Ca' Foscari Japanese Studies 19 Arts and Literature 6

# **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

**19**

**IMAGES FROM THE PAST**

—

NEGRI, TOMMASI

Ca'Foscari

Images from the Past: Intertextuality in Japanese Premodern Literature

# Ca' Foscari Japanese Studies Arts and Literature

Collana diretta da | A series directed by Luisa Bienati, Paolo Calvetti, Toshio Miyake, Massimo Raveri 19 | 6

# Ca' Foscari Japanese Studies Arts and Literature

#### **Editors-in-Chief**

History and Society sub-series: Toshio Miyake (Università Ca' Foscari Venezia, Italia) Arts and Literature sub-series: Luisa Bienati (Università Ca' Foscari Venezia, Italia) Linguistics and Language Education sub-series: Paolo Calvetti (Università Ca' Foscari Venezia, Italia Religion and Thought sub-series: Massimo Raveri (Università Ca' Foscari Venezia, Italia)

#### **Advisory boards**

History and Society sub-series: Nicola Bassoni (Università degli Studi di Genova, Italia) Sonia Favi (The University of Manchester, UK) Kimio Ito¯ (Kyoto University, Japan) Toshio Miyake (Università Ca' Foscari Venezia, Italia) Andrea Revelant (Università Ca' Foscari Venezia, Italia) Steffi Richter (Universität Leipzig, Deutschland) Sven Saaler (Sophia University, Tokyo, Japan) Hirofumi Utsumi (Otemon Gakuin University, Japan) Marco Zappa (Università Ca' Foscari Venezia, Italia) Arts and Literature sub-series: Tomoko Aoyama (The University of Queensland, Brisbane, Australia) Jaqueline Berndt (Kyoto Seika University, Japan) Luisa Bienati (Università Ca' Foscari Venezia, Italia) Katja Centonze (Università Ca' Foscari Venezia, Italia) Caterina Mazza (Università Ca' Foscari Venezia, Italia) Daniela Moro (Università Ca' Foscari Venezia, Italia) Carolina Negri (Università Ca' Foscari Venezia, Italia) Andreas Regelsberger (Universität Trier, Deutschland) Bonaventura Ruperti (Università Ca' Foscari Venezia, Italia) Silvia Vesco (Università Ca' Foscari Venezia, Italia) Pierantonio Zanotti (Università Ca' Foscari Venezia, Italia)

Linguistics and Language Education sub-series: Patrick Heinrich (Università Ca' Foscari Venezia, Italia) Hideo Hosokawa (Waseda University, Tokyo, Japan) Kikuo Maekawa (National Institute for Japanese Language and Linguistics, Tokyo, Japan) Marcella Mariotti (Università Ca' Foscari Venezia, Italia) Giuseppe Pappalardo (Università Ca' Foscari Venezia, Italia) Barbara Pizziconi (SOAS, University of London, UK) Aldo Tollini (Università Ca' Foscari Venezia, Italia)

Religion and Thought sub-series: Giovanni Bulian (Università Ca' Foscari Venezia, Italia) Andrea De Antoni (Ritsumeikan University, Kyoto, Japan) Federico Marcon (Princeton University, USA) Tatsuma Padoan (University College Cork, Ireland) Silvia Rivadossi (Università Ca' Foscari Venezia, Italia)

#### **Executive coordinator**

Marcella Mariotti (Università Ca' Foscari Venezia, Italia)

#### **Head office**

Dipartimento di Studi sull'Asia e sull'Africa Mediterranea Università Ca' Foscari Venezia Palazzo Vendramin dei Carmini Dorsoduro 3462 30123 Venezia

e-ISSN 2610-9425 ISSN 2610-8992

URL https://edizionicafoscari.unive.it/en/edizioni4/collane/ca-foscari-japanese-studies/

# Images from the Past: Intertextuality in Japanese Premodern Literature

edited by Carolina Negri and Pier Carlo Tommasi

Venezia **Edizioni Ca' Foscari** - Venice University Press 2022

Images from the Past: Intertextuality in Japanese Premodern Literature edited by Carolina Negri and Pier Carlo Tommasi

© 2022 Carolina Negri, Pier Carlo Tommasi for the text © 2022 Edizioni Ca' Foscari for the present edition

# cb

Quest'opera è distribuita con Licenza Creative Commons Attribuzione 4.0 Internazionale This work is licensed under a Creative Commons Attribution 4.0 International License

Any part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means without permission provided that the source is fully credited.

Qualunque parte di questa pubblicazione può essere riprodotta, memorizzata in un sistema di recupero dati o trasmessa in qualsiasi forma o con qualsiasi mezzo, elettronico o meccanico, senza autorizzazione, a condizione che se ne citi la fonte.

Certificazione scientifica delle Opere pubblicate da Edizioni Ca' Foscari: i saggi qui pubblicati hanno ottenuto il parere favorevole da parte di valutatori esperti della materia, attraverso un processo di revisione doppia anonima, sotto la responsabilità del Comitato scientifico della collana. La valutazione è stata condotta in aderenza ai criteri scientifici ed editoriali di Edizioni Ca' Foscari, ricorrendo all'utilizzo di apposita piattaforma.

Edizioni Ca' Foscari Fondazione Università Ca' Foscari | Dorsoduro 3246 | 30123 Venezia https://edizionicafoscari.unive.it | ecf@unive.it

1st edition August 2022 ISBN 978-88-6969-608-4 [ebook] ISBN 978-88-6969-609-1 [print]

The volume has been published with the financial support of The Japan Foundation.

Images from the Past: Intertextuality in Japanese Premodern Literature / edited by Carolina Negri and Pier Carlo Tommasi — 1. ed. — Venezia: Edizioni Ca' Foscari, 2022. xiv + 188 p.; 23 cm. — (Ca' Foscari Japanese Studies; 19, 6). — ISBN 978-88-6969-609-1.

URL https://edizionicafoscari.unive.it/it/edizioni/libri/978-88-6969-609-1/ DOI http://doi.org/10.30687/978-88-6969-608-4

### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# Abstract

This volume brings together scholars from different backgrounds and career stages to rethink the role and scope of intertextuality in the context of premodern Japan. From antiquity to the rise of modernity, originality through repetition persists as a staple in the literary, performative, and artistic traditions of this country. Nonetheless, rather than slavish recycling of pre-existing tropes, the redeployment of familiar motifs by patterns of borrowing, allusion, and imitation would become a means to explore untrodden creative pathways and craft a shared sense of cultural belonging. Stemming from an international symposium hosted at Ca' Foscari University of Venice in 2021 with the generous support of The Japan Foundation, the papers in this collection offer a thoughtful contribution to this debate by engaging texts from different historical periods, media, and genres – be it poetic, narrative, theatrical, visual, or religious. Although intertextuality may not be a new topic, the essays that follow attest to the enduring appeal of a concept whose explanatory power proves most effective when combined with other methods of inquiry, such as discourse analysis, social sciences, gender studies, and material culture. Thus, while opening new windows onto Japan's literary worlds, these cross-disciplinary approaches provide further insights into the uses (and abuses) of the past in a non-Western non-modern society.

**Keywords** Japan studies. Area studies. Literary theory. Intertextuality. Interdiscursivity. Intermediality. Interdisciplinarity. Cultural heritage.

# Table of Contents



#### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# Foreword

Carolina Negri Università Ca' Foscari Venezia, Italia

#### Pier Carlo Tommasi

University of Hawaiʻi at Mānoa, USA

This volume includes fuller versions of ten papers presented during the international online symposium *Images from the Past: Intertextuality in Japanese Premodern Literature* hosted at Ca' Foscari University of Venice, Department of Asian and North African Studies, from February 2-3, 2021 with the generous financial support of The Japan Foundation.

The symposium was originally scheduled to take place from March 9-10, 2020, but one week before this date we were forced to announce its postponement to reduce the risk of COVID-19 exposure and prevent any difficulties our international guests may have faced upon returning to their home country. In that period, Italy was experiencing the greatest outbreak of Coronavirus cases in Europe and our healthcare system was dealing with the overwhelming burden of mass casualties. Among the measures for stopping the spread of the pandemic, Ca' Foscari University of Venice resolved to suspend all in-person classes as early as the end of February 2020 and cancel a series of public events. At first, only the Northern regions of Lombardy and Veneto were designated as 'affected areas' for the virus and placed under unprecedented restrictions. However, on March 9, the very same day originally planned for our conference, Italian Prime Minister Giuseppe Conte took the extraordinary step of putting the entire country on lockdown as the number of cases was rapidly increasing nationwide. After a few months of uncertainties, we decided to follow the advice of The Japan Foundation and reorganise the symposium as an online event. We are immensely grateful for this opportunity to share our expertise with esteemed colleagues around the globe despite the trying times of physical distancing we all had to endure.

The present volume aims to rethink the role, meaning, and scope of intertextuality as displayed in a variety of texts ascribed to different genres and historical periods. In the context of premodern Japan, originality was arguably not the primary goal of writing. Contrary to our post-Romantic expectations, the repetition – or rather the reenactment – of conventional tropes and motifs was the hallmark of the premodern literary discourse. Yet, far from being due to a lack of inventiveness, this act of alluding to and borrowing from canonical sources gave way to the seeming paradox of rule-bound creativity. Over the centuries, the redeployment of familiar textual fragments would serve multiple purposes, from aesthetic appeal to community building, at court and beyond. This collection of essays aims to reassess such politics of intertextuality from a cross-disciplinary perspective.

The conference that inspired this volume brought together earlycareer and senior scholars specialising in a range of fields to promote critical engagements with the Japanese past and foster a more nuanced understanding of cultural diversity by means of a theoretically-driven approach. To the informed reader, the methodological framework of intertextual studies may sound worn out after a decades-long debate. However, we contend it can still be a productive one if we use it to filter the interlacing networks of knowledge and practice in a situated context. Accordingly, the authors have brought the category of intertextuality in dialogue with primary sources to reassess its efficacy in capturing the complexity of the socio-literary phenomenon.

In his paper, Ivo Smits explores the formation of the lore surrounding the *locus classicus* of the 'Riverside Mansion' by tackling how this place was reimagined in later poetic commentaries. Giuseppe Giordano follows the traces of *Genji monogatari* (The Tale of Genji, early eleventh century) in the *Shinkokinshū* (New Collection of Poems Ancient and Modern, 1205), emphasising the artistic quality and the novelty achieved through intertextual means at the dawn of the thirteenth century. Along those lines, Itō Moriyuki looks at the unstable boundaries of life and literature in the pseudo-fictional genre of female autobiography, uncovering its connections to Murasaki's masterpiece. Moving onto medieval narratives, Simone Müller's chronotopic reading of Abutsu's *Utatane* (Fitful Slumbers, ca. 1238) demonstrates how its intertextual layering encodes multiple temporal levels, with the protagonist stretching her emotional repertoire from nostalgic longing to impending anxieties, and all shades in between. Cristian Pallone brings us forward in time with his analysis of *Tsubosumire* (The Violet in Pot, 1794), which not only complicates the conventional approaches to this text but also offers a glimpse into the shape intertextuality took during the Edo period. In a similar mode, Matthew Chudnow sheds light on the intersections of literature, religion, and performance as seen through the retelling of Yang Guifei's legend in Japan by casting the spotlight on Zeami's homonymous nō play. Similarly, Yamashita Noriko proposes an intermedial study of a kabuki *piece* by looking at the relationship between its script and a popular seventeenth-century Chinese novel, while also bringing the visual dimension to the fore with the aid of contemporaneous illustrations. Matilde Mastrangelo concludes this itinerary into the dramatic traditions of Japan by tracking the intertextual dynamics of a famous ghost story as it migrated from the puppet theatre to other forms of storytelling. The two closing papers deal with the Chinese classical corpus – either Confucian or Buddhist – and its Japanese localisation. More specifically, Kōno Kimiko unveils the intertextual potential embedded in the very materiality of the sinographs used by Monk Kyōkai when penning his parables; whereas Aldo Tollini demonstrates how the reference to foreign sources may reflect a domestic agenda, as it happened with Dōgen's (mis)quotations of scriptural pericopes.

The virtual platform of our symposium allowed for the fruitful – and safe – encounter of people, providing the occasion to build and reinforce professional ties within the Japan Studies community. During the conference, we were delighted by the attendance of almost a hundred participants, whose active engagement sparked lively discussions. We would like to express our deepest appreciation to all those who took time out of their busy schedules to join the conversations. A special thanks for their inspiring contributions goes to those speakers who are not represented in this volume: Robert Campbell (former Director of the National Institute of Japanese Literature), Christina Laffin (University of British Columbia), and Jeffrey Niedermaier (Brown University). We are equally indebted to the representatives of The Japan Foundation for their cooperation and flexibility. We would also like to express our gratitude to Professor Tiziana Lippiello, Rector of Ca' Foscari University of Venice; Professor Marco Ceresa, Chair of the Department of Asian and North African Studies; the Organising Committee, particularly Giuseppe Pappalardo and Pierantonio Zanotti for their help in arranging the Zoom sessions; Professor Bonaventura Ruperti for his constant advice and assistance; the office administrators for their dedication and hard work. Last but not least, our heartfelt gratitude goes to the members of the Student Association Gesshin for sharing their energy and enthusiasm; and to all our colleagues for their interest in this initiative. Your unfaltering support gave us the strength to make this possible against all odds.

# Images from the Past: Intertextuality in Japanese Premodern Literature

**Images from the Past: Intertextuality in Japanese Premodern Literature** edited by Carolina Negri and Pier Carlo Tommasi

# Riverside Mansion Mythologies **Retextualising the Past in Poetic Commentary**

Ivo Smits

Universiteit Leiden, Nederland

**Abstract** This essay explores one reading of Roland Barthes' notion of myth and intertextuality as a potential approach to understand how what I call the Riverside Mansion myth may be a model for the creation of cultural memory in late classical and medieval Japan. I argue that medieval commentaries (so-called *kochūshaku*) to *Kokin wakashū* and *Ise monogatari* constitute a method of creation of a Barthesian myth. The case study is Kawara-no-in (Riverside Mansion), a historical estate created by the ninth-century statesman Minamoto no Tōru (822-895), which became famous, among other things, for its supposed recreation of Shiogama, a place in northern Japan. Specifically, this essay considers as a so-called 'base-text' not the classic texts but the image of Riverside Mansion, and looks at elaborations in medieval commentaries of the concrete features of Tōru's reconstruction as creative forms of creating an understanding of the cultural role and meaning of Riverside Mansion.

**Keywords** Roland Barthes. Commentaries. Ise monogatari. Kawara-no-in. Kokin wakashū. Metatextuality. Mythologies.

**Summary** 1 *Loci*. – 2 Myths. – 3 Commentaries. – 4 Colouring Riverside Mansion. – 5 Writerly Readings.

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 3** Submitted 2021-12-13 | Accepted 2022-02-06 | Published 2022-08-30 © 2022 Smits | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/001**

Composed when, after the Riverside Minister of the Left had passed away, he went to that mansion and saw how he had constructed his garden to look like the place called Shiogama. (foretext to *Kokin wakashū* poem no. 852; Kojima, Arai 1987, 256; Author's transl.)

Back then there was this Minister of the Left. He built himself a very nice house beside the Kamo River at Sixth Avenue, and there he made himself at home. … "When can it have been, | I came to Shiogama?" (*Ise monogatari*, section 81; Sakakura et al. 1957, 158; transl. Mostow, Tyler 2010, 172-3)

#### 1 *Loci*

The myth, to borrow Roland Barthes' term, of Riverside Mansion has one of its several anchors fixed to two ultra-short texts from the early tenth century, the ones quoted above. Known in Japanese as Kawara-no-in 河原院, Riverside Mansion is the name of an estate built by an historical figure, Minamoto no Tōru 源融 (822-895), a one-time statesman whose sobriquet was 'Riverside Minister of the Left' (*kahara no hidari no ohoimauchigimi* 河原左大臣<sup>1</sup> ). Perhaps rather than the estate as a whole, it is fair to say that the 'riverside' pointed specifically to its garden.

The presumably oldest of these two *loci classici* of the Riverside Mansion myth is the 'foretext' (*kotobagaki* 詞書) 2 to a poem in *Kokin wakashū* 古今和歌集 (Collection of Poems Ancient and Modern), Japan's first royal anthology of vernacular verse (*waka* 和歌), completed in 914,<sup>3</sup> two decades after Tōru's demise. We can think of *kotobagaki* as a form of paratext that directs the reader towards grounding the poem that follows in a particular situation; that is, it suggests a specificity of occasion that forces the reader's hand in interpretation. (The ubiquity of foretexts stating "Situation unknown" [*dai shirazu* 題しらず] underscores this function, I would argue.) Like all paratexts, the *kotobagaki* is an integral part of the entire text. The foretext to poem 852 in this anthology is the plainest statement that Tōru's garden somehow contained a version of Shiogama 塩釜 (Salt Cauldron), a geographi-

This essay is part of preparatory research for a monograph on the cultural history of Kawara-no-in, with the working title *Riverside Mansion*. It is to be taken in the original meaning of *essai*, an attempt at testing the usefulness of a certain conceptual framework.

<sup>1</sup> When quoting classical Japanese, I provide a transliteration of the original orthography (the so-called *kyūkanazukai*), rather than a transcription of the pronunciation in modern Japanese.

<sup>2</sup> I borrowed this translation of *kotobagaki*, usually translated as 'headnote', from Okada (1991).

<sup>3</sup> The order for its compilation was given in 905; the last poem was added in 914, arguably finishing its editorial process.

cal *locus* situated in the far north, in Japan's 'hinterlands' known as Michinoku. The poem by Ki no Tsurayuki 紀貫之 (?-945), a member of the anthology's editorial committee, may be translated as follows:

君まさで煙たえにししほがまのうらさびしくも見え渡るかな


The myth's second *locus classicus* is section 81 of the *Tales of Ise* (*Ise monogatari* 伊勢物語), a text that has undergone considerable editorial revisions and augmentations in the course of the tenth and eleventh centuries. This textual 'place' is enigmatic terrain: the section is quite hard to make sense of without recourse to a series of extratextual assumptions. Its setting is the Riverside Mansion, where a number of noblemen gather to drink and compose poetry. Seemingly out of nowhere, "an old beggar" (*katawi wokina* かたゐ翁) who, so the reader is suddenly told, had "been roaming around below their viewing platform", also contributes a poem:

塩釜にいつか来にけむ朝なぎの釣する舟はこゝによらなん


The narrator then cryptically adds:

The thing is, he'd been to the province of Michinoku and seen many exceptionally beautiful places there. The realm has sixty provinces and more, but there's nowhere else like Shiogama. That's why the old fellow praised the Minister's place by wondering in his poem when he could have come there. (Mostow, Tyler 2010, 173)

Traditionally, two things are inferred here: first, that "this Minister of the Left" is Minamoto no Tōru and that he had designed his garden in such a way that one might pretend it could have been Shiogama; second, that the mysterious "old beggar" is none other than the presumed protagonist of *Tales of Ise*, Ariwara no Narihira 在原業平 (825-880), the grandson of an emperor. That second, not unproblematic identification I will not address here. However, the first assumption is essential in order to make sense of the poem: then it can be read as an instance of 'elegant confusion', which was a widely used rhetorical device in poetry of praise.

Why Shiogama? This, too, is somewhat beside the point of this short essay, but it has been pointed out that the standard answer, that Shiogama was an *utamakura* 歌枕, a poetic toponym, and therefore an acceptable source of inspiration in garden design, is weak if not untenable. As a 'poetic place', Shiogama came to be associated with sadness (*urasabishi* うらさびし) and loneliness. Yet, contrary to what seems the general assumption, Shiogama was *not* a name that already early on had established itself as part of the poetic repertoire. The early tenth-century *Kokin wakashū* contains only two poems that mention the 'real' Shiogama (nos. 1088 and 1089); with Tsurayuki's poem, that adds up to a total of three. In fact, throughout the tenth century one mostly sees *waka* on Shiogama that bear on the garden at Kawara-no-in, and it is only *after* the tenth century that the original Shiogama really seems to catch on as an independent *utamakura* (Nishimura 1990). It is far from obvious, then, that Tōru chose Shiogama because it was so famous: during his lifetime in ninth-century Japan, it was not.

For this one anchor of the Riverside Mansion myth, then, we have four *loci*: two textual, and two geographical.

#### 2 Myths

This essay is a heuristic attempt to find an approach that can synthesise the different identities of Tōru's Riverside Mansion throughout the course of Japan's late classical and medieval periods (roughly, the eleventh through sixteenth centuries), of which the association with Shiogama is only one, albeit the best-known. Barthes' notion of a 'myth' may point towards such an approach, and here I would like to see where that leads me.

In his book *Mythologies*, Roland Barthes (1957) sketched his understanding of the essential working of sign systems and how his concept of myth figures in it.<sup>4</sup> He begins by pointing out how semiology differentiates between the two terms *signifiant* (signifier) and *signifié* (signified: the object referred to by its linguistic or visual representation). (The classic example, I was taught, is the word 'tree', or a photo of a tree, as a signifier that refers to, or expresses, the signified organism in the park or, if one is lucky, in front of one's house.) Together, these two constitute a third term: the sign (*signe*), which

<sup>4</sup> In the second part of his book, "The Myth, Today" ("Le mythe, aujourd'hui"), and specifically the section "The Myth as Semiological System" ("Le mythe comme système sémiologique") (Barthes 1957, 217-24).

Ca' Foscari Japanese Studies 19 | 6 **6** Images from the Past: Intertextuality in Japanese Premodern Literature, 3-16

is the relationship between the two. This is a so-called semiological system of the first degree. The myth, says Barthes, has the same three-dimensional outline, but constitutes a semiological system "of the second degree". That is to say, the synthesising sign of the first degree becomes a signifier in the second degree. This means that in a myth there are two semiological systems at work: an 'object language' (*langage-objet*) in which the myth expresses itself, and a 'meta language' (*méta-langage*) that is the myth itself. The signifier of the first degree can now be regarded from two perspectives: as final result of object language and as starting point of the 'mythical system'. As a result, what is the sign in the semiological system of the first degree is the significance (*signification*) in the system of the second degree: the sign can be given new meaning. Abstract? Yes, but the point is that it allows Barthes to think of a myth as a concept that has an almost unlimited number of 'signifieds' at its disposal: the fundamental character of a myth is to be 'appropriate', he writes. The myth is 'an empty, parasitic form' that can be given new meaning, because its original contents have evaporated. Implicit in Barthes' 1957 idea of the myth is his notion of the 'empty sign'. Famously, in *The Empire of the Signs* (Fr. *L'empire des signes*; Barthes 1970), Barthes stuffed his analysis of the 'system' that was Japan with the notion of the 'empty sign': signs can be given new significance because their relationship with the signifieds is severed and is constructed anew.

In the following section in *Mythologies*, Barthes proposes as an example of a myth the notion of 'Sinity' (*sinité*) to designate the totality of associations ('bourgeois' associations, he writes somewhat condescendingly) with an imagined 'China' (Barthes 1957, 228). In analogy, and in full recognition that Barthes was interested not in cultural practices of and the creation of meaning in the distant past but of the present, we may speak of a 'Kawara-no-innity'. Like 'Sinity', this is an ugly neologism, but it is, to echo Barthes, "constructed through a reasonable analogy" (1957, 228). The 'Kawara-no-innity', then, is a mythical concept that gives access to an almost unlimited number of signifieds. Like 'Sinity' and 'Japan', the 'Kawara-no-innity' may be thought of as an empty sign.

What I call the Riverside Mansion myth is not, in the strict sense, a Barthesian myth, but it is a functional concept. While the 'Kawarano-innity' may possess several characteristics of Barthes' myth and empty sign, it is for example not the case that there are unlimited signifieds or meaning-givers for it. As far as I can tell at this stage of my research, the 'Kawara-no-innity' is more of a quadruplicity: a myth of Shiogama, a myth of Daoist realms, a myth of bygone glories, and a myth of haunted pasts. These myths require specific languages: respectively, that of poetry and prose in Japanese (*waka*) and commentary, that of the Sinitic, that of *waka* (again), and that of prose in Japanese (as well as the poetic language of the nō theatre).

Riverside Mansion's first myth, the Shiogama myth, clad in smoke, fish, and *fūryū* 風流 (elegance), is what I briefly explore here.

#### 3 Commentaries

The creation of the Shiogama myth takes places primarily through medieval commentaries, or 'old commentaries' (*kochūshaku* 古注釈).<sup>5</sup> The genre of commentary is eminently and inherently intertextual: its method is the embedding of texts in other texts, its effect (if not its goal) to beget a new text that shows its textual DNA, so to speak. A commentary is as much a text onto itself (but with 'parents') as it is a 'signpost' pointing to the First Text or base-text. A commentary does not so much explain the First Text as it reveals itself. That is why commentary traditions are such rich material for reception history and by extension cultural history. Medieval commentaries tell us perhaps more about this historical period than about whatever it is they ostensibly comment on.

Recently, Rein Raud has pointed out how one way to conceive of culture is as 'textualities': "ordered sets of texts of different status that are related to each other and come with pre-arranged modes of interpretation" (2016, 55). In such textualities there is a category of texts, the so-called 'base-texts', that are a permanent marker in a given culture, some knowledge of which is seen as proof that one is part of that culture (examples for western culture could be: the Gospel or the Eiffel Tower – the latter a very Barthesian example of an empty sign). For medieval Japan, *Kokin wakashū* and *Ise monogatari* would certainly fit that bill. What Raud calls "[t]he operational mode of a textuality" is "the imaginary space in which the base-texts are continuously (re)interpreted and result-texts produced" (2016, 68). With 'result-texts' Raud seems in the first place to think of instances of a 'text' that many of us would still consider a primary text, even if it is inspired by a base-text, but one that is not in circulation very long (the 1956 sci-fi movie *Warning from Space* comes to mind, but I could be wrong). Nonetheless, I believe we can expand this notion to encompass commentaries. Cumulatively, commentaries form what Gérard Genette has called "metatextuality" (1977, 4, 8), which is at once a genre as well as a space in which the First Text or base-text is continuously given new meaning. Commentaries may be thought of as the manifestation of this discursive space. Rather than think of commentaries as a 'delayed response' to a base-text, I am interest-

<sup>5</sup> Here I make no formal distinction between 'old commentaries' (*kochūshaku*, lasting until roughly until the Ōnin war of 1467-77) and the subsequent so-called 'transitional commentaries' (*kyūchūshaku* 旧注釈) of the late fifteenth and sixteenth centuries.

Ca' Foscari Japanese Studies 19 | 6 **8** Images from the Past: Intertextuality in Japanese Premodern Literature, 3-16

ed in the corpus of commentary as discourse onto itself in which the base-text is a sign but not the object of inquiry per se. The *totality* of commentaries, that is, the commentary *tradition*, is a unit. To speak, for example, of *kokin denju* 古今伝授, (the transmission [of interpretations] of *Kokin wakashū*) is to point to a (perceived) coherence of the act of discussing *Kokin wakashū*, and not so much to an assortment of individual commentaries.

Two important aspects of the study of medieval Japanese commentaries that I will not elaborate upon here are the role that commentaries played in establishing genealogies of cultural authority, and the role of such texts in oral interaction.<sup>6</sup> Much important work has been done in historicising approaches to commentary. That the commentary traditions were part and parcel in confirming authority in matters of poetry and other literature is eminently clear, as is the importance of physical copies of such commentaries with attested authenticity, for example by signed certification of licensed transmission (Cook 2000, 53). In this context, the content of knowledge was perhaps less important than the legitimacy of access to that knowledge. Furthermore, in traditional Japan, we cannot assume that commentaries functioned as 'stand-alone' texts. That is to say, the commentaries are the written residue of an oral interaction between instructor and pupil; hence the term *kuden* 口伝, 'oral transmission', for written texts. These markers of exclusivity notwithstanding, I would like to treat the collectivity of commentaries as one corpus.

Moreover, rather than separate commentary traditions for *Kokin wakashū* and *Ise monogatari* respectively, I would like to weave these two strands together and take them as one metatext to explore the myth of Kawara-no-in, and to consider as base-text not the two textual *loci* but the signifier of that myth: the estate. In other words, Riverside Mansion is itself a base-text that is given new meaning through a totality of discursive lineages that incorporate the estate in their discussions, explanations, and elaborations.

#### 4 Colouring Riverside Mansion

The metatext I indicate above, that is, said commentary traditions for *Kokin wakashū* and *Ise monogatari*, constitutes a massive corpus. Here I will highlight only a few samplings in more or less chronological order.

Interestingly, a relatively early example for *Kokin wakashū* comments not so much on the Tsurayuki poem as on its foretext. *Kokinshū chū* 古今集註 (Notes to *Kokin wakashū*), one of several commentaries with that name, is a fourteenth-century Nijō commentary to the an-

<sup>6</sup> For more on this in English, see for example Cook 2000; Newhard 2013; Aoki 2021.

Ca' Foscari Japanese Studies 19 | 6 **9** Images from the Past: Intertextuality in Japanese Premodern Literature, 3-16

thology's reception within the once competing but by then basically defunct Rokujō house and may reflect Nijō Tameuji's 二条為氏 (1222- 1286) notes to the Rokujō edition of that anthology. On Tsurayuki's *kotobagaki* it has this to say:

Note. The Kawara Minister of the Left is Lord Tōru. In this lord's near garden, he had built [re-constructed] all [or: completely] the famous places from the [more than] sixty provinces. After the minister [passed away], it was converted into a palace for the Kanpyō retired emperor [Uda, 867-931]. This villa occupied four blocks located south of Rokujō Bōmon, north of Rokujō, east of Made no Kōji and west of the river bank. (*Kokinshū chū*; Yoshizawa 1935, 190; Author's transl.)

It is intriguing that it makes no attempt to provide information that might explain Tsurayuki's poem. What we do encounter here is a preoccupation with Riverside Mansion's exact location that recurs in almost all of the commentaries, both to *Kokin wakashū* and to *Ise monogatari*. The mapping of Tōru's villa on the existing grid of the capital becomes something of a fixed rhetorical gesture and underscores a grounding of the base-text in a historical and almost tangible reality. The address is already present in a commentary that goes back to Kenshō 顕昭 (1130-1209), a proponent of the Rokujō school in *waka* poetry. In 1183 Kenshō presented a commentary to *Kokin wakashū* to his patron, 'dharma prince' Shukaku (Shukaku Hōshinnō 守覚法 親王, 1150-1202). In this *Kokinshū chū* 古今集注, to which later commentators kept adding, a new feature of the Riverside Mansion is introduced, one that was already passed on earlier within the Rokujō school and apparently could be traced back to the eleventh century:

This house is Riverside Mansion. It lies south of Rokujō Bōmon, north of Sixth Avenue, east of Made no Kōji, and west of the river bank, and measures four blocks. They put thirty *koku* 石 [ca. 5,400 litres] of salt water in the pond every month, and sea fish and other salt water creatures were kept in it, as is noted in Lord Kiyosuke's<sup>7</sup> commentary. After the Minister [passed away], it was converted into a palace retreat for the Kanpyō retired emperor [Uda]. A private [note] says, originally it was called East Sixth Avenue Mansion. Now it is a temple. Lord Takakuni's<sup>8</sup> commentary notes that they made a model of Shiogama in Michinoku, and filled [its pond] with salt water.

<sup>7</sup> Fujiwara no Kiyosuke 藤原清輔 (1104-1177), a poet and commentator of the Rokujō school.

<sup>8</sup> Minamoto no Takakuni 源隆国 (1004-1077).

[A note at the top:] "Yoshimune<sup>9</sup> surmises that the salt water in the garden [pond] of that house was hauled by several hundreds of servants every day from Amagasaki Bay in Settsu". (*Kokinshū chū*; Kyūsojin 1980, 327; Author's transl.)

This is where the fish come in. The story that in his recreation in the garden of Riverside Mansion Tōru aimed for a degree of hyperreality is well-known. It is also a story that, the further removed in time from the historical Riverside Mansion it is, gets more elaborate and detailed. One way in which this apocryphal design strategy of Tōru's operated was to fill the pond, a central element in any garden of a Heian-era (794-1185) estate, with salt in mimicry of seawater; after all, Shiogama is situated on the sea coast. The details may vary: two, three, twenty, or thirty *koku* are mentioned; rather than seawater, the pond is filled with salt; a monthly or even a daily transport is recorded, etc. The oldest example in Japan of such a story that I have so far located concerns a different, mid-tenth-century estate in the Heian capital, but one that has a number of parallels to that of Riverside Mansion. The source is a Sinitic text by Minamoto no Shitagō 源順 (911-983), and the context is the conjuring up of an enchanting but now lost garden. This suggests that the, apparently later, story of Tōru's habit to fill up his pond with salt taps into an existing embellishment that bespeaks the wealth of the estate's owner as well as the fantastic nature of the estate itself.

The fish are, equally fantastic, sea fish such as *tai* 鯛 (sea bream) and *suzuki* 鱸 (sea bass), as is explained in a thirteenth-century commentary to *Tales of Ise*, the *Waka chikenshū* 和歌知顕集 (Collection of Manifest Knowledge about Poetry, pre-1260):

Minamoto no Tōru … At Rokujō Takakura he had [a garden] built to look like Shiogama. On the bank [of the pond] he built a salt hut such as divers use and had smoke rise up from it. At the foot of the boards of the fishing pavilion he organised boat races and such. Each month he sprinkled three *koku* salt in the pond and released fish in it that normally live in the sea and kept them there, so that sea bream and sea bass frolicked and leaped up at the boards [of the fishing pavilion]. It was fascinating and he invited several princes. If I have to name these princes, they were the prince of the Kaya palace, Prince Saneyasu, the prince of the Urin'in, Prince Koretaka, and such.10 The present Riverside Mansion is that Shi-

<sup>9</sup> It is not known who this Yoshimune was. He left annotations to a copy of Kenshō's notes to *Kokin wakashū*.

<sup>10</sup> These princes were associated with *Ise monogatari*. Saneyasu figures in section 78, Koretaka in section 82. This touches on a different discussion, namely the theme of representations of thwarted power, which I will not take up here.

ogama. He copied the actual Shiogama, but soon [Riverside Mansion] was called the actual Shiogama. (*Waka chikenshū*; Katagiri 1969, 176; Author's transl.)

Here we also see what was already intimated, that as a geographical *locus* Riverside Mansion replaced the actual or 'unmediated' (*dzika* 直) Shiogama in Michinoku. The signified Shiogama usually was Tōru's estate garden in the capital. The fourteenth-century Nijō commentary mentioned above even claims that one of the two poems in *Kokin wakashū's* section of "Songs from the East" (*adzuma-uta* 東歌, no. 1088) is, in fact, about the replica in Tōru's garden (Yoshizawa 1935, 240). The realism of this superimposition was supposedly heightened by having 'divers' or sea folk (*ama* 海人) operate salt kilns and catch sea fish, and marking the bank of the pond with sea shells:

His house in the Rokujō area is the present Kawara-in. There he built [a garden] in the appearance of Shiogama Bay and always had smoke rise up. Boats were set afloat and he had divers go fishing [from them]. Each month he put two *koku* salt in the pond and had sea fish such as bream and bass released in it and kept them there. On the banks [of the ponds] he scattered empty shells crushed by the sea and in this way it really was no different from Shiogama. (*Waka chikenshū* [Shimabara Bunko-bon]; Katagiri 1969, 275; Author's transl.)

These details are perpetuated in various *Ise* commentaries, including the authoritative fifteenth-century *Ise monogatari gukenshō* 伊勢物 語愚見抄 (Humble Views of *Ise monogatari*, 1460) by Ichijō Kaneyoshi 一条兼良 (var. Kanera, 1402-1481), in which once again a suspense of disbelief is made possible not only by the authority of commentary, but also by the precision of the identification of location:

The Kawara Minister of the Left built a spectacular house at Sixth Avenue Riverside, dug a pond, and filled it with water, and every day he transported thirty *koku* of salt and put it in [the pond], and had sea fish and shells from the bottom of the sea live in it. He had smoke rise from divers' salt huts and amused himself in this way. This is at East Sixth Avenue Mansion. (*Ise monogatari gukenshō*; Katagiri 1969, 561; Author's transl.)

The operative language of the commentaries as a rule is Japanese. However, Kenshō's *Kokin wakashū* commentary quoted earlier proceeds with quoting in full a description in Sinitic of Riverside Mansion:

In the Petition at the Behest of Retired Emperor Uda to Perform a Sutra Reading after the Death of the Kawara Minister of The Left, a work by Ki no Arimasa,<sup>11</sup> it says, "Riverside Mansion was the residence of the late Minister of the Left, Lord Minamoto. Forest and springs he selected for the neighbourhood; the clamour [of the city] was kept at bay. He chose the [suitable] grounds and built [the mansion and garden]. Although it was on the east side of the Eastern Capital [the eastern or 'left' section of Heian-kyō], when he entered its gates and resided [there], it was as though he had escaped to the north of the Northern Hills [Kitayama]". (*Kokinshū chū*; Kyūsojin 1980, 327; Author's transl.)

Such a crossover into a different language of scholarship is not unusual in medieval literary commentaries, but usually confined to shorter passages. Here we see a second crossover as well, one into the Riverside Mansion myth of haunted pasts (the petition's point is to placate Tōru's ghost). All Kawara-no-in myths are holistically connected, of course, but what this passage also does is to underscore the enchanted nature of Riverside Mansion, emphasising its sophistication (*fūryū*). To walk through its gates is to enter a different world, far removed from the bustle of the capital.

Other Sinitic texts can be evoked to the same effect, as in these fifteenth-century lecture notes (*kikigaki* 聞書) by *renga* master Sōchō 宗長 (1448-1532) recording his teacher Sōgi's 宗祇 (1421-1502) lectures on *Tales of Ise*, the *Ise monogatari Sōchō kikigaki* 伊勢物語宗 長聞書 (1491):

"He built himself a very nice house": The present Riverside Mansion is what is left of [that house]. It was a reconstruction of Shiogama Bay. "In its pond he set free whales and cetaceans; | in its mountains he kept tigers and wolves". (*Ise monogatari Sōchō kikigaki*; Katagiri 1969, 693; Author's transl.)

With the lines "In its pond he set free whales and cetaceans; | in its mountains he kept tigers and wolves" (*ike ni keigei wo hanachi, yama ni korau wo sumasu* 池放鯨鯢、山住虎狼) we enter the realm of the truly fantastic. Conversely, one might say, we enter deeper into the realm of literary embellishment and poetic hyperbole in the description of Tōru's reconstruction (*utsushidokoro* うつし所) of Shiogama. To my knowledge, this is a first instance in which commentaries quote from a now lost anonymous "Poetic Exposition on the Riverside Mansion" (*kahara no in no fu* 河原院賦), which is also quoted by Hosokawa Yūsai 細川幽斎 (1534-1610) in his *Ise monogatari ketsugishō* 伊勢物語闕疑抄 (Doubting Commentary on *Ise monogatari*, 1596; Katagiri 1969, 820).

<sup>11</sup> *Honchō monzui* 本朝文粋 (Literary Essence of Our Court, 1066), 14, no. 427 (Ōsone et al. 1992, 115-18, 375).

Ca' Foscari Japanese Studies 19 | 6 **13** Images from the Past: Intertextuality in Japanese Premodern Literature, 3-16

Rather than spell out the Shiogama myth of Tōru's estate, the commentaries in fact *create* it. The sparse formulations in *Kokin wakashū* and *Ise monogatari* allow for a dynamic construction of a Riverside Mansion base-text that readers actively colour in to meet their expectations.

#### 5 Writerly Readings

As stated, my aim here is not to squeeze Kawara-no-in into a Barthesian Procrustes' bed, but rather to find a way to understand how the Riverside Mansion myth may be a model of the creation of cultural memory in late classical and medieval Japan. That is a different thing. I am not engaged with charting how a reader today might engage in deviation or "derivation" (*dérive*, Barthes' term in *Le plaisir du texte*) to find pleasure in the act of reading. Rather, I am interested in a form of reconstruction. The value of such reconstruction is partly hermeneutical: it is an exercise in understanding other cultures and as such is a testing of an approach.

While Barthes' position seems to have been that intertextuality was a decidedly modern form of readers' interaction with a text, Haruo Shirane (2008, 9) has, in an apparent and creative use of the Barthesian distinction between 'readerly' (*lisible*, fairly straightforward) and 'writerly' (*scriptible*, demanding an effort) texts, argued that traditional Japan knew both 'readerly' and 'writerly' receptions of such classic works as *The Tale of Genji*. Commentaries are given as an example of readerly reception, and parodies and adaptations are examples of a writerly reception. However, when we think of the Riverside Mansion myth and the role that commentaries play in construing it, we may accept that commentaries, too, can be a writerly engagement with base-texts and have a fully active part in the production of their meaning, of which Barthes would say that there always is a plurality. That commentaries build on one another to weave new meanings suggests their creative potential.

#### **Bibliography**


#### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# Traces of *Genji monogatari* in *Shinkokinshū*

#### Giuseppe Giordano

Università degli Studi di Napoli "L'Orientale", Italia

**Abstract** In this paper, after mainly dealing with the poetics of Fujiwara no Shunzei and of his son Teika, above all in relation to *Genji monogatari*, I analyse some poems from *Shinkokinshū* that evoke, through the so-called allusive variation, the world of the Shining Prince, and I wish to point out the peculiarities of the *shinkokin*-period poets' composition techniques.

**Keywords** Japanese poetry. Waka. Fujiwara no Shunzei (Toshinari). Fujiwara no Teika (Sadaie). Genji monogatari. Shinkokinshū.

**Summary** 1 Fujiwara no Shunzei's Vision of *Genji monogatari*. – 2 Fujiwara no Teika's Vision of *Genji monogatari*. – 3 *Mumyōzōshi* on *Genji monogatari*. – 4 Traces of *Genji monogatari* in *Shinkokinshū*. – 4.1 Shunzei's Daughter. – 4.2 Fujiwara no Teika. – 4.3 Ex-Emperor Go-Toba. – 5 Conclusions.

#### 1 Fujiwara no Shunzei's Vision of *Genji monogatari*

*Shinkokinshū* 新古今集 (New Collection of Poems Ancient and Modern, 1205) appears to be an emblematic work when we look at the aptitude of Japanese scholars of the Heian period, and of the following ones too. Its peculiarity emerges when we focus attention on the literary production of the past which should be reinterpreted into something to be cherished, preserved and, to some degree, renewed.

It goes without saying that the golden age, which Heian and Kamakura poets turned their gaze to, was the so called *kokinjidai* 古今時代,

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 17** Submitted 2021-12-13 | Accepted 2022-03-21 | Published 2022-08-30 © 2022 Giordano | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/002**

the time when *Kokinshū* 古今集 (Collection of Poems Ancient and Modern, ca. 905) was edited and, likewise, the period in which *Genji monogatari* 源氏物語 (The Tale of Genji, early eleventh century) is set. The main tool the poets of the *shinkokinjidai* 新古今時代 (the period of *Shinkokinshū*), above all those who belonged to Go-Toba's 後鳥羽 (1180-1239) *kadan* 歌壇 (poetry circle), would use to express their devotion to the past was, no doubt, *honkadori* 本歌取り, the allusive variation, one of the most frequently used technique in *Shinkokinshū*.

In his encyclopaedia of Japanese poetry, Ariyoshi Tamotsu (1982, 347) states that the anxiety of medieval authors to follow the trail of their predecessors was rooted in the shock they had received when the violent Hōgen 保元 (1156) and Heiji 平治 (1159) disorders broke out. The aristocrats, after crashing down to reality, as Kamo no Chōmei 鴨長明 describes in his *Hōjōki* 方丈記 (An Account of My Hut, 1212), reacted by refuging in a world of romantic beauty. In other words, the nobles of the Imperial Court, by devoutly studying the literary masterpieces of the past, tried to revive after three centuries the splendour of the golden age, on paper at least.

As we know, it was Fujiwara no Teika 藤原定家 (1162-1241) the one most responsible for canonising the right use of *honkadori*. He took a cue from his father Shunzei's 俊成 (1114-1204) idea, "old words and new spirit" (*kotoba wa furuki o shitai, kokoro wa atarashiki o motome*  詞は古きを慕ひ、心は新しきを求め), and set guidelines for an excellent allusive variation (Hashimoto et al. 1975, 471). When talking of "old words", he is above all up to the diction of *Kokinshū*, since he was absolutely sure that the first poetic anthology edited by an imperial order was the model par excellence for contemporary poetry. In his famous *Korai fūteishō* 古来風躰抄 (Notes on Poetic Style Through the Ages, 1197, revised in 1201), Shunzei claims that this is due to the fact that, during that age, for the first time a distinction between good and bad poetry had been made. That is why poets would worship *Kokinshū* and take it as a reference model for the basic poetic styles (Hashimoto et al. 1975, 288; see also Royston 1968, 2). Therefore it is not so strange that in the *Shinkokinshū* there are several poems that allude in a way or another to a *Kokinshū* poem.

But Shunzei did love another masterpiece of the past as well: *Genji monogatari*. In his *Shōji ninen Shunzei-kyō waji sōjō* 正治二年俊成卿 和字奏状 (Lord Shunzei's Memorial in Japanese Script, submitted in 1200), the letter Shunzei wrote to plead his son's case with Go-Toba, begging the former Emperor to let Teika take part in the poetic contest the sovereign was planning those days, he chose, as evidence of his opponents' incompetence, their ignorance of *Genji monogatari*. Surprisingly, Fujiwara no Norinaga 教長 (1109-?) and Fujiwara no Kiyosuke 清輔 (1104-1177), in compiling the *Shūi kokin* 拾遺古今 (Gleanings Old and New, date unknown) had made a serious error:

In the first place, the poem "Of a spring night, neither shining brightly nor yet completely clouded…" they took to be on the subject of "summer nights" and placed it in the summer section. In the *Hana no en* chapter of *The Tale of Genji*, which takes place in the Second Month, it is this poem that the Chief Palace Attendant (Oborozukiyo) is depicted as quoting when she speaks of "the light of a misty moon". But Norinaga and Kiyosuke have not read *Genji*, and certainly have not read [*Hakushi*] *monjū*. In one of his poems Haku Kyoi writes, "Neither bright nor dark, the misty moon; neither warm nor chill, the gentle breeze". This is the Chinese poem upon which the Japanese poem is based. Knowing neither of them, they describe it as a poem on "summer nights" and place it in the summer section. Both Norinaga and Kiyosuke are a disgrace. (Harper, Shirane 2015, 164-5)

So, it is possible to affirm that at some point *Genji monogatari* was recognised as having the same artistic and literary quality as *Kokinshū*. As evidence of that, we can look at the way poets started using *Genji* as reference material for their *honkadori* or *honzetsudori* 本説取 (taking a foundation passage) (Minemura 1950).<sup>1</sup>

The statement once Shunzei made is pretty famous: *Genji mizaru utayomi wa ikon no koto nari* 源氏見ざる歌詠みは遺恨の事なり, that is to say, "writing poetry without knowing *Genji* is quite regretful" (Shinpen kokka taikan 1983-92, 5: 294). We will soon come back to it. In this regard, Minemura quotes a remark by Go-Toba, who was quite familiar with Shunzei's poetry, which he loved a lot, to stress how the same Go-Toba was a careful reader of *Genji*: "I recall Shakua's 釈阿<sup>2</sup> poetry as gentle and evocative [*en ni* 艶に], infused with deep feeling [*kokoro mo fukaku* 心も深く] and moving in its sensitivity [*aware naru* あはれなる]" (Brower 1972, 35-6).

This passage is illuminating, because in the "Azumaya" 東屋 (The Eastern Cottage) chapter of *Genji monogatari* we find the following words:

He was so kind [*en naru*], and he spoke to her so gently [*kokoro fukaku aware ni*], that despite bitter sighs over what her mother might be thinking, she plucked up the courage to get down as well. (Tyler 2003, 1003)

<sup>1</sup> The distinction between the two techniques is well known: the former is the allusive variation through the quotation of old verses, while the latter simply means that the author echoes an old prose passage in his/her verses.

<sup>2</sup> Shunzei's priestly name.

Even though we can't be sure Go-Toba quoted this exact point by heart, the similarities between the two passages are impressive (Minemura 1950).

As mentioned before, Shunzei drew inspiration from *Genji monogatari* not just for *aware* あはれ, but for the ideal of *en* (charming, fascinating) as well. According to Shunzei, a poet who bases his own poetry on *en* can reach the highest degrees of poetic expression; and one can get plenty of *en* only by closely reading *Genji monogatari*, avoiding to consider it as a simple entertainment medium, because the greatness of *Genji* is neither limited to its plot and stories, nor to its numerous characters' love intrigues, high hopes, burning desires and disappointing delusions. When Shunzei affirms that it is not possible to compose good poetry without having read *Genji monogatari*, he seems to mean that a real poet must be able to easily handle all those elements hidden between the lines of the text, the exquisite style of Murasaki Shikibu 紫式部 (973?-1014?), and her incredible literary sensibility.

In order to show how important *Genji monogatari* was for Shunzei to reach high levels of poetic splendour, we can read in full his judgement briefly touched on earlier. It's the thirteenth match of the summer section of *Roppyakuban utaawase* 六百番歌合 (Poetry Contest in Six Hundred Rounds, 1193), and the pre-assigned topic is the *yūgao*  夕顔 (Moonflowers). The opponents are Fujiwara no Yoshitsune 良経 (1169-1206) and Fujiwara no Iefusa 家房 (1167-1196).

**265** かたやまのかきねの日かげほの見えて露にぞうつるはなのゆふがほ<sup>3</sup>


**266** をりてこそ見るべかりけれゆふつゆにひもとくはなのひかりありとは


This match is quite interesting because, before the judge's decision, both opponents stated their points of view. Iefusa says: "It's true, the poem by the Left depicts the *yūgao* flower, but the spirit of the topic is quite weak". To this, Yoshitsune replies: "The poem by the Right clearly alludes to *Genji monogatari*, but I wonder if this is enough for a poet-

Ca' Foscari Japanese Studies 19 | 6 **20** Images from the Past: Intertextuality in Japanese Premodern Literature, 17-40

<sup>3</sup> The text and poem numbers used are those appearing in Kubota, Yamaguchi 1998 (Author's transl.).

ry contest". The last word is Shunzei's, who states: "The poem by the Left does not fully respect the essence of the topic, but I wonder why, instead of *yūgao no hana*, the author writes *hana no yūgao*. He may be aiming at the effect of singularity". As we can see, explicitly naming the *yūgao* flower does not guarantee success. At the same time, Shunzei declares that "not only does the poem by the Right allude passionately to *Genji monogatari*, but its style is just superb. For this reason, victory goes to the Right" (Kubota et al. 1998, 99; Author's transl.).

The chapter alluded to is clearly the fourth one, "Yūgao". In the poem by Iefusa, two poems seem to resonate. The first one is addressed to Yūgao by Genji, who is asking to meet her:

I see, Genji thought, it must be the young woman in service. She certainly gave me that poem of hers as though she knew her way about! She cannot be anyone in particular, though.

Still, he rather liked the way she had accosted him, and he had no wish to miss this chance, since in such matters, it was clearly his way to be impulsive. On a piece of folding paper, he wrote in a hand unlike his own,

*Let me draw and see whether you are she, whom glimmering dusk gave me faintly to discern in twilight beauty flowers.* (Tyler 2003, 57-8)

The second one is the poem Yūgao composes to answer Genji, later in the chapter:

"The place is eerie", he said, "but never mind: the demons will not trouble me". She was thoroughly offended that he still had his face covered, and he agreed that this was unnatural by now.

*The flower you see disclosing its secrets in the evening dew glimmered first before your eyes in a letter long ago.*

he said. "Does the gleam of the dew please you?" With a sidelong glance she murmured,

The light I saw the dewdrops adorning then a twilight beauty *was nothing more than a trick of the day's last fading gleam!* (Tyler 2003, 65-6)

We should also keep in mind that re-echoing *Genji monogatari* was not to be a mere mechanical reminiscence. In fact, these allusions had to widen the literary universe of Shikibu's masterpiece. In this regard, in his *Go-toba-in gokuden* 後鳥羽院御口伝 (Ex-Emperor Go-Toba's Secret Teachings, first half of thirteenth century), Go-Toba writes:

Shakua, Jakuren, and others have said that when composing poems for a poetry contest, a person may not be too free in expressing his originality. However, poems composed for such an occasion are actually put together no differently from other poems. "Give careful thought to the significance of the topic, make sure your poem is free of poetic ills, and use diction from the poems in the Genji and other romances if you like, but do not use poetic conceptions from such works", they said. (Brower 1972, 34)

In this passage the author reproduces the opposition between *kotoba* 詞 (words) and *kokoro* 心 (spirit) and his statement could be read with reference to the "old words, new spirit" motto. However, such a reading leaves some doubts in that a poet like Teika, on drawing inspiration from *Genji monogatari*, used to resort to *honzetsu* rather than to *honkadori*, that is to say, he focused more on the spirit than on the wording of the *monogatari*.

Minemura (1950, 36-7) cites another comment by Shunzei to a poem by Fujiwara no Takanobu written for *Roppyakuban utaawase* (turn eighteenth, right), which reads:



The judge says that the poem overlaps with the *monogatari*, which is not good as the lines do not open up new and original horizons, nor do they present any original resonances. According to Minemura, this shows once again Shunzei's great respect for *Genji monogatari*.

### 2 Fujiwara no Teika's Vision of *Genji monogatari*

In this respect, Teika seems to have well understood his father's teachings. As we have already said, in his theoretical works, Teika attached great importance to *honkadori*. Nevertheless, when it came to *Genji monogatari*, he didn't limit his inspiration to the poems interwoven in the text, but according to him the atmospheres and the *kotobazukai* ことばづかひ (wording) of some particular scenes were to be worshipped as well.

This kind of attitude is already startlingly clear in the poems he wrote for *Shōji shodo hyakushu* 正治初度百首 (First Hundred-Poem Sequence of the Shōji Era, 1200). This *utaawase* was the first court poetry contest Go-Toba organised for his *kadan*, and hereafter it turned out to be one of the most important sources for *Shinkokinshū*, with its 79 poems chosen for the anthology, second only to *Sengohyakuban utaawase*  千五百番歌合 (Poetry Contest in Fifteen Hundred Rounds, 1202-03).<sup>4</sup>

Obaishi (2014) focuses on the following *waka*, whose *dai* 題 (topic) is *sanka* 山家 (mountain hermitage):

**988** 浪の音に宇治のさと人よるさへやねてもあやうき夢のうきはし


In most cases, the true spirit of a *dai*, such as 'mountain hermitage', was thought to be fully respected only if verses would depict someone who had fled from the world to live in a poor and shabby hut. For this reason, it was not so unusual that such poems were set in Uji, or in other places famous for their desolating panoramas, such as Yoshino, Fukakusa or Ōhara. That is why the poem just quoted does not seem to stray too far from traditional diction. But, until that moment, when a poet chose to set his verses in Uji, the suitable elements used to describe the place would be river wicker fences, whitebaits, the fog on the river, the Uji bridge, or the Maiden of the Uji Bridge and so on. But Teika focuses on the sound generated by the river. This is important, because, at that time, this kind of sound image was an absolute novelty.

So, where did Teika draw inspiration from? Obaishi (2014, 34-6) thinks the answer is in "Hashihime" 橋姫 (The Maiden of the Bridge) chapter of *Genji monogatari*:

It was a sadder place than he had been led to imagine, and considering who His Highness was, everything about his life there suggested the drastic simplicity of the grass hut built to last little more than a day. There are other quiet mountain villages with an appeal all their own, but here amid the roar of waters and the clamor of waves one seemed unlikely even to forget one's cares or, at night amid the wind's dreary moan, to dream a consoling dream.

Surroundings like these undoubtedly stir thoughts of renunciation in His Highness, the Captain reflected, inclined as he is to seek a holy life, but now must they affect his daughters? (Tyler 2003, 834; emphasis in the original)

<sup>4</sup> On the importance of each *utawaase* 歌合 (poetry contest) for the *Shinkokinshū*, see Huey 2002; 2000.

<sup>5</sup> The poem number refers to the numeration in Kubota 1985-86 (Author's transl.).

The idea that the "roar of waters and the clamor of waves" prevent people from sleeping is something we find in "Ukifune" 浮舟 (A Drifting Boat) chapter as well, when the young Ukifune is tormented by Niou and Kaoru's love. In that particular scene, the river sound becomes a meaningful counterpoint to the girl's anxiety. Here, is Ukifune's mother who is speaking:

Outside, the river roared menacingly past. "Not all rivers sound like that. No wonder he has taken pity of her, when she has had to spend so long in a place so dismally wild!" her mother remarked with satisfaction.

One of the women described how fast and frightening the river had always been. "The only day, you know, the ferryman's grandson missed his stroke on the oar and fell in. That river has taken so many people!" With that anyone agreed. (Tyler 2003, 1032; emphasis in the original)

Once it was night again, she lay sleepless, planning a way to get out of the house without being seen. (1042; emphasis in the original)

At this point in the narration, we are approaching the end of the conflict between the calm and gentle Kaoru and the passionate Niou. Ukifune receives a letter from Kaoru, who is inviting her to the capital and her handmaids, ignoring their mistress' state of mind, keep on preparing for departure.

Here, the roaring river waves become the perfect soundscape for the turmoil of Ukifune's heart, torn on what to do. The suggestion that the river can kill is meaningful as well. Ukifune's inability to sleep seems to blend well with the word *ayauki* used in the fourth verse of Teika's poem, where we find an allusion to the last chapter of *Genji monogatari*, "Yume no ukihashi" 夢浮橋 (The Floating Bridge of Dreams), and to the maiden's pain who can't easily cross the "floating bridge of dreams".

The poems inspired by *Genji monogatari* that Teika composed for *Shōji shodo hyakushu* are in all six, but not in all of them *honkadori* is used. The following *waka*, for example, is not inspired by another poem, but just to a passage in prose of the *monogatari*, and gives life to a *honzetsu*:


The passage we can glimpse between Teika's lines is the following one from the *Akashi* chapter:

In this wilderness where I am a stranger, I have suffered every outlandish affliction, and yet no one brings me words of comfort from the City. Your fishing boat is a welcome refuge, when my only old friends here are the sun and the moon in their course across the sky! (Tyler 2003, 260)

Teika didn't make clear his devotion toward *Genji monogatari* only in his poetry, but in his theoretical production as well. In a text edited in 1228 by Fujiwara no Nagatsuna 藤原長綱 (unknown dates), *Teika-kyō sōgo* 定家卿相語 (Discussion with Lord Teika), we can find the following passage:

Lately, the way people read and annotate *Genji monogatari* has changed in some respects. To realise a *honkadori*, they tend to pick up some poems from it; or, posing as real experts, they make a dispute about the identity of lady Murasaki's mother, debating about her genealogical tree. In the past it was not like that. The way I see it, even though we avoid speculating about Murasaki's ancestors, even though we do not make any effort to realise a *honkadori*, since the words are used in an ineffable way, just enjoying the same Murasaki's style will clear our spirit up, and we will be stimulated to compose bewitchingly elegant poetry, refined in aspect and words. (Hisamatsu 1971, 335-6; Author's transl.)

To Teramoto (1961, 68), this passage is very interesting, because it lets us understand two different things. First of all, when Teika says "even though we avoid speculating about Murasaki's ancestors, even though we do not make any effort to realise a *honkadori*", he shows his intention to distance himself from this kind of historical investigations about the small plots of the narration, and to be interested just in the poetical aspect of the work.

On the other hand, Teika in his *Kindai shūka* 近代秀歌 (Superior Poems of Our Time, 1209) wrote some words doomed to become extremely famous:

<sup>6</sup> The poem number refers to the numeration in Kubota 1985-86 (Author's transl.).

If in diction you admire the traditional, if in treatment you attempt the new, if you aim at an unobtainable lofty effect, and if you study the poetry of Kampei and before – then how can you fail to succeed? (Brower, Miner 1967, 44)

We do find the same idea in the incipit of *Eiga taigai* 詠歌大概 (The Essentials of Poetry, ca. 1222) (Hisamatsu 1971, 299). There, while reaffirming that diction should be ancient but the spirit of poetry should be fresh and new, Teika stresses that it is the poetry of *Sandaishū*  三代集, the first three imperial anthologies of Japanese *waka* poetry, that has to be taken as a model:

In a poem, more than anywhere else, priority should be given to original inspiration. That is to say, one should compose poems in a spirit that cannot be found in poems written by others. As regards diction, it is necessary to use the words from ancient poems. These must never be different from the ones used by the great poets represented in *Sandaishū*. (Hashimoto et al. 1975, 493; Author's transl.)

Teramoto (1961), reasoning about the link between *Genji monogatari*  and Teika's poetry, analyses the following five poems Teika wrote on different occasions.<sup>7</sup>

**192** よそにてもそでこそぬるれみなれ棹猶さしかへる宇治の川長


#### **3075** さしかへる宇治の川長袖ぬれてしづくのほかにはらふ白雪


**2501** 如何せんさすがよなよなみなれざをしづくににごる宇治の川長


<sup>7</sup> The numeration of the following five poems refers to Kubota 1985-86 (Author's transl.).

<sup>8</sup> The real meaning of this poem is: now that I am accustomed to meeting you every night, what can I possibly do?

**(2501)** 浮舟のなにの契りにみなれ棹あだなる袖をくたしそめけむ


#### **2072** 花の色のをられぬ水にこすさをのしづくもにほふ宇治の川長


All these five poems refer to the following poems exchange between Kaoru and Ōikimi we find in the "Hashihime" chapter:

橋姫のこころを汲みて高瀬さす棹のしづくに袖ぞ濡れぬる


さしかへる宇治の川長朝夕のしづくや袖をくたしはつらむ


Teramoto (1961, 73) points out how these two *honka* fall into the category of the so called *jukkai* 述懐 (lament), but the first, the second and the fifth of the ones by Teika seem to be purely descriptive. Moreover, in the second and fifth ones, while quoting explicitly the *honka*, Teika introduces a paradigm shift in diction by using the image of white snow and flower scent along with the ferryman's oar, obtaining an absolutely new and fascinating effect.

9 Abe et al. 1970-76, 5: 141-2; Author's transl.

#### 3 *Mumyōzōshi* on *Genji monogatari*

Before starting the analysis of the poems of *Shinkokinshū*, it is impossible not to mention the *Mumyōzōshi* 無名草子 (Nameless book, early thirteenth century). As it is well known, *Mumyōzōshi* is an anonymous text, attributed to Fujiwara no Shunzei no Musume 藤原俊成女, Shunzei's daughter (1171-1252). It is a long dialogue between three or four highly educated ladies and an eighty-three-year-old nun who spends all the night talking to them.

To effectively present her material, the author takes as model the second chapter of *Genji monogatari*, "Hahakigi" 帚木 (The Broom Tree), in which we witness the famous scene of the discussion about the ideal woman during a long rainy night (*amayo no shinasadame* 雨夜の品定め). This is an important point of the *monogatari*, because Murasaki lays the theoretical foundation for further development of her characters.

Marra (1984, 124) claims that it is not by coincidence that *Mumyōzōshi* was written in a period during which other famous works of literary criticism were composed. During the late Heian period, the aristocrats had been weakened by the emerging military power, so they tended to tenaciously linger on the artistic values of the past. In that particular period *Korai fūteishō*, *Kindai shūka*, *Maigetsushō* 毎月 抄 (Monthly Notes, ca. 1219), and *Go-Toba in gokuden* saw the light.

The influence of such treaties on *Mumyōzōshi*, Marra goes on saying, is evident from beginning to end. The first thing we can notice is the critical approach to prose by means of poetry. In the essay on *Genji monogatari*, for example, the characters are judged on the basis of their poetical skills. The whole exile to Suma scene is judged commenting on the poems Genji exchanges with Murasaki, the woman he loves most, and Hanachirusato, sister of the concubine of Emperor Kiritsubo.

Talking about *Genji monogatari*, at a certain point, the author says:

Perhaps in the future someone will be able to write a novel superior to *Genji Monogatari* in the light of his knowledge of that work. But Murasaki knew only *Utsuho Monogatari*, *Taketori Monogatari*, and *Sumiyoshi Monogatari*, and so to have written such a masterpiece under such conditions cannot be the work of an ordinary being. (Marra 1984, 137)

This is the same idea Teika expresses in his *Meigetsuki* 明月記 (The Record of the Clear Moon, 1180-1235), when, talking about *Genji monogatari*, he says:

"Wild words and fancy phrases", though it may be, this is a work of extraordinary genius. "The more I look up to it, the higher it seems; the more I probe into it, the more solid it seems". How dare anyone discuss it thoughtlessly? (Harper, Shirane 2015, 169)

#### 4 Traces of *Genji monogatari* in *Shinkokinshū*

In *Shinkokinshū*, which contains 1,978 poems (with minor differences depending on the version), there are thirty-three poems that refer to a poem of *Genji monogatari*. If we include in the scope the poems that simply allude to the *monogatari* without clearly taking a particular poem as reference, or those that have some resemblance with a poem of the *monogatari*, the number grows to fifty-one.

These poems inside the anthology are so distributed: 2+2<sup>10</sup> in the books on spring; 6+3 in the book on summer; 6+3 in the books on autumn; 4+2 in the book on winter; 2+2 in the book on grieving; 3+0 in the book on travel; 7+0 in the books on love; 3+6 in the books of miscellaneous. As can be seen, most of these poems are concentrated in the books on four seasons, meanwhile there are none in the gratulatory book and in the last two books that contain religious poems. This fact is quite interesting, as almost all the *honka* (original poems) are not simply descriptive seasonal poems, but strongly express human feelings; so, this kind of re-contextualization seems to be consistent with Teika's concept of a perfect *honkadori*. Among other things, he thought that one of the fundamental prerequisites for a good allusive variation was that the theme or the atmosphere should be renewed in the new poem.

If we look at the poems of *Shinkokinshū* that, in one way or another, refer to *Genji monogatari*, we can see that some chapters of the work are preferred over others. The two that poets of *shinkokinjidai* seem to have appreciated most are "Kiritsubo" 桐壺 (The Paulownia Pavilion) and "Yūgao". We can possibly find the key of this in *Mumyōzōshi*, where, in the passage dealing with *Genji monogatari*, the author says:

"Which of the chapters do you think is the best and the most moving?" "How could there be a chapter superior to 'Kiritsubo'? From its opening words, 'In a certain reign…', down to Genji's comingof-age ceremony, an intimately sad mood permeates this chapter as regards tone and content".

[…] "'Yūgao' is a most touching chapter that arouses our sympathy". (Marra 1984, 137)

With regards to the poets whose poems in *Shinkokinshū* were most inspired by *Genji monogatari*, we can give the following numbers: Fujiwara no Teika: 4+2; Fujiwara no Ietaka 家隆 (1158-1237): 2+1; Fujiwara no Yorizane 頼実 (1155-1225): 1+1; Fujiwara no Yoshitsune: 4+2;

<sup>10</sup> The first digit refers to the poems that present a real *honkadori*; the second one indicates the number of poems that just present some kind of similarity with a poem contained in *Genji monogatari*.

Go-Toba: 3; Shokushi Naishinnō 式子内親王 (1149-1201): 2+2; Shunzei's daughter: 3+0.

It is interesting to notice how all of them were closely related with Go-Toba or his *kadan*. Three of them are particularly interesting in the way they handle poetic materials: they are Shunzei's daughter, Fujiwara no Teika, and Go-Toba himself. I chose not to put Yoshitsune in this number because, even though he proves to be quite fond of *Genji monogatari*, he is not so innovative in treating *honka* taken from Murasaki's masterpiece, at least if we confine our analysis to the poems selected for *Shinkokinshū*.

#### **4.1 Shunzei's Daughter**

The following *waka* is by Shunzei's daughter, and it is contained in the second book on autumn of *Shinkokinshū*.

**515** とふ人もあらしふきそふ秋はきて木の葉にうづむ宿の道芝


This *waka* was composed for the *Sengohyakuban utaawase* and in judging it Fujiwara no Teika affirmed that autumn melancholy was overwhelming and well linked to the idea of passion. Actually, the poem is based on two *honka*. The first is an anonymous poem of the third book of *Shūishū*:

**205** とふ人も今はあらしの山風に人松虫の声ぞかなしき


The second one is from the "Hahakigi" chapter of *Genji monogatari*:

11 The poem number refers to the numeration in Tanaka, Akase 1992 (Author's transl.).

うち払ふ袖も露けきとこなつに嵐吹きそふ秋も来にけり<sup>12</sup>


The verses by Shunzei's daughter are quite traditional in diction, but the way she combines elements from the two *honka* is really impressive. On the one hand, she takes from the poem of *Shūishū* the seasonal element but on the other she hides between the lines the grief of a woman who fears losing her lover.

Another interesting *waka* by Shunzei's daughter is the following one presented in *Shinkokinshū*:



Here, the locution *irokawaru tsuyu* (hue-changing dew) refers to the idea that crystal clear dew turns scarlet once condensed on autumn leaves, and subsequently suggests the word *kōrui* 紅涙 (tears of blood), a term traditionally used to express a heart-breaking sorrow.

Watanabe (1991, 71) thinks that in the verses of Shunzei's daughter we can find an allusion to a *waka* of "Shiigamoto" 椎本 (Beneath the Oak) chapter of *Genji monogatari*:

色かはる袖をばつゆのやどりにてわが身ぞさらにおきどころなき


This poem is one by Ōikimi, who is grieving for her father's death, Hachi no Miya. The new poem takes from the *honka* the image of the mourning dress on which tears of sorrow fall. Needless to say, those tears are represented metaphorically by dewdrops, semantically linked to the expression *uragareteyuku nobe* 枯れてゆく野辺 (the fields wither starting from the top of the trees).

12 Abe et al. 1970-76, 1: 159. 13 Abe et al. 1970-76, 5: 190.

> Ca' Foscari Japanese Studies 19 | 6 **31** Images from the Past: Intertextuality in Japanese Premodern Literature, 17-40

The *waka* by Shunzei's daughter is interesting in many respects. First of all, she introduces the image of the wind, which is quite typical of Uji in *Genji monogatari*, but not in former poetry. Moreover, withered fields make the reader think of the mountain village where Ōikimi lives. But there is more. The expression *okimayoi* をきまよひ contains an emotional element, *mayou* まよふ, reminiscent of Ōikimi's grieve nestled in the locution *okidokoronaki* おきどころなき.

Another detail that we can notice is the inversion in the second line of the words *sode* 袖 and *tsuyu* 露. Furthermore, the line of the *honka*, *irokawaru sode*, which refers to the mourning dress, in the new poem is changed into *irokawaru tsuyu* 色かはる露, a locution that, on alluding once again to the word *kōrui* 紅涙 (tears of blood), evokes the image of the dew condensing on autumn leaves. In so doing, the poetess can recall Ōikimi's sorrow very clearly.

#### **4.2 Fujiwara no Teika**

This capability of adding elements of novelty and giving new shape to classical diction is perfectly in line with Teika's poetics. In this regard, the following *waka* by Teika contained in the summer book of *Shinkokinshū* can be considered emblematic:

**247** 夕暮はいづれの雲のなごりとてはなたち花に風のふくらん


The cloud the poet is alluding to is the one produced by a funeral pyre. At the time people still believed that the smoke produced by the piles of wood incinerating corpses would become clouds.

In this case, Teika took two poems to compose his verses. The first *honka* is a poem from the "Yūgao" chapter of *Genji monogatari*:

見し人の煙を雲とながむれば夕べの空もむつましきかな

*mishi hito no keburi o kumo to nagamureba yūbe no sora mo mutsumashiki kana* When I stare at the clouds that seem to me the smoke from her funeral pyre, oh, I end up cherishing even the dusk sky.<sup>14</sup>

14 Abe et al. 1970-76, 1: 262.

The second poem Teika recalls is one by poetess Sagami included in the book on summer of *Goshūishū* 後拾遺集 (Later Collection of Gleanings, 1086):

**214** さみだれの空なつかしく匂ふかな花たち花に風や吹くらん


In this *waka*, Teika cleverly combines elements taken from the two quoted poems. From the one of *Genji monogatari*, he takes the image of the funeral pyre smoke, meanwhile, from the poem by Sagami he pulls up the scent of blossoming orange trees and, consequently, the summer setting. His highly personal touch is given by enriching the image of the funeral pyre smoke, traditionally linked to a feeling of nostalgic melancholy, with the fragrance of the *hanatachibana* 花た ち花, an evocative flower which reminds of a beloved person or of a lost lover, and that had a symbolic connection to the cuckoo, a bird considered a messenger to the afterlife world.

A synesthetic style is typical of Teika. In this regard, we can quote another poem of his, contained in *Shinkokinshū* as well. This poem, even though does not contain any reference to a particular poem of *Genji monogatari*, perfectly blends with Shunzei's enthusiastic opinions about Murasaki's masterpiece, particularly on the "Hana no en" 花宴 (Under the Cherry Blossoms) chapter: *hana no en no maki wa kotoni en naru mono nari* 花の宴の巻はことにえんなる物なり, "the chapter "Under the Cherry Blossoms" is particularly fascinating" (Shinpen kokka taikan 1983-92, 5: 294). These opinions had a great impact on the poetic production of the poets represented in *Shinkokinshū*, especially for what concerned the image of the blurred moon (*oborozuki* 朧月), an image doomed to become one of the most typical elements of the anthology.

One of the earliest examples of this image in classical poetry is found in a *waka* by Ōe no Chisato 大江千里 (ninth-tenth century), a *kokinjidai* poet, chosen three centuries later for *Shinkokinshū*:

<sup>15</sup> The number refers to the numeration in Kubota, Hirata 1994 (Author's transl.).

**55** てりもせずくもりもはてぬ春の夜のおぼろ月夜にしく物ぞなき


Nothing is more beautiful than the pale moon of a spring night, neither crystal clear nor veiled by clouds.<sup>16</sup>

The *kotobagaki* that precedes these verses reads: "A poem on the topic "pale moon, neither christal clear nor veiled by clouds", a line of a poem about a spring night in Jaling, contained in Bai Juyi's collection". Teika took this Chisato's *waka* as a *honka* for the following poem, chosen for the *Shinkokinshū* as well:

**40** おほぞらは梅のにほひに霞みつゝくもりもはてぬ春の夜の月


Tanaka (2004) points out that this *waka* ends a triplet of poems extremely coherent in terms of topic. In fact all of them focus on a vague spring atmosphere. And maybe, it is not a coincidence that the first of these three poems is Teika's famous poem on the floating bridge of dreams (*Shinkokinshū*, poem no. 38) which contains a strong allusion to *Genji monogatari*.

#### **4.3 Ex-Emperor Go-Toba**

If Teika did often show a strong sensibility for olfactive elements, Go-Toba seems to be more skilful in handling visual elements in his verses, particularly the light.

Let's take as an example the following *waka*, included in the book of mourning poems of *Shinkokinshū*:

<sup>16</sup> The poem is included in *Chisato shū* 千里集 (no. 71), Ōe no Chisato's personal collection also known as *Kudai waka* 句題和歌. In *Genji monogatari*, the last two lines are recited by lady Oborozukiyo herself even though with a slight variation: *niru mono zo naki* instead of *shiku mono zo naki*. See Abe et al. 1970-76, 1: 426.


**803** なき人のかたみの雲やしほるらんゆうべの雨に色はみえねど

This poem is based on the same *honka* of *Genji monogatari* we have already quoted (*mishi hito no*). Here Go-Toba uses once again a technique he is familiar with: he takes from the *honka* some elements but, at the same time, he introduces semantic deviations, creating a sort of short-circuit in the reader's mind. In the *honka* what the poet can't see is the image of the person he once loved, but in Go-Toba's *waka* it is the smoke itself that becomes invisible. In the *honka* this smoke is quite visible because it is silhouetted against the burning sunset sky, but in Go-Toba's poem the background is given by the rainy clouds that make the rising smoke almost indistinguishable. Terashima (2015, 615) points out in passing that Go-Toba wrote this poem in 1206, just one year later the official presentation at court of *Shinkokinshū*, and that was one of the last he wrote taking inspiration from *Genji monogatari*. And this makes us wonder if the reason is to be found in the breaking of the relationship of collaboration with Teika, who was so fond of Murasaki's masterpiece.

Another similar example of Go-Toba technique is the following poem, always contained in *Shinkokinshū*:

**433** 秋の露やたもとにいたく結ぶらんながき夜あかず宿る月かな


Needless to say, the dew on the sleeve is, once again, nothing more than the metaphoric image for the tears shed because of the autumnal melancholy. In most cases it was the poet who spent the night admiring the moon without ever getting bored of her, but Go-Toba inverts the elements and imagines it is the moon that does not get tired to reflect herself in dewdrops. Once again, the *honka* is a poem contained in *Genji monogatari*:

鈴虫の声のかぎりを尽くしても長き夜あかずふる涙かな


This poem is recited by a *nyōbō* 女房 (court lady), which the Emperor Kiritsubo had sent to Genji's grandmother, when weeping she takes leave from the old woman. For this reason, the tears mentioned in the *honka* are related to a tangible and deep sorrow felt in a concrete situation. On the contrary, in Go-Toba's poem, the tears in which the moon shines are just the tears shed, as in the previous example, because of the autumnal melancholy. In spite of that, thanks to the use of allusive variation, the pain of the *nyōbō* echoes through his verses. Moreover, it is useful to note that in *Shinkokinshū* this poem is in between two poems (no. 432 by Princess Shokushi and no. 434 by Minamoto no Michiteru) that contain an allusion to the love dimension, since they make use of the *topos* of the abandoned woman spending the autumn night in solitude.

We can find a different kind of inversion in the following *Shinkokinshū waka*:

**471** 野原より露のゆかりをたづねきてわが衣手に秋風ぞ吹く


In these lines we can find an allusion to a poem Fujitsubo sends to Prince Genji, in which she is actually talking of their illegitimate child:

袖ぬるる露のゆかりと思ふにもなほうとまれぬやまとなでしこ

*sode nururu tsuyu no yukari to omou ni mo nao utomarenu yamato nadeshiko* Even though it is for that Yamato dianthus my sleeves get wet with dew, I do not have the heart to scorn it.<sup>18</sup>

In Go-Toba's *waka*, maternal love becomes a melancholic feeling of frustration depending on the cooling of a lover's passion, the latter being represented, as usual, by the autumn wind. Once again here

17 Abe et al. 1970-76, 1: 108. 18 Abe et al. 1970-76, 1: 402.

> Ca' Foscari Japanese Studies 19 | 6 **36** Images from the Past: Intertextuality in Japanese Premodern Literature, 17-40

dew drops are nothing but the lady's tears. The reason being that in the natural scenery a metaphor hides: the autumn wind is nothing but a man and the dewdrops (or tears) represent a woman.

### 5 Conclusions

The love and respect often shown by the Japanese scholars to the works of the past reached, at the beginning of the Kamakura period, an undoubtedly remarkable level. The poets, feeling the pressing need to breathe new life into their art, looked at the masterpieces of the Heian period. But they did not limit their attention to the first *chokusenshū*  勅撰集 (imperial anthologies), in particular the first three ones, the *sandaishū*, or to the private poetic collections or *utaawase*. Under the vigorous thrust of the thought of Fujiwara no Shunzei, in the poetic circles of the era they began to turn with increasing determination to *Genji monogatari* in order to draw inspiration for their new compositions. This attitude is reflected quite clearly not only in the poetic contests of the time, but also in the eighth *chokusenshū*, the *Shinkokinshū*, considered a faithful mirror of the court poetics of that period.

There were many poets who, tempted by an easy job, turned to *Genji* to create new material. However, just few of them succeeded in managing this material in a truly original way, giving life to deep but enjoyable verses, classic in shape but fresh and original in spirit. In this respect, Fujiwara no Teika, Shunzei's daughter and the Ex-Emperor Go-Toba proved to have extraordinary abilities and incredible literary sensitivity.

#### **Bibliography**


cusing on *Shōji shodo hyakushu*). *Tōdaigaku kokubungaku ronshū*東京大学 国文学論集, 9, 33-47.


#### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# The *Sarashina Diary*: How Are Literary Quotations Woven into Reminiscences?

Itō Moriyuki Gakushūin Women's College, Tōkyō, Japan

**Abstract** This paper aims to clarify the characteristics of the *Sarashina nikki* (1060) whilst paying attention to the issue of intertextuality. However, first, I review the establishment of *kana* literature in the Heian period. After the *Kagerō nikki*, female writers emerged in succession, ushering in the height of Heian female literature. Then, the *Sarashina nikki* appeared at the end of this golden age of *kana* literature. The uniqueness of this diary is clear in that it pursues the metaphysical theme of the meaning of narrative in life. In this paper, I carry out a concrete analysis of the way literary citations are employed in the *Sarashina nikki*.

**Keywords** Kana literature. Classical Chinese literature. Ki no Tsurayuki, Kagerō nikki. Genji monogatari. Sarashina nikki.

**Summary** 1 Introduction. – 2 Until the Birth of *kana* Literature. – 3 Ki no Tsurayuki and Classical Chinese Text. – 4 From the *Kagerō nikki* to the *Sarashina nikki*. – 5 Citations in the *Sarashina nikki*.

### 1 Introduction

This paper is based on the keynote speech "The *Sarashina Diary*: How Are Literary Quotations Woven into Reminiscences?" delivered at the online international symposium *Images from the Past: Intertextuality in Japanese Premodern Literature*, which was held on February 3-5, 2021. My basic argument remains unchanged. However, I have add-

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 41** Submitted 2021-12-14 | Accepted 2022-02-13 | Published 2022-08-30 © 2022 Itō | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/003**

ed an explanation of the background of the symposium and the process of writing this paper to the beginning of this article, as well as reinforcing the argument in the latter part of the paper.

The lecture was given in Japanese, thus this paper was prepared by revising the Japanese lecture draft and then having it translated into English. Of course, this forced me to use English translations when citing classical Japanese works. When preparing a paper such as this, I sometimes feel frustrated that I cannot convey the form of the original Japanese in a foreign language, but unfortunately, there is no way around this problem. (Romanisation to indicate Japanese pronunciation is also used as necessary in this paper, but to limited effect).

Since analytical reading of the text is indispensable to literary studies, the question of whether a work can be analysed using a different language from that of the object of research is an important issue directly related to research content and method. It is a troublesome problem peculiar to literary studies, but the confusion surrounding the language of use has been avoided in Japan so far because Japanese literary research is fundamentally conducted in Japanese. However, the flip side of this has been the phenomenon of low interest in Japanese literary studies conducted in foreign languages, literally giving rise to a 'language barrier'. Now that international cultural interactions are expanding and deepening, it is by no means desirable that this situation persists. In the future, it will be necessary to explore ways to properly evaluate translations of Japanese literature and Japanese literary studies into foreign languages, also within Japan.

If I were to comment on how translated works are evaluated, I would say that the translation of a literary work is an intertextual act in itself. This is because various interpretations are added to the original Japanese text by a translator who freely goes back and forth between different languages. The act of translating between languages with different cultural backgrounds is a superbly creative one, and the translated works produced through such acts possess intrinsic value as new texts. Needless to say, it is through translation that Japanese literature has gained a high reputation as world literature among English-speaking readers, and Japanese literature, once transplanted to different cultural spheres through translation, has found new meanings in accordance with the cultural climate of the recipient countries. Having ascertained these basic facts, I will now consider the issue of intertextuality in classical Japanese literature. Before addressing the main topic, I would like to briefly mention the background to this symposium.

This online international symposium was held in February 2021, only after many twists and turns. Originally, it was supposed to be held in March 2020 with participants attending in person at Ca' Foscari University of Venice. However, that came to coincide with the first global wave of COVID-19, and a decision to cancel the symposium was made just before it was supposed to be held. A little more than six months later, I received a message concerning the online symposium from Ca' Foscari University, and as I reread the manuscript that I had intended to present in Venice, I realised that a Zoom presentation in Japanese to an audience, for most of whom Japanese would not be the native language, produces various adverse conditions, in particular the difficulty of reading the expressions of the audience. As a result, I completely rewrote the manuscript I had prepared in the spring of 2020. I broadened the horizons a little and paid attention to literary-historical viewpoints throughout my discussion, rather than only discussing the work in great detail. Therefore, the first half of this paper gives an overview of the history of diary literature whilst focussing on the relationships between Heian period (794-1185) diary works. In the second half, I add some thoughts from the perspective of intertextuality on the significance of the *Sarashina nikki* 更級 日記 (Sarashina Diary, ca. 1060) being written at the end of the golden age of *kana* literature, that is to say, *after* the *Genji monogatari*  源氏物語 (The Tale of Genji, ca. 1008).

The title of this paper makes clear that I will discuss the *Sarashina nikki* from the viewpoint of intertextuality, but it is because of the above circumstances that I will devote so much time to literary-historical explanations in the first half of the paper. Now, let us begin by reviewing the process of formation of the Japanese written language.

# 2 Until the Birth of *kana* Literature

Although my task given in this paper is to consider the issue of intertextuality in Japanese literature, if we consider the emergence of literature as a linguistic art form and reflect further on the facts of the process of human language acquisition, it is clear that virtually no literary work can be unrelated to intertextuality. As everyone knows, the basis of learning in language acquisition is imitation. Although we are engaged in the complex linguistic activity of considering Japanese literature, we did not ourselves invent that language that is at once our object of analysis and an analytical tool. There already exist spoken and written languages that just happened to emerge in this world, and we have acquired them by repeated learning based on imitation and iteration. The time and energy required to acquire a language thus is by no means small, but the path to mastering a written language is particularly long and complex. To begin with, when learning a language, it takes a considerable amount of time just to acquire all the knowledge of pronunciation, writing, and grammar, after which the acquisition of excellent writing ability requires further training in a different dimension. Memorising letters and grammar does not mean that everyone can write good texts immediately. When there is no model expression in the mind of the person writing, it is almost impossible for them to write a good text by producing something from nothing. To create sophisticated expression, one needs to refine one's writing style by coming in contact with various texts. Therefore, a good writer is always first a good reader, and the issue of intertextuality is a key issue in the discussion of literary works.

The first thing to be noted about the history of the Japanese written language is that when the Japanese people<sup>1</sup> began writing their spoken language, they did not invent a new script for this purpose but borrowed the Chinese characters used for the Chinese language. On the basis of the latest archaeological findings, it is believed that letters were brought to the Japanese archipelago by literate people from the continent at the end of the Yayoi period (tenth century B.C. to third century A.D.). It was then, that residents of the Japanese archipelago first became aware of the existence of writing, and the written language thus introduced from overseas was not isolated as a language used by a small number of immigrants. On the contrary, it continued to grow in importance, and by the time of the establishment of a centralised governmental system, all official documents were written in literary Chinese. In other words, officials from the seventh century onwards developed a dual-language environment in which they routinely conversed in *Yamato kotoba* 大和言葉 (the distinctive language of Japan) whilst writing in literary Chinese. Before long, not only official records but also private diaries were written in literary Chinese, but the Chinese script also began to be used to transcribe the *Yamato kotoba*. For example, the *Man'yōshū* 万葉集 (A Collection of Ten Thousand Leaves, ca. 759) is a long collection of about 4,500 *waka* 和歌 poems, but all of the *Yamato uta* 大和歌 (Japanese poems) are written in Chinese script. Moreover, since the *Man'yōshū* includes many difficult and rare Chinese characters, we know that the literati of the time had mastered an enormous literary Chinese vocabulary, meaning that they had already read many Chinese classics. Many history books were included among the titles of Chinese books recorded in the *Nihonkoku genzaisho mokuroku* 日本 国見在書目録 (Catalogue of Present Books in Japan, ca. 891), and government officials of the time also had a deep knowledge of Chinese history and culture.

Considering this formation of the Japanese written language, it is obvious that Japanese literature was deeply connected with the issue of intertextuality from the very beginning. By the time the inhabit-

<sup>1</sup> The first appearance of the appellation 'Japan' (*Nihon* 日本) in written sources was in the seventh century, but for convenience sake, I will use it in this paper also when talking about the sixth century and earlier.

Ca' Foscari Japanese Studies 19 | 6 **44** Images from the Past: Intertextuality in Japanese Premodern Literature, 41-58

ants of the Japanese archipelago first discovered the existence of the Chinese written language, the corpus of texts in history, literature, and philosophy produced on the Chinese mainland was already huge.

Reflecting on the history of Japanese literature, one will notice that, following the creation of the *Man'yōshū*, the largest anthology of *waka* poems in history, no other compilations of Japanese poetry were produced for a century. Instead, anthologies of classical Chinese poems were compiled one after another. In the first half of the ninth century, Chinese poetry collections such as the *Ryōunshū* 凌雲集 (A Collection from Above the Clouds, ca. 814), the *Bunka shūreishū* 文華秀麗集 (Anthology of Splendid Literary Flowerings, 818), and the *Keikokushū* 経国集 (Collection of National Polity, 827) were compiled in succession. Moreover, all of these Chinese poetry collections were compiled under imperial order and were publicly authorised poetry collections.

It is ironic that the heyday of the creation of classical Chinese poetry should come immediately after the completion of the *Man'yōshū*, but if one opens an old manuscript of the *Man'yōshū* and sees the enormous number of Chinese characters used for the 4,500 *waka* poems, one can also understand that the successive compilation of Chinese poetry collections that followed was a historical inevitability. Borrowing Chinese characters and matching them to Japanese pronunciation to record *Yamato uta* must be considered an inefficient and incomplete method both on the recording and the reading side.<sup>2</sup> It is no wonder that the literati of the time found it more beneficial to compile a collection of their own poems in literary Chinese rather than to compile Japanese poems using such a cumbersome and uncertain script.

That being said, these circumstances that lasted from the eighth century to the ninth century changed completely with the appearance of *kana* 仮名 (the phonetic syllabary developed from abbreviated forms of Chinese characters to transcribe spoken Japanese) in the latter half of the ninth century. Although the exact time of the creation of *kana* is unknown, there is no doubt that it began to spread before the writing of the *Kokin wakashū* 古今和歌集 (Collection of Poems Ancient and Modern, ca. 905), as that work is considered to have initiated *kana* literature.

The *Kokin wakashū* is an early work of *kana* literature and is known as the first imperial anthology of *waka* poems, but it is also distinct in literary history for having two prefaces: the '*kana* preface' and the '*mana* preface' (in literary Chinese). Why did the author of the *kana* preface, Ki no Tsurayuki 紀貫之 (?-ca. 945) write a preface in Japanese at a time when it was common sense to write such a work in Chi-

<sup>2</sup> In the *Man'yōshū*, there are also *waka* poems written in the form of Chinese poems, and it is not easy to decipher them as *waka* poems with 31 syllables.

nese? We can understand the reason for this by comparing the *kana* preface and the *mana* preface. The *kana* preface, making use of the new Japanese syllabary was able to discuss *waka* poems freely and concretely, whereas Ki no Yoshimochi's 紀淑望 (?-919) *mana* preface reveals how restricting it was to discuss *waka* poems in the foreign language of Chinese. The two prefaces are thus contrasting, but we can in fact see some similarities between them as well. That is, both the *kana* preface and the *mana* preface were written with the preface of the *Shijing* 詩経 (The Classic of Poetry, comprising poems dating from eleventh to seventh centuries BC), China's oldest essay on poetry, in mind. Especially in the case of the *kana* preface, a full-fledged theory of poetry is developed to discuss the origin, techniques, and history of *waka*, starting with a basic discussion about 'what *waka* is', but it does this by exploiting terms from the preface of the *Shijing*. In particular, the discussion about poetic technique employs the methodology used to discuss Chinese poetry in the *Shijing* by rearranging and adapting it for *waka* discussion. This approach to writing the *kana* preface proved that it was possible to discuss *waka* from the same vantage point as classical Chinese poetry is discussed in the *Shijing*. It can be understood as an expression of the poet Ki no Tsurayuki's strong determination to position *waka* on the same level as classical Chinese poetry.

Not only did Ki no Tsurayuki gather old and new poems together to create a comprehensive anthology in the *Kokin wakashū*, he also skilfully discussed the history of *waka* poetry and its techniques in the *kana* preface and thus became the pioneer of a new world called *kana* literature in terms of both verse (*waka*) and prose (*waka* theory). Moreover, thirty years after the completion of the *Kokin wakashū*, he created a new style of literature called the *kana* diary when he wrote the *Tosa nikki* 土佐日記 (Tosa Diary, ca. 935).

# 3 Ki no Tsurayuki and Classical Chinese Text

By the time Ki no Tsurayuki had finished his term as provincial governor and departed from Tosa, he was already in his mid-60s. One cannot but be astounded by the flexibility of his spirit as he pioneered a new literary style at such an age. He recorded his fifty-five-day journey from his departure from Tosa in the twelfth month of 934 to his arrival in Kyōto in the second month of the following year using *kana* prose and named his diary the *Tosa nikki*. The *Tosa nikki* is said to be the oldest *kana* diary, but the value of this diary is not simply due to its antiquity (if we are only talking about records of short-term events, there are some *kana* diaries that preceded the *Tosa nikki*). The high praise for the *Tosa nikki* stems not so much from it being the first as from its status as a literary work of excellence. This reputation arises from the interesting content which combines diverse thematic elements in a complex manner, the novelty of the subject matter in recording a trip at sea, the fact that the work was given a title by its author and the fact that the narrator is given a persona. Following the *Tosa nikki* and until the *Towazugatari* とはずがたり (1306- 1313) and the *Takemukigaki* 竹むきが記 (1349) in the medieval period, a wealth of *kana* diaries of high literary evaluation were written in Japan. In the discussion of Japanese literary history, they are commonly identified as belonging to a genre called *nikki bungaku* 日記文学 (diary literature). However, the term 'diary literature', which is often treated as a self-evident concept in Japan, is not so common internationally. Therefore, when foreign researchers run into the problem of the basic conceptual definition of 'diary literature', the example of the *Tosa nikki* emerges as a suitable work to explain the essence of diary literature, both through it being historically the oldest *kana* diary and because it uses elaborate literary devices to develop complex subjects within the loose diary form.

Given its high literary repute, there is actually one thing I find curious about the *Tosa nikki*. It is that the diary opens with the following strange sentence:

男もすなる日記といふものを、女もしてみむとてするなり。

*Otoko mo su naru nikki to iu mono o on'na mo shi te mimu tote suru nari.*

Diaries are written by men, I am told. Nevertheless, I am writing one to see what a woman can do. (Keene 1955, 82)

This short sentence serves as a preface to the *Tosa nikki* by itself. The sentence that follows says, "I departed at 8 p.m. on the 21st day of the twelfth month of that year [934]", and this clear indication of a date and a time for the start of the journey signals the beginning of the main text of the *Tosa nikki*. That is to say, the *Tosa nikki* has no other text than this first sentence that serves as a preface.

The *Tosa nikki* was written more than 1,000 years ago, but most speakers of Japanese today should have no problem understanding the opening sentence in its original form. It has neither difficult words nor complicated grammar. However, many readers would feel that there is something awkward about this sentence. There is a particular problem with the repetition of the verb *su* 'to do'. It appears in different forms three times: *otoko mo su*, *on'na mo shi-te*, and *suru nari*. Repeating the same word in such close proximity necessarily gives an impression of poor expression.<sup>3</sup> As a compiler of the *Kokin wakashū* and the author of its *kana* preface, Ki no Tsurayuki, was the leading literary figure of his day. A Japanese reader would wonder why such a brilliant pioneer of *kana* literature would employ such an odd sentence to stand as a preface for this new style of work he was attempting in his later years.

Would Ki no Tsurayuki write such a poor preface? It is enough to make one worry that not even Tsurayuki could fight off the debilitation of age, but, when one reads the *Tosa nikki*, as a whole, it becomes apparent that such a worry is groundless. From that perspective, it is obvious that the seemingly poor quality of the opening sentence is deliberate. So why did Tsurayuki do such a thing? Recently, Komatsu Hideo has suggested an interesting interpretation for this question (2006, 98-111). The point of Komatsu's theory is that it recognises the implied existence of two double-word pairs: *otoko mo su* (male characters) and *on'na mo shi* (female letters). In light of this, the reason why Ki no Tsurayuki used the unfamiliar expression 'diaries are done by men' instead of 'diaries are written by men' also becomes clear, and the curious expression *otoko mo su naru* also convincingly becomes a vehicle to introduce *on'na mo shi*. In other words, the main point of the Komatsu theory is that the problem in the opening sentence should not be the contrast between 'male' and 'female' but rather between 'male characters = classical Chinese' and 'female characters = *kana*'.<sup>4</sup> Thus, the meaning of this preface becomes the following: whilst men have a custom of keeping diaries in classical Chinese, the author (although a fictive narrator) announces that she/ he will keep this diary in *kana*. From the viewpoint of this paper, it is clear from this interpretation that Ki no Tsurayuki, who had written the *kana* preface of the *Kokin wakashū* in his youth with the preface of the *Shijing* in mind, was now writing the *Tosa nikki* in his later years with the classical Chinese diary in mind. Although clumsy at first glance, the opening sentence also reveals Tsurayuki's stance; he was pioneering this new terrain of *kana* literature on the basis of Chinese literature, which was something that had remained unchanged

<sup>3</sup> This English translation uses different verbs, 'write' and 'do', so it is difficult to appreciate the character of the original sentence. Sonja Arntzen suggested the following literal translation using 'do' for all three verbs by e-mail to me: "It is said that diaries are done by men, but I am doing this one because I thought I would like to see a woman try and do it". As Arntzen herself remarked, "It is not good English". This trial translation, however, reflects the oddness of the original Japanese well.

<sup>4</sup> Komatsu's theory, which discovered the 'secondary *kana* series' of *otokomosu* 男文字 and *on'namoshi* =女文字 in the opening sentence, concludes that the superficial 'primary *kana* series' is no longer meaningful to discuss, including the issue of female persona, but Hanzawa Kan'ichi's (2021, 7-34) scrutinises the opening sentence and re-evaluates the meaning of Tsurayuki's unique concept of trying his hand at a new literary genre: *kana nikki* using a fictive female persona.

throughout his life. Ki no Tsurayuki, who was the first and the greatest in the history of *kana* literature, kept searching for new expressions in *kana* whilst always being conscious of the world of Chinese poetry and literature for comparison.

# 4 From the *Kagerō nikki* to the *Sarashina nikki*

It took forty long years for a new *kana* diary to appear after the *Tosa nikki*. Considering the intellectually stimulating and attractive nature of *Tosa nikki*, it is strange that it took so long, so much so that it makes me want to call this a mysterious blank in Japanese literary history.<sup>5</sup> As men were so accustomed to keeping diaries in classical Chinese, writing *kana* diaries must have been something they resisted, but at any rate, everything changed with the appearance of the *Kagerō nikki* 蜻蛉日記 (The Kagerō Diary, ca. 974) by the woman author Fujiwara no Michitsuna no haha 藤原道綱母 (Fujiwara no Michitsuna's mother, ?-ca. 995).

Since both the length and content of the *Tosa nikki* and the *Kagerō nikki* are totally different except for the common aspect of being in the style of a *kana* diary, it is impossible to recognise any similarities between the two. The *Tosa nikki* records the provincial governor's trip back to the capital in a daily record format, which is the same format as for classical Chinese diaries, and the beginning of the *Tosa nikki* is written in a style that can be directly converted into classical Chinese. However, the *Kagerō nikki* does not contain such elements at all. It seems that Michitsuna no haha, who was not accustomed to writing diaries in classical Chinese, had no interest in the daily record format in the first place. This is evident from the fact that most of the entries in the *Kagerō nikki* are recorded as recollections.

The *Kagerō nikki* is a work written by Michitsuna no haha in an autobiographical style, spanning approximately twenty years from her young womanhood to middle age, a basic structure that is decisively different from the *Tosa nikki*, which is the daily record of a fifty-fiveday journey. In addition, the two had a contrasting influence on literary history. The fact that a male writer, Ki no Tsurayuki, wrote a *kana* diary must have had a great impact on the literati of the time but, as

<sup>5</sup> There are few clues to learn the actual circulation of the *Tosa nikki*. Despite one theory that doubts the secrecy of the original manuscript (Komatsu 2006), it is clear that the diary was read, as we know that images of the *Tosa nikki* were made from descriptions in the *Egyō hōshi shū* 恵慶法師集 (The Poetry Anthology of Monk Egyō, ca. 992) and that poems from the *Tosa nikki* were included in the *Gosen wakashū* 後撰和歌集 (Later Collection of Japanese Poetry, 951). In addition, although I will not discuss it here, there are interesting similarities between the *Sarashina nikki*'s journey to Kyōto along the Tōkaidō Road and the *Tosa nikki*.

mentioned earlier, there was a forty-year gap in the history of diary literature after the completion of the *Tosa nikki*. However, if we look at the forty years after the creation of the *Kagerō nikki*, diaries (or diary-like works) such as the *Makura no soshi* 枕草子 (The Pillow Book, ca. 1000; it has a diary-like character), *Izumi Shikibu nikki* 和泉式部日 記 (The Diary of Izumi Shikibu, ca. 1007), and *Murasaki Shikibu nikki* 紫式部日記 (The Diary of Murasaki Shikibu, ca. 1010) appeared one after another. Moreover, all the authors of those works were women. In other words, the *Kagerō nikki* pioneered the so-called Heian female literature.<sup>6</sup> The *Sarashina nikki* (ca. 1060) appeared just at the end of this golden age of *kana* literature that was led by female writers.

When reviewing the diaries of the Heian period, one notices that they are all unique with no commonalities in terms of subject matter. Nonetheless, the *Kagerō nikki* and the *Sarashina nikki* share an autobiographical character, and the two authors, according to the notes of Fujiwara no Teika 藤原定家 (1162-1241) attached to the *Sarashina nikki*, were related as aunt and niece. However, the contrast between these two autobiographies is striking in terms of both structure and subject matter. This contrast is obvious from their opening passages, so let us compare the two.

Thus the time has passed and there is one in the world who has lived such a vain existence, catching on to neither this nor that. As for her appearance, she can hardly be compared to others, and her intelligence – to say she has some is as good as saying she has none at all – so it is only natural that she has come to such a useless state she thinks again and again; it is just that in the course of living, lying down, getting up, dawn to dusk, when she looks at odds and ends of the old tales – of which there are so many, they are just so much fantasy – that she thinks perhaps if she were to make a record of a life like her own, being really nobody, it might actually be novel, and could even serve to answer, should anyone ask, what is it like, the life of a woman married to a highly placed man, yet the events of the months and years gone by are vague; places where I have just left it at that are indeed many. (Arntzen 1997, 57)

As a girl raised in the back of beyond, even further than the end of the road to the East Country, how rustic and odd I must have been. But, however it was that I first became enthralled with them, once I knew that such things as tales existed in the world, all I could think of over and over was how much I wanted to read them. At leisure times during the day and evening, when I heard my elder

<sup>6</sup> In the early period of *kana* literature, before the *Kagerō nikki*, it was male writers such as Ki no Tsurayuki who were most active.

Ca' Foscari Japanese Studies 19 | 6 **50** Images from the Past: Intertextuality in Japanese Premodern Literature, 41-58

sister and step-mother tell bits and pieces of this or that tale, or talk about what the Shining Genji was like, my desire to read these tales for myself only increased (for how could they recite the tales to my satisfaction from memory alone?) (Arntzen, Itō 2014, 90)

The opening sentences of both works are written in a way that emphasises a self-deprecating attitude, and whilst they both refer to 'tales', they show quite opposite attitudes concerning tale literature. The opening part of the *Kagerō nikki* can be regarded as an independent preface, and after commenting that there is "so much fantasy" in old tales, she explicitly states the reason for writing the *Kagerō nikki*, saying that she will write about her 'own life'. By contrast, the opening section of the *Sarashina nikki* depicts a girl who admires and pursues works of fiction. This extreme contrast at the beginning may puzzle the reader. This is because it does not make sense for the *Sarashina nikki* to depict a naïve girl who is enthralled with tales given the fact that this text was written after the dismissive critique of tales in the *Kagerō nikki*. However, I suggest that the author of the *Sarashina nikki* wrote this introduction with a clear awareness of the perplexity that it would occasion. This is because Takasue no musume 菅原孝標女 (1008-1059), who was born 'after' both the *Kagerō nikki* and the *Genji monogatari*, wrote her text after carefully reading those works and clearly determining her own stance towards them.

The fact that the *Sarashina nikki* begins by explaining the author's upbringing and her early infatuation with tales in a self-deprecating way displays a consciousness of the preface of the *Kagerō nikki*. Thus, she would naturally have been able to predict that any reader familiar with the preface of the *Kagerō nikki* might be puzzled by Takasue no musume's seemingly naïve admiration of tales. This is why Takasue no musume mentions the name "Hikaru Genji" at the beginning of the *Sarashina nikki*, as if to forestall the reader's doubts. Readers who started reading the *Sarashina nikki* and realised the similarity to the preface of the *Kagerō nikki* would have recalled how much the writing of tales had advanced with the appearance of the *Genji monogatari*.

No one took the criticism of the old tales in the preface of the *Kagerō nikki* more seriously than Murasaki Shikibu herself. In the *Genji monogatari*, she succeeded in employing techniques from diary literature to realistically depict the personalities and lives of many characters, and thus raised the level of narrative style in tale literature beyond the reach of the simple critique that tales were merely 'fantasy'. The creation of the *Genji monogatari* marked the emergence of a new kind of fictional narrative that would greatly change the history of Japanese literature. If we take this trend in literary history into account as we re-examine the relationship between the *Kagerō nikki*, the *Genji monogatari*, and the *Sarashina nikki*, then it is the *Kagerō nikki*'s criticism of 'old tales' that is 'simple', and the mention of Hikaru Genji at the beginning of the *Sarashina nikki* seems to declare that she was enamoured not with the 'old tales' that Michitsuna no haha had criticised but with the 'new tales'. Takasue no musume was living in a time very different to that of Michitsuna no haha's when one had no choice but to be exasperated by the 'fantasy' of 'old tales.' By reading the *Sarashina nikki*, we can gain a detailed appreciation of the literary environment in which the author lived.

Moreover, the opening section of the *Sarashina nikki* goes on to describe how the author came to travel from Azuma / the east to the capital as if in answer to her wish to be able to read tales to her heart's content. Interestingly, in this section, the author clearly states her age by saying that it was "the year I turned 13". Amongst contemporary diaries, only the *Sarashina nikki* mentions the author's age. When considering why Takasue no musume did this, we must not overlook the fact that her year of birth coincided with the beginning of the circulation of the *Genji monogatari*.

We know the birth year of Takasue no musume because Fujiwara no Teika notes that the journey to Kyōto took place in 1020, and if we use that to count backwards from Takasue no musume's own statement that she was thirteen at the time, we can know that her birth year was 1008. Although the year of the completion of the *Genji monogatari* cannot be precisely determined, an entry in the *Murasaki Shikibu nikki* tells us that a presentation copy of the *Genji monogatari* was produced in 1008, which (mainly for the sake of convenience) has been taken as an indication of the year of the completion of the *Genji monogatari*. In the *Sarashina nikki*, entries that show influence from the *Murasaki Shikibu nikki* appear frequently, so it is evident that Takasue no musume had carefully read this diary. If so, she must have felt something special when she learned that the *Murasaki Shikibu nikki* records events from her birth year as well as seeing that the nearly completed *Genji monogatari* made its public appearance that year.

We are unable to ascertain the birth year of any of the female writers of the time, including Murasaki Shikibu 紫式部 (973-1014) and Sei Shōnagon 清少納言 (966-1025), because they make no mention of their own ages in their diaries and private *waka* collections. Given that Takasue no musume was the only one to leave a record of her age, it may be surmised that she felt a strong connection between the year of her birth and the writing of the *Genji monogatari*. Her opportunity to live alongside the *Genji monogatari* is something that Takasue no musume came to see as more than pure chance, especially as she grew older. We can glean from many instances in the *Sarashina nikki* that the *Genji monogatari* was deeply meaningful to her throughout her life. For example, one notices in the *Sarashina nikki* a shift in the author's relationship to tale literature in her middle age, and the fact that she neatly records this midlife change is evidence that Takasue no musume was thinking about the relationship between fiction and life throughout her life. That is why she wanted to record as accurately as possible the moment of her encounter with the *Genji monogatari*. She wrote that she departed from Azuma in "the year I turned 13", which was tantamount to stating that she encountered the *Genji monogatari* when she was fourteen years old (which was also fourteen years after the completion of the *Genji monogatari*).

Now, considering the above-mentioned points, it is clear that reference to the *Genji monogatari* at the beginning of the *Sarashina nikki* was not merely part of a superficial depiction of a girl infatuated with tales but that it was an expression of deeper thoughts she had about the relationship between her life and the *Genji monogatari*. Although, unlike the *Kagerō nikki*, the *Sarashina nikki* does not declare the intention of the author directly in its preface, we can observe that a carefully ordered group of entries related to the *Genji monogatari* forms a central axis throughout the whole work and are integral to the way she depicts her life from various perspectives. If we re-examine the opening passage keeping in mind the particular character of this work, we notice that, at the same time the beginning of the *Sarashina nikki* tells us about the narrator's upbringing in Azuma, it also very subtlety implies what kind of work the *Sarashina nikki* is by conjuring up a literary allusion to the *Genji monogatari*.

Below, we will take a closer look at the techniques of literary citation employed at the beginning of the diary.

# 5 Citations in the *Sarashina nikki*

As has been shown so far, the short opening section of the *Sarashina nikki* contains important themes related to the history of diary and tale literature in Japan, but I would like to mention one more point about how that passage is related to the *Genji monogatari*. This is something that previous studies have repeatedly examined, so I would like to merely signal what is salient here on the basis of past research.

The issue is the expression 'As a girl raised in the back of beyond, even further than the end of the road to the East Country'. This is an expression that goes back to an ancient poem by Ki no Tomonori included in the *Kokin waka rokujō* 古今和歌六帖 (Six Quires of the Ancient and Modern Japanese Poetry, ca. 980):

あづま路の道の果てなる常陸帯のかごとばかりも逢ひ見てしがな


Ca' Foscari Japanese Studies 19 | 6 **53** Images from the Past: Intertextuality in Japanese Premodern Literature, 41-58

The *Kokin waka rokujō* is a collection of poems structured as a handbook for writing *waka* and was a must-have collection for poets at the time. Accordingly, most Heian readers would have recognised the allusion to "Azumaji no | michi no hate naru | Hitachi" and therefore connected "the end of the road to the East Country" with the placename Hitachi and assumed that Takasue no musume 菅原孝標女 grew up beyond Hitachi (present-day Ibaraki Prefecture). Even if this *waka* did not come to mind, it is an immutable fact that the country located at 'the end of the Azuma road', meaning the Tōkaidō Road, is Hitachi. Now, something that has previously been debated with regard to this embedded citation of an old poem (a recognised technique known as *hikiuta* 引歌) is that it might contain a geographical contradiction.

Takasue no musume introduces herself as 'someone born beyond the end of the Azuma road', but Hitachi itself was at the end of the Azuma road and the country beyond is Michinoku/Mutsu (the four present-day prefectures of Fukushima, Miyagi, Iwate, and Aomori). Heian readers therefore have thought of Hitachi and Michinoku when they read the opening sentence of the *Sarashina nikki*. However, as mentioned in the note by Fujiwara no Teika, the province where Takasue no musume actually lived was Kazusa (present-day central Chiba Prefecture).

There has been much debate over how to interpret this discrepancy between expression and reality. Some have argued that there is no need to see the opening as problematic because Kazusa was also situated in a faraway location in the central part of the Boso Peninsula, but this interpretation only applies to readers who already knew that Sugawara no Takasue was appointed to Kazusa. Indeed, whilst Takasue no musume might actually have felt that Kazusa Province was an isolated area beyond the Tōkaidō Road, as a sophisticated reader of the *Kagerō nikki* and the *Murasaki Shikibu nikki*, she would also have foreseen that future readers of her diary would not all be her relatives, that they would be of different generations and might have no knowledge of her father's career. A reader unfamiliar with her father's career who reads the expression "beyond the end of the Azuma road" would naturally associate it with Hitachi and Michinoku. Even more than being unnatural, it would be impossible for them to associate that expression with Kazusa.

Takasue no musume compiled *Sarashina nikki* in her mid-fifties when there would have been few people around her who could remember what had happened forty years ago. Therefore, it would have been obvious to her that most readers of the *Sarashina nikki* at her own time and into the future would assume mistakenly that she grew up in 'Michinoku as they read the opening. The fact that she nonetheless wrote it this way suggests that she did not mind being misunderstood. Why did she make such a judgement?

The most convincing interpretation to explain this question is that the image of Ukifune from the *Genji monogatari* is projected onto this opening passage (see Inukai 1969). The *Sarashina nikki* frequently contains the names of the characters in the *Genji monogatari*, but nobody appears more frequently than Ukifune, and she was also the character that Takasue no musume admired the most since her first reading of all volumes of the *Genji monogatari*. And indeed, Ukifune spent her girlhood in Hitachi and Michinoku. In other words, although 'someone born beyond the end of the Azuma road' appears to be a self-deprecating way of emphasising the author's rural upbringing, it in fact seems to implicitly also convey the idea that just like the character Ukifune, Takasue no musume had spent her girlhood in the wild eastern provinces and had therefore experienced the environment that was supposed to have shaped that fictional character. It was an unprecedented device to superimpose a fictional character over the narrator at the beginning of an autobiographical diary, but for Takasue no musume who happened to be born at the same time as the *Genji monogatari* and lived her life alongside it, this device must have been wholly appropriate for the opening of the *Sarashina nikki*. This stylistic choice would have been made after due consideration, but at the same time, it must have been a very natural choice for Takasue no musume.

In this way, right from the beginning, the *Sarashina nikki* incorporates intertextual references to earlier works. Furthermore, it uses the complicated technique of quoting Ki no Tomonori's 紀友則 (850- 904) poem directly and then using the associations evoked in the poem by the place name of Azuma (eastern region) to superimpose the image of Ukifune over the narrative. By using such sophisticated techniques, Takasue no musume constructed a fitting introduction for a work centred around the theme of the relationship between fiction and life. Just by examining the structure of this opening, we readers of a 1,000 years later can imagine what Takasue no musume experienced having been born just after *Genji monogatari*'s completion and living during the golden age of Heian literature when there were so many narrative works (both fiction and diaries) as well as *waka* from which to draw inspiration.

Takasue no musume, who was aware that the words of fictional narratives and poetry were just as meaningful in one's life as relationships with real people and things (and in some cases more so), employed intertextual references not only at the beginning but throughout the *Sarashina nikki*. Accordingly, she repeatedly worked in references to tales as well as recording many *waka* poems exchanged with relatives, friends, and colleagues, and even citing folklore that she heard on her travels. When considering the literary discourse within the *Sarashina nikki* from the viewpoint of the resonance between literary works, we will note that the use of indirect references or implicit suggestions, such as the evocation of Ukifune in the opening, stand out. Of course, specific titles and fictional characters as well as passages of *waka* poems and texts are sometimes directly quoted. Those direct citations and references have important historical value as evidence to trace the reading history of Takasue no musume, but their expressive effect is limited compared to the implicit method. For example, in the case of the opening sentence, the expressive effect of quoting the *waka* poem of Ki no Tomonori is limited to showing that Takasue no musume grew up in Azuma region,<sup>7</sup> but if we look at the opening sentence as a whole, it evokes the image of Ukifune by suggesting the two province names of Hitachi and Michinoku consecutively by alluding to the place name Hitachi as it appears in Ki no Tomonori's poem and adding the words *nao oku tsu kata*. Considering the importance of the author's identification with Ukifune in the *Sarashina nikki*, the effect of this implicit method is enormous.

Now, when we compare the direct citation with the implicit method in this way, we observe that, in the case of direct citation, the literary effect is produced and clearly understood with just the citation of text or the mention of place and proper names. However, for the most part, the literary effect is limited to what is visible on the surface of the text. By contrast, in the case of the implicit method, although it comes with the difficulty of conveying intentions 'implicitly', once that intention is successfully communicated to the reader, various semantic connotations and even whole images of suggested works and characters overlap with each other to create a multi-faceted and rich expressive effect. In the *Sarashina nikki*, we find the implicit method used in a range of contexts.

When the *Sarashina nikki* is analysed from the viewpoint of intertextuality, this diary appears like a treasure house replete with seeds of discussion. The reason for this richness of seeds of discussion is that Takasue no musume used all means at her disposal, whether direct or indirect citation, to continuously enquire into the relationship between literature and life.

For example, the following can be noted regarding the diary's relationship with the *Kagerō nikki*.

As we have already demonstrated, the opening sections of the *Sarashina nikki* and the *Kagerō nikki* both raise the topic of tale literature and express contrasting attitudes towards it. Furthermore, both texts feature pilgrimage-related travel entries frequently amongst the midlife entries. Despite some differences in the circumstances leading to the pilgrimages and the seriousness of their motives,

<sup>7</sup> Though further expressive effects may be expected for readers who have knowledge of the *Hitachi obi* (knowledge of the rituals at Kashima Shrine), I will not discuss it here as it falls outside the scope of this paper.

these entries share the same pilgrimage destinations (Ishiyama Temple, Hase-dera Temple, etc.), and similarities are also obvious in how they record in detail the pilgrimage journeys from Kyōto including depictions of specific natural features along the way. Of course, since the *Sarashina nikki* appeared later, such similarities are because the *Sarashina nikki* was written with the *Kagerō nikki* in mind. Thus, although the *Sarashina nikki* follows the *Kagerō nikki* with regard to the midlife pilgrimages both in terms of attitude to life and entry composition, between the lines, it also suggests a different and particular relationship with the *Kagerō nikki* through a completely different approach. The 'different approach' I mention here is simply 'not writing in the same way as the *Kagerō nikki*.' This is seen specifically in the choice of primary subject matter. The *Kagerō nikki* narrative is centred around the vicissitudes in her marriage and the birth and raising of her son. Although Takasue no musume was a wife and mother herself, she remains almost perfectly silent about this in the *Sarashina nikki*. Since it is inconceivable that a woman could forget to write about these events in her life, Takasue no musume's choice not to include them in her autobiography is not due to carelessness but obviously deliberate. Moreover, I believe that the existence of the *Kagerō nikki* is the only thing that can rationally explain the intention behind this silence.

In the first place, I must admit that, compared to discussing why an author has written an entry that we have in front of us, clarifying the reason why something was not written is a challenge akin to the 'devil's proof' (*probatio diabolica*). Nonetheless, it is reasonable to assume that the topics of marriage, childbirth and childrearing were excluded from the *Sarashina nikki* precisely because they were so central to the *Kagerō nikki*. In other words, lurking in the background of Takasue no musume's choice of topics for her diary was an intention to write a completely different autobiography from the *Kagerō nikki*.

As far as the *Sarashina nikki* is concerned, it can be said that this literary intention was successfully realised throughout. Whilst Michitsuna no haha produced an unprecedented autobiographical *kana* diary that stimulated the growth of women's writing Heian Japan by focusing on depicting her relationship with her husband Fujiwara no Kaneie, Takasue no musume succeeded in producing her unprecedented autobiographical *kana* diary by excluding her husband and son and drawing the *Genji monogatari* to the foreground so that she could build her narrative around the relationship between literature and life. Moreover, given that this relationship is such a deep and complex theme, when we start to discuss the issue of intertextuality in the *Sarashina nikki* we will never run out of 'seeds of discussion,' which is why I shall stop writing here.

#### **Bibliography**


**Images from the Past: Intertextuality in Japanese Premodern Literature** edited by Carolina Negri and Pier Carlo Tommasi

# Hidden Temporalities **Time and Intertextuality in the Medieval Court Diary**  *Utatane*

Simone Müller

Universität Zürich, Schweiz

**Abstract** This paper explores intertextual expressions of temporality by way of *Utatane* (Fitful Slumbers, ca. 1238), a medieval memoir describing the unhappy love affair between a young lady-in-waiting and a courtier of higher standing, and her vain efforts to get over her lover. Through a close reading of the work's beginning and end, it will be demonstrated how intertextual techniques are used in *Utatane* to inscribe the past into the present and to express the protagonist's temporal sensations. Hereby it will be argued that, while allusions underline the protagonist's dissatisfaction with the present and her longing for the past, they may also be read as encrypted expressions of nostalgia for the Heian period's court culture. At the same time, they demonstrate the author's sophistication, an important 'social capital' of court ladies at the time.

**Keywords** Intertextuality. Nun Abutsu. Utatane. Court Diary. Temporality.

**Summary** 1 Introduction. – 2 Plot Structure and its Implications for Intertextual Techniques. – 3 Beginning: Protracted Time and Nostalgia. – 4 End: Insight into Life's Ephemerality and Hopes into the Arts. – 5 Conclusion.

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 59** Submitted 2021-12-14 | Accepted 2022-03-23 | Published 2022-08-30 © 2022 Müller | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/004**

#### 1 Introduction

Japanese literature of the Kamakura period (1185-1333) deals intensively with time-specific issues of memory and nostalgia. This concern stems from the court's fall from power in the thirteenth century, spurring among the aristocratic elite a sense of deterioration and a nostalgic longing for the court's heydays in the preceding Heian period (794-1185).

Intertextual allusions are among the most effective devices to express this retrospective temporal sensation. References to previous texts allow for superimposing different temporal levels and referring to literary ancestors or bygone days. Intertextual techniques are omnipresent in the literary output of medieval Japan, spanning poetry, diaries, and fictitious tales.

In poetry, the technique of *honkadori* 本歌取 (allusive variations) (e.g., Nagafuji 1984, 51-65; Kamens 1997) plays a prominent role. The interweaving of the past by way of allusions to earlier poems conducts an intriguing temporal tension with the concomitant idealisations of the present fleeting moment, and its freezing into a picture of beauty against the backdrop of an ever-changing world, expressed by the aesthetic ideals of *yūgen* 幽玄 (mysterious depth) and *yōen* 妖艶 (ethereal beauty). Medieval court tales, so-called *ōchō monogatari* 王朝物語, show strong similarities to the diction and plot structures of novels from the Heian period, such that they even came to be called *giko monogatari* 擬古物語, 'tales that counterfeit the old style'. The literary ideal alluded to most prominently was the renowned *Genji monogatari* 源氏物語 (The Tale of Genji) from the eleventh century. In war tales, intertextual techniques are skilfully used to produce a common identity in the audience, thus 'construing reality' (Erll, Roggendorf 2002, 80) by way of collective or cultural memory (Assmann [1992] 2013) that is closely connected to space (Assmann [1999] 2006). Collective memory is reinforced by *utamakura* 歌枕 (poetic places, lit. 'poem pillows') – famous sites that appear in classical literature and poetry – and the technique of *michiyuki* 道行 (the road of courtiers into exile or imprisonment) rhetorically underlining the collapse of courtly order (Büyükmavi 2009, 178-83) and the traveller's sadness (Kubukihara 2007, 52). War tales also enforce legitimacy by interweaving the present with various historical precedents from the past, an intertextual characteristic that Huisman (2013) categorises as "equative sequences" characteristic of the so-

This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement no. 741166). I am very grateful for the valuable comments on earlier versions of this article by the peer reviewers and the editors of this volume.

cio-temporally narrative form of European epic literature. In travel diaries (*kikō* 紀行), intertextual devices are used to express temporal feelings associated with memory alike, as the places visited during the journey consist of famous sites (*meisho* 名所) that appear in classical literature and poetry: they thus link the past with the present by way of reminiscence (Plutschow 1981; 1982).

Intertextuality also plays a vital role in female diaries – *joryū nikki* 女流日記 – of the medieval period. Imazeki (1984; 1990; 2004; 2005; 2016) argues that diaries of medieval court ladies such as *Tamakiwaru* たまきはる (Fleeting is Life, 1219) or *Kenreimon'in Ukyō no Daibu shū* 建礼門院右京大夫集 (The Poetic Memoirs of Lady Daibu, ca. 1232) are characterised by a duality of a lost past and an unfulfilled present, a hallmark that also holds for other female diaries such as *Nakatsukasa no Naishi nikki* 中務内侍日記 (Nakatsukasa Naishi no Nikki, ca. 1292) (Laffin 2015, 275-6), or *Towazugatari* とはずがたり (Confessions of Lady Nijō, ca. 1307) (Brazell 1971, 223). Allusions to earlier poems and court tales are frequently used to stress these sensations of longing and loss.

In the following, temporal sensations generated through intertextual techniques by way of the medieval diary *Utatane* うたたね (Fitful Slumbers, ca. 1238) will be explored. In this work that has allegedly been written by Ankamon'in no Shijō 安嘉門院四条 (1226?-1283<sup>1</sup> ), better known under her later name Abutsu-ni 阿仏尼 (Nun Abutsu), intertextual techniques are notably abundant. It describes the unhappy love affair of a young lady-in-waiting and her futile endeavours to get her lover out of her mind. The piece is sometimes read as an early work of Abutsu-ni describing a juvenile liaison of the author (cf. Watanabe 1990, 168; Imazeki 1987, 163).<sup>2</sup> However, *Utatane* may well be a later work of fiction in the tradition of *ōchō monogatari* (Nagasaki 1986; Imazeki 1987; Tabuchi 2000, 83-112, 128-45; Imazeki 2002, 27; Tabuchi 2009, 43-4; see also Wallace 1988a, 397; Laffin 2013, 62; Negri 2021, 102-3).<sup>3</sup> Imazeki Toshiko (1987, 164), for example, sees the work's literary worth particularly in its dramat-

<sup>1</sup> These life data are suggested by Nagasaki Ken (1986, 4). For different theories about Abutsu-ni's life data, see Nagasaki 1986, 1-4.

<sup>2</sup> Abutsu-ni is said to have served Ankamon'in 安嘉門院 in her residence near Kitayama between seventeen and eighteen years of age (Watanabe 1990, 168). During this time, she appears to have been involved with a man of higher standing that became the topic of the work.

<sup>3</sup> Satō Shigeki (1991, 54; see also Laffin 2013, 65) as well argues that the love affair described in *Utatane* is fictional, deriving from Abutsu's imagination. As Laffin (2013, 60) remarks, it is, however, not possible to determine if the work represents the author's own experiences or if the described love affair is merely "a trope for depicting a romance".

ic features (*gekiteki yōso* 劇的要素).<sup>4</sup> The work is one of the most sophisticated literary pieces that medieval Japan brought forth. It excels through its accomplished plot structure and not least also by the virtuosity with which the author uses intertextual devices to express the protagonist's emotionality and worldview.

The following analysis will focus on the narrative's beginning and end, where the protagonist's temporal sensations 'thicken' (Bakhtin 1981, 84). It will be argued that, while intertextual references underline the protagonist's dissatisfaction with the present and her longing for a past in which her emotional experientiality was still fulfilled, they also involve nostalgia for the Heian period's court culture. At the same time, they demonstrate the author's literary education and cultural refinement, or – in Laffin's words – "her command of both poetic and narrative convention" (2013, 67), a significant 'social capital' of court ladies at the time. Lastly, the work's beginning and end exhibit conventional characteristics of female diaries of the Heian and Kamakura periods which may be considered another form of intertextuality.

#### 2 Plot Structure and its Implications for Intertextual Techniques

*Utatane* describes by way of seasons – beginning in spring and ending in winter<sup>5</sup> – the ill-fated love of a young lady-in-waiting for a man of higher standing who, after a short love affair, loses interest in her. Interwoven into this basic structure are two storylines in which the heroine tries to overcome her longing for her lover and to make her life somewhat self-determined: her attempt to become a nun at a temple in Nishiyama, and her journey to her stepfather's country estate in Tōtōmi.<sup>6</sup> Both actions end with the heroine returning to the capital. The story closes with the protagonist coming to terms with her situation and her anxious thoughts about her future.

*Utatane* thus has a clear narrative structure (Nagasaki 1990, 155‑6) in the line of western theories of narrative sequences (Labov, Waletzky 1967; Labov 1972, 362-75; Adam [1991] 2005), consisting of

<sup>4</sup> However, it is also criticised that the work lacks any dramatic feature (*doramasei* ドラマ性) (Watanabe 1990, 169-70).

<sup>5</sup> By way of this seasonal structuring in analogy to the love poems of the *Kokin wakashū* 古今和歌集 (Collection of Poems Ancient and Modern, ca. 905) and other narratives of the Heian period (cf. Walker 1977, 152; Laffin 2013, 63; Negri 2021 103), a 'poetic ideal' (Walker 1977, 182) in the existential sense of ephemerality is created, implying that, like the natural order of the passing seasons, love affairs are, by nature, destined to find an end.

<sup>6</sup> The protagonist's spatial deprivation may also be read as a test of the lover's heart (Imazeki 2002, 25) in order to gain his attention (Laffin 2013, 83; Negri 2021, 105).

an 'initial situation' in which the liaison is thematised in the form of a prologue, a *mise en intrigue* in which the 'complicating situation' unfolds, namely the man's increasing reluctance to visits and indifference, with the central part describing the protagonist's 'actions' to resolve the problem by way of two journeys, a 'resolution' comprising the heroine's return to the capital and a 'final situation' in the form of an epilogue describing the protagonist's insight into the irrefutable transience of all worldly things (Müller 2020, 237). Watanabe (1990, 155-6), by pointing out parallels to conventional plot structures of medieval diaries (1990, 169), divides the work slightly differently into a first part bearing features of a court diary including the prologue, the complicating situation, and the protagonist's refuge to the nunnery, and a second part, showing characteristics of a travel account including her retreat to the residence of her stepfather, her return to the capital and the epilogue. Using this structure, she analyses the intertextual allusions of the work.

*Utatane* contains numerous classical expressions (*kotengo* 古典語) (Wallace 1988a, 393; Watanabe 1989, 139; Shimauchi 1994; Wakabayashi 1998) and extensive allusions to poems and tales of the Heian period (Watanabe 1989, 148-58; Murata 1994; Laffin 2013, 67-78). Watanabe (1989, 148-58) highlights 65 allusions to former works or poems, which most probably does not cover the whole range of intertextual references the work plays with. Also, as Abutsu-ni uses many familiar tropes of court poetry and narratives of the Heian and Kamakura periods (Laffin 2013, 62), it is sometimes difficult to distinguish between deliberate allusions and conventional diction.

Particularly in scenes where the protagonist describes trysts with her lover or her yearnings for him, as well as in the account of her journey to Tōtōmi in the second part of the work, allusions are abundant (Watanabe 1989, 140; Wallace 1988a, 394), while exhibiting quite varied characteristics. In reminiscences of rendezvous with her lover in the work's former part, references to *Genji monogatari* and *Ise monogatari* 伊勢物語 (The Tales of Ise, ca. 950) cumulate (Imazeki 1987, 170; Tabuchi 2000, 84-5). They are a means of presenting the protagonist's lover as a Genji-like figure: a passionate though light-hearted man with many liaisons, supplying the narrative with features of a court tale.<sup>7</sup> When describing the heroine's loneliness and longing for the absent lover, allusions to poems predominate. In the narrative's second part, describing the protagonist's journey to the residence of her stepfather in the province, *utamakura* – poet-

<sup>7</sup> It is even argued that *Utatane* is so closely fashioned on the *Genji monogatari* that she needed to have a copy of the tale beside her when writing her work (Tabuchi 2000, 84; see also Laffin 2013, 60; Negri 2021, 109). For an inquiry of *Utatane*'s absorption of the *Genji*, see, e.g., Tabuchi 2000, 81-128; Laffin 2013, 69-78; Negri 2021, 102-10.

ic places<sup>8</sup> – as well as allusions to Genji's exile to Suma (Watanabe 1990, 181-2) cumulate. It is noteworthy that intertextual techniques are sparse in the description of the first journey to the nunnery – although there are some references to the Ukifune chapters in *Genji* (Tabuchi 2000, 33; Negri 2021, 104-5) –, while they are numerous in the narration of the second journey to the province. This is related to differences in the protagonist's temporal sensations during the two travels: while the first journey is accompanied by feelings of fear, hope, and expectation and is temporally directed into the future, the second journey is characterised by despair and a constant longing for a return to the capital, temporally directed into the past. The author's deliberate interweaving of allusions in the description of this second journey thus underlines the protagonist's longing for bygone days by intertextually paying reference to poetic epigones. The work as a whole therefore adeptly combines specific intertextual techniques of female memoirs, court tales and travel dairies.

# 3 Beginning: Protracted Time and Nostalgia

*Utatane*'s emotive chronography is prototypically expressed at the work's very beginning, where we find the heroine in her boudoir during sleepless nights, melancholically pondering a seemingly ended love affair and with anxious thoughts about her future:

もの思ふことの慰む(1a)にはあらねども、寝ぬ夜(2a)の友と慣らひにける月 (1b/2b) の光待ち出でぬれば、例の妻戸押し開けて、たゞ一人見出だした る、荒れたる庭の秋の露、かこち顔なる虫の音(3)も、物ごとに心を痛まし むるつまととなりければ、心に乱れ落つる涙をおさへて、とばかり来し方行 く先を思ひ続くるに、さもあさましく果無なかりける契り(4)の程を、など、 かくしも思ひ入れけんと、我心のみぞ、返す〴〵恨めしかりける。(Fukuda 1990, 158-9; English trans. by Laffin 2013, 62; emphases added)

It was not as though pondering things (1a) was any comfort, but I spent many sleepless nights (2a) in which I became accustomed to pushing open my door and gazing up, alone, waiting for the light of the moon (1b/2b) to appear. The autumn dew in my overgrown garden and the plaintive cries of the insects (3) invited my sadness. Suppressing the tears that welled up in my heart, I would think for a while about the past and [what would become of me],<sup>9</sup> of how

<sup>8</sup> For an enlisting and discussion of the poetic places the protagonist visits, and her allusions to Genji's exile to Suma, see Laffin 2013, 89-93.

<sup>9</sup> Laffin translates *yuku saki* 行く先 with "where things had gone", but as the expression rather refers to the future, I rephrased.

frail and fleeting those ties (4) to him where, and about why I had become so entranced. My heart was filled with countless regrets.

As the work's opening makes clear, the protagonist's emotionality is forcibly projected on the seasons. By beginning in autumn, the work emphasises the heroine's sense of sadness, ephemerality, and nostalgic state of mind. Seasonal experientiality is emphasised by the moon, the dew, and the insects that all figure in autumnal melancholia (Yamamoto 1989) and "the withering of relationships" (Laffin 2013, 62). The solemnity of the autumn scene is underlined – as Laffin (2013, 64, 208) points out – by the repetition of the consonances 'm' and 'n' in the first line (*mono omou* もの思ふ, *nagusamu* 慰む, *nenu* 寝ぬ, *narainikeru* 慣ひにける), which imitates the sound of the insects associated with sadness.

Conflicts between the protagonist's inner desires and reality as well as the longing for self-determination and liberation are further emphasised by the spatial setting in the socially confined 'chronotope'10 of the boudoir, and by the protagonist gazing yearningly into the garden (Würzbach 2004, 54), underlined by the moonlight, representing a longing for her lover and at the same time for Buddhist enlightenment. Time in the present is perceived as monotonous and repetitive, revealing gendered and closed notions of time (Müller 2020, 242). The protagonist's protracted perception of time is stressed by the use of iteratives such as *rei no* 例の (as usual), *nenu yo* 寝ぬ夜 (sleepless nights) or *kaesu-gaesu* 返す〴〵 (again and again), further underlined by the static verbs *matsu* 待つ (to wait) and *miru* 見る (to see), by attributive expressions such as *are-taru* 荒れたる (desolate) *kakochigao-naru* (doleful), as well as by the repetitious use of the verb *omou* (to think/to long for). The protagonist thus stages as a prototypical 'waiting woman' (*matsu onna* 待つ女), a popular theme in *waka* poetry that expresses the lonely state of mind of a court lady who waits in vain during the nights for a lover whose passion has faded out or who has died.

The protagonist's temporal sensations are further enhanced by a series of intertextual allusions to poems of the Heian and early Kamakura periods (underlined and numbered in the quotation), figuring a nostalgia for the past, when the love affair was still fulfilling.

Watanabe (1990, 171) quotes the following poem by Ōe no Tamemoto 大江為基 (?-?) as the first allusion in *Utatane*. The poem is included in *Shūi wakashū* 拾遺和歌集 (Collection of Gleanings, ca. 1005-07) and *Wakan rōeishū* 和漢朗詠集 (Collection of Japanese and Chinese Poems for Singing, 1013).

<sup>10</sup> For a detailed temporal analysis of *Utatane* in terms of its chronotopes, see Müller 2020.

#### 妻に遅れて侍りける頃、月を見侍りて ながむるに物思事のなぐさむは月は憂き世の外よりや行く

When his wife passed away, he gazed at the moon:


The foreword makes clear that the poem describes Ōe no Tamemoto's grief over the death of his wife and his striving for consolation through the gaze at the moon. The moon represents both the deceased wife and Buddhist enlightenment. It gives the lyrical I the impression of not belonging to this world and thus not being subject to the world's transience. The lyrical I's nostalgic and melancholic emotionality is stressed by the verbs *nagamu* 眺む (to gaze) – a 'pivot word' (*kakekotoba* 掛詞) that homophonically alludes to long rains (*naga-ame* 長雨) –, *mono omou* 物思 (to contemplate), and *ukiyo* 憂き世 (fleeting world). Abutsu-ni in her opening adopts the expressions *mono omou*, *tsuki* 月 (moon), and *nagusamu* なぐさむ (to console) from Ōe no Tamemoto's *waka* in order to stress the protagonist's loneliness. The temporal sensation alluded to in the text is thus nostalgia and an unfulfilled present.

Laffin (2013, 64) also quotes a similar poem by Saigyō 西行 (1118- 1190) from his collection *Sankashū* 山家集 (Collection of a Mountain Hut, ca. 1185), pointing out that the phrases used in *Utatane*'s opening were rather common tropes in the poetry and works of her time, associated with the fleeting nature of this world.

眺るに慰むことはなかれども月をともにて明かすころかな


As Stoneman (2005, 315-16) points out, Saigyō showed a deep affection for the moon, seeing it as a means of enlightenment and even as a friend and support to break the loneliness of his mountain hut. However, as this poem falls under the category of love poems, it may have a double-layered meaning as a 'waiting-love poem' (*matsu koi no uta* 待つ恋の歌) in the tradition of Chinese boudoir poetry (*guiyuanshi*  閨怨詩). Even though the gaze out at the autumn moon on sleepless nights is a typical image in medieval poetry and stories, the similarities between *Utatane*'s opening lines and Saigyō's poem are so striking, that a mere coincidence seems unlikely. Both describe a person spending sleepless nights seeking consolation from their loneliness by the sight of the moon, expressing unfulfilment with the present moment and implied longing for a fulfilled past. Therefore, it may also be the case that Saigyō' alluded to Ōe no Tamemoto, and Abutsuni alluded to Saigyō, or both poems.

The second allusion in *Utatane*'s introduction is, according to Watanabe (1990, 171), a winter poem by Uemon no Suke Michitomo 右 衛門督通具 (?–?) in *Shinkokin wakashū* 新古今和歌集 (New Collection of Poems Ancient and Modern, 1205).

霜むすぶ袖のかたしきうちとけて寝ぬ夜の月のかげぞさむけさ


This poem captures a person's solitary lying awake on a cold winter night. Again, the moon stands for an absent person, stressing the lyrical I's loneliness and longing. The author is male, but the poem may also be a waiting love poem, describing a lonely court lady that vainly waits for a visit from her lover. Abutsu-ni adopts the iterative expression *nenu yo no tsuki* 寝ぬ夜の月 (moon in sleepless nights) in order to highlight the protagonist's protracted awareness of time.

The third allusion in the prologue again refers to a poem by Saigyō from the anthology *Senzai wakashū* 千載和歌集 (Collection of a Thousand Years, 1187), which is also found in Saigyō's private collection *Sankashū* and in the famous *Ogura hyakunin isshu* 小倉 百人一首 (One Hundred Poems by One Hundred People, thirteenth century).

なげゝとて月やはものを思はするかこち顔なる我涙かな


(Katano, Matsuno 1993, 278, poem 929; English transl. by Mostow 1996, 395)

This poem grasps the grief of an unhappy lover who rhetorically delegates the cause of his pain to the moon. Abutsu-ni alludes to the expression *kakochigao nari* かこち顔なり (grieving face), emphasising the protagonist's unfulfilled present.

The last allusion in the work's opening is by the poetess Nijōin no Sanuki 二条院讃岐 (ca. 1141-1217) from *Shinchokusen wakashū* 新勅 撰和歌集 (New Imperial Waka Collection, 1234).

あはれあはれはかなかりけるちぎりかなただうたたねのはるのよのゆめ


The poem portrays a court lady's lament over a love affair experienced as being ephemeral, like a fleeting dream during a slumber (*utatane*) in spring. Abutsu-ni adopts the expression *hakanakarikeru chigiri* はかなかりけるちぎり (ephemeral affair). This allusion is particularly complex, as it spans several temporal layers and alludes to the work's title, *Utatane*. It is also noteworthy that Sanuki's poem, in turn, is an allusive variation, alluding to the famous Chinese poem *Gaotangfu* 高唐賦 by Song Yu 宋玉 (fl. 298-63 BC), in which King Huai of Chū 楚懷王 (r. 328-299 BC) has an erotic dream of a goddess on the mountain Gaotang during a spring-night. This allusion points to the transience of love. More importantly, Sanuki's poem also alludes to the following famous *waka* by Ono no Komachi 野小町 (?-?) included in *Kokin wakashū* 古今和歌集 (Collection of Poems Ancient and Modern, ca. 905):

うたゝねに恋しき人を見てしより夢てふ物は頼みそめてき


*Utatane* means 'unconscious napping in the afternoon', but traditionally refers to dreaming of a lover (Konishi 1986, 206). The poem of Ono no Komachi expresses the lyrical I's hope for the power of dreams to fulfil her longing, and at the same time expresses love's ephemerality: the phrase *utatane no yume* うたたねの夢 (dreams in fitful slumbers) is a metaphor for the transience of life and love. However, it can also be seen as a metaphor for the difficulty of a person tormented by lovesickness to sleep well. The allusion to the poem by Sanuki creates a threefold temporal level in which the transience of the love relationship is emphasised by comparing it to a fleeting dream.

Moreover, the hidden intertextual references to Ono no Komachi, a poet who is said to have had an unhappy love for a man of high standing and to have therefore escaped into a dream world, suggests that *Utatane*'s heroine is reconstructed as a Komachi-like figure. The poem articulates her inability to shape her life in a self-determined way. What remains is the feeling that the love relationship and life itself are nothing but an evanescent and sorrowful dream (Kubo 1989, 71-2).

With the exception of two allusions, all intertextual references in the opening scene come from Heian-period sources. They underline the protagonist's longing for a past in which the love relationship was still fulfilled, a present that is experienced as unfulfilled, and a general sense of transience.

The subjective semantisation of the chronotopical combination 'boudoir-garden' in the opening scene in order to express feelings of longing and *ennui* can typically be found in a number of female court diaries. For comparison, consider the opening of *Izumi Shikibu nikki* 泉式部日記 (The Diary of Izumi Shikibu, ca. 1007), where the protagonist sadly reflects on the recent death of her lover:

夢よりもはかなき世の中を、嘆きわびつつ明かし暮らすほどに、四月十余日 にもなりぬれば、木の下くらがりもてゆく。築土の上の草あおやかなるも、 人はことに目もとどめぬを、あはれとながむるほど、近き透垣のもとに人のけ はひすれば、たれならむと思ふほど、故宮にさぶらひし小舎人童なりけり。 (Fujioka 1994, 17; English transl. by Cranston 1969, 131)

Frailer than a dream had been those mortal ties for which she mourned, passing her days and nights with sighs of melancholy. And now the tenth of the fourth month had come and gone, and the shade beneath the trees grew ever deeper. The fresh green of the grass on the embankment – though most people would hardly give it a glance – somehow aroused an emotional awareness within her, and, as she sat gazing out at it, she noticed a movement at the nearby openwork fence. Who could it be, she wondered, only to discover a moment later that it was the young page who used to wait on the late Prince.

Although the temporal setting is not the same – in *Izumi Shikibu nikki* the prologue is temporally set in spring, as it precedes a new love affair between the protagonist and Prince Atsumichi 敦道親王 (981- 1007), the younger half-brother of her deceased lover Prince Tametaka 為尊親王 (977-1002) for whom she longs in this scene –, the protagonist's gaze into the uncared-for garden during sleepless nights, and her pondering about a lost love and the ephemerality of life, bears striking similarities to the opening of *Utatane*. A similar scene is found in chapter 34 of *Murasaki Shikibu nikki* 紫式部日記 (The Diary of Murasaki Shikibu, ca. 1010), describing the protagonist's visit to her parents' home and her memories of her former life there:

見どころもなきふるさとの木立を見るにも、物むつかしう思みだれて、年ご ろつれ〴〵にながめ明かし暮らしつゝ、花鳥の色をも音をも、春秋に行か ふ空のけしき、月の影、霜・雪を見て、その時来にけりとばかり思ひ分きつ ゝ、いかにやいかにとばかり、行く末の心ぼそさはやる方なき物から、[…]。 (Itō 1989, 285-6)

As I looked at the unattractive grove at home, I felt depressed and confused. [After my husband's death] I had spent the years by idly gazing out from morning to evening, watching the colours and sounds of the flowers and birds, the scenery of the sky that moved from spring to autumn, the moonlight, the frost and the snow, doing little more than registering that now their time had come, and I was unable to cast off worries about my future – what, yes what would become of me […]?<sup>11</sup>

Here, too, the melancholy gaze out into the garden awakens memories of the past, when the days idly followed one another, the protagonist's awareness of the passing of time, a feeling of *ennui*, as well as anxious thoughts about the future.

These analogies show that the gaze out into the garden is a conventional trope in female diaries implying a protracted awareness of time, feelings of longing, melancholy and anxiety. It can be concluded that the beginning of *Utatane* reproduces this specific trope intertextually to express the protagonist's sense of time by underlying it with allusions to earlier texts.

### 4 End: Insight into Life's Ephemerality and Hopes into the Arts

*Utatane* ends with the heroine returning to the capital and accepting her fate, but also – again – with anxious thoughts about her future. The spatial return to the starting point in the capital in the static chronotope of the boudoir stands for the protagonist's resignation and refers to 'closed' perceptions of time: time proves to be determined in the end (Morson 1994). At the same time, the heroine gains a deep insight into the transience of the world through her experiences and travels and shows a development or 'transformation' (Laffin 2013, 85) that bears characteristics of a coming-of-age novel:

<sup>11</sup> For the translations of this scene I drew on previous translations by Omori, Doi (1935, 110-11) and Bowring (1996, 95), but in order to render it more literal, I provided my own translation.

暮れ果つる程に行き着きたれば、思ひなしにや、こゝもかしこも猶荒れま さりたる心地して所〴〵漏り濡れたるさまなど、何に心留まるべくもあらぬ を見やるも、いと離れま憂きあばら屋の軒ならんと、そゞろに見るもあわれ なり。[…]

その後は、身を浮草にあくがれし心も、こり果てぬるにや、つく〴〵とかゝる 蓬が杣に朽ち果つべき契りこそはと、身をも世をも思ひ鎮むれど、従はぬ心 地なれば、又なり行かん果ていかが。

われよりは久しかるべき跡なれどしのばぬ人はあはれとも見じ

We arrived home at sunset. It was probably my imagination but everything seemed even more run down; here and there the house was damp and leaky. There was nothing really appealing about the place, but still, I felt it would be difficult to part with this poor ramshackle house, and I was moved as I gazed upon it […].

After this, perhaps because I was done hoping to be "beckoned" somewhere like duckweed, I utterly consoled myself and the world by the thought that it was probably my lot to rot away in this dilapidated, weed-ridden home. But my heart was not one to follow such things easily and I wondered what would come of me.


(Fukuda 1990, 177; English transl. by Wallace 1988b, 415 and Laffin 2013, 94<sup>12</sup>)

The final poem, in which the heroine 'entrusts her feelings' (Terashima 1992, 117) to a quotation from the anthology *Shokugosen wakashū* 続後撰和歌集 (Anthology of New Pickings, 1251, poem 1140), expresses the protagonist's fears for the future and implies her secret wish for her work to stand the test of time, thereby projecting her longing into art.<sup>13</sup>

The closing scene contains only one allusion, interestingly again to a poem by Ono no Komachi included in *Kokin wakashū*.

<sup>12</sup> In order to render the translation more literal, I slightly reformulated the translations of Wallace and Laffin.

<sup>13</sup> In the final poem, *hito* 人 (person) refers to the unfaithful lover, but it can also be read as an appeal to future readers (cf. Imazeki 1977, 212 cited from Laffin 2013, 95).


(Kojima, Arai 1989, 282, poem 938; English transl. by McCullough 1985, 206)

Ono no Komachi composed this *waka* in response to a poem by Funya no Yasuhide 文屋康秀 (?–?), in which he invites her to accompany him on an inspection of Mikawa county. Ono no Komachi deliberately answers him by expressing her response through natural phenomena. Abutsu-ni picks up the expression *mi o ukikusa* 身を浮草 (body like duckweed) from Komachi's poem to emphasise the protagonist's instability in the world and uncertainty about her future life. In contrast to the beginning of the work, in which her thoughts are rather directed into the 'past', the ending is filled with thoughts about the 'future' and expresses the protagonist's weak economic situation and her desire to gain stability instead of floating like 'duckweed'.

Analogous to its beginning, the ending of *Utatane* also bears similarities to the end of other female diaries. Like many other court ladies, the protagonist reveals her writing intention by pondering "on whether her written words will serve as a record to others" (Laffin 2013, 94). For example *Towazugatari*, written by Gofukakusain Nijō 後深草院二条 (1258-1307?), a work that describes the author's life at the court of Emperor Go-Fukakusa 後深草天皇 (1243-1304, r. 1246- 1260) and her later pilgrimages as a nun, closes with the following words:

深草の御かどは、御隠れの後、かこつべき御事どもも跡絶え果てたる心地 して侍しに、[…]さても、宿願の行く末いかゞなりゆかんとおぼつかなく、年 月の心の信も、さすがむなしかれあずやと思ひつゞけて、身のありさまを一 人思ひゐたるも飽かずおぼえ侍上、修行の心ざしも、西行が修行の式、う ら山しくおぼえて社思ひ立ちしかば、その思ひをむなしくなさじばかりに、 か様のいたづら事をつゞけ置き侍こそ。後の形見とまでは、おぼえ侍ぬ。 (Misumi 1994, 248-9; English transl. by Brazell 1973, 264)

After Go-Fukakusa's death I had felt as though there was no one with whom I could share my feelings. […] Now I am anxious about the outcome of my long-cherished desire, and I worry lest the faith I have kept these many years prove fruitless. When I attempted to live in lonely seclusion, I felt dissatisfied and set out on pilgrimages modeled after those of Saigyō, whom I have always admired and wanted to emulate. That all my dreams might not prove empty, I have been writing this useless account – though I doubt it will long survive me.

In this epilogue, very similarly to *Utatane*, the protagonist unconfidently expresses her wishes that her work will survive times, providing her life with some meaning. Here, too, the author thus concludes her work by revealing her writing intention and directing her thoughts into the 'future' (Misumi 1990, 58).<sup>14</sup>

A similar closing is found in *Takemukigaki* 竹むきが記 (From the Bamboo-View Pavilion, ca. 1349) by Hino Nako (Meishi) 日野名子 (?-1358), which is considered the last major work of female court diaries. The work covers the author's life at the court of emperor Kōgon 光嚴天皇 (1313-1364, r. 1331-1333) during the turbulent times of the Northern and Southern Courts period (1336-1392), anxious thoughts about her future and her desire to renounce the world. The work closes with poems in which the protagonist, now having become a nun, reflects about the transitoriness of the 'fleeting, agonising world' (*ukiyo* 憂世) and whether her name and work will endure the ages, thus, in analogy to *Towazugatari* and *Utatane*, creating a contrast between the ephemerality of historical time and the timelessness of art.

なき跡にうき名やとめんかき捨つる浦の藻屑の散り残りなば


(Iwasa 1990, 344; English transl. by Tyler 2016, 168)

A comparison of various diaries written by court ladies thus reveals conventionalised endings: in all the works presented here, the author's endeavour to create a personal legacy in the form of a work of art for future generations is expressed. This raises the question of the purpose for this kind of closure: why was it so crucial for court ladies to create an artistic legacy? One explanation could be that cultural refinement was essential for ladies-in-waiting, securing their social and economic positions at court. The declaration of intention to have created an outstanding work of art that can outlast 'time' and be read in the 'future' therefore indicates – even if this intention is often rhetorically uncertain and modestly formulated – the mastery of literature by its authors and is intended to increase their social capital. Unlike Sinitic diaries (*kanbun nikki* 漢文日記), which aim to demonstrate their author's competence in a particular court office and pass it on to their descendants in order to secure the family's position at court (Matsuzono, Kondō 2017, 4), in the case of the court ladies, the cultural asset of art itself seems to have served as eco-

<sup>14</sup> For a detailed investigation of the work's writing intention, see Mismui 1990.

nomic means. The importance of literary education for a court lady's career and economic stability is shown by the fact that in her late twenties, Abutsu-ni was introduced to the poet Fujiwara no Tameie 藤原為家 (1198-1275) – the son of the famous poet Fujiwara no Teika 藤原定家 (1162-1241) – as a scholar and poet. She acted as Tameie's copyist and assistant and eventually became his first wife, which in turn influenced her own career as a professional poet and scholar (Laffin 2013, 96-7). On the side of scholars who argue that *Utatane* was written when Abutsu-ni was already acquainted with Fujiwara no Tameie, it is assumed that Abutsu-ni wrote *Utatane* to prove her literary sophistication to Tameie (Matsumoto 1983, 138, cited from Negri 2021, 103).

#### 5 Conclusion

Abutsu-ni's use of conventional tropes of Heian and Kamakura-era poetry and narratives, as well as a wealth of intertextual allusions in her work *Utatane*, fulfil a complex function closely linked to temporality.

Intradiegetically, they underline the core messages of the work and the underlying temporal sensations. The core theme is the protagonist's conflict between her life design and social conventional and gender-specific life schemes at the time, illustrated by an unhappy love affair. The temporal sensations underlying this conflict are nostalgia (past), *ennui* (present), and fear (future). This gives the protagonist the feeling that life and love are nothing but a fleeting dream. The central chronotope of the work – the boudoir where the narrative begins and ends – represents the social confinement of the protagonist. It is therefore characterised by 'cyclical everyday time' (Bakhtin 1981, 247) and monotonous repetition and provides information about gender-specific and closed notions of time. *Utatane* thus prototypically expresses the female court's perception of time in the Kamakura period.

The core message of the work becomes evident in the prologue and epilogue, which, in addition to allusions to earlier poems, display conventional plot features of female court diaries, which usually begin with meditations on life and end with a kind of review of the protagonist's feelings. The heroine's view out at the garden from her boudoir at the work's very beginning – a typical scene in the literary output of court ladies – evokes memories of the past and produces a sense of protracted time. The protagonist's revelation of her fears concerning her future and the shift of her hopes into the arts in the epilogue discloses economic concerns regarding the 'future' and the importance of literary mastery – an essential social capital for a court ladies' career. This also holds true for the implementation of allusions as such, which are a means to demonstrate the author's literary education and sophistication: As Negri (2021, 109) points out, the work's numerous citations reveal Abutsu-ni's extraordinary poetic skill and thorough knowledge of literary epigones, both necessary prerequisites for becoming a full-fledged member of an illustrious family of poets such as that to which her husband Fujiwara no Tameie belonged.<sup>15</sup> Cultural capital thus was an essential means to enhance a court lady's prestige and thus her social capital, which eventually became a way of economic capital, fostering a court lady's chance to make a career at court or get married to a man of high standing. The protagonist's secret wish that her lover – and readers in general – will read her work in the future encrypts the novel's pragmatic purpose.

On a third level, intertextual techniques inscribe the past into the present. As Nagafuji Yasushi (1984, 51-65) has pointed out, allusions are often used in medieval literature as a means to evoke nostalgia for the Heian period. *Utatane* can thus be read as a coded expression of nostalgia for the court culture of the Heian period in the tradition of 'counterfeit tales' (*giko mongatari* 擬古物語), endowing the work with a political message: Lost love that is compared to a short dream during a fitful slumber can be read as a political-erotic allegory for the court aristocracy's loss of power in medieval Japan.

This gives the narrative a threefold meaning: one that reveals a young lady's longing for a past in which her love was still fulfilled, one that reveals a medieval court lady's longing for a past in which the power of the court and its cultural sophistication was still in bloom, and one in which a court lady flaunts her literary sophistication to improve her economic situation. The work's multiple intertextual techniques serve a vital function on all three levels.

Since the technique of allusive variation is a common feature of classical Japanese literature, which flourished in the Middle Ages, *Utatane*'s use of allusions does not seem notably different from other works of the period. Nevertheless, Abutsu-ni is remarkably virtuosic in her appropriation of earlier poems, stories, and diaries. As Negri (2021, 109) notes, the numerous quotations are an 'homage to the culture of a recent past that defined and supported the authority of the aristocratic class in a period of great political and social transformations', and at the same time, a testimony of the author's participation at the cultural discourses of her time. Abutsu-ni thus makes excellent use of intertextual techniques to encode her key messages, which are closely linked to temporality: longing for a lost past, dissatisfaction with the present, and plans for the future.

<sup>15</sup> Negri (2021, 110; see also Ratcliff 2009, 34-9) also points out that *Utatane*, along with her manual *Abutsu no fumi*, as well as other texts written by women, may have aimed to impart literary knowledge with other women at the time and thereafter.

#### **Bibliography**


Bowring, R. (ed.) (1996). *The Diary of Lady Murasaki*. London: Penguin Classics.


Büyükmavi, M.U. (2009). "Raum und Erinnerung – Zur Konstruktion kultureller Erinnerung im *Heike monogatari*". Wittkamp, R.F. (Hrsg.), *Erinnerungsgeflechte: Text, Bild, Stimme, Körper – Medien des kulturellen Gedächtnisses im vormodernen Japan*. München: Iudicium, 166-97.


Literature. The Case of *The Diary of Kenrei Mon'in Ukyō no Daibu*). *Nihon bungaku* 日本文学, 33(12), 1-10.


Kojima Yoshio 小島吉雄; Kazamaki Keijirō 風巻景次郎 (eds) (1967). *Sankanshū*, *Kinkai wakashū* 山家集・金槐和歌集 (Collection of a Mountain Hut, Collection of Gold Bars). Tōkyō: Iwanami Shoten. Nihon koten bungaku taikei 29.

Komachiya Teruhiko 小町谷照彦 (ed.) (1990). *Shūi wakashū* 拾遺和歌集 (Collection of Gleanings). Tōkyō: Iwanami Shoten. Shin Nihon koten bungaku taikei 7.


Kubukihara, R. (2007). "Various Aspects of Diary and Travel Literature" (English transl. by E. Sarra), in Sekine, E. (eds), "Aspects of Classical Japanese Travel Writing", *Review of Japanese Culture and Society*, 19, 30-56.

Labov, W. (1972). "The Transformation of Experience in Narrative Syntax". Labov, W., *Language in the Inner City. Studies in Black English Vernacular*. Philadelphia: University of Pennsylvania Press, 354-96.


Murata Noriko 村田紀子 (1994). "*Utatane* no koten sesshu no hōhō. Kanmatsu no jukkai o yomitoku" 『うたたね』の古典摂取の方法 – 巻末の述懐を読み解く (The Method of Adoptions of the Classics in Fitful Slumbers – Analysis of Recollections at the Book's End). *Kokubungaku kenkyū* 国文学研究, 112, 60-8.

Nagafuji Yasushi 永藤靖 (1984). *Chūsei Nihon bungaku to jikan ishiki* 中世日本 文学と時間意識 (Time Consciousness in Medieval Japanese Literature). Tōkyō: Miraisha.

Nagasaki Ken 長崎建 (1986). "Abutsu-ni *Utatane* kō" 阿仏尼『うたたね』考 (Considerations on Abutsu-ni's *Fitful Slumbers*). *Chūō daigaku bungakubu kiyō* 中央大学文学部紀要, 118, 1-24.

Nagasaki Ken (1990). "*Utatane* no kōsō to shippitsu ito" 「うたたね」の構想と執 筆意図 (Plot and Writing Intention of *Fitful Slumbers*). Imai et al. 1990, 151-67.

Negri, C. (2021). *Tra corte, casa e monastero. La vita di una donna nel Giappone del Medioevo*. Venezia: Edizioni Ca'Foscari. Ca' Foscari Japanese Studies 15. Arts and Literature 5. http://doi.org/10.30687/978-88-6969-536-0.

Omori, A.S.; Doi, K. (transl.) (1935). *Diaries of Court Ladies in Old Japan*. London: Routledge.


Ratcliff, C. (2009). "Telling Secrets: *Mumyōzōshi*, Abutsu, and the Transmission of Literary Expertise by Women". *Jinbun kenkyū* 人文研究, 169, 25-50.


Satō Shigeki (1991). "*Utatane* ni okeru kyokō no mondai: tsuki no byōsha to koigokoro" 『うたゝね』における虚構の問題 – 月の描写と恋心 (The Problem of Fictionality in *Fitful Slumbers* – Descriptions of the Moon and Love). *Hiroshima jogakuin daigaku Nihon bungaku* 広島女学院大学日本文学, 1, 45-56.

Shimauchi Keiji 島内景二 (1994). "*Utatane* no hyōgen yōshiki. Chūsei ni yomigaetta *Genji monogatari*" 「うたたね」の表現様式 – 中世に蘇った「源氏物語」 (Patterns of Expression in *Fitful Slumbers*. A Medieval Revival of *The Tale of Genji*). *Denki tsūshin daigaku kiyō* 電気通信大学紀要, 7(1), 103-32.

Shinpen Kokka Taikan Henshū Iinkai 編国歌大観編集委員会 (ed.) (1983). *Shinchokusen wakashū* 新勅撰和歌集 (New Imperial Waka Collection). Tōkyō: Kadokawa Shoten, 259-88. Shinpen kokka taikan 1.

Stoneman, J. (2005). *Constructing Saigyō: Poetry, Biography, and Medieval Reception* [PhD Dissertation]. New York: Columbia University.


#### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# Intertextual Intersections in Late Edo-Period Prose Literature

#### Cristian Pallone

Università degli Studi di Bergamo, Italia

**Abstract** This article aims to reconsider the intertextual relations surrounding *Tsubosumire* (The Violet in Pot), published in 1794 and reprinted at different times with different formats and titles. On the one hand, it seeks to stimulate a discussion on how intertextuality is to be intended when studying Edo texts, and to what extent *shukō* as a working tool for literary criticism is effective. On the other hand, while attempting an interpretive reading of *Tsubosumire*'s intertextual intersections, it traces the yet little explored links between *sharebon* narrative material and other literary manifestations of the late eighteenth century.

**Keywords** Intertextuality. Edo literature. Fantastic literature. Yomihon. Sharebon.

**Summary** 1 *Tsubosumire* as a Place of Intersections. – 2 *Tsubosumire* and 'Source Studies'. – 3 Intertextual Reading and 'Source Studies'. – 4 Reading *Tsubosumire*'s Intertextuality.

#### 1 *Tsubosumire* as a Place of Intersections

The objective of this paper is to analyse *Tsubosumire* 壺菫 (The Violet in Pot, 1794), which is acknowledged as being one of the first attempts to write a fantastic long-story or novel in Edo 江戸, from the standpoint of so-called intertextuality. Nowadays *Tsubosumire* is a little known text despite having been published under different titles at least three times in the thirty years spanning 1794 and 1823 (Kigoshi 2008, 28). Therefore, we can assume that it attracted a fairly consistent readership in the late Tokugawa era (1603- 1867) era before being relatively forgotten. The first edition was pub-

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 81** Submitted 2021-12-13 | Accepted 2022-03-21 | Published 2022-08-30 © 2022 Pallone | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/005**

lished as a *hanshibon* 半紙本 (half-folio book) text in five volumes by Edo Bakurochō sanchōme 江戸馬喰町三丁目 publisher, Wakabayashi Seibei 若林清兵衛, and notwithstanding the colophon announcing a sequel, this would appear to never have been issued. In the following year, it was reprinted as a three-city edition (in the colophon, the names of Kyōto publisher Yamamoto Heizaemon 山本平左衛門 and Ōsaka publisher Shibukawa Yozaemon 渋川与左衛門 follow Wakabayashi's name, cf. Ōtaka, Kondō 2000, 690). In 1797, it was published as a *chūhon* 中本 (middle size book) text in five volumes under the new title of *Kaidan konogoro zōshi* 怪談[コノゴロ]草紙 (A Booklet of Ghost Stories of Recent Times) (Kurashima 1992, 470-2). Later, a new version of this text was released in the *hanshibon* format under the title of *Kidan nasake no futasujimichi* 奇談情之二筋道 (Fork in the Road of Love: A Strange Story, 1823), with specially carved illustrations, and although considered a rare book, the slight differences amongst the remaining copies allow us to infer that different impressions of this version existed (Kigoshi 2008, 29-30). No information remains of who the author is: some propose his identity lies in his pen name of Minamoto Atsumoto 源温故 (as per the signature at the end of the preface in literary Sinitic), others suggest he could be a certain Aritomi 蟻登美, whose signature is on the Japanese preface. Nor can we ascertain the identity of the original illustrator, Shōjusai Tsunekage 松樹斎常蔭.

Looking briefly at its content, the first four volumes of the 1794 version recount the story of a young samurai, Heijirō 平次郎, who is intimately involved with Katsuura 勝浦, a courtesan, whom he promised to ransom. However, later he is told that Ofune お舟, the daughter of his lord, has fallen in love with him and now being critically ill, the only way he can save her life is to marry her. In the last chapter of the novel the *kaidan* 怪談 (ghost story) element is finally exploited. Notwithstanding Heijirō's reluctance, his comrades go to the brothel to talk to the courtesan and ask her to forget her lover. She loses her mind and curses Ofune, later appearing to Heijirō's comrades as a flying sphere of fire. After the courtesan's suicide, Heijirō's young wife is possessed by the spirit of Katsuura and falls ill and dies. Heijirō then decides to take vows.

*Tsubosumire* stands at many crossroads in Tokugawa literature: it contributes to blurring the boundaries of the *chūhon* and *hanshibon* formats, as well as the so-called *ninjōbon-sharebon* 人情本・洒落 本 (sentimental books-witty books) and *yomihon* 読本 (reading books) 'genres', and this can be demonstrated by the fact that this same text was republished in various formats under a variety of titles. The different titles present it respectively as a homage to classical Japanese culture (1794 edition), a true account of a ghostly story (1797 edition) and a summary of the strange developments of a fatal love liaison (1823 edition). These different editions certainly served the publishing house agenda in matching the exterior characteristics of the books with the publishing trends of the moment irrespective of what the actual content was, whilst at the same time offering readers various clues as to what to expect and the elements in the text on which they should focus. Along with the title, other elements of the paratext also change accordingly: 1794's prefaces are not included in subsequent extant versions, completely obliterating the explicit classical reminiscences from the book; as Kigoshi (2008, 31) also highlights, the 1823 version presents the first two lines of a Chinese poetic sequence by Tang (618-907) female poet Liu Yuan 劉媛 (dates unknown), entitled "Lament at the Changmen Palace" (*Changmen yuan* 長門怨, included in the 801st *juan* 巻 of the *Complete Collection of Tang Poetry* [*Quan Tang shi* 全唐詩]). The poem intones the sorrow of a woman abandoned by her lover and its lexicon is highly reminiscent of classical Chinese *topoi* on love, resentment and sorrow, such as the flood of tears that never ends and the image of the rain on the parasol tree, which brings to mind a famous verse from Bai Juyi's 白居易 (772-846) *Changhenge* 長恨歌 (Song of Everlasting Sorrow). The substitution of the allusions to *Genji* 源氏 and his love affairs and the ruminations on ghosts, in favour of the Chinese discourse on female abandonment and sorrow may just enhance the effect of styling the text as a *hon'an* 翻案, a transposition of a Chinese love story into a Japanese environment or, as Kigoshi (2008, 31) suggests, it may strengthen the focus on the sentimental theme, harmonising the title with the requests of the book market.

# 2 *Tsubosumire* and 'Source Studies'

The various titles and presentations of *Tsubosumire* and its consequent adaptability to the book market hint that deciphering the variety of themes and tones the novel is clothed in has been a crucial endeavour to understand how it works as a literary text. *Tsubosumire* can be defined as a place of intertextual intersections and, in fact, researches conducted in Japan mainly focus on the relationship the novel bears to other works of literature. Scholars have identified a wide number of sources that could have been drawn on by the author. To make a long story short, Ōhashi (2018) recently identified a *kobon* 小本 (small size book), *Rinbeki yawa* 隣壁夜話 (Night Talk from the Next Chamber, 1780). The first of these five short tales precisely resembles the last part of the plot of *Tsubosumire*, thus demonstrating that the author may have borrowed or even copied the plot from this source. Previous studies, however, found other similarities with other possible sources: Yokoyama (1974, ch. 1, § 1.2) links it with a Buddhist-themed story published in Kyōto, *Kaidan kien* 怪談奇縁 (Strange Liaisons. A Ghost Story, 1785); Miura (2002) later problematizes this link, focusing more on Chinese and classical sources, such as the *chuanqi* 傳奇 (transmission of the strange) *Huo Xiaoyu zhuan* 霍小玉傳 (Life of Huo Xiaoyu, ninth century) and *Genji monogatari* 源氏物語 (The Tale of Genji, early eleventh century). After all, the preface of the text pays an explicit homage to "Murasaki's *monogatari*" (*Murasaki no monogatari* 紫のものかたり, Ōtaka, Kondō 2000, 338). A relation has also been found by Ōtaka (in Otaka, Kondo 2000) with *Ugetsu monogatari* 雨月物語 (Tales of Rain and Moonlight, 1776) – also Miura (2002) supports the idea that *Ugetsu monogatari*, the tale "Kibitsu no kama" 吉備津の釜 (The Cauldron of Kibitsu) in particular, may have influenced *Tsubosumire* in adopting narrative elements from *Huo Xiaoyu zhuan* –<sup>1</sup> and with another *kobon*, entitled *Geisha yobukodori*  妓者呼子鳥 (Geisha and the Lamenting Cuckoo, 1778).

To sum up, in the last fifty years several texts have been considered by Japanese scholars as *shutten* 出典, *tenkyo* 典拠, or *genkyo* 源拠 (source) or *shukō* 趣向 (inspiration) for *Tsubosumire*. In these analyses, intertextuality is intended as a verbal and narrow relation between historical artefacts, namely as a conscious loan of semantic or expressive content, often referred to as 'tapping into a source' in Edo studies, and the recognition of possible *genkyo*s and *shukō*s for any text seems to have been the most appreciated methodology when studying Edo fantastic literature. This paper aims to cast doubt on the internal coherence of this terminology and the methodology it entails and further endeavours to explore a more profound reconsideration of this relation between texts so as to assess whether these relations and borrowings are significant when scrutinising *Tsubosumire* or any other text in Edo literature.

It then becomes clear that rather than undergoing an intertextual reading, *Tsubosumire* has been analysed within the framework of so-called 'source studies', or better '*shukō* studies', which certainly have a traditional significance and are highly relevant in the Japanese tradition of literary criticism. As is well known, the term *shukō* as used by Edo authors, derives its meaning both from the *shukō*polarity of the '*sekai* 世界 vs *shukō*' dyad of the theatrical lexicon and from the poetic terminology, where the term means the 'original idea' or 'gimmick' behind the poem. Accordingly, in Edo studies *shukō* came to mean two different and somehow quite opposite things: on the one hand, *shukō* is a new idea that enhances the perception of a literary text as being new within its 'genre' – incidentally, the term 'genre' and its usage within Edo studies is utterly prob-

<sup>1</sup> Tokuda (1995a; 1995b) has already put into relation *Ugetsu monogatari*'s stories "Asaji ga yado" 浅茅が宿 (The House in the Reeds) and "Kibitsu no kama" 吉備津の釜 with *Huo Xiaoyu zhuan* 霍小玉傳: the Chinese text is defined as a possible *funpon* 粉本 (sketch) or source of *shukō* 趣向 for the two ghost stories.

lematic as already pointed out by Nagashima (1999) –; on the other hand it is the 'literary borrowing of a narrative mechanism', thus, indicating not the 'original element of a story but the 'inspiration' which triggered the 'characteristic elements' of the story.<sup>2</sup> The first meaning, for example, is easily found in the reviews of literary texts, *gesaku hyōbanki* 戯作評判記, where the authors highlighted which *shukō* (or original idea) was worthy of being noted in each category. This nuance of the term may obscure the hypertextual relations at stake in the creative process, but it is evident that here, a dyad of '*sekai* vs *shukō*' is still active: the *sekai* on which the *shukō* is applied is the web of conventions of the 'genre'. Therefore, the *shukō* is the novel element that makes a particular recurrent arrangement of narrative content evolve into something new. One could even say that literary forms such as *kaiwatai sharebon* 会話体洒落本 (dialogic *sharebon*), urban *kibyōshi* 黄表紙 (illustrated short stories with yellow covers), and in later years, conventional *ninjōbon* and *hizakurigemono* 膝栗毛物 (reprises of *Hizakurige* narratives) are a series of sequential rewritings, where every new work is an adaptation of the previous one and should be analysed as such. Reading the definition of *shukō* attributed to Chikamatsu Monzaemon 近松門左衛門 (1653- 1724) in the preface of *Naniwa Miyage* 難波土産 (Souvenir from Naniwa, 1738), it leads us to believe that the dialogue between previous works and the new ones is the same:

*Shukō* works this way too: while resembling the original, a new work should also have sections that only roughly follow the original. This is after all what art is and what people find enjoyable. (Saltzman-Li 2010, 131)

However, Edo authors often make use of the term *shukō* with the other meaning of 'inspiration from a previous work', and this meaning is prevalent in today's Edo literary studies. For example, in the manuscript version of *Kaidan oi no tsue* 怪談老の杖 (Ghost Stories: The Old Man's Staff), Ōta Nanpo 大田南畝 writes this record about a story of a resentful second wife who cursed the first wife's grave: "The inspiration of *Geisha yobukodori* is based on this work [*Geisha yobukodori no shukō wa kore ni motozukite* 妓者呼子、鳥の趣向は、是に本づきて]" (Kokusho kankōkai 1913, 240). Several years later, in the incipit of *Kuruwa zōdan* 廓雑談 (Various Stories from the Pleasure Quarter, 1826), Hanasanjin writes: "This story of olden times functioning as a seed | source of inspiration" (*ko no ikkai no monogatari o shukō no tane toshite* 此一回の物語りを趣向の種として) (*Kuruwa zōdan* I, f. 6*v*),

<sup>2</sup> I agree with Moretti (2000, 59), who highlights the variety of nuances the different occurrences of the term *shukō* have in Edo materials.

which refers to some gossip about a love triangle that escalated into a vengeful ghost story. As these two examples show, the term is very often linked with others indicating a 'literary borrowing' (*motozuku* 本づく 'to be based on', *tane* 種 'seed') and its very meaning in some cases shifts from 'original mechanism' to 'mechanism inspired by something else' even when there are no other terms suggesting this idea of borrowing.

The main question here is that this concept of *shukō*, along with the other terms often used, such as *tenkyo* or *genkyo* or *shutten*, although inferring an idea of 'borrowing', do not imply a real intertextual effect for the text. In fact, being for the most part intended as a source of inspiration, the old text does not directly trigger a dialogue with the new text or plot and, furthermore, it is unclear as to whether the readers were aware of this. Hence, it pertains to the text as an historical artefact rather than as a literary piece.

### 3 Intertextual Reading and 'Source Studies'

All in all, it is expedient to think that previous studies on *Tsubosumire* analyse their object as an historical artefact, demanding it be a once forever determined unit of sense, which is historically included within a process of affiliation through semantic borrowings from one or more different sources. These analyses are not problematic *per se* except when one seeks to engage with a reading of *Tsubosumire* as a literary text. This opposition between historical artefact and literary text must not be interpreted as a difference in the quality of the texts or in their engagement with the plurality of sense – it does not equate the opposition between 'work' and 'text', or *texte lisible* and *texte scriptible* in Barthes;<sup>3</sup> nor does it necessarily adhere to the traditional opposition between artefact and aesthetic object styled by Mukařovský (cited in 2015, 300-3), a foundational one for modern reception theories, inasmuch as it ultimately entails a distinction between the idea of decoding the text *vis-à-vis* and that of interpreting the text, as Bottiroli (2006, 106) insightfully demonstrates. Ōhashi and the other scholars engaged with *Tsubosumire* do not seek to decode the stable meaning of the artefact but only to frame it in a process of affiliation, positing it in historically determined literary categories and relations with other texts, only described by criteria such as similarities between plots and other details. On the one hand, Yokoyama (1974) speaks about influence and similarities, and

<sup>3</sup> See respectively Young 1981, 31-47, although page 43 interestingly styles how historical artefacts are commonly studied before the development of textual analysis; Marrone 2016, 181-6 for an analysis; and Barthes 2002, 2-15 in particular.

Ca' Foscari Japanese Studies 19 | 6 **86** Images from the Past: Intertextuality in Japanese Premodern Literature, 81-94

in doing so his analysis may explain why the work is as it is; however, he does not address what it is like. Very much like Yokoyama, Ōtaka (2003) deals with superficial similarities and does not touch on how to read or interpret the text.<sup>4</sup> And the same methodology is evident in Ōhashi's article, which eminently addresses surface similarities and discrepancies among portions of the plot of the different titles she brings into discussion. Miura seems to differentiate between sources (*genkyo*) and inspirations (*shukō*) for *Tsubosumire*, and he explicitly states that *Genji monogatari* is to be considered the source of the *shukō*, functioning also as a source of signification invoked by the author himself:

it is made clear that, at least in the description of the mysterious phenomena caused by the vengeful ghost of Katsuura in the fifth volume, the word usage of *Genji monogatari*'s 'Yūgao' 夕顔 Chapter is applied, and the author himself demands his readers to read it as the *shukō*. (Miura 2002, 28-9)

In this way, although the relation between the text and *Genji monogatari* is credited by Miura with having the function of giving literary depth to the mysterious description and to complete the portrayal of Katsuura's sorrow and the inevitability of her vengeance (Miura 2002, 29), this is accepted by Miura only in light of the authorial words in the paratext where he explicitly alludes to the Heian (794- 1185) masterpiece. What happens then to the other main editions of the text, where the peritextual apparatus is effaced and substituted with one summoning a different literary discourse? Should the 'presence' of *Genji monogatari* in the text be considered as diminished or of less significance when there is no authorial voice pointing it out? In his preliminary definition of the concept of genre-function, Bawarshi (2003, 26-9) builds on an example by Heather Dubrow proving that a change in the title of a book may affect the reader's attitude towards a text, because genres (whereof titles are often transparent tags) "function as sites of action that locate readers in positions of interpretation". Anyway, although titles and other peritextual elements may direct the reader's reception of a text, casting light on some aspects and overshadowing others, the presence of other texts in the text is still active: it may be overlooked when reading the text, but not when interpreting it, to use Bottiroli's (2006) difference between – virtually infinite – doxastic readings *versus* interpretations.

<sup>4</sup> Reviewing a later collection of essays by Ōtaka, which include Ōtaka (2003), also Kigoshi (2011, 81) strongly disapproves of Ōtaka's superficial approach in comparing texts.

Ca' Foscari Japanese Studies 19 | 6 **87** Images from the Past: Intertextuality in Japanese Premodern Literature, 81-94

#### 4 Reading *Tsubosumire*'s Intertextuality

By and large, previous critical literature concerning *Tsubosumire* may apparently be styled as pertaining to traditional source studies or influence studies but it is evident that these studies mainly address the alleged history of its affiliation without dealing with any literary aspect whatsoever. Therefore, if they undisputedly fail to comply with a so-called intertextual theory,<sup>5</sup> they also do not completely attain to the 'source studies' trend styled by Hutcheon (1986) and apologetically by Quirk (1993; 1995), amongst others.

As I previously mentioned, according to Ōhashi, the principal source for *Tsubosumire* is the first story of *Rinbeki yawa* and this intertextual relation lies in the striking resemblances between the narrative structure of the last part of the novel with the plot of the short story. However, it must be said that the preface of *Tsubosumire* quite openly states that the text is just a rewriting of a story once heard in the past (Ōtaka, Kondō 2000, 338). It is not rare to find this sort of framing in prefaces and paratexts connected with ghost stories and, in fact, the prefacer of *Rinbeki yawa* himself confesses that he is just reporting ghost stories he overheard one night being recounted to others in the room next to his (Mizuno et al. 1978-88, 9: 321-3). All in all, to preserve the analysis of the perspective of the historical production of *Tsubosumire* as an artefact, these vague revelations of the source may not be completely fictitious considering that stories like this are said to be based on rumours of actual incidents which occurred in the past, such as the flying sphere of the soul of the wife cursing her rival recounted in the abovementioned *Oi no tsue*, which, according to the author, is based on a true story that took place in 1722. *Kuruwa zōdan*, another example, recounts the story of a love triangle, which clearly resembles that of *Tsubosumire*, and is said to have happened in Yoshiwara 吉原 during the 1770s. Other similar stories are detectable even in diaries and collections of mysterious anecdotes recorded in the previous century. And this is without making mention of the classical repertoire of stories of female vengeance against deceitful men. In light of all this, the debate on the search for the one text that inspired the *shukō* of *Tsubosumire* appears to be rather fruitless. From the point of view of how *Tsubosumire* function as a literary text, regardless of whether it is or not the source of inspiration on which the author of *Tsubosumire* built the last part of its story, the narrative structure of *Rinbeki yawa* is not involved in a dialogue with other elements and an analysis of the intertextual links triggered by *Tsubosumire* do not bring us to *Rinbeki yawa*.

<sup>5</sup> For a general overview of the critical stance of the principal spokespersons of intertextual reading toward source studies see Allen 2000, 36, 53, 69-74, 120-1, 135.

Ca' Foscari Japanese Studies 19 | 6 **88** Images from the Past: Intertextuality in Japanese Premodern Literature, 81-94

Considering the text from an historical or a contextual point of view, it must be noticed that the author, but even more the publisher, are orphans of the Kansei (1789-1801) censorships in search of a new field of fiction to explore. Wakabayashi Seibei and Kazusaya Ribei 上総屋利兵衛, as Ōtaka (Ōtaka, Kondō 2000, 679; on the latter also see Kigoshi 2002) precisely points out, were admonished for their activity of *sharebon* publishers in the nineties and therefore, they were struggling to find new products to sell to the same readership (see also Nakamura 1956; Tanahashi 2007; and Suzuki 2017, in particular 264-72). In this regard, *Tsubosumire* is not a work participating in a homogeneous trend but stands as a hybrid text with both conservative and innovative aspects when compared with other texts both from the *yomihon* corpus and *kaidan* literature. And the text, as *Ugetsu monogatari* before it, is enriched by several intertextual links that ennoble the characterisation of its protagonists and the love triangle they are involved in. Whether Miura's hypotheses that *The Tale of Genji* offered the *shukō* for the story is true or not, it is undeniable that the text builds on some famous scenes from *The Tale of Genji* to give depth to the story and strengthen the overall architecture of the narrative built on a new paradigm in the characterisation of the protagonists.

Ōtaka (in Ōtaka, Kondō 2000, 677) notices that, while *Ugetsu*'s and most of pre-1790s *kaidanshū*'s 怪談集 (collection of ghost stories) male protagonists are young men who are described as unworldly, or incapable of getting by in this society (*wataraigokoro no nai* 渡らひ心のない), their immaturity and propensity to follow their emotions without considering the consequences finally leads them to happen on a ghost vendetta and either die or redeem themselves by learning how to comply with the rules of society. On the contrary, an important aspect in *Tsubosumire* is the prudence of Heijirō and his propensity to comply with the rules instead of following his heart which eventually causes the ghostly phenomenon. This difference and the novelty are highlighted within the text precisely by intertextual comparisons of classical *exempla* of integrity and masculinity, such as those between Genji's restraint in famous scenes from the Heian novel to the behaviour of the young samurai Heijirō. Heijirō is, in fact, a samurai hero who, whilst being sometimes described as emotional and indecisive (Ōtaka, Kondō 2000, 344; see also Miura 2002, 28), complies with all the rules of the world and adheres to what society and his superiors ask of him and in so doing forgets his true feelings (*ninjō* 人情) and sets in motion his tragedy.

By characterising Heijirō's behaviours with references to Genji's words and actions and by positing him in modern tableaux inspired by famous chapters of the classical novel, the text suggests the reader to consider the protagonist in a rather positive way. This interpretation is also reinforced by the fact that prospective readers of the first version of the novel were probably *literati* or well-educated men, if we consider that titles published by Wakabayashi in those years predominantly address a public who could read classical Japanese and literary Sinitic. One could go as far as to point out possible *kokugaku* 国学 (national studies) sympathies within the group of people which produced and read the text. This also links with another innovative aspect of the plot: *Tsubosumire* overturns another convention of ghost stories, such as the cause-effect (*inga* 因果) mechanism and, in fact, functions in quite the opposite way to *Kaidan kien*, the Buddhism-inspired text Yokoyama considers as a source. Heijirō performs many good deeds in the first three volumes, such as helping a stranger who we later learn be the father of the courtesan Katsuura or demonstrating his intention to save Katsuura from her fate or devoting himself to his duties and to calligraphy. Anyway, these good deeds are not beneficial to Heijirō's destiny but only bring sorrow.

All this considered, although it may be true that the author and the publishers used *Rinbeki yawa* as a source of inspiration, this does not teach us anything about how *Tsubosumire* works as a literary text. When addressing a meaningful reference, while the reader must choose whether to keep reading or to explore it,<sup>6</sup> it is evident that whoever seeks to interpret the text has no alternative other than choosing both ways: triggering a dialogue or even bringing out a conflict between this displacement and the text hosting it. The case of *Tsubosumire* is fascinating not only because its web of meaningful references is wide and covers many fields of the literary world, with both *verbatim* loans from the classical repertoire, reconstructions of sentimental and witty dialogues as in eighteenth century urban Edo literature, and vivid descriptions of mysterious events which remind of some contemporary grisly illustrations of obscure phenomena. *Tsubosumire* adds to this variety of tones the intersection of different ways of reading the text suggested by its paratext, which was changed through time. For instance, the appearance of the book in the *chūhon* format (1797 edition), or with a more *au courant* title (1823 edition) certainly had the effect of widening the readership of the text, including emergent groups of readers, such as women.<sup>7</sup> In comparison with the prototypical heroes of the sentimental novels shaped in the nineteenth century for a female readership, Heijirō's prudent attitude may be interpreted as lack of will or strength, determining an overall negative characterisation of the protagonist. The intersection with this interpreting attitude is justified especially since *Tsubo-*

<sup>6</sup> I refer to these well-known lines notably written by Jenny (1982, 45): "each intertextual reference is the occasion for an alternative: either one continues reading, taking it only as a segment like any other, or else one turns to the source text, carrying out a sort of intellectual anamnesis where the intertextual reference appears like a paradigmatic element that has been displaced, deriving from a forgotten structure".

<sup>7</sup> I wish to acknowledge the anonymous referee's insightful suggestions that helped sharpen my focus on this point.

*sumire*, at least the first half of the novel (for instance, see chapter 2 and the affair between Katsuura and Kendayū, a wealthy but unpleasant customer of the Yamamoto house) is rich in witty or sentimental dialogues which, although expressed in classical Japanese, convey the informal atmosphere of the newborn sentimental narrative.

In lieu of a conclusion, I have summed up some considerations on how *Tsubosumire*'s web of literary allusions actually works: the text makes wide usage of *sharebon*-like dialogues enhancing a sentimental overtone; it modernises the characterisation of the male protagonist by mirroring him with classical examples of masculinity; it challenges and breaks the convention of the *inga* mechanism, by collecting false clues which do not lead to the expected results. A rereading of the intertextual web embedded in *Tsubosumire* helps us to consider this text not as a narrative which copies and expands to the long-story form of a short-story plot already exploited in the past but as a hybrid text challenging the expectation of its readership by including new and destabilising elements.

### **Bibliography**


Miura Ichirō 三浦一郎 (2002). "*Tsubosumire* kō. Sono genkyo, sakusha, shiteki igi nado ni tsuite" 『壺菫』考 – その原拠、作者、史的意義などについて (A Study on *Tsubosumire*. The Sources, the Author and the Historical Significance). *Nihon bungei ronkō* 日本文芸論稿, 27, 18-29.

Mizuno Minoru水野稔 et al. (eds) (1978-88). *Sharebon taisei*洒落本大成 (Complete Collection of *Sharebon*). 30 vols. Tōkyō: Chūōkōronsha.

Moretti, L. (2000). "Quando la creazione si fa allusiva: la retorica testuale dello *shukō* nelle forme narrative del periodo Tokugawa". *Asiatica Venetiana*, 5, 59-84.


#### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# The Dynamics of *Nyonin jōbutsu* in Zenchiku's *Yōkihi Honzetsu***, Poetic Allusion, and Sacred Space**

#### Matthew Chudnow

Central Connecticut State University, USA

**Abstract** Medieval nō theatre is built upon *honzetsu* (source texts), which are drawn from the premodern Japanese canon and critical for dramaturgical structure and meaning. This study focuses on Zenchiku's *Yōkihi*, a female-spirit play with an initial *honzetsu* of Chinese Tang the poem *Changhenge* that is further filtered through additional texts, primarily Heianperiod narrative *Genji monogatari*. Therefore, examination of *Yōkihi* involves consideration of all relevant *honzetsu* cited within the play, their earlier Heian receptions, and subsequent medieval textual lives. The dramaturgical structure and language of *Yōkihi*, with its variations from these *honzetsu*, emphasise thematic concerns of existential isolation, karmic clinging, and sacred space as critical to feminine ontological and enlightenment status.

**Keywords** Nō theatre. Premodern Japanese literature. Religion. Gender dynamics. Female enlightenment.

**Summary** 1 Introduction and Methodology. – 2 Brief Background on Religious Environment and Female Soteriology in Medieval Japan. – 3 *Yōkihi*: *Honzetsu* and Textual Analysis. – 4 Conclusion: Inability to Transmigrate or Incomplete Ritual Process?

#### 1 Introduction and Methodology

While focused on universal salvation of all sentient beings, Mahāyāna Buddhism displays practical incongruities concerning female gender and sexuality. One example is manifested in nō theatre, Japan's masked drama developed during the Muromachi period (1337-1557).

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 95** Submitted 2021-12-13 | Accepted 2022-03-21 | Published 2022-08-30 © 2022 Chudnow | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/006**

nō is a vector for elite Buddhist soteriological discourse and popular religious beliefs, providing a view into gender-based social and spiritual topics within medieval Japanese society, such as gender-bias within soteriological discourse. This example is presented in Third Category or female-spirit plays (*katsura mono* 鬘物, or 'wig plays') genre, works featuring dense religious language, utilisation of shamanic ritual, and dramatic conflict centred on ambiguous feminine soteriological status. This article focuses on *Yōkihi* 楊貴妃 (Consort Yang), a female-spirit nō play attributed to actor and dramatist Konparu Zenchiku 金春禅竹 (1405-1468). Influenced by Chinese Tang Dynasty (618-907) poet Bai Juyi 白居易 (Jp. Haku Kyoi, 772-846) and Heianperiod author Murasaki Shikibu's 紫式部 eleventh century narrative *Genji monogatari* 源氏物語 (The Tale of Genji), *Yōkihi* is a nuanced dramatic work that considers the issue of female soteriology within medieval combinatory Buddhist discourse through multiple layers of poetic influence. Its attributed author Zenchiku frequently utilised the *mugen* 夢幻 (dream-vision) plot structure, which was developed by his father-in-law, innovative actor and playwright Zeami Motokiyo 世阿弥 元清 (1363-1443). In *Yōkihi*, the *shite* シテ (main actor) plays the spirit of Yang Guifei 楊貴妃 who languishes in a Daoist heaven realm, clinging to memories with her karmic state in stagnation, experiencing emotional torment. Is this poetic emphasis within Zenchiku's dramaturgy to produce *yūgen* 幽玄 (ineffable beauty) or a manifestation of deeper medieval socio-religious issues? To answer these and other questions, my primary interdisciplinary methodology takes a threepronged approach to analysis of nō plays (*yōkyoku* 謡曲) in historical, social, and religious context. The first is predicated on nō's textual construction as densely layered rhetorical weaves of allusive poetic variation and textual citation via *honzetsu* 本説 (source texts). This wide range of texts from the classical (ninth century) through late medieval (fifteenth century) eras of the Japanese and Chinese literary canons include genres such as: poetry (*waka* 和歌, Japanese poetry; *kanshi* 漢詩, Chinese poetry; *renga* 連歌, linked verse), diaries (*nikki* 日記), narrative tales (*monotagari* 物語; *otogizōshi* 御伽草子), folk tales (*setsuwa* 説話), temple histories (*engi* 縁起), and religious texts (various genres). *Honzetsu* was central to nō textual composition and structure, and also dramatic performance, in that these textual sources:

typically provides the source material on which the composition [of the nō play] as a whole is based, the *shite*'s identity, and the outline of the plot coming from it. (Quinn 2005, 138)

Therefore, it is logical that interrelated gender elements, religious and ontological associations, and geographical locations with their associated sacred spaces located within these *honzetsu* will be vital factors in determining soteriological status for a play's female-spirit persona. As will be examined below, the demarcation and ritual purification of sacred space was critical in the *honji suijaku* 本地垂迹 (the Buddhas as original enlightenment, the kami as traces) paradigm (Teeuwen, Rambelli 2003, 4-7) that defined medieval religiosity, making a historicised analysis of geographic location and sacred space within *honzetsu* essential to understanding its soteriological dynamics.

This leads into my second approach, the examination of the femalespirit persona's geographic location, its religious associations, and their impact on soteriological status. I will unpack the religious and poetic references connecting a character to her physical surroundings and how they inhibit or encourage enlightenment. In almost all instances, nō plays take place in spiritually charged spaces directly affecting dramatic action, with each play's setting functioning like a central dramatic persona. A primary example of this methodology at work is demonstrated in *Yōkihi*'s ostensibly Chinese folk religion setting of Penglai 蓬莱 (Jp. *Hōrai*), which will be examined in detail.

Lastly, I systematically examined the densely layered rhetorical weaves of allusive poetic variation and textual citations in nō plays. This involves two distinct activities: tracking the multiple textual citations and *honkadori* 本歌取り (allusive variation) to canonical works for symbolic and allusive resonance and identifying the utilisation of *honji suijaku* language and textual sources for their meaning. Etymological allegoresis, or paronomasia, is a central concept to medieval understanding of language and plays an important role in delineating nō *honzetsu*, especially regarding soteriology.<sup>1</sup> Seemingly non-religious words and phrases, when cross-referenced with Buddhist texts and related writings, may reveal deeper Buddhological or kami worship meanings.

My analysis of *Yōkihi* reveals that achieving enlightenment for female-spirits in the medieval *honji suijaku* worldview of Muromachi Japan is complex and variable, with ambiguous soteriological potentialities. As I will demonstrate, the female *shite* in *Yōkihi* grapples with karmic constraints from previous manifestations of cosmetically Chinese literary origins, that are revealed to be deeply rooted in medieval Japanese literary and religious traditions. Additionally, she is located in an ambiguous sacred space with resonances that have shifted across space and time via disparate *honzetsu*. This is combined with an incomplete/missing ritual process for enlightenment, further complicating her soteriological potential.

<sup>1</sup> "Paronomasia analyzes graphs into their component parts and rearranges them into a sentence that reveals the 'true' meaning of the graph, including hidden identities. The methodology of associational identification derives from the rule of correspondence: that when two people, places, or things share a name, they must ultimately be the same (nondual)" (Klein 2021, 91).

Ca' Foscari Japanese Studies 19 | 6 **97** Images from the Past: Intertextuality in Japanese Premodern Literature, 95-116

#### 2 Brief Background on Religious Environment and Female Soteriology in Medieval Japan

Medieval Japan was governed by the *honji suijaku* paradigm where Buddhism, kami worship, and multiple religious systems were combined in an amalgamation under the ultimate authority of Buddhist institutions (Teeuwen, Rambelli 2003, 4-7). It is within the systems of this paradigm that the Buddhist doctrine of female enlightenment (Ch. *nuren chengfo*, Jp. *nyonin jōbutsu* 女人成仏) operated. However, in early Indic Mahāyāna Buddhism, the female body was a problematic topic due to cultural perceptions of physical and spiritual uncleanliness that necessitated extraordinary means for potential enlightenment.<sup>2</sup> In Japan's earlier Heian period, court commissioned rituals based on Ying-Yang divination (*onmyōdō* 陰陽道), adapted from combinatory Chinese traditions termed the Five Elements or Five Phases (Ch. *wuxing* 五行; Jp. *gogyō shisō* 五行思想), served to reinforce ideologies of ritual purity as central to spiritual health, personal safety, and financial prosperity (Faure 2003, 71). By the medieval era, *kegare* 汚れ (defilement), meaning 'filth' and 'exhausted energy', became central within society to preserving ritual purity, including the demarcation of sacred space within *honji suijaku* religious thought (Faure 2003, 69). A woman's menstrual cycle, blood from childbirth, and other feminine biological processes were viewed as desecration to sacred and even secular spaces. Medieval religious responses to these blood-based sources of *kegare* were not uniform but shifted in severity over time.<sup>3</sup>

When considering female soteriology, it is useful to address the core of Buddhist practice as cessation of *saṃsāra* (Ch. *lunhui*, Jp. *rinne* 輪廻), wandering through the six realms of rebirth (Buswell, Lopez 2014, 758). It is manifested in the medieval Japanese concept of the *rokudō* 六道 (Skt. *ṣaḍgatīḥ*, Ch. *liudao*), the six realms (literally, 'paths') of transmigration. The concept of *rokudō* permeated all forms of combinatory Japanese religious experience. While the original Sanskrit terminology refers to six destinies of rebirth with their associated realms of existence (Buswell, Lopez 2014, 315-14, 731), the premodern Japanese interpreta-

<sup>2</sup> The Parable of the Dragon Princess (Skt. *Nāgakanya*, Ch. *Longnu*, Jp. *Ryūnyo* 龍女): she gives the Buddha a jewel representing purity, changes into a human male, then an enlightened buddha (Hurvitz 2009, 184). While allegorical for female enlightenment potential, multiple aspects of this parable have been heavily debated and are outside the current study.

<sup>3</sup> *Onmyōdō* and several kami worship sects (such as Ise Grand Shrine 伊勢神宮) viewed *kegare* via female blood harshly, as did Nichiren 日蓮 (1222-1282), who cited menstruation as "the moral defilement of women". Other Buddhist leaders were positive, with Jōdo sect 浄土宗 founder Hōnen 法然 (1133-1212) denying it as *kegare*. Rinzai Zen 臨済宗 monk Mujū 無住 (1227-1312) also discounted female blood as defilement, proclaiming "if only the heart is pure, the body likewise is not defiled" (Faure 2003, 71-3).

tion emphasised the 'act of wandering' through all six realms (*rokudō rinne* 六道輪廻) and with the accruing of karmic sin or merit (Nakamura et al. 1989, 848). Central to *rokudō* were the many hell realms and their associated torments, depictions of which gained widespread influence in premodern literature and visual art. A primary option for salvation in this *honji suijaku* system was through shamanic ritual practised by an ascetic who communicated with the spirit realm. The role of the ascetic, in the form of the *waki* ワキ (side actor), is a central feature in female-spirit nō plays, linking the diverse threads of medieval combinatory Japanese religious practice with nō dramaturgy.

The ascetic is a principal figure in the *honji suijaku* religiosity of the medieval era, a conduit between the human world and spiritual realms inhabited by kami and Buddhas (Blacker 1999, 75 and 79). It also bears relevance to the role of the *waki* in the *mugen* nō genre. Within this dramatic structure, a Buddhist monk (*waki*) encounters a spirit disguised as a local resident (*maejite* 前ジテ, Part 1 lead actor) of a famous site who is suffering from karmic delusion, thereby trapped in the physical and spiritual pain of *rokudō*. This spirit slowly reveals their true identity via poetic exchanges filled with profound knowledge about the famous historical or literary figure connected to the famous site. The monk promises to hold funerary rites to assist in the spirit's transmigration, thereby breaking the cycle of *rokudō* and releasing the spirit into enlightenment. In Part II, the *shite* re-emerges in their true form having changed costume (*nochijite* 後ジテ, Part II lead actor) and acts out the character's karmic delusion in connection with the location, punctuated by a final dance. Intrinsic to the *waki*'s status as an ascetic are the following abilities within the dramaturgy of the play: the ability to undertake a dreamvision, contact spirits, and special ritual knowledge. In medieval Japan, only select initiates with esoteric knowledge and spiritual powers gained via physical austerities and religious disciplines were true aesthetics, primarily tasked with healing illness and exorcising malevolent spirits (Blacker 1999, 21-2). Fully ordained Buddhist monks of the Tendai, Shingon, and Nichiren sects frequently performed severe austerities within their religious life and specialisation in healing spiritual ailments (Blacker 1999, 165), aligning them with the Buddhist monks frequently seen as *waki* in female-spirit nō.

In Japanese shamanism, the spiritual journey by the ascetic is typically accomplished utilising three different methodologies: first, through a supernatural dream (dream-vision); second, divine possession by a spirit or god (*kamigakari* 神憑り); and third, in the mantic journey where the ascetic's soul (*tamashii* 魂) is guided across spacetime to another realm, often by a guardian divinity (Blacker 1999, 168). Dreams have a long history of supernatural, spiritual, and literary provenance in Japan. In the case of *mugen* nō, the dream-vision appears to be the most common type of shamanic communion between the *waki* and the spirit *shite*. The lines between dream-vision and mantic journey may be blurred in nō dramaturgy, as will be observed in the analysis of *Yōkihi*.

### 3 *Yōkihi*: *Honzetsu* and Textual Analysis

*Yōkihi* occurs in a heaven realm, yet the *shite* of Yang Guifei occupying this reality expresses intense emotions that are fully human, diverging from standard depictions of Buddhist heavens. Why this incongruity? This involves analysis of the multiple source texts of Zenchiku's work and their spiritual resonances. The play seems to take its influence from the final third of *Changhenge* 長恨歌 (Jp. *Chōgonka*), a narrative poem written by Chinese poet Bai Juyi. Active during China's Tang Dynasty, Bai's poetry exerted influence on the Heian period, with its subsequent medieval literary reception heavily filtered through this prior canon. Therefore, examination of *Yōkihi*'s dramatic operation involves consideration of the original poem, its Heian reception, and medieval era readings. From a critical perspective, the textual structure of Zenchiku's play, and its unique variations from Bai's text, emphasises *Yōkihi*'s thematic concerns of existential isolation, karmic clinging, and the static nature of the Yang persona. In many ways, the feminine suffering of the *shite* this work clarifies complimentary themes that pervade Zenchiku's other female-spirit plays: a deep textual connection to *Genji monogatari* as a primary *honzetsu*, female-gendered modes of karmic suffering, and the centrality of sacred space as critical element to the *shite*'s spiritual status (Chudnow 2017, 95). Before continuing, delineation of the following is required: first, the work's Chinese antecedents; second, its influence and connection with the Heian literary world informing it; and third, this overall literary cannon's further influence on the Japanese cultural world.

Bai Juyi's poetry exerted a considerable impact upon *kanshi* 漢詩 (Chinese-language poetry) among the Heian literati, with the synthesis of his works into the Heian poetic consciousness being essential in the development of *waka* 和歌 (Japanese-language poetry). Among his works (Ch. *Bai shi wenji*, Jp. *Hakushi monjū* 白氏文集), the *xin yuefu* 新楽府 (Jp. *shingafu*, 'new ballads') were especially popular in the Heian period, However, their Confucian didactic nature was generally ignored by most Japanese *kanshi* and *waka* poets, who preferred a depoliticised aesthetic focusing on the natural world's beauty as an expression of truth (Smits 1997, 174). This selective approach is a hallmark of Bai reception in Japan and essential to understanding later Japanese interpretations of his works as *honkadori* and *honzetsu*. Of all the subjects in Bai's collected poems, his poems detailing imperial consorts most enjoyed the greatest popularity with Heian and later medieval writers and were referred to collectively as the "Five Consorts" (Jp. *gohi* 五妃; Sasaki 2001, 89). The poetic images of these women be utilised as *honkadori*, *honzetsu*, and other forms of intertextual variations in multiple premodern Japanese literary and performance genres, with representative examples in Heian period *waka*  and *monogatari*, as well as later Muromachi period nō plays. However, the negative historical and political contexts originally depicted in Bai's original poems were frequently removed or glossed over in these Japanese interpretations, thus constructing the "Five Consorts" as idealised feminine poetic archetypes that typified poetic expression of tragedy, isolation, longing, and ephemeral beauty (Sasaki 2001, 87-8). Of all these poetic personae, *Changhenge*'s Yang Guifei was especially captivating to the Japanese artistic and popular imagination throughout the premodern era.

Reception of *Changhenge* during the Heian period was multifaceted across several artforms. Bai work was popular across all literate Heian social classes, including the highest-ranking members of the imperial court: Genji's author Murasaki Shikibu tutored Empress Shōshi 彰子 (988-1074) in Bai's poetry (Fukumori 2001, 109). In *Sarashina nikki* 更級日記 (The Sarashina Diary, ca. 1060), the author takes great pleasure in receiving a gift of an early *monogatari* (fictive narrative) version of *Changhenge* as a beautifully illustrated scroll (Inukai 1994, 303). The Heian court was immersed in works dedicated to Bai's work, as will be discussed further below. The most famous of these is the opening "Kiritsubo" 桐壺 (The Paulownia Pavilion) chapter of Murasaki's *Genji monogatari*, where *Changhenge* and the image of Yang play an essential narrative function, as will be explored further. The deep intertextual dynamic between Bai's work and Shikibu's tale resulted in a symbiotic relationship between the two texts in post-Heian reception, with *Genji* and *Changhenge* forever linked in the Japanese literary consciousness. Furthermore, this literary image of Yang established an independent life within the premodern Japanese literary canon, separate from Chinese historical fact or the original context of Bai's poem (Itō 1988, 508). Both of these conditions will be seen in full effect through analysis of Zenchiku's *Yōkihi*.

A basic summary and analysis of *Changhenge* is essential to contrast Bai's original Yang character with the hybrid *Genji*-influenced Yōkihi-persona of Zenchiku's play. The original is a pseudo-historical narrative poem based on popular aspects of the political scandal at the Tang Emperor's court during the later era of that dynasty: the deep infatuation of Emperor Xuanzong 玄宗 (Jp. Gensō, 685-762) with his primary consort Yang Guifei (Jp. Yōkihi, 719-756). Born into a family of minor provincial officials, Yang was introduced into court as a concubine for the emperor's son. After the emperor's previous favourite died, Yang was recommended, and he was instantly taken with her artistry in dance. She was raised to level of *guifei* 貴妃 (Jp. *ki-* *hi*, 'Precious Consort'). Popular legend and Bai's poem dictates that Xuanzong became more focused on Yang than government, culminating in the rebellion of general An Lushan 安禄山 (Jp. An Rokuzan, 705-757) in 755. During the evacuation of the capital Chang'an 長安 (Jp. Chōan), the emperor's guard refused to continue unless Yang, whom they blamed, was killed. Xuanzong then abdicated in favour of his son, who quelled the rebellion and reinstated Tang rule. Later, Yang was given the title of Empress posthumously while Xuanzong grieved in retirement (Jian, Luo 2000, 219-21).

Although *Changhenge* deals with these events, Bai Juyi is interested in portraying a tragic love affair through poetic narrative. While Bai was a Confucian scholar and an active provincial official, the complex political conditions surrounding the An rebellion are obscured to stage a romantic tragedy. The final two thirds of the poem is focused on the emperor's deep mourning, forming the poetic and thematic core. Tortured by sorrow, Xuanzong employs a Daoist wizard (an ascetic) to undertake a mantic journey to search for Yang's spirit, as she has inauspiciously not appeared in his dreams. The wizard transverses heavens, the earth, and underworlds but is unable to locate Yang. He finally arrives at Penglai, the Isle of the Immortals, where Yang has been reborn as a goddess. She grants the wizard an audience and speaks of her longing for the emperor. Before the wizard leaves, she entrusts him with a jewelled hairpin and filigree box to give Xuanzong as evidence of her existence and sign of their eternal love. She also tells of their promise in life, on the seventh night of the Seventh Month, during the Seventh Night (Ch. *Qixi*, Jp. *Tanabata* 七夕) star festival:<sup>4</sup>

[I]f in Heaven, may we become those birds that fly on shared wings; or on Earth, then may we become branches that twine together. (Owen 1996, 447)<sup>5</sup>

The closing line of Bai's poem iterates that even after Heaven and Earth cease to exist, "yet this pain of ours will continue and never finally end" (Owen 1996, 447).

According to Manling Luo, this ending in the heaven-like realm of Penglai, coupled with the eternal vow made on earth, completes the construction of *Changhenge* as a transcendental romance. This posi-

<sup>4</sup> Allusion to the Chinese legend of the Cowherd (Ch. *Niulang*, Jp. Gyūrō 牛郎, the star Altair) and the Weaver Girl (Ch. *Zhinu*, Jp. *Shokujo* 織女 or *Orihime* 織り姫, the star Vega), who can only meet once a year, on the seventh night of the seventh month of the lunar calendar, by crossing over the River of Heaven (the Milky Way) on a bridge of magpies (Luo 2005, 166).

<sup>5</sup> Owen's translation was used for this study while Kroll's (1990-91) translation was also referenced.

tions Xuanzong and Yang Guifei as a romantic couple on an equal hierarchical level via physical and spiritual separation, thereby circumventing the political and social pressures inherent in their roles in life (Tang Emperor/man and concubine/woman); with him in the human realm and her reborn as an immortal goddess in Penglai (Luo 2005, 166-7). It is also worth noting that Yang is reborn as a divine Daoist being, elevating her social and spiritual status to one worthy of the Tang Son of Heaven, something impossible while she was a mortal woman and concubine. While tragic, Luo's reading and this change of spiritual status brings a degree of equity to a fundamentally uneven relationship and social dynamic (Chudnow 2017, 100-1). However, this type of resolution on the ontological and soteriological level is complicated in the Japanese environment of Zenchiku's *Yōkihi* due to several mitigating socio-religious factors, as will be examined below.

When discussing the *honzetsu* utilised by Zenchiku for composition of *Yōkihi*, it is critical to note that Bai's original Chinese-language poem was not the primary source. As will be examined further, this plays a significant role in forming the soteriological status of the work's Yang persona (Chudnow 2017, 102). There is a strong possibility via textual analysis that *Yōhiki*'s primary *honzetsu* were summaries of *Changhenge*'s plot and descriptions of the Yang poetic mythos as described within medieval commentaries on *Genji monogatari* (Itō 1988, 508). Here again the deep connection between the reception of Bai's poem with *Genji* in the medieval era becomes evident but raises another question: why would a nō playwright like Zenchiku be using medieval commentaries as a source for textual composition?

By Zenchiku's lifetime in the mid-fifteenth century, the Heian period language of *Genji* was increasingly remote to all but elite specialists of the text. Furthermore, knowledge of *Genji* tended toward being based on *renga* (linked verse) handbooks, plot digests, and general summaries rather than the original text.<sup>6</sup> As the popularity of *renga* spread, literate people of disparate social and education levels sought knowledge of the poetic tradition necessary for composition. Therefore, simplified plot digests and *renga* linking manuals were produced to accommodate their interests. For example, *Genji* digests written in contemporary language contained short chapter summaries, the most famous poems, and brief commentaries were often used as *honzetsu*. Such works were excellent resources for *renga* poets and nō playwrights, who primarily required the following: the general location, dramatic concept, and poetic atmosphere for *Genji* chapters; main characters and their mythos; and critical poems to draw upon as *honkadori* (Goff 1991, 27-8). This type of textual filtering during the medieval era was common, especially for a literary Chinese-language

<sup>6</sup> Please view Goff 1991, 14-29 for a detailed history of *renga* and its influence on nō.

source such a *Changhenge* which would be linguistically inaccessible to the majority of poets and playwrights. It is likely that during Zenchiku's lifetime, Bai's original poem was mostly known from its intertextual status as a Chinese source text for *Genji*'s opening chapter, "Kiritsubo", which will be examined in detail below.

The dramatic action of *Yōkihi* begins with the Daoist wizard (played by the *waki*) travelling to and then arriving at Penglai. He summarises key points of *Changhenge*'s narrative before inquiring with an inhabitant (*aikyōgen* 間狂言) 7 about Yang. He learns of a female deity residing on Penglai's and visits her palace (represented by a veiled *tsukurimono* 作り物),<sup>8</sup> which contains Yang in her celestial form (the *shite*). The dramatic action in these passages is primarily constructed from a poetic weave of allusions derived from imagery distilled from Bai's original poem. The exchange of the jewelled hairpin and a variant of the iconic 'joined-wing/entwined branches' metaphor is featured prominently (Itō 1988, 409). The wizard announces his departure, causing Yang to be wracked with sorrow and yearning for her previous life. These are expressed through lines that combine nostalgic longing with Buddhist anxiety on reincarnation and impermanence. These emotional passages are dramatically punctuated by the *shite*'s slow *jo no mai* 序ノ舞 (slow tempo dance); the chorus states this is *Geishō ui no kyoku* 霓裳羽衣曲 (Ch. *Nichang yuyi qu*, 'Song of Rainbow Skirts and Feathered Vestments'), the piece Yang danced famously in life (Itō 1988, 412).<sup>9</sup> As will be explored below, Yang's meditation on the process of reincarnation is linked inextricably with her karmic clinging, unique ontological origins based on divergent *honzetsu*, and hybrid sacred space (Chudnow 2017, 103-4).

*Yōkihi* opens with the *waki* reciting a *michiyuki* 道行 (travelling chant). The *michiyuki* is a key passage located in *Dan* 段 1 of the *mugen* nō structure (Quinn 2005, 131), describing travel via famous landmarks and the *waki*'s arrival at his destination (2005, 132). In Zenchiku's play, the *michiyuki* portrays the wizard undertaking a mantic journey via dream-vision, traversing the phenomenological realm to reach Penglai:

<sup>7</sup> Actor specialising in comedic, satirical *kyōgen* 狂言 theatre roles, who also chants the narrative explanatory passages between nō play sections.

<sup>8</sup> *Tsukurimono*, 'built things', the minimal stage props used in nō. They are assembled and taken apart by hand before and after every performance.

<sup>9</sup> Itō's edited version of *Yōkihi* with its accompanying scholarship was primarily utilised for this study. However, I also cross-referenced Sanari's (1982) earlier version as well as Koyama and Satō's (1997) later edit.


This passage also establishes thematic elements essential to the play: the *waki* as ascetic utilising a dream-vision and immediate allusive variation to *Genji monogatari*, grounding the play directly to the poetic mythos of Shikibu's tale. The phrase "weave a traveller's grass pillow" (Itō 1988, 406) contains the overlaid pun "brief sleep" (*karine*  かり寝), indicating the poetic conceit of a resting traveller while also focusing on the ascetic's dream-vision employed to contact the spirit realm. This is predicated upon imagery of a ship sailing through both waves *and* the void of space-time: *namiji o wakete yuku fune no* なみ ぢを分けて行く舟の (Itō 1988, 406) with the phrase *namiji* containing the overlaid poetic meanings of "void" (無) on the single phoneme *na* and "waterways/ocean" (波路) on the entire phrase. This paronomasia is the wizard's mantic journey as he travels to the supernatural shores of Penglai, conceived as a physical location within the void of supernatural space (Chudnow 2017, 105-6).

In Japan, conceptions of Penglai diverged from original Chinese Daoist legends of an isle of immortality, combining with indigenous beliefs in supernatural maritime realms such as Tokoyo 常世 and the *Ryūgūjō* 龍宮城, the Dragon King's Palace (Itō 1988, 406; Nakamura et al. 1989, 731). Diverse definitions of Tokoyo existed in the *honji suijaku* beliefs of premodern Japan. One was as a realm of immortal beings similar to Penglai, although at ocean's floor, similar to the mythical Dragon King's Palace from Chinese myth (Ōno et al. 2002, 934). In other definitions, it became almost a general catch-all 'other realm' for supernatural existence:

The locus of Tokoyo can vary: in a foreign land, under the sea, in heaven, under the ground, or in a place beyond the ocean. Various kami and spirits of ancestors are believed to live there. In ancient times people perceived Tokoyo as an utopia of eternal youth, long life, bounteous wealth, and pleasure across the sea. Tokoyo also implied the land of the dead, the nether worlds of *yomotsu kuni*, or *ne no kuni*. (Nishioka 2012)

With shared features such as the sea, immortality, and the domain of sacred beings, the combination of Penglai with the indigenous subterranean world of Tokoyo may have appeared natural to Muromachi Japanese. The distinct ontological and sacred space resonances of Chinese Penglai and Japanese Tokoyo should be considered when analysing *Yōkihi*'s dramatic action, especially the potential for female enlightenment in both non-Buddhist sites.

The Daoist wizard's *michiyuki* provides more crucial information via an allusion to the "Kiritsubo" chapter of *Genji*. The line "If only there were a wizard who could go and inquire" (Itō 1988, 406) intoned during his *michiyuki* is a direct allusion to "Kiritsubo" from *Genji*. This poem is spoken by the Kiritsubo Emperor while in deep mourning for his recently deceased love, the Kiritsubo Intimate:

たづねゆくまぼろしもがなつてにても魂のありかをそこと知るべく


This poem implicitly refers to *Changhenge* as the Kiritsubo Emperor is directly viewing screen paintings illustrating scenes from Bai's poem at this point in the chapter. The *Changhenge* screen paintings (*byōbu-e* 屏風絵) cited in "Kiritsubo" were commissioned by Emperor Uda 宇多天皇 (867-931, r. 887-897) and documented in the personal collection of early Heian poet Ise 伊勢, who composed a series of poems with these screens as subject (Inukai et al. 1994, 18-19). In this scene from "Kiritsubo", the Kiritsubo Emperor is viewing Uda's screens, reciting Chinese verse (presumably *Changhenge*) and also Ise's poems, obsessing "only on that theme, of which he spoke of day and night" (Abe et al. 1970, 104).

Like Xuanzong, who sends the wizard to Penglai to bring back Yang's hairpin, the Kiritsubo Emperor sends a lady-in-waiting on a journey, this time to the home of the Kiritsubo Intimate's mother. She also returns with a hairpin and a letter with information about his son, Hikaru Genji. He composes the above poem holding his former love's hairpin, almost playacting as Xuanzong while gazing at the screens depicting Yang, seeing the lotus blossoms of Taieki Lake and the willows of Biau in her features (Abe et al. 1970, 111) drawing a direct connection between his sorrow and the imagery of *Changhenge* (Owen 1996, 445). This first section of the chapter concludes with dramatic emphasis directly from the final lines of Bai's poem:

Day and night, he had said that they would fly side by side as two birds who share a pair of wings or be as two trees with branches intertwined, so now the unfulfilled life was full of inextinguishable regret. (Abe et al. 1970, 111; Author's transl.)

This assimilation of *Changhenge* into Heian literature and the *Genji* specifically is critical for understanding the symbiosis of Bai's work with Shikibu's by the Muromachi period. When *Yōkihi* was originally performed, it may have been more closely identified with the *Genji*  mythos and the Heian poetic than the Chinese Tang. This is further confirmed by additional allusive variation to *Genji* throughout the play, such as frequent citation of the "Yūgao" 夕顔 (Evening Face) chapter with its meditations on sorrowful memory as intrinsic to the poetic construction of *Yōkihi* (Matsuoka 2005, 100). This deeply overlaid image of *Genji* within *Yōkihi* illustrates that the play has two primary *honzetsu*: the mythos of *Changhenge* and of *Genji*. As will be explored further, the dynamics of karmic suffering within *rokudō* is the focus of *Yōkihi* rather than the tragic transcendental romance of *Changhenge*.

It has been observed that among Zenchiku's plays, *Yōkihi* creates an atmosphere of complete isolation and focuses on the stasis of the *shite* (Atkins 2006, 174). While the play *Yōkihi* takes poetic cues from the original Bai's poem and the *Genji*, drawing heavily from the *Genji*'s broader theme concerning female karmic suffering due to love relationships and social status. This is emphasised in the following:


In this passage is an allusive variation on *Changhenge*'s famous "birds that fly on shared wing" poetic image. However, in *Yōkihi*, the karmic sin of attachment appears to break this transcendental vow, replacing it with a solitary, loveless bird on one wing and withered branches (Itō 1988, 410).

The line detailing entwined branches is allusive variation on Bai's line of "may we become branches that twine together" and is of special significance to the Yang spirit's soteriological status, as it references the *shorea robusta* trees (Jp. *sara sōju* 沙羅双樹) that withered and turned white upon the death of the historical Buddha, Siddārtha Guatama (Ch. *Shijiamouni*, Jp. *Shakamuni* 釈迦牟尼; Itō 1988, 410, headnote 4). It is uncertain if *Yōkihi* is referring to the Buddha's entry into *nirvāṇa* (Ch. *niepan*, Jp. *nehan* 涅槃, 'extinction: the cessation of suffering'; Nakamura et al. 1989, 647) to indicate enlightenment potentiality for Yang's spirit in Penglai/Tokoyo. However, given the work's overall poetically melancholy tone, this image may be an allusion to the sadness and loss felt upon the Buddha passing, exemplified by even plants turning white to express their mourning. Here, the image of dying vegetation serves as the first Buddhist sign of decay in *Yōkihi* and underscores Yang desolate mourning. In fact, Yang's spirit seems acutely aware of *rokudō* and the inevitability of suffering through endless transmigration:


In this passage, the Yang spirit (voiced simultaneously by *shite* and chorus) invokes the inevitability of endless transmigration (Itō 1988, 411) emphasising that *all* sentient beings dwelling within the Twenty-Five Stages of Existence (Skt. *pañca-viṃśati-bhava*, Ch. *ershiwu you*, Jp. *nijūgōu* 二十五有) <sup>10</sup> are bound to Buddhist Law of Mortality (Ch. *Shengzhe bi mie*, Jp. *Shōja hitsumetu* 生者必滅; Itō 1988, 411)

Yang may be referring to herself, which is clarified as she references the Five Signs of Decay of a Celestial Being (Ch. *tianshang wushuai*, Jp. *tenjō no gosui* 天上五衰). This is *Yōkihi*'s second Buddhist sign of decay: the symptoms when a celestial being (*tenjō*) is approaching death: their flower-crown falling from above their head, heavenly robes becoming filthy, sweating from the armpits, eyes

<sup>10</sup> Three realms of existences that are sub-divided into twenty-five realms where all sentient beings will transmigrate (Itō 1988, 411, headnote 15).

suddenly spinning, and finally, being unable to return to their heavenly realm (Nakamura et al. 1989, 275). She concludes that across the cosmos, from celestial beings in heavens to the denizens of faraway Mount Shumi and the Northern Continent, all will eventually die (Chudnow 2017, 113-14).<sup>11</sup> These Buddhological references on mortality and ephemerality further define the Yang spirit as far removed from *Changhenge*'s divine goddess and closer to the *honji suijaku* worries of medieval women, positioning these issues central to the play's dramatic action.

This is due to the *honji suijaku* problem of rebirth and karmic clinging, as is revealed in the next passage:

[*kuse*] chorus At that time, I also was a celestial being, yet because of a karmic bond, I was temporarily born into the human realm, and was raised in the inner chambers of the Yō clan. (Itō 1988, 412, Author's transl.)

This passage leads the play's *kuse* section. The *kuse* narrative style was pioneered by medieval female *kusemai* 曲舞 performers whose distinctive vocal style was adopted by nō. Performed by the chorus (*ji* 地), the *kuse* is typically where the *shite* reveals their inner psychology and karmic conflict. In *Yōkihi*, Yang reveals the root of her karmic suffering: an undefined karmic bond from a prior existence. She is reborn into the human realm from a celestial existence, with the chain of events involving her relationship with the Tang Emperor resulting in the female-gendered karmic sin of attachment. Paul Atkins has noted that Yang's status as a celestial being was first delineated by nō scholar Wang Donglan who cites the medieval Japanese *kanbun* text *Chōgonka jo* 長恨歌序 (Preface to the Song of Everlasting Sorrow), which uses language referencing karma, and Yang celestial origin and rebirth in the human realm that appears in *Yōkihi* (Wang 1994, 15-16; Atkins 2006, 171). This utilisation of native Japanese reinterpretations of the Yang mythos is corroborated by Itō Masayoshi, who notes texts such as *Chōganka den* 長恨歌伝 (also cited by Wang 1994) along with *Genji*, as major sources for *Yōkihi* (Itō 1988, 411). These native-Japanese *honzetsu* share the concept of Yang as celestial being of divine origin and ontological status suffering from karmic clinging, which complements the socio-religious thematic within the *honzetsu* of *Genji*. Therefore, *Yōkihi*'s Yang spirit

<sup>11</sup> Both mysterious locations cited in the late-Muromachi dictionary *Ikyōshū* 伊京集: Mount Shumi (Skt. *Sumeru* or *Mahāmeru*, Ch. *Xumishan*, Jp. *Shumisen* 須弥山) is surrounded by Four Continents at the cardinal directions and its citizens live to five hundred. People of its satellite Northern Continent (Skt. *Uttarakuru-dvīpa*, Ch. *Beijulu-zhou*, Jp. *Hokkuru-shū* 北倶盧洲) live to one thousand (Itō 1998, 411, headnote 17).

is highly complex, as she appears aware of her past and current ontological and soteriological statuses while also struggling with concepts of mortality and suffering from the karmic sin of attachment within a heaven-like realm.

Following the *shite*'s slow *jo no mae* (a dance piece typical of Third Category plays), the focus on Buddhist decay and ephemerality is transformed to outright sorrow and mourning with the drama's closing lines:

[*nori ji*] chorus As for my Lord, I shall never meet him again in the world… although drifting in an ephemeral world, how I long for the past, and our fleeting parting. The goddess sinks down to the floor of the Palace of Tokoyo, to remain. (Itō 1988, 412; Author's transl.)

These closing lines ground the Yang spirit in the hybrid supernatural realm of Penglai/Tokoyo, where she continues to endure the karmic suffering of attachment. Also of interest is the line, "As for my Lord, I shall never meet him again in the world…", indicating that the transcendental romance achieved in *Changhenge* is not only impossible but further meetings through rebirth may be unrealised. Instead, her attachment and clinging to Xuanzong seems to have bound her to the sacred space of Penglai/Tokoyo. But what of the ascetic wizard who has visited her? Why is he unable to provide aid? The answers may be within the play's rhetorical and dramaturgic structure.

While *Yōkihi*'s thematics and its Yang persona align with the *mugen* nō, the work fundamentally lacks important structural divisions typical of the genre. As previously noted, this is typified by the following: two overall parts of dramatic action, further subdivided into five *dan* of dramatic action, accentuated by a change of costume and persona for the *shite* on their re-emergence in Part II.<sup>12</sup> *Yōkihi*  does not feature this costume or identity transformation for *shite*; rather the Yang persona manifests at Penglai to the *waki* in her true form throughout the entire dramatic action. Additionally, despite occurrence within the supernatural sacred space of Penglai/Tokoyo, the play's dramatic action occurs in linear progress, i.e., 'real time'. While this aligns *Yōkihi* closer to the dramaturgic operation of *genzai* 現在 nō, dramatic action grounded in linear time, categorising the play as purely in the *genzai* dramatic mode is problematic. This is due to major thematic and dramaturgical differences between works classified as *genzai* nō and *Yōkihi*. For example, plays considered as *genzai* nō focus on familial and sociological conflicts pertinent to Muromachi society located in the real world, rather than the super-

<sup>12</sup> Again, refer to the *mugen* nō structure chart (Quinn 2005, 131).

natural or mythical. The genre's masterwork may be the Fourth Category (*yonbanme-mono* 四番目物, Miscellaneous Piece) play *Sumidagawa* 隅田川 (Sumida River) by Zeami's son Kanze Motomasa 観世 元雅 (1394-1401), depicting a mother driven mad with grief searching for her abducted child. With grief and mourning are also major themes of Motomasa's work, the poetic axis centres on mother-child love, with Pure Land Buddhism used as ritual means of assuaging a mother's grief and enacting a child's enlightenment.

This is sharply contrasted in *Yōkihi*, where the Yang persona appears static in karmic clinging while marked by signs of Buddhological decay. However, the play is also missing a critical rhetorical and religious element common to other female-spirit *mugen* nō. This is the *shite*'s request for enlightenment and the ascetic *waki*'s later performance of religious rites for her. These passages are usually located following the *kuse* passage in *Dan* 4 at the conclusion of Part I (*shite*'s request) and the beginning of *Dan* 5 to start Part II (*waki* intones *kuyō*  供養, funerary or memorial services). These passages are missing from *Yōkihi* when cross-referenced with other Zenchiku-attributed *Genji*based female-spirit *mugen* works such as *Tamakazura* 玉鬘 (The Jewelled Chaplet) and *Nonomiya* 野宮 (The Shrine of the Fields), or even Zeami-attributed female-spirit *mugen* nō.13 In these plays, the *shite* directly requests the *waki* to hold a memorial service for their true form (hinted at throughout Part I). The *waki* then frequently utilises the verbal *tomurafu* 弔ふ (to hold a memorial service), the consonant alternate form of *toburafu* (訪ふ/弔ふ; Ōno et al. 2002, 951) to indicate that proper religious rites have been performed. Use of the specific character 弔ふ to write the phrase is also typically employed in *mugen* nō and is explicitly defined as holding *kuyō* to pacify and commemorate the dead (Ōno et al. 2002, 947-8). This reoccurring textual example of providing funerary services for the spirit's enlightenment embedded within the dramaturgic fabric of the plays indicates the intrinsic *honji suijaku* role of the *waki* as ascetic involved in pacification of karmically troubled spirits. This ritual interaction between spirit *shite* and ascetic *waki* is noticeably lacking in *Yōkihi*. Instead, these two interactions are replaced by the following lines in the same textual positions. Rather than the *shite*'s request for enlightenment, there follows:

[*mondō*] shite Truly it is as you have said, my body is fragile as dew, although it seems his August love had you search so far, this visit is painful, like chrysanthemums that wither when scattered, how I resent the wind that may bring news of my love!

<sup>13</sup> This includes plays such as *Izutsu* 井筒 (The Well Curb) and *Matsukaze* 松風 (Pining Wind).

Yet even still, I cry tears of longing,

my spirit disappears as I remember my past life in the human realm. waki I must return quickly and report to the Emperor, with that being the case, may you grant me an August memento [to return with]? (Itō 1988, 408-9, Author's transl.)

This exchange details two critical points. First, the Yang spirit appears extremely bound by the karmic sin of clinging. With her "body is fragile as dew", an ephemeral poetic image that is allusive variation to an Izumi Shikibu 和泉式部 (976-1030) poem that describes the female body both dew and tears (Itō 1988, 409, headnote 13). Usage of this type of imagery was common in Heian women's poetry to describe the insubstantial nature of romantic relationships and their emotional impacts. Additionally, the Yang spirit describes that the *waki*'s visit causes great pain due to causing reminiscences of her former life, even to the point of her own annihilation. Here she again avails upon the imagery of withering foliage, a poetic theme in *Yōkihi* that sharpens its soteriological focuses. Rather than providing enlightenment potentiality, the *waki* is only causing additional karmic pain. Finally, the *waki* himself is only tasked with information gathering: "report to the Emperor" (*sōmon* 奏聞) is the unique phrase. This strictly follows the role of the wizard in the play's *honzetsu*, regardless of origin. Furthermore, the *waki* never performs any religious rites for the Yang spirit. He announces his departure thusly:


It is important to note that the *waki* used no Buddhist language for memorial rituals, as discussed above, although he appears moved by the *shite*'s sorrow. Furthermore, the Yang spirit delays the wizard on his mantic journey back to the human realm by desiring to

<sup>14</sup> Allusive variation to *Manyōshū* 万葉集 (Collection of Ten Thousand Leaves, compiled approx. 759), poem no. 13, describing becoming so thin due to stress over love that the author can wrap their belt or sash (*obi* 帯) around their waist three times (Itō 1988, 410, headnote 7).

recreate the evening's entertainments (*yayū* 夜遊) from her time as Xuanzong's consort. This is manifested by her dance of the "Song of Rainbow Skirts and Feathered Vestments" as a slow *jo no mae*, the main dance piece closing the play.

This lack of these specific female-spirit *mugen* nō rhetorical and dramatic hallmarks, replaced by passages highlighting karmic clinging, may be critical in explaining why the *waki* of Yōkihi is unable to facilitate enlightenment to the *shite* despite his ascetic nature. Another possibility may be the multiple *honzetsu* of the play:


These unique quirks of the text may delineate why the Yang persona is denied enlightenment: her hybrid Daoist/kami worship sacred space lacks the dharmic power required, she does not request or receive the proper ritual means, and her attachment remains too great.

# 4 Conclusion: Inability to Transmigrate or Incomplete Ritual Process?

Zenchiku's play *Yōkihi* closes with intense focus on the soteriological suffering experienced by the *shite*. This stands in direct opposition to Luo's interpretation of Yang's reincarnation as a goddess in Bai's conception of Penglai, which creates a transcendental romance levelling the social and spiritual hierarchies of Tang China. As we have seen, delineating the layers of *honzetsu* in *Yōkihi* places intense focus on Japanese issues such as transmigration and the aesthetic of longing. Also of crucial interest is hybrid Penglai/Tokoyo local and its associated sacred space, along with the role of Daoist wizard as *waki*. As detailed above, the Yang persona seems trapped by the femalegendered sin of karmic attachment, with the combinatory Chinese/ Japanese space of Penglai/Tokoyo functioning more as a netherworld than a heaven realm, making enlightenment problematic. Finally, the Daoist wizard *waki* does not possess the Buddhist knowledge to offer her soteriological aid despite his powerful ascetic ability (Chudnow 2017, 117). Furthermore, he was only tasked by the Tang Emperor with bringing back evidence of her existence, not with releasing her karmic constituents into the void of enlightenment.

#### **Bibliography**


*Genji*). Matsuoka Shinpei (ed.), *ZEAMI – Chūsei no geijūtsu to bunka* ZEAMIー 中世の芸術と文化 (Zeami: Medieval Performative Arts and Culture), vol. 3. Tōkyō: Shinwasha, 89-102.


#### **Images from the Past: Intertextuality in Japanese Premodern Literature** edited by Carolina Negri and Pier Carlo Tommasi

# Evil Women of the Lower Classes **A Study of Tsuruya Nanboku's Use of Chinese Novels in the Kabuki Play** *Osome Hisamatsu ukina no yomiuri*

#### Yamashita Noriko

National Institute of Japanese Literature, Tōkyō, Japan

**Abstract** The kabuki play *Osome Hisamatsu ukina no yomiuri* was written by kabuki playwright Tsuruya Nanboku IV and first performed at the Morita za theatre in Edo (Tōkyō) in 1813. The plot of the play includes a fraud scene with a corpse, which is based on seventeenth-century Chinese popular novel *Jingu qiguan*. One of the features of *Osome Hisamatsu ukina no yomiuri* is that it showcases the attempted fraud by a woman of the lower classes, Dote no Oroku, first performed by *onnagata* actor Iwai Hanshirō V. Oroku belongs to the kabuki type cast known as *akuba*, which realistically depicts the life of women of the lower classes. This type of role was first made popular by *onnagata* actor Iwai Hanshirō IV's performance in 1792, though lead actor Onoe Matsusuke I already performed evil female fraudsters as early as 1789. There is a possibility that the kabuki actors and playwrights were made aware of this particular female image in the Chinese novel by Dutch scholar and writer Morishima Chūryō. This paper discusses the social interactions between Tsuruya Nanboku, Onoe Matsusuke and Morishima Chūryō, and how Iwai Hanshirō V's enacting of Dote no Oroku was influenced by Hanshirō IV's and Matsusuke's evil old women.

**Keywords** Tsuruya Nanboku IV. Kabuki. Chinese novels. Morishima Chūryō.

**Summary** 1 Introduction. – 2 Special Features of the Kabuki Play *Osome Hisamatsu ukina no yomiuri*. – 3 The Plotline of the Fraud in *Osome Hisamatsu ukina no yomiuri*. – 4 Episode 29 in *Jingu qiguan*. – 5 Incorporation of Episode 29 of *Jingu qiguan* in Japanese Novels. – 6 Special Features of *Osome Hisamatsu ukina no yomiuri*'s Fraud Scene. – 7 Tsuruya Nanboku, Onoe Matsusuke and Morishima Chūryō.

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 117** Submitted 2021-12-14 | Accepted 2022-02-08 | Published 2022-08-30 © 2022 Yamashita | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/007**

#### 1 Introduction

It has been argued that Tsuruya Nanboku 鶴屋南北 (1755-1829) based his 1813 play *Osome Hisamatsu ukina no yomiuri* お染久松色読販 (The Latest News of Osome and Hisamatsu's Love Affair) on the play *Daigashira kazumi no iromaku* 台頭霞彩幕 (Rising Mist and Coloured Curtains), written by Sakurada Jisuke II 桜田治助 (1768-1829) and others, performed in the first month the previous year at the Nakamura za theatre (Furuido in Furuido, Ichiko, Noma 1983, 491). Both plays feature a scheming head shop clerk, a vegetable selling farmer who gets involved in a fight, and an unconscious man's body being mistaken for a corpse.

The plots of these plays are based on the 29th episode of a seventeenth century Chinese popular novel (Ch. *huaben*, Jp. *wahon* 話本, *hakuwa shōsetsu* 白話小説) *Jingu qiguan* 今古奇観 (Jp. *Kinko kikan*, Mysterious Spectacles Today and in the Past). This episode has been included in the Japanese translation of Chinese stories *Shōsetsu suigen* 小説粋言 (Novels with Elegant Words) by Confucian scholar and publisher Sawada Issai 沢田一斎 published in 1758, and the plotline was used in the *yomihon* 読本 (novels in eighteenth-nineteenth of Edo era) novel *Sekijō kaki negusa* 席上奇観垣根草 (Mysterious Spectacles at the Banquet – Grass by the Fence) by Sōkan Sanjin 草官山人 in 1770 and again by Kyokutei Bakin 曲亭馬琴 in his 1807 novel *Sono no yuki* 園の 雪 (Snow in the Garden) so it was clearly a widely known plot device.

However, one of the novelty features of *Osome Hisamatsu ukina no yomiuri* is that it showcases the attempted fraud by a woman of the lower classes, Dote no Oroku 土手のお六 (Oroku of the Riverbank), first performed by *onnagata* 女形 (female role specialist) actor Iwai Hanshirō 岩井半四郎 V. Oroku belongs to the kabuki typecast known as *akuba* (悪婆, evil old woman), which realistically depicts the life of women of the lower classes. This type of role was first made popular by *onnagata* actor Iwai Hanshirō IV's performance of *Mikazuki Osen* 三日月おせん (Three Day Moon Osen) in 1792, though lead actor Onoe Matsusuke 尾上松助 I already performed evil female fraudster as early as 1789. There is a possibility that the kabuki actors and playwrights were made aware of this particular female image in the Chinese novel by Dutch scholar and writer Morishima Chūryō 森島 中良. This paper discusses the social interactions between Tsuruya Nanboku, Onoe Matsusuke and Morishima Chūryō, and how Iwai Hanshirō V's enacting of Dote no Oroku was influenced by Hanshirō IV's and Matsusuke's evil old women.

I would like to extend my deep appreciation to everyone at the Senshū University Library who gave me permission to publish the illustrations of the picture book *Osome Hisamatsu ukina no yomiuri* of the Mukai Nobuo Collection.

#### 2 Special Features of the Kabuki Play *Osome Hisamatsu ukina no yomiuri*

One of the prominent features of the play *Osome Hisamatsu ukina no yomiuri* is that lead *onnagata* actor Iwai Hanshirō V played seven roles, and his quick changes between the roles became the showcase of the play **[fig. 1]**. Amongst the acts, the first act at Koume tobacco shop and middle act's extortion scene at the Aburaya pawn shop are considered to be of high dramatic quality (Matsuzaki in Matsuzaki, Urayama 1961, 10). These two acts depict the realities of the lower classes in Edo realistically, a kabuki genre called *kizewa* 生世話. This particular play was not only popular in Edo, but also gained fame at the kabuki theatres in Ōsaka. It was often staged in both cities, and in 1831, eighteen years after the first performance, the script was published with pictures by the publisher Kawachiya Tasuke 河内屋 太助 in Ōsaka **[fig. 2]**.

#### 3 The Plotline of the Fraud in *Osome Hisamatsu ukina no yomiuri*

The outline of the part of the play *Osome Hisamatsu ukina no yomiuri* which is considered to be based on the Chinese novel *Jingu qiguan*, is the fraud scene, and goes as follows.

The head clerk Zenroku of the Aburaya pawn shop in the Kawaramachi quarters in Edo plots to take over the shop by forcing the owner's daughter Osome to marry him. To this purpose, he convinces Osome's brother Tasaburō to steal the certificate of the pawned sword of the Chiba family to get him disowned. Zenroku tricks the shop assistant into going on an errand so that it will look like he's run away, and thereby will be suspected of the theft. Zenroku proceeds to hide the stolen certificate in a bunch of greens the farmer Kyūsaku has brought to sell. Zenroku intends to buy the vegetables of Kyūsaku, but Kyūsaku insists that he has promised the vegetables to another buyer and can't sell them to Zenroku. A fight ensues, in which Kyūsaku is kicked by the pawn shop staff, but as the owner enters, the fight is stopped and Kyūsaku is given compensation for medicine and a new kimono, upon which he goes home **[fig. 3]**.

In the next scene we see Dote no Oroku, who is running a tobacco shop in the Koume quarters. She used to work for a warrior household but eloped with her lover Kimon Kihē. Her former employer, Lord Takegawa, orders her to get her hands on 100 gold *ryō* coins (about 10,000,000 yen) so that he can retrieve the famous sword of the Chiba family from the pawn shop. This is repayment for the 100 *ryō* she and Kibē stole from him when they eloped. Oroku also does odd sewing jobs, and just as she accepts a sewing order from an

**Figures 1-2** Nanboku/Kunisada, *Osome Hisamatsu ukina no yomiuri*. 1831. NIJL Nihon Kotenseki Sōgō D.B., 22.3 × 15.4 cm. Senshū University Library, Kawasaki-city (Kanagawa-ken). © Senshū University

**Yamashita Noriko Evil Women of the Lower Classes**

**Figures 3-4** Nanboku/Kunisada, *Osome Hisamatsu ukina no yomiuri*. 1831. NIJL Nihon Kotenseki Sōgō D.B., 22.3 × 15.4 cm. Senshū University Library, Kawasaki-city (Kanagawa-ken). © Senshū University

**Figure 5** Nanboku/Kunisada, *Osome Hisamatsu ukina no yomiuri*. 1831. NIJL Nihon Kotenseki Sōgō D.B., 22.3 × 15.4 cm. Senshū University Library, Kawasaki-city (Kanagawa-ken). © Senshū University

itinerary hairdresser, Kyūsaku appears, and asks Oroku to fix his torn kimono too. He also asks the hairdresser to fix his topknot, as his hair is in disarray after the fight. He tells the hairdresser about the fight at the pawn shop, while Oroku and Kibē listen intently, thinking of a plan of how to extort money from the pawn shop using Kyūsaku's torn kimono as evidence. After Kyūsaku leaves, Kibē places the corpse of the runaway shop assistant from the first act, who apparently died of blowfish poisoning. Kibē makes a wound on the corpse's face **[fig. 4]**, and after that, has it carried to the Aburaya pawnshop, where Oroku loudly laments that the shop clerks killed her younger brother in a fight and demands compensation. But just then, Kyūsaku, who is supposed to be dead, appears. To that, the apparently dead shop assistant is revived with the help of some medicine and the fraud unravels **[fig. 5]**.

**Figure 6** Mokumokugyōin/Kunisada, *Gekijō ichikan mushimegane*. 1830. NIJL Nihon Kotenseki Sōgō D.B., 22.2 × 15 cm. NIJL, Tachikawa-city (Tōkyō). © NIJL

#### 4 Episode 29 in *Jingu qiguan*

The above storyline is based on episode 29 in the Chinese novel *Jingu qiguan*, which can also be found in the Chinese story collection *Pai'an jingqi* 拍案驚奇 (Jp. *Hakuan kyōki*, Slapping the Table in Amazement), but it is the former work which became popular in Japan (Marui 2019, 13). The outline of episode 29 is as follows:

On a spring day, Wang Jie 王杰 returns drunk to his home in Zhejiang. In front of his house, he finds his servants fighting with a ginger tradesman Lü Da 呂大, who is a very stubborn man. Wang Jie, being drunk, ends up hitting Lü Da, who is unconscious. Wang Jie hurriedly revives him, and after inviting him for a drink and food, gives him some white silk and sends him home. But that evening, ferry man Zhou Si 周四 comes and claims that Lü Da died on his boat, and that Wang Jie is to blame. Wang Jie sends one of his servants to investigate, and indeed the servants find a body. Wang Jie gives 20 gold coins to the ferry man and asks him to bury the corpse during the night. The ferry man wants to have 100 gold coins, but after some haggling, agrees to do it for 60. Wang Jie sends his servant along to help. Later, Wang Jie's daughter falls ill, and Wang Jie tells his servant to go and fetch the doctor. However, the doctor never arrives and Wang Jie's daughter dies. The servant claims the doctor wasn't at home when he called, but later it is revealed that the servant in fact was too busy drinking and never went to fetch the doctor. Upon learning this, the enraged Wang Jie flogs the servant, who out of resentment tells the authorities of Wang Jie's murder and has him arrested. He is tortured and confesses, upon which he is imprisoned and almost dies. But half a year later, the ginger tradesman Lü Da reappears at Wang Jie's house. The servant, thinking he has seen a ghost, flees, but Wang Jie's wife convinces Lü Da to go and explain things to the authorities. In the end, Wang Jie is released, and the boatsman and servant are arrested and executed.

#### 5 Incorporation of Episode 29 of *Jingu qiguan* in Japanese Novels

Chapter four of the *yomihon* novel *Sekijō kikan kakinegusa* (席上奇観 垣根草) is a rewrite of episode 29 in *Jingu qiguan*. It faithfully follows the original plotline but stages them in a Japanese setting, making sure that each villain gets his just punishment as the story unfolds. Kyokutei Bakin, on the other hand, uses the story differently in his 1807 novel *Sono no yuki*. The storyline centres on a revenge drama between princess Usuyuki 薄雪姫 and her brother Sanewaka 実稚. Sanewaka is planning to get his hand on a fortune by marrying off princess Usuyuki to the enemy. As a drunk warrior accosts the Princess, Sanewaka manages to force him off her, but in the process the drunk warrior falls as if he was dead. When the drunk warrior later regains consciousness, he hires a beggar to go and show his clothes to Sanewaka and claim the warrior is dead. By doing so, Sanewaka again hires the beggar to bury the body. The next day, the beggar comes back and demands money not to expose the body in a fraud scene, which structurally overlaps with the fraud scene in *Osome Hisamatsu ukina no yomiuri*.

#### 6 Special Features of *Osome Hisamatsu ukina no yomiuri*'s Fraud Scene

The special character of the fraud scene in *Osome Hisamatsu ukina no yomiuri* lies in the humoristic setting displaying the bold attempt of fraud by a woman of the lower classes, and the coincidences which leads to her failure. Women who commit crime like Dote no Oroku are referred to as *akuba* 悪婆 (evil old woman) as a kabuki role type. This role type was made popular by *onnagata* actor Iwai Hanshirō IV's enacting of *Mikazuki Osen* in the play *Ōfuna mori ebi no kaomise* 大船盛蝦 顔見勢 (Auspicious Face Showing of Shrimps Heaped on the Big Ship), performed in the 11th month, 1792 (Kawatake 1960, 25). *Mikazuki no Osen* was enacted as a real low-class prostitute. The word *akuba* infers the meaning old, but *akuba* doesn't necessarily need to be advanced in years. The root of inferring the age to the word comes from Onoe Matsusuke's enactment of evil old women as enemies. In the 1831 kabuki journal *Gekijō ichikan mushi megane* 戯場一観顕微鏡 (The Theatre Through the Looking Glass), Onoe Matsusuke's career is described as follows: "He started out as an *onnagata* female impersonator, but later started to play sexy villains. When he advanced to be the leading actor of villains (敵役 *kataki yaku*), he started enacting the important villains as old women" (Mokumokugyōin, Utagawa 1973, 331-2).

For example, he is the actor who created the role of an evil lady in waiting Iwafuji appearing in the play *Kagamiyama kokyō no nishiki-e* 加賀見山旧錦絵 (Coloured Picture of the Mirror Mountain of Yore) **[fig. 6]**, repeatedly reenacting this role throughout his lifetime. However, Matsusuke did not only enact plotting high-ranking ladies-inwaiting, but also of the lower classes, acting as partners for villains, tricking and extorting people with the use of clever words. This use of verbal skills makes them different from previous evil women, who had mostly been displayed as murderers, thieves and evil mothers-inlaws. This eloquent scheming woman also appears in several episodes of *Jingu qiguan* (7, 13, 23 and 37). Especially episode 7, "Monopoly of the Oilselling Prostitute", has had a big influence on several Japanese works, being basis for the eloquent old brothel manager, and it is my hypothesis that Matsusuke's acting was inspired by this Chinese novel.

#### 7 Tsuruya Nanboku, Onoe Matsusuke and Morishima Chūryō

Tsuruya Nanboku 鶴屋南北 IV's use of the *Jingu qiguan*-inspired fraud scene probably stems from Nanboku's connection to Dutch scholar Morishima Chūryō 森島中良. This connection is not a direct one, but mediated by Chūryō's teacher, the scholar and playwright Hiraga Gennai (1727-1779), who also had close connections to Nanboku and Matsusuke. The details of these relationships have been clarified by Furuido Hideo (1998, 134; 2018, vol. 1: 154, 681-2; vol. 2, 853; 2020, 24, 41), who, on the basis of the essay collection *Ukiyo zōdan* 浮世雑談 (Various Talks of the Floating World, by Mimasuya Nisōji, 1854), shows that Tsuruya Nanboku took part in the making of a showcase in the shape of a bull for the Ryōgoku area show stalls together with Hiraga Gennai, Tatekawa Enba. Further, he points to that Tsuruya Nanboku himself testifies to receiving material concerning the trouble in the Date family household from 'Lord Katsuragawa', aka Morishima Chūryō, in his essay collection *Fukiyose zōshi* 吹寄草紙 (Collected Writings, 1842). At the same time, Matsusuke was also involved with both Gennai and Chūryō. In the comic novel *Yakusha myōmyō nochi no masayume* 役者妙々後の正夢 (True Dream of Later Day Actors, 1833) by Hanagasa Bunkyō 花笠文京, we read that Matsusuke "was friendly with Hiraga Gennai and often visited the Katsuragawa's estate in Tsukiji, gathering all sorts of unusual information" (Hanagasa 1894, 839).

The Katsuragawa referred to in the text is Morishima Chūryō, who was the second son of leading Dutch scholar and Doctor Katsuragawa Hosan, who was employed by the Tokugawa government. Chūryō started studying the art of fiction writing for Hiraga Gennai at the age of nine, and in 1779, he penned the ningyō jōruri puppet play *Meguro hiyoku tsuka* 驪山比翼塚 (Love Bird's Mound in Meguro). In the story collection *Kogarashi sōshi* 凩草紙 (Leaves in the Cold Wind, 1792), he points out that writers such as Tsuga Teishō 都賀庭鐘 used several Chinese novels such as *Gujin xiaoshuo* 古今小説 (Stories Old and New), *Jingu qiguan, Jingshi tongyan* 警世通言 (Records of Remarkable Things), *Pai'an jingqi* 拍案 驚奇 as his sources, and in the *Kogarashi sōshi* 凩草紙 (Leaves in the Cold Wind, 1792), showing his proficiency in the field of Chinese literature.

In conclusion, Tsuruya Nanboku used episode 29 from the Chinese novel *Jingu qiguan* as source material for the fraudulent scene in the play *Osome Hisamatsu ukina no yomiuri*. Further, Iwai Hanshirō V's enacting of the scheming Dote no Oroku as a contemporary woman of the lower classes were based on two distinct acting traditions, one being the evil women enacted by his predecessor Hanshirō IV, and the other Onoe Matsusuke I's aged female villains. This second tradition was again also probably inspired by Matsusuke's connection to Dutch scholar Morishima Chūryō, through whom the world of kabuki gained access to rare material such as Chinese novels.

### **Bibliography**


**Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# Ghosts in Intertextuality *Sarayashiki* **Between** *wagei* **and Local Tradition**

#### Matilde Mastrangelo

La Sapienza Università di Roma, Italia

**Abstract** *Sarayashiki* (The Dish Mansion) is one of the four best-known Japanese ghost stories with a female character playing the leading role. The analysis will focus on intertextual elements shared by the many variations and adaptations in different genres and periods, looking particularly at the *wagei* theatrical repertory of the Edo period and some elements in local productions of the story. This paper does not seek to examine the origins or the narrative itinerary of the story, instead, it aims to detect some of the more interesting factors of intertextuality, which will allow some speculation on their meanings and literary results.

**Keywords** Sarayashiki. Baba Bunkō. Layers of narration in intertextuality. Katsura Bunji I. San'yūtei Enchō.

**Summary** 1 The Origins of the Story. – 2 *Sarayashiki* in the *wagei* Repertory – *kōdan*, *rakugo*, San'yūtei Enchō. – 3 Ghosts in Intertextuality: Local Productions and Some Further Elements. – 4 Conclusions.

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 129** Submitted 2021-12-13 | Accepted 2022-02-08 | Published 2022-08-30 © 2022 Mastrangelo | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/008**

#### 1 The Origins of the Story

*Sarayashiki* 皿屋敷 (The Dish Mansion), together with *Yotsuya kaidan* 四ツ谷怪談 (The Yotsuya Ghost Story, 1825), *Kaidan Botandōrō* 怪談牡丹灯籠 (The Peony Lantern Ghost Story, ca. 1861-64) and *Shinkei Kasane-ga-fuchi* 真景累ヶ淵 (The True Vision of Kasane Swamp, 1859), is one of the four best-known Japanese ghost stories with a female character in the leading role.

The most popular part of the story, which tells of the mistreated Okiku お菊 appearing as a ghost from an old well, desperately counting dishes but never able to find and count the tenth dish, whose loss is one of the causes of her murder, is, of course, present in all the variations. The shifting elements of the story are the events narrated before and after the ghost's appearance, and it is in the two sections of narrative prologue and epilogue where the most interesting instances of intertextuality are located. Indeed, the prevalent use of Okiku's death and the scene where her ghost materialises, whether in theatre or *ukiyoe* 浮世絵 (woodblock print), can somehow dim the considerable creative process surrounding *Sarayashiki*.

Two main strands are shared in the several adaptations of the story: one produced in Kansai and the other in Kantō, both of which have distinct aspects. Starting with the first performance of the story, as a ningyō jōruri 人形浄瑠璃 (puppet theatre) entitled *Banshū sarayashiki* 播州皿屋敷 (The Dish Mansion at Banshū), played at the Ōsaka Toyotakeza in 1741, we can examine the nature of the plot that preludes the apparition of Okiku's ghost. The story is set in Banshū, in the Harima province, where the Akamatsu 赤松 family have built a castle in Himeji. After a period of decline, Lord Akamatsu's heir Akamatsu Masanori 赤松政範 regains control of the territory and hands the castle over to Lord Kotera Toyomoto 小寺豊職. After a while, Aoyama Tessan 青山鉄山, a man in service with the Kotera family, decides to join a conspiracy against them. Some of the loyal retainers of the Koteras ask a servant, Okiku, one of the orphan daughters of a retainer, to act as a spy, but Aoyama's conspiracy ultimately succeeds. Once Aoyama becomes lord of the castle, strange ghostly phenomena begin, such as the manifestation of Osakabe hime 刑部姫, a female divinity and guardian of the castle, a figure unique to this adaptation. Alarmed by these phenomena, Aoyama decides to live outside the castle in a residence, a *yashiki* 屋敷, taking Okiku with him, but once he discovers Okiku's role as a spy, he plots to kill her. His retainer, Chōnotsubo Danshirō 町坪弾四朗, however, asks to deal with the matter himself, although his true purpose is to force Okiku into marriage. The girl rejects him, and, in revenge, he steals one of the ten precious dishes that have been entrusted to Okiku. Aoyama asks Danshirō to punish Okiku, and once again, he tries to force the girl to yield to him, but after her firm refusal, he tortures and kills her before throwing her into a well. Fear breaks out in the *yashiki* upon the apparition of Okiku's ghost, who is trying to count the dishes (*hitotsu* 一つ, *futatsu* 二つ, and so on), to the sound of broken crockery. In this adaptation, Okiku's sisters avenge her, killing Danshirō and bringing the box of dishes for the ghost to see, thus appeasing her.

### 2 *Sarayashiki* in the *wagei* Repertory – *kōdan*, *rakugo*, San'yūtei Enchō

### **2.1** *Kōdan*

Some scholars have noted that Akamatsu, the name of the lord who builds the castle, is also the name of the first known *kōdan* 講談 (historical narration) performer, Akamatsu Hōin 赤松法院 (1688-1704) straddling legend and history. In 1758, *kōdanshi* 講談師 (master of *kōdan*) Baba Bunkō 馬場文耕 (1718-1759) adopted the genre of oral narrative to produce his own variation of the story, *Sarayashiki bengiroku* 皿屋敷辨疑録 (The Sad and Suspicious Chronicle of the Dish Mansion), which became the standard reference plot for many variations set in Edo from the eighteenth century onwards, most of which bore the title *Banchō sarayashiki* 番町皿屋敷 (The Dish Mansion at Banchō) (Konita 1987, 49). Baba Bunkō's *Sarayashiki bengiroku* is a *jitsuroku* 実録 (a documented story), a genre of *kōdan* that combines real and invented facts in a plot where two main storylines intersect. The story is organised into ten chapters, and Okiku appears from the fifth one onwards. *Sara* 皿 and *yashiki* 屋敷 are the pivotal words in Bunko's story since everything begins with the allocation of the territories around Edo castle under *shōgun* Tokugawa Ieyasu 徳川家康 (1543-1616). It is widely known that the most faithful *hatamoto* 旗本 (retainer of the *shogun*) were distributed in six circles, or *bangumi* 番組, located in districts (*banchō* 番町), and this may be one of the origins of the first part of the title *Banchō sarayashiki*. At the time of *shōgun* 将軍 Tokugawa Iemitsu 徳川家光 (1604-1651), the chief of the *shōgun*'s retainers, Yoshida Daizen no Nosuke 吉田大膳亮, was transferred elsewhere, and his estate became an empty land, *akichi* 空き地 or *sara-yashiki* 皿屋敷, a place (and a residence) that can be used 'once more', *sara ni* さらに, so here we have a wordplay based on the homophony of *sara*. The land was given to *shōgun* Iemitsu's sister Tenjuin 天樹院 (1597-1666), presented as follows:

As was widely known, she would shower her attention on many men. Any young samurai, or any handsome young man even from the lower classes passing by the gate of her residence at Kōjimachi, would be invited in by seductive maids, and many a lustful guest accepted the invitation. This is why a children's rhyme of the time ran, 'If someone passes by the Yoshida residence, he will surely be invited in'. The song was already famous before then, and at that time became so popular that old people would memorise and sing it. Even in the kabuki repertory, I know that about forty years ago, the late lamented actor Yamanaka Heikurō 山中平九郎 (1642- 1724) played the role of Yoshida's widow in the play *Ushi no gozen*  牛の御前 (Ushi no gozen) that tells of a gang invited into the residence. (Baba 1929, 4)<sup>1</sup>

Baba Bunkō adds several elements, quoting true and false sources about Tenjuin and the legend around her (Hirosaka's notes in Yokoyama et al. 2015, 18), using references to popular songs and kabuki 歌舞伎 to weave his intertextual web. Yamanaka Heikurō was actually a kabuki actor famous for his terrifying roles as ghosts and vengeful spirits. Baba Bunkō mentions him in his *Kinsei Edo chomonshū* 近世江都著聞集 (Early Modern Stories Heard in Edo, 1757), quoting the play *Tsuma goi Sumidagawa* 嬬恋隅田川 (The Lovely Couple at Sumidagawa) in order to associate some other frightening images to Tenjuin (Hirosaka 2015) in an intriguing intertextual game.

Tenjuin certainly deserves her scary description. In a fit of jealousy, she tortures and kills her servant Takeo 竹尾, guilty of having an affair with her lover Hanai Iki 花井壱岐. She then kills Hanai and throws them both down the old Kojimachi well, which came to symbolise suspicious disappearance and death. From then on, she uses the same old well to eliminate any person or lover who opposes her, turning the house into a place of terror. After the death of Tenjuin, strange phenomena appear from the well:

In the middle of a drizzly night, a blue, bright flame appeared and disappeared from the well many times. People passing by were amazed to see it, and after a while, everybody said it was a haunted house. (Baba 1929, 6)

For a while, the land remained uninhabited, also due to the need for accessible spaces for the construction of mansions in Edo; it would be used again in the future as it was still a *sara-yashiki*. The estate was divided between three lords, and the Kojimachi well was buried. The *kanji* 漢字 for *sara* 皿 in Baba Bunkō's title *Sarayashiki bengiroku* is, of course, the one referring to 'dish', but the underlying meaning of 'once more' is characteristic of the Kantō variation of the story. A land that is divided and reused with newly constructed houses is a place that can somehow be 'deconstructed' and 're-written'. In this sense, I think we can actually consider *sarayashiki* an 'intertextuali-

<sup>1</sup> All translations of the proposed passages are by the author of the contribution.

ty *yashiki*', a highly explanatory model of intertextuality where many layers of memory and meanings overlap time after time, like the numerous lovers thrown into the well by Tenjuin. It is split into several 'fields' of literary inspiration, spawning new literary phenomena and adaptations in a process typical of intertextuality.

The *sara-yashiki* we are discussing, one of the three estates formed by Tenjuin's property division with a new well, becomes the official residence of Aoyama Shuzen 青山主膳, the classic villain of *jitsuroku* stories. This character is not based on a historical figure, but the name is another example of intertextuality; in *Banshū sarayashiki* too, as we saw, the role of villain is ascribed to a man named Aoyama. In *Sarayashiki bengiroku*, he is actually a *machibugyō* 町奉行 (magistrate), but his approach to the administration of justice is inappropriate as he is aggressive towards the weak. He arrests Mukōzaki Jinnai 向崎甚内, a famous swordsman and a famous thief and gambler, which leads him to fall prey to Aoyama, who jails him and puts him to death. His daughter Okiku becomes the Aoyama family's slave, and – with no rights or possibility of defending herself from Aoyama's insolence and his wife's jealousy – every moment of her pitiful life is lived in fear.

Naturally, Aoyama Shuzen's wife imitated her husband's violent behaviour. […] She was cruel to her servants and was also very jealous. Since the recently hired maid Kiku was so beautiful, and suspecting that her husband was fond of her, she nurtured constant hatred for the maid, punishing her harshly for nothing, making her suffer awfully. […] There was not a day Kiku lived without soaking her sleeves with her tears, suffering as if she had been exiled to Kikaigashima. (Baba 1929, 13)

Kikaigashima is a highly cultured reference to the traditional Japanese theatrical world, with famous plays such as the nō *Shunkan* 俊寛 (Shunkan) or the *jōruri* 浄瑠璃 *Heike nyogo no shima* 平家女護島 (The Heike and the Island of women), which shows the breadth of the *kōdan* audiences' literary knowledge.

One day, Okiku breaks one of the dishes from the precious family collection and is tortured; interestingly, the dish is not stolen or broken by others, as in other variations. Her middle finger is cut off; as Shuzen's wife says: "Wretched woman! You carelessly broke one of my precious dishes, and now my collection is incomplete, so I'm going to cut one off one of your fingers, and your ten fingers will be incomplete" (Baba 1929, 15). Before throwing herself to her death in the old well, she curses the baby that Aoyama's wife is about to bear, and the child is, in fact, born missing a middle finger. From that moment, as in all the other variations, Okiku's tearful voice counting the dishes, "*hitotsu, futatsu*…", counting also her *fingers* in this version, scares all the servants or visitors in the night, and they all decide to leave the house.

From that moment on, every night, from the hour of the rat until morning, a light would rise from the bottom of the well together with a woman's sorrowful voice counting "one, two, three, four", and then "eight, nine", sadly and as though grief-stricken, bemoaning the absence of one dish. It was terrifying. No one dared doubt that it was Kiku's ghost, and all the people living at the Aoyama residence were quite afraid, with their hair standing on end […] In the end, none, neither maids nor retainers, remained in service there. (Baba 1929, 18)

Thanks to Okiku's revenge, Aoyama's bad reputation finally became known to the *bakufu* 幕府 (the shogunate government), causing the downfall of his family. The authorities asked the famous monk, Mikazuki Ryōyo Shōnin 三日月了誉上人, a character who recalls the famous fourteenth-century monk Ryōyo Shōgei 了誉聖冏 (1341-1420), to placate the ghost, and – we might add – to allow a precious piece of land in Edo, now no more than a wasteland, *nohara* 野原, to be used once more, *sara ni*. Narrating some of the cases in which Ryōyo goes to emancipate the vengeful soul of a ghost, reference is also made to the story of Kasane, a female ghost who widely inspired the *wagei* 話芸 repertory until the famous San'yūtei Enchō's 三遊亭圓朝 (1839- 1900) *kaidanbanashi* 怪談話 *Shinkei Kasane ga fuchi* 真景累ヶ淵 (Mastrangelo 2012). In a final scene shared with other works, the monk listens to Okiku's voice, but in the end, he says the number ten out loud, and the ghost finally finds relief and disappears.

There was a reason why the monk made up for the missing number ten. In fact, it is written in the *sūtra* that if the body has some physical defect, one cannot become a Buddha, and this was no doubt why Kiku mourned her severed tenth finger. There was a missing finger as well as a missing dish, but luckily, by saying the number ten aloud, the monk used his gift to placate her soul. (Baba 1929, 21)

Another interesting point in this variant is the inclusion of a character who is also the main character of another *kōdan*. Okiku's father, named Kōsaka Jinnai 高坂甚内, is, in fact, a protagonist in another *kōdan* belonging to the thief-story repertory (Takarai 1971, 135, 148‑50), but it has a different title depending on the period and storytellers. Bunkō was famous for being sentenced to death for his narration, which perhaps sought to underline the complex social makeup within which corruption is present at all levels of society and harshly prevails over morals. He seems to suggest a doubt as to where the difference between a thief and the authority who gains advantage from his position might lie or whether the moral strength of a thief is not greater than that of a corrupt official.<sup>2</sup> The *kōdan* engages with political issues through history, calling into question the *shōgun* family and overbearing, bullying public officials. Being a *jitsuroku*, he remains attached to reality and political situations. In Bunkō's case, intertextuality can be found in his narrative works and *wagei* repertory but also in the number of quotations related to the world of theatre. The numerous *kōdan* variations bearing the title *Banchō sarayashiki* that sprang up under Bunkō's influence and developed basically along the same plot lines are often very long. They offer much greater opportunities to enrich the text with intertextual references, providing a most fascinating investigation of external references, even if the part of the story featuring Okiku appears only from the second half onwards, although her portrait is well detailed (among many Baba 1955; Kyokudō 2006).

### **2.2** *Rakugo*

Returning to the Kamigata area, *Sarayashiki* also became an inspiring plot in the *rakugo* 落語 (comical stories) repertory. The storyteller attributed with the creation of this variation is Katsura Bunji 桂文治 I (1773-1815), who, with his master Matsuda Yasuke 松田弥助 (1781- 1801), started performances in their little theatre, making their art a form of pure entertainment (Hida 1971, 101-2). Katsura was also one of the first *rakugo* storytellers to collect and print the stories he told on stage. An important collection is *Otoshi banashi Katsura no hana* 落とし噺桂花 (Katsura's Best Comic Stories), in two volumes and containing 600 titles, reprinted in *Kamigata hanashibon shū* 上方話本集 (Kamigata Comic Stories Collection). The plot of *Sarayashiki Okiku ga yūrei* 皿屋敷お菊が幽霊 (The Ghost of Okiku at Sarayashiki), as it was originally known – in today's repertory often referred to as *Okiku no sara* お菊の皿 (Okiku's dish) or *Sarayashiki* –, takes place in Banshū where the maid with a leading family broke one of ten dishes making up a collection, in an opening similar to other narratives. However, in *rakugo* the sad voice of the girl that counts dishes "*ichi mai, ni mai*" does not scare people away, and the well from which the voice comes out becomes a tourist attraction, like the haunted house in a modern amusement park. Many people gather in front of the residence, stands are built, an entrance fee is fixed – turning the place into the setting for a show – and *bentō* and drinks are sold.

<sup>2</sup> Farge (2016, 56-9) argues that another reason why Bunkō describes Mukōzaki Jinnai as 'a holy sage' is to present him as a Christian martyr in order to be offensive towards civil authorities and Buddhist establishment.

What makes the 'performance' exciting is that it is said that anyone who hears the ghost count until the end will die; this is why people listen to the ghostly voice get ready to escape before the ninth dish.


The *rakugo* derives from *Banshu sarayashiki* but was partly influenced by kabuki productions too, and the novelty of this *rakugo* seems to be the new way the dishes are counted: "*ichi mai, ni mai*". The final punchline, *ochi* 落ち, of this *rakugo* makes the ghost similar to the audience, with human habits, smart enough to find a way to take free time next day, and the coexistence of a ghost in the human world gives the story a comic turn (Nagashima 2006, 106-7). In the modern *rakugo* repertory, the reason the ghost decides to take a day off is sometimes a cold or the need to have a rest.

I found the *rakugo* variation quite amusing, and I think we can call it a sort of 'reverse intertextuality': in the world of *rakugo*, where everything can be different and opposite, there is no need for a monk to free the ghost as the ghost manages to free herself by choosing to count the nine dishes twice, and at the same time, she can mock the way humans can even use fear for their own gain, affirming the right not only to take a rest but also to leave behind an old intertextual subject to suggest a hint for a new possible line of intertextuality.

#### **2.3 San'yūtei Enchō**

The famous storyteller San'yūtei Enchō, active in Kantō, who, as Katsura Bunji I, has the merit of having definitively and successfully introduced *wagei* into the written literary repertoire, offers intertextual links with the traditional repertory through his own variation *Kiku moyō Sarayama kidan* 菊模様皿山奇談 (The Suspicious Story of Sarayama and the Chrysanthemum Pattern). Enchō produced this story in 1860, and it was performed again years later and published in 1871. At the beginning of the first *seki* 席 (episode or performance), the storyteller narrates that in a moment of creative crisis writing new episodes for newspapers, he had been repeatedly asked to produce a text: "Something old might be good too". He then wrote *Kiku moyō Sarayama kidan*, composed as a narration complete with stage sounds (*shibai narimono iri* 芝居鳴り物入り) at the age of twenty-two, when he was still "a good and beautiful young man" (San'yūtei 2014, 115). The story starts with the presentation of *sara*, part of a treasure collection made up of thirty dishes, of which only nine remain. They belong to the Higashiyama 東山 family, a powerful lineage descended from shogun Ashikaga Yoshimasa 足利義政 (1436-1490). In this variation, the homophone *sara* さら is shared with the name of the place, Sarayama 皿山 (now the northern part of the Okayama Prefecture), where the story is set, and with which Enchō creates an original and interesting intertextual link. In fact, he quotes a *kyōka* 狂歌 (comic poem) without specifying whether it is his own creation, but it is quoted as if it were a well-known poem. It parodies a poem from *Kokinshū* 古今集 (Collection of Poems Ancient and Modern, ca. 905): *Mimasaka ya | Kume no Sarayama | sarasara ni | wa ga na wa tateji | yorozuyo made ni* 美作や久米の佐良山さらさらにわが名はたてじよろづ世 までに (In Mimasaka | Sara Mountain of Kume | towers but never | in ten thousand long years will | I ever let my name rise) (Rodd, Henkenius 1984, 369, no. 1087). In the *Kokinshū* poem, the name of the place anticipates the sound of *sarasara ni* さらさらに (never ever).<sup>3</sup> The *kyōka* quoted by Enchō is: *Mimasaka ya | Kume no Sarayama | sarahodo no | manako de mitemo | minokoshita uta* 美作や粂の皿山皿ほど の眼で見ても見のこした歌 (In Mimasaka | Dish Mountain of Kume | although I open my eyes wide | to look round like a dish | I didn't notice that poem) (San'yūtei 2014, 115). The *uta* 歌 (poem) that the author of the *kyōka* overlooked might also be the *Kokinshū* one, in a reference thought to be easily understandable to an audience, who, despite their literary background, had to catch the joke quickly.

The narration goes on, introducing the actual dishes and the story around them before introducing the characters.

Among these dishes were some called white chrysanthemum (*shiragiku* 白菊) because of the chrysanthemum pattern on them; ten more were called wild chrysanthemum (*nogiku* 野菊) as they had the same charm as the flower. The dishes were a great treasure handed down by the ancestors of the Higashiyama family and had therefore been well looked after. Along with the dishes, the forebears had left a testament in which they wrote that anyone who

<sup>3</sup> In the Italian translation, the interesting solution is "mai e poi mai" (never and ever) in order to keep the repetition of the sound *sarasara ni* (Sagiyama 2000, 647).

damaged them should have their finger severed, even if he or she was a family member. A foolish idea, but a long time ago, there were many things like that. (San'yūtei 2014, 116)

The threat of a severed finger is a reference taken from Baba Bunkō's *Sarayashiki bengiroku* and thus from the same world of *wagei*. It has been noted that in some kabuki and jōruri plays there is also the threat of losing a finger as a punishment for damaging objects or trees (San'yūtei 2014, 438).

Chiyo 千代 is the female main character of the first part, and among the numerous characters there is also Okiku, but she has nothing to do with dishes or ghosts; she is a poor girl taking care of her sick mother. In order to buy medicines for her, she decides to work for Lord Higashiyama Sakuzaemon's 東山作左衛門 family, even if the conditions (broken dish/severed finger) are frightening. Chiyo's good behaviour causes no problems until Sakuzaemon's son Chōsuke 長助 tries to seduce her. Disappointed by her refusal, the young man waits until the *sekku* 節句 (seasonal festival) of the ninth month to take his revenge. This was the time his father used to proudly show the precious collection to his guests. On this occasion, Chōsuke accused Chiyo of breaking one of the dishes and asked his father to punish her not only by cutting off her finger but also by scarring her face, so it would become difficult for her to find a husband. The girl was desperate, but, unexpectedly, another attendant, Gonroku 権六, comes to her rescue, trying to convince Sakuzaemon through persuasive means. Firstly, he says that he was the one who had broken the dish; then he smashed all twenty dishes in the box with the following admonition:

My Lord, what do you think these dishes are made of? From clay, I think. What a stupid testament your ancestor wrote, ordering the amputation of fingers or feet to injure human beings, who are so important, just for some careless breakage […] I felt so sorry for her having her finger cut off that I thought I would save twenty people with my death alone if I broke all twenty dishes. […] If it became known that the Higashiyama family are so cruel as to consider a lump of earth and human beings on the same level, this would also damage the reputation of *shōgun* 将軍 Yoshimasa, and I don't think you, my Lord, would be happy about that. I wanted to avoid it, and I could not think of anything else but to eliminate the root of the problem by breaking the dishes. (San'yūtei 2014, 147, 149, 150)

Appreciating Gonroku's courage and loyalty, Sakuzaemon forgives him. Chōsuke admits that he broke the dish out of revenge for Chiyo's refusal, and at the end of the first part, the Higashiyamas allow Gonroku and Chiyo to marry. The story is then set in Edo and continues amid family struggles, moving away from the usual *Sarayashiki* plot.

As always in intertextuality, some elements are taken from previous stories and transformed, such as the vulnerability of the frail female main character, but in Enchō's case, the intertextual references are also new, like the quotation from *Kokinshū* poem. The dish is present in many variations, but in Enchō's story, the wordplay on *sara* takes on new meanings, and the narrative trajectories move away from ghost stories to that of the *oiesōdō* お家騒動 (family trouble). What is also very interesting in *Kiku moyō Sarayama kidan* is Enchō's original interpretation, which takes the story away from the traditional narrative pattern: the girl he depicts is defended and protected without becoming a vengeful ghost.

#### 3 Ghosts in Intertextuality: Local Productions and Some Further Elements

Of course, the kabuki or jōruri variations are very interesting, with many specificities, and we can observe that the Kansai and Kantō titles are particularly different in terms of the environments they narrate. In Kansai, the story very often tells of plots, *oiesōdō*, and power struggles, while Kantō tends to feature corrupt officials abusing their power over others. In Kansai, the ghost aims to show loyalty and fidelity and, perhaps, to take revenge. In Kantō, the ghost aims to avenge an ill-used woman who throws herself into a well and also to avenge the defenceless, under the thumb of crooked officials. Since the kabuki and jōruri variations have several characteristics related to the genre, I will not include them in this article.

Instead, I think it would be more germane to analyse some elements of the local productions that have been formed orally, namely a narrative typology closer to *wagei*. The spread of local Okiku stories throughout various places in Japan is remarkable, but although Okiku is a vengeful ghost, people seem more inclined to commemorate her sad story than be frightened of her. Perhaps this might be related to the theory that the local popularity of the story may be explained by the presence of the Buddhist monk who, at the end of the story, is the only person able to placate the ghost; this simplified the narration of the story, whose aim, in particular in some genres like *kōdan*, was to highlight the greatness of Buddhism. Looking at the collection of thirty-five local variations compiled by Itō Atsushi (2002), the interesting point is the presence of the well in only ten legends, mainly in the Shimane Prefecture, although the voice counting dishes is always present, and the servant (going by different names) is always the victim of a bullying lord. Water, a very familiar element to ghosts, is never absent, as the main character throws herself, or falls, into a river or a pond, and so on. Another element is that very often, the woman is neither killed nor tortured but commits suicide.

As we have seen, in a story like *Sarayashiki*, elements of intertextuality from previous stories may be found, as well as elements of the story that will become intertextual in later works. The elements which I find very often become the object of intertextuality in *Sarayashiki* are the dish, with an interesting use of homophony, as we have shown, and then the well and the monk, all of which can be seen as marking intertextuality with previous works, and, being inserted into *Sarayashiki*, became new sparks for future intertextuality. The well is full of meaning; as we know, it is very often found in ghost stories, and, of course, it is one of the places servants need to go to for their household tasks. From the numerous examples, I think that one of the most interesting intertextual links could be with the nō play *Izutsu* 井筒 (The Wooden Frame Around the Well), where the main character recalls her love for her husband, telling a monk how they used to play by the well as children; she then disappears, and the villagers tell the monk that the woman was, in fact, a ghost and ask him to pray for her soul. The role of the well is not related to ghostly apparitions in the same way as it is in the Edo ghost repertories, of course, but it is a way of bringing the past to light and connecting the human sphere with the other world. In a modern *kōdan* adaptation, it is a monk, and not Shōnin, who meets a woman who offers him a delicious meal and a place to stay for the night, but the following day he discovers that the lovely woman was Okiku's ghost (Kyokudō 2006, 162-5), following a pattern similar to the nō repertory.

Regarding the monk figure, Japanese ghost stories of the Edo period, for example, typically employ this figure essentially to protect the person tormented by the ghost, suggesting he or she should recite a sutra or hang a sutra strip at the windows; otherwise, they should use some powerful Buddhist talisman. In *Sarayashiki*, on the other hand, the monk's ultimate role is to help the ghost leave the human plane and all the rancour that still binds it to the worldly plane behind, as in the nō play; of course, the monk's action also brings relief to people who are frightened by the ghost or feel sorry for her – as in *Banshū sarayashiki* – but not to the one who caused her death: the aim is to free the ghost and perhaps also the land.

#### 4 Conclusions

The adaptation of a story across numerous genres and media is, of course, not peculiar to *Sarayashiki*; indeed, it is a widespread cultural phenomenon in Japan. Starting from the eighteenth century, we can identify many titles and genres, ranging from theatre to *ukiyoe*, from narrative to film, including the famous *Ringu* リング (*Ring*, 1998 by Nakata Hideo 中田秀雄), all of which can in some way be considered derivations of *Sarayashiki*. As I mentioned earlier, more than to detect changes in the story in the adaptations, I have aimed to find traces of intertextuality relating to *Sarayashiki*, especially in the world of *wagei*. This repertory can be considered the starting point of the Edo variations of *Sarayashiki*. Nevertheless, it is a repertory with different solutions, not only as narrative choices but one that offers fresh takes on the plot: the ghost makes fun of the living in *rakugo*; values are restored and human life is more important than a dish, however precious it may be, in Enchō's version. Of course, it is essential to consider that the fortune of the stories analysed also depends on the fame of the three storytellers Baba Bunkō, Katsura Bunji I, and San'yūtei Enchō. It would not be correct to affirm that the last two narrative choices have prevailed over later adaptations, perhaps also because the effects of ghost stories on stage are difficult to set aside, given their utility in engaging the public. Both seem to lead *Sarayashiki* towards modernity, with the two main female characters becoming freer and less vengeful than other Okiku stories. The *wagei* repertoire shows how intertextuality can provide a boost to creativity.

#### **Bibliography**


**Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# *Nihon ryōiki*: Its "Letters", and their Relevance for the History of Literature

#### Kōno Kimiko

Waseda University, Tōkyō, Japan

**Abstract** Compiled in the early ninth century, *Nihon ryōiki* is a collection of accounts involving karmic retribution and miracles. The compilation was intended to guide readers on the Buddhist path. Much of the phraseology in *Nihon ryōiki* was quoted from or influenced by texts from the Chinese Buddhist canon or non-Buddhist Chinese classics. The text overall reveals a metadiscourse that associates certain personages with a saintly lineage, embodying the ideal form of the human being. However, this metadiscourse was not always faithfully recreated in subsequent writings that drew upon the text. This paper dwells on the classification of books like the *Nihon ryōiki*, which are today known as *setsuwa* anthologies. From an intertextual perspective, the study reexamines the relation between *Nihon ryōiki* and other texts, and discusses its historical significance. In light of this discussion, this article attempts to reevaluate the nature of ancient texts written in Chinese characters and the study thereof.

**Keywords** Nihon ryōiki. Inner scriptures. Outer writings. Text. Book indexes. Intertextuality.

**Summary** 1 Introduction: *Nihon ryōiki*. – 2 The Textual Structure of *Nihon ryōiki*: Inner Scriptures and Outer Writings. – 2.1 Drawing from the Inner Scriptures: The *Lotus Sutra*. – 2.2 Drawing from the Outer Writings: *Mao's Book of Songs*, *Selections of Refined Literature*. – 3 How was *Nihon ryōiki* Read? – 3.1 *Nihon ryōiki*'s Metadiscourse and the Ideal it Presents. – 3.2 How *Nihon ryōiki* Was Reproduced in Heian-Era Texts. – 4 How *Nihon ryōiki* Was Reproduced in Medieval and Early Modern Texts: Book Indexes, Kariya Ekisai. – 5 Conclusion: *Nihon ryōiki* and Intertextuality.

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 143** Submitted 2021-12-14 | Accepted 2022-03-21 | Published 2022-08-30 © 2022 Kōno | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/009**

#### 1 Introduction: *Nihon ryōiki*

Ancient literature teems with mystery and mystique, and this is the case of the source discussed in this article, the *Nihonkoku genpō zen'aku ryōiki* 日本国現報善悪霊異記 (Miraculous Stories of the Reward of Good and Evil from the Country of Japan), commonly known (and henceforth referred to) as *Nihon ryōiki* 日本霊異記. *Nihon ryōiki* contains untapped insights that warrant detailed study from multiple angles. Such studies could reveal, for example, the relation between the text and other Sinosphere texts (ancient East Asian texts written in Chinese characters), the nature of Japanese lexicography of that time, religious and historical perspectives reflected in the text, and its reflections about nationhood and myth. Although *Nihon ryōiki* has been extensively studied,<sup>1</sup> an intertextual approach – the theme of this volume – promises to unlock further insights and poses some intriguing, even if challenging, questions. Having personally researched *Nihon ryōiki* over the years, I use this article to reexamine my findings through an intertextual lens. In this way, I aim to clarify the value of intertextuality and cast fresh light on *Nihon ryōiki*'s idiosyncrasies.

The *Nihon ryōiki* was compiled in the early ninth century by Kyōkai 景戒 (also read Keikai, dates unknown), a *shamon* 沙門 (Buddhist monk) of Yakushi-ji. It consists of three volumes with a total of 116 folktales known as *setsuwa* 説話, which are arranged mostly in chronological order. Kyōkai selected them from the oral legends he had heard, rather than creating the tales himself (*Nihon ryōiki*, vol. 3, epilogue). The question of textualising and editing these oral traditions is relevant to the issues of intertextuality.

How is *Nihon ryōiki* generally defined today? One Japanese dictionary defines it as, "[t]he oldest collection of Buddhist *setsuwa* in Japan" (Izumoji 1984). Another dictionary gives a similar definition: "[t]he first collection of *setsuwa* in Japan" (Ikegami 1990). Such definitions are consistent with Kyōkai's claims in the preface. Specifically, he stated that, whereas China had texts such as the *Record of Invisible Works of Karmic Retribution* (Ch. *Mingbaoji*, Jp. *Myōhōki* 冥報記) – that is, texts containing accounts of karmic retribution for good and evil to guide people to in the Buddhist faith –, Japan had no such texts; hence, *Nihon ryōiki* was to be the first of such texts in Japan.

If we inquire into the matter, we find that the inner, or Buddhist, writings and the outer, or non-Buddhist, writings were first transmitted to Japan in two groups. Both of them came from the country of Paekche, the latter in the reign of Emperor Homuda [Ōjin,

<sup>1</sup> A pioneering study of the *Nihon ryōiki* in English is Nakamura 1973.

r. 270‑312], who resided at the Toyoakira Palace in Karushima, and the former in the reign of Emperor Kinmei [r. 539-571], who resided at the Kanazashi Palace in Shikishima. However, it was customary for those who studied the non-Buddhist writings to denigrate the Buddhist Law, while those who read the Buddhist writings made light of the other works. But they are ignorant and foolish, embracing fatuous beliefs and disbelieving in the consequences of evil or good action. People of true wisdom regard both types of writing with seriousness and have faith in and are fearful of karmic causation. […] If the disposition of good and evil were not known, how could we straighten these tangles and make clear the right and wrong of them? And how, without evoking karmic causation, could we mend evil hearts and advance the path of goodness? Long ago in the land of China, the *Myōhōki* was compiled; and during the great Tang dynasty, the *Hannyagenki* was written. But why should we respect only these records of foreign countries and not credit the miraculous stories that occur in our own land? Since these events occurred here and I saw them with my own eyes, I cannot let them go unrecorded. After pondering them for a long time, I can no longer remain silent. Therefore, I have written down what I have chanced to hear, entitling it *Nihonkoku genpō zen'aku ryōiki* [Miraculous Stories of the Reward of Good and Evil from the Country of Japan], compiling it in these three volumes and handing it down to future times. (*Nihon ryōiki*, vol. 1, preface)<sup>2</sup>

As discussed later, the Chinese Buddhist canon was not the only inspiration behind the *Nihon ryōiki*. Moreover, Kyōkai never envisaged the work as a *setsuwa* anthology, as there was no such genre or literary category as '*setsuwa*' or '*setsuwa* anthology' in Japan at the time. In the preface, Kyōkai refers to the *Myōhōki* – the aforementioned *Record of Invisible Works of Karmic Retribution* – and to the *Hannyagenki* 般若験記, the latter being the abbreviated title of the *Collection of Recorded Miracles Concerning the Diamond Sutra* (Ch. *Jingang bore jing jiyan ji*, Jp. *Kongō hannya-kyō shūgenki* 金剛般若経集験記). We can infer that Kyōkai modelled his work after these references to specific Chinese texts. However, we must also note that China at the time categorised the *Record of Invisible Works of Karmic Retribution* as a history book, listing it under the heading "History: Sundry Books" (*Shilu zazhuan lei* 史録雑伝類) in the Bibliographic Treatises (*Jingji zhi* 経籍志) of the *Old Book of Tang* (Ch. *Jiu Tangshu*, Jp. *Kutōjo* 旧唐書).<sup>3</sup> Given the fact that the *Nihon ryōiki*'s title uses the same character

<sup>2</sup> All quotations in the English translation follow Watson, Shirane 2013. For the original, see Nakada 1995.

<sup>3</sup> I was unable to determine how the *Hannyagenki* was classified in China.

(*ji* 記 'record') that appears in the title *Record of Invisible Works of Karmic Retribution*, and also because *Nihon ryōiki* displays an underlying historical consciousness by chronologically arranging its tales, it is evident that the compilation of the *Nihon ryōiki* was shaped by such Chinese historical literature and was not, therefore, based only on discourse in Japan. Thus, to unpack the contents of *Nihon ryōiki*, we must consider how it reflects other literature in China and elsewhere in East Asia.

Kyōkai considered many sources apart from the *Record of Invisible Works of Karmic Retribution*. In the preface, Kyōkai states that Buddhist writings (Ch. *neijing*, Jp. *naikyō* 内経, lit. 'inner scriptures') and non-Buddhist writings (Ch. *waishu*, Jp. *gesho* 外書, lit. 'outer writings') came from China to Japan via the Korean kingdom of Paekche. Inner scriptures refer to the Chinese Buddhist canon, while outer writings refer to non-Buddhist Chinese classics, such as Confucian classics. Kyōkai states that both sets of writings came to Japan and were studied there, and argues that those with 'true wisdom' (*shinchi*  深智) gained a sense of reverence and a deep understanding of the non-Buddhist as well as the Buddhist literary traditions. This reverence is reflected in the *Nihon ryōiki* itself, though its *setsuwa* pertain to incidents of karmic retribution, good and evil deeds, and miraculous wonders that occurred *in Japan*, the text nonetheless incorporates the inner scriptures and outer writings from China.

The *Nihon ryōiki* predated the hiragana lettering system and was therefore written entirely in Chinese characters. An analysis of its textual characteristics suggests that Kyōkai adopted the phraseology used in Chinese texts to showcase Japan's progress in the path of Buddhism. Kyōkai referred to texts from a number of Chinese literary traditions for this phraseology. The texts included those from the Chinese Buddhist canon, such as the *Lotus Sutra* (Ch. *Fahua jing*, Jp. *Hokekyō* 法華経); those from the Confucian canon, such as *Mao's Book of Songs* (Ch. *Maoshi*, Jp. *Mōshi* 毛詩; also known as the *Book of Songs*, Ch. *Shijing*, Jp. *Shikyō* 詩経); and those included in the *Selections of Refined Literature* (Ch. *Wenxuan*, Jp. *Monzen* 文選). Accordingly, this article explores *Nihon ryōiki* to identify the literary sources that inspired the text and the influence of these sources in shaping the textual content and overall structure of the compilation. It also aims to determine how *Nihon ryōiki* resonated with later generations, and its contextualisation and classification by scholars of Japanese literary history. As it touches upon the disciplines of religion, history, and literature, *Nihon ryōiki* offers valuable subject matter for an intertextual analysis that provides insights into the realities of premodern Japan and the East Asian cultural sphere, where Chinese characters prevailed as lingua franca.

### 2 The Textual Structure of *Nihon ryōiki*: Inner Scriptures and Outer Writings

# **2.1 Drawing from the Inner Scriptures: The** *Lotus Sutra*

Section 2 explores the pre-existing textual elements in *Nihon ryōiki* and their incorporation. This subsection focuses on the inner scriptures – the Chinese Buddhist canon. *Nihon ryōiki* contains numerous *setsuwa* concerning Buddhist scriptures, and they are laced with scriptural references. The *Lotus Sutra* is the most quoted Buddhist text in *Nihon ryōiki* (Kōno 2016). An example of this is shown in vol. 1, *setsuwa* 19, titled "On Ridiculing a Reciter of the *Lotus Sutra* and Getting a Twisted Mouth as an Immediate Penalty". The tale goes as follows. Once upon a time, a self-ordained (*jido* 自度) monk was playing *go* 碁 when a mendicant (*kossha* 乞者) came begging for alms while reciting the *Lotus Sutra*. The novice laughed at the mendicant and started mimicking his recitation with a twisted mouth. The monk then started losing every game of *go*. Later, his mouth became stuck in its contorted position and no medicines in the land could cure him of the affliction. After recounting this tale, the *setsuwa* provides a quotation from the *Lotus Sutra*: "If anyone disparages or laughs at that person, then in existence after existence he will have teeth that are missing or spaced far apart, ugly lips, a flat nose, hands and feet that are gnarled or deformed, and eyes that are squinty" (若有軽咲之者、当世牙 歯疎欠、醜唇平鼻手脚繚戻眼目角睞). Next to this quotation is the phrase "This is what it means", or, "As it is written, so shall it be done" (其斯謂 之矣), implying that the sutra's warning would indeed occur. The *setsuwa* concludes by saying that it would be better to be possessed by demons than to badmouth the devotees of the *Lotus Sutra*.

This *setsuwa* is a simple tale of karmic retribution: a character who scoffs at the *Lotus Sutra* receives the same comeuppance warned of by the sutra. Many of the *setsuwa* in *Nihon ryōiki* have a similar pattern: they recount a miraculous event and then cite Buddhist scripture to explain the morality of the story – the reason the event occurred and an exhortation to readers. Kyōkai advocated the Buddhist faith to the readers by relating incidents in Japan to Buddhist teaching.

Two other *setsuwa* in *Nihon ryōiki* also involve people who are punished with a contorted mouth for speaking ill of a devotee of the *Lotus Sutra*. In all three cases, the *setsuwa* expresses the contortion using the character pair 喎斜 (Jp. *kasha*, Ch. *waixie*):

• *Nihon ryōiki*, vol. 1, *setsuwa* 19: "On Ridiculing a Reciter of the *Lotus Sutra* and Getting a Twisted Mouth as an Immediate Penalty" (呰読法花経品之人而現口喎斜得悪報縁)


This character pair 喎斜 appears in the *Lotus Sutra* as part of the following phrase in Chapter 18 ("The Merits of Joyful Acceptance"): "Their lips will not be drooping, pursed or twisted; and their lips will never have cankers, scabs, be cracked, misshapen" (脣不下垂、亦不 褰縮、不麁渋、不瘡胗、亦不缺壊、亦不喎斜) (Kubo, Yuyama 1993, 260; Takakusu 1925, 47a). However, it rarely appears in other Chinese literary sources. Here, we can identify an important textual characteristic: whereas Kyōkai could have chosen Chinese characters such as *qu* 曲 or *wai* 歪 to express 'twisted' or 'contorted', for the *setsuwa* concerning the *Lotus Sutra*, he repeatedly used a word that is particular to the sutra.

The following is another *setsuwa* that reveals a link between the *Nihon ryōiki* and the *Lotus Sutra*.

"On Taking a Fox as a Wife and Producing a Child"

This look place long ago in the reign of Emperor Kinmei (Emperor Amekuni-oshihiraki-hironiwa no mikoto, who resided at the Kanazashi Palace in Shikishima). A man of the Ōno district of Mino province set out on his horse in search of a good wife. At that time in a broad field, he came upon an attractive woman, who responded to him. He winked at her and asked, "Where are you going, pretty miss?" She answered, "I am looking for a good husband". "Will you be my wife?" he then asked. She replied, "I will". So he took her home, and they married and lived together.

After a time, she became pregnant and gave birth to a boy. At the same time, on the fifteenth day of the Twelfth Month, their dog gave birth to a puppy. The puppy constantly barked at the wife and threatened to bite her. She became so frightened that she asked her husband to kill the puppy, but in spite of her request, he would not do so.

Around the Second or Third Month, when the annual quota of rice was being hulled, the wife went to where the female servants were pounding rice to give them some refreshment. The puppy ran after her, trying to bite her. Startled and frightened, she changed into a fox and jumped on top of a hedge. (*Nihon ryōiki*, vol. 1, *setsuwa* 2)

This *setsuwa* has no Buddhist elements in its narrative. However, in its textual structure, Kyōkai uses the four-character compound *gaisai gōbei* 睚眥嘷吠 to describe the dog's action of 'glaring and barking' at the shapeshifting fox (睚眥 'glare', 嘷吠 'bark'). This obtuse compound was derived from a phrase in the *Lotus Sutra*. The phrase appears in Chapter 3 ("A Parable") during a passage in which ravenous foxes and dogs fight over carcasses.

There were foxes, wolves and vermin Devouring, trampling and gnawing on corpses, Scattering bones and flesh about; And a pack of dogs, Forcing each other out of the way, Rushed to the spot – Frightened and exhausted from hunger, They were searching everywhere for food, Fighting among themselves, snatching at food, Biting, snarling and barking at each other.

狐狼野干、咀嚼践蹋。䶩齧死屍、骨肉狼藉。 由是群狗、競来搏撮。飢羸慞惶、処処求食。 闘諍[齒+査]掣、啀齜嘷吠。 (Kubo, Yuyama 1993, 70; Takakusu 1925, 13c-14a)

The above passage uses the four-character compound 啀齜噑吠 to describe the action of canines 'snarling and barking' at each other (啀齜 'snarl', 噑吠 'bark'). In the first of these two character pairs (啀齜), the intended meaning of 'snarling' is ideographically evident from the use of the 'mouth' (口) radical in 啀 and the 'teeth' (歯) radical in 齜. As Nakamura Munehiko argues, in the *Nihon ryōiki*, the first character pair (睚眥) replaces the mouth and teeth radicals with the 'eye' (目) radical, thereby changing the meaning from 'snarl' to 'glare' (Nakamura 1985). This ocular rendering can also be found in Chinese literary sources, including in the biography of Fan Ju 范雎 in the "Ranked Biographies" (*Liezhuan* 列伝) of the *Records of the Grand Historian* (Ch. *Shiji*, Jp. *Shiki* 史記).<sup>4</sup> Here, Kyōkai's choice of this particular four-character compound and ocular rendering for the first character pair is unclear. However, it is clear that the characters originated in the *Lotus Sutra* and that Kyōkai therefore was referring to this sutra.

The *Nihon ryōiki* contains another story of a dog barking and the barking action is again represented by the character pair 噑吠.

Ca' Foscari Japanese Studies 19 | 6 **149** Images from the Past: Intertextuality in Japanese Premodern Literature, 143-166

<sup>4</sup> *Records of the Grand Historian*, Ranked Biographies, Biography of Fan Ju: 一飯之徳 必償、睚眥之怨必報 (Sima 1982).

On Killing Living Creatures and Suffering Revenge, Being Reborn as a Fox and a Dog, Hating Each Other, and Incurring a Penalty

[…] Yōgō continued to chant the formula, whereupon the sufferer, possessed, said, "I am a fox, […]" […] The spirit answered, "He killed me in his previous life, and so I am taking revenge on him. If he eventually dies, he will be reborn as a dog and will kill me". Hearing this, Yōgō, surprised, tried to instruct him, but he would not cease and finally the patient died.

A year later, one of Yōgō's disciples lay sick in the same room where the man had died. At that time, a visitor tied up his dog at the master's room and came to see him. The dog barked and tried to break free from his leash. […] As soon as he was freed, he ran into the room where the sick disciple lay and came out with a fox in his mouth. Although the master tried to stop him, he would not stop but chewed the fox to death. (*Nihon ryōiki*, vol. 3, *setsuwa* 2)

To summarise, the story goes as follows. The spirit of a fox who was slain by a man took revenge on its killer by possessing him. The possessed man was eventually killed by the fox and reborn as a dog. The fox then took possession of another man. The dog started barking at this fox-possessed man and then gnawed the fox to death. In this *setsuwa*, the action of the dog barking at the fox-possessed man is represented by the character pair 噑吠, the same character pair in the previously discussed *setsuwa*. The action of gnawing the fox to death is represented by the character 䶩. This obscure character appears in the previously cited passage from Chapter 3 of the *Lotus Sutra*. It is clear from these connections that this part of the *Nihon ryōiki* was influenced orthographically and phraseologically by the *Lotus Sutra*.

In both these stories that involve a fox and a dog (vol. 1, *setsuwa* 2, and vol. 3, *setsuwa* 2), the narrative is unrelated to the *Lotus Sutra*. However, the textual structure features character compounds that are indelibly associated with the sutra, such that readers who are familiar with the sutra can make the connection between the texts. We have no way of telling how deliberate the use of these words was, or the extent to which readers recognised the overlaps. Even if the overlaps are considered unintentional, their occurrence could be considered an intertextual phenomenon.

### **2.2 Drawing from the Outer Writings:** *Mao's Book of Songs***,**  *Selections of Refined Literature*

This subsection focuses on the outer writings – Chinese classics outside of the Buddhist canon. In the tale "On a Deaf Man Whose Hearing Was Restored Immediately, Owing to His Faith in a Mahayama Sutra" (*Nihon ryōiki*, vol. 1, *setsuwa* 8), the story goes as follows. For many years, a man suffered from a severe illness that had left him deaf and with blotches all over his body. After summoning a monk and devoting himself to the sutras, he was healed of the sickness. The anecdote concludes with the following message: "So we know that tales of a mysterious correspondence are not false" (是知、感応 之道、諒不虚矣). The *setsuwa*'s use of the *chuan* 遄 character is of interest as it appears in the protagonist's statement during his affliction: "Rather than living a long life and being hated by others, it is better to do good now and quickly die!" (長生為人所厭、不如行善遄死) (Kōno 2013). The 遄 character gives the meaning of 'soon' or 'quickly'. This character is rare and seldom used today, and is barely featured in ancient Japanese literature. Despite its rarity, the character appears as many as six times in *Nihon ryōiki* (in *setsuwa* 6 and 8 of vol. 1; in *setsuwa* 19, 20, and 21 of vol. 2; and in the preface to vol. 3). One possible source for this character is the Confucian classic called *Mao's Book of Songs*. Here, the character appears in one of the Odes of Yong Feng, titled "Behold the Rat":

Behold the rat; he has a body | A man with no propriety A man with no propriety | Oh, why doesn't he quickly die?

相鼠有体、人而無礼。 人而無礼、胡不遄死。 (*Mōshi*, vol. 3; Hoshino 1975, 11)

The stanza claims that the rat has a body, and then mentions a man without propriety. The final line suggests that such a man should "quickly die" (遄死). In other words, it means that a man without a meaningful purpose in life should die sooner than later, which is consistent with the context in which *Nihon ryōiki* uses the phrase. Given this contextual overlap and that the 遄 character is rarely used, it is likely that the above phrase inspired this part of *Nihon ryōiki*.

Although *Nihon ryōiki* was written in Chinese characters, it sometimes deviates from formal Chinese syntax, as it was written by a Japanese person. In this respect, the text is an example of *washū kanbun*  和習漢文, 'Japanistic Chinese literature', a Japanese term for Chinese literature with some Japanese linguistic features. However, no such deviations are found in the key sentence of *setsuwa* 8 ("Rather than living a long life and being hated by others, it is better to do good now and quickly die!" 長生為人所厭、不如行善遄死). In incorporating the "quickly die" (遄死) phrase, Kyōkai takes great care to make the wording consistent with Chinese syntax and grammar. In this way, he achieves elegant writing from a Chinese literary perspective, particularly in terms of the passive voice (人所厭 'being hated by others') and the typical Chinese construction '*a* 不如 *b*' ('rather than *a*, *b*'). In the preface to volume 2, Kyōkai expresses shame for his inelegant and disorderly writing: "I string out my lines of characters, but alas, they do not flower (連居字不華). […] I write, yet cannot get my phrases into order (編造文乱句)". Contrarily, this admission indicates Kyōkai's strong regard for the elegance of Chinese phraseology. Thus, it is likely that classic Chinese phraseology was deliberately incorporated out of a desire to create elegant and orderly Chinese writing.

Still, we should avoid leaping to the conclusion that Kyōkai sourced the "quickly die" phrase directly from *Mao's Book of Songs*, as the phrase was quoted in the *Records of the Grand Historian* (in the Biography of Shang Yang 商君) and in the *Collection of Discussions on Buddhist Teaching Past and Present* (Ch. *Ji gujin fodao lunheng*, Jp. *Shū kokon butsudō ronkō* 集古今仏道論衡) compiled by Daoxuan 道宣. Either source could have inspired Kyōkai's phrase. This issue illustrates an important point that, when exploring Chinese literature, we must be mindful of the cross-pollination and complex blending of the texts in ways that span the literary categories of Buddhism and Confucianism.

A similar example of reuse of a phrase from the outer writings appears in the story titled "On a Man Who Vowed to Copy the Lotus Sutra and was Saved from a Pit Devoid of Sunlight by the Power of His Vow" (*Nihon ryōiki*, vol. 3, *setsuwa* 13). The tale goes as follows. A man was working in an iron mine when the mine entrance suddenly caved in. The man wanted to copy the *Lotus Sutra*, and his failure to complete this filled him with regret. Inside the sealed mine, the man prayed that he would copy the sutra if he were saved. The pit then opened a little, letting in a ray of sunlight. Eventually, the man was rescued unharmed. The *setsuwa* concludes with the following message: "This took place owing to the divine power of the Lotus Sutra and the favour (Jp. *hiki* 贔屓) of Kannon. There can be no doubt about it!" (是乃法花経神力、観音贔屓、更莫疑之矣). Of interest here is the character pair 贔屓. In modern-day Japanese, the character pair 贔屓 is pronounced *hiiki* and typically means 'to favour someone' (to treat someone better than you treat others). However, this usage differs from the original usage as the usage has changed in Japan. Originally, 贔屓 was pronounced *hiki*, not *hiiki*, and it meant 'to exert force'. Searching Chinese classics for examples of this word revealed that it occurs in a passage from Yang Zheng's 張衡 "Western Metropolis Rhapsody" (西京賦), included in the *Selections of Refined Literature*.

The first capital of the Han dynasty | Lay on the banks of the Wei River | The Qin had dwelled to the north | This place was named Xianyang.

To the left lies the double rises of Yao and Han | And the Taolin Fortress | Both of which are connected by the two Hua peaks.

The Great Spirit, exerting its force | Reached up high with its hands and stretched out its legs | Thereby allowing the winding river to flow.

漢氏初都、在渭之涘、秦里其朔、寔為咸陽。左有崤函重険、桃林之塞、綴 以二華。巨霊贔屓、高掌遠蹠、以流河曲。 (Obi 1974, 120)

According to this, at the place where the two Hua peaks (*Taihua shan* 太華山 'Great Flower Mountain' and *Shaohua shan* 少華山 'Small Flower Mountain') meet, the Great Spirit (a river deity) had exerted its force to cleave the mountain into two, creating a channel for the river to flow through. Here, the word *hiki* 贔屓 is used to describe the divine force that moved mountains. This word is used in *setsuwa* 13 of *Nihon ryōiki* in the context of creating a small opening in a collapsed mine. Additionally, the force is exerted by the all-compassionate Kannon (Avalokiteśvara), as opposed to a gigantic spirit. Importantly, this use of the word in *Nihon ryōiki* marks the first instance of the word's usage in Japan, according to Shōgakukan's *Nihon kokugo daijiten* 日本国語大辞典 (second edition), a contemporary Japanese dictionary.

The *Selections of Refined Literature* was an anthology compiled by Crown Prince Zhaoming 昭明太子 of the Liang Dynasty. It has been studied extensively in China and Japan as a literary text. The works included in the anthology are obtuse and replete with rhetorical devices, and have an elaborate textual structure. The fact that *Nihon ryōiki* reflected such writing intriguingly illustrates how a monk from the Nara and early Heian period consulted not only the Buddhist canon but also the non-Buddhist Chinese classics in his literary career.

### 3 How was *Nihon ryōiki* Read?

#### **3.1** *Nihon ryōiki***'s Metadiscourse and the Ideal it Presents**

So far, we have seen how *Nihon ryōiki* incorporated textual elements from existing literature, including the Chinese Buddhist canon and non-Buddhist Chinese classics. The next question to consider is whether readers at the time, when they read the expressions derived from the *Lotus Sutra*, *Mao's Book of Songs*, or the *Selections of Refined Literature*, resonated with the milieu and ideals these texts represented. Also, leaving aside these connections, how did readers of the time understand and respond to these fantastical and farfetched stories in the first place?

*Nihon ryōiki* was the first book in Japan to compile Japanese tales of karmic retribution, good and evil deeds, and supernatural phenomena, and then lace them with Buddhist instruction. Given this information, one imagines that readers at the time would have regarded *Nihon ryōiki* as a groundbreaking text, the likes of which had never been seen. However, it had some features to accustom readers to a text that was entirely new to Japan.

The first feature is that the 116 *setsuwa* follow certain formulas and patterns (Kōno 2020, 79-80). For example, take the first *setsuwa* I discussed, *setsuwa* 19 in volume 1. After recounting an incident about someone receiving karmic retribution for mocking a devotee of the *Lotus Sutra*, the *setsuwa* quotes the relevant scripture from the *Lotus Sutra* and then concludes with the words, "This is what it means" (其斯謂之矣), implying that the scriptural warning is not an idle threat. As many as 38 other *setsuwa* recount an incident, provide a scriptural justification for the incident, and end with the same stock phrase, "This is what it means".<sup>5</sup> The *setsuwa* about the sick man getting healed through Buddhist devotion (vol. 1, *setsuwa* 8) concludes with the following sentence: "So we know that tales of a mysterious correspondence are not false" (是知、感応之道、諒不虚矣). As this example illustrates, a frequently occurring pattern in *Nihon ryōiki* is to give words of commentary prefaced with the adverbial "thus, forsooth, we know" (rendered as 是知, or as 誠知 in some cases). This pattern occurs in 62 *setsuwa* (the majority of *setsuwa* in *Nihon ryōiki*).<sup>6</sup> *Nihon ryōiki* evokes in readers a sense of familiarity, the feeling that they read this before in the very same book, by frequently reiterating the same moral message. In other words, reiteration creates an intra-textual effect.

*Nihon ryōiki* had another feature designed to make the text resonate with readers of the time. It presented the history of Japanese Buddhism by weaving anecdotes about famous historical personages such as Prince Shōtoku 聖徳太子 and Gyōgi 行基. Furthermore, it created a collective memory that resonated with people of the time and was imprinted in their memory (Kōno 2018). This point will be explained further.

The preface to volume 1 of the text states that those with *shinchi*  深智 (true wisdom) studied the inner scriptures and outer writings reverently and believed in karmic cause and effect. Thus, *Nihon ryōiki* inherently highlights the importance of *chi* 智, 'wisdom'.

In *setsuwa* 7 of volume 2, we read of a monk described as "second to none in knowledge" (智恵第一) and named Chikō 智光, '*wisdom*-light'.

<sup>5</sup> In volume 1, the phrase occurs in *setsuwa* 11, 13, 18, 19, 20, 27, 29, and 30. In volume 2, it occurs in *setsuwa* 2, 5, 7, 9, 10, 13, 17, 18, 19, 21, 22, 24, 30, 31, 32, 38, 39, 41, and 42. In volume 3, it occurs in *setsuwa* 2, 4, 14, 17, 18, 20, 23, 24, 27, 34, and 39.

<sup>6</sup> In volume 1, the pattern occurs in *setsuwa* 3, 4, 5, 6, 8, 9, 13, 18, 20, 22, 28, 29, 32, and 33. In volume 2, it occurs in *setsuwa* 1, 3, 4, 6, 7, 8, 9, 10, 12, 13, 14, 15, 17, 20, 21, 22, 27, 28, 31, 32, 36, 37, 38, 40, and 41. In volume 3, it occurs in *setsuwa* 1, 2, 3, 4, 5, 6, 8, 10, 11, 12, 15, 16, 17, 18, 22, 26, 28, 29, 30, 32, 34, 38, and 39.

In the story, Chikō descends into hell because he envied the monk Gyōgi.

On a Wise Man Who, Out of Envy, Abused an Incarnated Sage and, as an Immediate Penalty, Visited the Palace of King Yama and Underwent Suffering in Hell

Shaku Chikō […] was innately intelligent and ranked first in knowledge. He wrote commentaries on the Urabon-kyō, Daihannyakyō, Hannya Shin-gyō [Heart Sutra], and other works, and lectured on Buddhist teachings to many students.

There was at this time a monk named Gyōgi. […] He gave up lay life, freed himself from desire, and spread the Law, converting the deluded masses. He was highly intelligent and seemed to be guided by inborn knowledge.[…] Dharma Master Chikō, envious at heart, spoke ill of him, saying, "I am the wise one; Gyōgi is a mere novice! Why does the emperor ignore my wisdom and put faith only in this novice?" […] As they went forward, they came to an extremely hot pillar of iron.

"Embrace it!" commanded the messengers. When Chikō embraced the pillar, all his flesh melted away, and he was left with nothing but bare bones. […] Again they started northward, and they came to a copper pillar that was much hotter than the one before. […] "Embrace it!" commanded the messengers, and he did so, and all his flesh was melted away. […] As he faced the burning heat, he asked, "What place is this?" "This is the Avichi Hell, which is here to burn you up!" […] The two men at the gate said, "The reason that you were called here was that you spoke ill of Bodhisattva Gyōgi of the country of Ashihara. […] Go back now as quickly as you can!" […] From this time on, the Venerable Chikō put his faith in Bodhisattva Gyōgi, realising clearly that he was a sage. (*Nihon ryōiki*, vol. 2, *setsuwa* 7)

Chikō was a real historical character, an eminent monk who wrote a number of commentaries on Buddhist scriptures.<sup>7</sup> Gyōgi was also an eminent monk, and he features as a character in seven *setsuwa*. Among the characters in *Nihon ryōiki*, Gyōgi appears the most frequently, which suggests his importance. Although Chikō is comparatively less important, he was still considered a venerable scholar-monk, so it is rather shocking to read of his damnation. In the *setsuwa* 7, Chikō subsequently repents his envy and finds faith in Gyōgi, declaring the latter to be a 'sage' (*seijin* 聖人). This episode gives the

<sup>7</sup> These commentaries include *Jōmyō genron ryakujutsu* 浄名玄論略述 (Abbreviated Comments on the Profound Treatise on the *Vimalakirti Sutra*) and *Hannya shingyō jutsugi* 般若心経述義 (Observations on the Meaning of the *Heart Sutra*).

impression that a saint represents the ideal form of human beings and is superior to a wise person. In what respect, then, did Gyōgi exceed Chikō? The answer lies in the opening lines of the *setsuwa*, which introduce Gyōgi.

Gyōgi is described as a devoted and compassionate soul, who "spread the Law, converting the deluded masses" (弘法化迷), and who possessed "inborn knowledge" (生知). *Nihon ryōiki* also describes Prince Shōtoku, who was also a real-life person, in such glowing terms. The relevant passage in volume 1, *setsuwa* 4, describes him as "born by nature so wise" (生知). It also states that the prince was named Shōtoku (lit. 'sacred virtue') because he "spread the Way and brought profit to the nation" (弘法利物). Thus, according to *Nihon ryōiki*, to qualify as a sage – the ideal form of a human being – one must have 'innate' (生) 'intelligence' (知) and spread the Dharma to guide the people. In other words, one must combine 'wisdom' (智) with 'good deeds' (行). By using similar language to describe the historical personages of Prince Shōtoku and Gyōgi, both of whom would have likely occupied a prominent place in popular memory during that time, *Nihon ryōiki* links the two characters together as if they belong to a singular saintly lineage.

The incidents in the final *setsuwa* of volume 3, the latest episode chronologically, and thus the most recent episode at the time, are the most interesting. The *setsuwa* is titled "On a Monk Who Excelled in Both Wisdom and Practice and Who Was Reborn as a Prince".

(1) Meditation Master Shaku Zenjū's secular name was Ato no muraji, […] After he was ordained, he worked very diligently, studying the doctrine and excelling in both wisdom and practice. […] He worked to spread knowledge of the Law and guide others, […] the Most Reverend Zenjū lived two lives, first as Zenjū and then as a prince. […]

(2) […] His name was Bodhisattva Jakusen. The people of the time, both clergy and laypersons, praised him with the name bodhisattva because of the purity of his conduct. […] Meditation Master Jakusen realized that he was about to die. He therefore put his written records in order and transmitted them to his disciples, saying, "Twenty-eight years after my death, I will be reborn as a prince with the name of Kamino. You may know that the prince is I, Jakusen!" […] a prince named Kamino was born to the emperor. This is the present Emperor Kamino [Saga, r. 809-823], […] Therefore, we know that he is surely a sage. (*Nihon ryōiki*, vol. 3, *setsuwa* 39)

This *setsuwa* recounts two reincarnation episodes. The first episode (1) concerns a monk named Zenjū, who excelled in wisdom (智) and practice (行) and spread knowledge of the Law and guide others (弘法導人). Zenjū was reborn as a 'prince' or 'human ruler' (人王), Emperor Kanmu. The second episode (2) concerns Jakusen, a man whose pious actions were renowned. Jakusen was reborn as Prince Kamino, who was the son of Emperor Kanmu and who by the time of the *setsuwa* had himself become the emperor under the title Emperor Saga, the 'sage' or 'saintly ruler' (*seikun* 聖君). In this way, the *setsuwa* claims that the reigning emperor of the time embodied the ideal form of man, in that he has the blood of a saint, one who excels in wisdom and good deeds. In summary, *Nihon ryōiki* uses common phraseology to link the histories of Buddhist saints like Prince Shōtoku and Gyōgi, who were presumably commonly known at the time. It then portrays the reigning emperor as the rightful heir of this saintly lineage and lauds his reign as the fulfilment of a Buddhist ideal: a nation governed by a saintly emperor. In this way, *Nihon ryōiki* was, if nothing else, a politically charged text.

So far, we have seen that *Nihon ryōiki* has several intertextual layers; while primarily narrating Japanese episodes of karmic retribution and miracles, it incorporates textual elements from the Chinese Buddhist canon and Chinese classics, as well as elements from Japanese history and the worldview of the time. However, such an evaluation is based on the impressions of a twenty-first century reader; it is hard to determine how well this evaluation matches the intentions of the author of *Nihon ryōiki* intended the text to be, or its reception by contemporary readers.

However, we can get some idea of how *Nihon ryōiki* was read. The text survived intact throughout the medieval period, early modern period, and until today. Accordingly, the next section discusses how the text was read by subsequent generations and the position it occupies in Japanese literary history.

#### **3.2 How** *Nihon ryōiki* **Was Reproduced in Heian-Era Texts**

During the Heian era (794-1185), content from *Nihon ryōiki* was quoted and preserved in texts such as *Sanbōe* 三宝絵 (The Three Jewels, a collection of Buddhist *setsuwa* selected by Minamoto no Tamenori 源為憲 in 984) and *Fusō ryakki* 扶桑略記 (a historical text compiled by Kōen 皇円 some time after 1094).

In the case of *Sanbōe*, for example, the scholar Minamoto no Tamenori intended the text to serve as an introduction to Buddhism for Princess Sonshi 尊子, who was also a priestess. To that end, he drew primarily from *Nihon ryōiki*. Kōnoshi Takamitsu's analysis of the text provides valuable insights (Kōnoshi 1973). According to him, for the first four *setsuwa* in *Sanbōe*, Tamenori selected four Buddhist role models from *Nihon ryōiki*: 1. Shōtoku Taishi 聖徳太子 (Prince Shōtoku), 2. E no Ubasoku 役行者 (En no Gyōja), 3. Gyōgi Bosatsu 行基菩薩 (Bodhisattva Gyōgi), and 4. The 'Lump' nun of Higo Province 肥後国シシムラ尼 (Shishimura, nun of Higo Province). This illustrates that there were other characters besides Prince Shōtoku and Gyōgi who were hailed as sages in *Nihon ryōiki*, one of them being En no Gyōja. In volume 1, *setsuwa* 28, of the text, En no Gyōja is described as a 'Japanese sage' (我国聖人). The other was the nun of Higo Province, named the 'Lump' nun of Higo Province in *Sanbōe* (in *Nihon ryōiki*, she is named Saru Hijiri [Monkey Sahe]). In volume 3, *setsuwa* 19, of the text, the nun is described as "she had the understanding of sage" (聖化). Thus, the first four tales in *Sanbōe* indicate that the readership upheld the tradition of reversing the saintly lineage hailed in *Nihon ryōiki*. However, the *Sanbōe* effectively ignores the final *setsuwa* of *Nihon ryōiki*, the tale of Emperor Saga, which was part of current affairs during the compilation of *Nihon ryōiki*. Insofar as Tamenori, in his compilation of the *Sanbōe*, represents *Nihon ryōiki*'s readership, we can say that the readership made no attempt to faithfully recreate the original context of *Nihon ryōiki*; instead, the readership omitted some aspects and adapted the content to suit its own context.

Another textual idiosyncrasy of *Nihon ryōiki* was that it borrowed expressions from other texts and their stock narrative formulae. These features were frequently altered or omitted in the texts quoted from *Nihon ryōiki*. One example is *Nihon ryōiki*'s story of a sick man who heals after reciting Buddhist scripture (vol. 1, *setsuwa* 8). *Sanbōe* quotes this story, but it alters the part where the original had borrowed the 'quickly die' (遄死) phrase from *Mao's Book of Songs* ("Rather than living a long life and being hated by others, it is better to do good now and quickly die!" 長生為人所厭、不如行善遄死). In *Sanbōe* version (vol. 2, *setsuwa* 5), this part is expressed as follows:

Then he realized, "[…] Rather than live a long life in which I shall be despised, I should quickly generate merit and die".


The adverbial 'quickly' (遄) from *Mao's Book of Songs*, likely to be read in Japanese as *sumiyaka ni* 遄やかに, is replaced in *Sanbōe* with the character 早, read as either *hayaku* ハヤク or *toku* とく. In 1007 Minamoto had produced an anthology of contemporary idioms, titled *Sezoku genbun* 世俗諺文 (Popular Idioms). The anthology includes the very expression from *Mao's Book of Songs* 毛詩 that had inspired the "quickly die" phrase in *Nihon ryōiki*: "A man with no propriety | Oh, why doesn't he quickly die?" (人而無礼、胡不遄死). It also explicitly cites the *Book of Songs* as the source of the idiom.

人而無礼、胡不遄死。 毛詩云、相鼠有体、人而無礼。人而無礼、胡不遄死。注云、遄、速。 (Tenri Central Library 2017, 86)

As the above entry proves, Tamenori was familiar with *Mao's Book of Songs*, the alleged source of this idiom. However, when referencing the idiom in *Sanbōe*, he avoids using the 遄 character that appears in the original formulation. Given the peculiarity of this character, he probably replaced it with the more familiar word 早 for the benefit of his readership, Princess Sonshi.

A similar alteration was made to *Nihon ryōiki*'s cave-in episode (vol. 3, *setsuwa* 32). Both *Sanbōe* and *Fusō ryakki* quote this story, but neither use the character pair *hiki* 贔屓 discussed earlier.

[…] this was made possible through the power of his faith in the *Lotus Sutra*.


(*Sanbōe*, vol. 2, *setsuwa* 17; Kamens 1988, 233; Koizumi, Takahashi 1980, 198-9)

……是乃経王威力、観音霊験矣。 (*Fusō ryakki*, Empress Genmei; Kuroita 1965, 81)

*Nihon ryōiki* attributes the man's deliverance to "the divine power of the Lotus Sutra and the favor [Jp. *hiki* 贔屓] of Kannon". However, *Sanbōe* (Kanchi-in manuscript) omits the latter part, attributing the deliverance simply to "the power of his faith in the *Lotus Sutra*" (是法華経ノ願力也). Likewise, *Fusō ryakki* quotes this *setsuwa* from *Nihon ryōiki*, but avoids the word *hiki* in favour of alternative wording.

This article has cited only a few examples, but they illustrate a broad trend where later generations of readers ignored or altered some of the most idiosyncratic elements of *Nihon ryōiki*, such as the way it interwove the inner scriptures and outer writings and the way it constructed a Chinese elaborate text. Thus, later generations who read and transmitted *Nihon ryōiki* added their own new forms to the text.

### 4 How *Nihon ryōiki* Was Reproduced in Medieval and Early Modern Texts: Book Indexes, Kariya Ekisai

This section examines *Nihon ryōiki*'s prominence during the medieval and early modern period. For this, we can briefly look at the classification of the text by book indexes of the time and by Kariya Ekisai 狩谷棭斎 (1775-1835).

• *Honchō shojaku mokuroku* 本朝書籍目録 (compiled 1277-1294?):

*shinji* 神事 / t*eiki* 帝紀 / k*ōji* 公事 / s*eiyō* 政要 / s*hizoku* 氏族 / c*hiri* 地理 / *ruijū* 類聚 / *jirui* 字類 / s*hika* 詩家 / *zasshō* 雑抄 / *waka* 和歌 / *wakan* 和漢 / *kangen* 管絃 / *isho* 医書 / *inyō* 陰陽 / *hitobitoden* 人々伝 / *kan'i* 官位 / *zatsuzatsu* 雑々 / *zasshō* 雑抄 / *kana* 仮名 (Wada 1936)

\* In this index, a three-volume version of *Nihon ryōiki* (日本霊異 記三巻) was listed under the category *zasshō* 雑抄.

• Fujiwara no Sadamoto 藤原貞幹, *Kokuchō shomoku* 国朝書目 (published 1791):

*seishi* 正史 / *hennen* 編年 / *zasshi* 雑史 / *gyosensho* 御撰書 / *hinamiki* 日次記 / *seiji* 政事 / *reigi* 礼儀 / *kan'i* 官位 / *shizoku* 氏族 / *jinden* 人伝 / *tenmon* 天文 / chiri 地理 / *densya* 殿舎 / *fusetsu* 鋪設 / *ifuku* 衣服 / *inshoku* 飲食 / *kankai* 勧誡 / *kojitsu* 故実 / *ruijū* 類聚 / *jisho* 字書 / *rinchi* 臨池 / *gazu* 画図 / *kangen* 管絃 / *iyaku* 医薬 / *yōsen*  鷹鸇 / *kemari* 蹴鞠 / *kunkō* 薫香 / *jingi-jō* 神祇上 / *jingi-ge* 神祇下 / *butsuji* 仏事 / *bussetsukyūbun* 仏刹旧文 / *zassho* 雑書 / *bunshū* 文 集 / *shishū* 詩集 / *shibunbesshū* 詩文別集 / *shibunzassho* 史文雑 書 / *waka* 和歌 / *renga* 連歌 / *monogatari* 物語 (Masamune 1979)

\* In this index, a single-volume version of *Nihon ryōiki* (日本霊異 記一巻) was listed under the category *zassho* 雑書.

• Ozaki Masayoshi 尾崎雅嘉, *Gunsho ichiran* 群書一覧 (published 1802):

*kokushi-rui* 国史類 / *shinsho-rui* 神書類 / *zasshi-rui* 雑史類 / *kiroku-rui* 記録類 / *yūsoku-rui* 有識類 / *shizoku-rui* 氏族類 / *jisho-rui* 字書類 / *ōrai-rui* 往来類 / *hōjō-rui* 法帖類 / *monogatari-rui* 物語類 / *sōshi-rui* 草子類 / *nikki-rui* 日記類 / *wabun-rui* 和文類 / *kikō-rui*  記行類 / *senjū-rui* 撰集類 / *shisen-rui* 私撰類 / *kashū-rui* 家集類 / *uta'awase-rui* 歌合類 / *hyakushu-rui* 百首類 / *senshu-rui* 千首類 / *ruidai-rui* 類題類 / *wakazatsu-rui* 和歌雑類 / *senka-rui* 撰歌類 / *kagaku-rui* 歌学類 / *shibun-rui* 詩文類 / *isho-rui* 医書類 / *kyōkunrui* 教訓類 / *shakusho-rui* 釈書類 / *kangen-rui* 管絃類 / *chiri-rui* 地理類 / *meisho-rui* 名所類 / *zuihitsu-rui* 随筆類 / *zassho-rui* 雑書 類 / *gunshoruijū* 群書類従 (Kan 1984)

\* In this index, a transcript of a three-volume version of *Nihon ryōiki* (日本霊異記 写本 三巻 沙門景戒) was listed under the category *shakusho-rui* 釈書類.

• Hanawa Hokiichi 塙保己一, *Gunsho ruijū* 群書類従 (1819 edition):

*jingibu* 神祇部 / *teiōbu* 帝王部 / *buninbu* 補任部 / *keifubu* 系譜部 / *denbu* 伝部 / *kanshokubu* 官職部 / *ritsuryōbu* 律令部 / *kujibu* 公 事部 / *shōzokubu* 装束部 / *bunpitsubu* 文筆部 / *shōsokubu* 消息部 / *wakabu* 和歌部 / *rengabu* 連歌部 / *monogataribu* 物語部 / *nikkibu* 日記部 / *kikōbu* 紀行部 / *kangenbu* 管絃部 / *kemaribu* 蹴鞠部 / *takabu* 鷹部 / *yūgibu* 遊戯部 / *inshokubu* 飲食部 / *kassenbu* 合戦 部 / *bukebu* 武家部 / *shakkebu* 釈家部 / *zatsubu* 雑部

\* In this collection of old Japanese books, Kariya Ekisai's annotated version of the *Nihon ryōiki* (校本日本霊異記) was included under the category *zatsubu* 雑部.

In *Honchō shojaku mokuroku*, a book index compiled in the late thirteenth-century, *Nihon ryōiki* is listed under the category *zasshō* 雑抄, 'miscellaneous excerpts'. *Kokuchō shomoku*, published in 1791, categorises the text as a *zassho* 雑書, 'miscellaneous book'. Moving to the nineteenth century, *Gunsho ichiran* includes the text under a slightly different category, *shakusho-rui* 釈書類 or 'exegetical books'. However, *Gunsho ruijū*, a collection of old Japanese books, returns to the earlier trend by including the text under the *zatsubu* 雑部 'miscellaneous' section. Thus, *Nihon ryōiki* did not have a stable classification persisting across the medieval and early modern periods. We should also note that *setsuwa* remained absent as a literary category during these periods.

Of interest here is a study of *Nihon ryōiki* undertaken by Kariya Ekisai. Ekisai was a leading bibliographer of Chinese classics in the late Edo period (1603-1867); technically, he practised 'evidential scholarship' (Jp. *kōshō-gaku*, Ch. *kaozheng-xue* 考証学), a form of textual criticism that sought to identify the meaning of ancient texts. Ekisai studied *Nihon ryōiki* and produced an annotated version of the text, titled *Kōhon Nihon ryōiki* 校本日本霊異記. This version was included in *Gunsho ruijū*.

Kariya Eikisai, *Kōhon Nihon ryōiki*, epilogue (1816)

此書立言雖出浮屠氏而文辞古樸可喜。又間有下糾史之謬及証明他書、則 古書之最善者也。……又参以『扶桑略記』、『法華験記』、『今昔物語』諸 書補正譌脱。 (Masamune 1978)

In this epilogue, he states that his interest lies less in the text's literary qualities or its *setsuwa*, and more in its value as a resource for evidential scholarship, one that can help 'rectify erroneous histories' (糾史之謬) and 'corroborate other literary sources' (証明他書). Ekisai also states that, in analysing and annotating *Nihon ryōiki*, he referred to sources such as *Fusō ryakki*, *Hokke genki* 法華験記 (Record of *Lotus Sutra* Miracles, included in *Honchō Hokke genki* 本朝法華験記 compiled by Chingen 鎮源, circa 1043), and *Konjaku monogatari* 今昔物語 (Tales from the Past and Present, compiler unknown, early twelfth century). Here, *setsuwa*, first introduced by *Nihon ryōiki* and now established as a ubiquitous fixture of Japanese literary history, was finally entered in a distinctive literary category.

In the twentieth century, *Nihon ryōiki* began to be included in collections of classic Japanese *literature*. However, books that we would today regard as *setsuwa* became tied to disciplinary frameworks that were established during the modern period, namely the disciplines of literature, history, and philosophy, despite the fact that such books spanned such disciplines. For example, the version of *Fusō ryakki* read today is typically the version in *Kokushi taikei* 国史大系 (Anthology of Japanese *History*). For the *Hokke genki*, it is typically the print edition in *Nihon shisō taikei* 日本思想大系 (Anthology of Japanese Thought). Additionally, *Konjaku monogatari* is included in both *Kokushi taikei*  and various literary anthologies. Scholars of contemporary times tend to categorise such books under the name *setsuwa* or treat them as literature in the narrow sense of the word. This tendency reflects the disciplinary silos of the modern period. However, we could, and indeed we should, emancipate *Nihon ryōiki* from these disciplinary fiefdoms and consider the text in alternative contexts. For instance, we could start viewing *Nihon ryōiki*, along with pre-modern texts as a whole, as literature in a broader sense – text that spans philosophy, history, and literature as a discipline. This approach could open up new avenues for scholars of literature, one in which we still draw on the achievements of modern and contemporary academia, but not exclusively on them, and in which we explore the texts in an interdisciplinary manner or with the willingness to challenge our understanding of premodern scholarship and challenge epistemological frameworks. When approached in this way, *Nihon ryōiki* should yield new insights and provide valuable contributions to the history of literature.

# 5 Conclusion: *Nihon ryōiki* and Intertextuality

When we compare one text to another, we can encounter various scenarios of intertextual relationships. In some cases, a quotation may have been adapted and blended intricately into the text to such an extent that it is all but unrecognisable. Conversely, there are cases where a portion of the text is clearly traceable to a particular source, and this is likely to occur in the case of Chinese characters. As we saw in the case of the character 遄 or character pairs 喎斜, 嘷吠, and 贔屓, just a single character or character pair may be sufficient to evoke the image of a particular text or particular ancient source wherein that character or character pair appears. If the other cases can be likened to blended fruit juices, then these cases can be likened to *aemono* 和え物 (cold dishes with dressing): although the fish or vegetable is chopped up and dressed, its original form remains obvious. Text based on Chinese characters has this characteristic.

The Sinosphere has another important characteristic: a deeply entrenched tradition that continues to cherish and idealise the classical, no matter how much time passes; that values something for how much it connects with a classic text; and that has no desire for originality or novelty. For example, the *Selections of Refined Literature* has a section titled *zani* 雑擬, 'mimicry', that contains poems modelled after existing poems. In an annotated edition of the anthology recently published by Iwanami Bunko, Kawai Kōzō gives the following commentary on the 'mimicry' section:

For scholars of the classics, who valued adherence to tradition, retaining the original form was a requirement for any literary pursuit. Accordingly, the poems in the mimicry section were considered very different from mere counterfeits. Their mimicry was the very source of their value, and they constituted a distinct literary genre. (Kawai et al. 2019, 6)

We see the same values reflected in texts such as *Nihon ryōiki*. Such texts were modelled after Chinese texts. Rather than aiming for originality, the authors created a form of literature that was weaved in mimicry and quotation.

How can we effectively deploy the concept of intertextuality in a study of such Japanese texts and other texts in East Asia? Analysing Chinese-character texts from an intertextual perspective should unlock new possibilities regarding both the main text being studied and the texts it is being compared with.

#### **Bibliography**


#### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# Buddhist *Sūtras* in Dōgen's *Shōbōgenzō* **Intertextuality and Re-Interpretation**

# Aldo Tollini

Università Ca' Foscari Venezia, Italia

**Abstract** The essay deals with intertextuality in the field of religious studies and specifically with the quotation and re-interpretation of authoritative sources of Buddhist sacred texts (or *sūtras*) in the *Shōbōgenzō* written by the Japanese Zen master Dōgen in the thirteenth century. In particular, the purpose is that of highlighting the fact that Dōgen, through a personal and creative interpretation of excerpts from the *sūtras*, proposes an innovative vision of Buddhist doctrine. After a general introduction to religious intertextuality, the essay examines Zen master Dogen's peculiar form of exegesis of sacred texts in order to legitimise his personal vision of Buddhist doctrine. Five concrete examples of manipulative interpretation of quotations from the *sūtras* are presented in order to concretely show how Dogen does not hesitate to twist the meaning of quotations with great linguistic skill with the aim to promote and legitimise his opinion.

**Keywords** Zen. Buddhism. Dōgen. Sūtras. Re-interpretation.

**Summary** 1 Intertextuality and Religion. – 2 The *kanjin* Strategy. – 3 Dōgen's *Shōbōgenzō*. – 4 *Shōbōgenzō* and Quotations. – 5 Re-Interpretation of *sūtras*. – 6 Original Interpretations. – 7 Concluding Remarks.

#### 1 Intertextuality and Religion

Intertextuality is the interplay of mutually related texts by means of quotations and allusions. No text is an island and contrary to structuralist theory, it cannot be understood in isolation. It can only be understood as part of a web or matrix of other texts, themselves only to

**Ca' Foscari Japanese Studies 19 | Arts and Literature 6** e-ISSN 2610-9425 | ISSN 2610-8992 ISBN [ebook] 978-88-6969-608-4 | ISBN [print] 978-88-6969-609-1

**Peer review | Open access 167** Submitted 2021-12-13 | Accepted 2022-03-21 | Published 2022-08-30 © 2022 Tollini | cb 4.0 International Public License **DOI 10.30687/978-88-6969-608-4/010**

be interpreted in the light of other texts. Normally, it is intended as a literary device that generates related understanding in separate works, in a kind of interplay between two or more texts co-present in various forms. This often also creates an interplay of meanings, whether apparent or hidden, by which the author can play on different semantic levels.

As a literary device, it has been employed widely and from ancient times. When I say 'widely' I mean that it is not restricted to literary works, though this is probably the field in which it is most frequently used. The case I present in this essay concerns a rather particular field: that of religion where it has been studied since the second half of the 1980s, especially in Christian and Muslim contexts. I think that intertextuality applied to religious texts has some kind of special feature because in most cases it has an exegetical character, that is, it is intended to operate a critical interpretation of sacred texts in order to reach the comprehension of their correct meaning.

In this case we can speak of 'exegetical intertextuality': passages from sacred texts are quoted – verbatim quotations – and subjected to philological critical analysis in order to provide a correct interpretation. Rather than aiming at providing multiple contemporary semantic levels, allusions and references, this type of intertextuality aims at applying philological and linguistic methods to discuss the genuine meaning of sacred texts.

The two texts, the analysed and the analyser, are on different levels: the first, quoted, belongs to an ancient text, often a sacred text, is authoritative and the object of undisputed veneration and authority, the other, the analyser is subordinate and tentative. The search for the 'genuine meaning' of a text may have multiple objectives: one of which may be the intention to use authoritative texts in order to legitimise one's own original position.

This is often the case presented here in Dōgen's 道元 (1200-1253) *Shōbōgenzō* 正法眼蔵 (Treasury of the True Dharma Eye, thirteenth century). In fact, the Japanese Zen master was well aware of the fact that his teaching had very original traits that may have been misunderstood or opposed in his country where innovative ferments were not well accepted by the traditional schools. The support of the revered and unquestioned authority of the Chinese *sūtras* could enable him to sustain the novelties introduced by his teaching, although this often forced him to bend the Chinese texts to original or even forced linguistic interpretations.

As a matter of fact, the practice of reinterpreting authoritative texts from the past in a blatantly manipulative way was quite common in Buddhist circles in medieval Japan and widespread in all schools. All the more so because the search for original interpretations, even at the cost of philological transgressions was considered an accepted practice.

In the following pages I will show, by means of a few examples, how master Dōgen, who had a superior linguistic ability, employed his competence in this field in order to turn and adapt, or even, without hesitation, to change completely the meaning of *sūtras* and texts of Chinese masters in order to propose his own point of view.

#### 2 The *kanjin* Strategy

In medieval Japan, as in many other countries of East Asia, the production of commentaries on Buddhist sacred texts with exegetic purposes was a wide and intensive activity. One of the main locations of this activity was Mount Hiei 比叡山, the seat of one of the oldest schools in Japan, the Tendai 天台.

Traditionally, there were four types of interpretation of *sūtras* (especially the *Lotus Sūtra*): *innen* 因縁, *yakkō* 約教, *honjaku* 本迹, and *kanjin*  観心. The last one is the most interesting: *kanjin* literally means 'to observe one's own mind' and *kanjinshaku* 観心釈 was a personal interpretation based on *kanjin* meditative practice (observing one's 'self' – *shin* or *kokoro* 心 – in order to reach intuitive comprehension); an intuitive flash, an insight based on one's own understanding and practice (Stone 1999, 153; Sekiguchi 1973, 5). By observing one's own mind one can come to grasp the true essence of his own being and consequently of the whole of reality. However, this term was later used in the epistemological field to indicate a personal and intuitive interpretation of Buddhist canonical texts. In fact, rather than interpreting the texts of the *sūtras* in a philological way, people preferred to give a free and original interpretation based on their own contemplative experience.

In simple terms, *kanjinshaku* consisted in the practice of translating Chinese texts by means of *kanbun kundoku* 漢文訓読 (Japanese rendition of the Sinitic language), using additions in the native language to the Chinese text in such a way as to give an original translation, not corresponding to the real meaning of the text, but according to a forced interpretation that gave legitimacy to a personal and pre-constituted doctrinal vision.

The canonical text was never questioned: therein lies the truth. However, this truth was sometimes believed to be hidden, or inaccessible to most people because it was scarcely transparent or esoteric and should therefore be revealed through a personal intuitive interpretation rather than by strictly philological tools. In this way, originally elaborated doctrines were legitimised through the canonical texts and then disseminated as 'true interpretations' of the Chinese text. This kind of elaboration was often possible because all the words of the original text remain even in the native translation, and it is only with the addition of functional – i.e. interpretative – linguistic parts that the translation takes on an unconventional interpretation.

Many of these examples can be given to show how in pre-modern Japan translation by means of *kundoku* favoured the development of new epistemes within Buddhism in particular. Most of the reformers of the Kamakura period (1185-1333), when an indigenous Buddhism emerged, resorted to this translation strategy which they had learned on Mount Hiei, the seat of the Tendai school. The authority of the Chinese texts – and of the Chinese culture from which they came – could not be questioned by masters of a peripheral culture suffering from an inferiority complex. However, the innovative drive could not be curbed and the best way to give legitimacy to their thinking was devised in hermeneutics based on linguistic interpretation. In this way, Japanese peripheral culture, without denying Chinese dominant culture, found a way to elaborate its own specificity through translation: an indirect but effective way to rework and filter mainland culture according to its own perspective.

That of *kanjin* is obviously an interesting but unique borderline case. However, it tells us how the strategy of translation by means of *kundoku* in which two languages overlap in the same text is also the place where two epistemological dimensions overlap and where the boundaries between the proto-text and the meta-text are not clearly distinguishable. Just as the language of translation is a kind of hybrid between two languages, meaning also becomes a hybrid between two different cultures. Hence, translation in pre-modern Japan has for centuries continued to search for a hybrid terrain, a kind of search for a partly artificial intermediate language to act as a bridge between two different worlds. Therefore, while in Europe translation was considered a passage between different cultural and linguistic dimensions, a pouring of meanings from one container into another, a transformation, a change of form while trying to maintain the same content, in Japan a different strategy was implemented: the search for an intermediate means, the language of *kundoku*. In fact, Sino-Japanese was a tool to mediate a culture that never felt completely foreign. The language of *kundoku* is actually an 'interpretation' of Chinese, basically a variant of it. In the same way, the culture conveyed by this language was understood as a Japanese interpretation of Chinese culture by those who shared its values. The process of hybridisation and consequently of contamination has led in Japan to the elaboration of a syncretic semiotic universe with blurred and uncertain boundaries.

# 3 Dōgen's *Shōbōgenzō*

Among the many texts written by Dōgen, I have focused on *Shōbōgenzō* which nowadays is considered by scholars and monastics his most representative masterpiece. *Shōbōgenzō* is an important text of Zen Buddhism in medieval Japan and is also considered at present as one of the very relevant texts in the history of Japanese thought.

*Shōbōgenzō* is the fundamental reference for Dōgen's Buddhist thought and teaching, since it introduces in a rather systematic way his view of practice, enlightenment, and Buddha-nature. It was written during a span of time which extended for many years in the first part of the thirteenth century, after the return of Dōgen from his study journey to China in 1227.

### 4 *Shōbōgenzō* and Quotations

Dōgen makes extensive use of quotations from many Chinese Buddhist *sūtras* in *Shōbōgenzō*. The strategy implemented by Dōgen is that of presenting *sūtras* as an unquestionable reference for Buddhist truth and, in many ways, Dōgen finds the exposition of his teaching on the interpretation of those texts. In many passages Dōgen, in stark contrast to one of the cardinal principles of Zen of 'not depending on written sources' (*furyū monji* 不立文字), makes appreciative considerations about *sūtras* and their role in the search for the Buddha-Dharma. For example, in the chapter "Bukkyō" 仏経 (Buddhist *sūtras*), he writes:

しかのごとくの長老等、かれこれともにいはく、「仏経は仏道の本意にあら ず、祖伝これ本意なり、祖伝に奇特玄炒つたはれり」。

かくのごとくの言句は、至愚のはなはだしきなり、狂顛のいふところなり。 祖師の正伝に、またく一言半句としても、仏経に違せる奇特あらざるなり。仏 経と祖道と、おなじくこれ釈迦牟尼仏より正伝流布しきたれるのみなり。

Therefore, such veteran monks all say: "Buddhist sūtras do not conform with the original intention of the Buddhist Way. It is the transmission of the patriarchs which is the original intention. In the patriarchs' transmission the mysterious and the subtle has been transmitted. Such words are extremely stupid. It is talk of madmen. In the transmission of the patriarchs and masters there is not even a word or half a word which differs from Buddhist sūtras. The Buddhist sūtras and the Way of the patriarchs in the same way, have been correctly transmitted and spread from Shakyamuni Buddha". (Etō 1961, 2: 264-5)<sup>1</sup>

In a sense, we can almost say that in *Shōbōgenzō*, his view of Buddhist doctrine is presented as a personal and original commentary to famous *sūtras*.

The largest part of such quotations is from the *Lotus Sūtra* (Jp. *Hokekyō* 法華経, Chinese transl. in the fifth century CE) and from the *Jingde chuandeng lu* (Jp. *Keitoku denkōroku* 景德傳燈錄, Record of

<sup>1</sup> All the translations from Japanese to English are by the Author.

the Transmission of the Lamp of the Jingde Era, eleventh century CE), and other Mahayana texts. What is written there is the truth since they were written by buddhas and patriarchs. Nobody can question their authority; therefore, he considers the Buddhist written tradition with unsurpassed veneration. He estimates tradition as the most important leading guide in the Path of enlightenment.

# 5 Re-Interpretation of *sūtras*

However, Dōgen is convinced that texts are open to personal interpretation, or better, that though *sūtras* 'the truth is told,' still this truth is not always apparent and often needs to be 'correctly' interpreted. 'Correctly' means, in the mind of Dōgen, according to his own view of Buddhist doctrine, that it is not lingering at a mere linguistic level, or at the ordinary interpretation of language as a conventional tool, but going deeper, beyond the ordinary common interpretation. As Leighton maintains:

Dōgen uses [texts] to proclaim his own subjective teachings and to encourage the primary text as a vehicle for the self-interpretation of his audience, as well as for himself. […] texts are open to an abundance of meaning as appropriate to the diverse worlds of each interpreter. (Leighton 2007, 20)

Dōgen says in the chapter "Zazenshin" 坐禪箴 (Lancet of zazen) of *Shōbōgenzō*:

自己の所見を自己の所見と決定せざるのみにあらず、万般の作業に参学す べき宗旨あることを一定するなり。しるべし、仏をみるに仏をしらず、会せざ るがごとく、水をみるをもしらず、山をみるをもしらざるなり。眼前の法、さ らに通路あるべからずと倉卒なるは、仏学にあらざるなり。

We should not think that what we see is what should be seen, but we must be convinced that there is a deep meaning to investigate in everything. We should know that seeing a Buddha, we could not recognize him and meet him. [In the same way] seeing water we could not recognize it, and seeing mountains, we could not recognize them. To limit ourselves to the in-mediate fact which is in front of us and to not search more deeply, is not the study of Buddhism. (Etō 1961, 1: 400)

According to Dōgen, to take for granted the doctrine is not the true spirit of Buddhism, which, on the contrary, consists in "striving in search" of the true meaning which is concealed. The Way of Buddhism is a quest which implies the complete engagement of ourselves and of our energy.

### 6 Original Interpretations

In many cases, Dōgen gives new and original interpretations of texts quoted in *Shōbōgenzō*, by way of which he makes *sūtras* say what he wants to be said. These are, as a matter of fact, linguistic and/or semantic manipulations, also called in Japanese *tenshaku* 転釈, or 'different interpretation', and are considered by the Japanese master as a quest to 'unveil' reality and true meaning.

Indeed, in medieval Japan, there are many instances of this kind of semantic manipulation in the Buddhist world, not limited to the Zen sect (for example, in Shinran 親鸞 [1173-1263]). However, the high linguistic ability of Dōgen makes him one of the prominent examples. What follows are five examples taken from chapters of the *Shōbōgenzō*.

### **6.1 Buddha-Nature**

In the *Mahāparinirvāṇa Sūtra* 大般涅槃経 (The Sutra of the Great Nirvana; probably translated into Chinese around the fifth century CE) there is written:

一切衆生悉有仏性 (SAT, vol. 9, no. 0270)

In *kundoku*:

一切衆生悉く仏性有り *Issai shujō kotogotoku busshō ari*

The translation is: "All sentient beings without exception have the Buddha nature". This means that all human beings have Buddha-nature, that is, the potentiality that developed through practice, allows access to enlightenment. Practice is the means by which the spark of Buddhahood can be brought to full development.

Of course, in view of the fundamental teaching of Dōgen, that "practice and enlightenment are one" (*shūshō ichinyo* 修証一如), the statement of the *Sūtra* is unacceptable. For Dōgen, Buddha-nature cannot be a latent potentiality to be fully developed. Therefore, with a linguistic strategy, he completely changes the meaning of the sentence of the *Sūtra* and turns it consonant with his teaching.

Dōgen interpretation of 一切衆生悉有仏性 is as follows. He divides the sentence in three parts (Etō 1961, 1: 315):

一切衆生 (*issai shujō*) – 悉有 (*shitsuu*) – 仏性 (*busshō*) [なり] 一切衆生 (all sentient beings) – 悉有 (all beings) – 仏性 (Buddha-nature) – [なり] (are)

Which can be read as: "All sentient beings completely are Buddhanature". In this way, Dōgen, by means of a strategic reorganisation of the original sentence, radically changes the meaning of the *Sūtra*: Buddha-nature is not something that we 'have', but something that we 'are'. In other words, Buddha-nature is not a latent potentiality, but a fully developed manifestation, and we also are part of it.

In this way, Dōgen makes the sentence of the *Sūtra* consistent with his teaching of *shūshō ichinyo*. In fact, he negates that practice is a 'means' in view of enlightenment, and that Buddha-nature must be enacted by practice.

### **6.2 Again Buddha-Nature**

In *Daie Fukaku zenji goroku* 大慧普覺禪師語録 (twelfth century), the sayings of the Chinese master Daie Sōkō 大慧宗杲 (1089-1163), we find the following sentence:

識佛性義當觀時節因縁、時節若至其理自彰。(SAT, vol. 47, no. 1998A)

Which means: "If you want to know the meaning of Buddha-nature, observe the temporal causes and conditions. When the time comes, the principle (Buddha-nature) will manifest by itself". According to Buddhist doctrine, phenomena have no substance of their own, but are the result of the interaction of causes and conditions which from time to time join together giving rise to temporary formations, precisely phenomena. The first part of the phrase "If you wish to know the meaning of Buddha-nature, observe temporal causes and conditions" means that Buddha-nature constantly manifests itself in phenomena. Therefore, knowing the phenomena, one also knows the Buddha-nature.

The meaning of the second part in the Chinese texts is: "Buddhanature, that is, enlightenment will manifest itself only when time and conditions are ripe". So keep practising, then there will come a day when enlightenment will reveal itself. This conception of practice and enlightenment (or Buddha-nature) is very common, that is, the merits accumulated with constant practice will lead one day to maturation and to the expected outcome. This, however, is very far from that of Dōgen's view. Therefore, in the chapter "Busshō" 佛性 (Buddha-nature) of *Shōbōgenzō* we find the sentence:

仏言、「欲知仏性義、当観時節因縁。時節若至、仏性現前」。

That is a manipulation, or modification of the sentence of Daie's *sūtra*:

(欲)識(=知) 仏性義、当観時節因縁。時節若至、其理自彰。

Which becomes:

欲知仏性義、当観時節因縁。時節若至、仏性現前。

Dōgen interprets this sentence as follows:


In this way Dōgen overturns the conception of the *Sūtra* and gives a completely different interpretation, consistent with his thought that enlightenment is immanent and practice is not separate from enlightenment itself.

### **6.3 Flowers in the Space**

The *Śūraṅgama Sūtra* 大佛頂首楞嚴經 (Heroic March sutra, eighth century CE) says:

亦如翳人見空中華,翳病若除,華於空滅

It is like a person who has clouded eyes, seeing flowers in space. If the sickness of clouded eyes is cured, flowers vanish in space. (That is, a person with an eye-disease may see a mirage of flowers in the air, but once the disease is cured, the flowers he saw in the air will disappear.) (SAT, vol. 39, no. 1799)

This means that illusion (flowers in the sky) is considered a kind of 'disease'; in this case, a disease of the eyes which see a kind of mirage, or something whose existence is illusory and deceptive. In order to eliminate visual distortions, we should cure the disease and see reality as it is, that is to say, in its true aspect. By doing so, the flowers in the sky (or void) disappear and our view becomes clear. The *Sūtra* considers illusion a kind of optical disease, a distortion leading people far from the clear view of enlightenment. The ordinary status of men is sickness while enlightenment is health. A very simple statement, which however, Dōgen cannot make his own.

In the chapter "Kūge" 空華 (Flowers in the Sky) of *Shōbōgenzō*, denying the common reading of this sentence, he states:

```
迦牟尼仏言、「《また翳人の空中の華を見るが如し、翳病若し除こほれば、
華空に滅す》」。 この道著、あきらむる学者いまだあらず。
```
Shakyamuni Buddha said, "Again, it is like a sick person in the eyes who sees flowers in the sky. If the disease disappears, then the flowers of the sky also disappear". There is no scholar [of the Way] who has yet understood this expression.

眼翳によりて空花ありとのみ覚了して、空花によりて眼翳あらしむる道理 を覚了せざるなり。

They only understand that the flowers of the sky exist because of the diseases of the eyes, but they do not understand the truth according to which the diseases of the eyes exist because of the flowers of the sky.

```
しるべし、仏道の翳人といふは、本覚人なり、妙覚人なり、諸仏人なり、三
界人なり、仏向上人なり。
```
We should understand that, the person with the eye disease of the Buddha Way, is a person of original enlightenment, is a person of mysterious enlightenment, is a person of all Buddhas, is a person of the triple world, a person who goes beyond the dimension of the Buddha.

His conclusion is that:

```
諸法実相なれば翳花実相なり。
```
Since all dharmas are the true aspect of things, [also] the flowers seen with eye disease are the true aspect of things. (Etō 1961, 2: 165-74)

Dōgen disputes the claim that illusions are not part of our world and argues that they too are part of the dimension of enlightenment. Buddhahood is the whole reality, including the so-called illusory reality.

It does so by arguing that scholars do not fully understand the phrase of the *Sūtra* (この道著、あきらむる学者いまだあらず), even if in reality the sentence is very clear and the meaning unquestionable. However, Dōgen gives the *Sūtra*'s statement a completely original interpretation, and again, consistent with his teaching.

### **6.4 Do not Do Any Evil**

In the chapter "Shoaku makusa" 諸悪莫作 (Do not Commit Any Evil) there is a quotation from *Mahāyāna Mahāparinirvāṇasūtra* 大般涅槃經 (SAT, vol. 12, no. 0374): 「諸惡莫作」 which is to be read in Japanese: 諸悪作る莫れ or 'do not do the various evils' (negative imperative form). However, Dōgen reads this sentence differently (Etō 1961, 1: 147-56). Instead of reading 莫作 as 作る莫れ (do not do), he reads it 莫作 or *makusa* as a single word which actually does not exist and is an invention by Dōgen. He interprets this word as 'evil not be done', in the sense that evil cannot be done. As a consequence, Dōgen reads 諸惡莫作 as 諸惡は莫作なり or, 'the various evils, are not done'. Dōgen writes:

諸悪なきにあらず、莫作なるのみなり。諸悪あるにあらず、莫作なるのみな り。諸悪は空にあらず、莫作なり。諸悪は色にあらず、莫作なり。諸悪は莫作 にあらず、莫作なるのみなり。

It is not that evil does not exist, but it is nothing other than *makusa*. It is not that evil exists, but it is nothing other than *makusa*. It is not that evil is the void, but it is *makusa*. It is not that evil is phenomena, but it is *makusa*. It is not that evil is *makusa*, but it is nothing other than *makusa*. (Etō 1961, 1: 150)

That is: evil is not this and that. It does not exist as such, cannot be objectified. It does not exist as something concrete, definable, and abstract. Evil is just 'not doing it': it is the action of not committing it, that is to abstain from it. Therefore, evil is not a characteristic of reality, it is just what cannot be done. If action is pure evil cannot be done.

諸悪さらにつくられざるなり、莫作の力量見成するゆへに。諸悪みづから 諸悪と道著せず、諸悪にさだまれる調度なきなり。

The various evils cannot be committed because the power of not committing is realised. Therefore, the various evils do not express themselves as the various evils and there is no established implementation for the various evils. (Etō 1961, 1: 148)

#### **6.5 Phenomena and Void**

The famous *Hannya haramita shingyō* 般若波羅蜜多心経 (Sutra of the Heart of the Perfection of Wisdom, Chinese transl. in the seventh century CE), also known as the *Heart Sūtra*, dealing with the subject of wisdom in the opening sentence says:

觀自在菩薩。行深般若波羅蜜多時。照見五蘊皆空。

The bodhisattva Kanjizai (Avalokiteśvara) practising profound wisdom (*prajnā-paramita*) saw clearly that the Five Aggregates are all empty. (SAT, vol. 8, no. 0251)

However, Dōgen quoting this sentence writes:

観自在菩薩の行深般若波羅蜜多時は、渾身の照見五蘊皆空なり。

The bodhisattva Kanjizai (Avalokiteśvara) practising profound wisdom (*prajnā-paramita*) saw clearly that the entire (own) body is the empty Five Aggregates. (Etō 1961, 1: 79)

Therefore, Dōgen adds the word *konshin* 渾身, literally 'the whole body',<sup>2</sup> but understood as 'one's own body', and makes it the subject of the sentence which follows. While the *sūtra* states that the bodhisattva saw that the Five Aggregates are empty, Dōgen, with the addition of another subject, says that the bodhisattva saw that his own body is formed by the Five Aggregates that are empty.

In other words, while the *sūtra* considers wisdom as the vision of emptiness in a generic sense, Dōgen proposes a conception of wisdom as the vision of oneself as emptiness. In fact, a little later in the same chapter, he uses the expression 渾身般若なり, that is, 'the whole (one's) body is wisdom'. Wisdom, therefore, for Dōgen is intrinsic to the human being, it is already given in his existence and is not to be sought outside. It is the original enlightenment, present in man *a priori*.

In the same chapter, Dōgen about the supreme wisdom writes:

色即是空なり、空即是色なり、色是色なり、空即空なり。百草なり、万象なり。

Phenomena are emptiness and emptiness is phenomena, phenomena are phenomena and emptiness is emptiness. They are the hundred herbs (the various things) and the ten thousand phenomena. (Etō 1961, 1: 79)

<sup>2</sup> In the chapter "Kokū" 虚空 (Void), Dōgen says: 皮肉骨髄の渾身, which means "the whole physical body of skin, flesh, bones, and marrow".

Clearly, the first part "Phenomena are emptiness and emptiness is phenomena" is taken from the famous sentence of the *Hannya haramita shingyō* which says: 色即是空。空即是色 'phenomena are emptiness and emptiness is phenomena'. However, Dōgen is not satisfied with positing the relationship between phenomena and emptiness and adds that 色是色なり、空即空なり 'phenomena are phenomena and emptiness is emptiness'. This addition that makes a dilemma into a tetralemma has a deep meaning in Dōgen's thought, which is also made explicit in the sentence which follows: "They are the hundred herbs (the various things) and the ten thousand phenomena" (Etō 1961, 1: 79). That is, the principle shown in the tetralemma manifests itself in everything that exists, as it exists, and this is true profound wisdom.

Dōgen suggests that it is not enough to relate phenomena and emptiness, but that one must also admit that phenomena as such exist, just as emptiness as such also exists. This does not mean considering phenomena and emptiness as substantial. Rather, they are in accord with the principle developed by the Tendai school (remember that Dōgen was a monk of this school), of the Triple Truth or *santai* 三諦, according to which phenomena have three aspects: the relative and provisional substantial existence *ke* 仮, that of the true nature of phenomena which is insubstantial emptiness *kū* 空, and the intermediate aspect that combines both relative and provisional substantiality and the true nature of emptiness at the same time, called *chū* 中 which is beyond sayability and thinkability. These last two aspects are co-present in such a way that every phenomenon is real and unreal, substantial and empty at the same time.

Therefore, Dōgen, somehow forcing the text of the *sūtra*, makes a modification or addition to it and argues that the two poles of the dilemma, phenomena and emptiness, besides being only mutually related and dependent, also have an individual stance, raising the dilemma into a tetralemma.

We can summarise the above discourse in a scheme:


#### 7 Concluding Remarks

Re-interpretations of quotations take various different forms as: word-play, creative reading (*yomikudashi* 読み下し), modification of syntactical markers, phonetic play, reshuffle of word order, word addition, creative change of word function (verbs for nouns and so on), in order to produce Japanese readings of Chinese sentences radically different in meaning, but often (though not always) 'technically' possible, because the words in the original Chinese sentence are generally present in the Japanese outcome. In particular, the syntactical gap between *kanbun* (Chinese) and its *yomikudashi* (Japanese) with the same sinograms (*kanji* 漢字) allows a certain freedom of interpretation of the Chinese source.

It seems incredible to us that a hermeneutical tool as *kanjinshaku* or *tenshaku* may have played such an important role in the development of Japanese Buddhist doctrine, but we must remember that the *sūtras* were considered the words of Buddhas and patriarchs. In the exegetical approaches mentioned above, *kanjishaku* – especially in the field of Zen – was highly valued as a means to reach comprehension. Therefore, intuitive comprehension of sacred texts, together with a philological approach, was generally appreciated.

The search for enlightenment in Zen is a process founded on intuitive insight, a personal experience based on practice. This means that individual quest outside fixed frameworks, beyond conventionality and logic and in a sense, transgressive, is the true path. To question the words of masters in order to find one's own truth is not denigrative but may be a mandatory route.

What has been presented in this essay by means of a few examples is a particular form of intertextuality largely employed in medieval Japanese Buddhism to help the formation of new Buddhist schools during the Kamakura period with a rich variety of new approaches to Buddhist doctrine. In the spread of Chan from China to Japan, a very important role was played by transmission via texts. In fact, a great quantity of Chinese texts and *sūtras* were imported and studied in Japan. The widespread exegetical approach involved quotation with interpretation and re-interpretation; an impressive work where texts of different origins mixed.

The Japanese researched in the canonical Chinese texts as an authoritative source for establishing a local Buddhism congruent with the socio-political and spiritual needs of Japan at that time. They sometimes used those texts in a transgressive manner in order to support their views of a renovated Japanese Buddhism.

#### **Abbreviations**

SAT = *SAT Taishō Shinshū Daizōkyō* 大正新脩大藏經テキストデータベース (The SAT Daizōkyō Text Database). 85 vols. http://21dzk.l.u-tokyo.ac.jp/ SAT/satdb2015.php.

#### **Bibliography**


### **Images from the Past: Intertextuality in Japanese Premodern Literature**

edited by Carolina Negri and Pier Carlo Tommasi

# Notes on Editors and Contributors

**Matthew Chudnow** is Assistant Professor of Japanese at the Central Connecticut State University. He received his PhD in East Asian Languages and Literatures from the University of California, Irvine in 2017. His current research is an interdisciplinary analysis of female soteriology in nō theatre. He is also interested in the medieval history of the Kōfukuji-Kasuga multiplex, as well as in the history and development of traditional Japanese martial arts, especially their religious and political aspects.

**Giuseppe Giordano** is Lecturer at the University of Naples "L'Orientale", Italy. He is interested in classical Japanese literature and particularly in court poetry of the Heian and Kamakura periods. Selected publications: *Go-Toba. Cento poesie dalle isole lontane* (Aracne, 2018); *La natura in versi. La flora e la fauna nella poesia giapponese classica*  (Aracne, 2021).

**Itō Moriyuki** is Professor at Gakushūin Women's College of Tōkyō, Japan. He is interested in women's literature of the Heian period, with an emphasis on the characteristics of *nikki* literature and *monogatari* literature. Selected publications: *Sarashina nikki kenkyū* (Research on the Sarashina Diary) (Shintensha, 1995); *Sarashina nikki enkinhō* (Perspective in the Sarashina Diary) (Shintensha, 2014); *The Sarashina Diary: A Woman's Life in Eleventh-Century Japan* (Columbia University Press, 2014, with Sonja Arntzen).

**Kōno Kimiko** is Professor at Waseda University, Tōkyō, Japan. She received her PhD from Waseda University. She specialises in the study of ancient Chinese books in Japan. Selected publications: *Nihon ryōiki to chūgoku no denshō* (*Nihon ryōiki* [Record of Miraculous Events in Japan] and the Chinese Tradition) (Benseisha, 1996); co-editor of *Nihon 'bun'gakushi* (A New History of Japanese "Literature"), vols 1-3 (Bensei Shuppan, 2015-19).

**Matilde Mastrangelo** is Professor at the Sapienza University of Rome, Italy. Her main fields of research are Japanese *wagei*, with an emphasis on their cultural role from the Meiji period on, and the writer Mori Ōgai (1862-1922). Selected publications: *La lanterna delle peonie. Storia di fantasmi* (Marsilio Editori, 2012); *Mori Ōgai, Il romanticismo e l'effimero* (Aracne, 2014); *L'Intendente Sanshō di Mori Ōgai* (Marietti, 2019); "Twentieth-Century *wagei* in Tōkyō: Stages and urban places" (*Rivista degli Studi Orientali*, 93, 3-4, 2020).

**Simone Müller** is Senior Lecturer of Japanese studies at the University of Zurich, Switzerland. Research interests: Japanese literature and intellectual history. Her recent publications include "The 'Debate on the Literature of Action' and Its Legacy: Ideological Struggles in 1930s Japan and the 'Rebirth' of the Intellectual" (*The Journal of Japanese Studies*, 41, 1, 2015); "A Young Lady's Longing for a Lost Past. A Chronotopic Analysis of the Medieval Memoir *Utatane* ('Fitful Slumbers')" (*BmE*, special issue 7, 2020); and *Zeit in der vormodernen japanischen Literatur / Time in Premodern Japanese Literature* (de Gruyter, 2021).

**Carolina Negri** is Associate Professor at Ca' Foscari University of Venice, Italy. She is interested in women's literature of the Heian and Kamakura periods, with an emphasis on the social and cultural role of ladies-in-waiting. Selected publications: *Le memorie della Dama di Sarashina* (Marsilio Editori, 2005); *Diario di Murasaki Shikibu* (Marsilio Editori, 2015); *Tra corte, casa e monastero. La vita di una donna nel Giappone del medioevo* (Edizioni Ca' Foscari, 2021).

**Cristian Pallone** is Assistant Professor of Japanese Language and Literature at the University of Bergamo, Italy. He is interested in Edo-period Japanese literary culture, with a particular focus on late-eighteenth century sentimental and fantastic fiction. Selected publications: *Il rovescio del broccato* (Atmosphere, 2019); *Erbe senza radici* (Aracne, 2020).

**Ivo Smits** is Professor of Arts and Cultures of Japan at Leiden University, The Netherlands. He specialises in traditional Japanese literature, with a focus on classical court poetry in both Japanese and Chinese. Recent publications include: "La dynamique sinojaponaise (*wakan*) à l'époque Heian" (*Médiévales*, 72, 2017); "Genji's Gardens: Negotiating Nature at the Heian Court" in *Oxford Studies in Philosophy and Literature: Murasaki Shikibu's The Tale of Genji* (2019).

**Aldo Tollini** is a retired Professor of Ca' Foscari University of Venice, Italy. His field of research is Japanese language and culture, especially of the medieval period. He is also active in the field of translations of Japanese and Chinese Buddhist texts from both the ancient and pre-modern periods. Selected publications: *Lo Zen. Storia, scuole, testi* (Einaudi, 2012); *L'ideale della Via. Samurai, monaci e poeti nel Giappone medievale* (Einaudi, 2017); *Eihei Dōgen. Poesie* (Bompiani, 2019); *Alla ricerca della mente. Testi del buddhismo chán cinese di epoca Tang* (Ubaldini, 2021).

**Pier Carlo Tommasi** is Assistant Professor in Japanese Literature at the University of Hawaiʻi at Mānoa, USA. He specialises in medieval poetry and prose, with an emphasis on warrior culture and its local reconfigurations. More broadly, his research questions pertain to how writing practices shaped gender, agency, and selfhood in premodern Japanese society.

#### **Notes on Editors and Contributors**

**Yamashita (Takahashi) Noriko** is Emeritus Professor at the National Institute of Japanese Literature, Tōkyō, Japan. She is interested in kabuki, *ukiyo-e*, and printed books with pictures, with an emphasis on the literary nature of kabuki, *yakusha-e*, and *gesaku* in the late Edo period. Selected publications: *Kusazōshi to engeki* (Kusazōshi and Performing Arts) (Kyūko Shoin, 2004); *"Mitate" to "yatsushi"* ("Correspondence" and "Transformation") (Yagi Shoten, 2008); *Edo no "hyōgen": ukiyo-e, bungaku, geinō* (Edo's "Expressions": *Ukiyo-e*, Literature, and Performing Arts) (Yagi Shoten, 2014); *Zaigai eiribon: kenkyū to mokuroku* (Overseas Picture Books: Research and Catalogue) (Miyai Shoten, 2019).

# Ca' Foscari Japanese Studies


Per acquistare | To purchase: https://fondazionecafoscari.storeden.com/shop


This volume brings together scholars from different backgrounds and career stages to rethink the role and scope of intertextuality in the context of premodern Japan. From antiquity to the rise of modernity, originality through repetition persists as a staple in the literary, performative, and artistic traditions of this country. Nonetheless, rather than slavish recycling of pre-existing tropes, the redeployment of familiar motifs would become a means to explore new creative pathways and craft a shared sense of cultural belonging. The essays in this collection engage with a broad spectrum of texts from different perspectives to shed light on the Japanese past and its literary representations.

**IMAGES FROM THE PAST**

NEGRI, TOMMASI

Università Ca'Foscari Venezia