. Bashp’s use of the term kojin is discussed in more detail later. 12. Mori, Ware mo mata, 77–78. 13. Mori is never drawn into speculation about what constitutes the “truth” of the journey—whether it is the outward facts or inward “poetic” reality. This is clearly not a relevant question for him and does not reflect his sense of what literature is. 14. Examples of Mori’s arguments can be found in almost every section; they deal often with changes in the order of events and have been discussed frequently by other scholars. One example will suffice here to reveal Mori’s approach. Sora records in his diary that on their visit to Nikkp they stayed at Hotoke Gozaemon’s inn on their last night, visiting the shrine and temples on the mountain first. Bashp presents three separate Nikkp haibun, placing the dramatization of their stay with Hotoke Gozaemon first. Mori suggests that he does this to set up a correspondence between the man, Hotoke Gozaemon, his direct words, and the narrator’s response to him in this section with Sora’s quoted words and the narrator’s response to them in the previous Muro no Yashima passage. With this structure Bashp is also able to line up or compare and contrast three ways of thought: 1) Shintp (the shrine and shrine history at Muro no Yoshima; 2) Confucian (Hotoke Gozaemon, whose character stimulated the narrator to recall a key passage from The Analects [which, incidentally, is “echoed” later by other key words from The Analects remembered by the narrator in Iizuka see Mori, Ware mo mata, 23 and 65-66]; and 3) Buddhist (in the two Nikkp passage the narrator describes first their pilgrimage to the sacred mountain on which, he explains, Kukai had founded a temple, and second, after introducing Sora and his religious preparations for the journey, he describes their improvised Buddhist “summer purification retreat”—backing into a cave under the famous waterfall). 15. Hirota Jirp, Bashp to Koten—Genroku Jidai (Tokyo: Meiji Shoin, 1987), 429–714. 16. Mori, Ware mo mata, 31 and 39–40. 17. Ibid., 56–57. 18. Keesey, “Intertextual Criticism,” 255. 19. In the discussions of Oku no hosomichi’s opening haibun here and in the conclusion of this chapter I draw myself into Mori’s responses to Bashp’s work by borrowing in some detail form my own dissertaion on Oku no

M   A     ’ W A R E

M O M ATA ,

OKU

NO HOSOMICHI

197

hosomichi (“Masuo Bashp’s Oku no hosomichi, a Critical Study,” University of Indiana, 1974), see especially pp. 172–216. Translations of Oku no hosomichi passages are my own. For Mori’s discussion of the opening passage see Ware mo mata, 9–16. For English translation of Oku no hosomichi see David Barnhill, Bashp’s Journey (State University of New York Press, 2005), “The Narrow Road to the Deep North,” 49–77. 20. This translation of the preface is based in part on Mori’s rendering into modern Japanese (9–10) and in part on the original Chinese. Li Bo’s preface is included in the collection of Chinese prose and poetry popular in Bashp’s day, Kobun shinpp goshu (“Prefaces,” Xu Lei). Direct quotations of the first three phrases of Li Bo’s preface are found in works by two of Bashp’s contemporaries: Ihara Saikaku in Nippon eitaigura (Eternal Storehouse of Japan, 1688) and Pyodo Michikaze in Nippon angya bunshu (Collection of Pilgramiges Through Japan, 1690). Ogata and others have noted that Saikaku and Michikaze do not, as Bashp does, alter their Chinese source. See Ogata Tsutomu, Kaishaku to Kansp, May 1964, “Tabidachi,” 150–151. 21. Mori, Ware mo mata, 9. 22. The first passage continues: Just as I—when did it all begin?—a solitary cloud lured off by the wind, unable to silence the will to roam— drifted along far sea coasts, and last autumn, back in my tumbledown hut on the river, brushed the cobwebs away as the old year drew to its close, and with spring came mists in the sky and the thought: “go, cross the Barrier of Shirakawa!” the God of Wanderlust possessed all things around me and my heart was bewitched; the God of the Road called to me, my hands kept to nothing; so I patched the holes in my trousers, replaced the cords on my bamboo hat, applied moxa to my shins, and then . . . the moon at Matsushima fill my heart as I handed my hut to another and moved to Sampu’s cottage, kusa no to mo sumikawaru yo zo hina no ie

Even my thatched hut has its season for moving! a household of dolls

I left a renku’s first eight on a pillar of the hermitage. The renku’s first eight (omote hachiku) are the first eight verses of a one-hundred-verse linked verse. Mori suggests that the act of leaving the verse reveals the poet’s determination to complete the linked verse with his life, that is to say with his own body on this journey through the north country. Ibid., 8.

198

E    K    

23. A point made by Ogata, Kaishaku to Kansp, May, 1964, “Tabidachi,” 157. 24. Sora’s verse alludes to two well-known phrases; one is from the “Nineteen Poems in Ancient Style” in Wen-xuan, Vol. 29: “Going on always on and on / alive, but parted from you” (translation by Charles Hartman, Sunflower Splendor [Bloomington, IN: Indiana University Press, 1975], 30); and the other from a verse by Saigyp: izuku ni ka / nemuri nemurite / taorefusan to / omou kanashiki / michishiba no tsuyu (Where will it be / that I fall / sleeping after night night on the road? / so go my thoughts / dewdrops on the lonely roadside). And both Sora and Bashp allude more generally to Saigyp’s desire to die under the cherry blossoms and the legend that he did so. 25. The phrase the narrator and Sora would have written in their travel hats at the beginning of their journey, kenkon muju dpgyp futari (no place is our home in heaven or earth; two companions on the Way), is one traditionally copied by pilgrims. Dpgyp futari signifies my companion and I, the Buddha and I, or myself and my Buddha nature—all following the Buddhist Way. 26. Mori, Ware mo mata, 14. 27. Ibid., 126. 28. The Gassan (Moon Mountain) passage centers on images of the heavens, the sun, and moon: On the eighth we climbed Gassan. We hung sacred laced rope around our necks bleached cotton hoods over our heads and were led by a mountain pilgrim’s guide; the climb, over ice and snow into clouds, mists, and cold mountain air, was eight li up; I felt as though we had stepped through the cloud barrier on to the pathway of the sun and moon; breath was short and body chilled, and on the summit the sun sank away and the moon appeared. We spread bamboo grass for a bed, bamboo leaves for a pillow, and lay down to await the dawn. The sun rose, clouds disappeared, and we climbed down to Yudono. 29. Mori, Ware mo mata, 126–128. 30. Hamaguri (clams) are the special product of the Bay of Futami in Ise. We know from earlier haibun that Bashp associated Saigyp with Ise and especially with Futami (Two Views and two bodies), where the earlier poet is thought to have lived in retreat. Mori presents the following waka by Saigyp as the verse with which Bashp’s hamaguri no futami (two bodies, two views, of the clam) quietly “intersects”: ima zo shiru / Futami no Ura no / hamaguri o / miawase tote / punarikeri (now you know, / the clam shells / of Futami no Ura, / saying “they’re for a matching game” / I enclose them). 31. Mori, Ware mo mata, 182.

P  III

T P  P

This page intentionally left blank



B¯   H  Joan O’Mara

W

hen the creative output of Matsuo Bashp is thought of, as it most often is, in literary terms alone, he is recognized for his achievements in the composition of individual haikai verses, his participation in haikai no renga (haikai linked verse), and his contributions to the development of haibun, the composite form in which Bashp’s prose accounts of his travels were interspersed with the haikai he composed while making those journeys. If the idea of ‘creative output’ can be stretched to encompass not only literary forms but also visual ones, then Bashp can be perceived as an even more varied and interesting creative personality. Without changing, but simply by being slightly redefined, he becomes one who can speak to an even wider audience. These visual forms all relate to Bashp’s poetic practice in one way or another and include simple sketches, illustrations based on his journeyings, and the genre to be considered here, the haiga. In its purest form, the haiga consists of a painting ( ga) that is accompanied by the inscription of, and is related to the content of, at least one seventeen-syllable verse. The painting is usually executed in monochrome ink, or ink and light color; its style is usually abbreviated. In other words, the painting should be carried out in a manner that is consistent with the poetic form it accompanies. The connection of the painted part of a haiga to the content of its verse is particularly important to the overall impact of the work. This connection, or linkage, in a haiga is most often simply illustrative, with the painting depicting only what is presented in the verse. More challenging, and often more rewarding for the viewer, is a use of linkage that can be termed associative; that is to say, the painting not only relates to the content of the verse, but takes the viewer’s understanding a step further by the addition of some new detail, or several steps

202

J  O’M

further by the introduction of an associated subject that the verse suggests. It is generally the use of such associative linkage that will bring the haiga form to its fullest expressive potential, as a type of mixed media variant on the haikai no renga process. Bashp used both sorts of linkage in his production of haiga. Some of Bashp’s paintings were little more than sketches and in the absence of any accompanying verse or consequent linkage should not really be regarded as haiga. His visualization of the snore of one of his disciples, for example, is wonderfully suggestive1; its shape swells and then subsides, and it is accompanied by Bashp’s humorous inscription of the dimensions of the snore at its beginning and middle, and a comparison of the subsiding of the snore (wavy lines at the end) with the jolting vibrations of a wheeled clothes chest (nagamochi). The sketch is also a superb example of Bashp’s use of synesthesia, or mixing of the senses (here, providing visual dimensions to an auditory phenomenon), and it is replete with haikai flavor in the sense of “humorous” or “playful.” Wonderful though it may be, however, it is not accompanied by or linked to any particular haiku, and therefore should not, technically, be regarded as a haiga. Neither should Bashp’s illustrations of his travels be casually lumped together with his haiga.2 Such scenes may describe visually some of the places visited by Bashp, and pick out some of the specific things that he saw on those travels; the details may even be handled in a way that is stylistically consistent with the abbreviated forms associated with the haiga genre, but the illustrations generally have little relation to the content of the verses that he composed while journeying. The haiga form was one of a number of variants on the practice of haikai that were evolving in the late seventeenth century. Bashp, who scorned the frivolity of most of these variants, evidently considered the creation of haiga to be a loftier pursuit, connected as it was to deeply rooted sources in the poetic and artistic traditions of both China and Japan, for he produced several dozen haiga, both alone and in collaboration with various disciples. His attitudes are clearly revealed in a brief prose work of 1693, Saimon no ji (“The Rustic Gate”), in which he said of his disciple Morikawa Kyoriku, also known as Kyoroku (1656–1715): Talented as he is, he loves both painting and haikai poetry. I asked him once as a test why he liked painting, and he said it was because of poetry. “And why do you love poetry?” “Because of painting.” Two things he studied for one purpose. Indeed, since it is said [in the Analects, IX, 6] that it is shameful for a gentleman to have many accomplishments, it is admirable that he makes one use of the two arts.3

B ¯   H 

203

Although Bashp also numbered among his followers others who were painters first and poets second, such as Hanabusa Itchp (1652–1724) and Ogawa Haritsu (1663–1747), it was not those artists but Kyoriku whom he called master. By his own account, also in Saimon no ji, he goes on to say: In painting he was my teacher; in poetry I taught him and he was my disciple. My teacher’s paintings are imbued with such profundity of spirit and executed with such marvelous dexterity that I could never approach their mysterious depths.4

It may be difficult to imagine anyone performing up to levels that might inspire such high praise, but Kyoriku was in fact a competent painter in the style that he had learned from Kanp Yasunobu (1613–1685). Traces of the crisp detail and mannered brushwork for draperies that were characteristic of Kanp style in the Edo period may be seen in Kyoriku’s painting of Bashp as poet–pilgrim–traveler, following in the footsteps of his revered predecessor Saigyp (1118–1190) with sedge hat and walking stick in hand (figure 9.1).5 In his haiga, Kyoriku showed an ability to select and abbreviate, even to the point of abandoning the Kanp line altogether in favor of a “boneless” (mokkotsu) use of wash, as in the collaborative work ( gassaku) in which he illustrated Bashp’s inscription of his well-known hokku about a crow sitting on a withered branch (figure 9.2).6 The association with Kyoriku came late in Bashp’s life. Although Kyoriku had known and admired Bashp’s poetry for some years, his active duties as a samurai of the Hikone clan had prevented him from actually meeting the famous master until the autumn of 1692, when he came up to Edo for a period of service there. An instant mutual rapport seems to have sprung up between the two men, and the next nine months saw the quick ripening of a close relationship, as Kyoriku had the opportunity to work directly with the poet he had so long admired from a distance, and Bashp found a disciple after his own heart (at least, that is the way it was reported by Kyoriku). According to Kyoriku’s account, Bashp had been delighted with the verses he had brought to show him, had been amazed to find such verses coming from one who was self-taught, and had felt that here, finally, was a disciple who understood the philosophical and theoretical underpinnings of his haikai. Even allowing for a certain amount of hyperbolic self-promotion, there were at least some grounds for Kyoriku’s claim to be Bashp’s true successor and favored pupil, among all the similar claims put forth by other disciples after the master’s death in 1694.

204

Figure 9.1 Kyoriku, Portrait of Bashp (as the “new” Saigyp), ink on paper, 99  28.5 cm, Kakimori Bunko, Itami.

205

Figure 9.2 Bashp (verse) with Kyoriku (illustration), Crow on a Withered Branch (“Kara eda ni ”), ink on paper, 107  31.2 cm, Idemitsu Museum of Arts, Tokyo.

206

J  O’M

A number of haiga by Bashp, including a depiction of a solitary traveler in the rain and several versions of plum blossoms and the moon,7 are known to have been created during the course of these visits, and there are probably others that were produced then, in a similar context. This period also saw the production of at least three gassaku with verses by Bashp, as the senior poet present, and paintings by Kyoriku: yellow rose (yamabuki) blooming next to a waterfall; the crow on a withered branch, mentioned earlier (figure 9.2); and two sparrows in flight (also done in the boneless manner).8 In the summer of 1693, Kyoriku had to return to his post in Hikone, and it was his departure that occasioned Bashp’s words in Saimon no ji about the relationship between painting and haikai, as well as his artistic debt to Kyoriku. Bashp’s statement there should not necessarily be taken to mean that his study with Kyoriku had prompted his first attempts at painting, and that all of his haiga should therefore be assigned to the last three years of his life. Such a narrow interpretation does not fit the evidence, for although Bashp’s haiga are not generally dated, a number of them are known to have been created prior to 1692. It seems clear that Bashp’s words in Saimon no ji were meant to honor Kyoriku as a parting gift, rather than to provide an all-inclusive history of his own involvement with painting. If he had also studied painting with another, in an earlier year, would it have been necessary or appropriate to mention such a fact in the context of a farewell tribute to Kyoriku? What seems more likely, after comparing works from Bashp’s “pre-Kyoriku” days with those known to be later, is that he was mostly self-taught as a painter, and his association with Kyoriku may have marked his first systematic, formal study of the discipline. It was not his first direct contact, however, with an artist working in the Kanp style. There is a gassaku, produced probably in the early 1680’s, of two compositionally unrelated scenes of crows on a bare tree (right half of a short handscroll) and a lone traveler in a complete landscape setting (left half ), with verses and a prose passage inscribed by Bashp.9 This haiga was not the casual product of a collaboration at a social/literary gathering, but a full-fledged, detailed two-part composition painted on silk. Tradition once held that the whole work was by Bashp, but it is now recognized that the painting is too much a product of formal Kanp training to be entirely from his hand. It is much more likely to be a gassaku requested of Bashp and a professional Kanp painter whose name, however, does not appear. While Bashp’s part in this work was limited to the calligraphy of the verses and the accompanying prose text, its existence is evidence that

B ¯   H 

207

he had been exposed to the Kanp style well before the period of his association with Kyoriku. That he did not absorb much from the exposure can be seen in a haiga of the 1680s that is entirely from his own hand. This work (figure 9.3), which takes gourd-beating (hachi-tataki) for its subject,10 was created in late 1689, when Bashp was visiting his disciple Mukai Kyorai (1651–1704) at his estate in the Saga district west of Kyoto. The purpose of the end-of-the-year visit was to hear the gourd-beaters as they made their seasonal rounds, and the occasion was marked by the composition of verses and the use of one of them in this haiga. The form of Bashp’s gourd-beater, glancing over his shoulder at the moon, is not without a certain naive charm, but the painting is not especially strong. The flaring, jagged hemline of the robe indicates a degree of familiarity with late Kanp brush conventions, but the puppetlike figure is almost childishly drawn when compared with the Kanp figure of a traveler in the gassaku of several years earlier, or with Kyoriku’s Kanp-influenced portrait of Bashp (figure 9.1). It has been suggested that the image of the gourd-beating monk may reflect the influence of Hanabusa Itchp (1652–1725) on Bashp’s early painting style, and a painting by Itchp of a samurai’s servant holding a hand drum has been offered in support of that suggestion.11 Itchp’s painting is undated and may be later than 1689. It is not monochromatic, there is considerably more detail, and the brushwork is much more self-assured and accomplished than Bashp’s, but there are some affinities in the jagged lower hemlines, tilt of the heads, and general attitudes of the two figures. Itchp’s association with Bashp may go back far enough that he could have influenced the painting of his teacher in 1689, the date of Hachi-tataki. One of his verses appeared in an anthology published by the Bashp school in 1683. There is also a gassaku (figure 9.4) with Itchp’s painting of a bagworm (minomushi) suspended in its cocoon from a bare tree, accompanied by Bashp’s inscription of his own verse: Minomushi no ne o kiki ni koyo kusa no io

Voice of the bagworm— come and listen to its cry at my thatched retreat.

The visual image has little to do with Hachi-tataki stylistically, except that it uses no color and is in some sense Kanp-related. The work may well have been produced before 1689, since Bashp’s verse first appeared in 1687, in the context of an invitation to an autumn poetry gathering. The evidence, therefore, shows that Itchp could have been in a position to influence Bashp’s early painting. It is not conclusive,

208

Figure 9.3 Bashp, Gourd-beating (Hachi-tataki), ink on paper, 34  19 cm, Wataya Bunko, Tenri Central Library, Tenri.

B ¯   H 

209

Figure 9.4 Bashp (verse) with Itchp (illustration), Bagworm (“Minomushi no”), ink and light color on paper, 27.7  33.9 cm, Wataya Bunko, Tenri Central Library, Tenri.

however, for nowhere did Bashp acknowledge an artistic debt to Itchp, as he did to Kyoriku. The difference that Bashp’s association with Kyoriku made in his painting style is evident when the gourd-beater of 1689 is contrasted with another of Bashp’s figures (figure 9.5) from a later handscroll that depicts scenes from his travels.12 There, a squat little traveler (Bashp himself ?) is sheltered by a sedge hat as he hurries through the rain toward a thatched cottage among some trees; the form of the traveler by Bashp is still not a masterpiece of figural painting, but it does exhibit greater confidence in its brush handling, especially evident in the more detailed definition of hat and robe, than is seen in the earlier image of the gourd-beater. The traveler may not have the assurance of Kyoriku’s own portrait of Bashp (figure 9.1), but a comparison of those two paintings makes Kyoriku’s influence especially apparent. Of Bashp’s mature haiga it may be said that although there may not be many that are troubled by major problems in handling or linkage, there are also few that are individually memorable. Most are

210

J  O’M

Figure 9.5 Bashp, Solitary Traveler in the Rain; detail from Handscroll of Travel Sketches, ink on paper, 22.5  517 cm, Kakimori Bunko, Itami.

haiga of a fairly prosaic type in which verses about bamboo, grasses, flowers, and the like are linked illustratively with images of those subjects. A straightforward example of the latter may be seen in Yellow Rose ( Yamabuki) (figure 9.6). In a number of late works, Bashp followed Kyoriku’s example (as seen in figure 9.2) by dispensing with the Kanp-derived line in favor of a boneless use of wash, often with tones of pale color, and many of these later haiga are characterized by a light, delicate touch. Bashp may have seen in them a visual counterpart to his late emphasis on karumi (“lightness”) in haikai. In one such work (figure 9.7), a bashp tree stands next to a rustic gate, evidently the entrance to the Bashp-an, the poet’s residence at Fukugawa in Edo. The painting, an example of the influence of Kyoriku’s brush technique, is executed lightly in ink and wash, with a pale blue–green used for the boneless definition of the broad ragged leaves of Bashp’s favorite tree. Its verse is the familiar one about the bagworm, already encountered earlier in the discussion of Bashp’s gassaku with Itchp, that was originally composed in 1687 as an invitation to an autumnal poetry gathering at the Bashp-an. Here, visual image and verse are only obliquely related. The bagworm is not seen, and neither is the thatched hut. The viewer stands outside the gate as the invited guest; beyond it are the Bashp-an and the world of the poet who hears unspoken cries, perceiving what others do not. The verse beckons the visitor to open the gate and enter into that world. In this associatively linked late work, the verse and the image come together to form a strong, integrated haiga. If many of Bashp’s haiga can be seen as fairly standard, albeit pleasant examples of illustrative linkage, Banana Plant by the Gate to the Bashp-an stands as a masterpiece of synthesis and lyricism in the history of haiga practice.

211

Figure 9.6 Bashp, Yellow Rose (Yamabuki), ink on paper, 42.8  27.2 cm, Kakimori Bunko, Itami.

212

Figure 9.7 Bashp, Banana Tree by the Gate to the Bashp-an(“Minomushi no”), ink and light color on paper, 97.3  29.7 cm, Idemitsu Museum of Arts, Tokyo.

B ¯   H 

213

In addition to the contribution made by this and others of Bashp’s individual haiga to the development of the haiga form,13 his contribution should perhaps also be measured in a second way: in terms of his influence in establishing the genre. The importance of the first should not be overly exaggerated, while that of the second should not be underestimated.14 Looking at the intrinsic merits of Bashp’s haiga (as distinguished from a consideration of the literary merits of the respective verses involved) yields a mixed evaluation. On the positive side, Bashp did seem to be more systematic about keeping his visual imagery selective and abbreviated than some of his predecessors had been, and thus he helped to advance the genre toward stylistic maturity. He also placed a strong emphasis on the social nature of the haiga, as is evidenced by the notable number of gassaku produced with other poets and painters in his circle. His attitude reinforced the idea that the creation of haiga was an appropriate activity for a gathering of haikai poets. A painting subject could be assigned in much the same way as a topic for hokku could be announced.15 A collaborative work, especially one with associative linkage, could almost be regarded as a mixed media version of the linking process involved in a haikai no renga sequence.16 Finally, the sheer bulk of Bashp’s haiga would be enough to establish his place in the history of haiga practice, even without any consideration of who he was. On balance, it must be said, however, that some of those haiga are of interest primarily because they are by Bashp. Had they been produced by anyone else, they would occasion far less comment. While some are conceptually quite stimulating, just as often they are linked illustratively, with the linkage not very demanding or interesting. Moreover, in terms of visual appeal, Bashp’s early haiga can be painfully weak, and there are not many from any period that are exciting; he seldom strayed far from the Kanp-derived style he had learned from Kyoriku. Disrespectful though it may be to offer such opinions about so prominent a figure in the haikai tradition, Bashp’s haiga do not in general measure up to the strength of his verses. As much as he valued painting, he was still a poet first and a painter second. Nothing, however, that Bashp set his hand to has escaped the attention of those who came after him, from his immediate followers through succeeding generations of admirers and scholars. The mere fact that he created haiga, and not just on isolated occasions, therefore, gave to the genre a prestige that it might not otherwise have attained until the following century. It has even, sometimes, been Bashp who

J  O’M

214

has been given credit for the origin of the haiga form, in spite of the fact that examples of the genre were produced by a number of earlier haikai poets, including Arakida Moritake (1473–1549), Yamazaki Spkan (1464–1552?), and Hinaya Ryuho (1595–1669). Although some of Bashp’s individual haiga may fall short in terms of the execution of their visual imagery or the ingenuity of their linkage, his contribution was nevertheless tremendously important for the history of the haiga genre because his interest in the form gave it a legitimacy for future generations in a way that haiga by Moritake, Spkan, or Ryuho never could have done. Although the creation of haiga did not by any means become a universal practice among haikai poets, dozens of Bashp’s followers used the form, including eventually the poet–painter Yosa Buson (1716–1784), who is now generally regarded as the foremost practitioner of the genre. The abundance of haiga produced in the generations following Bashp, especially when compared with the frequency of their occurrence in the works of his predecessors, marks a perceptible shift in attitude and should be seen as one more measure of Bashp’s influence in making haiga an accepted, even an expected byproduct of the haikai discipline.

N 1. In the collection of the Bashp-p Kinenkan; reproduced in Okada Rihei, Zusetsu Bashp (Tokyo: Kadokawa Shoten, 1972), plate 84. 2. See, e.g., Imoto Npichi, Bashp. Haijin no shoga bijutsu, Vol. 2 (Tokyo: Shueisha, 1979), plates 19–25. 3. William Theodore de Bary, Carol Gluck, and Arthur E. Tiedemann, comps., Sources of Japanese Tradition, Vol. II: 1600–2000, 2nd ed. (New York: Columbia University Press, 2005), 351. 4. Ibid., 351. 5. For an example of a traveling figure painted by an anonymous Kanp school artist, see the collaborative work ( gassaku), with calligraphic verses and inscription by Bashp, that is reproduced in Leon Zolbrod, Haiku Painting (Tokyo, San Francisco, and New York: Kodansha International, 1982), 8–9 (plate 3). 6. The verse, with translation by Stephen Addiss, is as follows: Kara eda ni karasu no tomarikeri aki no kure

Crow perching on a withered branch— autumn evening

For further information, see Stephen Addiss, Haiga: Takebe Spchp and the Haiku-Painting Tradition (Honolulu, HI: University of Hawai’i Press, 1995), 30–31.

B ¯   H 

215

7. The former is in the collection of the Masaki Art Museum, Osaka, and is reproduced in Zolbrod, 11 (plate. 7). One of the latter haiga, of plum blossoms and the moon, is also reproduced in Zolbrod, Haiku Painting, 34 (figure 4). 8. Addiss, Haiga, 28–29 (plate 2). 9. This gassaku is discussed more fully, and Bashp’s inscription is translated in Zolbrod, Haiku Painting, 8 (plate 3). 10. Gourd-beating was a popular devotional practice which commemorated the anniversary of the death of Kuya (903–972), monk and traveling evangelist for the Pure Land sect of Buddhism in the Heian period. The commemorative practice was carried out in Kyoto during the last forty-eight days of the year and therefore had strong seasonal connotations that appealed to the practitioners of haikai. 11. Cleopatra Papapavlou, “The Haiga Figure as a Vehicle of Buson’s Ideals with Emphasis on the Illustrated Sections of Oku no Hosomichi and Nozarashi Kikp,” PhD dissertation, University of California, Berkeley, 1981, 27. 12. This detail from a longer handscroll with scenes from Bashp’s travels lacks a poetic inscription, and therefore is not, technically, a haiga, but the figure is very similar to the one found in a dated haiga of 1692, tenth month, that is in the Masaki Art Museum, Osaka; the latter has been cited earlier in note 7. 13. A number of these works, including several under consideration here, are reproduced and discussed in Zolbrod, Haiku Painting, 8–12, 33–35. 14. Current questions about authenticity for several haiga formerly believed to be by Bashp (a recently challenged example is Hibiscus/Mukuge, in the Idemitsu Museum of Arts; reproduced in Zolbrod, Haiku Painting, 11, plate 4) can be related to this question of Bashp’s influence in establishing the haiga genre. Had he been less highly regarded as a haikai poet, or less interested in the haiga genre, there would certainly have been less motivation for the production of haiga purported to be from his hand. 15. For example, at a gathering at Kyoriku’s in early 1693, the first full moon of the year prompted the choice of plum blossoms and the moon as the theme for poetry and paintings. Several haiga on that subject were produced before the night was over. One of them by Bashp is reproduced and discussed in Zolbrod, Haiku Painting, 34 (figure 4). 16. Relationships between the haiga form and Japanese poetic forms are further discussed in Joan Hertzog O’Mara, “The Haiga Genre and the Art of Yosa Buson (1716–84),” PhD dissertation, The University of Michigan, 1989, chapters I and II, and especially 105–108.

This page intentionally left blank



I  T  I   H  Stephen Addiss

T

he form of poem–painting called haiga has been a form of visual–verbal expression almost since the emergence of haikai poetry.1 There is no direct evidence of haiga’s first occurrence, but since a great number of haikai represent visual images in words, and also considering that poems and images in Japan were created with the same brush and ink, it must have seen a natural step to add paintings to haikai. Yet they are two very distinct forms of expression, with different capabilities and potentials, leading to a number of interesting questions. How did Japanese masters reconcile, or play upon, these differences? Most important, how did the addition of visual images reinforce, add to, or change the meanings of the poems? In order to explore these questions, the history of haiga will be dealt with here only briefly. Among early haikai masters who depicted visual images, Nonoguchi (Hinaya) Ryuho (1595–1669) is said to have studied painting with the famous artist Kanp Tan’yu (1602–1674). Ryuho is sometimes cited as the originator of haiga, in part because he inscribed upon one of his later scrolls that around the age of sixty he began a new practice of painting; whether or not he specifically meant haiga is unclear.2 His works are simple and relaxed, but in one respect they are not typical of later haiga because he usually added considerably more text than just the haikai poem, and his calligraphy is seldom visually integrated with the image. Instead, his introductions and poems often remain somewhat separate from the painting, such as at the top of a hanging scroll, or on the side of a horizontal format.3

218

S  A 

Other early haikai poets also created haiga, although less frequently than Ryuho. Like haikai itself, haiga became a major form of artistic expression with Matsuo Bashp (1644–1694). His paintings, which are discussed in more detail in chapter 9, have sometimes been criticized by art historians for lacking the purely visual values, such as sophisticated brushwork and bold compositions, of most Japanese professional painting. In addition, Bashp’s haiga have disappointed some critics by not seeming to add secondary meanings to his poems or by showing images beyond those mentioned in the haikai. Certainly, his modest and subtle haiga do not seek to impress viewers with technical prowess, bright colors, or dynamic brushwork, any more than his calligraphy dazzles viewers with bold or powerful forms. Nor does Bashp supplement the words of the poem with further imagery, as later haiga masters would often do. Nevertheless, his simple but flavorful paintings set the tone for most haiga that was to follow in the next half-century. Several followers of Bashp continued his lead in haiga as well as haikai, each adding an individual voice. Morikawa Kyoriku (1656–1715) had extensive training in the professional Kanp School tradition, and Bashp gave him credit for teaching him painting even as Bashp taught Kyoriku haikai. It may be well to take this kindly comment by Bashp with some caution, but there is no question that the master admired his pupil’s skill. The two did several joint works, with Kyoriku painting the image and Bashp adding the poem or poems.4 The restraint of Kyoriku’s haiga images makes them more interesting than his purely Kanp-style paintings, which are standard for artists in this tradition. This certainly suggests that Bashp’s influence extended to Kyoriku’s haikai paintings, not in artistic skill, but in the crucial element of artistic taste. Of all his many pupils and followers, Bashp seems to have been most fond of Enomoto (Takarai) Kikaku (1661–1707), despite their differences in personality. Where Bashp was serious, Kikaku was lighthearted, and his poems are highly admired for their karumi or lightness of touch. The son of a physician, Kikaku became Bashp’s pupil in his early teens, also studying painting, medicine, Chinese-style poetry, and Confucianism. His spirit was free and bold, and Bashp once commented that Kikaku “deals with trifling matters in a most eloquent way.”5 In one haiga Kikaku depicts the most tawdry of objects, a section of melon rind floating down a stream (figure 10.1). Uri no kawa mizu mo kumode ni nagarekeri

Melon skin— spider-legs floating on the water

I  T  I   H

219

Figure 10.1 Enomoto Kikaku (1661–1707), Melon Skin, ink on paper, 31  44.4 cm, Shpka collection, Midlothian, VA.

The painting helps to illuminate as well as interpret the scene, as Kikaku shows that someone had cut the melon in sections in order to eat its fruit; now these sliced sections of skin gradually spread out and bob on the water. But there is more in the poem: kumode means spiderlegs, but it can also mean crosswise or in various directions, so a second possible translation might be: Melon skin— the water also flowing in crisscross directions

Perhaps no other form of visual art has had such modest ambitions— who but a haikai poet would paint a discarded melon skin? And yet this unglamorous subject, simply garnished with freely spontaneous calligraphy that flows over the image, allows Kikaku to express his fresh poetic vision of ordinary life. The image does not depict anything not mentioned in the poem, but it adds a fresh view to the subject, a visual “take” that contributes another level of meaning to the haikai.

220

S  A 

Nakagawa Otsuyu (also called Bakurin, 1675–1739) was another follower of Bashp who created a number of haiga, including a simple image of a Deer (figure 10.2) with the following poem: Shika no ne no todokanu yama wa mada aoshi

The mountain no deer’s cry has reached is still green

Otsuyu’s haikai can be compared to a much earlier tanka by the courtier Pnakatomi no Yoshinobu (active c. 990): Momiji senu tokiwa no yama ni sumu shika wa onore nakite ya aki wo shiruramu

The deer who live without maple leaves on old pine mountain can know that autumn has come only by their own cries

In the tanka, the scene is clearly spelled out: with no maple leaves to turn red, the deer create their own sense of autumn. The haikai, on the other hand, suggests more than it defines. Is Otsuyu implying that the cries of the deer are not responses to autumn colors, but part of

Figure 10.2 Nakagawa Otsuyu, (Bakurin, 1675–1739), Deer, ink on paper, 27.7  37.6 cm, Beckett collection, Denver Colorado.

I  T  I   H

221

their cause? Botanists would object, but as an expression of feelings of loneliness and loss as the season changes, the cries of deer are certainly a powerful image. Otsuyu has merely painted a deer and a simplified form that can be read as a mountain, but the manner in which he depicts them is significant. First, the forms are created with the same blunt, tonally varied brushwork. Second, they seem to merge into each other; the body of the deer could easily be understood as part of the mountain but for the darker brushstrokes that define its antlers and legs. Third, these lines relate to both the somewhat more gossamer calligraphy just to their left, and the more strongly inked signature in the lower left. Thus, the mountains become the deer, the deer becomes the poem, and the deer and poem become Otsuyu. To emphasize the interconnection, the poet–artist has unified the entire work into a bowl-shaped composition that could also be seen as leading the eye in a pendulum movement. Once again, nothing has been painted that is not mentioned in the haikai, but by his unification of poetic and visual imagery, Otsuyu brings us a moment in which we may sense the interaction and interpenetration of deer, mountain, poem, and ourselves. A third follower of Bashp who often painted haiga was Kakujp (1664–1747), who had been adopted as a youth by Senna (1651–1723), the haikai poet, painter, and abbot of Honpuku-ji in Shiga. Bashp had often visited Senna at his temple, and may have learned some aspects of painting from the artist–monk while sharing his knowledge of poetry. Kakujp, continuing Senna’s tradition, became a noted painter–poet and the thirteenth abbot of Honpuku-ji. In his image of Sailing on the Blue Sea (figure 10.3), Kakujp has added a new element, a figure that strongly resembles the god of good fortune, Hotei. The poem is largely celebratory: Aoumi ya kyp no suzumi mo iso senri

The blue sea— Kyoto’s coolness also 1000 miles of seashore

The composition is masterfully arranged on a strong diagonal, with the boat moving rapidly through space while the calligraphy flutters in the breeze on the right. Hotei’s mouth, open seemingly in wonder, is cleverly echoed by the artist’s kao (cypher-seal) in the lower right-hand corner. The signature over the kao is also amusing since it literally means “above the corner.” However, these amusements are kept in check by the strong composition, as well as by the bold and confident brushwork of both image and poem. Perhaps most significant

222

S  A 

Figure 10.3 Kakujp (1664–1747), Sailing on the Blue Sea, ink on paper, 26  36.7 cm, Beckett collection.

is the use of empty space in the center of the format, which allows us to enter into the scroll and enjoy the cool breeze for ourselves. Kaga no Chiyo (1703–1775), the daughter of a mounter of painted scrolls, began composing haikai in her youth, so she was wellprepared for creating and sometimes joining images and words. She studied with two of Bashp’s leading pupils, Kagami Shikp (1665–1731) and Otsuyu, but when she was in her thirties, the deaths of her parents and older brother left her the responsibility of carrying on the family business. As she neared the age of fifty, Chiyo adopted a married couple and turned the scroll-mounting shop over to them. She then became a nun in 1754, and was once again able to devote much of her time and attention to haikai. She also became a master of haiga, combining image and text in complex and subtle ways such as in Late Spring (figure 10.4): Fuji no hana nagpte tsure ni okurekeri

As wisteria grows long, meeting companions is delayed

223

Figure 10.4 Kaga no Chiyo (1703–1775), Late Spring, ink on paper, 96.3  17.7 cm, Yabumoto collection, Tokyo.

224

S  A 

The poem creates a nice ambiguity—is the poet so fascinated with the flowers that she is late meeting her friends, or is she describing a tendril of wisteria that has not caught up with its fellows? Chiyo’s composition is equally multilayered. The flowers fan out in the center of the composition with calligraphy above and below interacting with the image, as tones of ink ranging from black to grey in the leaves contrast with the deeper black of the writing. The long trailing vine in the right center divides the final word of the poem, as though it were entwining both the poet and the blossoms in order to keep them from meeting their companions. In examining these four works by followers of Bashp, we can note a number of common features. First, they seldom go beyond the text in their visual images, although Kakujp added a figure not specifically mentioned in his poem. Second, the brushwork tends to be informal and modest, rather than precise and polished. Third, these works are executed completly in ink; there are some exceptions in the haiga of Bashp and his followers, but even in these, the colors are usually soft and unobtrusive instead of bright and bold. Fourth, the calligraphy of the poem is always a prominent feature of the composition, completing the visual design rather than merely adding to it; this is a change from the earlier haiga tradition of Ryuho. Finally, simplicity is vital to these images, with empty space an important element in the total design. For fifty years after Bashp’s death, haiga generally followed the lead that he had established. However, when in the late eighteenth century the Bashp poetic tradition faltered in the face of pervading tendencies toward wit and cleverness, haiga also reached a crossroads. New trends in art such as literati painting, as well as new attitudes toward poetry, were to combine with changes in society to challenge the modest and refined haiga tradition of Bashp and his followers. The poet–painter Yosa Buson (1716–1784), although he wrote often about reviving Bashp’s poetic spirit, took haiga into entirely new directions. Like Takebe Ayatari (1719–1794), Buson followed the lead of his mentor Sakaki Hyakusen (1698–1753) in studying literati painting; he also learned the elaborate Nagasaki School techniques of colorful and naturalistic bird-and-flower painting, partly as a means of earning a living. By the time he reached sixty, Buson was a master artist, with a variety of styles, subjects, and techniques at his service. It may not be surprising, then, that in his haiga Buson tended to give more space and attention to the images than had earlier poet painters, and the great fluency of his brushwork is always apparent.6 Some of

I  T  I   H

225

his scrolls go beyond the simplicity and modesty established by Bashp, turning haiga into a more purely painterly medium. This led to two consequences: first, haiga became a form of art that was enjoyed by a larger audience than before, including lovers of the visual arts who were not necessarily conversant with haikai traditions. Second, haiga developed into a genre that could be practiced by professional artists, whether or not they were themselves poets. Buson utilized haiga in several significant ways. He created his own place in the haikai succession by painting many portraits of Bashp, as well as several versions of the master’s travel journeys. He copied the complete Oku no hosomichi several times with haiga-style paintings, usually on handscrolls but also once upon a screen. To many viewers, these delightful works represent a high point in the haiga tradition, although they might more accurately be termed haibunga (poetic prose paintings). In depictions of Bashp’s travels, Buson almost always depicted the master himself rather than natural images in his prose and poetry.7 In addition, Buson honored other haikai poets of previous decades by painting an album that was later used as the model for the woodblock book Haikai sanjurokkasen (Thirty-Six Haikai Masters). First printed in 1799, it includes a portrait of the nun Chigetsu (1632–1706), which will be discussed later. In all of these works, Buson was utilizing visual art to establish his position as interpreter of haikai history and successor to Bashp. Although the field of haiga had now been opened to professional artists and some haikai masters also undertook professional training as painters, the tradition was certainly not lost to amateur poets. Most of the finest haikai masters of later ages continued to add images to their poems, often in a seemingly artless manner, and therefore after the time of Buson two main streams of haiga emerged: those by painters and those by poets. These two streams sometimes joined in the works of individuals who were highly skilled in both arts, but viewers can usually detect whether a haiga has been created primarily as a painting with an accompanying poem, or remains a poem that has given rise to an image. Stylistically, most earliest creators of haiga had followed the painting practices of the dominant Kanp School tradition of the seventeenth and early eighteenth centuries, with its Japanese codifications of Chinese ink-painting techniques.8 Buson took haiga brushwork in a new direction, in which a free calligraphic style associated with literati painting replaced the more formal and descriptive brushwork of the Kanp tradition. However, after Buson’s death, his most important

226

S  A 

pupil Matsumura Goshun (1752–1811) came under the influence of the naturalistic painter Maruyama Pkyo (1733–1795) and began to paint in a softer and more impressionistic style.9 Goshun’s work had a strong influence on later artists, so in the early nineteenth century the mainstream of professionally painted haiga changed once again, this time from calligraphic brushwork to a more naturalistic mode that featured forms created by washes of tonally varied ink or soft colors. Professional artists of other schools, such as the boldly decorative Rimpa school, also created haiga from time to time, so it may be observed that nineteenth-century haiga reflected a variety of artistic traditions.10 Nevertheless, certain stylistic features particularly characteristic of the haikai painting tradition remained. Prime among these was the simplicity mentioned earlier, which allowed viewers to add their own interpretations to the paintings as well as to the poems. The history of haiga, its artistic features, and its integration of painting and calligraphy are all worthy of further study.11 The focus of the remainder of this chapter, however, will be on the question of how visual images interact with written text in the work of poet–painters. Just what do the paintings add to the haikai, and how do they do it? Continuing to use images from a 1995 haiga exhibition seen at five venues in the United States,12 we can trace a number of different potential amalgams of text and image in a cross-section of Japanese haiga. What seems to be the least complex combination occurs when the image is that of the poet whose haikai is written next to his or her portrait. Of course, poet-painting was nothing new to Japan. Paintings of famous tanka (waka, thirty-one-syllable verse) poets had been created for hundreds of years, especially by artists associated with the court who featured delicate brushwork, detailed compositions, and bright colors. Images of the “thirty-six” or “one hundred” great tanka masters became so popular that such portraits even appeared on card and shell games, where participants had to match the poets with their most famous verses. Partly in emulation and partly in parody of this tradition, portraits began to be painted of haikai poets, but without such refined colors and brushwork. Instead, the haikai masters were almost invariably shown very simply in free brushwork, using ink or light colors. Rather than appearing as noble monks, refined courtiers, and exquisite court ladies, they were depicted as everyday people. Even the most famous and frequently portrayed haikai master has usually been shown without visual honorifics, as can be seen in more than 250 portraits of Bashp gathered in the Inui Collection in Shiga, which have been published by the collector in two volumes.13

I  T  I   H

227

We may see many typical characteristics of Bashp portraits in a small painting by Ki Baitei (1734–1811), an artist who had studied with Buson. Here, Bashp is portrayed in his most typical pose, seated facing the viewer with his hat and traveling box, ready for a poetic journey (figure 10.5). The master is depicted in relaxed outlines and tones of grey, but Baitei gives equal space to calligraphy, writing out a pair of introductions and poems referring to trips Bashp took with or without his haikai friend Etsujin. Curiously, despite Baitei’s introduction, Etsujin actually accompanied Bashp on the first journey as noted in Oi no kobumi (Notes from a Knapsack).14

Figure 10.5 Ki Baitei (1734–1811), Portrait of Bashp, ink and light colors on paper, 30  26.9 cm, private collection.

228

S  A 

This time, they say he didn’t take Etsujin along: Samukeredo futari tabine wa omoshiroki

Even when it’s cold it’s more interesting to travel together

Intending to send it to Etsujin, the next year he composed this: Futari mishi yuki wa kotoshi mo furikeru ka

The snow we saw together does it fall this year as well?

The two poems about friendship both contain the word futari, literally meaning “two people” but here signifying “together” or “the two of us.” Yet, Baitei has only pictured the single figure of Bashp. Who is the invisible second person in this haiga, the friend of the poet? It may be Etsujin, but if we can infer the presence of the artist, it might also be Baitei, who honored Bashp with this portrait. But couldn’t it also be any one of us? Bashp’s direct gaze helps to involve the viewer in the scene, a feature of many of the best haiga. Baitei’s painting, on one level, adds nothing to the poems but the image of their author. On another level, however, there are a number of visual features that both supplement and suggest interpretations of the haikai by stressing certain aspects of their meanings. For example, the two Chinese characters “two people” which can be pronounced futari are twice written in large size at the top of calligraphic columns, stressing their importance. Furthermore, the inscription hovers next to the poet, echoing the diagonal of his shoulder, and Baitei’s brushwork is very similar in both painting and writing. Finally, the pose of Bashp facing the viewer helps to suggest that the poems are not only about his relation to Etsujin, but extends to all empathetic readers of his haikai. Comparing another poet portrait, this time by Baitei’s teacher Buson, may help to point out the similarities and differences within the genre (figure 10.6). From his album of thirty-six haikai masters Haikai sanjurokkasen, Buson integrated the calligraphy of one of Chigetsu’s haikai poems with the curving outlines of her face and robe: Toshiyoreba koe mo kanashiki kirigirisu

When it grows old its voice becomes plaintive— katydid

At first this image may look very similar to that of Baitei, but there are small differences that are significant as artistic expression. Unlike

229

Figure 10.6 Yosa Buson (1716–1784), Portrait of Chigetsu, 1799, woodblock print, ink on paper, 26.8  19.3 cm, private collection.

230

S  A 

Bashp, Chigetsu looks down, as though listening to the weakening call of the insect as summer ends. She seems less substantial, less grounded upon the earth than Bashp, due to the curving lines that define her and the lack of internal modeling. Both portraits convey a feeling of melancholy appropriate to the haikai, but while Baitei interprets Bashp as relating to human friends, Buson shows Chigetsu in a more solitary guise and evanescent sensibility. If images of poets can do more than merely illustrate the writers of the haikai, what about images that are created both in the poem and the painting? Do they merely reinforce each other, or can they add new shades of meaning? Let’s first consider haiga in which the pictorial aspect seems to mirror the poem. Two minimalist images can take us into the intertextuality of poem and painting. The first is by the Nagoya doctor and poet Inoue Shirp (1742–1812) who has evoked an autumn evening (figure 10.7). Yorozu yo ya yama no ue yori kyp no tsuki

Through the ages rising above the mountains— tonight’s moon

The final word of the poem, tsuki, both symbolizes and pictorializes the moon. Shirp slightly separates this word from the rest of the

Figure 10.7 Inoue Shirp (1742–1812), Moon, ink on paper, 31  50 cm, private collection.

I  T  I   H

231

text, enlarges it, and bends it at an opposite diagonal to the shape of the single-line painting. But what is this image? Can we be sure what it represents? Let’s look more closely. The contour moves up from Shirp’s signature at the lower left, curves, moves back downward, curves again, and bends gently as it descends to the lower right. By creating a smaller and then a larger diagonal, it gives a sense of movement to the composition. Furthermore, the brushstroke is not limited to a single tone of ink, instead revealing darker and lighter greys within its single stroke. Dividing the total space roughly in half, this image allows just enough space for the calligraphy in the upper right. Because Shirp has given us both a verbal and a visual image of the moon rising over the singlestroke painting, it can become a mountain in the mind of the viewer, fully uniting text and image. An even simpler haiga on the theme of the moon was created by Takebe Ayatari (1719–1794). Known as a painter by the name Ryptai, he was one of the most versatile poet–artists of early modern Japan.15 Born to a samurai family in northern Japan, he left home at the age of twenty, reportedly because of a scandalous affair with the wife of his older brother; forced to fend for himself, he spent his life devoted to the arts, and became a significant figure in the world of poetry and art in the second half of the eighteenth century. Like Buson, he followed the lead of his friend and mentor Sakaki Hyakusen in studying Chinese literati painting, but in Ayatari’s case, he maintained an extremely simple style in his haiga, never letting image take precedence over the text. In a work from Ayatari’s middle years, a single, relaxed curving brushstroke in varied grey inktones creates a moon in one corner of the composition, while the boldly meandering calligraphy of the poem fills the rest of the space (figure 10.8). Yoru naraba abunaki hashi ya kyp no tsuki

In the evening the bridge becomes dangerous— tonight’s moon

Ayatari leaves much to the viewer: we may well ask, why exactly does the bridge become dangerous? Visually, is there anything in the image that enriches the range of our possible responses? Does the fact that Ayatari has painted nothing but the moon—no bridge, no people— leave us no choice but to imagine them? Does the faintness of the image lead us to look all the harder (rather than watching where we are going)? Does the slightly unsteady tilting of the calligraphy indicate

232

S  A 

Figure 10.8 Takebe Ayatari (1719–1794), Moon, ink on paper, 30.5  48.6 cm, Shpka collection.

the inebriated state of the viewers? Why does the poem end with fading brushstrokes, as though the moon itself may disappear? Haikai poets (haijin) and professional painters were not the only ones to create haiga. The man who has been generally acknowledged to be the most important Zen Master of the past five hundred years, Hakuin Ekaku (1685–1768), also united haikai and image in several of his Zen teachings through brushwork. One of his more complex scrolls suggests the presence of Hotei, the happy-go-lucky god of good fortune, through his attributes of bag, fan, and staff (figure 10.9). The poem, however, seems to question the image: Neta uchi wa kami ka hotoke ka nonobukuro

While sleeping, a Shinto god? a Buddha? —just a cloth bag

Humor, not often encountered in religion, is a special feature of Zen, and Hakuin’s work is full of verbal and visual puns. The first line of the poem means “while sleeping,” but it can also mean “a leaning round fan.” Therefore, the main image can be seen not only as Hotei’s bag, but also as a round fan put down to rest at an angle. Furthermore, as Joshua Mostow has pointed out, “Mr. Nono” (nono-sama) is Japanese baby talk for gods and buddhas.16 Hotei’s name itself can

I  T  I   H

233

Figure 10.9 Hakuin Ekaku (1685–1768), Bag of Hotei, ink on paper, 42  51.2 cm, private collection.

mean “cloth bag,” but where is Hotei himself? Sleeping behind the bag? Or is this the kind of Zen riddle called koan? Behind the series of meanings in the scroll is the understanding that even Hotei, whom we may consider a god or a Buddha, is really just an ordinary human who has discovered his inner Buddha nature. Hakuin confronts us with a close-up view of a paradox, and therefore while this work qualifies as a haiga, it is even more a Zen painting. Some critics would even dispute that combinations of haikai and image done by Zen Masters are haiga at all, since the spirit is somewhat different. While haiga are usually subtle, suggesting more than they define, Zenga tends to present images that are a great deal more powerful and insistent. Since most Zenga have inscriptions in Chinese rather than haikai, this question does not come up very often. Nevertheless, if our definition of haiga is the combination of one of more haikai with a visual image, works by monks such as Hakuin must qualify, and there is no reason that a scroll may not be both Zenga and haiga simultaneously.17 So far we have been studying examples of haiga in which there is a clear interconnection between poem and painting, but there are other

234

S  A 

works in which the relationship is less obvious. Two examples of how images can add to the humor of a haikai come from the works of Takebe Spchp (1761–1814), a leading haijin in Edo during the early years of the nineteenth century. One of Spchp’s most delightful haiga shows four old men along with two women and a baby clustered under a mosquito net, a multigenerational Japanese slumber party (figure 10.10). The poem emphasizes the possible disadvantages of a rural lodging: Susuki kara ka no deru yado ni tomarikeri

From the pampas grass mosquitoes came into the inn where I stayed

The painting adds considerably to the haikai. Spchp has arranged the figures in a semicircle extending into a horizontal like the shape of a ladle; the lumpy form of the final body is echoed in visual parody by the lamp and the sedge hat. Behind them, the women are asleep, but the baby looks on as the four old geezers smoke, wave a fan, and carry on their conversation. The purpose of both text and image is to show an ordinary moment and make it special: this haiga evokes the special flavor of summer at a Japanese inn. Another painting by Spchp (figure 10.11) was inscribed by his friend Tsuji Rangai (1758–1831) with a poem by Spkan (1465–1553),

Figure 10.10 Takebe Spchp (1761–1814), Discussion Under a Mosquito Net, ink and light colors on paper, 34  55.8 cm, Masuda collection, Tokyo.

235

Figure 10.11 Takebe Spchp (1761–1814), Strange Figure, ink and light colors on paper, 117  26.1 cm, Masuda collection, Tokyo.

236

S  A 

one of the first major figures in the transition from renga to haikai. Credited with inventing the form of linked verse called haikai renga, Spkan has also been admired for studying Zen, retiring from his life as a samurai, and living as a monk at his modest hermitage. Here, his haikai shows a lightness of touch that may seem surprising for a man of artistic and spiritual perception: Tsuki ni e o sashitaraba yoki uchiwa kana

If you put a handle on the moon— a good round fan

Typical Zen images often show a monkey trying to catch the reflection of the moon in the water, a symbol of the futility of trying to cling to the pleasures of life. However, if the moon can never be grasped, how can it be given a handle and made a fan? When we come to Spchp’s painting, further complications arise. Who is this strange figure? The most obvious answer would be Spkan, since many haiga show the poet rather than an image from the poem. But Spkan was a monk and this strange figure is not. Furthermore, he is looking rather dismayed, and the fan he is holding is not the round Chinese-style uchiwa type but the more common Japanese folding fan. With his body composed of wriggling lines, his toes curled up in concentration, and his long nose, he almost looks like a tengu, the semi-humorous Japanese demon. Could this be a poet of Spchp’s generation, trying to puzzle out the meanings in Spkan’s poem, or wishing to compose a haikai as delightful and multilayered? Might Spchp and his friend Rangai be adding one more layer of humor to Spkan’s famous verse? Another oddly humorous haiga was created by Inoue Shirp (figure 10.12). Instead of a clearly defined image, we are offered a strange shape that is difficult to read. Is it the profile of a face with whiskers hunched over a stubby body? Is it, in fact, a self-portrait of Shirp? Could it even be the portrait of his sneeze? If so, the artist is certainly being self-depreciating, and his poem echoes this playful mood. Iro mo ka mo nakute hana miru kazahana kana

No color or scent when flower-viewing— stuffy nose

The word kazahana is written as “stuffy nose,” but with different characters it could also mean “snowflakes.” The haiga may simply

237

Figure 10.12 Inoue Shirp (1742–1812), Self-Portrait, ink on paper, 114.4  27.6 cm, private collection.

238

S A 

indicate that when Shirp has caught a cold he cannot appreciate the flowers, but it also suggests that pure white snowflakes are also like tiny blossoms that have no color or scent. By making the visual image ambiguous, rather than depicting any direct motif of the poem, the poet–painter has offered us a range of interpretations that is greater than what either a poem or a painting could offer singly. In the most complex haiga relationships, the visual image is not encountered in the text at all. This gives rise to questions about the relationship of painting and poem, further involving the viewer in the artistic process and transmission of meaning. One fascinating example is by Kobayashi Issa (1763–1827), with one of his best-known poems (figure 10.13). Niwa no chp ko ga haeba tobi haeba tobu

Garden butterfly as the baby crawls, it flies— crawls close, flutters on

Figure 10.13 Kobayashi Issa (1763–1827), Garden Butterfly, ink on paper, 28.6  37.6 cm, Shpka collection.

I  T  I   H

239

The young child chasing the butterfly that always flutters just beyond reach suggests both the beauty of nature and the futile pursuit of pleasure. As such, the poem can be taken for a Buddhist lesson in not grasping. The haikai spirit, however, suggests rather than proclaims, which may be why Issa depicted neither the child nor the butterfly in his painting. The hut sets the scene, and perhaps the movement of the calligraphy may suggest to us the flight of the butterfly, but there is no further need to pictorialize—if we try to catch the butterfly in our minds, it is gone. A final example of a painting that does not directly depict an element from the haikai occurs in the work of Fujimori Spbaku (1758–1821), a merchant friend of Shirp, who became a Buddhist priest. Spbaku enjoyed creating haiga as well as haikai, being especially noted for his figure paintings (figure 10.14). Here, he has shown an old couple at their little brazier; the caricature of the faces is one of the characteristics that appears in many of Spbaku’s haiga, helping to keep them from becoming banal or pretentious. The title Spbaku gave this scroll, Dawn of the Thirteenth Night, relates to the passion for moon-viewing on the thirteenth night of the ninth month, which was considered to have the brightest moon of the year. Kakugo shite yami wa akarushi jusan ya

Once you know that darkness is also bright— the thirteenth night.

From the painting we might not be able to guess the meaning of the poem, and vice versa, since the two arts bring forth images that complement, rather than describe, each other. We certainly need not insist upon one composite meaning in the total work, but there are several ways in which the interaction between text and image can be seen. On the simplest level, the old couple is willing to go out in the dark night because they know the moon will be bright. On another level, their brazier creates enough light for them, and they don’t really need the moon. More broadly, in the world of haikai as in Zen, brightness is in the eyes and mind of the beholder. The history of haiga does not end in the early nineteenth century, but the main trends noted here have not significantly changed until the present day. Western artistic influences have been only peripherally felt in this artistic genre, although a tendency to brighter colors and fuller compositions can sometimes be seen. Haiga are still being produced in Japan, and there are even “how-to” books that inform amateurs how to add images to their haikai poems. The future direction

240

Figure 10.14 Fujimori Spbaku (1758–1821), Dawn of the Thirteenth Night, ink on paper, 114 28 cm, private collection.

I  T  I   H

241

of haiga is uncertain, however, since it depends on the future of haikai itself, which must contend with powerful social influences that lead potential poets away from empathetic observations of nature and the special flavors of the four seasons. Instead of the traditional cycles of rural existence, contemporary urban and industrial life is deluged with mass entertainment through radio, photography, recordings, television, film, and video. Perhaps, as haikai poetry continues to serve as a counterweight to the pressures of the modern world, haiga will continue to be a creative form of visual–verbal art based on the extraordinary awareness of ordinary life.

N 1. Portions of this essay are adapted from the catalogue by Stephen Addiss, Haiga: Takebe Spchp and the Haiku-Painting Tradition (Richmond: Marsh Art Gallery, University of Richmond, and Honolulu: Hawaii University Press, 1995). 2. See Akira Morikawa, ed., Haijin no shoga bijutsu, Vol. 1 (Tokyo: Shueisha, 1978), image 13 and 96–97. 3. Ibid., images 13–23. 4. Two of these joint works are illustrated in the haiga catalogue, Addiss, Haiga, 28–31 and discussed by Joan O’Mara in chapter 9, 202–207. 5. Quoted in “Conversations with Kyorai” in (Kyoraishp) by Mukai Kyorai (ed.) Donald Keene, Anthology of Japanese Literature (New York: Grove Press, 1960), 379. 6. See, e.g., the images in Okada Rihei, Bashp, Buson (Tokyo: Benseisha, 1976); and in Haiga no bi: Buson. Gekkei (Kyoto: Hpshobp, 1973). 7. I noticed this especially when I was asked to create my own ink paintings for a translation of this text by Sam Hamill, published by Shambhala as Narrow Road to the Interior. I found that I most often painted visual images suggested by the text rather than the poet himself, and this made me realize that Buson had a point to make: he was the true successor to Bashp. 8. Bashp’s paintings show some traces of Kanp School influence, and his poetry pupil and possible painting teacher Morikawa Kyoriku (1656–1715) studied with Kanp Yasunobu (1613–1685). 9. Since Goshun lived on Shijp (Fourth Street) in Kyoto, his school became known as Shijp, and the combined tradition of Pkyo and Goshun coalesced into the Maruyama–Shijp School. 10. In particular, the Rimpa master Sakai Hpitsu (1761–1828) was a fine haikai poet, and he created a number of haiga. 11. The scholar Audrey Yoshiko Seo is currently working on some of these questions; she presented a paper entitled “Painting Calligraphy

242

12.

13.

14. 15.

16. 17.

S  A  Interrelationships in Haiga” at a symposium on haiga at the University of Richmond on March 25, 1995. The exhibition “Haiga: Takebe Spchp and the Haiku-Painting Tradition” was shown at the Marsh Art Gallery of the University of Richmond, the Spencer Museum of Art at the University of Kansas, the University Art Museum at the University of California in Santa Barbara, the New Orleans Museum of Art, and the Smith College Museum of Art in 1995 and 1996. See Inui Norio, Bashp-p no shpzp hyakuei (One Hundred Portraits of Bashp); (Tokyo: Kprinsha, 1984) and Bashp-san no kambase iroiro (Many Faces of Bashp); (Hikone: Muboan Bunko, 1966). I am indepted to Eleanor Kerkham for this information. Why Baitei thought that Etsujin had not been along on this journey is unknown. For a thorough study of Ayatari, see Lawrence E. Marceau, Takebe Ayatari: A Bunjin Bohemian in Early Modern Japan (Ann Arbor MI: Center for Japanese Studies, University of Michigan, 2004). Joshua Mostow pointed out this pun during a lecture at the Seattle Art Museum in 1991. Among other major Zen Masters who created Zenga–haiga are Sengai Gibon (1750–1837) and Nakahara Nantembp (also known as Tpju Zenchu, 1839–1925).



B’ B  : ¯ T E   I  Eri F. Yasuhara

I

f Bashp is revered as the premier poet of haikai, for many, Yosa Buson (1716–1783) occupies a similar, only slightly less exalted position as “number two.” To this duo is usually added Kobayashi Issa (1763–1827) to make up the requisite triumvirate of Tokugawaperiod haikai literature. However, it is the first two, Bashp and Buson, who are most frequently paired in discussions, usually in comparison. The most famous such discussion was also the first in the modern period: Masaoka Shiki’s essays “rediscovering” Buson and comparing him favorably to Bashp.1 Perhaps because this kind of pairing and comparison are so standard and pervasive in Japanese scholarship, and because Buson’s debt to Bashp is considered almost a matter of historical fact, the relationship has not often been subjected to critical analysis. What, exactly, was Bashp to Buson? What did the former represent to the latter? Did he really “influence” Buson’s poetry, and if so, how? In general, the study of influence—defined in one literary handbook as “the impact that a writer, a work, or a school of writers has on an individual writer or work”2—has not occupied an important place in Western literary studies for some time. It is, however, part of the dominant discourse of Japanese scholarship on haikai literature, in which influence studies may be said to be subsumed under a much larger “search for origins” mode of scholarship. Such things as preexisting texts, the biography of the writer, and background—familial, historical, cultural, political, social, economic—are scrutinized to determine where something “came from,” whether it be a particular phrase, image, idea, or an entire work. The search for allusions, on the one hand, and lineage studies, on the other, are two of the more

244

E  F. Y 

familiar activities in this mode of inquiry, which is by no means restricted to haikai studies. Given this context, and given the tendency in traditional Japanese literary studies to privilege a writer’s own words on such subjects, it is no wonder that the Bashp–Buson influence relationship is more or less taken for granted in Japanese scholarship. Both Buson’s own words as well as the slogan of the mid-eighteenth century haikai revival movement of which he was a major figure explicitly acknowledged the debt to the earlier poet.3 The general view of haikai in the late seventeenth and early to mid-eighteenth centuries is that it was a period of overall decline following the death of Bashp. Haikai poets split into sometimes quarrelling factions, each touting his own style as the “true” successor to the master, Bashp. Two major trends were the vulgarization of haikai in the cities and a kind of reductionist simplification leading to banality in the rural districts. There were signs of resistance to these developments as early as the 1730s, and by the 1770s efforts by many poets in different regions of the country to revive haikai were in full flower. They called for the return of haikai to the “true way of Bashp” (shpfu the “true style,” which was defined as Shpfu, the “Bashp style”), that is, haikai of seriousness and depth. One constituent of the historical context within which Buson wrote, therefore, was the explicit acknowledgment of Bashp as the standard to which poets ought to aspire. Within this context, moreover, Buson himself explicitly acknowledged the importance of the earlier poet in the formation of his own poetics. He told his disciple Kitp, for example, that during the period of his travels through the eastern provinces as a young man, he “sought the depths of Bashp’s spirit and composed refined poetry after the lofty manner of Minashiguri (Empty Chestnuts, 1683) and Fuyu no hi (Winter’s Day, 1684).”4 For this, however, he was regarded as a heretic and reviled for being a young man with “old” poetic taste. This conscious positioning of Bashp as his poetic ideal continued even as Buson came into his own as a poet. In the 1774 preface to Bashp-p tsukeaishu (A Collection of Master Bashp’s Linked Verse), he declares that the best way to learn how to link verses properly is to memorize Bashp’s works. To drive home this point, he then says, “Should one go three days without reciting the Master’s verse, one’s mouth would sprout thorns.”5 If Buson revered Bashp as poetic ideal or inspiration, some of his haiku and haibun (prose in haikai style) express a concomitant sorrow at what he perceived to be his distance from the earlier poet’s times, lifestyle, and art. Two haibun, written about six years apart, take up

B ’ B  ¯

245

such themes of despair at the inability to live a life completely dedicated to poetry, as Bashp supposedly did. The first is Saimatsu no ben (Year’s End), dated 17766: We make even more miserable our short existences by running helterskelter after fame and profit, drowning ourselves in a sea of greed and avarice. Our behavior on the last night of the year is particularly abhorrent. We go around banging on people’s doors, cursing them and raising an unholy ruckus—truly hateful!7 Yet, how can this fool escape from the vulgarities of this world? Bashp said, “On the road / with straw hat and sandals / I meet the year’s end.”8 Reciting this verse to myself in a quiet corner, my heart is calmed and purified, and I find myself wishing I, too, could live the life he did, beyond the concerns of this world. This poem has become my personal sacred text. Bashp is gone and there is no other Bashp; the years just come and go. Bashp sarite sono nochi imada toshi kurezu

Bashp is gone, and ever afterward, no year has ended as his did.

He has not been able to end any year the way Bashp did, on the road in pursuit of art. Another haibun, entitled Hinokigasa no ji (The Cypress Hat) and written in 1782, echoes even more darkly this sorrow at not being able to live a life dedicated to art, detached from the mundane.9 “I will show you the cherries, my cypress hat!” I have not followed the way of the Master [Bashp] as he hurried on his journey to Yoshino with those words. It is common enough for people to end up just staying at home, suffering the hardships of this sorrowful world, wondering “What shall I do about that? How shall I handle this?” never accomplishing the plans they first had, and finally even turning away from the beauties of nature. Yet I feel it is I alone who have thus failed and haven’t the courage to face anyone. Hana chirite mi no shita yami ya hinokigasa

The blossoms have fallen— Around me all is darkness, My cypress hat!

Whether we take such protestations literally or more figuratively, as a literary pose, no one would ever characterize Buson as a “failed Bashp” or even as a pale imitation of the poet whom Buson held as his ideal. It need not even be said that Buson is a premier poet in his own right, who left a body of work as interesting, engaging, and complex

246

E  F. Y 

as Bashp’s. There are, in fact, passages that suggest that Buson’s stance toward his acknowledged ideal was a more complex one than some of the earlier examples might at first suggest. In the preface to the collection Mukashi o ima (The Past in the Present) written in 1774, Buson tells of a memorable lesson learned from his master Spa (Hayano Hajin, 1677–1742) when as a young man Buson lived with him as uchi-deshi or live-in disciple.10 One night, Master Spa drew himself up and said to me, “The way of haikai does not lie in always following one’s teachers’ methods. One must change one’s style with the times. [The creation of ] haikai must be spontaneous, without regard for what went before or what is to come.” With these words I gained sudden enlightenment as if under the blows of a Zen master’s stick, and felt I had grasped to some degree the self-sufficient freedom of haikai. Therefore, my teaching to my own disciples is not to imitate Master Spa’s unaffected style, but to venerate Bashp’s sabi (subdued style) and shiori (delicate style) and think only of returning to those long ago days. This may appear on the surface to be going against my master Spa, but in reality it is the way of conforming to his wishes. This is called the zen of haikai and the technique of direct communication of mind with mind.

On one level, this seems to be another example of Buson’s placement of Bashp at the pinnacle of poetic achievement as someone to be emulated, even at the risk of denying his immediate master Spa. The wording, however, opens up the possibility of validating a similar gesture vis-à-vis Bashp himself. If a disciple may outwardly deny his immediate master in order to conform to that master’s true, inner teachings, may not the same disciple deny the poet he reveres as his ultimate master, Bashp? Buson, in fact, does do so, although “deny” is perhaps too strong a word. In his preface to the New Year’s collection Yahanraku (Midnight Music, 1777) the following is found11: The music and dance of the Gion Festival Do not harmonize with the tones of the autumn wind. The subtle depths [sabi and shiori] of Bashp’s school Should not appear at a celebration of spring.

The Buson who declared Bashp’s sabi and shiori mode to be the ultimate style to be venerated, here declares that there are times when these aesthetic ideals are inappropriate. Buson’s view of Bashp, even in his own words, is thus a complex, sometimes shifting one. A different model seems called for

B ’ B  ¯

247

to characterize the relationship beyond Bashp’s complex positioning as poetic ideal for Buson. In his 1973 work The Anxiety of Influence, Harold Bloom offers a Freudian model for the workings of influence in the relationships between poets and their precursors.12 Though the subject of Bloom’s analysis are the English Romantic poets, the book does provide a useful structure for considering the relationship between Buson and his great precursor Bashp. To summarize very briefly, Bloom characterizes the relationship between a poet and his predecessor in terms of a father–son antagonism: a poet must struggle against and overcome his strong precursor, denying influence and in some instances deliberately misreading the precursor, “so as to clear imaginative space” for himself and his own creativity.13 Needless to say, this is not the paradigm generally used to describe the many master/student, precursor/successor relationships in Japanese art and literature, but it can be useful nonetheless. Buson’s gesture with respect to Bashp may be described as the exact opposite of that outlined by Bloom, but having the same effect. Rather than deny or struggle against Bashp in an agony of poetic anxiety, Buson freely and explicitly embraces his precursor, and in that embrace paradoxically clears the imaginative space needed for his own creativity. He accepts and acknowledges Bashp, pays proper homage, and then goes on to create his own poetry, in effect using Bashp for his own ends. One small example of such a gesture is this very early haiku (1744) by Buson, the first in which he uses the haigp, or haikai name, “Buson”14: Furuniwa ni uguisu nakinu hi mo sugara

In an old garden a nightingale was warbling, all day long.

The echoes of Bashp’s frog poem are unmistakable: nightingale in an old garden, frog in an old pond. But there the similarities end. Where the earlier poem focuses on the single sound of the frog jumping into the water and the reverberating silence that follows, Buson’s poem is filled with the warbling of the nightingale as it flits to and fro all day long in the garden. In the very first haiku that launched him as “Buson,” the poet begins with a clear nod to Bashp, and then takes the poem in a completely different direction with different nuances and mood. An even better example of Buson’s embrace and subsequent use of Bashp for his own ends is the famous essay commemorating the

248

E  F. Y 

rebuilding of Bashp’s hut in eastern Kyoto, Rakutp Bashp-an saikpki (An Account of the Rebuilding of the Bashp Hut in Eastern Kyoto, 1776).15 It is a fairly long and elegantly written work in which Buson describes the site of the hut (on the grounds of the Buddhist temple Konpukuji) and the area around it, gives an account of some of Bashp’s travels, recounts how there came to be a hut there and how it came to be rebuilt by him and several of his circle. Given the subject matter and occasion for writing, there is no question that this is at the very least a fine example of Buson’s willing embrace of his great precursor Bashp: it is, after all, an essay on the rebuilding of the “Bashp Hut.” Indeed, it even contains echoes of Bashp’s famous essay Genjuan no ki (Record of a Phantom Dwelling 1690). But a careful reading would reveal that there is only the most tenuous connection between this hut and the historical Bashp. No one who lives in the area knows why it came to be called “Bashp Hut,” and the best Buson can do is recount the story of the monk Tesshu (fourth head priest of Konpukuji) who admired Bashp’s poems and who, grieving the poet’s death, named his modest room Bashp-an. But because of this connection and the beauty of the site, a group including Buson decided to build a hut at Konpukuji and to have poetic meetings there at least twice a year to honor Bashp. It is as if the “Bashp connection” were a pretext for the group to have an elegant meeting place and for Buson to compose a wonderful essay. In the background, too, may have been the desire to respond to the rebuilding of the “real” Bashp-an, Bashp’s actual dwelling, in Fukagawa, Edo, at around the same time. Regardless of the actual motivation for the rebuilding and the writing, the end result is that the very act of paying homage to Bashp and acknowledging his precursor status opened up the imaginative space for Buson to create his essay and for his circle to acquire an elegant physical space for their meetings. When we turn to some of Buson’s activities as painter–poet, the “use” of his predecessor takes on the quality of outright appropriation. As a young man in Edo, Buson had studied both painting and calligraphy in addition to Chinese poetry and haikai. After the death of his master Hajin (Spa) in 1742, he seems to have devoted himself more to the study of painting, though he always wrote poetry. Even after his return to the Kyoto–Osaka area in 1751, Buson concentrated on becoming established as a painter first, and only returned to poetry in earnest in the late 1760s after his ability to earn a livelihood as a painter had been assured. Though known primarily as a Nanga (Chinese Southern School) painter, Buson also perfected the hybrid

B ’ B  ¯

249

poem/painting genre known as haiga, or haiku painting. Depending on one’s point of view, these can be described as a simple painting and its accompanying haiku, or a haiku accompanied by an illustration. Balance and harmony between the two elements were deemed essential. Buson created a number of haiga, using not only his own poems, but also the literary works of others, including Bashp. Perhaps the most striking examples of these are also ones that expand the earlier mentioned definition of haiga from a simple painting plus one haiku, to multiple paintings depicting selected portions of a long prose work: Buson’s numerous renderings of Bashp’s Oku no hosomichi (Narrow Road through the Deep North). Beginning in 1777 and over a span of three years, Buson completed no fewer than ten illustrations of Bashp’s masterpiece.16 Judging from available reproductions of the four that are extant, this was no small undertaking. One is a large double screen (bypbu) with three panels each, for a total of six panels filled with the entire text of Narrow Road in Buson’s own hand and interspersed with illustrations of nine selected sections.17 Three others in scroll format also depict the entire text, with twelve to fourteen illustrations.18 The standard explanation for this prodigious output of Bashprelated works (Buson also produced a scroll painting of Nozarashi kikp [Skeleton by the Roadside] in 1778 and participated in the “rebuilding” of the Bashp Hut in 1776) is that the haikai revival movement with its “Back to Bashp” motto was beginning to gather steam around this time.19 Thus, these extended haiga were Buson’s homage to the master as well as his contribution to the cause of reestablishing Bashp’s aesthetic ideals in the contemporary era. Whatever Buson’s intentions, however, the result was that, for those with access to a scroll or screen, the “master” they encountered was, literally, Buson’s creation. Though wood block print editions of Oku no hosomichi had been available since the beginning of the century, it would be through Buson’s calligraphy and paintings that poets and connoisseurs would henceforth be reading and, most importantly, visualizing Bashp’s journey. In the act of paying tribute to his predecessor, Buson created some of his most memorable haiga, forever linked his own name to that of Bashp’s greatest work, and even ensured that at least where visual interpretation was concerned, his reading of Bashp would be fixed as the standard. This particular embrace of the predecessor resulted in an almost total appropriation of him. In another haiga, Buson weaves a complex tapestry of allusions that places him in the lineage of not only Bashp, but of those even more

250

E F. Y 

ancient poets of Japan and China who were Bashp’s ideals. The gesture to Bashp is very subtle, almost lost amid the allusions to those who might be considered precursors of the master himself. The work is variously titled Yugyp yanagi jigasan (Pilgrim’s Willow Haiga) or Ishi jigasan (Rock Haiga); it is a hanging scroll depicting six rocks in the bottom half and a three-part text above (see figure 11.1). The text consists of two prefatory notes and a haiku: In the two Prose Poems on the Red Cliff, every word is exquisite, but among them these lines are particularly fine: “The mountains were very

Figure 11.1 Yosa Buson (1716–1784), Pilgrim’s Willow Haiga (Yugyp yanagi jigasan), hanging scroll, ink and light reddish brown and grays on paper, 58.5  36.5 cm, courtesy of Itsup Art Museum, Itami.

B ’ B  ¯

251

high, the moon small / The level of the water had fallen, leaving boulders sticking out.”20 They stand out like a solitary crane among a flock of chickens. Long ago when I was on my pilgrimage through the northern provinces, I recalled these lines when I came to Pilgrim’s Willow Tree and so composed the following: Yanagi chiri shimizu kare ishi tokoro dokoro

The willow is bare, the spring has dried up— rocks here and there.

The painting is thought to have been completed after 1777,21 but there is also an earlier version painted on a fan, with six rocks and the haiku alone, without preface, thought to have been produced in the early 1770s.22 The haiku first appeared in the collection Hogobusuma (Wastepaper Screen) in 1752, with a preface similar to the one in the painting that also identifies it as having been composed in 1743 during a pilgrimage through the northern provinces, making it one of Buson’s earliest published poems. The nod to Bashp is clear in the mention of michinoku (the northern provinces making up today’s Tphoku region), angya (long journey or pilgrimage), and yugyp yanagi (Pilgrim’s Willow Tree). These words immediately evoke Bashp’s earlier journey, the resulting work Oku no hosomichi, and the poem that Bashp composed under this same tree: Ta ichimai uete tachisaru yanagi kana

As I stood in the shade of Saigyp’s Willow—look! An entire field was planted.

Bashp’s poem in turn was a gesture to the twelfth-century poet–priest Saigyp, who composed this waka (thirty-one-syllable verse) at the same site23: Michinobe ni shimizu nagaruru yanagi kage shibashi tote koso tachidomaritsure

In the shade of a willow by the road side, where pure spring water flows, I stopped “for a moment” to rest— and look! How many “moments” have passed.

In the second part of the preface, therefore, Buson has embraced not only Bashp, but through him also Saigyp, an important predecessor for Bashp. In the first part, moreover, Buson pays even greater tribute

252

E  F. Y 

to the Song dynasty poet Su Dong Po (Su Shi, 1037–1101) by quoting two lines from his important work Prose Poems on the Red Cliffs, 1082, and praising them. By invoking these many illustrious predecessors, therefore, Buson appears to be placing himself in that poetic lineage, claiming for himself the heritage not only of Bashp, but also of his great precursor Saigyp, and even beyond Saigyp to the Chinese poet Su Shi. The relationship between Buson and Bashp in this work, however, is more complex. It will be noticed, first, that the poet whom Buson actually quotes in his haiku is not Bashp but Saigyp, in the word shimizu, “pure water.” While Bashp is unmistakably present in the second part of the preface, it is by implication and not directly. The yugyp yanagi “Pilgrim’s Willow” is more directly connected to Saigyp, being the name of a np play in which both Saigyp and his poem are mentioned.24 It can be argued that, with respect to the precursor Saigyp, Buson may even have been creating a space for himself that was equal to that of Bashp. Having traveled to the same willow tree as his predecessor, Buson could now stand with Bashp and acknowledge Saigyp as their common poetic precursor—a proprietary, even presumptuous gesture. A similar gesture is made toward both Bashp and Saigyp in the even more prominent invocation of Su Shi in this haiga. Indeed, there is both visual and textual evidence here that Buson claims the heritage of Su Shi directly, not through Bashp or Saigyp, and even that he at least partially contradicts the heritage of his Japanese predecessors. The second part of the preface, which contains the allusions to Bashp and Saigyp discussed earlier, is visually dwarfed by the first part, which contains the even more direct and prominent references to Su Shi. Not only are the lines about Su Shi’s Prose Poems on the Red Cliffs encountered first, the calligraphy is also much bolder and larger and the entire section longer than the second part. The lines of the second part with the references to Bashp and Saigyp are also presented at a level below that of the lines referring to Su Shi. This second, visually weaker and subordinated part also refers back to the first part, with the words “I recalled these lines [by Su Shi] . . . and so composed the following.” The inspiration for his haiku, Buson seems to be saying, was neither Bashp nor Saigyp, but Su Shi. Indeed, the second line quoted from Prose Poems sets the subject matter of both Buson’s haiku and the painting: “The level of the water had fallen, leaving boulders sticking out.” This image of rocks exposed by the receding waters of a river is exactly what Buson takes

B ’ B  ¯

253

up in his haiku and painting, which is of six rocks of various sizes that take up the bottom half of the scroll. In the haiku, moreover, Buson goes so far as to contradict the image presented in his predecessors’ poems. In Saigyp’s waka, the willow tree was flourishing (it had enough foliage to provide shade) beside a spring flowing with fresh water (shimizu nagaruru). Bashp does nothing to contradict that scene, in fact reinforcing the image of a full spring and green willow leaves by saying that he rested there while an entire field was planted before leaving to resume his journey. Even Su Shi’s Prose Poems has a flowing river, though its water level is so reduced that the rocks at the bottom are exposed. Buson’s haiku, however, specifically denies and contradicts the picture painted by his predecessors. His willow has shed all its leaves and the branches are bare. The spring is not only at a low water level; it has completely dried up. All that is left are the rocks strewn “here and there” (tokoro dokoro) and across the bottom of the painting. The poetic tradition that seemed to flow in an unbroken, luxuriant stream from Su Shi to Saigyp to Bashp to himself has ended here, Buson seems to be saying. This could be taken as another comment on the deterioration of Bashp’s kind of elevated poetic practice during the decades following his death in 1694.25 I submit, however, that taken as a whole, the complex text of this haiga declares that this “end” is not a termination, but a brilliant culmination. In this multilayered and sophisticated work, Buson pays homage to his poetic predecessors, places himself in their lineage, and then contradicts them in a complex gesture that brings his own poetry and painting to the fore. It is an instance in which Buson’s “embrace of influence” approaches the Bloomsian model of poetic patricide—multiplied three times. I would like to close with an idiosyncratic reading of the Bashp–Buson relationship as it unfolds during a typical visit to Konpukuji, site of both the Bashp Hut (renovated since the time of Buson) and of Buson’s grave. The temple, a modest one not generally overrun with tourists, is set in the eastern hills of Kyoto not too far from the Shisendp (Hall of Poetic Immortals). After paying the entrance fee, one enters a small hall in which several rooms have been given over to displays of articles associated not only with Buson, but also with other figures prominent in temple history. One spends a due amount of time inspecting these, then buys—or not—appropriate souvenirs, and finally descends into the small garden in front of the hall. One follows a path that leads up into the hill behind the hall, and comes to the Bashp-an, Buson’s Bashp Hut. It reminds one of a

254

E  F. Y 

chashitsu, or tea ceremony room, so modest are its proportions. One inspects its rooms from the outside, takes pictures if one wishes, and gazes down on what must have been a lovely view of Kyoto long ago. One might even pause to sit on the veranda and imagine eavesdropping on a gathering of Buson and his circle. Finally, one follows the narrow path further up into the hillside, to a forested area containing several graves, including Buson’s and those of some disciples, as well as others. But Buson’s is the most prominent and has pride of place high up the slope—if memory serves, no others have been permitted above his. After paying due respect, one can turn around and look down upon the roof of Bashp-an, the Bashp Hut that Buson helped construct. This tour of Konpukuji has taken us steadily uphill, from the temple hall, to the Bashp-an, to Buson’s grave. While the Bashp Hut is encountered first, the pinnacle of the tour is Buson, represented by his grave. The chronological—and many would say artistic—priority of Bashp has been displaced by the geographically higher elevation of Buson, in a temple where Bashp is an artifact constructed by Buson. An interesting relationship, indeed.

A: Y B Buson was born near Osaka in 1716, twenty-two years after the death of Bashp. Nothing else is known for certain about his early life, but there is some evidence suggesting that he was born into the family of a well-to-do farmer. He went to Edo as a young man and studied haikai, Chinese poetry, painting, and calligraphy. After the death of his haikai master Hayano Hajin in 1742, Buson spent about ten years in the Kantp area (eastern Japan) traveling, studying, and painting. He then returned to western Japan, to Kyoto, in 1751. However, he spent further long stretches of time away from Kyoto, first in Tango (north of Kyoto) for about three years, then on the island of Shikoku for about two. He finally settled in Kyoto in 1768. From the days of his wanderings in eastern Japan, Buson devoted most of his energies to the study and production of Nanga painting. Although he always wrote poetry, he made his reputation first as a painter, and even after he became more active as a poet from the late 1760s, that is how he made his living. He is better known in this country as a Nanga painter, while in Japan, literary scholars often describe his poems as “painterly” and art historians refer to his paintings as “lyrical.” He died in the third year of Tenmei (1783). The period of Buson’s greatest activity as a poet is considered to be the 1770s until his death.

B ’ B  ¯

255

Thus, approximately ninety years separate him from Bashp’s most active years in the 1680s.

N 1. Masaoka Shiki, “Haijin Buson,” in Masaoka Shiki shu, Gendai Nihon bungaku zenshu: 11 (Tokyo: Kaizpsha, 1928), 484–505. Originally published in 1898 as a series of newspaper articles. 2. C. Hugh Holman and William Harmon, A Handbook to Literature, 5th ed. (New York: Macmillan Publishing Company, 1986), 256–257. 3. Approximately nine decades separate the period of Buson’s peak literary activity from that of Bashp’s. See appendix for a brief biographical sketch of Buson. 4. Kitp’s 1787 remarks attached to a manuscript of Momosumomo (Peaches and Plums), composed in 1780. Quoted in Muramatsu Tomotsugu and Kiyoto Noriko, Kinsei bungaku no sekai, Wabun koten: IV (Tokyo: Hpspdaigaku Kypiku Shinkpkai, 1988), 152. 5. Ptani Tokuzp, Okada Rihei, and Shimasue Kiyoshi, eds., Buson shu, Koten haibungaku taikei: 12 (Tokyo: Shueisha, 1972), 313. 6. Ibid., 336. 7. In traditional Japan, it was considered important to settle all accounts by the last day of the year in order to start the New Year with a “clean slate.” This meant that creditors went around at year’s end to collect debts as the cash-strapped tried desperately to elude them. Saikaku’s Seken munezan”yp (Reckonings at Year’s End) is a collection of stories telling how the poor managed to get through this time. 8. The hokku quoted by Buson here (toshi kurenu / kasa kite waraji / hakinagara) was composed by Bashp on his 1684–1685 journey and is found in Nozarashi kikp (Skeleton by the Roadside); see David Landis Barnhill, Bashp ’s Journey: The Literary Prose of Matsuo Bashp (Albany: State University of New York, 2005), 19. 9. Ptani, Okada, and Shimasue, Buson shu, 326–327. For translation of the Yoshino section in Pi no kobumi in which Bashp’s hokku can be found, see Barnhill, Bashp’s Journey, 36. 10. Ibid., 312–313. 11. Ibid., 326. 12. Harold Bloom, The Anxiety of Influence: A Theory of Poetry (London, Oxford, New York: Oxford University Press, 1973). 13. Ibid., 5. 14. Ptani, Okada, and Shimasue, Buson shu, 31. 15. Ibid., 324–326. 16. Okada Rihei, Haiga no bi: Buson, Gekkei (Kyoto: Yutaka Shobp, 1973), 186–188. 17. Yamagata Bijutsu Hakubutsukan; reproduced in Sasaki Shphei, and Kodama Kpta, Bashp, Buson, Zusetsu Nihon no koten, 14 (Tokyo: Shueisha, 1978), 66–67 and in Itsup Fine Arts Museum and Kakimori

256

18. 19. 20. 21. 22.

23. 24.

25.

E F. Y  Collection, eds., Botsugo nihyaku niju nen: Buson (Buson: Two Hundred and Twenty Years after his Death) (Kyoto: Shibunkaku, 2003), 28–30. For full color reproductions of the three Buson Oku no hosomichi scrolls, see Botsugo nihyaku niju nen, 12–27. Okada, 188–189. Burton Watson, trans., Su Tung-p’o: Selections from a Sung Dynasty Poet (New York: Columbia University Press, 1965), 92. Leon Zolbrod, Haiku Painting (New York, Tokyo: Kpdansha, 1982), 14. Private collection. Reproduced as “Rock Haiga” in Calvin L. French, The Poet–Painters: Buson and His Followers (Ann Arbor, MI: University of Michigan Museum of Art, 1974), 75. Shinkokinwakashu, #262. Translated as “The Priest and the Willow” in Donald Keene, ed., Twenty Plays of the Np Theatre (New York: Columbia University Press, 1970), 219–236. A common explanation for why the youthful Buson left Edo after his haikai teacher Hajin’s death in 1742 is that he was, indeed, disgusted by the state of poetic practice in the city at the time. This view is reinforced by the purported date of composition of this haiku, 1743, during Buson’s ten-year sojourn in Kantp studying painting and poetry after leaving Edo the year before.

C

Stephen Addiss had a first career as a composer and musician, traveling extensively in Asia and Africa with Bill Crofut in the State Department’s cultural exchange program, and becoming especially interested in Asian music and art. He went to the University of Michigan in his mid-thirties for an MA and PhD in art history and musicology, and then taught for fifteen years at the University of Kansas. Since 1992 he has been professor of art and TuckerBoatwright Professor in the Humanities at The University of Richmond. His own paintings, calligraphy, and ceramics have been exhibited in Japan, China, Taiwan, Korea, Germany, England, and the United States, and his publications include Nanga Paintings, 1976; Tall Mountains and Flowing Waters: the Arts of Uragami Gyokudp, 1987; The Art of Zen: Paintings and Calligraphy by Japanese Monks, 1600–1925, 1989; A Haiku Menagerie, 1992; Haiga: Takebe Spchp and the Haikai-Painting Tradition, 1995; A Haiku Garden, 1996; How to Look at Japanese Art, 1996, and The Art of Chinese Calligraphy, 2005. David Landis Barnhill is director of environmental studies and professor of English at the University of Wisconsin Oshkosh. In addition to several earlier articles analyzing Bashp, he recently published a two-volume translation: Bashp’s Haiku: Selected Poems of Matsuo Bashp and Bashp’s Journey: The Literary Prose of Matsuo Bashp, both from State University of New York Press. He also does research on American nature writing, including the influence of East Asian culture on contemporary nature writers. He teaches a course on “Japanese Nature Writing.” Cheryl Crowley is assistant professor of Japanese language and literature in the Department of Russian and East Asian Languages and Cultures, Emory University. She teaches a variety of courses on premodern and modern Japanese literature and visual culture in Japan. Her research interests include early modern Japanese literature and art, classical Chinese poetry, and women’s studies. She has published articles on Yosa Buson (1716–1783), Uejima Onitsura (1661–1738), and women haikai poets; she is working on a book on Yosa Buson and the Bashp Revival movement.

258

C 

I. Leopold Hanami is assistant professor of Japanese literature and language in the Department of East Asian Languages and Literatures, George Washington University, Washington, DC. He received his BA and MA from UCLA and Ph.D from Stanford University. He teaches modern and classical Japanese language as well as a variety of courses in classical and modern/contemporary Japanese Literature, and Japanese film and culture. His research interests focus on the influence of texts and contexts on reading, particularly as they pertain to late Heian and early medieval Japanese court poetry and to orthodox linked verse. Eleanor Kerkham, associate professor of Japanese literature at the University of Maryland, Department of Asian and East European Languages, Literatures, and Cultures, is a specialist in premodern and early modern Japanese literature and has taught at Colby College and Tenri University in Japan. She has served as chair of the Committee (now Center) on East Asian Studies and teaches premodern and modern Japanese language and literature. She has published in English and Japanese on the prose, comic linked verse, and literary career of the haikai poet Matsuo Bashp, on Japanese women writers, and on the theory of the body and Korean “comfort women” stories of the modern novelist Tamura Taijirp. Her current research includes a study of Matsuo Bashp’s early apprenticeships, focusing on his comic linked verse and interaction with the Tokugawa literary world as haikai master, and a translation and critical study of Tamura Taijirp’s World War II literature, focusing on his examination of the intersections between war, the will to survive, sexuality, and violence. She is coeditor of and contributor to a collection of essays on war, occupation, and creativity in Japan and East Asia, War, Occupation, and Creativity, Japan and East Asia 1920–1960 (2001). William R. LaFleur is the E. Dale Saunders Professor in Japanese Studies in the Department of East Asian Languages and Civilizations at the University of Pennsylvania. His PhD is from the University of Chicago. He has taught at Princeton, UCLA, and Sophia University in Tokyo. In 1989 he was the first non-Japanese recipient of the Watsuji Tetsurp Culture Prize for scholarship. His books include The Karma of Words: Buddhism and the Literature Arts in Medieval Japan (University of California Press, 1986); Buddhism: A Cultural Perspective (Prentice-Hall, 1988); and Liquid Life: Abortion and Buddhism in Japan (Princeton University Press, 1992), published in Japanese as Mizuko: “chuzetsu” o meguru Nihon bunka no teiryu (Aoki Shoten, 2006). He edited Zen and Western Thought: Essays by Masao Abe

C 

259

(Macmillans, 1985), which received a prize from the American Academy of Religion; edited Dpgen Studies (University of Hawaii Press, 1985); and is the principal editor of Dark Medicine: Rationalizing Unethical Medical Research (Indian University Press, 2006). He is a senior fellow at Penn’s Center for Bioethics and is currently completing work on a volume that studies Japanese critics of American biotechnology and bioethics. Horikiri Minoru is professor emeritus, Japanese literature, College of Education, Waseda University, Tokyo, Japan. He lectures extensively in Japan, has participated in a number of international symposia, and is a leader within the Haibun Gakkai Association. Dr. Horikiri has edited a number of Bashp, haiku, and hairon dictionaries and study guides, and has written on a wide variety of haikai topics, including studies on the haikai poets Kagami Shikp, Buson, and Issa and monographs on Bashp’s haibun, on Oku no hosomichi, the Bashp School hairon, and hokku expression. His books include Shpfu hairon no kenkyu: Shikp o chushin ni, 1982; Hypgen to shite no haikai: Bashp, Buson, Issa, 1988; Bashp no monjin, 1991; Oku no hosomichi: eien no bungaku-kukan, 1997; Bashp no Saundosukeepu: haikai hypgenshi e mukete, 1998; and Bashp to haikaishi no tenkai, 2004. Hori Nobuo is professor of Japanese literature, School of Education, Kobe University, Kobe, Japan. He too is an active leader within the Haibun Gakkai and has specialized in and written on Bashp’s haibun, renku, and haiga, and on his relationships with disciples and contemporaries, and the philosophical/literary influences on his thought and life style. He has coedited and provided commentary on several collections of Bashp’s complete works, including the Bashp volumes in the Koten Haibungaku Taikei series (volume 5) and the Nihon Koten Bungaku Zenshu series (volume 41). Dr. Hori’s contributions to the somewhat less well-covered area of haiga research include co-editor of and contributor to volume 3 (Shpmon shoga, Prominent Painters of the Bashp School) of the twelve volume Haijin no shoga bijutsu (The Art of Haikai Painters/Calligraphers) series and contributor to the authoritative Bashp zenzufu (reproduction and discussion of Bashp’s complete calligraphic works and paintings). Joan O’Mara received her BA from Carleton College and her MA and PhD from the University of Michigan in Ann Arbor. She is currently associate professor of art history and director of the East Asian Studies Program at Washington and Lee University. She has also recently completed a term as chair of the Executive Board of ASIA Network. Her dissertation on the haiga of Yosa Buson (1716–1784)

260

C 

has been followed by additional scholarly research and publications on various aspects of the relationship between word and images in the arts of Japan. Peipei Qiu is associate professor of Japanese and chair of the Department of Chinese and Japanese at Vassar College. She is the author of Bashp and the Dao: The Zhuangzi and the Transformation of Haikai (University of Hawai’i Press, 2005) and her studies of Japanese poetry and modern Japanese novelists have been published in English, Japanese, and Chinese. Her current teaching and research interests include comparative study of Japanese and Chinese literature, women in Japanese and Chinese literature, and Japanese language pedagogy. Haruo Shirane, who is Shincho Professor of Japanese literature and culture at Columbia University, has done research on premodern and early modern fiction, poetry, and literary criticism, and is interested in cultural theory. His books include The Bridge of Dreams: A Poetics of The Tale of Genji (1987), Traces of Dreams: Landscape, Cultural Memory, and the Poetry of Matsuo Bashp (1997), Inventing the Classics, Canon Formation, National Identity, and Japanese Literature (2000), Early Modern Japanese Literature, An Anthology, 1600–1900 (2002), and Classical Japanese, A Grammar (2005). Eri F. Yasuhara received her PhD from UCLA in 1982, from the Department of Oriental Languages (now Department of Asian Languages and Cultures). Her dissertation was Buson and “Haishi”: A Study of Free-Form “Haikai” Poetry in Eighteenth Century Japan. She was professor of Japanese at California State University, Los Angeles, teaching Japanese language, literature, and culture; in 1995, she became associate dean of the School of Arts and Letters at Cal State Los Angeles, and in 2000 she moved to California State University, San Bernardino, as dean of the College of Arts and Letters, where she remains today. In addition to her interests in early modern haikai poetry, Dr. Yasuhara has published in the field of Japanese American literature, focusing on the work of first generation (issei) authors writing in Japanese in the early twentieth century.

I

“Account of Eighteen View Tower” (Juhachirp no ki, 1688, Bashp), 39 “Account of Pure Washed Temple” (Sharakudp no ki, 1690, Bashp), 49 “Account of the Rebuilding of the Bashp Hut in Eastern Kyoto” (Rakutp Bashp-an saikpki, 1776, Buson), 248 “Account of the Unreal Dwelling” (Genjuanki, 1690, Bashp), 39, 52, 248 aisatsu (poetic greeting) 117, 123, 159, 177 Alter, Robert, 90 Amino Yoshihiko, 82 anthropomorphism, 44–45 Anxiety of Influence (Bloom), 247 atarashimi (newness), 119 Ayatari see Takebe Ayatari Azuma nikki (Diary of the East, haikai collection, 1681, Gonsui) 26, 165 Backpack Notes, see Oi no kobumi Bag of Hotei (Hakuin Ekaku), 233 Bagworm (Bashp), 207, 209 Baitei, Ki, 227–28, 230, 242 Bakhtin, Mikhail, 12, 107, 122, 124 Banana Tree by the Gate to the Bashp-an (Bashp), 210, 212 Bashp, Matsuo Buson and, 243–44, 247–55 celebration of, 1–8 Chinese literati tradition and, 36–40, 188–95 creativity and, 47–50 on death and dying, 92–99 criticism of, 79–81 literary style, 141–42, 162–65, 176, 181–84

portrayal of landscape in travel journals, 67–71 scholarship on, 8–17, 159–60 on social conditions, 85–90 shpypyu (carefree wandering) and, 61–67 travels to northern Japan, 81–85 and use of legends in his works, 184–86 zpka (the creative) and, 27, 43–50, 71–74 and Zhuangzi, 27–30, 61–77 Bashp-p tsukeaishu (A Collection of Master Bashp’s Linked Verse, Buson, 1774), 244 bashp plant (banana plant, Japanese plantain), 47, 49, 58 Bashp Revival movement, 244, 249 Bay, Alexander, 82 Beardsley, Monroe, 133, 138 Beneath the Cherry Trees see Ko no moto Benjamin, Walter, 12, 118–19 Bloom, Harold, 16, 117, 247, 253 Bo Juyi, 36, 42, 59, 171 Bonchp, Nozawa, 128, 141–44, 146–49 Brower, Robert, 128 bunka gypsei (cultural politics), 4 Buson, Yosa, 4, 15–16, 109, 168 artistic ideal and, 244–47 Bashp and, 243–44, 247–55 haiga and, 214, 224–25, 227, 228–31 haibun of, 244, 245, 246, 248, 250–53 Butchp Oshp, 178, 185, 187 calligraphy, 36, 53, 54, 206, 252, 254 Ayatari and, 231

262

I

calligraphy––continued Baitei and, 227–28 Bashp and, 218 Buson and, 248–49 Chiyo and, 224 haikai and, 217–19, 226–28 Issa and, 239 Kakujp and, 221 Kikaku and, 218–19 Ryuho and, 217 Charcoal Sack see Sumidawara Chigetsu, 225, 228–30 Chinese literati tradition, Bashp and, 36–40 Buson and, 224 Chiyo, Kaga, 222–24 chpnin (townspeople), 105–10, 112–13 Chpshpshi, Kinoshita, 153, 157 Confucianism, 8–9, 100, 196 Bashp and, 34, 37–38, 50, 55, 57 Daoism and, 45, 58 Kikaku and, 218 zaohua (zpka) and, 43, 56 Cormin, Cid, 18, 81 cultural colonization, 84–85 creativity artistic, 10, 33, 36, 42, 44, 46–47, 68 Bashp and, 50 nature’s, 47–50 Crow on a Withered Branch (Bashp), 205 “Cypress Hat” (Hinokigasa no ji, 1782, Buson), 245 Dainichi Nyprai, 112 Danrin school, 9, 24, 63, 109–10, 112, 114, 121, 141, 167 Daoism, 7–10, 25, 34, 37–39, 43, 45–47, 50, 55, 57 Bashp’s kikpbun and, 63–67 Confucianism and, 45, 58 zpka and, 67–74 Dawn of the Thirteenth Night (Spbaku), 239, 240 Dazai Osamu, 142–49 de Man, Paul, 138 death, 92–99 defamiliarization, theory of, 114 Derrida, Jacques, 115, 139, 155

Discussion Under a Mosquito Net (Spchp), 234 disposition, 43–44, 92–93 Dog Tsukuba Collection see Inu Tsukuba shu dokugin (solo linked verse), 123 Du Fu, 36, 38, 53, 54, 87, 111, 115, 191 Eight Views, motif of, 39–40 Eisenhower, Dwight, 80 Empty Chestnuts (Minashiguri, Bashp School, 1683), 36, 244 En no Gypja, 178, 185, 193 Erfahrungen, 118 Erlibnisse, 118 Etsujin, 227–28, 242 Freud, Sigmund, 121, 247 fueki ryukp (the unchanging and the ever-changing), 116, 119, 162, 179 fuga (poetic elegance), 49, 193 fukyp (madness in poetry), 116 furyu, 91–92, 112 Fujiwara Kiyosuke, 24, 25–26, 152 ga (classical refinement and elegance), 110, 116 gago (classical diction), 106–07 Gaikotsu (Skeletons, Ikkyu), 95 Garden Butterfly (Issa), 238 Gassan (the novel, Mori Atsushi, 1973), 19, 174, 184, 195, Gassan (Moon Mountain), 192–93, 195, 198 Gates, Henry Louis, 12, 112 Genji monogatari See Tale of Genji Genjuan no ki See “Account of the Unreal Dwelling” Genroku period, 13, 116, 159 Girardot, Norman, 41, 55 Goshun, Matsumura, 15, 226, 241 Gotoba, ex-emperor, 151–52 Gourd See Hisago Gourd-beating (Bashp), 207–8 gugen (yuyan, allegorical, metaphoric words), of Zhuangzi, 25, 27, 65, 73, 75, 114

I Great Clod, image of, 41, 47–48, 55 Guangwen, 64 Hachi-tataki See Gourd-beating Hakuin Ekaku, 232–33 haibun, 1, 10, 14, 17, 176 Bashp and, 23, 27, 35, 37, 38, 46–48, 51, 58, 59, 201 Buson and, 244–45 Chinese religious tradition and, 37–40 explanation, 52 Oku no hosomichi as haibun collection, 185–87 shpypyu (carefree wandering) and, 61–62, 66 zpka (the creative) and, 35 haibunga (poetic prose paintings), 225 haidan (haikai community), 159 haiga, 23, 167, 224–41 Ayatari and, 231–32 Baitei and, 227–30 Bashp and, 201–3, 206–8, 210, 213–14, 215 Buson and, 224–26, 249–53 Chiyo and, 222–24 explanation of, 201–2 Hakuin and, 232–33 Issa and, 238–39 Otsuyu and, 220–22 portraits and, 225–27 Ryuho and, 217–18 Shirp and, 230–31, 236–38 Spbaku and, 239–40 Spchp and, 234–36 haigon (haikai words), 67, 116 haikai debates on, 26–27 explanation of, 23–24 Zhuangzi and, 24–26, 61 Haikai sanjurokkasen (Thirty-six Haikai Masters, 1799), 225, 228 Hajin See Spa Han shu (Annals of the Han, first century CE), 23 Hanami, Leopold, 12, 176 Harunobu, Suzuki, 110 Hass, Robert, 1, 17 Hayashi, Eiichi, 33, 52

263

Heian period, 83, 91, 106, 108, 113, 121, 141, 151–52, 215 Heike monogatari See Tale of Heike Hibbett, Howard, 128 “Hinokigasa no ji” See “Cypress Hat” Hirota Jirp, 87, 183 Hirsch, E.D., 132–33, 137 Hisago (Gourd, Bashp School, 1690), 120, 162 Hishikawa Moronobu, 110 hon’i (poetic essence), 70, 113, 117–19, 162 honkadori (borrowing from a source, allusive variation) 109–10, 160 Horikiri Minoru, 8, 13 Horton, H.M., 134 humor, 9, 13, 17, 23, 40, 64, 108–9, 114, 116–17, 120–22, 141, 159–60, 232, 234, 236 Hutcheon, Linda, 12, 117 Hyakusen, Sakaki, 224, 231 Ima wa mukashi (The Past in the Present, haikai collection, Buson, 1774), 246 Ichinaka wa (In the City, Bashp School kasen, 1690), 12, 140–50 Ideologiekritik, 81, 85, 89, 92 Ikkyu, 95 Imoto Npichi, 50, 67 intertextuality, 12, 34, 50, 51, 57, 64, 75, 112, 115–17, 120–22, 132, 142, 183, 186, 230 Inu tsukuba shu (Dog Tsukuba Collection, Spkan, 1532), 108, 116 Ishi jigasan (Buson), 250 Issa, Kobayashi, 238–39, 243 Ise Monogatari See Tales of Ise Itchp, Hanabusa, 203, 207–10 jisei (one’s “last” verse), 95 jo-ha-kyu (introduction-developmentswift ending) pattern, 177, 179 Journal of Bleached Bones in a Field see Nozarashi kikp, Journey to Kashima (Kashima kikp) see Kashima no ki Jun, Ishikawa, 112

264

I

Kagami Shikp, 163 kakawari (literary intersections), 183 Kakujp, 221–22, 224 kanjaku (profound solitude and silence), 72–74 Kanp Yasunobu, 203 Kanp Style, 203, 206–08, 213–14, 218, 225, 241 Kant, Immanuel, 132 Kara eda ni See Crow on a Withered Branch karumi (lightness), 125, 141, 156, 162, 210, 218 kasen (thirty-six verse linked verse), 120, 142–50 Kashima no ki (Record of a Journey to Kashima, Bashp, 1687), 61 Kawamoto Kpji, 160 Kawazu awase (Frog Contests, Bashp School, 1686, 124, 164) keiki (scenery, environment) style, 114, 161–62 keisen jpgo (environment first, then feeling or keiki style), 161–62 see also shisen jpgo kigo (season word), 62, 114, 118 Kerkham, Eleanor, 33, 52, 166 ki-shp-ten-ketsu (opening-receivingtransforming-tying together) pattern, 176–77, 180 Kikaku, Enomoto, 113, 123–24, 218–19 Kita Pu (northern provinces), 82 Knapsack Notebook (Oi no kobumi, Bashp, ca. 1690), 9, 27, 33, 35, 36, 43–45, 62, 65–69, 76, 97, 160, 227 Ko no moto (Beneath the Cherry Trees, Bashp School kasen, 1690), 120 Kobayashi Tadashi, 110 kojin (poets of old), 111, 190 Kokin wakashujo, 111 Kokinshu, 24, 51, 107, 111, 113, 117, 125 kokkei (comic, playfulness), 23, 159–60 kokkei no hito (comic person), 23–24 Konishi, Jin’ichi, 73, 80–81, 130, 133, 135, 154 Kpshoku ichidai otoko See Life of a Man of Love Kurozuka (np play, Black Mound), 185

Kuspshi emaki, 94 Kuwabara Takeo, 86 Kyohaku, 187 Kyorai, Mukai, 8, 23, 128, 141–42, 145, 147–50, 156, 162, 207 Kyoraishp (Kyorai’s Gleanings, Kyorai, 1702–1704), 23, 162 Kyoriku, Morikawa, 15, 37, 202–8, 210, 213, 215, 218 Kyoroku. See Kyoriku, Morikawa LaFleur, William, 8, 10–11, 167 Late Spring (Chiyo), 222–24 Li Bo, 36, 53, 87, 111, 188–94, 197 Life of a Man of Love (Kpshoku ichidai otoko, 1682, Saikaku), 112 literary tradition, Japanese, 34, 50 Liu Xie, 39 Lu Ji, 42, 46, 58 maeku-zuke (single-verse capping) 109 Mair, Victor, 87, 101 Makura no spshi See Pillow Book Man’ypshu, 113 Marxism, 11, 81 Matsuda Yoshiyuki, 7, 19 Matsushima, 2, 35, 37, 40, 44, 46, 48, 69, 179, 187, 192 Matsuyama Seikai, 47 Mei Yaochen, 42 Melon Skin (Kikaku), 219 Mencius, 88 Michikaze, Pyodo (1639–1707), 189, 197 Midnight Music see Yahanraku miira (mummy), 83 Minase sangin (Three Poets at Minase, 1488, Spgi, Spchp, Shphaku), 12, 127–28, 129–37, 139–40, 151–53 traditional reading of, 129–32 Minashiguri See Empty Chestnuts Miner, Earl, 128, 305 Minomushi no See Bagworm mitate (“seeing by comparison,” visual allusion), 109–10

I mitate-e, 110–11 Moon (Ayatari), 232 Moon (Shirp), 230 Mori Atsushi, 14, 173–94, 195 literary theory, 175 on Oku no hosomichi, 175–80 see also Ware mo mata, Oku no hosomichi Moritake, Arakida, 214 Mostow, Joshua, 232, 242 Mukai no oka (Yonder Hill, haikai collection, 1680), 165 Mukashi o ima (The Past in the Present, haikai collection, Buson, 1774), 246 Nagasaki School, 224 Nakagawa Otsuyu (Bakurin), 220–22 Nakamura Shunjp, 146, 256 Nanga school, 248–49, 254 Knapsack Notebook see Oi no kobumi Narrow Road Through the Deep North See Oku no hosomichi New Criticism, 81, 195 NHK (Japan Broadcasting Corporation), 173, 174, 180 Nijp Yoshimoto, 130, 140 Noh drama, 106, 111–12 Npin, 111, 112, 115, 117, 191 Nozarashi kikp (translated as Skeleton in the Fields, A Weather-beaten Journey, Journal of Bleached Bones in a Field, Bashp, 1684), 39, 61, 63–64, 72, 75, 85, 94, 96, 98–99, 100, 249 Pemaru, 163 Oi no kobumi (translated as Knapsack Notebook, Rucksack Notebook, BackpackNotes, Essay in My Pannier, Notes from a Knapsack, Bashp, ca. 1690), 27, 33, 35, 45, 51, 58, 62, 65–7, 68–9, 76, 97, 101, 160, 227, 245, 255 Ogawa Haritsu, 203 Ogata Tsutomu on Bashp’s style, 64, 75, 76, 115, 123, 141, 179, 197 on celebrations of Oku no hosomichi 3, 17, 18, 19

265

on creativity, 57 on role of the np drama waki/narrator in Oku no hosomichi 184 on opening passage of Oku no hosomichi, 177 on zaohua (zpka) 43, 53, 55, 56 Okazaki Yoshie, 43, 54, 57 Oku no hosomichi (translated as The Narrow Road to the Deep North, Back Roads to Far Towns, Narow Road to the Interior, Bashp, 1694), 11, 14, 37, 40, 44, 62, 83, 112, 166, 171, 215, 225 Mori Atsushi on, 173–94, 195, 196 celebration of, 2–8, 17, 18 cosmogony in, 56 expressive elements in, 160–62 influence on Buson, 225, 249, 251 influence on other poets, 80–81 karumi and, 141 opening passage, 188–95 place and, 119–20 social conditions in, 89–90 soundscapes in, 164 symposia on, 3, 7, 19 translation of, 52 zpka in, 35, 69, 72–73 see also Ware mo mata Oku no hosomichi “On Mount Fuji” (“Shihp no san,” Bashp, 1684), 37, 39, 46, 69–71 Ptani Tokuzp, 159, 167 Otsuyu see Nakagawa Otsuyu (Bakurin) 220–22 Owen, Stephen, 39, 42, 44, 46, 58 Pyamazumi, 45, 56 parody, 12, 109, 111, 116–17, 120, 144, 160, 226, 234 Past in the Present see Mukashi o ima Pilgrim’s Willow (Yugyp yanagi jigasan also Ishi jigasan, Buson), 250–53 Pillow Book (Makura no spshi) and Bashp, 163, 165, 168 Portrait of Bashp (Kyoriku), 204 Portrait of Chigetsu (Buson), 229–30 poverty, poets and, 90–92

266

I

Problems of Dostoyevsky’s Poetics (Bakhtin), 124–25 Prose Poems on the Red Cliffs (Su Shi), 57, 250, 252–53 qi (Japanese ki, energy) 38, 43, 55 Qiu, Peipei, 56–57, 58, 74 rajio haikuingu, 3 Rangai, Tsuji, 234, 236 Ranran, Matsukawa, 37 “Record of a Phantom Dwelling” see “Account of an Unreal Dwelling” renga, 1, 8, 12, 15, 25, 33, 36, 48, 62, 68, 74, 111, 176–77, 236 intention in, 132–38 language and, 107–8 meaning in, 138–40 structure of, 127–29 traditional reading of, 129–32 freeing the links, 151–54 waka and, 140–42, 145–50 renku, 1, 17, 23, 154, 159, 168, 179, 194, 197 Rexroth, Kenneth, 80 Richter, Jean Paul, 121–22 Riffaterre, Michael, 12, 114–15 Rimpa school, 226, 241 Ryptai see Takebe Ayatari Ryuho, Nonoguchi (Hinaya), 214, 217–18, 224 “Rustic Gate” (Saimon no ji, Bashp, 1693), 202–03, 206 Saga nikki (Saga Diary, Bashp, 1691), 127, 153, 161 Saigyp, 33, 49, 62, 161, 167, 203–4 Bashp and, 66, 111–12, 115, 117, 174, 251–53 influence on Bashp, 16, 67–68, 97 moon imagery, 144 in Oku no hosomichi, 177–79, 186, 191, 195, 198 poetic tastes and, 36 travel writings, 94–95 Saikaku, Ihara, 110–12, 123, 189 “Saimatsu no ben” See “Year’s End” Saimon no ji see “Rustic Gate”

Saitp Yoshimitsu, 130 Sampu, Sugiyama, 26 Sano Kiyohiko, 163, 168 Sarashina kikp (Journey to Sarashina, Bashp, 1688), 62 Sarumino (Monkey’s Straw Raincoat, haikai collection, 1691, Bashp), 12, 125, 128, 141, 143, 162 Saussure, Ferdinand de, 139, 155 Schafer, R. Murray, 13, 163, 168 Schwartz, Benjamin, 88 Sei Shpnagon, 91, 152, 163 Self-Portrait (Shirp), 237 Senna, 221 Seven Samurai, 147 shigure, 26, 111, 115 Shiki (Records of the Grand Historian, first century BCE), 23 Shiki, Masaoka, 109, 243 Shinkokinshu, 49, 151–52, 156 Shintp, 10, 45, 51, 56, 57, 196, 232 Shirane, Haruo, 11–12, 34, 56 Shirp, Inoue, 230–31, 236–39 shisen jpgo (form first, then feeling or the keiki style), 161–62 Shpfu (Bashp Style), 162 Shphaku, 12, 129, 131, 134–36, 140 shpypyu (C. xiaoyaoyou, carefree wandering), and the haikai traveler, 61–67 Shuichi, Katp, 130 “Sleeping Alone in a Grass Hut” (Hitorine no kusa no to, Bashp, 1681), 38 Snyder, Gary, 80 Spa, 246, 248 Spbaku, Fujimori, 239–40 Spchp, Takebe, 12, 129, 134–35, 140, 234–36 Sodp, Yamaguchi, 37, 153 Spgi, 66, 127, 131, 134–35, 137, 140, 151–53 Spin, Nishiyama, 24–25, 110, 160, 167 Spkan, Yamazaki, 108, 214, 234, 236 Solitary Traveler in the Rain (Bashp), 206, 210 Song Di, 40 Sora, Kawai Bashp and, 2–5, 8, 17, 174, 184, 187–88

I Mori’s use of diary, 178, 181–82, 191, 196, 197–98 Tokugawa shogunate and, 84 Sora zuikp nikki (Travel with the Master—Sora’s ournal), 72 soundscape poems, of Bashp, 163–166, 168–171 Su Che, 42 Su Shi (Su Dong Po), 37, 38, 41–42, 53, 54, 57, 58, 252–53 Sukeroku, 112 Sumidawara (Charcoal Sack, Bashp School, 1694), 125, 162 Taiheiki (Chronical of Medieval Japan), 105 taip (intersections), Mori’s theory of, 175–81, 184, 186, 188, 194 Takebe Ayatari, 224, 231–32 Tale of Genji, 107, 111–12, 117, 150 Tales of Ise, 107, 111 Tale of Heike, 105, 170, 171, 175, 183–84 Tanaka Yoshinobu, 87 Tang Hou, 42 Tani Saimaro, 26 Teimon School, 63, 108, 167 Teitoku, Matsunaga, 106, 108–9, 110, 123, 160, 167 Teruoka Yasutaka, 146, 256 tenmei (C. tianming, “heaven’s mandate”), 88–89 Tokiwaya no kuawase (Tokiwaya Verse Contest, Sampu, 1680), 26 Tokugawa period, 3, 8, 17, 45, 57, 106–7, 111–12, 116, 188, 243 Tokugawa shogunate, 11, 84, 88 Tosa Mitsuoki, 28 Tosa nikki (Tosa Diary, Ki no Tsurayuki, ca. 935), 62 travel poems, Bashp, 62–63 tsukeku (attached verse), 108, 120 Tsunoda Tadanobu, 163, 168 Tuning of the World (Schafer), 163 Ueda, Makoto, 33, 52, 130–31, 142, 166 Ukiyo-e, 110, 163 Ulysses (Joyce), 117 utaibon, 111 utamakura, 62, 117, 119 Validity in Interpretation (Hirsch), 132

267

wabi, 38, 115 waka, 8, 14, 94, 117–18, 145, 151, 183, 184–86 explanation of, 107–8, 129–31 haikai and, 23–26 parody and, 110 renga and, 140–41 senryu and, 109 Mount Fuji and, 70–71 Ware mo mata Oku no hosomichi (Mori), 173–98 examination of Oku no hosomichi’s opening passage, 188–95 examination of structure in, 176–80 intersections in, 186–88 Mori’s approach to translation in, 180–83 Mori’s theory of literature in, 175–76 retelling of stories and folk legends in, 183–86 Watson, Burton, 31, 40–41, 54, 75 Welch, Lew, 79 Wen fu, 42–43, 46 Wen Tong, 42 Whalen, Philip, 80 Wimsatt, W.K., 133, 138 Winter Days, (Fuyu no hi, haikai collection, 1684, Bashp School), 125, 244 Yahanraku (Midnight Music, haikai collection, 1777, Buson), 246 yamabushi (mountain priest), 6, 178, 193 Yamabuki (Yellow Rose, Bashp), 210, 211 Yamagata Prefecture, 3–4, 7, 18, 173, 195 Yamamoto Kenkichi, 86, 123–24 Yamato, 82–83 Yayoshi Kan’ichi, 67 “Year’s End” (Saimatsu no ben, Buson, 1776), 245 yobikake (address), 160 yohaku (empty space), 15, 28–29, 222, 224 Yonder Hill see Mukai no oka Yoshitsune, 95, 174, 178–79, 184 Yuasa, Nobuyuki, 33, 35, 52, 88 Yugyp yanagi jigasan see “Pilgrim’s Willow”

268

I

zaohua (the creative), 10, 35, 44, 55, 56, 67, 69 Chinese use of, 40–43 See also zpka Zen Buddhism, 1, 34, 50, 64, 170 Zhang Dunjian, 42 Zhuangzi, 9–11, 47, 53, 55, 56, 59, 75, 87 Bashp and, 27–30, 61–63 Daoism and, 37–38, 47–49 haikai and, 24–26 influence of, 28–30, 96–97 landscape and, 67–71 Nozarashi kikp and, 63–66 Oi no kobumi and, 66–67 zaohua and, 40–43 zpka and, 72–74

Zpga, 112 zpka (the creative) Bashp’s use of, 27, 43–47, in Bashp’s writings, 35–36 being one with, 71–74 landscape reenvisioned, 67–71 zoku (popular, contemporary), 108, 110, 121 zokugo (popular language), 106–07 Zoku minashiguri (Sequel to Empty Chestnuts, Bashp School, 1687), 165 Zoku yama no i (Sequel to a Mountain Well, Kitamura Koshun, 1667), 105 zokutan heiwa (common place stories in ordinary language), 162