Oral Epic Traditions in China and Beyond 1032191821, 9781032191829

Table of contents :
Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
List of illustrations
Part I Critical reflections on epic studies
Chapter 1 Homer to Arimpil: The paradigm shift in international epic studies
Chapter 2 The history of epic research
Chapter 3 Current issues in epic research
Chapter 4 John Miles Foley and recent research trends on oral traditions
Chapter 5 Gregory Nagy: From the Homeric Question to Homeric Questions
Chapter 6 Lauri Honko: The identity function of epic poetry
Part II Theories and methods of oral poetics
Chapter 7 Oral poetics and the Oral Formulaic Theory
Chapter 8 Field investigations of oral epic transmission
Chapter 9 Oral poetics and Chinese epic research: Interview with the author
Chapter 10 Types of oral epic texts: A Mongol case study
Chapter 11 “Returning to the voice” : Textual research of oral epics as a starting point 
Chapter 12 “How long is long”: Epic length
Part III Indigenous research on Mongolian oral poetics
Chapter 13 Mongolian oral epic poetry
Chapter 14 The Oirat epic cycle of Jangar
Chapter 15 Analysis of Mongolian epic formulae
Chapter 16 Analysis of Mongolian epic prosody
Part IV Comparative study of four epic traditions
Chapter 17 Challenges in comparative oral epic (co-authored with Miles Foley)
Appendix: Basic concepts in the study of oral epic
Bibliography
Index

Citation preview

Oral Epic Traditions in China and Beyond

This volume is the masterpiece of Chao Gejin, one of the best-known Chinese scholars of epic studies, during the last several decades between the 20th and 21st centuries. The discussion ranges from Homeric and Indo-European epics to renewed discoveries of age-old African and Asian epics. The author details developments in research from Parry and Lord’s work on Homeric epics and Serbo-Croatian oral poetry to his own research on the Mongol heroic epic. The book traces the formation of theoretical systems such as Oral Formulaic Theory, Ethnopoetics, and Performance Theory, and ends with the author’s explorations of the 20th-century Mongolian singer Arimpil’s singing of his native epic poetry. By combining China’s theoretical concerns in verbal art and Western theories in folklore, Chao illustrates the nature and feature of oral epic in many ways, and is heading for constructing an oral poetics in a broader sense. Students and scholars of epic studies, literature, folklore, and anthropology will find this an essential reference. Chao Gejin is a member of the Chinese Academy of Social Sciences. Being a folklorist and literary critic, he focuses on folkloristics and literature, oral traditions in particular.

China Perspectives

The China Perspectives series focuses on translating and publishing works by leading Chinese scholars, writing about both global topics and Chinarelated themes. It covers Humanities & Social Sciences, Education, Media and Psychology, as well as many interdisciplinary themes. This is the first time any of these books have been published in English for international readers. The series aims to put forward a Chinese perspective, give insights into cutting-edge academic thinking in China, and inspire researchers globally. To submit proposals, please contact the Taylor & Francis Publisher for China Publishing Programme, Lian Sun (Lian​.Sun​@informa​​.com) Titles in literature currently include: Seven Lectures on Wang Guowei’s Renjian Cihua Florence Chia-Ying Yeh A Companion to Shen Congwen Sihe Chen, Gang Zhou, Jeffrey Kinkley A Comprehensive Study of Tang Poetry I Lin Geng A Comprehensive Study of Tang Poetry II Lin Geng A Thematic Exploration of Twentieth-Century Western Literature Jiang Chengyong Oral Epic Traditions in China and Beyond Chao Gejin For more information, please visit https://www​.routledge​.com​/China​-Perspectives​ /book​-series​/CPH

Oral Epic Traditions in China and Beyond

Chao Gejin Translated by Liang Yanjun

First published in English 2022 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2022 Chao Gejin Translated by Liang Yanjun Translation revision by Daniel A. Kister The right of Chao Gejin to be identified as author of this work has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. English Version by permission of China Social Sciences Press. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record has been requested for this book ISBN: 978-1-032-19182-9 (hbk) ISBN: 978-1-032-19183-6 (pbk) ISBN: 978-1-003-25800-1 (ebk) DOI: 10.4324/9781003258001 Typeset in Times New Roman by Deanta Global Publishing Services, Chennai, India

Contents

List of illustrations PART I

vii

Critical reflections on epic studies

1

  1 Homer to Arimpil: The paradigm shift in international epic studies

3

  2 The history of epic research

32

  3 Current issues in epic research

40

  4 John Miles Foley and recent research trends on oral traditions

51

  5 Gregory Nagy: From the Homeric Question to Homeric Questions

61

  6 Lauri Honko: The identity function of epic poetry

65

PART II

Theories and methods of oral poetics

77

  7 Oral poetics and the Oral Formulaic Theory

79

  8 Field investigations of oral epic transmission

86

  9 Oral poetics and Chinese epic research: Interview with the author

91

10 Types of oral epic texts: A Mongol case study

96

vi Contents

11 “Returning to the voice”: Textual research of oral epics as a starting point

102

12 “How long is long”: Epic length

112

PART III

Indigenous research on Mongolian oral poetics

119

13 Mongolian oral epic poetry

121

14 The Oirat epic cycle of Jangar

130

15 Analysis of Mongolian epic formulae

150

16 Analysis of Mongolian epic prosody

168

PART IV

Comparative study of four epic traditions

183

17 Challenges in comparative oral epic (co-authored with Miles Foley)

185

Appendix: Basic concepts in the study of oral epic Bibliography Index

215 220 229

Illustrations

Figure  6.1 Tradition-culture-identity

67

Tables   1.1 Text types of oral poems   1.2 Classification of oral poems 11.1 Text types of oral poems 11.2 Classification of oral poems 12.1 The variety of epithet embellishments in Mongolian epic 14.1 The differences between Oirat hero-tale and epic 15.1 Selected formulaic names in Jangar 15.2 Instances of the epithet “giant red Hongor” 15.3 Epithet formulae in Arimpil’s singing and Homer’s poems 15.4 The analysis of formulaic phrases in the Classic of Poetry

12 19 108 108 117 143 153 154 158 164

Part I

Critical reflections on epic studies



1

Homer to Arimpil The paradigm shift in international epic studies

  But come, be you Propitious, Apollo, and Artemis also:   farewell, all of you maidens; and me then, even hereafter,   call to your memory, when someone among men on the earth, some   much-tried suffering stranger, arrives here making inquiry:   “Maidens, for you which singer is it of men wandering hither   who is the sweetest in song, and by whom you most are delighted?”   Then do you all, each one, make answer and tell him about me:   “It is a blind man dwelling in Chios, rugged and rocky,   Whose songs, every one, are the best both now and hereafter.” —Homeric Hymn to Apollo: 165–170 DOI: 10.4324/9781003258001-2

4  Critical reflections on epic studies The history of epic research can be traced back to Aristotle. His comments on Homer’s epics initiated the diachronic study of epic on at least three fronts. First of all, the Homeric Question has been the focus of the classical debate on epic that continued through the 19th century and has affected the trends of epicology in the 20th century as well. Secondly, the study of the Homeric epics followed the classical philological approach from the Alexandrian Period up until the Oral Formulaic Theory became the core approach in the 20th century. For over 2,000 years, this tradition has experienced the baptism of fresh literary discussion from Medieval times through Renaissance, Neoclassicism, Romanticism, historicism, symbolism, structuralism, and deconstructionism to postmodernism and garnered countless accomplishments with great influence on what followed. No literary scholarship can be compared to Homeric studies in linking periods of literary thought. Thirdly, in the recent upsurge of oral and intangible heritage protection, Homeric poetry and related studies have once again become a focal point for scholars seeking the historical roots of cultural narrative. From questions like “Who is Homer?” and “Who killed Homer?” (Hanson and Heath 1998), we can trace the course of epic research as centered on the ancient Greek poet. The status and roles of epic bards differ in different nationalities and traditions. To begin with, the processes of becoming a singer are different. People with the title of “family singer” are highly respected in ethnic groups like the Mongols and the Yi of southern China. There are special singer schools to cultivate epic singers from childhood, for example in Uzbekistan. In Tibetan areas, people believe the legend of miraculously becoming a “dream-taught singer” (gnyid lam bab). Professional bards are different from amateurs; but some professionals, and amateurs as well, live only by chanting epics, while some rely mainly on singing but also have another profession. In some traditions, epic singers have other social roles, such as officiants of religious folk rituals. In recent years, singers have been found with an extensive repertoire and high artistic and linguistic proficiency, such as the Yugoslav singer Avdo Međjedović, West African Dickey Szokor, Kalmyk Eela Ovlaa, Kirgiz Jüsüp Mamay, Tibetan Drakpa and Samdrup, and the Oirat Mongol singer Arimpil named in the title of this chapter. Many can sing their people’s epics that are several times as long as the Homeric poems. Rich in performance experience and creative skills, they can improvise or adjust their sung content and narrative strategy according to the reactions of the audience in live performance and gradually form their own unique style as they actively participate in the transmission and development of their oral tradition. Looking back on the history of the study of epic by international scholars, Chinese scholars reflect on our own problems in the study of indigenous Chinese epic so as to respond to the situation faced by the local epic tradition and thus better participate in the international academic dialogue. In line with Western and Eastern scholarly traditions, the final concerns of our questioning must be 21st-century Chinese epic research and its future in relation to the multi-ethnic living forms of oral epic that are still found in China. In this, the Oirat singer in Xinjiang, Arimpil (1923–1994), has become a symbol of the Chinese indigenous epic tradition.

Homer to Arimpil  5 In the limited space of this chapter, the author attempts to discuss the source and flow of epic scholarship over the centuries. Focusing on main points, we can scan the whole evolution of epic research from micro to macro perspectives. The chapter sketches the modes of study of six epic bards, from Homer in ancient Greece to Arimpil in contemporary China, in order to highlight benchmark changes in this evolution as centered on the Homeric Question. This kind of academic history is far from being a matter of simply ordering material and opinions according to a time scale. Its strong and weak points are left to judgment and correction on the part of the international scholarly community.

The Homeric Question: From authorship to the scholarly tradition Let us start with the blind singer from the island of Chios. The Homeric Hymn to Apollo was recorded by the historian Thucydides, who accepted that the poem was written by Homer himself as the poet’s self-evaluation (Thucydides 2004:190–191).1 Chios has an advantage in the dispute of the seven cities competing for Homer in that it is famous for the group of epic bards called the Homeridae (Homer’s offspring), whose names are known from the festival of the Panathenaea. The Homeric Hymns are thirty-four odes to deities, which later scholars have regarded as anonymous. When it comes to the author of the Iliad and the Odyssey, however, the situation is quite different. Since the time of Alexander, there have been many disputes about this question. Aristotle systematically discussed characteristics of the epic; and Xenon and Hellenicus, who were called “separators” among ancient Greek scholars, stated that there were differences and internal inconsistencies between the Iliad and the Odyssey and that the Odyssey was not composed by Homer (Murray 1988:11). Later, the Jewish historian Flavius Josephus (born 37/38 CE) concluded that Homer was an oral poet. Though all these writers lived relatively close to the Homeric era, they failed to provide us with much concrete information about the poet. As is well known, the study of Homer in the 18th century centered mainly around the so-called “Homeric Question”; and that question continued to dominate epic studies in the 19th and even the 20th century. Scholars explored the identity of the author or authors of Homer’s epics, other aspects of his relationship to the poems, and questions about the spread of Homeric epic throughout the whole Western world, expanding from the Homeric Question to “Homeric Questions” (Nagy 2008:92–93). This exploration sums up the trend of international epic studies and reflects the most important scholarly developments in the field. Conservative classicists have tended to think that Homer lived around the 8th century BCE; and for a long time, this supposition dominated the study of Homeric poetry and its authorship. Concentrating on popular short narrative poems and folk tales, however, the Romantic movement gradually came to hold the view that Homer’s epics must have undergone a stage of oral transmission before being written down. The Italian Enlightenment philosopher Vico insisted that epics are the cultural achievements of nations, not the works of talented

6  Critical reflections on epic studies individuals. Regarding the Homeric epics, Vico stated that the glory of Homer as a citizen of many Greek cities is due to the fact that all these cities see that some words, phrases, and even some scattered sayings in Homer’s epic are from their own localities. There are many different opinions on the period of the early transmission of the Homeric epics, the range reaching 460 years. It is estimated that the earliest time was the Trojan War and the latest the reign of Numa, the legendary second king of Rome, c. 700 BCE (Vico 1997:416). The British archaeologist Robert Wood in An Essay on the Original Genius of Homer, published in 1769, held that Homer was illiterate and that, in any case, his epics were handed down by word of mouth. In 1795, the German F. A. Wolf published Prolegomena ad Homerum, which quickly became a catalyst triggering the debate between “Analysts” and “Unitarians” in the 19th century and the rise of Oral Formulaic Theory in the 20th century. Generally speaking, the debates of the two opposing camps of Analysts and Unitarians revolved around the topic of a single poet Homer or many Homers. Scholars in line with Vico and Wolf think that Homer’s epics come from many poetic creators. Contradictions in the epics make it difficult to think that they were composed by one person; the dialects they use come from several different areas of Greece; and the time span manifest in Homer’s language is far longer than a single person’s life cycle (Vico 1997:439). Because of their analysis of the content and structure of Homer’s epics, these scholars are called Analysts. In this camp, the two early 19th-century German scholars Johann Hermann and Karl Lachmann are noteworthy. In his kernel theory, Hermann proposed that the earliest Homeric poetry constitutes only the core of the Iliad and the Odyssey, an Ur-Iliad, on the basis of which, after continuous additions, revisions, and deletions, we have the Homeric epics that we read today. In his Liedertheorie, Lachmann maintained that Homeric poetry, like the German The Song of the Nibelungs, is composed of eighteen ancient short lays. In the same line, others believe that the Odyssey is composed of the Telemacheia and four or five independent epics about nyki (ghosts). For Unitarians, however, the author of the two Homeric epics was a single individual. This was proposed by Gregor Wilhelm Nitzsch in the early 19th century and his successor, the early 20th-century American scholar John A. Scott, along with others. They maintained that Homer’s epics are complete works created by an individual genius alone, with a unified structure and a centralized concept of dramatic conflict, such as Achilles’ anger. Because they defend the integral unity and originality of Homer’s epics, they are called Unitarians; but their theory is based mainly on subjective assumptions and has few supporters. In the heated argument between the two camps, both have sought to answer the Homeric Question threading through the whole 19th century, but from different philological and literary standpoints and approaches. Of course, there are some approaches that take a middle ground, maintaining that Homer’s epics are not the work of one author alone but that “he” played a considerable role in their construction and shaping. In his weighty Homerische Untersuchungen (1884), Ulrich von Wilamowitz-Moellendorff fully manifested his support for the Analysts with careful and detailed arguments in the classical manner. On the basis of his

Homer to Arimpil  7 meticulous textual research and excellent grasp of the history, dissemination, and stylistic changes of the epic, he made a thorough analysis of the Odyssey; but his open academic vision to some extent narrowed the long-term gap between the two sides of the debate. With the passage of time, moreover, especially with the publication of Homer: The Origins and the Transmission (1924) by Thomas W. Allen, the Unitarians forced the Analysts to face up to and partially accept some viewpoints of the other side. The two groups of scholars thus began to adjust their positions and absorb each other’s opinions, gradually moving toward a kind of constructive academic convergence as Neo-analysts and Neo-unitarians.2 As a result, the tit for tat that had long plagued Homeric scholarship tended to ease. Scholars in the classical mold admit that Analysts are proceeding in a complex and multi-faceted manner, while the Unitarians have become in essence a historical legacy. An appreciation of the implications of orality theories could actually deprive Unitarians of any foothold, but both Analysts and Neo-analysts have paid scant attention to oral theory (Foley 2000b:11–12). Regarded as the father of Western literature, Homer and his works have remained intimately linked with the history of literature. His status is very different from the modern oral epic folk notion that “everyone here is a poet, because everyone can sing.” In the ancient Greek tradition, there was a relationship between aoidós (singer) or rhapsōidos (singer) and poiētēs (poetry maker). Being all of these, Homer can be listed among the dēmioergoi, “those who serve the dēmos (people)” (Homer 1994:15–41). In the Odyssey itself, the bards mainly mentioned are Phēmios (etymologically perhaps “prophet”) and Dēmodokos (man of public respect). Gregory Nagy has convincingly reconstructed the narrative tradition that constitutes the background of Homer’s poems, with possible modes of text formation, performance, and evolution (Nagy 2008). Citing a number of examples of relationships between songs and singers, poems and poets, he provides a fulcrum to get a glimpse of the ancient Greek bards behind the text. When our thoughts travel from ancient times to present reality and from Olympus to the Himalayas or Tianshan Mountains, we find that Dēmodokos and Phēmios are not far away from us. Outstanding contemporary singers and oral poets in China, like Samdrup, Junai, and Jüsüp Mamay, are the Homers of our time. The Homeric Question and Homeric Questions are some of the oldest topics in the humanities. From “Who is Homer?” to “Who killed Homer?” we have here briefly outlined the progression of international epic studies and the questions that steer the directions of research. From questions of authorship to text collation, from bilingual translations to multi-disciplinary investigations, generations of scholars have responded to Homer’s age-old challenges with rapidly growing academic achievements in a never-ending spirit of exploration, asking “Why do people always need stories? Why do we want similar stories?” As long as epic exists, Homeric questioning will not stop. It will always inspire people to understand the essence of human oral artistry and the roots of expressive culture. An outstanding 20th-century example is the young Milman Parry, who deeply loved Homer’s epics and in the 1930s joined the ranks of thousands of questioners from the past, bringing a fresh voice to classical research and the whole field of traditional humanities.

8  Critical reflections on epic studies

Avdo: From singer to oral poetics As curiosity about the Homeric Question continued, Parry, together with his student and collaborator Albert B. Lord, found a solution in the Parry–Lord Theory or Oral Formulaic Theory, based on an anthropological response to the Homeric Question in terms of philology and data from ancient Greek, Old English, and South Slavic epic. Inspired by V. V. Radlov (alias Friedrich Wilhelm Radloff) and Matija Murko’s field investigations, Parry found that neither the Analysts nor the Unitarians touched the essence of the problem. In his view, Homeric epic is rooted in oral traditions; and to support this inference, he and Lord carried out fieldwork in many areas of Yugoslavia starting in the 1930s. Through on-site testing, they confirmed Radlov’s perception that in certain types of folk narrative singing, no two performances are exactly the same. Every competent singer regularly improvises his songs according to the situation of the moment, not singing the same song twice in the same way, yet not thinking that this kind of improvisation amounts to a new creation (Radlov 1885). By comparing recorded texts of the same story sung by different singers in the same region as well as the same story sung by the same singer at different times, Parry and Lord became convinced that what these Yugoslav folk singers sing each time is actually a new story. What they chant, however, is not simply a new song related to older songs; it is the epic song itself. Singers learn, compose, and disseminate it on the basis of poetic and narrative formulae. They do not just recite or perform an epic poem word for word; they rely on formulaic themes, typical scenes, and stylized story-patterns to re-structure the work. Thus, generally speaking, their epic singing is like aligning playing cards. Parry and Lord came to the conclusion that Homer’s poetry was the product of the same process, not the work of an individual gifted poet. Of the singers whom Parry and Lord met, Avdo Međjedović was the most outstanding. He had very high performing skills and was called “the Homer of Today.” In 1935, Lord wrote an article introducing his performances (Lord 2004:320–330). Though he had received no schooling, Avdo had stored up about fifty-eight epics in his memory, including The Wedding of Smailagic Meho with 12,323 verses (Heroic Songs Vols. 3 and 4) and Osmanbeg Delibegovic and Pavicevic Luka with 13,331 verses (Vol. 4). Each of these is as long as the Odyssey. According to Lord’s account, Parry once invited Avdo to attend the performance of a singer named Mumin, during which the singer chanted a song that Avdo had never heard. When the singing was over, Parry asked Avdo if he could sing the same song immediately, perhaps better than Mumin. Mumin accepted the friendly contest, and it was his turn to sit and listen. Avdo sang what he had just learned from Mumin, but in 6,313 verses, almost three times the length of the original.3 In a second investigation a decade later, Lord himself recorded some of Avdo’s epics, including The Wedding of Smailagic Meho. Though he was ill at the time, Avdo completed the recording of up to 14,000 verses in about a week. “After Avdo’s death, no one can sing as he did,” said Niko La Vujnovic, one of Parry’s assistants (Foley 2000b:95). The poems of the Yugoslav singers

Homer to Arimpil  9 represented by Avdo were an important fulcrum for the development of the oral poetics of the Parry–Lord Formulaic Theory.4 Of all Lord’s achievements, his 1956 article “The Role of Sound Patterns in Serbo-Croatian Epic” deals with the most basic aspects of oral traditions. He points out that not only syntactic parallelism but also patterns of alliteration and assonance play a leading role in the formation of poetic formulae. The repeated sounds do not necessarily create a rhythm, but they and other acoustic means play a supporting role as sound clusters centered on keywords, which form a bridge between meaning and sound and fix a verse as an independent unit. Yugoslav singers often produce “large words” on the basis of smaller word clusters. In the article, Lord discusses the “Song of Bagdad” as sung by Salih Ugljanin (SerboCroatian Heroic Songs Vols. 1 and 2), elaborating on the composition of phonetic patterns and outlining how they are connected with the basic meaning of a formula. The sound is not extrinsic to the formula; it is another (auditory) dimension of a singer’s traditional method of establishing the creative layout (Lord 2004). This relatively short article by Lord demonstrating the major significance of sound is important for focusing on the oral and auditory nature of traditional narrative song, and it had a profound impact on later theories. Three years later, in his article “The Poetics of Oral Creation” (1959), Lord once again explored phonetic patterns and their function in the creation of oral epic. After discussing formulae, theme, sound sequences, and syntactic balance, he went on to address the enduring continuity of myth in epic. Myths handed down through the ages have maintained their vitality through the art of singers and have had a long-standing influence on oral epics. The article expresses Lord’s notion that the oral epic tradition is diachronic in nature. As long as an epic is significant for the people who spread it, it will continue to develop in an evolutionary process. Parry and Lord’s Oral Formulaic Theory has been universally influential. According to incomplete statistics issued several years ago, there were then at least 2,207 references to the theory, involving more than 150 different languages and cultural traditions around the world. The references cover various types of analysis and conceptual tools—from song to text to performance as well as terminological schemas that include formulae, typical scenes, and story-patterns. This has made the theory one of the most powerful interpretive theories in the field of folklore. As a theoretical proposition, the presumption that Homeric poetry is a record of what was composed in the course of oral performance has led, moreover, to new approaches to textual analysis. The Oral Formulaic School emphasizes the connection between text and tradition, between an individual singer and a group of singers, and between field observation and cross-genre analysis. Analogical studies have kept the theory alive and ever new. The study of oral tradition in Yugoslavia has thus been of great significance in academic history. Soon after the emergence of The Singer of Tales in 1960, several disciplines implicitly or explicitly followed its focus on oral poetics, including two other influential schools of folklore thought: Ethnopoetics and Performance Theory.

10  Critical reflections on epic studies The Ethnopoetics School was co-founded by Dennis Tedlock and Jerome Rothenberg, starting with the publication of the journal Alcheringa: Ethnopoetics in 1970, which became a symbol of the rise of the school. Tedlock undertook an in-depth investigation and analysis of Zuni Indians’ oral poetry. His ethnographic poetics theory focuses on the rediscovery of sound, including the linguistic transmission of features of Indian poetry from the inside—pauses, voice, tone, volume control, etc. (Chao Gejin and Bamo Qubumo 2004). The linguistic anthropologist and founder of ethnography Dell Hymes broadened the scope of Ethnopoetics in another direction, namely the representation of form. Hymes conducted a field survey among Indian tribes on the northwest coast of the United States, focusing on polyphasic elements in the structure of indigenous poetry, such as lines, verses, stanzas, scenes, actions, meters, etc. Later, Elizabeth C. Fine put forward the notion of text-making patterns. Through the exploration of text presentation and operational patterns, along with a concern for expression and rhetoric in oral communication and the explanation of cross-cultural traditional aesthetic issues, Ethnopoetics can provide Analysts with a set of valuable tools for understanding the communication of expression while deepening people’s understanding and appreciation of oral transmission within their own communities or ethnic groups. David Antin, Stanley Diamond, Gary Snyder, Nathaniel Tarn, and others also joined the school of Ethnopoetics. For its part, the school of Performance Theory has been called a “methodological revolution.” Differing from earlier concepts and analytical practices in several ways, it brought to the fore a new understanding of the rules of folklore research. It focuses on folk customs as event and pays greater attention to the interaction between text and context and to improvisation and creativity, zeroing in on individuals rather than the collectivity. It does not seek universal systems of classification or functional schemas but pays more attention to situational practices in terms of ethnography. On the whole, this school has fundamentally changed the traditional way of thinking and research perspective (Yang Huili 2004a). In recent years, the thrust of oral poetics has come to be reflected mainly in the theoretical contributions of two scholars. One is the standard bearer of Oral Formulaic Theory, John Miles Foley, whose work epitomizes the theoretical advances in the study of the performance arena, traditional referentiality, and singers’ “large words.” The other is Gregory Nagy. As inheritor to the approaches of Parry and Lord, he became the leader of the fifth generation of oral poetics scholars at Harvard following Lord. He developed a sophisticated demonstration of the Homeric epic tradition and its textual process that has played an important role in advancing epicology, proposing such notions as cross-reference; the tripod of composition, performance, and diffusion; the integration of diachronic and synchronic perspectives; and the evolutionary model of the Five Ages of Homer. According to this model, the oral composition and dissemination of the Homeric texts went through a long period of development and became finalized in about 550 BCE (Nagy 2008). Oral poetics has benefited immensely from Parry and Lord’s field studies of Avdo’s performances. Their studies have become a beacon illuminating the path

Homer to Arimpil  11 toward advances in folklore theory and opened up new horizons. Today’s epic research, from Africa to South America, from India to China, has greatly changed; and the mainly philological perspectives and methodologies of classical studies are gradually being replaced by more comprehensive and objective methods and technical approaches. Beyond question, the research of Parry and Lord constituted a paradigm shift in epic studies that has led to their flourishing today.

Lönnrot: From oral folk texts to epic text classification The Finnish philologist and collector of traditional oral poetry Elias Lönnrot (1802–1884) is famous for compiling the Finnish national epic Kalevala. He was a medical student and practiced medicine for many years in central Finland. During this time, he visited many areas, collecting folk narratives, and successively published them. His early achievements include Kantele (1829–1831) (kantele being a traditional Finnish string instrument) and Kalevala (1835–1836), also known as Old Kalevala. His later publications include the collection of Finnish folk poetry Kanteletar (Maiden of the Kantele, 1840), Sananlaskuja (Proverbs, 1842), the expanded version of Kalevala known as New Kalevala (1849), and the FinskeSvenskt Lexikon (Finnish-Swedish Dictionary, 1866–1880). Of these works, it is the collation and editing of Kalevala that brought him the greatest fame. He created the epic from a wide range of folk narratives that he had collected from local people, some of which are believed to have a history of thousands of years and that include myths and legends, lyric and ritual poetry, and incantations. Today, Kalevala has become one of the world’s literary classics. It has been translated into major languages of the world, and there have been as many as thirty translations in English. Despite being a member of the educated class, Lönnrot had strong feelings for Finnish folk culture; and he showed an extraordinary interest in folk wisdom and practices. All his life, he paid close attention to magic thought and practices; and according to the scholar Jouni Hyvonen, 17%–18% of Old Kalevala comes from incantation materials. Lönnrot was keen to explore all aspects of human consciousness and unconsciousness and always disapproved of science’s indifference to magic. The process by which Lönnrot composed Kalevala is worth summarizing. In the course of editing Old Kalevala, he created a set of formats for sorting out folk poetry, using multiple-voiced dialogues to construct the epic text. Generally speaking, he favors a simple style, and he serves in the text as narrator. He does so from three positions: mythological narrator, who has dwelled for a long time in the past; intermediary, who organizes and creates the epic text; and interpreter, who explains the mindset of the Finnish people through insights into their myths and folk beliefs. Some scholars argue that as narrator, Lönnrot speaks in his own voice; and in New Kalevala, one recognizes the narrator as one raised in Lutheranism who is a supporter of the Enlightenment and Romanticism. Given these social and ideological perspectives, New Kalevala unfolds as a warm Finnish scroll, reminding Finns of their history and describing their future. At a time when Finland was under the harsh reign of Nicholas I of Russia, Finns began

12  Critical reflections on epic studies under Lönnrot’s pen to strive for a better future, striding forward in accord with the laws of freedom and progress defined by the Enlightenment while vigorously pursuing the moral code of Christianity. In brief, Lönnrot is both recreator of the past and visionary of the future. In the compilation of his epic, he combines folk poetry from mainly eastern Finland and Karjalais with the social and cultural trends of western European thought.5 Lönnrot’s efforts as epic compiler brought him great prestige. One of the reasons for his great popularity is that his practices conformed to the trend in Finland at the time toward increasingly awakened national consciousness and identity. The Finnish Society of Literature sanctified and mythologized him; and he became an icon and symbol of Finnish national identity, his head even appearing on 500-markka banknotes. It has been argued that it was he and Sibelius, whose music was greatly influenced by the Kalevala, who together put Finland on the world map. As far as epic poetics go, with him, epic construction was closely connected to the shoring up of ethnicity and even fostering the country’s independence.6 In many ethnic groups, epic always appears as a song tradition, not just as a work. This points to the complex formation of epic texts. They take many forms, including manuscripts, woodblocks, lithography, modern printing, adaptations, collations, dictations, recordings, and video and audio texts. Thanks to collators and editors, the texts of classical epics like those attributed to Homer and The Song of the Nibelungs have been transmitted down to this day, with some commonly accepted texts long having given the impression of being great books. As long as an epic remains in the continual process of living transmission, it is unrealistic to try to seek or construct an authoritative, normative text. Nevertheless, that process is not without limits; for the long history of the sung tradition restricts the direction and scope of textual variations. The composition of Kalevala differs from that of classical epics. This posed new challenges for epic researchers and triggered new ways of thinking about epic poetry. Foley and the Finnish folklorist Lauri Honko have made theoretical explorations into the classification and definition of epic text types. From the angle of text sources, they argue that epics can generally be divided into three main types: oral text, oral-derived text, and tradition-oriented text.7 This basic tripartite classification is based on the features and contexts of the texts as manifest in their composition and dissemination, with the text types of oral poems being further defined in terms of composition, performance, and reception (Table 1.1). Table 1.1 Text types of oral poems From composition Composition Performance Reception to reception

Example

Text type Oral text Oral-derived text Tradition-oriented text

Oral Oral Oral/written Oral/written Written Written

Aural King Gesar Oral/written Homeric poetry Visual Kalevala

Homer to Arimpil  13 The most fundamental research object of oral poetics and textual interpretation is based on this tripartite classification without being bounded by transmission forms like writing. The Yi Chinese epic scholar Bamo Qubumo (2003) explains as follows. Oral text: Oral tradition refers to folk customs and events transmitted orally rather than in writing. In Honko’s view, in the field of folk literature, the oral transmission of epic is the work of thousands of folk artists and singers. They have a pattern in mind, which Honko calls a “mental text”; and these mental texts are the cornerstones for constructing the storyline in a live presentation. In most cases, as with the living epic texts of Yugoslavia, this textual construction process is confirmed empirically in live oral performances through rigorous fieldwork—observation, collection, recording, and description. Oral texts are conservative but alterable, the same narrative producing different texts in different performance contexts, which results in a large collection of divergent texts. China’s Three Great Epics—Tibetan Gesar, Mongol Jangar, and Kirgiz Manas—are oral epics. Oral-derived text: This sub-genre is also called “oral-connected” or “oralrelated” text and refers to written texts closely related to the oral traditions of a community. The narratives are fixed in written words, but they retain the characteristics of oral poetry. The contextual features of their oral origins are often beyond the scope of on-the-spot investigation. As in the case of the Homeric epics, however, they can be surmised from stylistic and aesthetic aspects of the written text brought to light by specific parsing processes. The Naxi people’s creation epics Genesis and Wars between the Black and the White and the Yi epics King Zhenu of Esuo and King Zhiga Alu have been handed down to this day through ancient books, and the cultural context of their oral performance has mostly disappeared in modern life and cannot be observed and verified on the spot. Nonetheless, textual analysis shows that the finalized texts of these southern Chinese people still adhere to the basic features of the oral traditions of their indigenous ethnic groups. Tradition-oriented text: In accordance with Honko’s definition, this kind of text is compiled on the basis of the oral texts of a certain oral communication tradition. It is often seen that the components or subject matters of several texts are put together, edited, processed, and modified so as to display aspects of the tradition. The motivation of such text formation is often nationalistic or patriotic, the best example being the Finnish national epic compiled and organized by Lönnrot. Honko has repeatedly stressed, however, that the Kalevala is not the creation of any one author but a compilation of a large collection of oral texts from the national tradition. The name of Lönnrot is affixed to the epic not as author but as one who has created an integrated amalgamation of aspects of the ethnic tradition. In this sense, Honko has classified the work as a tradition-oriented text. It has had a profound impact on the awakening of the Finnish nation.

14  Critical reflections on epic studies An epic is invariably seen as telling the stories of great persons and major events in a lofty tone; and in Honko’s research, especially in his definition of epic, he emphasizes that it has played a major role in national identity. He and Lönnrot are both Finns, and they both observed and experienced the close connection between the epic Kalevala and the process of strengthening national identity. The process of building an epic tradition is actually a process of finding a people’s cultural fulcrum. Honko also positively evaluates another important role of this kind of written oral medium: It gives a dying oral tradition a new lease of life through the written text carrier and the reading of what were originally oral works as written literature. Homer’s epics, no matter how one looks at them, are clearly a successful and long-term exemplar of this second life. Of course, in creating an epic text from pre-existing folk materials, one must bear in mind that this mode of literary construction is by no means just a matter of arbitrary adaptation or editing of oral folk material. Current forms of editing, such as compilations with additions, deletions, sequential adjustments, and formatting in line with a preconceived concept, have often resulted in a departure from the scientific spirit and academic principles (Bamo Qubumo 2003/2004). Bamo Qubumo sees the formatting of a hard disk as a way to illustrate the problems that commonly arise in the process of producing and editing a poetic text. The formatting tends to read and write disk tracks repeatedly in a way that to some extent damages the hard disk and may lead to the deletion of data. When processing previously created epic texts in line with a set format, a similar problem readily arises. The analogy with computer formatting can aid in discussions of the disadvantages of traditional processes of folk narrative and provide lessons that can clear up questions about the history of epic research. At the same time, it can prod the present generation of scholars to debate issues on how to maintain clear and objective standards of evaluation. In any case, from the classical philological tradition of investigations on Homeric and European Medieval epic texts to the nature of Kalevala, epic study has made a significant leap forward. Viewing epic from the comprehensive perspective of folk customs has now become the dominant research orientation.

Vyāsa: From the compilation of great epic to the reconstruction of epic traditions In the ancient East, the great Indian epic Mahābhārata is believed to have been formed between the 4th century BCE and the 4th century CE over about 800 years and as was the case in ancient India, handed down orally. The text is dynamic, and it reached its current size and form only through the continuous processing and expansion by court singers or folk minstrels known as sūta (Huang Baosheng 2006:8). A sūta was usually the male offspring of a man of the Kshatriya warrior class married to a Brahman woman. He would often serve as a rajah’s singer and compile heroic songs to praise his achievements. Scholars think that the Mahābhārata took form in three stages: first entitled Jaya, with 8,800 śloka (couplets) or over 17,600 verse lines; next as Bhārata, with 24,000 śloka; and finally, as

Homer to Arimpil  15 Mahābhārata, with 100,000 śloka. Vyāsa, the epic writer as we know him today, is probably just a legendary figure whose identity, like that of Homer, remains obscure. The name Vyāsa itself means “division,” “expansion,” or “arrangement” (Huang Baosheng 2006:5–11). Like the Greek Hómēros of Homer, the name suggests the compilation of songs but leaves the actual authorship a mystery. Regarding the etymology of the name Hómēros, Nagy explains that hom comes from homo (together) and ēros from ararískō (fit, connect), suggesting splicing songs and poems together. The name of the Greek poet Hēsíodos is also intriguing, hēsí being derived from híēmi (emitting), described by the Muse as emitting a beautiful, immortal, charming voice (óssan hieîsai, Thesaurus, verses 10, 43, 65, 67). Like Hēsíodos, the name Hómēros fits the Muse’s description, homo and ararískō merging into homēreûsai, or phōnêi homēreûsai (Thesaurus, verse 39). Nagy believes that in both cases the poet’s name suggests the Muse’s overseeing of poetic creativity. F. Bader has attempted to connect the root *seH- (sew) with the hom of Hómēros; and Murray sees the root’s metaphorical scope as extending to “hostage,” suggesting that Homer was probably an alien captive (Murray 1988:6). Nagy accepts the possibility of Murray’s inference, but thinks that the combination of homo (together) and ararískō (fit, connect) is etymologically more reasonable.8 In ancient traditions, it was probably common to use such meaningful names to refer to a group of singers who developed the original narrative into a huge story cycle over the years. It is said that Vyāsa taught Jaya to his five disciples (eponym for sūta bards), who roamed the world chanting it. They gradually expanded the contents, transforming Jaya into the various versions of Bhārata and finally, Mahābhārata as we have it today. Vishnu S. Sukthankar (1887–1943), the initial editor in chief of the critical edition of Mahābhārata (Pune: Bhandarkar Oriental Research Institute, from 1927), convincingly demonstrates that the Brahman Boricua once usurped the Bhārata, which originally sang the praises of Kshatriya royalty, and tried to transform it for the sake of promoting Brahman thought and elevate the status of the Brahman class. Since then, however, the original Bhārata has been lost; and it is Mahābhārata that has been transmitted down to this day (Murray 1988:6). In ancient India, Vyāsa was often called Krsna Dvaipayana (island born) Vyāsa, with reference to the belief that he was born on an island. As in other oral epic traditions, he is also a character in the epic. In many epic traditions, there are legends about the author. Of special significance regarding the formation of written epic from earlier oral traditions is the Persian epic Shāhnāma or Shahnameh (Book of Kings), written by the Medieval poet Ferdowsi (940–1020), who first wrote narrative poetry in Dari Persian. Consisting of 120,000 verses in fifty cantos, the epic records myths, legends, warrior tales, and historical accounts of fifty Persian mythical and legendary Sassanid rulers. Although it is a literary epic, it has the artistic characteristics of oral poetry; and previous to Ferdowsi’s version, there were five works, now lost, with the same title. In reconstructing the textual tradition of Homer, Nagy recounted a story found in the Shāhnāma. A noble Vizier called together wise men from all over the kingdom. The wise men were all experts in the Zoroastrian Code, and

16  Critical reflections on epic studies each brought with him a remnant of the Book of Kings. They were asked to recite what they brought, one by one; and from these recitations, the Vizier collected ancient books that had been lost, which became the model for Ferdowsi’s own composition (121:156–161). This self-contradictory myth clearly delineates the comprehensive relationship of oral and written traditions (Nagy 2008:92–93). We have seen how the Homeric poems exemplify this in the inextricable relationship between the written texts and their oral sources. The text of Mahābhārata has long been recorded in writing; and there are as many as 700 collations in the Bhandarkar critical edition (1927–1966). In the fertile times of its epic formation, however, the study and transmission of the work would have been by word of mouth. Although the epic has been reorganized and revised by numerous Sanskrit poets and bards, it is still oral poetry in nature, with a strong oral coloring that readers can readily sense. The Indian scholars engaged in its text compilation have been committed to restoring the original epic, going as far back as possible. These efforts themselves amount to an epic journey. For their part, Chinese scholars of Sanskrit literature have been concentrating on the work’s translation for more than ten years. With regard to the formation of the great oral Mongolian epic Jangar, the singers of the epic, called jangarch (or jangarchi), commonly do not sing all the sequences of a particular version of the work; and they may add some new content or forget verses because they have not sung it for several years. This shows that the epic’s contents increase and decrease with dynamic fluctuation. The Mongolian epic singer Arimpil would sometimes chant nine, fifteen, or even seventeen cantos.9 For scholars doing textual analysis, it is necessary to understand this flexibility in the text as manifest in the singer’s repertoire. Though it is not easy to reach a clear conclusion about how many songs a particular singer may know, the scale of the repertoire is the main sign of the maturity of a singer’s art and mastery of the form. Once a singer has accumulated a reserve of traditional creative units like formulae and typical scenes, it is easy to learn new sequences; and the more cantos make up a work, the easier it is for the singer to improvise. Modern theoretical constructions of epic textual traditions are based on numerous interdependent multi-dimensional aspects: the production of the text and the relationship between it and what is found by field investigation work, between text and context, between performance event and communication within the community, between inheritor and audience, and between the epic community and academic research. Whether it be Greek or Indian, the present form of an ancient epic, after countless generations of compilation and collation, has become a written canon, with records of its live performance in ancient times being mostly lost. Taking precedents from contemporary oral epic performances as understood with the aid of Performance Theory and Ethnopoetics, oral poetics seeks to bring to life the poetic music of the original text and at the same time, in the ecological chain of epic transmission and inheritance, give it what Honko has called the epic’s “second life.” In investigations of the relationship between ancient epic texts and living epic traditions, there is an intimate linking among performer, recorder, collector,

Homer to Arimpil  17 editor, producer, reviser, translator, researcher, and reader. These investigations have freed themselves from the tradition of classic textual research and are now gradually approaching a more comprehensive understanding of the traditions of epic performance as a whole. The reconstruction of classical epic traditions by scholars like Nagy marks a significant advance in epic poetics, as well as a profound achievement in humanistic studies on the whole. Nagy’s research on the etymological semantics of hómēros and the construction of the evolutionary model of epic texts has reversed and transcended the conservative view of the older tradition and injected new life into the study of classic works. It is through this process of inquiry and response, construction and deconstruction, affirmation and negation that current international epic research makes the silent music of the ancient Homeric texts ring out for contemporary lovers of the spirit of poetry and our cultural heritage. We may never know exactly who Homer was, but we can confidently ask: “Who can kill Homer?”

Ossian: From criticism for forgery to oral poetry interpretation In 1758, the Scottish poet and translator James Macpherson (1736–1796) published his poetic collection The Scottish Highlander, which aroused little response among readers; but his Fragments of Ancient Poetry, collected in the Scottish Highlands, translated from Gaelic, and published in 1760, was a sensation. He then published two epics, Finga in 1762 and Temora in 1763, and the two together in 1763 as The Poems of Ossian. He claimed that the book was the work of Ossian, a 3rdcentury CE Scottish singer; but in fact, he just gathered together legends about Ossian and redid them in free verse in the style of the 1611 King James Bible. Rich in metaphors, melancholic sentiment, and love of nature, the work became very popular, had a great impact on early European Romanticism, and aroused a strong interest in ancient heroic legends. It influenced many European authors, including Goethe, as well as the young Walter Scott; and it led Madame de Staël to divide European literature into two branches, Homer and Ossian. As is well known, beginning with Macpherson’s contemporary Samuel Johnson, scholars came to doubt Macpherson’s identity as the author; and modern scholars have demonstrated how he added his own personal poems to the original Gaelic narrative poems by modifying the original characters and concepts to fit contemporary sensibility and interests. As for the historical Ossian, it is speculated that he was a legendary Gaelic hero and minstrel in Scotland around the 3rd century CE and son of the hero Finn Mac Cumhail. Legends and narrative poems about this hero are widespread throughout Ireland and the Scottish Highlands; and Ossian and his hero father became the protagonists in the Irish Fenian Cycle, in which Ossian, as narrator, sings tales of his pioneering father and legends of the father’s war band, the Fianna Éireann. Though usually described as old and blind, Ossian was said to have outlived his father. The early 16th-century The Ballads of Ossian is authentic Gaelic lyric and narrative poetry, but the Poems of Ossian that Goethe read is the creation of Macpherson. In 1996, a confrontation between science and the humanities once gave rise to the mention of Ossian. The postmodern academic journal Social Text published

18  Critical reflections on epic studies an article submitted by the physics professor Alan Sokal proposing that quantum gravity is a social and linguistic construct. Sokal later stated that he wrote the article to challenge modern journals’ intellectual integrity; and in a review entitled “Sokal’s Hoax” in the New York Review of Books on August 8, 1996, the Nobel Laureate in physics Stephen Weinberg said that Sokal’s prank reminded him of famous cases of academic cheating like Dawson’s pseudo fossil Piltdown Man and Macpherson’s Ossian poems.10 A few years later, like Charles Dawson, a highly regarded Korean scientist lied about a cloning achievement. Macpherson’s literary hoax, however, had a creative dimension that makes it quite different from this kind of bogus scientific achievement. In the history of ancient Western literature, creative epic imitation was a literary tradition. The ancient Greek poet and scholar Apollonius (about 295–215 BCE), who was born in Alexandria and called himself Alexander, moved to Rhodes after quarreling with the Alexandrian poet Callimachus. Callimachus maintained that the era of long epic poetry was over and that it was useless for a poet to want to imitate Homer and write a new epic. Apollonius strongly opposed this idea; and among his many poems, he wrote the four books of the Hellenistic epic Argonautica, telling the story of Jason and the ship Argo searching for the Golden Fleece. He wrote in the Homeric epic style, but his new treatment of the old plot, enlightening similes, and excellent description of nature has long had the power to grip readers (Murray 1988:400). Later, Apollodorus of Athens (around 140 BCE) began to create in Chronika a poetic account of Greek history from the fall of Troy in 1184 to 119 BCE; and Virgil later creatively imitated Homer’s poetry in the Aeneid, weaving ancient Roman legends together with political exigencies of the early Roman Empire. Later still, Tasso’s Jerusalem Liberata, Dante’s Divine Comedy, and Milton’s Paradise Lost were composed in the same way. Such cases are ubiquitous. In Shepeardes Calendar (1579), the lyric marriage poem Epithalamion (1595), and the allegorical The Faerie Queene, with its initially mock-heroic character (1590–1596), Edmund Spenser creatively absorbed the nutrients of earlier works. A poet may regard his work as parody or claim it to be an authentic imitation of Homer or the rediscovery of Ossian. These are all quite different; and we need to consider the social and cultural context behind his claims. We will return to Ossian later; but for now, let us discuss a case from Serbia. The early 19th-century Montenegrin poet, bishop, and monarch Petrović Niegoš spent much of his time as a child in the Serbian countryside. From the beginning of his life, he was immersed in the South Slavic oral tradition and heroic stories; and he learned epic singing from local guslar (singers). In 1830, he was chosen as heir to the throne of Montenegro by his uncle Peter I (also a poet); and he also served as Bishop of Montenegro. He carved himself a place in history by leading Montenegro to a bright future and is recognized as the greatest poet of Serbian literature. His Voice of Mountains (1833), The Serbian Mirror (1845), and poetic drama The Mountain Wreath (1847) show that poetry occupied the supreme place in his heart. Even as prince, he never gave up singing. His poetic compositions exhibit four elements. They combine both Muslim and Christian oral traditions, manifest a proficiency in two

Homer to Arimpil  19 languages, include old traditional songs and new poems, and incorporate oral traditions in written texts. Foley has noted that some researchers see Petrović Niegoš’ poetic imitations of the tradition as parody, questioning their quality and authenticity. But Foley maintains that in fact, Niegoš truly sang on paper, creating oral poems with a pencil in hand for the consumption of literary-minded readers. Foley analyzed how Niegoš’ imitative poetic creation involves almost all forms of expression. Some poems amount to the re-performance of traditional stories that people knew well; some are completely new songs; and some draw upon traditional literary conventions for creating new poetic works. The new poems are localized and have contemporary themes, but they are always created according to the stylized form with ten metrical feet. Looking through Niegoš’ works, moreover, we can see that his comprehension of register—his expressive strategy—is rooted in his profound appreciation of both the oral tradition and literary texts. His life itself reflects the interweaving of oral and written culture, the epitome of 19th-century Montenegro. All in all, Niegoš’ case tells us that to grasp the diversity and significance of oral poetry, we need to enter into the tradition, the genre, and especially the carrier, the particular poetic medium (Foley 2002:50). In line with this approach, Foley suggests in How to Read an Oral Poem four models for interpreting oral poems according to their particular medium (Foley 2002:50; see Table 1.2). These models can be applied to many classic epics and other works of oral poetry. To return to Macpherson’s Ossian, the reason why people remember the ancient folk figure Ossian is due to Macpherson’s creative method of concocting a forged body of knowledge. As Wellek and Warren have pointed out: It is important to identify fake works and expose literary fraud in the history of literature; but we need to go further and acknowledge that the controversy caused by Macpherson’s fake collection from Ossian has actually attracted many people’s interest to study Gaelic folk poetry. (Wellek and Warren 2005:67) Table 1.2 Classification of oral poems Medium

Composition Performance Reception

Example

Oral performance Oral

Oral

Aural

Voiced texts Voices from the past Written oral poems

Written Oral/written

Oral Oral/written

Aural Oral/written

Tibetan paper singer (singing from written notes) Chicago slam poetry Homer’s Odyssey

Written

Written

Written

Bishop Niegoš

20  Critical reflections on epic studies As the Chinese thinker Laozi has said, “Under each blessing, a scourge crouches; with each scourge, a blessing approaches.” It is said that the reason why Macpherson collected ancient ballads was because the last Prince Charlie of the Stuart royal family had fled, which led to bleak prospects for Scottish culture. His collection and creative use of the old poems turned that threat into a blessing. Foley took Macpherson as a specific case study and juxtaposed his mode of poetic composition with those of Lönnrot and Niegoš. Shunning commonly used critical terms like “forgery,” “fake works,” and “deception”; he analyzed the nature of Macpherson’s poetry in terms of resources and creative imitative skills. In Foley’s view, Macpherson’s creative works have three main components: records of actual oral performances; poems and verses copied from extant manuscripts of the period; and his own pure creations. All of these are present in Macpherson’s rich panorama as he explores the cultural and political power of the oral tradition and expresses it anew through the voice of his own text. To a great extent, he filtered the folk poems he collected and added to the verses far more elements garnered from the field than did Lönnrot, but his imitative works still have the creative status of written oral poetry (Foley 2002:52). The unfolding series of Foley’s works indicates that his vision went beyond Oral Formulaic Theory. In his view, the most important theories of folklore in the 20th century became creatively fused into the comparative study of oral traditions. Theories of the comparative principles of oral tradition, the performance arena, and traditional referentiality have been put forward systematically to form a unique system of oral poetics and analytical method for the study of oral poetic texts. Foley himself set out a theory of narrative ethnography or ethnographic poetics. He traveled many times to the Šumadija area of Serbia for investigations and translated the epic poem The Wedding of Mustajbe’s Son Bećirbey as performed by the Yugoslav singer Halil Bajgorić and collected by Parry and Lord in 1935. He also incorporated modern Chinese storytelling into his cross-cultural perspective, thus bringing this ancient but everlastingly new oral art form into the framework of international oral traditions. Objectively evaluating earlier theoretical presuppositions of the oral/written dichotomy put forward by Walter J. Ong and others, Foley believed that the analytical pattern of binary opposition—oral and written—is the first step to correctly understanding and appreciating oral traditions and their diversity. He proposed that a deeper understanding of oral traditions has led to the opening up of the barriers between orality and writing and established a link between the two that gives a correct understanding of modes of human cultural expression. He reiterated the theoretical views of traditional referentiality and stressed the explanatory power of tradition in such a way as to remind us of the need to explore the poetic principles of oral traditions rather than remain locked into the poetic concepts of written literary criticism. It is important to question the divergent destinies of Macpherson and Niegoš. Their differing cases provide us with information about how the concept of oral poetics has expanded from epic study to the interpretation of oral poetry. It is thus necessary to understand the epistemological spectrum of oral poetic interpretation

Homer to Arimpil  21 as proposed by Foley in the intercommunication and interaction of oral transmission and writing, memory and knowledge, small and grand traditions, so as to generate a pluralistic, open vision that returns to the original wisdom and poetic spirit of human expressionist culture. Robert Frost’s poem “Mending Wall” quotes the proverb “Good fences make good neighbors.” As one scholar has said regarding Chinese researchers of our indigenous oral poetic heritage, China is a country where wall culture prevails both tangibly and intangibly. How many walls are there that form barriers between academic disciplines? We need a more open vision to deal with the objects of our research and understand and share our oral poetry and its artistic essence. Then, we may say, “Open the gates and know your neighbors.”

Arimpil: From oral folk narrative to scholarly research The English term “epic” is derived directly from the Greek epikos and Latin epicus, which are etymologically related to the ancient Greek epos, with its original meaning of “word” or “discourse” that later extended to refer to early oral narrative poetry and oral epic fragments. The concept of epic was introduced into China in the late 19th century. In 1879, the Qing Dynasty official Guo Songtao wrote about the Trojan War in Homer’s epic in his diary when he was assigned to England. According to available data, however, Zhang Taiyan (Zhang Binglin) was probably the first person to use the term in China. He did so in his index to the woodcut Qiushu (Miscellaneous Names) compiled in 1899 by Liang Qichao and revised in 1902. Zhang explains: Probably before the creation of written scripts, some stories or records could only be passed on orally, which means some content will definitely be missing or completely lost. To preserve contents, people have polished language by making it rhyme so that it could be easily chanted and remembered. I guess that before Ju Yong and Cang Ji created our Chinese characters, the only literature was epic. … The written form of language emerged only after the language had been made to rhyme, and poetry took shape only after epics had become quite sophisticated. For Zhang, epic apparently includes national epics, legends, stories, short songs, historical songs, etc. Later, in 1918, Hu Shi translated “epic” as “story poem” (Hu Shi and Ouyang zhesheng 1998:150); and in 1922, Zheng Zhenduo introduced Homer’s epics in Novel Monthly, believing that Homeric poetry expressed the freshness and simplicity of the early Greek nation. In 1923, in the 87th issue of Literature Weekly, he published “The Epic,” in which he argued that epic is a kind of narrative poetry, adding, “There is no epic poetry in China; if we follow the strict definition of epic, there are only sporadic narrative poems” (Zheng Zhenduo 1998:362–365). Later, however, his position is said to have become less clear (Chen Yongchao 2005:164–165). In 1929, Fu Donghua published his prose translation of the

22  Critical reflections on epic studies Odyssey. Chinese readers then began to fully appreciate Homer’s poetry; and scholars began to debate whether there had been epic poetry in the history of Chinese literature, a debate that lasted until around 1985 (Ye Shuxian 2005:76). The exploration, collection, recording, collation, and publication of Chinese ethnic epics that began in the 1950s put an end to this debate and refuted Hegel’s early assertion that there is no epic poetry in China (Wen Yiduo 1997:209). As we know today, among the minority ethnic peoples of northern and southern China, there exist thousands of epics. No one to date has offered detailed statistics or specific figures; but considering that there are hundreds of epics in Turkic languages, at least 550 records of Mongolian epics, and numerous epic traditions in the southern nationalities, thousands is a conservative estimate. Among these, the Tibetan Gesar, the Mongol Jangar, and the Kirgiz Manas have become the world-renowned Three Great Epics. In addition, hundreds of heroic epics have been handed down among the Altaic-language peoples in the north of China; and creation epics, migration epics, and heroic epics sung in simple styles have been handed down among many ethnic groups in the south. All these oral epics, passed down from time immemorial, have converged into a mountainous intellectual treasure of oral traditions, of which the Chinese nation as a whole is proud and which has become a rare and magnificent part of the landscape of world culture. The earliest scientific research activities on the epics of China were carried out mostly by foreign scholars, dating back to the 18th century. The Russian traveler P. S. Pallas first introduced Gesar in foreign countries in 1776, and in the 19th and 20th centuries, the epic poetry of ethnic minorities in China gradually caught the attention of foreign scholars as a branch of Oriental studies.11 The study of Mongolian and Tibetan epics in Western Europe has been quietly booming ever since about 1851, when the monographs of W. C. Schott and many others began to be published in Abhandlungen der Berliner Akademie. The first translation in a foreign language of the Mongolian epic Geser (Tibetan Gesar) appeared several years earlier in 1839. It was translated from the 1716 Beijing Woodblock Geser by L. J. Schmidt and published in St. Petersburg as Die Thaten Bogda Gesser Chan. Later, there were various published versions of Geser, of which A. H. Franke’s A Lower Ladakhi Version of the Kesar Saga (1901–1902) is the most famous. In China itself, traces of epic research can be found almost 300 years ago. In 1779, Sumpa Yeshé Penjor (1704–1788), an eminent Tibetan or Mongolian monk in Qinghai, spoke of Gesar in communications with the Sixth Panchen Lama (Zhao Bingli 1990:286–290), and he collected relevant topics in Questions and Answers about Gesar.12 The Tibetan Gesar and Mongolian Geser are generally considered to have originated in the same epic, which came to have divergences in narrative content, pronunciation, and distribution. The voluminous work later spread to other ethnic groups, with texts today that include Gesair among the Tu and Kizir among the Tuwa of Xinjiang. The epic also spread by word of mouth among the Pumi, Naxi, and Lisu peoples of Yunnan, who live in close contact with Tibetans. Nevertheless, the first publication to introduce the epic in Chinese was that of Ren Naiqiang in 1930. He spent the year 1929 in the Tibetan area of Xikang; and in December of the following year, he published two articles in

Homer to Arimpil  23 the Sichuan Daily entitled “Tibetan Three Kingdoms” and “A Brief Introduction to the Tibetan Three Kingdoms.” Ren believed that Gesar of Ling (called also Tibetan Three Kingdoms in Chinese and Gesar Lante in Tibetan) is the Gesar epic because it is poetry (Ren Naiqiang 1945; Jiangbian Jiacuo 1986). Large-scale epic collection and editing in China commenced in the 1950s, albeit with ups and downs. This activity roughly clarified the main texts of our national ethnic epics and their distribution. However, systematic epic study took shape only in the early 1980s, forming a body of comprehensive materials for in-depth discussion. The series Chinese Epic Research, edited by scholars of the Institute of Minorities Literature of the Chinese Academy of Social Sciences (now known as the Institute of Ethnic Literature), represents the general research level of this period. Representative works include A Study of the Mongol Epic Jangar by Rinchindorji, A Study of the Kirgiz Epic Manas by Lang Ying, A Study of the Tibetan Epic Gesar by Jiangbian Jiacuo, and Studies of the Epics of Southern China by Liu Yahu. These monographs comprehensively and systematically expound on the general features of Chinese epic, key epic texts, important singers, and other major issues; and they put forward the goal of establishing a theoretical system for the study of Chinese epics. In this process, a transformation is now taking place from oral narrative to written epic texts; and Chinese creation, migration, and heroic epics are now finally coming to enrich the world’s epic heritage. At the same time, scholars are breaking out of a focus on the texts of classic works and our written poetic heritage and pointing our vision toward the living epic of folk ethnic folk cultures. As for folk bards, Tibetan epic singers can be divided into at least six types in terms of their place in their local social and cultural context (Jiangbian Jiacuo 1994; Yang Enhong 1995). There are dream-taught bards, disciple bards, recitation bards, prasenā bards, and hidden treasure-text bards. It has been said that P. Arimpil (1923–1994) may have been the greatest epic singer of our time (Jamcha 1996; Dawataya 1999; Rinchindorji 1999: 41–42). This legendary illiterate jangarch could sing most of Jangar and was the first singer interviewed by Professor Choijinjab when he visited Hoboksar County in Xinjiang in 1980. Later, Arimpil’s nephew Dr. Dawataya contributed to the collection and preservation of his oral performances; and case studies, textual interpretations, and reconstructions of Arimpil’s output, along with other empirical research, have led to a new perspective on Chinese epic and oral poetics (Chao Gejin 2000). This amounts to a paradigm shift in Chinese epic research from the work of earlier scholars on written texts to a focus on our oral traditions. As Zhong Jingwen has said, moreover: It has transformed epic scholarship from the collection of basic data to the scientific organization of literary facts, from general surveys and the accumulation of data under a subjective framework to the restoration and exploration of the epic tradition in accord with objective historical perspectives.13 In the past, Chinese epic studies were quite alienated from oral folk traditions. They focused on epic as a literary text, not as part of the oral narrative tradition

24  Critical reflections on epic studies and without attention to epic poetry as a dynamic, living folk event and speech act. We are now learning much as we explore the different mechanisms of textual and oral epic inheritance. With the syntactic analysis of oral formulae based on Arimpil’s performances, we have come to realize that there are great epistemological differences between oral and written poetry. One cannot simply apply the methods of studying written works to the interpretation of oral composition and transmission or interpret oral epic in light of the poetic canons of written literature. This awareness is the starting point for correcting earlier misreadings of the Chinese epic tradition and enhancing the methodological consciousness of the epic research community in China (Chao Gejin 1999a). Beginning with the end of the 20th century, we have thus moved from text to the field, from tradition to inheritance, from collective wisdom to individual talent, from inheritor to audience, and from others’ perspectives to indigenous scholars’ views. Several cutting-edge topics in the field of Chinese folk literature and art have been launched as a result of this profound change in the research patterns of Chinese epic. As regards the transition from tradition to inheritance, more attention is now being paid to inheritance and diffusion in epic performance traditions and the development trajectory and ever-evolving internal drive of epic. In some traditions, epic performance is combined with ritual activities. For example, through long-term field observation and folk-life experience, Chinese scholars found that epic performance among the Nuosu branch of the Yi people of southern China has always been an integral part of oral folk exchanges and has taken place at weddings, funerals, and ancestor rituals (Bamo Qubumo 2004a). The movement away from collective wisdom to individual talent provides a correction for a limited focus on the collectivity and anonymity of folk narration. Talented individuals play a special role in the development of their group’s poetic traditions. In today’s Chinese arenas of living epic performance, there are many such remarkable individuals, including the Tibetan Drakpa and Samdrup, the Kirgiz Jüsüp Mamay, and Mongolians like Pajie, Jianbajams, Junai, Arimpil, and Qumo Yinuo. All have made definite contributions to the development of traditional oral poetics with their distinct performing personalities and styles. Responsive scholars have been moving to the field, investigating living folk practices, and tracking down specific epic bards.14 At the same time, research on singers who are the inheritors of a folk tradition is being accompanied by a focus on the audience as the receptors of the tradition in an emphasis on epic performance as a whole process of transmission and reception. The role of the audience is not that of mere passive listeners but of active participants who constitute an inseparable link in the performance process, creating and transmitting meaning along with the singer. Research by non-native scholars is being enriched, moreover, by the perspectives of the cultural consciousness of indigenous scholars. As native speakers, they can master more complex aspects of epic texts and deal with indigenous conceptual systems like the Tibetan hidden treasure texts (gter ma) (Norbu Wangden 2003) and Yi “male” and “female” texts and Black Yi and White Yi narratives

Homer to Arimpil  25 (Bamo Qubumo 2004b). Investigations carried out, moreover, from the objective perspectives of outside researchers in tandem with those from inside the poetic tradition produce a higher degree of reliability and depth. Finally, research based on field investigation augments that pursued through the reading of books and documents in establishing the canons of oral poetics (Chao Gejin 2002; Chao Gejin and Foley 2003:33–97). First-hand field investigations have led to a realization that in oral literary research, there is a need for enhanced cultural elaboration and new modes of criticism. Due to the work of a constellation of epic scholars with solid foundations, a wide vision, open and creative practices, and the support of cross-linguistic research efforts, the indigenous reconstruction of Chinese poetics has gradually become integrated into international academic dialogue.15 Over the years, the theoretical achievements of Western oral poetics and folklore have been introduced in China through the systematic translation and introduction of the three main current schools of folklore research: Oral Formulaic Theory, Ethnopoetics, and Performance Theory.16 At the same time, local research in China has played an important role in the development of epic theory and criticism in general. In their reflections over the years, local Chinese scholars have made remarkable advances in several respects: The creation of analytical patterns for epic syntax; the establishment of field re-certification of working models for extant texts (Chao Gejin 2000); and the scrutiny of format disadvantages in the production of folk texts based on the presuppositions and field framework of the “five presences” of epic tradition, performance events, performers, audiences, and researchers (Liao Mingjun and Bamo Qubumo 2004). These five modes of “being present at the same time” are needed to establish the narrative and performance context as an interactive field of vision for appreciating the inter-subjectivity of the research terrain and building a workable relationship between the research object and the researcher. Another advance made by Chinese scholars lies in the re-examination and fresh description of the ancient Chinese classics from the perspective of oral traditions and the implications of oral folk performance (Yin Hubin 2002). All this demands detailed depictions of a specific singer or groups of singers after long-term tracking, such as the Institute of Ethnic Literature of the Chinese Academy of Social Sciences has pursued by setting up ten traditional oral field research bases among ethnic peoples in western China. Worth noting, too, are the designing of institutional and individual work guides for scholars17and the establishing of a database in line with new technical specifications and concepts for collecting, sorting, and digitally processing audio-visual documents.18 In recent years, great advances have been made in epic philology in terms of the collation and publication of scientific editions, many of which have won acclaim by peers in China and beyond.19 These academic achievements have had a positive influence on international epic circles. On the whole, advances in epic research based on specific folk cases as the technical route and the ideas of oral poetics as the conceptual reference have emerged one after another, indicating that the rationale of Chinese epic research is

26  Critical reflections on epic studies becoming increasingly clear.20 The direction is toward empirical research drawn from both field and text. Most studies are grounded in accurate cultural description and oral poetic interpretation that dig deep into the epic inheritance and oral communication traditions of Chinese ethnic minorities in the current period of social transformation. They are establishing a local, indigenous foundation for multi-interactive investigations of ethnic narrative traditions, folk practices, and the present situations of epic inheritors. The current aims have been stated as follows: 1) Understand how to study features of the traditional cultural systems, reality contexts, and evolutionary tracks of oral narrative through fieldwork and ethnographic expression in line with the perspectives of contemporary Western folklore studies. 2) Become familiar with the conceptual tools and theoretical backgrounds of Western epic scholars studying Chinese epic and understand typical cases of overseas epic research in terms of the fusion of multi-disciplinary perspectives, the conceptual framework of oral poetic analysis, and the evidence for new theories. 3) Become aware of the fact that only when we go to the field and take oral folk texts as our focus do we realize the transformation in research from written to oral and truly establish the independent status of folk literature (Liu Zongdi 2004). Present-day Chinese scholars are having success in achieving these aims through the various current transformations in research perspectives and methods: Goaded on by both oral and textual problems, we have emerged from the pattern of literary research with the written text as the reference frame. Taking epic traditions, rather than epic works, as the basic starting point of oral narrative research, we have broken out of the historical research mode of Soviet folklore studies. Using oral poetics and formulaic syntactic analysis as the interpretative framework, we have broken away from research on the structure and function of narrative that earlier Western epic scholars favored in the analysis of Chinese epic texts. Based on a bottom-up approach, present-day analytic interpretation rests on the metanarrative of inheritors, audiences, communities, and even the family of the epic inheritors and the field experience of the researchers. The recent efforts of Chinese scholars have made it possible to shift from being consumers of Western theories to producers of indigenous theoretical approaches. In dialogue, discourse, field research, and text production, our scholars have fostered scholarly directions based on native practices. Translating basic concepts of Western oral poetics and then transforming them by combining them with the classification system of local oral ways of thinking and the traditional expressive vocabulary of folk narrative, we are refining the terminological system and conceptual tools of Chinese epic research. At the same time, in terms of methodology, we have achieved rational operational procedures for field research and text

Homer to Arimpil  27 production (Liao Mingjun and Bamo Qubumo 2004). The present author has used the conceptual framework of the three modern schools of folklore for explaining the traditional modes of expression of Mongolian epic. Yin Hubin has produced multi-faceted insights on the integration of Western epic concepts in Chinese traditional oral traditional practices. Bamo Qubumo has refined our understanding of textual properties and narrative structures in terms of the “five presences” by drawing upon an integrated perspective of folk wisdom and academic concepts of semantics and pragmatics in a way that produces an integrated East–West construction of epic theory.

The study of Chinese epic poetry in the 21st century Jangar’s Bumba, An earthly paradise, The people always young, Just twenty-five years old, No aging, no withering, no dying. (Prelude to Mongolian epic Jangar) Few countries in the world boast oral epics as rich, complex, diverse, and ancient as those handed down in China. Epic bards are not only inheritors and discriminators but also creators and preservers of our epic traditions. Due to human and material limitations, however, and the irreplaceable losses of time, we have failed to get to know many outstanding singers and hear their lovely voices. Oral epic has been battered by the huge shock of China’s rapid modernization and the development of the country’s western regions; and the number of epic singers has now become sharply reduced. At present, most of those who can chant the Three Great Epics Jangar, Gesar (Geser), and Manas suffer from old age and infirmity. There are many talented inheritors of the traditions among the folk; but for various reasons, the epics they sing have not been recorded. Some famous singers have died; and with no successors, epic transmission is fast weakening. If not quickly rescued, many historic poems of our ethnic nationalities that have been passed on for centuries will disappear forever, with irreparable loss to our national cultural heritage. It is urgent that we give due attention to the protection of epic traditions, support singers, recover texts, and promote epic research. Epic is often regarded as the bearer of national spirit and the source of a nation’s literary tradition. The Encyclopedia of Traditional Epics published in the United States in 1994 lists nearly 1,500 kinds of epics worldwide (Jackson 1994); and this is, no doubt, a very conservative and incomplete estimate. It is reported that the epic heritage of Vietnam is quite rich, with some sixty-two volumes of seventy-five kinds of epics of the Central Highland collected and published by the Vietnamese Academy of Social Sciences from 2002 to 2007; but as yet, it has not been brought to light in the outside world. Study of the epic traditions of Africa

28  Critical reflections on epic studies is confined to just a few regions. China’s epic census is still underway. When more complete information is gathered, the richness of Chinese epic will become apparent. As the eminent folklore scholar Zhong Jingwen once said at the turn of the century: We have an extensive collection of philological discussions of some special topics; but the overall theoretical exploration is not systematic or in-depth, and it is here that we can continue to make progress. In particular, in the past two or three decades, we have focused our work mainly on the collection, recording, sorting, publishing and other basic research tools; and we have zeroed in on specific works. There is still a lack of vertical and horizontal linkages and macro thinking. This lack restricts the vision of theoretical research, with the result that our perception of Chinese epic is like “seeing trees, but not the forest.” If this situation does not change, we will slow down in the disciplined construction of Chinese epic.21 With relation to international epic research, the development of research in recent years in China is far from satisfactory. We have not yet completed the nationwide survey of singers and performers or come up with a systematic, indepth description and analysis of inheritor types, genealogies, and the distribution of inheritors. The number of inheritors and the construction of archives for directional tracking are far from meeting strict requirements. As for epic texts themselves, there is much room for improvement in the existing links between collation, translation, preservation, publication, and exposition. As regards enlightened theories, the analytical explanations that our discussions have commonly come up with are not commensurate with the diversity of the Chinese epic tradition. Descriptions and interpretations from the cross-cultural spectrum and rooted in multiple resources are far from being in place. As yet, there are no objective theoretical definitions and interpretations of the three major types of epic in China—creation, migration, and heroic—to say nothing of sub-types like flood epics. The systematization of conceptual tools, terminologies, theoretical methodologies, and research paradigms needs to be further refined. The thrust of epic studies in the north and south is uneven. The momentum of increased case studies far exceeds any sort of comprehensive, macro understanding. The mapping of an epic hierarchy and the cultivation of cross-lingual talents are far behind the goals we have set. As regards controlled institutional evolution, much work remains to be done in the field base, digital filing, information sharing, integration of resources, cooperation mechanisms, graduate education, international exchanges, etc. For the formation of a productive epic academic community, we still need to break out of barriers between academic disciplines and further strengthen dialogue and exchange among researchers of folklore, folk literature, classics, linguistics, and anthropology so as to open up the scope of vision and inclusiveness and truly realize the transformation of the academic paradigm. The systematic structuring of Chinese epic is a long-term project. How can we make the voices of Chinese

Homer to Arimpil  29 scholars heard internationally? How can we encourage more epic inheritors of all ethnic groups to maintain the traditional epic cultural ecosystem? How can we prevent the polyphonic melodies of Chinese epic from withering with age and even dying? These questions urgently need answers. This chapter was originally published as an article in the Journal of the Institute of Literature of the Chinese Academy of Social Sciences (Beijing: China Social Sciences Press, 2008). 1–39.

Notes 1 Here and elsewhere in this book, some references to or quotations from an author are from a Chinese edition of the author’s work, sometimes published after his or her death; and some quotations are not from the original English text but a re-translation from the Chinese. In some such cases, the page and other bibliographical data are lacking in the Chinese edition and cannot easily be located. 2 In this vein in China is the Preface to Chen Zhongmei’s translation of the Odyssey (Huacheng Press, 1994), which maintains that in view of some differences between Homer’s two epics, we can also make the assumption that the Iliad is a work shaped by Homer himself, while the Odyssey is a work arranged, supplemented, and pruned by some of his descendants (Homeridae) according to the words and basic patterns that Homer shaped and passed on to them. 3 For a comparative analysis of the two performances, see Albert B. Lord, The Singer of Tales, chapters four and five. 4 Serbo-Croatian Heroic Songs, about 1,500 hours of recordings collected by Parry and Lord, is kept in the Milman Parry Collection of Oral Literature at Harvard University. 5 Maria Vasenkar 2002. “A seminar commemorating the bicentennial of Elias Lönnrot’s birth, April 9, 2002.” FFN 23, April 2002: 2–4. 6 Ibid. 7 The American scholar Mark Bender has given an enlightening introduction and discussion of this in “What to Do about Meige: A Tradition-oriented View of Chuxiong Yi’s Nationality Texts,” Folklore Studies, No. 4, 2002, Chuxiong Yi being the name of an Autonomous Prefecture in central Yunnan Province of China. 8 See Gregory Nagy, Greek Mythology and Poetics (Ithaca: Cornell University Press, 1990, pp. 47–48): Pindar’s Homer: The Lyric Possession of an Epic Past (Baltimore: Johns Hopkins University Press, 1990, pp. 47–48, 52–81); Poet as Performance: Homer and Beyond (Cambridge: Cambridge University Press, 1996, pp. 74–78). For the semantic and etymological research of Hómēros, aoidós, and rhapsodes as all pointing to the meaning of “compiling songs,” see Homeric Questions (Austin: University of Texas Press, 1996), chapter three. 9 See Batnasan, “An Introduction to Arimpil,” Jangar Materials (Urumqi: Xinjiang People’s Publishing House, 1988, p. 6); Jamcha, “The Epic Jangar and Jangarch in Hoboksar” in The Anthology of Jangar (Urumqi: Xinjiang People’s Publishing House, 1988). 10 www​.physics​ nyu​.edu​/faculty​/sokal​/index​ html 11 Among these, the scholars whose achievements had the greatest influence in their own countries were: in France, David Neal and R. A. Stein; in Germany, L. J. Schmidt and V. V. Radlov (long-term residents in Russia), W. Heissig, K. Sagaster, and K. Reichlof; in Russia, G. N. Potanin, S. A. Kozin, A. Rudnev, Zhamcarano, N. Poppe (who later emigrated to the United States), B. Y. Vladimirtsov, V. Zhirmunsky, and S. J. Nekljudov; in Finland, G. J. Ramstedt; in the United Kingdom, C. R. Bawden and Nora K. Chadwick.

30  Critical reflections on epic studies 12 See Collected Works of Sumpa (Gansu: Labrang Temple, 11–16). 13 It is quoted from the “preface” written by Zhong Jingwen for Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing (Nanning: Guangxi People’s Publishing House, 2000, p. 7). Zhong Jingwen has proposed forward-looking ideas on the construction of Chinese epic based on the study of southern epic traditions in “Research and Pattern of the Northern and Southern Epics of China.” See National Arts, 2002, Vol. 4. 14 Field research works on the inheritors of epic singing and their groups include: Yang Enhong, The Muses of Folk Poetry: Studies on Performers of Gesar (Beijing: China Tibetology Publishing House. 1995); Yang Enhong, On the Road: An Account of the Visit to Tibetan Singers of King Gesar (Nanning: Guangxi People’s Publishing House, 2007); Adil Jumaturdu and Tokon Isak, Critical Biography of Jüsüp Mamay (Hohhot: Inner Mongolia University Press, 2002); Chao Gejin, Heroic Songs: Past Fieldnotes on the Oirat Mongolian Epic Tradition (Nanning: Guangxi People’s Publishing House, 2004). See also Bamo Qubumo’s recent series of articles on Yi epic performers and oral interlocutors: “Narrative Context and Performing Arena: A Case Study of Nuosu Yi’s Oral Argument and Epic Tradition” and “Narrative Pavilion Construction, Text Boundary and Narrative Domain: Traditional Referentiality.” 15 In 2003, the journal Oral Tradition published “Collection of Essays on Chinese Oral Tradition,” which included thirteen research papers from three generations of scholars from the Institute of Ethnic Literature (Chinese Academy of Social Sciences [CASS]), discussing the epic traditions and oral narrative of Mongol, Tibetan, Manchu, Naxi, Yi, Kirgiz, Miao, Dong, and other ethnic groups. This was the first time the international academic community had published a collection of the oral narrative art of Chinese ethnic minorities in English, and it attracted attention at home and beyond. 16 In the 1980s, the Institute of Ethnic Literature (CASS) compiled two volumes focusing on epic research entitled Translated Collection of Ethnic Literature. This was followed by other Chinese translations of epic research: S. J. Nekljudov 1991, The Heroic Epics of the Mongolian People, trans. Xu Chenghan, Gao Wenfeng, and Zhang Jizhi, Inner Mongolia University Press; R. A. Stein 1994, Studies of Tibetan Epic and Singers, Tibet People’s Publishing House; John Miles Foley and Chao Gejin 2000, The Theory of Oral Composition: History and Methodology, Social Sciences Academic Press; Albert B. Lord 2004, The Singer of Tales, trans. Yin Hubin, Zhonghua Book Company; Gregory Nagy 2008, Homeric Questions, trans. Bamo Qubumo, Guangxi Normal University Press. The Institute of Ethnic Literature, Chinese Academy of Social Sciences launched a collection of translations of research on oral traditions by the Institute of Ethnic Literature of North American in a special issue of Ethnic Literature Research in 2000. 17 In September 2003, for example, the Institute of Ethnic Literature (CASS) sponsored the Interpretation of the Urgent Entrusted Project of UNESCO; and for the research findings, see Cultural Diversity and Identity of Ethnic Groups in West China and Beyond: Reports on Oral Traditions along the Silk Road, ed. Chao Gejin, 2008. Beijing: Social Sciences Academic Press. 18 See comments on “Building the Lifeline of Chinese Ethnic Cultural Heritage,” China Ethnic News, April 11, 2006. 19 Examples are: The Complete Book of Geser, Vols. 1–5, ed. Sechenmunhe, Beijing: Ethnic Publishing House, 2002–2008; Kalmyk Jangar (annotated edition), ed. Daburjav, Beijing: Ethnic Publishing House, 2002; Jugai Mijid, Huder Altai Khan, Naren Khan Hubun, ed. Rinchindorje, Beijing: Ethnic Publishing House, 2007; Mongolian Heroic Epics, Vols. 1–4, ed. Rinchindorji, Chao Gejin, and Sechinbatu, Beijing: Ethnic Publishing House, 2007. Beginning in 2006 under the direction of Lang Ying, Tsewang Jikmé, and Yang Enhong, the Tibetan Ancient Books Publishing House has a project of publishing over forty volumes of Gesar as Performed by Samdrup, nearly half of which have been published. Under the direction of Jiangbian Jiaicuo, Ethnic Publishing House has a project of publishing over forty volumes of Selected Books of Gesar, over ten of which have been published.

Homer to Arimpil  31 20 An example is the research results of younger and middle-aged scholars of the Institute of Ethnic Literature (CASS): Chao Gejin, Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing, Nanning: Guangxi People’s Publishing House, 2000; Yin Hubin, Ancient Classics and Oral Traditions, Beijing: China Social Academic Press, 2006; Sechinbatu, Mongolian Epics: From Formula to Metaphor, Beijing: Ethnic Publishing House, 2006; Adil Jumaturdu, Research on Epic Singers of Manas, Beijing: Ethnic Publishing House, 2006; Daburjav, Research on the Oirat Heroic Story, Beijing: Ethnic Publishing House, 2006; Huang Zhongxiang, Kazakh Heroic Epics and Grassland Culture, Beijing: Central Compilation & Translation Press, 2007; Bamo Qubumo, Epic Tradition and Field Research; Li Lianrong, Academic History of the Tibetan Epic Gesar (to appear). Other achievements in the new trends in research on epic and oral traditions are those of Dr. Dawataya on Mongolian epics and Chen Ganglong on the manggus and storytelling art. 21 It is quoted from the “preface” written by Professor Zhong Jingwen for Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing (Nanning: Guangxi People’s Publishing House, 2000, p. 7).

2

The history of epic research

The term “epic” as used by Western literary critics refers to a poem that generally conforms to the following standards: It is a long narrative poem that deals with serious subjects in a noble style; the protagonist is a hero or semideity whose behavior embodies the life of a tribe, a nation, or even the whole of mankind; epic stories have the magical feel of fantasy, but some draw directly from or describe real historical events. In recent years, the academic community has gradually formed a consensus on epic classification as falling into the following three categories: oral epic; or oral-derived epic; and tradition-oriented, semi-literary epic, which has its slot between the first two.1 Though the earliest epic text we have today is Gilgamesh, Homer’s epic poetry is generally recognized as the origin of Western literature. The texts of the Iliad and Odyssey are said to have been formed in the 6th century BCE (Nagy 2008) and came to be widespread throughout the Greek Peninsula. In the annual Panathenaea, bards competed in singing them; and there were probably hundreds of versions popular at that time in their gradual, complex process of finalization (Foley 1990:24; Chao Gejin 1999b:181). In the Western tradition, no one has ever gained such a high reputation as Homer; but Beowulf, the Song of the Nibelungs, and the Song of Roland have also come to be ranked as masterpieces. The close relationship between Indian and European languages and cultures has also aroused the interest of Western scholars in the Indian Mahābhārata and Rāmāyana; and with the Far East, Africa, the Americas, and Oceania becoming a focus of scholarly study following the European Enlightenment, the epic traditions of Altaic-language peoples and the civilizations of East and Southeast Asia all became objects of research. Accordingly, international scholarly perspectives on epic as transcending the traditional Western models gradually gave birth to new norms, new scales, and a new study atmosphere.

Classic Western epic studies Beginning with ancient Greek commentaries on the Homeric epics, the Western study of epic over the centuries has made great strides in understanding both the nature of epic poetry in general and the social history of specific epics, mainly through textual and philological research. Both Socrates and Plato discussed DOI: 10.4324/9781003258001-3

The history of epic research  33 Homeric epic, but Aristotle was the first to give a systematic discussion of the characteristics of epic poetry. His Aporēmata Homērika (Homeric Questions) has been lost, and other such studies are extant only in catalogues. In Peri Poiētikēs, however, Aristotle compares the two most popular poetic genres of the time, epic and tragedy. Several later critics from the 1st to the 4th century CE expressed their admiration for literary works of “epic imitation,” but most of their discussions lack new insights. Critics of the 4th century like Maurus Servius Honoratus and Macrobius Ambrosius Theodosius gave high praise to Virgil and later literary epics. Critics in the Middle Ages likewise have left few incisive insights on epic; in some respects, they even went backward. Isidore in Etymology (6th century) and the author of the Suda literary encyclopedia (10th century) had no clear point of view on the genre. They either confused epic with history or regarded it as a mixed type of drama and narrative. In the Italian Renaissance, Vida’s Art of Poetry (1527) also manifested no progress in epic theory. It was not until Aristotle’s Peri Poiētikēs was rediscovered that scholars began discussing theoretical questions about epic. In his Poetics, Trissino (1528–1563) quoted important passages from Aristotle as the standard for elaborating on the achievements of Italian literature up to that time. Later, Castelvetro’s Interpretation of Aristotle’s Poetics (1576), Tasso’s On the Heroic Epic (1594), and other works discussed epic from many aspects. Although these discussions offer some insights, on the whole, they are relatively narrow and made little contribution to a deep understanding and interpretation of the genre. In 17th- and 18th-century France, the achievements of the Jesuit Mambrun’s On Epic Poetry (1652), Rapin’s Reflections on Aristotle’s Poetics (1674), Boileau-Despréaux’s Art of Poetry (1674), and Bossu’s On Poetics (Traité du poème épique, 1675) should be mentioned. These discourses were influenced by the idealism of that time and aimed to draw up norms for epic while emphasizing moral principles and religious ideals. The only work that had substantial influence on later generations is Voltaire’s Essay on Epic Poetry (1733). Among English works of that era, scholars have recognized the importance of Hobbes’ Preface to his translation of the Iliad and Odyssey (1675), Dryden’s “Apology for Heroic Poetry and Poetic License” (1677), and Mulgrave’s “Essay on Poetry” (1682). A few scholars also mention Blackwell’s An Enquiry into the Life and Writings of Homer (1736) from time to time, and very few mention Lord Kames’ Elements of Criticism (1762) and Hayley’s An Essay on Epic Poetry (1782). As for the study of epic theory in Germany in the 18th and 19th centuries, the achievements and influence of Goethe, Schiller, Schelling, and Hegel are the most outstanding. Goethe and Schiller co-wrote On Epic and Dramatic Poetry (1797), but Schelling’s Philosophy of Art (1802) and Hegel’s Lectures on Aesthetics (1835) offer more in-depth explanations of epic. Quotations from Hegel are often found in 20th-century Chinese studies on epic theory, some of which are based on the theoretical foundation that he laid. However, some Chinese scholars have been haunted by his assertion that there is no epic in China (Hegel 1997:170). Though China has a multitude of epics from its minority ethnic peoples, China

34  Critical reflections on epic studies as a whole has no national epic, because our main form of literary reflection is prose and because religious views are not generally thought suitable for poetic expression. To make up for this lacuna, some legends and later novels are highly developed, vividly depicting situations of public and private life in ways that are delicately and colorfully expressed, especially in describing aspects of women’s psyche.

Modern epic studies As we saw in the previous chapter, since the start of the 20th century, with the deepening and refining of the study of humanities, epic research has taken on new dimensions. The classic philological approach has further developed and yielded a great deal of valuable results; but new approaches have begun. As early as 1908, the Danish folklorist Alex Olrik (1864–1917) started enunciating his Epic Laws of Folk Narrative, a set of basic principles of folk narratives formed in the light of a large number of different types of materials. In mid-century, C. M. Bowra’s large-scale summaries of the world’s epic tradition in From Virgil to Milton (1945) and Heroic Poetry (1952) are quite remarkable; but his research hardly gets out of the rut of the traditional literary perspective. More representative of the new trend is Lord’s The Singer of Tales (1960), marking the formation with Milman Parry of the Oral Formulaic Theory. The London Seminar on Epic, which gathered together the world’s most influential epic scholars from 1964 to 1972, yielded a wealth of theoretical achievements, such as Traditions of Heroic and Epic Poetry (1980). The journal Oral Tradition, the core publication of research on oral traditions, was first published in 1986 in the United States. North America has taken the lead in this trend, with generations of scholars from Harvard forming a continuous line: Milman Parry, Albert Lord, Gregory Nagy, and Indian epic, the founder and editor-in-chief of Oral Tradition. In Europe, the Finnish epic expert Lauri Honko and the German scholar Karl Reichl took the leading position in this direction. As we have seen, in the 1930s, Milman Parry (1902–1935) began seeking to answer the eternal question of who Homer was and when and how he created what we call Homer’s poetry. His investigations into the Homeric Question along with his student Albert Lord (1912–1991) caused a major controversy in the field of classical studies. The achievements of philology in the 19th century, especially for German literature, and the methods of Western anthropology, especially the fieldwork of the German-Russian scholar V. W. Radlov and Matija Murko, opened up Parry’s way of thinking He realized that Homer’s poetry must be oral. To prove the reliability of this inference, Parry and Lord conducted their field surveys in Yugoslavia and confirmed Radlov’s statement that in certain types of narrative folk singing, no two performances are exactly the same. They came to realize that every time folk bards sing, they generate a new work, but by relying on traditional formulaic themes, typical scenes, and story-patterns; and they concluded that Homer’s poetry is the product of such a tradition.

The history of epic research  35 The middle of the century was a period of vitality in the humanities. About two years after the appearance of Lord’s The Singer of Tales, scholars in France, Britain, the United States, and Canada published in less than twelve months four important works, which coincidentally focused on the nature of oral culture and its relationship with writing technology: The Gutenberg Galaxy (1962) by the communication scholar Marshall McLuhan; La Pensée Sauvage (1962) by the structural anthropologist Claude Lévi-Strauss; The Consequences of Literacy (1963) by the social anthropologist Jack Goody and writer Ian Watt; and Preface to Plato (1963) by the classicist Eric A. Havelock. These discussions on the relationship between oral tradition and writing technology provided a fresh theoretical fulcrum and a broader perspective for the later study of oral epic. On October 29, 1963, a seminar entitled “History and Epic: Their Mutual Relationship” was held at Queen Mary College, University of London, at which a series of lectures on epic were given; and the London Seminar on Epic was founded. In a triangular epic research mechanism established among London, Oxford, and Cambridge, scholars studying different traditional epics came to read their research results and communicate with their peers. The founder and chairman of the seminar, Arthur T. Hatto, professor of Germanic Studies at the University of London, later published a review of its activities and achievements (Auty and Hatto 1980). The seminar linked the academic work of predecessors in the epic community, including the founder of Turkic studies, V. W. Radlov, the classical scholar Domenico Comparetti, the Soviet scholar V. Zhirmunsky, and the British Cecil M. Bowra. It was not possible to include all outstanding epic scholars in the seminar; but it had twenty-three members and invited guests, including Parry and Lord as senior scholars. Participants presented thirty-one academic reports covering a territorial range described as Iceland-UgandaThailand-Mongolia, which included most of Europe and Asia.2 Today, the seminar’s academic activities remain representative of the level of epic studies in the 1960s, in particular in the establishment of academic paradigms and the tracking of important issues. Harvard University is not only the center of American classical scholarship but also the cradle of folklore studies and birthplace of the Oral Formulaic Theory. Since the middle of the 19th century, the study of oral literature there has formed a century and a half of academic tradition through the inheritance of four generations of scholars, including Francis James Child, George Lyman Kittredge, and of course, Parry and Lord. The baton is now in the hands of Gregory Nagy (b. 1942), professor in the Classics Department, well-known Greek literature expert, oral poetry theorist, and Indo-European phonologist. A representative of the fifth generation of Harvard scholars and a major theoretician in interdisciplinary research on oral traditions, his research involves ancient Greek literature, mythology, rhetoric, philology, comparative literature, oral narratology, and many other fields, especially those based on the interaction between the ancient classics and oral traditions. He pays close attention to the comparative study of the oral inheritance of ancient Greek poetry and the restoration of its performing tradition; and he has made remarkable academic achievements in the expansion of the theory of oral

36  Critical reflections on epic studies poetics, which has had far-reaching influence beyond the scope of classical studies and Greek cultural studies. The founder of the academic journal Oral Tradition in 1986, John Foley (1947–2012), also shared in the authoritative American tradition of research on oral literature and epic. His work covers the range of ancient Greek epics, Medieval Anglo-Saxon literature, and the Yugoslav oral tradition. In 1986, he established the Center for Studies in Oral Tradition at the University of Missouri and for many years was the leader of international research on oral traditions. In his early years, he was keen on the Oral Formulaic Theory and put forward many new ideas. In his later years, he devoted himself to improvements in the laws of oral poetics. He creatively applied the most important folklore theories of the century, such as narrative ethnography, performance theory, and ethnopoetics, to the comparative study of oral traditions; and he systematically put forward a comparative law of oral tradition, performance arena, traditional referentiality, etc. with the intention of constructing an original system of oral poetics and an analytical method for the study of oral poetry texts. Continuing the work of the London Seminar, Oral Tradition has taken up the seminar’s mission in the study of oral traditions. The Finnish professor of folklore studies and comparative religion Lauri Honko (1932–2002) was another epic expert active in the 20th century. Much of his early work was concerned with comparative religious and folk literature, including folk medicine, folk beliefs, myths and rituals, elegies, and the history of Finnish folklore; and his later fieldwork and research in India focused on analytical issues of oral poetry. As we saw in Chapter 1, he reevaluated the role of the Kalevala as a traditional epic. His research on this semi-literary epic, the mental texts of epic singers, and the relationship between epic and national identity has been widely recognized by the academic community.

Chinese epic studies This section briefly recalls some of what was related in detail in the final section of Chapter 1. The first scholar to discuss epic in China may have been the Mongol or Tibetan monk of Youning Temple in Huzhu County, Qinghai Province, China, Sumpa Yeshé Penjor (1704–1788). In 1779, he and the Sixth Panchen Lama (1737–1780) exchanged letters about Gesar; and he later collected relevant topics into a volume entitled Questions and Answers about Gesar (Zhao Bingli 1990:286–290). He once said that “The Legend of Geser is not a Buddhist classic, but a biography of historical figures, Geser himself being a king in Ando.” In modern times, Zhang Taiyan (alias Zhang Binglin, 1869–1936) was probably the first to comment on epic. Later, the essayist Hu Shi (1891–1962) referred to epic as “story poem.” Among those who published early introductions to epic was Ren Naiqiang (d. 1989), who visited the Xikang area of Tibet for a year in 1929 and in a December 1930 issue of the Sichuan Daily published “Tibetan Three Kingdoms”

The history of epic research  37 and “Story of the Tibetan Three Kingdoms” to introduce Gesar. These 20th-century writings, however, actually came to light relatively late. Since the 18th century, Western scholars have studied the epics of Chinese ethnic minorities. The earliest mention appears in Compilation of Mongolian Historical Materials (St. Petersburg 1771–1776), Vol. I by the naturalist P. S. Pallas. Based on time spent in Kalmykia from 1804 to 1805, the German traveler Bergmann later recorded two fragments of Jangar in The Travels of Benjamin Bergmann among the Kalmyks (Benjamin Bergmann’s nomadische streifereien unter den Kalmuken in den jahren.C. J. G. Hartmann, 1804–1805, 4 vols). Since about 1851, the study of Mongolian and Tibetan epics in Western Europe has been booming. In brief, monographs of W. C. Schott and others were published successively in Abhandlungen der Berliner Akademie; and later, more advanced research came out in Britain and Russia, including the works of Radlov, Potanin, and Ramstedt. In the early 20th century, representative scholars of the epics of ethnic peoples in western China were the Russian V. Zhirmunsky and later, Chadwick and the founder of the London Seminar, Hatto. In recent years, the German scholar who has made the most outstanding contribution to the discipline is Karl Reichl (b. 1943), whose exemplary work is Turkic Oral Epic Poetry: Traditions, Forms, Poetic Structure (1992). In the vein of traditional Turkic research, he attaches importance to language materials and fieldwork; but he has also absorbed rich nutrients from the Oral Formulaic Theory and other recent theories, thus forming a broad vision for discussion based on solid data. In the field of Altai Mongolian epics, most of the world-renowned figures are from Russia: first Nicholas Poppe (1897–1991), who moved to the United States during the Second World War, and then S. J. Nekljudov and others. Among scholars based in western Europe, the first was the German Walther Heissig (1913–2005), whose in-depth research on Mongolian epic motif series has been far-reaching. Proficient in Mongolian, he was very familiar with native materials; and he was awarded the title of Academician of the British Royal Asiatic Society for his outstanding contributions to Mongolian studies. The study of the Chinese epic provides valuable insights for the construction of international epic theory in several respects. First of all, it enriches the body of epic types with, for example, the creation and origin epics of various ethnic peoples in southern China. Moreover, the diversity of cultural types and modes of epic inheritance, as well as the large number of still living examples of epic transmission, have aroused a new awareness, new questioning, and new efforts to interpret epics. Taking Tibetan epic as another example, the acquisition of the ability to perform epics as “dream-taught” singers and the diversity of text types like the “hidden treasure texts” have contributed to a more comprehensive understanding of epic. Similarly, there are the Yi people’s “male” and “female” epic texts, and Black Yi and White Yi narrative have added to the depth of explorations of oral epic performance. In short, the Chinese epic tradition provides a variety of fresh

38  Critical reflections on epic studies materials through which to observe the formation and dissemination of epics as well as variations and transformations in textualization, epic styles, and the roles of epic singers, thus helping to firm up the foundation for the development of epic theory.

The 20th-century sea change in epic studies We have here deliberately neglected mention of research on the epic traditions of specific ethnic groups. We have simply highlighted the exemplary figures who have contributed to epic theory and the methodologies of grappling with epic as a major form of human expressionist culture. There is as yet no approximate estimate of the number of epics sung around the world, and research findings are too numerous to cite individually. From a macroscopic perspective, in 20th-century epic research, we find a refreshing shift in the interpretation of some basic issues. Since the London Seminar on Epic, the scope of the definition of the genre has become more open. The past notion of taking Western standards as the ultimate criterion has been corrected. In addition, the approach to epic from the literary angle has been replaced by a variety of new perspectives. From the point of view of folklore and anthropology, epics are now regarded more as a tradition than as just individual works. Oral epic performance, moreover, is understood as a kind of comprehensive folk activity, with complex operational modes and social mechanisms, not just simple performance designed to please the audience. Our present deeper understanding of epic texts is due in large part to the incorporation of the concept of oral poetics. What is the relationship between the performance of an epic narrative and the singer’s mental text? What is its relationship with other epic performances of other singers? What is the relationship with other genres, such as myth, lament, elegy, lay, or legend? What is the intertextual relationship among these genres, and what are the characteristics of intertextuality itself? On the one hand, the pursuit of these issues and the reconciliation of conflicting notions about them have been strongly colored by 20th-century academic trends in general. On the other, the rethinking of basic ways of judging the significance of epic and oral folktales has subverted the study of classical poetics. What are “large words,” poetic formulae, and a verse line? What in a typical scene and story-pattern allows it to be channeled and re-positioned in the course of a performance? Such questions are now put forward more strongly and directly than before. Understanding the whole process of oral epic creativity, from the generation of meaning to transmission and audience acceptance, as taking place in the same time and space brings new perspectives and opportunities for the interpretation of oral epic production. An earlier version of this chapter originally appeared in World Literature, Vol. 5. 2008.

The history of epic research  39

Notes 1 As is well known, the most internationally recognized oral sourced epics include the Babylonian Gilgamesh, the Indian Mahābhārata and Rāmāyana, the Iliad and the Odyssey, the Anglo-Saxon Beowulf, the Germanic Song of the Nibelungs, the French Song of Roland, the Iceland Edda of Iceland, and the West African Epic of Son-Jara. Among literary epics are Virgil’s Aeneid, Tasso’s Jerusalem Liberata, Dante’s Divine Comedy, and Milton’s Paradise Lost. A typical semi-literary epic is Finland’s Kalevala. 2 The areas covered included Serbo-Croatia, ancient Greece, Russia (Europe and Asia), Spain, Finland, ancient Sumer, ancient France, southeastern Europe, Romania, the Middle East, ancient Arabia, Swahili-speaking regions of Africa, ancient India, Thailand, Mongolia, Japan, and others.

3

Current issues in epic research

Epic study As we have seen, epic studies have made great progress in the decades since the London Seminar on the Epic in the 1960s–1970s. This is also the case in China, where the situation has changed a great deal since research on Chinese epic began to emerge in the 1980s. The main aim of epic research in general is to analyze and throw light on epic poems’ cultural connotations, artistic characteristics, methods of transmission and scope of distribution, etc. In recent years, international epic research has undergone substantial developments by using the analogical approach, studying the ancient based on analogies with the present. In the study of the Homeric epics, Beowulf, The Song of Roland, The Poem of the Cid, The Song of the Nibelungs, and Shi Jing (Classic of Poetry), we lack sufficient information on the past to assess their creative inheritance from historical angles; but we can learn about them from their more recent equivalents in the genre. The study of living epics such as Manas and Gesar from China can help us understand more deeply aspects of the Odyssey, Beowulf, and other epic classics and answer questions that are difficult to solve through the classic literary approach. The classic “Homeric Question” and “Homeric Questions” have a long history. Western scholars have for centuries conducted epic research with the tools of philology, textual analysis, and literary criticism; but these tools have their limitations. It is not always easy to fully answer questions by using literary methods to analyze the epic tradition simply as a body of poetic works. In the study of the Homeric poems, for example, classical theory always takes the view that “the plot starts from the middle,” while “the opposite is true of characterization.” This type of literary research to a certain extent answers how and why Homer’s poetry has had such a profound impact on the whole of European literature. Despite its being the mainstream of epic studies for centuries, however, due to limitations of technical means, cognitive level, and methodology, purely literary approaches to the study of epic have not been able to fully solve some problems, most especially the issue of Homer as a single poet versus several Homers. Arguing from their own classical standpoints, neither Analysts nor Unitarians have been able to deal with the crux of the problem. One reason is that they have not taken the analogical approach (Foley 2000b). The revolutionary application of DOI: 10.4324/9781003258001-4

Current issues in epic research  41 analogy in the field of epicology benefited from developments in cultural anthropology and has borne fruit in combination with the classical literary and philological approaches. Outstanding in pioneering the application of analogy is Milman Parry, who used an analogy with present-day epic singing in Yugoslavia to demonstrate that Homeric poetry must have grown out of an oral tradition and that Homer was not a genial poet working alone but a large group of bards. This “Oral Homer” view was a thunderbolt that shocked classical scholars who adore Homer with the need to accept the idea of a great blind poetic genius being replaced by a group of obscure, illiterate singers. Modern research shows that no matter what the linguistic tradition, there are individual bards with outstanding skills who become distinguished representatives of their special group and worthy of laurels. Albert B. Lord could thus praise Avdo in the Balkans as the “Homer of our time” (Lord 2004:69). Recent epic studies have also given rise to sometimes mind-boggling answers to questions in other disciplines, such as how the brain stores, synthesizes, invokes, and yields information. Some scholars initially believed that the memory limit of the human brain is limited to about 4,000 verses of poetry and on this basis asserted that Homer’s poetry, which adds up to about 30,000 verses, could not possibly be the product of an oral tradition. However, Parry and Lord, by their field studies of heroic songs in Yugoslavia in the 1930s, showed that an illiterate singer can perform thousands of verses smoothly without the help of a written text. Through the study of rhyme in narratives thousands of verses long, moreover, scholars have been able to understand how oral folk bards can use formulae, typical scenes, and story-patterns as memory units—the “large words” of Yugoslav folk poetry—to remember particular verses and compose large-scale narratives on the spot (Foley 2000b). Such in-depth epicological study tackled the mystery of the oral arts developed by human beings in the long evolution of civilization, triggering the debate about oral and written in the 1960s. Research on human thought processes and culture, especially the relationship between thinking and its material manifestations, has not yet found thorough explanations for the questions involved in that debate. Nevertheless, the fact that human beings have two paths for the transmission, accumulation, and evolution of knowledge is well established. One is the writing systems that began around the world perhaps 7,000 years ago; the other is human language, with a history of about 200,000 years (Kress 1996). The latter is even more amazing in its quantity, the total amount of information contained in any one language being far beyond the reach of writing technology. Recent studies have shown that human beings have not paid due attention to such issues as the co-evolution of spoken language and thinking and the heights achieved by oral arts in the development of civilization, let alone conducted in-depth research on them. The recent studies of oral epic have made a major contribution to dealing with these issues. The study of oral epic prods us, moreover, to understand literature itself in a fuller way. Textbooks tell us that literature is the art of language and that its main function is to understand and grasp the world artistically while giving aesthetic

42  Critical reflections on epic studies pleasure. Myth scholars argue, however, that although mythic literature is full of fanciful beauty, its function is mainly to explain “why is it so,” why there is only one sun in the sky or why the sun and the moon need to appear alternately. In other words, mythology charts the existing world order, explaining the source and status of myriad phenomena. For its part, the study of the epic suggests that in most cases, epics are first of all literature, but they are often something else as well. The chanting of the southern China Miao people’s King Yalu at funerals makes it indispensable at a turning point in the life process; and the singing of the Yi epic Hnewo at a wedding gives it the status of a symbol of the transformation of life. In such cases, epic is regulated and restricted by ritual activities. Sometimes, too, as with the Jingpo Dance of Munao, epic chanting is a group performance, the basis of large-scale dance. In that way, collective physical group movement bolsters understanding, memory, and communal participation, bringing to life the complex cultural symbols handed down in the community. The complex forms and functions of these epics tell us that epic cannot simply be understood as a series of symbols printed on paper for enjoyment as literature. In many traditions, epic is life itself, a mode of life and an indispensable substance of human existence.

The definition of epic The Oxford Dictionary of Literary Terms (Baldick 2001), the more professional The New Princeton Encyclopedia of Poetry and Poetics (Preminger, Brogan, and Warnke 1993), and even some recent literary research publications continue to define epic as a literary genre in the way academic circles have long understood it. Western epic studies have had great achievements; but in the definition of epic, they have mainly focused on static, morphological descriptions, defining it as sacred narratives about gods or heroes in a sublime style and therefore sacred. Of course, an epic is a story. Of course, it is about sometimes god-like heroes, halfhuman and half-divine, not the everyday lives of ordinary people. It deals with grand themes like the destiny of a nation or all human beings and so adopts a high, commonly poetic style; and it is performed by professional bards. Nevertheless, in defining epic, it is necessary to go beyond purely literary aspects and take into consideration the relation of epic to myth, history, and various social factors. Like myth, epic is a cross-breed kind of narrative. In the 800 or so years that it took to complete the Mahābhārata—from the 4th century BCE to the 4th century CE—legends, folk tales, folk songs, genealogies, and other genres suited to royalty became wrapped into a large, not so clearly defined “crossgenre” or “super-genre.” European epics like the Iliad, The Song of Roland, and The Poem of the Cid have distinctive historical imprints. The archaeological discovery of the ruins of the Trojan War in Turkey proves that the Iliad depicts events that more or less actually occurred; and later, discoveries of giant wooden containers indicate that the Trojan horse story may have some historical basis. If we examine the concept of epic from the hundreds of living forms of epic in China, we find that consideration needs to be given to content, narrative scale, social dimensions, cultural significance, performance strategies, and modes of

Current issues in epic research  43 transmission. Epics from the various regions of China are extraordinary in their diversity of form, range of distribution, and overlapping of the three main epic types—creation (with its mythic dimensions), migration (with its historical resonances), and heroic. There are epics from the fishing and hunting traditions of the Tungus Manchu in the northeast, from the Mongol nomadic traditions on the Mongolian plateau, and from the Turkic traditions of Xinjiang and the QinghaiTibet and Yungui Plateaus. There is the mega-epic Gesar of Tibetan areas and the creation and origin epics of the mountainous farming areas of Yunnan, Guizhou, and Guangxi (Chao Gejin, Yin Hupin, and Bamo Qubumo 2010). The overlapping of creation, migration, and heroic is clearly exemplified in the epic traditions of southwestern China. In the recently discovered Miao epic King Yalu, the first part focuses on creation myth—the birth of cosmic stars and origins of gods and human beings; the second part recalls what were originally historical migrations—how King Yalu avoided disputes with his brothers and led his people through the hardships of migration into many diverse areas. As the epic continues, moreover, it recounts the hero’s struggles with foes who chase after him along the way and fierce battles with other enemies (Chinese Folk Literature and Art Society 2012). Hegel has said that war is the most suitable soil for heroic epic; and early Chinese scholars thought of epics as poems about history, the 20th-century essayist Hu Shi once translating “epic” as “history poem.” However, this definition surely does not fit Mongolian epics, which are perhaps the least historical of epics. Some scholars have argued that the name of the protagonist of Jangar is derived from the hero Jahangar, who was active in western regions centuries earlier; and some maintain that the prototype for the protagonist of the Tibetan epic Gesar is a historical tsenpo (emperor). But whatever historical figures may be reflected in the heroes of such works, they have been deliberately abstracted and symbolized; and the enemy is a demon manggus with multiple heads. Similarly, Mongol heroic epics contain specific place names like the Altai Mountains and the Manas River; but the action takes place in a kind of Viconian poetic geography that cannot be plotted in terms of historical realities. Jangar speaks of “the wood of the Eming River, which is made into the wings of the saddle”; but that river’s name is clearly fabricated purely for poetic rhyme, with no such river being found in Mongolia. Mongolian epic treats everything in a virtual mode in tales based on antipodes of good and evil, friends and enemies. Its purpose is to generalize the heroic spirit without having to adhere to actual characters or events. Which generals of history went where and which nations were defeated form a distant vague backdrop in the collective memory of the people; but on the epic stage, they fight only semi-deified heroes and their demonized adversaries. In a word, epics exist in multifarious complex forms, and any definition and research must be based on a wide range of materials; otherwise, conclusions will be one-sided. Among the major oral art genres—myth, legend, and folk tale—the American folklorist William Bascom has said that myth is most sacred, tales are the most entertaining, and the legend is in between (Bascom 2006). The many epics of the Mongol language family consist mainly of tales, while Gesar is most

44  Critical reflections on epic studies sacred. In any case, to fully understand the epic genre, one should remember that epic encompasses a huge spectrum of heterogeneous factors, including mythic resonances and narrative material ranging from the closest to the farthest from historical facts.

Comparative epic poetics To study a complex creative work like epic, a well-thought-out set of tools and operational models is needed. In recent decades, Western epic scholars have brought forward a multitude of new research tools and points of focus. These include epic formulae, typical scenes, and story-patterns as advanced by Parry and Lord; the “large words,” traditional referentiality, performance arena, and on-the-spot composition in the course of performance stressed by Foley; Honko’s “superstory” and mental texts; Nagy’s cross-reference, three-in-one schema of composition-performance-diffusion; and the textual evolution pattern of the Five Ages of Homer. Underlying the creation of these terms lies a complex set of humanistic and scholarly concepts that form the basis of epistemological and philological investigation and anthropological field methodologies. With the fresh visions and insights that the terms open up, we can now carry out analysis with a large bag of scholarly tools. At the same time, however, detailed case analysis cannot ignore the aid of comparative poetics. Scholarly criteria in the fields of the humanities are often inseparable from the description of particular characteristics, but a grasp of these characteristics always comes from comparison. This is eminently the case in epic research. When we say that Mongolian epics are the least historical of the genre, we are comparing them with Homer’s Iliad, The Song of Roland, and the Mali Sundiata; and when we say that the Miao King Yalu is the most functional in its content, we are comparing it with more purely entertaining forms of epic. We make such comparisons in terms of at least two dimensions: time (diachronically) and space (synchronically). Comparing today’s living traditions with fixed texts, we can take a diachronic look at differences and similarities between today’s oral epics and historically fixed epics like Gilgamesh, which is carved on clay tablets. On the other hand, when seeking a more accurate understanding of a particular epic tradition, we can make synchronic comparisons between different regional traditions in terms, for example, of sacredness, size, and the complexity of social functions. In the Balkans, epics are sung mainly in public places like cafés during Ramadan. In southern India, the Siri Epic of the Goddess Siri is performed in the context of ritual sacrifice and near a temple but in a relatively less holy area (Chao Gejin 2012a). By comparing different traditions with each other, we can correct assumptions we make from the epic text itself. The richness of world epics opens wide poetic vistas and thus creates the need for us to broaden our scholarly horizons. If we seek an in-depth understanding of epic in Europe, Africa, the South Pacific, Mongolia, or southern China, or among the Ainu or Kirgiz, we cannot limit our perspectives to the readily observed characteristics and norms of that specific poetic tradition. Tibetan epic has been studied

Current issues in epic research  45 for decades but with little exchange between the Chinese academic community and international epic circles. We tout Gesar as the longest epic in the world and stress the sacrality of the bards who chant it. The huge volume of text is certainly amazing, but artistic value is never determined by length. The phenomenon of “god-inspired singers” has been relished for the aura of mystery it gives to an epic performance, but it should not lead us to neglect other modes of investigation and analysis. Gesar has many unsolved questions as an epic that need to be answered with the help of comparison with other epic traditions. Why is there no Tibetan community centered around Gesar, as is often the case in other traditions? How should we interpret the epic in terms of its social and cultural context? How is it related to the people’s belief system as a supremely meaningful narrative? What role do its open-ended structure and absorption of numerous other genres play in its inheritance and transmission? Why has this narrative of the hero Gesar spread so far and wide and had such an influence on surrounding ethnic groups? Why are there so many different types of singers in this performance tradition? Reflection on these questions constitutes a difficult challenge for scholars who study Tibetan epic in the light of today’s epic theories. Convincing answers can benefit from drawing upon international epic studies. Comparative poetics is often persuasive in answering basic theoretical questions. For example, in an article entitled “Challenges in comparative oral epic” that the present author co-authored with John Foley (Chao Gejin and Foley 2003:33–97), we were able to answer some of the basic questions in oral poetics by comparing four epic traditions. These included, for example, the nature of a poetic verse line. On the printed page, poetic lines can be problematic. Anyone with even a little experience can transcribe verses in accordance with the rhythm and spacing after listening to a recording; but comparing Old English, South Slavic, Ancient Greek, and Mongolian poetry, we found that in the different traditions, the understanding and use of poetic verse differ greatly. In some traditions, lines of poetry constitute strands of information helpful in the process of oral performance; but they do not look natural on a printed text. When chanted, they sound very rhythmical; but on paper, they may appear long and short without any particular reason. In some traditions, vowels are lengthened, or accompanying musical instruments are used to enhance the poetry; and verse length and meter give it a pleasant symmetry. But this readily gets lost in a line-by-line written text. Different poetic traditions provide different ways of understanding what constitutes a scene, a poetic formula, or even the nature of a poem or the makeup of an epic. In Jangar, sequences like the Mighty-armed Sabar and Hongor’s Wedding are themselves individual poems; but together, they all form the epic poem. To what extent is Jangar himself actually a character? He dominates the entire epic cycle but does not appear often and is not even very much mentioned. The real heroes may very well be Hongor, Hara Sanal, or Sabar. To answer such questions and at the same time explore the roots of expressive human culture, it is helpful to use normative scholarly formulations based on comparisons made with the help of versatile poetic concepts.

46  Critical reflections on epic studies

Epic singers, audiences, transmission, and ethnic identity The role of epic singers varies in different traditions. In the Jangar tradition, the role is inherited, sometimes by family line. Zhirmunsky and Chadwick speak of specialized schools for training singers in some traditions, such as are found in Uzbekistan. In the Siri Epic of India, the singer is the incarnation of the goddess. In the Yi region of southern China, a singer has dual identities: When singing the epic Hnewo, he is a storyteller; when presiding over a ritual, he is a bimo (shamanic priest). All these differences have a major influence on how we understand and define epic. In general, epic singers fall into two categories—professional and amateur— the degree of which depends on the local tradition. There are many differences between the two in terms of artistic attainment, singing standards, epic length, range of performance activity, audience response, and even text form. There is a broad spectrum from highly professional bards to extreme amateurs, with a large range between; and the transmission of an epic differs according to a singer’s position in the spectrum and mode of performance. Some, like singers of the Boutaga clan in Qinghai, perform from a scribbled text; some are “paper singers,” who sing from a written text. Some are said to be god-inspired, such as the Tibetan “dream-taught singer” (gnyid lam bab) mentioned in Chapter 1. Still others, like Manchu storytellers, are inheritors of written texts. The social and cultural functions of epics can also be understood from multiple angles. Some epics have a very small range of transmission, and their orientation is mainly entertainment. Other epic performances are grand and highly sacred, some accompanied by solemn ritual activity or strict taboos. The performance atmosphere may range from relief from pressure and fatigue on a journey to the solemnity of an indispensable life ritual. The Russian historian Boris Vladimirtsov and others have noted that in the Oirat and Uriankhai areas of western Mongolia, people will invite a singer to perform Geser when they encounter a disaster or plague. In some places in Xinjiang, Oirat Mongols believe that Jangar is magical and can only be sung after an accompanying ritual has been well prepared and a sheep has been sacrificed. They think, too, that no singer may sing the entire epic; if he does, his life will be shortened. In epics chanted to lead a deceased loved one to the other world, the atmosphere is solemn, and the ritual action centers on the person who has just died, with a member of the audience playing that person’s role. In the Yi epic Hnewo and the Miao King Yalu, the sound of chanting lingers around night and day. After the audience leaves, the donglang (priestly singer) guards the coffin along with two or three others in a procedure that invites people to ponder the function and reception of the epic (Chinese Folk Literature and Art Society 2012). All this raises the question of how singers and audience work together to create meaning. The composition, transmission, communication, generation of the meaning, and even completion of an epic are jointly brought about by multiple factors as what is in the mind of the singer interacts with what is in the minds of hearers in the specific space and time of the performing arena. All this constitutes

Current issues in epic research  47 the process of generating meaning. In epic research, it is necessary to take fully into account the role of the audience in the creative process. Examples of how an audience can influence the content of an epic are familiar to scholars. It is said that in the early years of singing the Kirgiz epic Manas, it included a sequence in which the manaschi (Manas singer) praised the Tsarina Catherine the Great at length in the presence of Russian officers. Later, in the 1950s, a bard in Akqi County, Kirgiz Autonomous Prefecture, Xinjiang singing for Han cadres went on at length describing how rich the Han area is. Once, when a singer had sung half of Jangar, the audience suddenly asked how the heroes could be fighting for such a long time. Didn’t they have time to eat or drink? The singer then quickly improvised a poetic sequence depicting the heroes eating. When Serbs and Croats were fighting in the Balkans, a singer would even change the ending of an epic in accord with the ethnic composition of the audience. Regarding the transmission of epic poetry, the Tibetan Gesar (Mongolian Geser) is exceptional. The 1716 Beijing Woodblock Geser is the earliest printed text; but there is evidence that the epic spread orally in various dialects and languages to far-away places over a large area, across barriers of ethnic groups, languages, and beliefs. In marked contrast, there are also epics with a relatively small geographical spread, such as the Miao King Yalu, which is mainly disseminated only in the Mashan area of Guizhou. Dialectal barriers and the boundaries of epic distribution are clear (Chao Gejin 2012). In between are regional epics like those of the Horqin region of Inner Mongolia and some Oirat epic traditions. Of course, “large” and “small” are relative terms; and the extent of epic transmission can expand from small to large, from local to national. In his discussion of Indian epics, Lauri Honko has emphasized that a deity praised in a local epic can be gradually upgraded to a national god. Based on fifteen living examples of the Indian epic tradition, Stuart Blackburn has divided them into four levels related to areas of circulation: Local (range of 10–100  miles); sub-regional (100–200 miles); regional (200–300  miles); and hyper-regional (400 miles and above) (Nagy 2008:67). The ancient Mahābhārata and Rāmāyana were gradually upgraded to the level of the entire Indian hyper-region, spreading far beyond national borders to many areas of Southeast Asia, such as Malaysia. Some epics may be gradually upgraded from a small local tradition to a national treasure, but this is not always the case. Some epic traditions came into being later in history, with no further chance for this sort of amplification. Influenced as it is by grassland tales and stories from books, Horqin epic has been called “quasiepic” (Xiao Bing 1989), the term applied to works of literary authors like Virgil and Voltaire that simulate the ancient epics. In any case, Horqin epic has no chance of being upgraded to a work treasured by Mongolia as a whole, because it has already ceased to grow and entered the stage of decline. Such upgrading is impractical, moreover, with today’s changes in social mores, like urbanization, the standardization of education, and the influence of modern media, all of which squeeze the space used by ethnic mother tongues to an unprecedented degree. Regional language and artistic traditions in many areas have been lost, moreover, and with them the possibility of a whole nation having a share in local ethnic treasures.

48  Critical reflections on epic studies The issue of epic ethnic identity has been raised by Western scholars relatively recently. In early epic research, scholars were not concerned to establish a connection between a particular epic and the national literature, sensibility, and identity. Recently, however, Finnish scholars became concerned about discussing such issues. Honko conducted an in-depth study of Elias Lönnrot’s Kalevala, which brought to light relations between epic construction and nation-state identity (Honko 1996:21). Comparing examples of epics from India, the Balkans, Mongolia, and Tibet, the present author has shown how epic identity can have graded levels. Among the Tibetan people, the identity of Gesar is that of the whole nation, a Tibetan proverb saying that every Tibetan has a Gesar. In some traditions, however, like Mongol Barag-Buryat and Horqin epic, a work’s identity is limited to that of a particular tribe or ethnic group. With their scope limited to a specific group or locale, such epics have not spread to other areas. The fact that the Mashan Miao sing King Yalu when someone dies sets their epic apart from other Miao. Of course, as a grand narrative with a serious theme, high artistic level, and long heritage, epic commonly condenses and sublimates the emotions and mindset of a whole nation and fully demonstrates a people’s oral artistic talents. The most representative oral artistic achievement of a nation is thus often epic. As a pinnacle of Greek literature, the Homeric epics are the most widely printed, translated, and read works in the world. There are also several peaks of early Mongol literature in which epic plays an important role; and squeezed between two powerful neighboring countries, the standard bearer for the national spirit of Finnish identity and independence was none other than the Kalevala (Vasenkar 2002:2–4). It is not surprising that it took Jan Sibelius and Elias Lönnrot’s song to make Finland appear on the world map. Without Sibelius’ music and Lönnrot’s poetry, there would have been no successful building up of the modern Finnish nation.

Epic authenticity and its nature as an object of research Some scholars have discussed the relationship between innovation and an epic’s traditional characteristics, arguing that folk singers’ newly created epics should not enter the tradition, because it is doubtful that they have an authentic connection with the themes of the old classics. This urgent question involves the definition of epic and the identification of what epic creativity is all about. The notions of “new” and “old,” newly created and traditional, are relative in the history of folklore. For citizens of the People’s Republic of China, Qing Dynasty mores were traditional. For scholars engaged in collecting epics after the founding of the People’s Republic of China in 1949, material from the period of the Republic of China (1912–1949) was authentic. For those collecting epics at the end of the 20th century, what was found in the second half of the century was thought to be more authentic. Moreover, what is recorded from illiterate epic singers can be considered closer to tradition than that from modern educated singers.

Current issues in epic research  49 During the evolution of a nation from oral to written culture, the complicated interaction and linkage between written and oral communication can make the question of authenticity even more ambiguous. Some scholars have put forward the concept of “writing inheritors” in an attempt to sum up what may happen in epic creation when a nation has become more script oriented. A folk creator steeped in tradition then receives a formal education and combines written skills with the oral traditions of his or her folk culture. What can be said about the authenticity of that singer’s written works? The Kalevala, written by the physician and literary author Lönnrot, is regarded as a tradition-oriented epic bearing the standard of the national identity of the Finnish people. If so, why should we regard as unauthentic the poetry written by the Kirgiz epic bard Jüsüp Mamay or by a Torghut Oirat jangarch (Jangar singer)? Answers to such questions must not be separated from the constraints of specific historical conditions, according to which judgments of epic authenticity are relative. There have been famous cases in history, such as the works of James Macpherson, for example, his Ossian (Thomson 1952), and those of Niegoš in the Balkans. Both serve as notable links in the process of transforming oral folk poetry into written poetry as heirs of traditional folk culture who received a systematic education and became written poets. In the history of Chinese poetry, the first great literary poet Qu Yuan (c. 340– 278 BCE, formerly translated as Chu’s Yuan, minister of the State of Chu and one of China’s earliest poets) composed poems like Li Sao (Encountering Sorrows), Jiu Ge (Nine Songs), and Tian Wen (Inquiries of Heaven), which contain many folktale elements, serving as an early bridge between two worlds. Today, there are more and more such examples all over the world. Should we stick to classical literary judgments, or should we accept today’s living folk traditions as being as authentic as the ancient texts, regarding folk culture as a continually flowing and evolving life process? There is no need to be bound by rigid models. This leads us to another question: Granted the currently developing trends in various humanistic disciplines, what is the nature of epic as an object of research? This remains a matter of opinion. Admittedly, fundamental epic research consists in the literary analysis of texts. As we have seen, however, it is far more than that. It is also an object of anthropology, folklore, oral communication arts, and other disciplines as well. The top Western scholars who have pioneered new paradigms and led the epic research community in new directions have all been highly educated in the ancient classics. The mastery of many a classic text from different cultural traditions constitutes the solid foundation for any epic investigation. Foley’s ability to connect the ancient and the modern, Nagy’s deep grasp of both classical and contemporary traditions, and Reichl’s in-depth skills in Turkic and Middle English oral traditions are the bases on which they have scaled the heights of academic research. Without the classical groundwork and extensive exploration of epic texts as literature, research cannot solve major problems. Though groundwork in the field is necessary, it is impossible to construct new theoretical approaches to Gesar and Jangar on the basis of brief fieldwork expeditions. Without a complete understanding of the whole picture, literary and living, no progress can be made.

50  Critical reflections on epic studies From 1964 to 1972, the London Seminar on epic made major contributions to international epic research. Distinguished scholars participated, including A.T. Hatto, J.B. Hainsworth, C.R. Bawden, Zhirmunsky, Walther Heissig, and Karl Reichl, but no Chinese scholars. The results were later published in two volumes; and since its first number in 1986, follow-up work has been carried forward by Oral Tradition, the flagship journal of research on epic and oral traditions published by the Center for Studies in Oral Tradition, University of Missouri, for fostering in-depth discussion on human knowledge theory, information and language technology, and thinking and psychological functions. The great strides in research on oral traditions since the 1960s have recently come to be closely integrated, moreover, with the international protection of our human intangible cultural heritage. Many experts in the world’s oral traditions are actively engaged in this protection in their respective countries, and China is an important player in this regard. The future of epic research can no longer be separated from research on oral traditions, which began with strong ties with epic studies. The scholar well known for combined research in these two areas is John Foley. Though he has gone, the fire of his thought is still burning. Many times before his death, he expressed great hope for the Chinese academic community. China has not only a rich oral tradition but also excellent research teams studying this tradition. China’s current trends in social and cultural development present excellent opportunities for research, with the vast western region still based mainly on traditional agriculture with rich examples of folk culture, while major cities throughout the nation provide urban centers of academic learning. An earlier version of this chapter was originally published in National Art, Vol. 1, 2013.

4

John Miles Foley and recent research trends on oral traditions

As can be seen from previous chapters, John Miles Foley (1947–2012) was an internationally renowned epic scholar, classical scholar, and expert in comparative studies of oral traditions. During his lifetime, he was William H. Byler Distinguished Professor of Humanities at the University of Missouri, Professor of Classics and Old English, director of the Center for Studies in (CSOT), founder and editor-in-chief of the journal Oral Tradition, director of the Center for eResearch, moderator of The Pathways: Oral Tradition and the Internet (Channel Project), founder of the International Society of Oral Tradition Research (ISSOT), and also chief academic advisor to the Oral Traditions Research Center of the Institute of Ethnic Literature, Chinese Academy of Social Sciences.

From classical epic to oral epic: Standard bearer of Oral Formulaic Theory As we have seen, Foley was devoted to comparative studies of oral traditions, his main research subjects being ancient Greek epics, medieval English (AngloSaxon), Yugoslavian, and folklore. In the expression of condolences written for Fabula, the German journal of comparative folklore studies, the present author summarized his professional attainments and academic contributions as follows: Professor Foley was proficient in French, Yugoslavian, Greek, Latin, German, Italian, Spanish, Old English and Middle English. During the more than thirty years of his academic career, he wrote or edited twenty books and published more than 200 papers in the fields of literature, folklore, classics, linguistics, oral traditions and information technology. He was a man of considerable culture, great attainments and broad horizons. No one could rival him in these respects. (Chao Gejin 2012b:111–113) Foley was born into a medical family, and his parents expected him to be a top doctor. As a teenager, he was not allowed to ride a bicycle for fear that a broken finger might affect his future in precision surgery. He received solid classical training, however, at the University of Massachusetts Amherst; and he later DOI: 10.4324/9781003258001-5

52  Critical reflections on epic studies became acquainted with Parry and Lord’s Oral Formulaic Theory and was influenced by Lord to shift his academic vision from Latin and Greek to contemporary oral epics. His first epic achievement in both classical studies and living oral traditions was based on direct fieldwork and is widely referenced in the Yugoslav materials in The Parry Collection of Oral Literature in the Widener Library of Harvard University. Chinese scholars’ initial acquaintance with Foley’s work began mainly with his history and epic research, and most are familiar with his studies on oral poetics.1 In the 1980s, drawing upon his considerable linguistic skills, Foley began collecting materials and compiling a detailed research index on Oral Formulaic Theory. On this basis, he wrote the academic history of that theory and then published several important books on epic research to carry forward the ideas of his predecessors.2 In these works, he expanded and dug deeper into the Parry–Lord Theory, which led to a paradigm shift in international epic research. Reflecting Honko’s idea of superstory, he came to see oral performance as a generalization of immanent art. Thanks to his overall understanding of the main features of epic art, the principles of performance, the modes of transmission, and the role of inheritors, he successively proposed conceptual terms that came to serve as a more complete schema of epic terminology. Among these, the most academically acclaimed include performing arena, large word, epic register, traditional referentiality, and legendary singer. Foley’s epic research is grounded in literary and classical studies but makes broad use of methods from anthropology, folklore, and other disciplines. He thus created an analytic system for the study of both Western classical and living oral epics in an open-minded manner. Breaking out of the bonds of epic research based on classical literary studies, Foley, along with the eminent Finnish folklorist Lauri Honko, created a dazzling array of epic research. Although Foley’s field base was located in Yugoslavia and Honko’s in Finland and India, the work of the two was highly compatible in how they classified epic text forms. Their concept of tradition-oriented text has been widely accepted by the academic community today because it aptly summarizes how literary-minded persons can collect folk materials and then patch them together to form a national epic. This provides the basis for the legitimacy of the Finnish Kalevala and similar cases around the world. Chinese scholars use the concept when discussing Manchu traditional storytelling. Foley and Honko were also in perfect accord on how to define epic. In his article “Analogues: Modern Oral Epic” in A Companion to Ancient Epic, Foley quotes Honko’s definition: Epics are great narratives about exemplars, originally performed by specialized singers as super-stories which excel in length, power of expression and significance of content over other narratives and function as a source of identity representations in the traditional community or group receiving the epic. (Honko 1998:28) Foley comments that Honko’s definition does not revere the ancient Iliad and Odyssey as the standard but puts emphasis on the symbolic function of epic for

John Foley and recent research trends  53 specific group identity. This leaves space for the investigation of the diversified styles, themes, characters, singers, ways of performance, religious or secular content, rhyme or prose, and other parameters of epic (Foley 2008:199). If epic research in the 20th century is a series of peaks, the theoretical contributions of Foley and Honko can be called twin peaks, the main marker of the current summit of epic studies.

From large words to traditional referentiality: Master of oral poetics Foley greatly surpassed the theoretical achievements formed on the basis of the Parry–Lord Oral Formulaic Theory. This can be seen from his studies of oral poetics in terms of some of the key terms that he himself coined. As was noted in Chapter 1, Foley traveled many times to the Šumadija region of Serbia for fieldwork; and he translated and edited the epic The Wedding of Mustajbeg’s Son Bećirbey by Halil Bajgori (2004). He was very familiar with Yugoslav materials and drew the concept of “large word” from field evidence there. The idea comes directly from the local people and in a sense, represents an abandonment of the term “formula.” Foley found that for a Yugoslav singer, a word is very different from a simple text unit, a dictionary term, or an abstract linguistic definition. In the mind of the singer Kukuruzović, the word in oral poetry is a complete unit of expression, an inseparable atom that constitutes speech, and a specific speech act. It may be a short phrase, but it is often longer. In that singer’s performance, the phrase “miserable captive” is a word; but a word may also be an entire poetic verse line, like “Mustajbeg of the Lika is drinking,” or even a multi-verse unit. Kukuruzović believed that some scenes or motifs, such as a description of a journey, a hero’s preparation of pre-war equipment, or the mustering of an army, can each be a word. In the world of sung performance, a word can thus be expanded to a large word (Foley 2002:13–14). In incorporating the folk term “large word” into academic research, Foley showed his unique vision of oral poetics. In its basic dimensions, the boundaries of the conceptual tool formula in the Parry–Lord Theory coincides with “large word.” So, why use the term? First of all, a formula is a phenomenological summary of a given text, whereas large words are morphological summaries from the singer’s standpoint. For Foley, they amount to a special way of narrating, a unit of expression. In the mathematics of language, large words are integers, not fractions. No part of a large word can be arbitrarily divided; otherwise, the meaning of the narrative will be lost. Secondly, large words are shaped by tradition, not created by an individual. They are forged and shared by singers throughout different generations; they go beyond specific songs to have a life within the larger oral tradition. Thirdly, we must not arbitrarily segmentize the singer’s words or interpret them in the light of principles of literacy; for this often misreads and mistranslates folk poems (Foley 2002:14). Another conceptual term proposed by Foley is “traditional referentiality.” As he explains, when the expression “fatty hands” or “green fear” appears in Homer’s

54  Critical reflections on epic studies epics, modern readers are often at a loss; and some even think that this is a corruption inevitable in long-term transmission. They may think that there was a clerical error in the initial copying, which led to an incomprehensible expression. After careful research, however, Foley found that these so-called corruptions are a common phenomenon in specific performance traditions; and he calls this “traditional referentiality.” That is, the literal meaning of a particular expression often differs greatly from the meaning it conveyed in the register of a particular tradition; but the narrators and their audience readily understand the reference. In Immanent Art: From Structure to Meaning in Traditional Oral Epic, Foley explains that this notion was inspired by reception aesthetics and reader-response criticism (Foley 1991:38–60). From the perspective of interpretation, traditional referentiality accordingly emphasizes that the interpretation of an expressional unit must not only be rooted in tradition and context but also requires attention to the cognition of the composer and the receptance of the receiver in the idiomatic context of the particular epic register.3 In the ancient Greek tradition, “fatty hands” meant that someone does something bravely and “green fear” that God will intervene in human affairs. In Yugoslav poetic tradition, “black cuckoo” means that a woman has or will become a widow. In the traditional Mongolian epic register, the birth of an unusual foal often means that a great hero will be born. Such instances are numerous in epic traditions around the world. Along with making major contributions in the study of the classics, epics, folklore, literature and art, communication, Slavic studies, and other related disciplines, Foley also became a leading figure in the field of oral traditions, bridging past and future. Based on the comparative study of epic traditions, he made many significant contributions to the theory and methodology of oral poetics. The theoretical insights that he developed as he expanded his horizons from epic toward the broader field of oral poetics can be appreciated only by combing some of the key words that he created.

Advocate of a new discipline: Leader in research on oral tradition In a narrow sense, oral tradition covers all categories of traditional language and art; in the broad sense, it includes all forms of oral communication. In recent UNESCO documents, the world’s oral traditions have come to have special importance. In the Convention for the Safeguarding of the Intangible Cultural Heritage adopted in 2003, oral traditions constitute the first of five main areas of the intangible human cultural heritage. Their importance is self-evident. As early as 1986, Foley established CSOT at the University of Missouri, of which he was the director. In the same year, he also founded the academic journal Oral Tradition, which Hatto immediately lauded as serving to pass down epic learning from generation to generation. Due to its intensive peer review process and impressive editorial board, the journal has become increasingly influential, especially through its exclusive online publication beginning in 2006. Moreover, it implements the spirit of academic democracy and sharing more thoroughly than

John Foley and recent research trends  55 any other journal in the field of the humanities. If research in oral traditions has developed into an interdisciplinary endeavor today, this is largely due to the multiple efforts and painstaking management of this Center and the journal over the past thirty-five years. As Foley’s academic field of vision came to transcend that of the Parry–Lord Oral Formulaic Theory, he creatively integrated the most important folklore theories of the 20th century, such as narrative ethnography, performance theory, and ethnopoetics, into oral tradition research, systematically putting forward a theory of the traditional comparative principles of oral poetry so as to construct a unique academic system of oral poetics and an analytical method for the study of oral poetry texts. The 2002 monograph How to Read an Oral Poem is his representative achievement in this. As early as 1988, in “Oral Poetics: Parry–Lord Theory,” Foley clearly proposed the existence of oral tradition as a discipline; but in this book, he traced the formation of its academic history and looked forward to its development (Foley 2002). He cheered it on along the path pioneered by Parry and Lord as he continuously explored further in the field, always regarding epic, especially oral epic, as a most important academic resource. As he traveled this path, he followed a clear trajectory from historical academic research to the study of the singer of tales in performance and then on to discussions of how to understand oral poetry. In the 1995 book The Singer of Tales in Performance, Foley discusses fresh aspects of the singer in performance in vivid detail. He begins with modes of performance and how they give meaning. He then discusses how oral texts evolve into librettos through a specific process, different for each culture, and how the initial text process unfolds. Turning to issues of performance and receptance, he deals with the definition of performance arena, verbal interchange, and communicative economy. He proposes how to interpret traditional strategies and rules in the use of metonymy, etc., as giving words power; and he discusses the power of spells, with instances from traditional Serbian magical practices. A singer’s long years of personal interpretation of a text and the process of communication and receptance constitute the main content of the book. In accordance with his analysis in that work, Foley then went on to develop and systematize further ideas about the characteristics and principles of oral tradition that formed the backbone of theoretical interpretations in the 2002 monograph How to Read an Oral Poem. Generally speaking, research on oral traditions is focused mainly on language arts and so has a natural connection with the study of oral folk literature. In recent developments in European and American folklore circles, however, more and more scholars have become aware that oral tradition as a discipline has boundaries that extend far beyond the scope of folk language and art—from the formation of the Old Testament to contemporary black religious preaching, from the text of Homer’s poetry to modern American Slam Poetry (Somers-Willett 2009). This latter genre was a wave of American poetry in the 1980s that started in bars frequented by Chicago’s white-collar class and became popular with youth. It aimed to revive and enhance the art of spoken language through competitions and encouraged anyone to improvise original poetry. The judges came from among

56  Critical reflections on epic studies the audience, thus breaking down the barrier between poets and listeners, between critics and audiences.4All such manifestations of language art have become the focus of oral tradition studies, and they have all benefited greatly from Foley’s disciplined discussions and advocacy. As he saw it, the eons of human beings’ use of oral media showed wonderfully how different media have been continuously discovered and used in the long evolution of our species to produce, transmit, communicate, and store information. As for the two major inventions of language and writing, the history of oral language is much longer than that of writing in terms of scale, influence, meaning, and function; and its scope is much larger. Without looking back into history, there are still existent on the earth today about 6,000 languages, not even onetenth of which are widely used. Living in a world of words, we deal with them every day; but we do not pay attention to the problems to which they give rise. We must not forget that in many parts of the world, the common tools for information exchange are still mouth and ears. Today’s world is still made up of both nonliterary and literary societies, as well as transitional societies in which the use of script is inadequate to meet people’s communication needs. According to the latest census, China’s illiteracy rate (the proportion of illiterate people aged fifteen and above to the total population) is 4.08%, a decrease from 6.72% in 2000. The reduction of a nation’s illiteracy rate takes time. Half a century ago, China could not have been called a literate society because it still had a high illiteracy rate. In Foley’s How to Read an Oral Poem, we have further evidence that his research perspective and ways of explaining matters were revolutionary. He selected four traditional singers as models, describing each one’s different performance traits: a Tibetan paper singer who uses written notes; a North American Slam recitation poet; a South African hymn poet; and an ancient Greek singer. Foley proceeds from their standpoints and uses their own words to examine how their particular traditions interpret their oral poems and songs. He then discusses the poetic and literary issues involved in oral poetry, the oppositions between oral and written, and the media vitality of oral performance, audio texts, past voices, the writing of oral poetry, and so on. In subsequent chapters, he discusses links in the oral transmission and performance that have so far remained almost untouched, including special codes, figurative language, parallelism, special formulae, appeals to tradition, and singers. He pays attention to elements taken from Richard Bauman’s observations on performance theory but notes that such elements are not necessarily present in all oral poetry traditions and appear in different forms in different traditions. These discussions of the core features of oral poetry point to a theoretical construction of oral poetics, though Foley himself did not particularly emphasize this point. An important value of How to Read an Oral Poem lies in its strong reminder that the humanities need to deal with poetry and other literary genres from oral traditions in terms of the principles of oral composition, transmission, diffusion, and acceptance. Otherwise, many misunderstandings arise, which make it difficult to understand the essence of oral communication. This reminder is particularly important because the international study of oral traditions has long

John Foley and recent research trends  57 been based mostly on the concepts and tools of general literary theory, which is inadequate if we are to go deeply into the essence of oral literature and explore its core characteristics and social cultural functions.

Oral tradition and the internet: Pioneer in the pathways of minds Edited by Foley and published in 1998, Teaching Oral Tradition included his own paper “The Impossibility of Canon,” which gives evidence of his consistent academic agility and searching spirit. In the early years of electronic media and the internet, he linked ways of thinking among the oral tradition that included the Library at Alexandria and the internet (Foley 2000a); and ever since, his oral tradition research and digital practices have complemented each other. Foley and his team worked to expand this interdisciplinary academic space and international exchange forum at the node between oral tradition and the internet. As we have noted, since 2006. the journal Oral Tradition, which he founded, has been fully digitized and shared online (http://journal​.oraltradition​.org). Recent statistics show that by 2012, 40,000 independent domain names had visited and downloaded the electronic version of the journal, which stands as an excellent model for the spread of professional information in the internet age. From 2005, moreover, Foley’s CSOT was committed to implementing and advancing the Pathways Project (www​.pathwaysproject​.org) and initiated the International Society for Studies in Oral Tradition, an online, free-of-charge academic platform for communication among researchers worldwide. Foley practiced academic democracy in the true sense and went to great lengths to establish this cross-cultural, crossborder sharing mechanism for oral tradition research. The book Oral Tradition and the Internet: Pathways of the Mind, published posthumously in 2012, is his final legacy. The book cover is composed of many random points and connections between points, symbols of the author’s attempt to describe how human thinking combines characteristics of text technology (TT), oral tradition (OT), and information technology (IT). The book confirms what was said thousands of years ago by a philosopher in the Mahābhārata: “What is faster than the wind?” Answer: “Thinking.” Only a few of the most talented, sensitive, and mission-minded scholars have had the will to enter into such a topic, and Foley gives his own unique answers. The pioneering feature of the book is reflected first of all in the compilation of the table of contents, which exemplifies Foley’s way of thinking. From the preface onward, the traditional chapter structure has become a range of parallel nodes. The author first explains the reading method of the book, that is, how to treat the book’s network and use the link map, etc. The text itself replaces the usual chapter structure with a format in which nodes are arranged in alphabetical order. The nine nodes that start with the letter A, for example, go from “A Foot in Each World” (p. 33) to “Audience Critique” (p. 49). There is only one node under the letter B. The rest go on in the same way until the end node, Wikipedia (p. 271). In fact, the book is part of the author’s Pathways Project, which focuses

58  Critical reflections on epic studies on the oldest and the latest technologies for human thought and underscores the similarity and unity between oral tradition and the internet (Foley 2012:5). The originality of the book also lies in its simultaneous appearance in print and online in the Pathways Project, which also contains links and contents that a book cannot include. The printed book itself, moreover, has a morphing character that allows it to be accessed, browsed, and read in many different ways. The book uses, moreover, an enlightening analogy—the ancient Greek agora— to show how oral tradition and information transmission follow the rules of the internet. In ancient Greece, the agora (market) was a forum built with bricks and mortar for engaging in various large-scale public activities. The Athenian Agora northwest of the Acropolis was built in the 5th–4th century as the venue for political, commercial, and religious activities as well as a public space for social interaction (Foley 2012:40). The Pathways Project is a verbal marketplace for virtual exchange, a public space and forum that allows ideas and knowledge to be adopted by the community. Foley refers to the tAgora (text market), oAgora (oral market), and eAgora (electronic market) to describe the dynamic mechanisms of their respective creative origin and dissemination. The ways we browse for electronic information, the links between information nodes, and the flow of information through channels are all connected with the generation, organization, propagation, and even the existence of oral information. The digital presentation of the world’s intangible cultural heritage, especially that of oral traditions, has difficulties but also incomparable advantages and almost unlimited possibilities with the growth of the internet and electronic technology. Along with the Pathways Project, Finland’s Culture Sampo semantic web and the Archives Project of Audio-visual Images of Oral Traditions of Chinese Ethnic Groups initiated by the Institute of Ethnic Literature of the Chinese Academy of Social Sciences have begun transnational cooperation based on Foley’s ideas.

Lasting memories: Forever Singer of Tales With Albert Lord as mentor in his early years, Foley did postdoctoral research at Harvard and the University of Belgrade. Foley’s respect for his mentor and the expansion of the oral school is encapsulated in the title of his book The Singer of Tales in Performance. After decades of fieldwork, he was like a singer himself, in love with oral tradition. However, heaven’s jealousy made this tireless singer of tales leave us in his prime. What we lost is not only an eminent scholar but also an old friend of China’s Institute of Ethnic Literature and of the present author as well. What we lost is a sharp-eyed, well-versed navigator in the boundless sea of learning.5 In addition to his many academic responsibilities in the United States, Foley held various academic positions in institutions in Asia, Africa, and Europe; and he was a presenter in several high-end international training programs. As a member of the Advisory Committee of the Folklore Fellows Network of the Finnish Academy of Humanities and Chair Professor of the Epic Workshop of the Folklore Fellows Summer School, Foley trained numerous epic scholars and

John Foley and recent research trends  59 oral tradition research experts from various countries. He came to China for a lecture in 1997, traveled to Inner Mongolia for fieldwork, and included Mongolian epic traditions and singers in his comparative research vision, including discussions of the legendary Mongolian epic singer Choibang and a Tibetan treasuretext singer, etc. On the cover of his 2002 book How to Read an Oral Poem is a photo of the Tibetan paper singer Drakpa Senge holding a sheet of paper as he performs an epic.6 After the Chinese translation of his monograph The Theory of Oral Composition: History and Methodology (2000) was published, ideas of Oral Formulaic Theory and Oral Poetics came to have a profound impact on the study of the oral traditions and construction of epics in China. In the opposite direction, the journal Oral Tradition published a collection of the oral traditions of Chinese ethnic minorities, which introduced the research of Chinese scholars to the international academic community. On the website of Oral Tradition, Foley posted materials that he thought would be beneficial to the Chinese academic community; and these have been gradually translated into Chinese for wider access. In June 2009, Foley was invited to China to serve as Chair Professor of the first Institute of Ethnic Literature International Seminar Series on Epic Studies and Oral Tradition Research Workshop; and he gave enlightening lectures on the themes of understanding diversity, oral tradition, and the internet. In May 2011, he came to Beijing to attend the China Social Science Forum: Interdisciplinary Research on the World’s Endangered Languages and Oral Traditions, where he delivered a speech on the Digital Project of the Oral Traditions Research Center, highlighting the basic foundation and details of the Pathways Project. Before his death, he was planning to come to China once more to participate in academic conferences on epic research while coordinating the tripartite cooperation of the Society of Literature of Finland, the Institute of Ethnic Literature of the Chinese Academy of Social Sciences, and the University of Missouri to promote the scientific theoretical construction of folklore materials in line with new technologies and ideas. In contrast to his fame, Foley was extremely modest and low-key, eloquent but never arrogant. His care and support for young researchers, including many from China, and his attention and support for cultural and academic resources outside Europe and the United States were admirable. Under his guidance, young scholars of our Institute are continuing to visit CSOT of the University of Missouri. A generation of gurus has left us. Their spiritual heritage will remain in the world, but the development of the discipline will be affected by their early departure. Their loss will become more apparent with the passage of time. On May 3, 2012, a tornado cloud system visited Missouri; and in the course of its dark clouds and storms, John Miles Foley passed away peacefully at the age of sixtyfive. The following day, the present author came to his residence on the outskirts of Colombia with Chinese students to pay tribute to an old friend, colleague, and mentor and present our condolences to his wife and children. I wish here to express my cherished memories of Foley’s academic cause. I ended the expression of condolences that I wrote for the German folklore journal Fabula with this sentence: “He will be missed internationally.” Yes, the memory of John Foley extends far beyond national boundaries.

60  Critical reflections on epic studies An earlier version of this chapter was originally published in Northwestern Journal of Ethnology, Vol. 2. 2013.

Notes 1 These include The Theory of Oral Composition: History and Methodology (1988); Teaching Oral Tradition (ed. 1998); How to Read an Oral Poem (2002); and the highly praised reference work A Companion to Ancient Epic (2008). 2 These include Traditional Oral Epic: The Odyssey, Beowulf, and the Serbo-Croatian Return Song (1990); Immanent Art: From Structure to Meaning in Oral Traditional Epic (1991); The Singer of Tales in Performance (1995); Homer’s Traditional Art (1999). For a list of Foley’s academic publications, see Scott Gamer, “Annotated Bibliography of Works by John Miles Foley,” Oral Tradition, Volume 26, Number 2 (Festschrift for John Miles Foley). http://journal. oraltradition​.org​/issues​/2​6ii/ garner 2. 3 For discussions of epic register, see Chapters 16 and 19. 4 http://en​.wikipedia​.org​/wiki​/Slam poetry 5 For Foley’s academic achievements, see Commemorative Issue for John Miles Foley released by the Chinese Ethnic Literature Network. http://iel​.cass​.cn​/html​/37​/index​ html. 6 The photo was provided by Chinese epic scholar Yang Enhong.

5

Gregory Nagy From the Homeric Question to Homeric Questions

As we have seen, the Homeric Question has puzzled generations of classical scholars. Through in-depth research from comparative linguistics and anthropological perspectives, the Harvard professor Gregory Nagy drew up a new historical pattern for answering questions about how, when, and where the Iliad and the Odyssey were produced and circulated, why they were preserved in the form of written texts, and what caused their wide dissemination over more than 2,000 years. Nagy’s research on Homer’s poetry is quite out of the ordinary. With his foundation, like Foley, in classical studies and a new folklore perspective on those studies, he opened up new paths and became a leader in the field of North American classical and epic research. As we saw in Chapter 1, the authorship of the Iliad and the Odyssey has been debated since the Alexandrian period. Aristotle had thought that Homer was the oral poet, but the “separators” Xenon and Hellenicus pointed out that there are differences and inherent inconsistencies in the two works and thought that the Odyssey was not created by Homer. Like Aristotle, Josephus at the beginning of the Christian era also thought that Homer was the author of the works attributed to him. Unfortunately, however, none of these ancients provided us with much actual information about Homer and his works. Centuries later, the Romantic movement gradually formed the view that Homer’s epics must have experienced a stage of oral transmission before they were written and that they were most likely written down much later than Homer’s time. In 1725, Vico argued in The New Science that epics are not the works of individual genius poets but rather, the cultural achievements of nations. In 1769, the British archaeologist Robert Wood proposed in An Essay on the Original Genius of Homer that Homer was illiterate and that his epics had always been transmitted by word of mouth. Then in 1795, the German scholar F. A. Wolf’s Prolegomena ad Homerum triggered the 19th-century debate between Analysts and Unitarians that led to the rise of Parry and Lord’s Oral Formulaic Theory in the early 20th century. The Analysts and Unitarians debated whether there was a single poet Homer or many “Homer” poets; and some scholars came to take a position between the two that although the Homeric epics were not completed by Homer alone, he must have played a considerable role in their formation. DOI: 10.4324/9781003258001-6

62  Critical reflections on epic studies Milman Parry’s answer to the Homeric Question in the 1930s triggered an earthquake in the field of classical studies. Making a precise linguistic analysis of the Homeric texts, he concluded that they were rooted in oral traditions. To support his conclusion, he and his student and collaborator Albert Lord conducted well-designed field investigations in the then Yugoslavia and found that epic singers there never recite poetry verbatim but rely on formulaic themes, typical scenes, and story-patterns to produce their stories. In the same way, they maintained that the giant system of Homeric poetry could not have been the product of one individual inspired genius; the poems arouse out of a great oral folk tradition. Their Oral Formulaic Theory rapidly became an influential doctrine, and its conceptual tools have gradually deepened from song to text and then to performance. As a spokesperson for classical studies as influenced by this theory, Gregory Nagy meticulously studied the Homeric tradition and its textualization process and came up with a threefold composition-performance-diffusion model and evolutionary pattern of the Five Ages of Homer. His investigations transcended the limits of traditional classical studies and injected into them a new vitality by bringing them into the broader landscape of contemporary research in the humanities. Extending the Homeric Question to Homeric Questions in chapter 3 of his 1996 monograph Homeric Questions, Nagy used his specialized literary skills to dig out the etymological meanings of Greek bards, poetic composition, and the name Homer itself; and he convincingly reconstructed many possible links underlying the Homeric narrative tradition, text composition, and evolutionary process. Exploring that tradition in the light of modern oral traditions and anthropological data, he brought to light internal relationships between the ancient songs and singers, poetry and poets; and he provided a fulcrum for gaging the realities of the ancient Greek singers behind our texts. In his search for the compiler of Homer’s poetry, Nagy did not restrict himself to Greek documents; he turned our attention to the Indian epic tradition. On the question of heroes and hero worship, Nagy argued that in the celebration of the local hero in Greek epic, there lies a subtext of hero worship at the pan-Greek level and that the connection of the two is crucial to understanding the spread of the Homeric tradition. He found a striking analogy with forms of the living epic tradition in Rajasthan as exemplified by its two major epics, Pabuji and Devnarayan. The persistent custom there of worshiping powerful spirits of the dead provides a traditional framework for the epics that allows them to maintain their significance and vitality (Nagy 1996:62–64). Applying an analogical etymological approach to the composition of Greek epic and the name Hómēros, Nagy discovered that the metaphorical use in ancient Indo-European languages of “joining” as applied to a well-crafted chariot wheel was also used in Indian and Greek traditions to express poetic composition. In the Indian tradition, “carpenter” or “craftsman” is a metaphor for master; in a stanza of the Rigveda, the terms “join” and “wheel” appear as a metaphor for a chariot; and in the oldest poetic traditions, verbs of joining that were used to refer to a carpenter’s craftsmanship were applied to the voice of a poet (Nagy 1996: 121).

Gregory Nagy  63 Nagy found an etymological analogy with the name Hómēros, which can be interpreted as “he who puts things together.” As regards the textual evolution of Homer’s poetry, Nagy convincingly demonstrated the model he developed over the years from dual diachronic and synchronic perspectives to a threefold schema of composition-performance-diffusion and the Five Ages of Homer. He found that the process of the oral composition and dissemination of the Homeric works was quite long and argued that the set text was finalized in 550 BCE at the latest. His model brilliantly answers questions of how, when, where, and why Homeric epic was finally set down in the form of a written text and then diffused for more than two millennia. In line with the comparative poetics and analogical perspectives of Parry and Lord, Nagy’s evolutionary model draws upon analogical evidence provided by numerous living oral epic traditions—India, West Africa, North America, and Central Asia. As he himself says: Given the strong parallelisms between textuality and certain patterns of evolution in oral poetic traditions, I have been arguing that the fixation of Homeric poetry as a text can be viewed as a process not necessarily an event. Textfixation becomes an event only when the text finally gets written down. But there can be textuality – or better, textualization – without written text. I have been arguing further that the Homeric tradition of epic provides an example of such textualization: in the process of evolution in composition, performance, and diffusion, the Homeric tradition of epic became increasingly less fluid and more stable in its patterns of recomposition, moving slowly ahead in time until it reached a relatively static phase. We may refer to this static phase as an era of rhapsōidoί`rhapsodes’ (Nagy 1996:109) Nagy based his reflections and theoretical construction on in-depth considerations of text production and recording, focusing on relationships between field findings and text, between text and context, between the performance event and community communication, between inheritor and audience, and between text community and academic research. He emphasizes these elements because, whether for Homeric or Indian epic, the text has become a written canon after compilation and collation by countless generations. Unfortunately, however, live performance data of ancient times have mostly been lost in the vicissitudes of time. The efforts of oral poetics rooted in anthropology and folklore are needed to reconstruct the sound of the original texts and provide analogies from the methods of textual production of contemporary oral epic in line with the two schools of Ethnopoetics and Performance Theory. In the ecological chain of epic inheritance and transmission that the Finnish scholar Honko says can give it second life, the layered relationships between the solidified canon and the living tradition have a close, sometimes mysterious relationship. Between the ancient epic text and the living tradition, we need to take into account the

64  Critical reflections on epic studies performer, collector, editor, producer, collator, translator, and researcher, all the way to the reader; for all are important links. The study of epic is thus increasingly shedding the limited tradition of classical textual research to deal with the comprehensive mien and general characteristics of the tradition of epic performance as a whole. The scholarly reconstruction of the classical epic tradition, with Nagy as the outstanding representative, marks not only a significant advance of epicology but also a key achievement of the humanities in the fruit it has borne in related disciplines. Nagy’s etymological research on the original meaning of hómēros and his construction of the evolutionary model of epic texts amounts to a reversal and transcendence of the conservative view and gives new life to classical studies. To reiterate what was said in Chapter 1, it is through continued academic inquiry and response, construction and deconstruction, affirmation and negation that the international epic academy has been making the silent Homeric texts ring out as the chorus of many voices and allowing the ancient Homer and his epic heritage to live forever. Nagy’s words at the end of chapter 3 of Homeric Questions reverberate with a sense of sadness and comfort in the heart of a classical scholar who is ever searching for truth: And yet, this envisioning of Homer in evolutionary terms may leave some of us with a sense of aching emptiness. It is as if we have suddenly lost a cherished author whom we could always admire for the ultimate achievement of the Iliad and the Odyssey. But surely what we have really admired all along is not the author, about whom we never did really know anything historically, but the Homeric poems themselves. To this extent, the evolutionary model may even become a source of consolation: we may have lost a historical author whom ical author who is more than just an author: he is hómēros, culture hero of Hellenism, a most cherished teacher of all Hellenes, who will come back to life with every new performance of his Iliad and Odyssey. (Nagy 1996:111–112). An earlier version of this chapter was originally published in China Reading Weekly, 9, March 11, 2009.

6

Lauri Honko The identity function of epic poetry

Since Aristotle, epics have occupied a key position in Western poetic theory; and in the hierarchical order of genres, the high ranking of epics has been gradually established. Scholars have discussed epic issues extensively, including the relationship of epic with history, its social and cultural value, its artistic structure and techniques, etc. It is regrettable, however, that while many are busy explaining the content and form of epic narrative, they have given little attention to its function of shaping and expressing the identity of the peoples in which epics have arisen. The issue of identity as an academic topic is a recent phenomenon, but the identity function of epic poetry has never been absent. Generally speaking, in the oral traditions of some communities, epics have greatly outdone other narrative categories in length, seriousness and richness of content, strength of expression, and comprehensive influence. Thus, they can rightly be called “superstories” (Honko 1998:28). However, the aesthetic value of epic is only one aspect of its complex value system. The significance of epic goes far beyond the message conveyed directly by the words of the poetic text. The traditional connotations of many epic texts in terms of group identity, norms of behavior, core social values, and symbolic structure are all closely related. An epic that may seem long and inexplicable to outsiders, full of redundancies and repeated verses, may have a very high status in the specific group to which it belongs, becoming a grand narrative and superstory that the group relies on to identify itself (Honko 1996:21–22). A people blessed with an epic tradition finds its identity through epic characters and events.

The identity-bearing thrust of epic In numerous reference books, the definition of epic is based on its character as a literary genre; but it is necessary to focus as well on social and cultural functions as an indispensable dimension of its definition. The compilers of such books generally divide epics into two categories: oral folk epics like Sundiata in Africa and King Gesar in China, and literary epics like Virgil’s Aeneid. In academic nomenclature, the oral epics are commonly called “original” and written epics “secondary” (Baldick 2001:82). Of course, it is not always the case that an original epic is first passed down orally and then followed by the secondary form, its DOI: 10.4324/9781003258001-7

66  Critical reflections on epic studies written imitation (Jackson-Laufer1996: ix). For example, in some traditions in the Far East and the Middle East, many written narratives appear first and are then followed by oral developments. Based on the protracted considerations of the national identity expressed in the Finnish epic Kalevala and in-depth study of the connections between Indian epic and the Indian people, Lauri Honko concluded that epic is an important resource in promoting national identity and that this identity function is one of the defining elements of epic. On this basis, he gives his concise definition of the genre: Epics are great narratives about exemplars, originally performed by specialised singers as superstories which excel in length, power of expression and significance of content over other narratives and function as a source of identity representations in the traditional community or group receiving the epic. (Honko 1998:28). This definition is revolutionary. Previous epic research had focused too much on the analysis of the epic text; it lacked consideration of situating an epic work in a specific time and place for a comprehensive, multi-dimensional grasp of the genre. According to Honko’s understanding, epic is an integral part of a certain cultural tradition, and it needs to undergo two stages of a selective process to become transmuted from a local traditional cultural component to a symbol of group identity. At the first stage, specific elements in an epic are gradually selected to enter the general system of the group’s cultural traditions, for example, the gradual upgrading of some local deities to the status of national gods. Honko points out that tradition is a concept based on general accumulation, like a library that gathers and stores all kinds of materials. The collected materials include both physical and non-physical items; but they are initially not carefully integrated and arranged, nor can they be fully presented in a particular traditional event or activity. Some materials in the accumulated tradition are like icebergs partially exposed in the sea, while a large number of other materials are waiting to be actively absorbed. On the other hand, because they have not been activated or used for a long time, some materials are destined to be forgotten and never have a chance to let their original function play itself out in the national area (Honko 1996:19). To put it more simply, various factors of a tradition converge to form a unified cultural system in a gradual and orderly process, each serving the whole with its own particular function. Selection is key. Those factors that are in line with the system of cultural interests and values become closely connected with the community and integrated into its social life as a whole. Culture thus exists not so much in things as in the process of people’s observation, reflection, and integrated use of elements of the tradition. It lies not so much in content as in function and application value (Honko 1996:19).

Lauri Honko  67 At the second stage in the selection process, the movement is from cultural system to national identity. In the cultural interaction of this stage, some traditional epics are selected as representative symbols of the group’s characteristic identity. This amounts to a process of sublimation and concentration by which selected currents of language, music, dance, customs, architecture, history, myth, stories, rituals, and other traditional materials no longer refer to a particular content or form but are given symbolic meaning as an expression of group identity that far exceeds their original meaning. In Honko’s view, group identity can now be defined as “a set of values, symbols, and emotions joining people, through constant negotiation, in the realization of togetherness and belonging, constituting a space for ‘us’ in the universe (as well as distinguishing ‘us’ from ‘them’)” (Honko 1996:20). Honko calls those aspects of tradition that are selected for integration into a people’s identity as an expressive system “super tradition” or “focused tradition.” They are seen as especially representative of the group’s culture and a focus of its identity as welded from its traditions and culture. It is through this complex selection process that epics often become a medium of a people’s identity, a sign and symbol that unites a specific group and strengthens its cohesion and internal consistency, a banner emblem and bugle call that inspires and summons the people. Honko cites the position of the anthropologist Susan Wadley, who advocates the study of epics for the unique connection they have with the group in which they are performed: Epics are “our stories,” which distinguishes them from other songs and stories in the same community due to their identifying relationship with the community (Honko 1996:40). Honko himself stresses that epics do not just carry forward historical memories and indigenous mythological concepts like the origin of the universe from chaos; in the process of selection, they come to constitute a symbolic embodiment of national culture and identity.

Figure 6.1  Tradition-culture-identity.

68  Critical reflections on epic studies It has been pointed out that Honko argues, furthermore, that epic is not just a long story of identity; it is also a long story about alterity (Connelly 1986:25). It highlights otherness and difference in line with UNESCO’S 2001 Universal Declaration on Cultural Diversity, which states: “Culture has different manifestations in different times and in different places. The specific manifestation of this diversity is the uniqueness and diversity of the characteristics of the groups and societies that make up humanity.” The diverse manifestations of human culture allow us to see clearly that the creation and existence of epics in different ethnic cultures are quite distinct from one another and to recognize that people’s methods of understanding and interpretation must also be different. Before the 18th century, the understanding of epic in Western academic circles was based mainly on Aristotle’s Poetics and Homer’s epics and rooted in philology. As the trend toward Romantic nationalism then began to sweep across Europe, however, scholars presented by the Grimm Brothers focused readers’ attention on folk poetry and promoted the rediscovery of the diverse panorama of alternative oral epic traditions. By the mid-19th century, oral epic as a part of folk cultural heritage entered scholars’ field of vision; and by the mid-20th century, orality itself came to the fore as Western epic scholars introduced multi-faceted research methods of linguistics, anthropology, and folklore along with classical philology to gradually reveal the richer cultural connotations of oral epic. The rediscovery of epic traditions in Africa, Asia, and other areas of the planet promoted, moreover, the understanding of epic diversity by scholars all over the world. The classical Western concept of epic has thus gradually expanded from the traditional scholarly discourse to a more cosmopolitan, regional, and localized understanding of epic based on the consensus of scholars all over the world.

Personal and community epic identity The diversity of epic traditions naturally leads to diversity in epic manifestations of identity. Not all such manifestations reach the national level. The identity function of some important epics serves smaller communities and even individuals. The Dravidian Tulu- and Kannada-language Siri Epic of Karnataka in southwestern India provides Tulu-speaking women with a symbolic resource for personal identity. It is not about war but about peace and tranquility and about local social customs and rituals. It is a story about women’s self-esteem, self-reliance, and morality that vigorously extols the moral values of justice and dignity, loyalty and purity, independence and freedom, with a strong feminist color. The protagonist is not a brave male hero but a strong-minded, independent woman, Siri, whose behavior serves as a model for women in the local community. Siri’s bravery is reflected in how she challenges the ruling male order in her village, boldly undertaking actions that ordinary village women have never done. She leaves her heretic husband and abiding by the moral standards of a woman, lives a relatively independent life. Later, she chooses a second marriage, has a son, Kumara, and eventually becomes a goddess. Every year, Tulu people hold a festival in

Lauri Honko  69 her honor, in which her devotees, mostly women, chant her legend and become possessed; and the women find a sense of personal identity as members of the Siri cult. Once a woman is selected as a medium for Siri, she loses her original identity and obtains a new personal identity as the goddess’ regular attendant, protected by her and supported by her other mediums. She thus forms a group identity with them, but they do not usually have contact with one another. Only Kumara, the male leader of the group, will visit Siri’s devotees from time to time to help solve difficulties and strengthen their position in the family and the community (Honko 1998). Somewhat similar is the sense of personal identity held by the deity-imparted bards and several other types of performers of the Tibetan epic King Gesar. The famous Gesar bards Drakpa, Yumé, Tsering Wangdü, and Samdrup all claimed that their performance ability was somehow granted by the gods in one way or another. They have all said that in their youth they had special dreams or strange experiences of illness, etc., and could then recite and sing King Gesar without learning it from a master. This group differs from other professional Tibetan groups like the oral storytellers, recitation bards, prasenā divination bards, and treasure bards mentioned in Chapter 1. For some Gesar bards, the identity-shaping power of the epic is reinforced by being blessed by a tulku (reincarnate custodian of Buddhist teachings) and given a drupzha hat, which distinguishes them from other Gesar singers. A lama blessed and gave a drupzha to the singer Drakpa, and Juma Rinpoché once opened the mind’s eyes and gave the drupzha for singing King Gesar to the singer Yudruk. The drupzha is the most important gear for a deity-imparted bard. Its shape and structure symbolize Tibetan cosmology, and it is regarded by singers as a source of revelation. They believe that while they wear it, poems will naturally pour out of their mouths. Before a formal performance, some bards therefore hold a drupzha while reciting the “Ode to the Hat,” stating its symbolic meaning. They then put it on to begin chanting. In another rite, before singing King Gesar, the bard offers juniper incense. Holding Buddhist beads, he sits for a moment, meditating to achieve spirit possession, and then speaks and sings. Before every main segment of the epic, moreover, a bard will chant a verse praising the gods or praying for the gods’ blessing (Chao Gejin, Yin Hubin, and Bamo Qubumo 2010). The Tulu and Kannada epics of Karnataka and Tamil Nadu also play a role in constructing and maintaining the self-identity of a whole community. During religious festivals, these regional epics are often performed before a temple along with sacred rituals. The temple’s host is a Brahmin with the sacred power of arranging epic performance venues and rituals and directing regional epics to be performed before the threshold of the temple. In the hierarchical order of the deities, those within the temple have a higher status, while the farther the gods are from the temple, the lower is their status. This hierarchical system is reflected in the epic poems. The higher the status of the deity celebrated in the epic, the closer to the temple it is chanted, along with the ritual for that god. Because of the low status of the goddess Siri, the Siri Epic and its related ritual can only be performed

70  Critical reflections on epic studies not in the temple itself but in a less sacred area in front of the temple entrance. Nonetheless, it is performed, not just as an engaging story but as a means of community protection and healing. After the ritual performance, the goddess Siri lives in the rural community temple, blessing villagers’ well-being for the coming year. The performance of the Siri Epic thus serves as the bearer of the Tulu people’s communal spirit, ideas, and emotions and a source of their religious and distinguishing regional identity. In western India’s Rajasthan, the epic Pabuji celebrates bhomiya, deified local heroes who gave their lives to resist the looting of cattle. The intricate storyline presents gods and incarnations of deities, promises and fulfillment of promises, feuding, and final revenge. At the end, Prince Pabuji fulfills his promise to retrieve stolen cattle but is later killed; and eventually, his nephew Rupnath avenges the family and becomes a sage (Blackburn 1989:25–26; Jackson-Laufer 1996). Pabuji’s celebration of local heroes gives it a sacred function in the community. Summoned by a bhopa (psychic medium), the spirits of long-dead bhomiya honored in the local temple appear and begin to deal with people’s problems. The well-controlled, vivid display of the bhomiya possessing the medium attracts people from all around; and the temple serves as an important ceremonial site for the celebration of the heroic story (Nagy 2008:65). The ritual purpose of performing the Pabuji epic is not only to worship the bhomiya heroes. More importantly, by chanting the story to summon the spirits to come and protect the local community, the performance reinforces community unity and identity. One distinguishing characteristic of some Indian epics is that they center on a band of brothers. The heroes of Dravidian Brothers Epics are neither warriors nor classic Indian ascetics but craftsmen and farmers; and the stories give verbal expression to value judgments held by local communities that are far from the political center, mocking many pan-Indian norms and behavior patterns. Brenda Beck has called them superstories with an identity function. They describe only the concerns of the community, and they have traits that clearly show the high status that this particular genre has in the community. The epics are told over a broad region by professional folk bards. They are more heroic than other regional tales, and figures of the story are enshrined in the local temple. The epics have connections as well with a wider realm of myth and culture and are believed by both narrators and audiences to depict real historical events (Honko 1998:27). Velcheru Narayana Rao once conducted fieldwork on six Telugu folk epics describing heroes fighting against death and divided them into two types: war epics and sacrificial epics, each type with its own audience. They are closely linked with the life and culture of their particular community, and the social and economic structure of that community directly affects the personality of the heroes and the spirit of the story (Rao 1986:162–163). The Telugu people believe that the stories depict real events and that there is a deep correspondence between their conceptual world and the real world of the community. On the level of community or regional identity, Indian epics are a lofty narrative with an authentic flavor. They bring to life the community’s socio-economic

Lauri Honko  71 structure, core values, feelings, and ideals. They depict the origin and relationship of different groups in the community and give explanations of the various groups’ power and resources. At the same time, they confirm the distinguishing identity of the community as a whole—the status of its various inhabitants and the rational legitimacy of various rights and obligations. India is a country where people live in countless subculture circles. Epic performances reinforce internal identity; at the same time, they separate subcultural circles from other such small cultural groupings. In China, there are many corresponding traditions. To give just one example, the Miao people’s large-scale epic King Yalu, which was discovered and published not long ago, has a similar identity function. The migration history of the Miao nationality spans out in a tree-like way. The inner identity of each branch has clear but complex markers. Not only surnames and families but also larger community relationships are carefully woven into a complex pedigree and identity distinctions. Small-branch identities seem to be juxtaposed, but they have relationships of unity and division on the time axis that are vividly manifested in different versions of the same epic narrative as it gives concrete expression to the specific identity scope of each branch. People in the Mashan area of Guizhou Province who speak the western dialect of the Miao language have inherited the narrative tradition of this epic. The boundaries of their particular subculture are relatively clear; and each performance strengthens their own internal identity (Chao Gejin 2012a).

National identity epic In the evolution of epic around the world, some works were originally disseminated within a local area but then spread to other areas and became elevated to a work able to be shared by other communities or regions and even by the whole nation. Of course, in the process of this expansion, the content and form of a regional epic will change to varying degrees. As Stuart Blackburn has explained regarding Indian epic, as the work begins attracting audiences beyond the small local group, it breaks out of its original base site and comes to acquire new elements. If the original group celebrated a dead hero, the death motif may be weakened in favor of deification. If at the local level, an epic tells of the birth of a god, at the national level, the god achieves pan-Indian identity; and the incarnation of a local hero god evolves into a transcendent divinity. When the hero god is recognized as a pan-Indian figure, the process is complete (Nagy 2008:67). As was noted in Chapter 3, taking fifteen living Indian epic traditions as examples, Blackburn describes four levels of epic circulation areas: (1) local (10 to 100 miles); (2) sub-regional (100 to 200 miles); (3) regional (200 to 300 miles); and (4) super-regional (400 miles and beyond) (Nagy 2008:67). Mahābhārata and Rāmāyana are super-regional epics. In the process of expansion, they have been modified and sculpted by different community traditions. Regional historical accounts have gradually decreased, while fictional narrative elements have gradually increased; and the stories became increasingly alienated from the daily

72  Critical reflections on epic studies life of the original audience and evolved into a symbol of identity for the entire Indian people. In China, King Gesar belongs to this super-regional category. During the long course of its evolution and distribution, it gradually developed into a symbol of Tibetan national identity and then spread abroad to other ethnic groups, with complex textual forms and variations in the storyline. The grand sacred narrative celebrating the life of the ancient Tibetan hero Gesar contains within its uniquely linked structure numerous ancient myths, legends, tales, songs, proverbs, riddles, and other oral literary forms, all merged into a superstory that has dynamic momentum and rich connotations.1 It flourished mainly among Tibetans and Mongols, most notably among the Tu, Yugur, Moinba, Lhoba, Naxi, Pumi, and Bai peoples in Tibet, Qinghai, Gansu, Sichuan, Yunnan, Xinjiang, and Inner Mongolia. However, it has also spread to regions of Nepal, Bhutan, India, Pakistan, and Mongolia, and in Russia among the Kalmyks, Buryats, and Tuvans, becoming the common cultural heritage of numerous ethnic groups sung in the local mother tongues. Such a phenomenon of cultural sharing is relatively rare. In the process, the original internal identity function of the legendary King Gesar became greatly weakened, while its entertainment function has been well preserved in localized developments. In the Mongol Buryat Abai Geser, for example, at a great distance from the Tibetan region, the original Tibetan cultural aspects are weak, while features of Mongol narrative artistry are highly distinctive. If we look at different epics within a particular group, we find that even in similar cultural environments, the identity function of an epic can evolve in complicated ways. The expanding range of epic identity and the degree of influence are not simply equivalent to its stage of historical evolution. Looking at the inherited forms of Mongolian epic in China in the middle of the 20th century, we find that the Mongol Geser has been sung everywhere in Mongolia but has lost some of its marks as a national epic, while the Oirat Mongols’ Jangar is on the way to becoming their national epic. As for more regional or tribal Mongolian epics, Horqin epic remains at the stage of a regional epic, and based on long-term investigations, it seems unlikely to be upgraded to an epic of the Mongol people as a whole. In sum, the course of an epic is evolutionary but not unidirectional. Palnâtiviracaritra, Mahābhārata, Rāmāyana, and King Gesar all developed naturally, whereas the Finnish Kalevala, the Estonian epic Kalevipoeg, the Latvian Lāčplēsis, and the Indian Tegulu Palnâtiviracaritra are national epics consciously compiled from traditional material (Honko 1998:37–43). Both types of epic are forms of poetic creativity, but the latter tend to have a stronger political motivation. In the 19th century, many Romantic nationalist-minded scholars, such as Herder, the Schlegel brothers, and Hegel, argued that the origin of literature was rooted in epic and that the emergence of epic marked the formation of a nation. They praised epic as a national opus, the “bible” and “exhibition hall of specimens of the national spirit” (Hegel 1997:108). As we have seen, inspired by this trend and driven by the strong realistic appeal of the awakening of Finnish nationalism, Lönnrot gathered together numerous ancient Finnish poems, tales,

Lauri Honko  73 and folk materials to create Kalevala, the revered symbol of Finnish national and cultural identity. It has rightly been said that an epic can have vital relevance for the construction of national character and even the independence of the state (Maria Vasenkar 2002:2–4). Driven by the inspiration of Kalevala and the power of nationalism, Friedrich R. Kreutzwald (1803–1882) compiled the Estonian national epic Kalevipoeg from traditional folk songs and legends. Compared with the Finnish work, traditional oral materials take up less space in this epic; and the work is marred by a somewhat incoherent plot and anomalies in Kreutzwald’s combination of verse and prose. He uses the rhythms of old Estonian alliteration, but only about one-eighth of the verses come directly from folk poetry; the rest are imitations. In neighboring Latvia, Andrejs Pumpurs compiled what became that country’s national epic, Lāčplēsis, based on the indigenous-origin narratives, folk tales, local legends, wedding songs, and other oral embodiments of traditional wisdom. Published in 1888, this epic describes the achievements of the halfman/half-beast hero Lāčplēsis, who defeats monsters and enemies of the mythical world as well as all kinds of human invaders. The work has had a profound influence on Latvians. Many writers, artists, and politicians have drawn nourishment from it; and many streets and shops are named after characters in the work. Widely accepted as a symbol of national identity, Lāčplēsis was given political significance in the drive for national independence of the 1990s as a banner of nationalism. In terms of expression of national identity, The Poems of Ossian by the Scottish poet James Macpherson (1736–1796) had a different fate from Kalevala. Macpherson claimed that his poems were the works of the 3rd-century Scottish poet Ossian; but Samuel Johnson and many others doubted this. In the end, modern scholars demonstrated how Macpherson fashioned his own works grounded on old Gaelic narrative poems by modifying the original characters and concepts and injecting many of his personal ideas so as to cater to the times and interests. In terms of embodiment of national identity, the antiquity and authentic folk character of narrative materials are most important. In this, Lönnrot was admired as a national cultural hero, while Macpherson was widely criticized. Although epic is not the only kind of narrative capable of expressing the identity of a particular group—myths, for example, often have a similar function— epic is undoubtedly the best genre for doing so. Epic has attributes such as huge volume, serious and lofty content, great artistic charm, and a long history, which make it superior to other genres from this perspective. In Finland, no folk narrative can stand comparison with Kalevala; and the Tibetans have no other narrative of the stature of King Gesar. Though there once were other such narratives, they have not been handed down like this great epic. With its incomparable indigenous spirit, Gesar has absorbed all kinds of folk tales to become a great narrative tradition; and as with the Kalevala, its character as a carrier of national identity is clear. Although originating through different paths, both works have the status of a superstory that sustains the self-identification of its people. The significance of

74  Critical reflections on epic studies an epic lies in its power not only to tell a story about a hero artistically but also to carry forward the spiritual style and emotional leanings of a group or nation. It not only educates the people; it also strengthens their internal coherence through consciousness of common ancestry, a feeling of mutual belonging, and a sense of historical continuity. Epic performances connect stories inherited from perhaps thousands of years with the mores of daily life in a particular time and space. Persons living in the present gain self-identity in the process of repeatedly reinforcing their connectedness and dialogue through a traditional story that is both artistic and sacred. Looking back to India, the national Telugu epic Palnâtiviracaritra compiled by Akkiraju Umakantham differs somewhat from these northern European epics. It is based on the song Palnâtiviracaritra written by the famous poet Srinathudu at the turn of the 14th and 15th centuries. When published in 1911, the epic immediately became the spiritual banner and national symbol of the Telugu people. Gene H. Roghair once commented, The epic was important to a Telugu-speaking people in search of a Telugu identity and local models in a past that seemed too often dominated by foreign rulers and adopted foreign models for behaviour (Honko 1998:41). In line with what can be found in Kalevala, Kalevipoeg, and Palnâtiviracaritra, Honko has summed up four marks of a consciously constructed “tradition-oriented” national epic: First, there is the individual intellect to define the needs of the people by following an international model at a particular historical moment; secondly, there is the poetic material of ancient origin; thirdly, there is the literary élite to receive and interpret the epic; and fourthly, the whole process amounts to a transference of tradition from one type of environment to another, totally different (Honko 1998:41).

The Indian epic of national identity Mahābhārata The philosopher and linguist Wilhelm von Humboldt has acclaimed this ancient epic as perhaps the most profound and sublime work the world has ever produced; and it is natural that a nation like India, with its ancient civilization, can freely produce such a super-regional cultural miracle. Archaeological data tell us that in the 25th century BCE, the Indus River Valley entered the Bronze Age and produced the seal script texts, which, unfortunately, have not yet been successfully interpreted. From about the 15th to the 6th century BCE, the Vedic Age then became the key period in the formation of the Indian cultural system; and during that period, the Sanskrit epics Mahābhārata and Rāmāyana emerged. Mahābhārata, the story of the great emperor Bhārata, was formed over about 800 years from the 4th century BCE to the 4th century CE, confirming a universal cultural phenomenon that a great narrative tradition often takes form in an ancient

Lauri Honko  75 civilization: the Babylonian Gilgamesh, Greek Iliad and Odyssey, Anglo-Saxon Beowulf. The creators of Mahābhārata were poetically talented sūta troubadours and court singers. They created large-scale heroic narratives from the nutrients of myriad Indian cultural and narrative traditions—from the chaos of the first creation myth to the complex lineage of rajahs, from the ups and downs of nations to tales of wars, from legendary heroes to specific historical events, from religious beliefs to philosophies of life, from moral laws to folk experience. As in other great narrative traditions, the massive work in the end came to stand as a symbol of identity for the whole Indian people. For a long time, Western scholars believed that Mahābhārata was the longest epic in the world.2 There are about 100,000 śloka (couplets), amounting to nearly seven times the sum of the Iliad and the Odyssey. It has been estimated that singing just the grand narrative dialogue between the two main characters at one śloka per minute would take twenty-five days and nights! The epic’s narrative structure is like a group of huge buildings with connected halls and adjacent courtyards surrounded by interconnecting pathways. There are 300 or 400 storytellers in total; and innumerable dialogues—independent or interrelated—often have their own narrators, interlocutors, and respective audiences. Story is set in story and dialogue in dialogue, which creates a narrative web of intertwined clues in fascinating twists and turns in the plot. It is not correct, however, to call Mahābhārata the world’s longest epic. The Tibetan epic King Gesar is much larger in scale. The publication of the famous Tibetan singer Samdrup’s recitation of the work has so far taken forty-five volumes of 400 pages each, which is only about two-thirds of the total inventory of his output. The ability of folk bards to remember and sing such superstories is truly amazing. The process of composing and disseminating the Mahābhārata is extremely complicated. As was noted in Chapter 1, there are as many as 700 emendations listed in the Bhandarkar critical edition, which shows that efforts to shape the work never stopped in the course of its history. However, in the era of its initial formation and flourishing, it was disseminated orally. Though it was later revised by innumerable Sanskrit bards, it is essentially an oral poem with a strong color of oral poetry that can be felt even by today’s readers in many ways. To begin with, there are many formulaic phrases and typical scenes, as is common in largescale poetry. As with Homer’s works and other originally oral epics, the use of character epithets and detailed lists of things is common. Bhishma (Book 6 of Mahābhārata) gives detailed lists of 147 rivers and hundreds of famous regions. No literary author writes poems like this, except to some extent Walt Whitman in Leaves of Grass. As is common in oral poetry, too, the composition, dissemination, and reception of Mahābhārata took place at the same time and in the same performance site, with active interaction between singers and listeners. The story is thus highly contextual, its interpretation inseparable from the background Indian culture. Judging from the large-scale living singing traditions found in various parts of the world today, we can infer that this kind of long-standing ethnic narrative performance must often have occurred in set cultural spaces as an organic part of traditional ritual activities.

76  Critical reflections on epic studies Such an epic is not a static work but a flowing tradition. The tradition not only cultivates countless highly talented performers from generation to generation but also shapes the cultural spirit and identity of the entire people. It lets them know how to deal with the universe, life, and all things in the world; it teaches them how to think and act in the world, what is right and wrong, what is just and evil. People must have listened to the magnificent Mahābhārata stories countless times on numerous occasions, deeply impressed by the heroic feats they display and inspired by the work’s profound thoughts and high aspirations. In it, they had an encyclopedia of folk knowledge and wisdom that rightly deserves its high renown. The performance of Homer’s poetry in the ancient Panathenaea was very popular; and ever since, there have been hundreds of written texts of the poems circulated throughout history down to the numerous translations of this day. This is also true of Mahābhārata. In the history of its transmission, there have been many transcripts, which shows that the singing activities have been active and extensive. It is an extremely arduous task, however, to transpose an ancient flowing narrative tradition into a printed text and establish a critical edition for the use of the academic community. Indian scholars took from 1927 to 1966 to produce the six-volume critical edition with the aim of restoring the original form of the epic as much as possible. For their part, Chinese experts in the field of Sanskrit literature have devoted more than ten years of hard work to the translation. In the Mahābhārata, the Hindu God of Justice Yamaraja, in charge of the world, turns into a River Spirit, which asks a question of children who come to drink the water: “What is faster than the wind?” The answer: “Thinking.” As the poetic heritage of the Indian people, the ancient epic embodies their wisdom; and Indian scholars believe that it has cohesive moral teaching power for India today. Its influence, however, has transcended national boundaries. It has become a cultural treasure of all humankind. Most of the material in this chapter was originally published in Folklore Studies, Vol. 5. 2012; that about Mahābhārata in the China Reading Weekly, 9, February 15, 2006.

Notes 1 The “Epic World Heritage Application” group of the Institute of Ethnic Literature, Chinese Academy of Social Sciences has published two texts of Gesar (Geser), in 2005 and 2009. 2 See the New Princeton Encyclopedia of Poetry and Poetics, 1993 edition.

Part II

Theories and methods of oral poetics



7

Oral poetics and the Oral Formulaic Theory

Oral poetics We have seen that research on oral poetics has made great progress in the West. As a research methodology, it has been inextricably linked since its birth with investigations on oral traditions in general; for oral poetry has a close relationship with other folk oral literary styles and performances. In areas where oral communication is the basic channel for the transmission of indigenous knowledge, the concepts and classifications of “sung” and “written” are quite different. The function of poetic rhythm, for example, is sometimes more a matter of mnemonics and learning methods than of considerations of phonological beauty. Twentiethcentury research on oral poetics was first initiated from within the discipline of classical studies as scholars reexamined the long-standing Homeric Question. It began by questioning the norms of classical poetry and focused on the unique attributes of oral poetry. Anthropologists and folklorists joined literary scholars in the questioning; and together, they gradually developed a complete set of theoretical systems, including Oral Formulaic Theory, Ethnopoetics, and Performance Theory. Oral Formulaic Theory takes oral poetic formulae as the main research object and finds that the expression of set formulae constitutes the core feature of oral poetry. The notion of a formula has different definitions in different poetic traditions, but a basic feature is that it is an expression that is used repeatedly. Its function, however, lies not in repetition itself but in the construction of verses of poetry. Formulae are ready-made expressions with fixed meaning (often in accord with a specific prosody format) and formed within the tradition, handed down from generation to generation. A qualified singer needs to learn and preserve numerous such expressions. In research practices, the frequency of the emergence of formulae often becomes an index for determining whether or not a poem has an oral origin. Assessments of early European works like Homeric poetry and the French Song of Roland to a large extent use this method. The meaning of a formula is often much more complicated than first meets the eye. Behind the literal meaning is traditional referentiality, the poetic register. When Homer describes what an elegant woman does with “fatty hands,” the expression seems strangely inappropriate. In fact, however, in the poetic register of the Greek epic tradition, DOI: 10.4324/9781003258001-9

80  Theories and methods of oral poetics “with fatty hands” means “courageously.” This traditional connotation cannot easily be got from reading the text or from a dictionary definition. The interpretation of oral poetry relies more on the cultural soil of the poetic tradition in which an expression is rooted than on just textual analysis. This is the case with the traditions of many ethnic peoples in China. In Mongolian epic, for example, we find many expressions that readers find surprising; but for an experienced audience, these expressions are no problem at all. The local audience knows the traditional implications or register behind the expressions. Of course, the marks of oral poetry exist on more than the level of words. They exist on other levels as well. The first is the guiding role of sound patterns. In Mongolian poetry, rhyme is an important feature; and many mistakes made in a performance are related to it, especially when a singer improvises. Another characteristic of oral poetry is the use of parallelism. In Chinese traditions, parallelism, couplets, and similar techniques are popular methods used by folk singers. Another characteristic, too, is the low density of newly created formulae in the sentence patterns. Scholars have found that Homer’s skill manifests itself in the way he makes sparing use of newly created expressions, relying, rather, on readymade expressions. If a poet has these for ready use, he/she will not need new expressions or formulae. Still another characteristic of oral poetry is redundancy. In reading, repetition is hard to accept; but in listening, it is not. On the contrary, it is a strong point. The relationship of expressions repeated in an orderly fashion in the timeline is reinforced by their recurrence. In many aspects, oral poetics deals with the problem of how to attract people’s interest and attention. In different traditions, this is achieved by different ways of aligning expressive units. In Western written poetry, for example, enjambement is frequent, whereas in oral traditions, bards use verse lines to measure out units of expression. According to Mongolian epic singers, however, a spoken line can involve the overlapping of one or more words. When printed, the line may seem not neatly aligned; but in performance, it constitutes a clear unit in close combination with the rhythm, meter, and melody. The key is performance. Most of the time in a performance, a verse has a relatively strict rhythm in accordance with the requirements of the meter. When written, it looks highly irregular, but the sung verses sound relatively neat. Singers often lengthen vowel tones to accord with the appropriate rhythm so that seemingly irregular verses do not sound irregular at all. There is a great degree of elasticity between the number of words in a line of verse and the beat. Large-scale oral poetries are commonly improvised on the spot. Though the text of a particular performance has a direct relationship with the same storyline in previous performances, it is a new text. The song is not limited to a set narrative text; it is full of new elements. The storyline in each performance serves as an inner text that is enhanced in performance. The process of composition, dissemination, and reception of oral poetry forms a coherent unity. The audience’s emotions and reactions affect the singer’s performance, deciding the length of the narrative, the complexity of the details, the elaboration of words, and even the structure of the storyline. The make-up of the audience itself may affect the

Oral poetics and the Oral Formulaic Theory  81 identity of the protagonist. When telling a story to audiences from different ethnic groups, the performers of Serbo-Croatian narrative poems adjust the main characters and the storyline to cater to each group. Recent studies have shown that there is not always an insurmountable gap between oral and written poetry. As a matter of fact, in many societies, singers follow the local tradition while being influenced at the same time by modern education and written culture. The narration may employ some rules of written literature and have characteristics of written literature. In a word, the relationship between oral and literary poetry is not so much polar but rather, a genetic relationship with borrowings from both sides as a kind of intermediate hybrid. With the in-depth study of oral poetics, moreover, the long-used yardstick for judging the value of works of literature has been questioned. Can the aesthetic principles of written literature be used to explain oral literature? What kinds of poetry can be selected for the list of works in the treasure trove of world literature? Why are some great oral works not included in the list?

Homeric poetry and oral tradition Some ancient writers credited Homer as being the author not only of the Iliad and the Odyssey, but also of the lost Greek epics Cypria and Iliad Mikra. As we have seen, as early as the 1st century CE, the Jewish historian Flavius Josephus pointed out that Homer’s poetry was an oral form of poetry composed somewhere and then remembered, with internal inconsistencies that later puzzled generations of Homer researchers. In the 18th century, Vico maintained: As for the contest among the Greek cities for the honor of claiming Homer as a citizen, it came about because almost all of them observed in his poems words and phrases and dialectical locutions that belonged to their own vernaculars (Vico 1948: 273). He noted also: “We find support for the opinion of those who place him long after the Trojan War. The time runs to the extent of 460 years, or until the period of Muma” (Vico 1948: 275). This line of thought gave rise to the debates between the Analysts and the now generally less favored Unitarians, to which some in China still adhere.1 In the first half of the 20th century, however, Milman Parry came to realize that neither Analysts nor Unitarians touched upon the essence of the problem. Through a precise classical linguistic analysis of the Homeric texts along with what he learned about oral poetry from fieldwork among Serbian and Croatian bards, he became certain that the poems ascribed to Homer were originally oral works. Of course, there is no authoritative text of an oral epic. This was the case with Homer’s epics. It is said that the Athenian ruler Peisistratos (about 600–527 BCE) first compiled his poems. The Athens (Peisistratos) text was the standard text used at the Panathenaea. In the 4th century BCE, both Plato and Aristotle quoted his poems extensively, and some words were quite different from those of the common texts today. By the time of Alexander, the following four texts

82  Theories and methods of oral poetics were circulated in Greek localities: (1) the rigorously transcribed and generally accepted text; (2) texts of various regions or city states; (3) texts edited and collected by certain individuals; (4) texts used by bards themselves. On the basis of these texts, mainly the first, the 4th-century BCE scholar Zenodotus revised and sorted out the “standard text” of the Iliad and the Odyssey; and in the 3rd century, Aristophanes of Byzantium and Aristarchos of Samothrace contributed to the set text with commentary and division into books. It is generally believed that the text from their hands is the direct predecessor of the modern Homer. Venetus Marcianus A is the earliest extant copy of the Iliad, written in the 10th century CE; the earliest extant Odyssey is Laurentianus, written in approximately the same period or later. There are also many epic fragments of varying lengths, some perhaps from the 3rd century BCE. There are three manuscripts of The Song of the Nibelungs,2 and a great many other epics lack a single authoritative text.

Parry’s determination of the oral nature of Homer’s poetry How did Parry come to establish that the Homeric epics were originally transmitted orally? When faced with the written text, how can a person determine whether it is basically a written literary epic or an oral folk epic? The information available is so fragmented and contradictory, where does one start? Beginning with detailed textual analysis, Parry analyzed the non-epithet formulae woven into the poetic syntax of the Iliad and the Odyssey and realized that they have the function of a phonological form to be used for the process of composing the works. Homer’s hexameter constitutes a complex metrical network that can accept only certain words and phrases implanted in specific positions. It thus plays the role of a mechanical selection mechanism. Over time, with repeated use by generations of singers, it evolved into a dimension of the poetic syntax adopted by each singer and passed on to his successor. Parry showed exactly how this was traditionally systematized. He noticed two characteristics of the poetry’s constituent idiomatic units: simplicity and efficiency of length. For example, when Homer says “then X replied,” he uses the same phrase whether X is “Queen Hera of the cow’s eyes,” “Nestor, breaker of horses,” “gray-eyed Athena,” or “Diomedes of the loud battle cry.” Parry selected twenty-seven subjects and twenty-four predicates in Homer’s poetry that have a formulaic dimension and found the evolution of 648 combinations that are derived from these two sets of components. This clearly shows the power of Homer’s rhetorical means (Parry 1971:1–239). However, such combinations are limited. Granted the restraints imposed by the meter and the need to express particular actions or concepts, it is extremely rare that multiple formulae are available for the singer’s selection. Parry thus speaks of Homer’s “thrift,” meaning that he uses such formulaic expressions very economically. As indications of the basically oral nature of Homeric poetry, Parry explained that they free an oral poet from the trouble of having to make choices. Parry then went on to study the nature of run-on lines, or enjambement, in Homeric poetry. By comparing the form of his lines with those written works by liturature figures like Apollonius and Virgil, he found that in as much as half of

Oral poetics and the Oral Formulaic Theory  83 Homer’s poetry, the poet does not use natural run-on lines. Their frequency is half of that found in those literary works. Homer depends, rather, on formulae, along with the omission of conjunctions, to create poetic units limited to the metrical line. Parry found this to be evidence of oral poetic syntax. A singer has no time to conceive beautiful balances and contrasts. He must use this syntactic style for freedom in arranging the sequence of sentences on the spot. To test his inferences about oral traits in Homeric poetry, Parry wanted to investigate a living oral epic tradition. Influenced by the field reports of Matija Murko and others, he selected Serbo-Croatia as his test site. After short preliminary soundings, Parry, together with his student Albert Lord, conducted a sixteen-month investigation there in 1934 and 1935, recording nearly 1,500 epic texts, including many different versions of the same work, and established the forward-looking database of the Milman Parry Collection of Oral Literature at Harvard University. Regarding singers’ memory, he found that the outstanding illiterate singer Avdo Međjedović had fifty-eight epics stored in his memory. Avdo is not a special case. Similar cases have been found from oral folk traditions around the world. Parry’s fieldwork investigations showed that such epic bards do not recite an epic word by word but rely on formulaic themes, typical scenes, and set story-patterns to structure works according to inherited oral poetic canons. He came to realize that the Homeric poems must have been formed in accord with such oral canons. Together with Lord, Parry thus came to establish the Oral Formulaic Theory (Parry–Lord Theory). It has rightly been said that Parry’s rigorous, creative analysis of the epic texts is a brilliant example of epic research. While others base their investigations on a large variety of examples, Parry makes a very thorough analysis of only one kind of illustration and uses it as the best way to verify his theoretical insights. Whereas others seek to describe a particular law, he seeks to demonstrate a system governed by the fluctuations of laws. Whereas others are content to point out a particular aspect of Homer’s style, he seeks to give a comprehensive interpretation. (Foley 1988:26)

Lord’s contributions to Oral Formulaic Theory After Parry unexpectedly passed away in 1935, Lord continued and deepened research in this field. His contributions to the development of the Oral Formulaic Theory include studies on the learning process and singing methods of Yugoslav epic bards, an intensive analysis of the concept of formula in oral epics, thorough discussions of formulaic themes and typical scenes, extensive coverage of the relationship between written and oral traditions, and comparative studies of ancient Greek, Medieval, and contemporary living epic traditions. Greatly expanding and advancing Parry’s early work, Lord created a systemic theory of oral formulae. His The Singer of Tales (1960) is considered the bible of the School of Oral Formulaic Theory and has had a profound influence. More than 1,000

84  Theories and methods of oral poetics published studies have benefited from his research, involving oral traditions in more than 100 of the world’s languages. Like Parry, Lord studied Homeric poetry, but he focused more on “performance units” as textual divisions and on textual inconsistencies and run-on lines, etc., regarding all of which he had his own insights. He maintained that a rich accumulation of formulae will lead to a higher level of variation in poetic creation and re-creation and that within limits, an accumulation of themes and storylines will lead to a large number of similar variants. In the composition of an epic work, sound patterns formed by syntactic parallelism, alliteration, and assonance play a leading role in the poet’s application of formulae. In all this, Lord emphasized the orality and heard dimensions of traditional poetry and stressed that the poetic syntax of oral epic is based on the formulae. With great subtlety, he suggested that there is a significant correlation between the density of formulae and the orality of a work. In line with this idea, some scholars argue that ancient French narrative poetry can be thought to have oral sources if the formula density exceeds 20%. In the Song of Roland, the density is 35.2%; so, it is easily judged to be oral. Regarding poetic composition overall, Lord concluded that the Serbo-Croatian Return Song is related to the decaying phase of the vegetation cycle. Historical events enter and are reflected in oral epics; but it is myth, not history, that gives birth to the ancient epic paradigms.

The significance of the Oral Formulaic Theory The Parry–Lord Oral Formulaic Theory establishes a model for the analysis of oral epic texts in terms of a holistic interpretation based on the living situation, using rigorous methods and detailed procedures. The school takes formulae as the core concept that captures the peculiarities of oral narrative literature, especially verse literature, and opens the way to solving many problems in the composition and dissemination of folk literature. The theory is applicable, moreover, not just to epic study; its approach to the analysis of the syntactic structure of poetry provides us with a strict, scientific model open to academic research in the field of poetics at large. It sheds light on various kinds of narrative folk songs and their formation, the formation of the Bible, jazz improvisation, African-American folk preaching, and other improvisatory genres. The theory is closely related to formalism and structuralism, but it is not a kind of new academic fad. Its rigorous procedures based on empirical findings have withstood the test of historical and cultural verification and ensure its long-term academic vitality. Under the entry “Folklore” in the Concise Encyclopedia Britannica, Lord is presented as a representative of “humanism” on the level of the anthropological school of Tylor and Boas and Freudian psychoanalysis. In the long history of Chinese ethnic literature, magnificent epic texts have become a treasure house of oral literature. There is no doubt that Mongolian epics have an oral origin and have been spread orally among the people, conforming to the principles of oral poetics. Nonetheless, records show that jangarch (Jangar singers) in Xinjiang can be divided into two types according to their way of

Oral poetics and the Oral Formulaic Theory  85 learning to perform: the one, more traditional, purely verbal; the other by reciting from a written copy. Among the latter, some stick faithfully to the written text, while others are more freewheeling (Rinchindorji 1999:39–41). The oral composition of Mongol heroic epic appears in various forms and can be quite complicated. Some works simply follow the plot lines of oral traditions, folk myths, tales, and legends; but some elements almost certainly come from printed texts. Jangar, for example, contains some storylines from the 1716 Beijing woodblock version of the Mongolian Geser. In any case, Mongol heroic epic is highly stylized in plot and structure, with the regular arrangement of fourteen motif series identified by the eminent Mongolist Walther Heissig: (1) the time; (2) protagonist’s origin; (3) his homeland; (4) his appearance, temperament, and property; (5) special relationship with his steed; (6) expedition; (7) helpers and sworn brother; (8) threats; (9) enemies; (10) encounter and fights with enemies; (11) protagonist’s strategy and magical power; (12) marriage quest; (13) wedding; (14) return home. The achievements gained from the empirical propositions and concepts of the Oral Formulaic Theory attest to the vital validity of such soundly based empirical approaches in the field of the humanities. Nevertheless, in their early work, Parry and Lord did not take into account all the factors of oral performance. In the composition of oral epic, creation and performance are complementary facets of the artistic activity. In performance, the creative activity involves both eyes and ears in the interaction between performer and audience; and good performers (creators) take full advantage of this interaction. Their styles rely not only on the deployment of vocables but on facial expressions and bodily gestures. When we read the recorded text of an oral epic in accordance with modern ways of reading written literature, no matter how interesting it may be, we miss these important factors in the creative process. The first section of this chapter was originally published in National Arts, Vol. 4. 2000; the remaining sections in China Folklore. Shanghai Literature and Art Publishing House, 1999. 176–189.

Notes 1 See Chen Zhongmei’s preface to his translation of the Odyssey (Guangzhou: Huacheng Publishing House, 1994). 2 Manuscript A is preserved in Munich, Manuscript B in Saint Gall, and Manuscript C in Donaueschingen.

8

Field investigations of oral epic transmission

Writing and written script are fairly recent human inventions. Compared with the long process of human evolution, the history of writing is quite short. This is so even in comparison with the history of human civilization as manifested in animal domestication, grain cultivation, and permanent settlement, which appeared in the Near East and North Africa around 9000 BCE. In 3100 BCE, Sumer invented hieroglyphics; from about 1500 BCE, China used ideographic characters; from at least 1450 BCE, Crete and Greece used Mycenaean linear B syllabic script; the Anatolian Hittites used hieroglyphs; and a few hundred years later, the Phoenicians developed the alphabetic script, which is the basis of the current Western script. There is a considerable time gap, moreover, between the invention of script and the extensive use of writing. In early history, writing was used to keep records, and readers were very rare, usually experts. Even later, in the European Middle Ages, reading and writing were the prerogative of a very small number of people. Manuscripts were extremely difficult to make and use; and in the entire population, the number of readers was very limited. Oral communication and dissemination of thought occupied the dominant position in the realm of information exchange. In today’s world, even in highly literate societies, people are very dependent on oral communication; and in a large part of the world, the populace mainly use oral means to communicate. It is easy to see this in western Africa, among native Americans, and as the main medium among those ethnic minorities in China who do not have a written script. In several nations that have used written words for centuries, moreover, the creation and transmission of oral literature still occupy the main position in literary activities. There are many ways to study oral culture. The purpose of this chapter is not to make a theoretical analysis of oral cultural phenomena but to clarify the unique nature of oral epic. We do so by analyzing oral epic fieldwork methods in two specific cultural traditions—Serbian-Croatian and Mongolian oral epic— so as to gain some local understanding and through that, improve our fieldwork. Through the pioneering work of Parry and Lord, the former came to the fore in international oral epic research. As for the latter, its academic history can be traced back almost 200 years; and it, too, has become a focus of international attention. DOI: 10.4324/9781003258001-10

Field investigations of oral transmission  87

Serbian and Croatian oral epic Prior to the investigations of Parry and Lord in the 1930s, oral epic fieldwork in the region of Serbia and Croatia was conducted by the Slovenian scholar Matija Murko (1861–1952) of the University of Prague. In his footsteps, the two Harvard researchers, after a brief field survey in the summer of 1933, began collecting material in earnest from June 1934 to September of the following year. Based in Dubrovnik, they gradually expanded the scope of their investigations to the diffusion of epics in Macedonia, Novi Pazar, Bijelo Polje, Kolašin, Gacko, Stolac, and Bihać, seeking representative oral epic singers from both Christian and local Muslim traditions. To determine general principles about the basic characteristics of oral epic performance, Parry thought it necessary to look at samples from differing traditions. It is more difficult, but much more valuable, to find similar laws in different cultural traditions than in the same or similar traditions. For recording their samples, along with the conventional method of transcription from oral dictation, they used what was at that time an advanced electronic recording device: aluminum disks. Through the interactive use of these two recording methods, they found that there are differences between acoustic recordings and orally dictated texts. In the latter, various factors are more abundant, and details are more fully described. The means of recording thus obviously affects the results. This salient discovery, even today, half a century later, is rarely noticed by epic field investigators. Parry and Lord focused on the best singers, the most talented inheritors of the oral epic tradition, in order to guarantee faultless observation of the typical form of the tradition. We have seen that the illiterate singer Avdo Međjedović (c. 1875–1953) made a deep impression on them with the fifty-eight epics he had in his repertoire, two of which, at 12,323 and 13,331 verses, each reached a length similar to that of the Odyssey. Unlike the one-time surveys and recordings that we often use now, they recorded not only a great line-up of songs by the same singer but also a goodly number of different versions or performances of the same song by the same or different singers for purposes of comparison. Avdo’s famous The Wedding of Smailagic Meho was recorded in 1935 and then again in 1951; and a song performed by the singer Demail Zogic exists in three versions: 1934 recording, 1934 dictation, and 1951 recording. This approach set an example for solid epic fieldwork. The two also conducted a series of field experiments. To expand on one that was mentioned in Chapter 1, after singing for several weeks, Avdo recognized that the pair thought highly of his performance. At this moment, there appeared Mumin Vlahovljak, a singer who also appeared to have good singing skills; and Parry learned that there was something in Mumin’s repertoire that Avdo had never heard before. Without disclosing the purpose of the experiment in advance, Parry asked Mumin to sing the poem. He then turned to Avdo and asked him what he thought about it and whether he could sing what he had just heard. Avdo replied that Mumin’s singing was very good, but that he could sing better! So, Avdo began to sing. Parry noticed that he greatly lengthened a party episode from

88  Theories and methods of oral poetics Mumin’s 176 lines to 558 lines, with not only modifications and added metaphors but even Avdo’s own understanding of the hero’s soul. The whole of Avdo’s version reached 6,313 verses, almost three times the length of Mumin’s original (Foley 1988:40; Lord 1960:103). In their fieldwork, Parry and Lord asked several questions: Are there set performance segments in a large-scale narrative? Are there specific rules to be followed for rest breaks in the singing? How do narrative inconsistencies and errors arise? What do singers think of their works? How do they understand “word”? Since the length of a song varies with each performance, what factors determine the length of a sung text? Parry and Lord found that epic singers in Yugoslavia may pause anywhere, either to rest their voices or for the sake of the audience. There is no direct parallel between narrative segments marked by intervals in an oral performance and those marked off in written texts of, for example, The Song of the Nibelungs or Poem of the Cid. Concerning inconsistencies and errors in narration, these arise because a singer constructs the storyline in accord with set motifs. Some highly patterned traditional motifs may fit the storyline but contradict or disagree with one another in some details. Preoccupied with the performance, a singer may not notice inconsistencies. So-called “textual errors” are unavoidable in oral performance, and they may not have a causal relationship with the unfolding temporal sequence or convergence of poetic elements. Moreover, the folk artists that Parry and Lord met did not know academic terms like “tensyllable verse”; and in the artists’ minds, “word” can refer to the “large word” of a complete line of poetry or an even longer poetic unit (Foley 1997). This means that in the minds of these singers, the smallest poetic unit is a phrase, not a single word. As for factors other than the text, Parry and Lord also paid careful attention to the huge role that the audience plays in a performance. Their response to the song determines how and to what extent the singer shapes it. “Leaving out of consideration for the moment the question of the talent of the singer, one can say that the length of the song depends upon the audience” (Lord 1960:17). In addition, Parry and Lord noticed in their field investigations that in those Balkan villages, printed texts of the songs have existed for at least a century. They are widely valued, and some are regarded as extremely sacred. In a strict sense, however, the artists dedicated to chanting these printed songs can only be regarded as performers of what were originally orally inherited narrative poems rather than real folk singers. Folk singers are those who have mastered all the traditional techniques and methods of epic singing and who construct each composition according to the fixed story-patterns and themes, being highly dependent on formulae, which are often fixed phrases of whole or half lines. Only by such real folk singers can the typical forms of oral narration be fully displayed. The following question then naturally arises: How do real folk singers come into being? Through what channels do they become such bards? Through investigations of numerous singers, Parry and Lord came to the conclusion that in the local societies, they do not form a special class or group. With the exception of a small number of paupers who earn their living by singing, most only sing to get paid during the Muslim holy month of Ramadan. To become a singer, moreover,

Field investigations of oral transmission  89 the learning process can be roughly divided into three stages. In the first stage, that of apprenticeship (usually from adolescence), they listen and absorb techniques; in the second stage, they practice; at the third stage, they have formal performances before an audience. The sign of completion of the third stage is that the singer has mastered a certain number of sung repertoires so that he can perform for several consecutive nights. However, Parry and Lord found that some singers seem never to complete this third step. A real folk singer is a person who can skillfully use the traditional narrative models and motifs, a master who can narrate a story smoothly and fluently under the restrictions of the traditional poetic meter format.

Mongolian heroic epic The collection of Mongolian oral epics began almost 200 years ago and has yielded considerable information and valuable materials (Nekljudov 1991:1). As far as Mongolian epics in China are concerned, since the 1950s, their collection has been carried out one after another, with translation and publication implemented intermittently. As our main achievement, we now have a certain amount of written records and electronically recorded materials, on the basis of which we have published, collated, and revised epic texts and carried out research. The situation of singers in numerous areas has also been established. Nevertheless, compared with the epic fieldwork carried out in Yugoslavia, our fieldwork has many rough and incomplete aspects, which will be detailed in the following chapter. It needs to be guided more by theory. It is not like the collection of folklore materials by travelers and explorers. Due to a lack of theoretically guided fieldwork, we are helpless when facing problems related to basic characteristics of Mongol epic: For example, how do Mongol bards remember and master the singing of large epic works? If, as in Yugoslavia, they create a new work with every performance, how do performances differ from one another? Regarding the mastery of their singing skills, the Russian scholar Vladimirtsov has pointed to “common passages” in Mongolian epic (Vladimirtsov 1993). Closely related to Parry and Lord’s concept of formula, they indicate that Mongolian epic singers manipulate traditional units to structure the narrative and move the plot forward. In an actual performance, how do they do so? What components are readily changeable, and which are the stable ingredients? It is not uncommon at home and abroad for handwritten texts to be circulated among bards and for printed texts to flow back to the people. As was noted earlier, singers who specialize in simply retelling handwritten or printed texts are performers but strictly speaking, not epic bards. Real folk singers do not just recite texts; they absorb and modify traditional textual materials to create new stories. How do we distinguish between them? Is this phenomenon the result of modern developments in Mongolian epic? As with other folk customs, the study of epic singing needs to start with specific live performances and their relationship with other performances. Is the text of a particular Mongol performance a unique creative work or just a replay of a

90  Theories and methods of oral poetics standard model? Can the recorded text give evidence of the general style of the whole sung tradition? If so, on what theoretical basis? If not, how can we claim that we have recorded an example of epics of a particular area? The epic text and the sung context, the “narrative event” of the performance, jointly create the meaning, with the context comprising numerous factors. Some epics are sung during the course of a special ritual and subject to specific taboos. If singing is interrupted at times by cheers and laughter, what is the cause? Is it due to the singer’s new improvised mode of expression, or because he or she is faithfully following the traditional text and performance paradigm as the audience knows it should be? On the other hand, if there are no interruptions at all, what is it that makes the audience keep silent? Is this a traditional taboo? The Russian scholar S. J. Nekljudov has noted that there is still a lot of work to be done in the fieldwork of Mongolian epic in China in terms of the classification and comprehensive analysis of groups of artists (Nekljudov 1996:133–143). In particular, he speaks of the folk tales of famous epic artists and the description of the “social context” of their singing: The population of the specific region, the average age of the audience, the local tradition of epic singing, the natural geographical form, the area and scope of the artists’ activities, and the social function and cultural status of the epic sung. As for analysis of the epic text, he stresses the division of verse and prose styles in a performance; the difference between manuscripts and oral performance transcripts; the attitude of artists and audience toward the text; the roles of musical instruments and tunes in the singing of the epic, etc. More recently, Nekljudov has called attention to the fact that Mongolian epic singers can be roughly divided into two categories: one who adheres closely to the tradition and the other who is adept at on-the-spot improvisation. This chapter was originally published in Studies of Ethnic Literature, Vol. 3. 1999.

9

Oral poetics and Chinese epic research Interview with the author

Interviewer Liao Mingjun: What are the current problems scholars face in research on China’s heritage of oral epic? Chao Gejin: From the start to current developments, domestic epic research has been similar to that in other disciplines. Theoretical thinking is relatively weak. There are certain advantages in classifying the material, but these advantages are in fact deficiencies. We in China have done a lot of descriptive work; but there is a lack of theoretical spin off and sound methodology in our academic consciousness. Liao: What are the theoretical problems that must be solved? Chao: Of course, the first is how to correctly understand the basic characteristics of oral epic. For many years, we have generally applied the methods of studying written literature to oral epic. This can be helpful, but there is still a long way to go before we can fully and accurately get a hold on the object of our research. In the past, it was inevitable that we misread oral works. This is the case not only with the study of the Three Great Epics—Gesar (Geser), Manas, and Jangar—but also with other epics of ethnic minorities, as well as with the study of foreign literary works and theory. When dealing with a printed text, persons often fail to realize the great difference between oral folk epic and written literary epic; so it is difficult to grasp the meaning behind the text in terms of the relevant context. Liao: How can we analyze and understand folk oral epic literature properly? Chao: The basis of the study of written literature is commonly an authoritative text edition. But oral literature is different, especially large works like epic. The living tradition is always changing, so there cannot be an authoritative version. The epic fieldwork of Parry and Lord in Yugoslavia in the 1930s made it clear that in the local Muslim sung epic tradition, one cannot find the same performance twice. A singer may say, “Every time I sing, it’s the song.” Scholars have come to realize that though we regard the epic songs of Homer that we have today as the Homeric epics, they are just relatively late and wellpolished versions from a long sung tradition of innumerable Homers. Liao: How do Western scholars trace the oral source behind the Homeric texts? Chao: Many epic traditions have disappeared, and a researcher can only work with the written texts. This is the only way to study Homeric poetry and Medieval DOI: 10.4324/9781003258001-11

92  Theories and methods of oral poetics European epic. This is why from the very beginning, the study of Western epic has focused on the text. Nevertheless, meticulous scholars have found that some features in earlier versions are quite different from those of contemporary text editions of the Homeric epics, including commonly repeated formulaic phrases like “fleet-footed Achilles” and “grey-eyed Athena.” The basic sentence patterns of Homer’s poetry are quite economy, but a few different combinations can change into hundreds of variants. Liao: In studying the living oral epics of ethnic peoples in China, what is the significance of confirming the oral tradition behind a text? Chao: Unlike the study of the Homeric poems, we don’t need to confirm whether our epic traditions are oral. That is obvious. Our job is to analyze the text along with the surrounding atmosphere, recognizing that it is the sung text and context that together create meaning. We need to attach great importance to the oral tradition behind the text and form a clear and conscious understanding of the oral characteristics of folk epic. At the same time, we need to correct methodological misreadings and clarify conceptual errors and ambiguities. Liao: What is lacking in our study of oral texts, singers, performance, and inheritance? Chao: The textual forms of oral epic are very diverse. In the past, we didn’t know enough about the differences of particular forms and so had some confusion in the study of the texts. The Mongolian epics that I am familiar with have been transcribed by literate singers, some handed down from unknown historical sources taken from performers hired by nobles, some made from onsite recordings with modern technical devices. Western scholars have found from fieldwork that in oral performance, narrative elements are abundant and details fully described. What is recorded by a modern device surely comes closest to the original presentation of a sung text, but we lack an accurate assessment of the potential impact on singers of using these devices. On the other hand, some scholars base their interpretations on the published texts, but then, other questions arise: What are the principles on which printed versions were produced, and how many facets of the content were taken into account in their publication? Some printed versions of Jangar have been published as literature for reading without consideration of faithfulness to the original sung performance. Others are compilations that have lost the true character of the oral text itself and are thus useless for research. Liao: What needs to be emphasized in the study of epic contexts? Chao: The Oral Formulaic Theory rightly focuses on textual analysis, but the more recent School of Performance Theory emphasizes the process of living expression and its significance. It recognizes that the meaning of folk narrative lies not just in the text itself but in the text as it appears in the local performance. In Story, Performance, and Event (1986), Richard Bauman distinguishes among narrative texts, narrated events, and narrative events. This latter refers to the situations in which the narratives are told, which are integral to the total research object of understanding them.

Oral poetics and Chinese epic research  93 Liao: What specific research approaches does Performance Theory emphasize that are helpful for us? Chao: We should not simply accept whatever approaches foreign scholars emphasize, but Performance Theory is based on a huge amount of fieldwork and provides excellent answers for basic problems, such as how to gather all the relevant information about the narrative process; what kinds of experiential frameworks are needed for sound fieldwork research; what can be found about the internal relationship between folk narrative and performance. To be specific, the theory emphasizes the identity and interaction among the singers of a performance; it pays attention to the performance’s potential for meaning; it evaluates the relationship among various factors in the performance; it traces the sequence of the various activities and behaviors that constitute the event; it analyzes the essence of the narrated event and the narrative event; and on the basis of all this, it then interprets the story. Epic research in China has not yet fully followed these guidelines. Of course, common sense tells us that epic singing is never a one-way transmission of events. As they participate in the process, the entertainer and audience share not only the inside knowledge of the epic register but also extremely complex modes of communication and interaction. In our research, we need to recognize the importance of this narrative event. Liao: How can we break free from focus on epic as simply a literary text? Chao: The main problem is methodology. Folklore consists of studying the knowledge of the people. For some works, we have descriptions of how the local people think of the epic and of the function it has in their minds, etc.; but such descriptions remain at this level. We have achieved little in the way of explanations of this kind of popular knowledge; and whatever explanations we possess have no empirical or theoretical framework. Mere detailed descriptions cannot provide comprehensive theoretical perspectives. Liao: Theoretical development is related to the level of fieldwork. What level has our fieldwork reached? Chao: Our work of folk epic collection has produced a wealth of recorded texts. But field investigation is not about taking a tape recorder, recording a singer’s singing, and coming home to sort it all out. However, the value of such an isolated text is relatively limited. We need to give more attention to the narrative events surrounding the text as it is sung; and we need the guidance of theory to do so. Without that, we have no disciplined controls and so cannot unearth the various problems in a live performance and achieve in-depth analysis. We need a theoretical framework and methods determined according to the characteristics of the specific research object. Liao: What success have we had in deepening theoretical thinking in accordance with our epic fieldwork? Chao: We can talk about this from two aspects. One is fieldwork methods, especially empirical investigations conducted to verify certain assumptions. The other is the post-processing means of dealing with the text. Regarding methods, what principles did Parry and Lord use for choosing their area of field

94  Theories and methods of oral poetics research? In the first place, they wanted to investigate oral traditions at a far remove from the heritage of Christian culture so that the results would be more convincing. They thus chose the Muslim epic tradition of Serbian and Croatian areas. It is better to find similarities between very different traditions than between those of cultural kinship. The research results are then more generally applicable. Secondly, investigations should come to a judicious understanding of the singer’s personal style and the singing tradition of the locality. Thirdly, the epic sung by a particular singer must be recorded by other singers as well into order to know different versions of the same story; and it should also be recorded by the same singer at different periods to verify whether he or she simply repeats it exactly or chants a new text every time. Using technical means for recording helps to assess differences between these versions and the reasons for the differences. Personal information about the singer needs also to be recorded in detail; and attention should be paid to how the audience has an active influence on the singer’s performance. Finally, empirical models are used to verify one’s inference of basic canons of epic creation. Liao: What models do we have for improving our fieldwork research principles and skills? Chao: Parry and Lord’s fieldwork can serve as a model. As professor at Harvard, Lord personally presided over the collation and publication of their fieldwork materials. The result is that since 1953, many volumes of Serbian and Croatian hero songs have been published. About 1,500 oral epic texts that they recorded from the Novi Pazar region have been collected in the Milman Parry Collection of Oral Literature of Harvard University with a catalog index. We can see their procedures in the first two volumes of the hero songs. The original sung texts are printed along with an introduction to their fieldwork experience, as well as conversations with some of the original singers and helpful abstracts and notes. Lord clearly explains the selection criteria and detailed principles for the compiling of the songs. To summarize, when examining an oral epic tradition, it is necessary to start with an investigation of the singing of individual singers and then broaden the investigation to other singers in the same area so that we understand the singer, not as an individual, but also in his or her relationship with the group of singers to which he or she belongs. Texts should be organized by region, the songs of each singer grouped together. This enables epic researchers engaged in various fields to have a unified, centralized reference system for investigating morphological variations in formulae, themes, and story-patterns of the oral tradition of a defined cultural and geographical area, song by song, singer by singer. Liao: What misreadings of the complex body of folk literature and art in China have you found in our research on oral epic? Chao: Sometimes, we have turned a blind eye to significant phenomena. For example, we have not seriously asked how mostly illiterate singers can remember epic patterns and fluently sing thousands of lines of poetry. Investigations have shown that bards do not memorize the text word by word but from the

Oral poetics and Chinese epic research  95 beginning, learn the rules of epic composition. They need to master the multitudinous structural units of formulae and larger segments like typical scenes and story-patterns. In the singer’s eyes, the smallest narrative unit is often not lexical words but the “long words,” sometimes a whole sentence, sometimes a formulaic paragraph. The term “formula” itself refers to a set expression, regularly used to express a particular basic idea under the same prosody conditions, that helps the singer improvise on the spot. He or she does not have to focus on piecing words together but can direct the flow of the song to fuse together traditional sequences. Liao: How can we distinguish written from oral epic in the heritage of China’s ethnic minorities? Chao: There have been controversial works, such as the 1716 Beijing Woodblock Geser. Some people regard this as a novel; and scholars who see it as a folk oral work have sometimes not offered a strong argument, even though the work manifests such characteristics of oral works as formulaic density, typical scene, etc. In the case of Jangar, for example, a feasting scene begins and ends almost every canto; and there are traditional formulae for expressing horses, weapons, and battles. Liao: How does epic differ from other forms of oral poetry? Chao: Epic is different from other oral folk genres in that it is a mirror of a nation’s spirit and ideals, of its historical life and cultural heritage. The styles of epic and folk songs are also quite different. As a long form of narrative art, the rhythmic progress of an epic is relatively slow. Its length demands, moreover, not only thrift in the use of phrases and formulae but also limited variations, which spring from the fact that the performer does not memorize each and every line. Liao: If oral epic has no authoritative text, where should scholars start? And how can they interpret the work correctly? Chao: The interpretation of the meaning of an epic text is based on the context of the activity as a whole, composed by an individual singer, but a creative action shared with previous generations of singers and audiences. The analysis of an isolated text can only yield limited meaning, but it is important because the general exists through its individual manifestation. The numerous particulars of a sung performance together constitute the oral tradition that makes it so fascinating. This chapter was originally published in National Arts, Vol. 1. 1999.

10 Types of oral epic texts A Mongol case study

Mongolian heroic epics are all oral works created and inherited by the folk collective; and in the course of their long development, they have gradually appeared in written records and transcripts. The Mongol studies scholar Walther Heissig has noted that to date, more than 350 kinds of Mongol heroic epics have been found and recorded in one way or another in China and beyond, of which one-third have been published. In China, most epic researchers are literary scholars familiar with the methods of analyzing written texts. As we have noted, however, study of the verbal text of a sung performance is only one aspect of epic research, and it has not always been pursued at a high level. Our Chinese folk epic traditions are those of the performing arts; and just as a playscript cannot equal a dramatic performance, so the study of an epic text is not the whole of epic research. The present author here follows Foley’s use of the term “text,” taking the definition of traditional linguistics and literary research as the benchmark but aligned with the meaning of oral poetic morphology. Foley notes that this notion of a text does not use the semiotician’s broad interpretation as an object of analysis. As applied to epic, it can have two meanings: either clearcut and written, or oral and created in the middle of performance (Foley 1995, 2000b). The Mongolian epic texts known to the present author can be classified into the following five types.

Retold, paraphrased texts The large-scale Mongolian epic Jangar first became known to the outside world through paraphrased retelling of the story text. The German traveler Benjamin Bergmann was an important source. In The Travels of Benjamin Bergmann among the Kalmyks in 1802–1803, published in 1804,1 he disclosed to the outside world for the first time, in German, the story of the hero Jangar and his warriors among the Mongol Kalmyks. It is speculated that he understood neither written nor spoken Kalmyk and had probably never seen a written text of Jangar, let alone had such a text in his possession. He very likely just heard folk artists singing the epic. Since he may not have been able to record the verses or even summaries of Jangar, what he brought to the world’s attention may have been translated into Russian for him by someone; but he does not mention the name of the singer or DOI: 10.4324/9781003258001-12

Types of oral epic texts  97 the translator (Burchinova 1983:250; Bormanshinov 1984:146–147). The exact source of the text thus remains a mystery. Another important recounting of an episode of Jangar is Hongor, the story of one of Jangar’s warriors having that name, written much later in the 1940s by Bian Yuan, a Han national who had heard it told by a Mongolian fellow prisoner, Man Jin, while in prison. Bian Yuan wrote: “I am deeply impressed by this story. In 1942, I began to write it according to my memory. In terms of plot structure, I kept the story as it was; I did not add or modify” (Bian Yuan 1958:7). Since the motivation for his retelling was not for academic research, its worth in terms of completeness and accuracy is greatly reduced. Nonetheless, such a text has information value even if it is not usable for poetic analysis.

Orally dictated texts We know little about the formation of orally based transcriptions; but before electronic recording devices appeared, travelers and scholars in most cases used written transcriptions of sung or orally dictated texts in collecting epic materials. In the middle of the 19th century, the Russian Mongolist Bobrovnikov obtained two “original” episodes of Jangar in the Kalmyk language, one from G. L. Mikhailov and the other from O. Kovalevskij. Most likely Mikhailov’s text was transcribed from an orally communicated text, not copied from a manuscript. Bobrovnikov translated it into Russian and published it in 1854 in St. Petersburg as Jangar: A Kalmyk Folktale. In 1857, this Russian version was translated into German; and Jangar then began to attract much attention in Europe. For the original transcription from the Kalmyk, it is said that Mikhailov “probably received the help of a native Kalmyk who was proficient in the language. This person would have been G. Sanjeev” (Bormanshinov 1983:235). He was a local intellectual teaching at a Kalmyk boarding school in Astrakhan Province and an enthusiastic collector of Kalmyk folklore. With his aid, later, in 1862, another Mongolist from the University of St. Petersburg, F. K. Golstunskij (1831–1899), took charge of the transcription of the oral transmission of Jangar and in 1864, published two episodes of the epic in Kalmyk (Golstunskij 1864). This was the first time that an original epic text of Jangar had been published. Neither Golstunskij nor his student Pozdneev (1851–1920) personally did the transcription, but Pozdneev later republished the two episodes along with one of his own, which was likely transcribed by Sanjeev in the 1860s. WI.L. Kotwicz (1872–1944), another favorite student of Golstunskij, received training from him in the 1890s and also became interested in epic. When the University of St. Petersburg recruited a group of Kalmyk students in 1905, Kotwicz learned through them that there was a folk singer named Eela Ovlaa (1857–1920) who was famous for singing Jangar in Dörbet, Astrakhan Province. On August 28, 1908, one of Kotwicz’ Kalmyk students, Nomtu.O. Ochirov, met Ovlaa and made a phonograph recording of his singing of a canto about the handsome Mingyan. In mid-December, at the appointment of the Oriental Circle of the Imperial Russian Geographical Society and the University of St. Petersburg, he

98  Theories and methods of oral poetics visited Ovlaa again; and on December 18–19, he enthusiastically and efficiently transcribed nine cantos in Cyrillic. WI.L. Kotwicz carefully proofread his phonetic transcription and stated that it was accurate beyond reproach. He made only minor changes and marked the titles of ten songs (Bormanshinov 1984:157–159). From Golstunskij to Pozdneev to Kotwicz, no other transcriptions have been left to us, which shows that transcribing from an oral source is extremely difficult. Part of the reason is that until the middle of the 19th century, there was a lack of Kalmyk-speaking scholars. Golstunskij tried to have more poems transcribed while being read at a slow pace; but except for the two episodes mentioned here, his efforts were unsuccessful. The reason is not that performers could not be found. It is said that he actually met many folk singers on the grasslands, but they were not comfortable with their singing being interrupted for a time to record the words as they sang them (Bormanshinov 1984:150). Judging from the clarity and purity of the transcription of Ovlaa’s Jangar as we have it today, it must have been an orally dictated text that was carefully arranged by Ochirov, not a transcription made as Ovlaa was actually singing; for there are no obvious mistakes in the text, which is often unavoidable during the process of transcribing live singing. As can readily be seen, an orally transmitted text is often poorly grounded. It needs to be transcribed directly as heard from the singer’s lips, but this is very problematic. Speaking of a collection of Tibetan epics, R. A. Stein once argued that they are only half oral, because for the purpose of transcription, the Tibetan bards were asked to chant slowly. This inevitably led to hesitations in their singing (Stein 1993:73). The same is true of Parry and Lord’s fieldwork in Yugoslavia in the 1930s; but in their view, slowly recorded “oral notebooks” are rich in detail. For the singer, there is an advantage in recording from dictation rather than from singing. Due to the demands of story-patterns or rhyme, a singer may mistakenly plant words in a particular slot in a sentence or sentence combination, which is worth studying; but the slow pace of narration tends to lead to a reduction of mistakes. It gives time for the singer to stop and ponder each verse, to make the best of his or her art as performer before a small group of hearers, and to furbish the text properly as poet and storyteller. From the point of view of the researcher, however, slowing down the transmission of a text is thus not just a matter of speed; it leads to editorial alterations. Some scholars assert that the Homeric epics are based on an “orally dictated text” rather than live singing (Foley 2000b). Granted their length and narrative richness, they must have been written down at the slow pace of oral dictation. In general, orally transcribed texts have historically played an important role in literary history and made possible the preservation of important epic texts. Nevertheless, they are not ideal for studying issues related to the oral characteristics of a performance.

Handwritten texts Investigations in China have shown that the Mongol Oirats have the custom of keeping handwritten texts. They often hire literate persons to copy the lyrics of

Types of oral epic texts  99 Jangar; and literate jangarch (Jangar singers) themselves may also record the lyrics in order to preserve them or pass them on to others. This is also a form of transcription; but the written texts retained, inherited, and used by such locals differ in purpose from those recorded for scholarly research and are here simply called “handwritten texts” in distinction from the “orally dictated texts” discussed earlier. Of course, this distinction is relative. Historically speaking, no matter what the motive or persons who supervised our transcripts of Jangar, most have been written by literate locals. The Russian anthropologist Nebolsin (1817–1893) believed that both Kovalevskij and Bobrovnikov once possessed an original text of Jangar; but we have no accurate account of the type of the transcript. In the summer of 1854, a Russian junior civil servant named Kornilov told a traveler that he had received a handwritten text of Jangar as a gift from a native of Zaisang named Arash Chiluun in Bag Chohor Törbed of the Soviet Union;2 but the report provides no details about this, and no further information has been found. There is information as well about a handwritten text of Jangar in Mongolia, but this also does not amount to a significant report or major discovery. In China, there are many stories and traces of handwritten copies of Jangar. It is said that seven such texts have been collected in Xinjiang in recent years (Jamcha 1996), the number perhaps being as many as ten. Some of these are the same in language and content as episodes in the Kalmyk version, and some are similar to the those sung by local jangarch. The scholar of Mongolian epic Rinchindorji once classified such texts into three types: 1) traditional handwritten texts left from predecessors; 2) orally dictated texts circulated among the people; 3) texts re-transcribed from a Russian translation (Rinchindorji 1999:10–11). The present author has learned that there are more than ten different kinds of handwritten texts in depositories of ancient books, such as the libraries of the Xinjiang Folk Artists Association and the Inner Mongolia Academy of Social Sciences. Those in the latter depository were collected in the mid-1950s in Xinjiang by the Inner Mongolian scholar Mergen Batur. Experts hold that some of them are ancient, some having been written shortly after the Buddhist monk Zaya Bandita created the Todo Oirat script in 1648. In other words, Jangar has been written down and edited at least since the middle of the 17th century; and this tradition continued into the 1950s (Jangar Manuscript 1996:5). Rinchindorji found a handwritten copy of Hara Hunas from Xinjiang in the library of the Inner Mongolia Academy of Social Sciences; and in May 1966, the present author compared it with the Legend of Jangar published in Hohhot. The latter is exactly the same as Hara Hunas except for a few incorrect words and a short paragraph added at the end, which is exactly the same as the opening stanza of Shara manggus (lines 1–12, p. 311). The dates of these two handwritten texts have not been determined. They may be based on what Golstunskji published in 1864 or copied later from the transcript he used (Rinchindorji 1988:67). Based on a survey by J. Batnasan, in the past, many families in Mongol settlements in Xinjiang, especially aristocratic families, collected handwritten copies of Jangar. Some have been handed down from private sources (Jamcha 1988:26).

100  Theories and methods of oral poetics There are two kinds; those formally commissioned, such as texts transcribed by scribes hired by the royal family; and those written by literate jangarch themselves or transcribed at their behest. For the latter, their written culture can have an influence on their conceptual world and verbal manipulation. They sometimes modify printed texts as new cantos, as is the case with texts found in Xinjiang made from audio recordings.

Live recorded text As Stein has pointed out, Bards generally cannot repeat their lyrics slowly. They become confused when they do so, losing their cues and their ability to improvise. To obtain a true oral version, that is, a pristine text, the language should be recorded directly by a recorder, like that of a rap artist performing rap in a free environment. (Stein 1993:130) Although Stein is speaking of the Tibetan epic Gesar, the argument is applicable to the recording of any living oral epic. Recording with a recorder is the most convenient and reliable method we can use at present, but the singing and recording affect each other. In the mutual interaction between the investigator and the singer, mutual influence necessarily occurs. A singer who is unfamiliar with the recording device may feel nervous, and this may directly affect the mode of singing. To give an extreme example, during the performance of Jangar in Usu County, Xinjiang, the jangarch Arimpil became extremely frightened when a foreign scholar suddenly put a black microphone in front of him, causing him to faint. He said that something went wrong with his heart at the time (Dawataya 1999:458). The earliest electronic recording of Jangar we know of is a phonograph recording of ten Jangar poems by Eela Ovlaa in August 1908. This audio recording is not only the earliest but also one of the most important of our recorded texts. The extensive use of tape recorders is a relatively recent development, but in the past thirty years, the epic collection in Xinjiang has been greatly augmented with their aid. This has left us with a large number of live performances. However, storage is a problem; and although recording itself is easy, its use is limited. The poetic register and dialect used can make even scholars proficient in the language flustered when facing difficult problems. Moreover, conversion from tape to text is much more complicated than people think. As far as China’s practices are concerned, the most problematic is the arbitrary addition, deletion, and combination of texts.

Printed texts There is not much to say about printed texts. They are a means for copying and transmitting texts in batches. Among the texts of Jangar published in China, there are reference texts edited for scientific research purposes as well as textbooks

Types of oral epic texts  101 for reading and study that are unsuitable for academic research. There is a complex relationship between a printed text and an oral folk text. There are reports of the impact of the Beijing Woodblock Geser on the folk singing of Jangar (Vladimirtsov 1983–1984) and also reports of printed texts being turned into a manuscript and circulated among local people. In the 1940s, a man named Bold of the Hure Banner of Inner Mongolia went to Tashkent for study and returned to China with twelve cantos of Jangar published in the former Soviet Union, which have been circulating in some regions (Rinchindorji 1999:11). In the past, Chinese scholars did not pay enough attention to distinguishing between different text types because of our insufficient understanding of the unique characteristics of oral literature. As a result, some ambiguities have surfaced in basic text research. Along with a theoretical understanding of text types, oral tradition text research demands sound empirical observation in fieldwork and a grasp of scientific principles in subsequent text collation. Otherwise, it is not only difficult to understand the significance of particular text types but also easy to run into epistemological and methodological problems when selecting and defining research objects. The idea of this chapter was originally published in Studies of Ethnic Literature, Vol. 4. 2000.

Notes 1 Benjamin Bergmann’s nomadische streifereien unter den Kalmüken in den jahren 1802 und 1803, Riga, C.J. Hartmann, 1804–1805. Vol. 4. 2 This report can be found in Kornilov’s 1854 Selected Manuscripts of AstrakhanKyrgyz Kalmyk Grassland, Bulletin of the Russian Geographical Society, Vol. 12-3.

11 “Returning to the voice” Textual research of oral epics as a starting point

In the Western academic tradition, the study of poetics commenced with Aristotle and was from the very beginning pursued in terms of both narrative art (Homeric poetry) and performance art (drama). Later, however, studies on poetics focused almost solely on the canons of written literature, with some fortunate exceptions like Lessing’s Laocoon: An Essay on the Limits of Painting and Poetry (1766). In the mid-20th century, Lord published Poetics of Oral Creation (1959) and a decade later reiterated the basic concepts of oral poetics with regard to the study of Homeric poetry: Surely one of the vital questions now facing Homer research is how to understand oral poetics, how to read oral traditional poetry. Its poetics is different from written literature because its technique of composition is different. It cannot be treated as a flat surface. All the elements in traditional poetry have depth, and our task is to plumb their sometimes hidden recesses, for there will meaning be found. We must be willing to use the new tools for investigation of multiple forms themes and patterns, and we must be willing to learn from the other oral poetries. Otherwise, “oral” is only an empty label and the “traditional” is devoid of sence. Together they form merely a facade behind which scholarship can continue to apply the poetics of written literature (Lord 1968:46). The term “oral poetics” has now come to be commonly used by scholars; but even at the beginning of the present century, there is no entry for “oral poetics” in important reference books and professional sources like The Oxford Dictionary of Literary Terms (2004) or the fourth edition of The New Princeton Encyclopedia of Poetry and Poetics (1993). There is no such entry, either, in the 1993 Chinese Dictionary of World Poetry; and in writings on the history of Chinese literature, there is commonly no special discussion of oral traditions. This chapter attempts to offer a systematic summary and definition of oral poetics in terms of oral epic, focusing on text and voice and discussing the tools and guidelines of oral epic research. Compared with other literary genres, epic has a longer history and boasts rich scholarly achievements and deep theoretical reflection. DOI: 10.4324/9781003258001-13

“Returning to the voice”  103

Oral Formulaic Theory and oral poetics As I understand it, the systematic construction of oral poetics began in the 1960s. The American scholar Rosemary L. Zumwalt has noted that during the great theoretical period of the 18th and 19th centuries, scholars like Herder gave important summary accounts of the existence and meaning of oral tradition (Zumwalt 2000). However, these discussions can only be regarded as the prehistory of oral poetics. As we have seen, the mid-20th century was the critical period for the formation of the concept of oral poetics. Lord’s The Singer of Tales (1960) marked the appearance of the Oral Formulaic Theory; and there were heated debates in western Europe and North America about the role of written and oral culture in promoting the progress of human civilization. Several eminent scholars from various fields participated in the debates: the communications expert Marshall McLuhan, the structural anthropologist Lévi-Strauss, the social anthropologist Jack Goody, and the classical scholar Eric Havelock. The London Seminar on Epic from 1962 to 1974 and the two-volume publication of its achievements (Traditions of Heroic and Epic Poetry 1980, 1989) reflect the historical process of the transformation of the epic research paradigm from literature to oral poetics. In 1970, the Ethnopoetics School emerged in North America; and its influential journal Alcheringa: Ethnopoetics was launched. The American anthropologist Dennis Tedlock gave a representative statement of the School’s key approach when he stated: “Oral poetry begins with voice, while oral poetics returns to voice” (Tedlock 1977:57). Dell Hymes’ notion of “The Ethnography of Speaking” is closely related to Ethnopoetics as well as the influential work of several scholars who did not actually join the Schools. Such was the work on African oral literature of Ruth Finnegan of the British Open University and the cultural and philosophical discussions of Walter J. Ong of St. Louis University. Of course, in the 1980s, John Miles Foley became the new leader of oral tradition research, standing at the front line of integrated research on oral poetics and gathering a team of researchers. Regarding oral poetics, two points need to be kept in mind: Firstly, the problem remains that the rules of composition, distribution, and reception of oral poetry and, in fact, of the entire oral tradition need clear formulation; and they must be different from the ideas, tools, systems, and methods of written literary theories. Secondly, oral poetics is an important wing of poetics as a whole, not an independent category. Theories of poetics have long neglected it, just as literary studies have long neglected oral folk literature. In recent years, oral poetics has become the main direction for a batch of dissertations and research topics in China. Most, however, use theories associated with oral poetics to analyze specific texts or traditions; few focus on the systematic construction of theory.

Oral and written poetic texts There are notable differences between written and oral poetics in understanding and analyzing a narrative text. For written literature, text is understood as the

104  Theories and methods of oral poetics fabric of language and is always in the process of being woven (Dong Xiwen 2006:1). Some scholars think that the textual analysis of a written work of literature needs to be carried out on four levels: (1) analysis and description of linguistic structure; (2) observation of the structure in terms of its essential make-up; (3) analysis of the intertextual relationship between texts; (4) identification of various forms of citation—quotation, pasting, allusion, and parody—and their relationship to intertextuality. Analysis at these four levels can reveal a text’s cultural value, though historical and ideological factors also demand consideration. Keeping these roughly in mind, let us compare differences between oral and written poetic texts from several perspectives. As we all well know, in the criticism of written works, it is common to indicate which version of the text is used, usually the standard critical edition. As we have seen, however, the situation is quite different with oral poetry, where there is no standard, authoritative text. Each time a performer presents a given story, it is his or her one and only text, an on-the-spot composition both related to and different from the same story as it has previously been told and may be told many times in the future. As Lord has pointed out, a singer’s maturity is often measured by the richness of his or her repertoire and of the various traditional structural units (formulae, typical scenes, motifs, etc.) he/she has mastered (Lord 2004). When a scholar begins to study an oral text, it is thus necessary to explain and define the formation of the text, who the speaker is, in what context (time, location, audience, etc.) it is delivered, how the present version has been produced (live transcript, audio or video recording), and who has been involved in the text production (interviewers, collaborators, etc.). As was noted in Chapter 1, the whole process of an oral epic’s creation and reception can be summed up in what Bamo Qubumo has called the “five presences”: epic tradition, performance events, performers, audiences, and researchers (Liao Mingjun and Bamo Qubumo 2004). If first-hand data are not available in the case of a historically handed-down text, there is a need to carefully identify and explain the precise kind of version being used—handed-down transcript, engraved or printed text, manuscript, version quoted in a reference work, shorthand copybook, abbreviated copy, illustrated text, etc. The printed text that a reader reads is a string of symbols that are generally fixed and depending on the reading environment and readership, do not usually change and can be re-read ad infinitum. Quite the contrary, in oral poetics, the text is a series of sound symbols propagated linearly in the air; and the told story is a one-time event that cannot be repeated in exactly the same way. With the end of the performance, the text vanishes into thin air. A written text is tangible, conveyed through written symbols, while the oral text is intangible, transmitted orally through sound waves. Today, we can document performance activities in video and audio recordings or transcribe the narrated text with written symbols; but in essence, an oral text is a unidirectional, irreversible vocal act. In a written work of literature, the writer’s personal linguistic style is the mark of his artistic attainments, with a unique aura that others seek to imitate. In an oral performance, the singer’s style is related to a specific inherited tradition, region, dialect, and

“Returning to the voice”  105 storytelling school. It is difficult to say that a folk singer has a truly distinctive personal linguistic style. The structure of a written literary work reflects both the ingenuity of the author and the aesthetic and psychological ideas of his or her literary traditions—the classical unities of drama, the polyphonic structure of a novel, the starting from the middle of the epic plot, and the principles of folk epic, such as have been summarized by the Danish folklorist Alex Olrik (Olrik 1992). Generally speaking, the creative directions of a particular writer are more difficult to predict than those of an oral storyteller because they tend to avoid anything like the oral formulaic patterns. In marked contrast, oral poets rely highly on a stylized structural model. Many ethnic groups have structurally very simple epic types, such as those that the Mongolian scholar Rinchindorji found to dominate early Mongolian epic: “the hero’s quest for a wife” and “the hero’s struggles against a demonic figure” (Rinchindorji 2001:383). As regards character portrayal, epic fully embodies the antagonistic style that is said to be common in oral traditions (Ong 2000). In Epic Laws of Folk Narrative, Olrik has pointed to the law of contrast as one of the important rules of epic poetry (Olrik 1999:83–98); and in Poetics of Mongolian Heroic Epic, the Mongol scholar B. Bürinbeki has referred to the strong common contrast between hero and demon as a “black-and-white image system” (B.·Bürinbeki 1997). Few novelists, moreover, give the reader the direction and end of the whole story from the very beginning; but this is common in an epic performance. Foley has called the motif at the outset of a Mongolian epic that commonly predicts the direction of the whole story its “road map.” A messenger or a hero’s dream, for example, indicates that the action will focus on war. In short, it is common for oral epic to have a set schema for the plot development; and singers perform according to the canons of specific narrative types or subtypes. Creative storytelling skills can cross the boundaries between written and oral literature, influence each other, and learn from each other. There are distinctive oral narrative features, for example, in the famous Chinese classical literary works Sanguo Yanyi (Romance of the Three Kingdoms) and the 14th-century novel Shuihu Zhuan (Outlaws of the Marsh, also known as Water Margin). Nonetheless, in the main aspects of composition, diffusion, and receptance, the differences between written and oral literature are great. According to the Oral Formulaic Theory, oral poetry is not just poetry chanted orally; it is poetry that is actually created in the course of an oral presentation. Its composition, diffusion, and receptance are carried out and completed in the time and space of the performance. Very differently, the creation, circulation, and reception of a written work of literature can be separated by a huge span of time and space. A modern reader’s bookshelf can contain the ancient Encountering Sorrows of Qu Yuan (c. 340–278 BCE), Whitman’s Leaves of Grass, and the recent Holy Snow Mountains of the Yi poet Jidi Maga about three snow-capped mountains in Sichuan. The author’s creative activity and the reader’s reading activity are carried out in different spacetime dimensions; and the reader’s response does not directly affect the finished work. This is not the case with oral storytelling, where listeners’ responses, verbal

106  Theories and methods of oral poetics interaction, shouting, and even the audience’s very composition affect the process and content of the performer’s creativity. As regards the integrity of a literary text, once a writer’s work is finalized, its basic meaning is complete. Reader-response criticism stresses that readers naturally have different understandings of a work in line with their different ages, levels of education, basic knowledge, and life experience; but they do not participate in the author’s process of writing the work or help direct its intrinsic thrust of meaning. Quite differently, in an oral folk performance, the process of generating and transmitting meaning is one in which the audience participates with the performer through its responses in shaping a work and completing its meaning. Moreover, every act of storytelling by a folk singer is a new creation. Though a literary author’s work has an end, that of a folk singer is open-ended. When told to audiences from different backgrounds, environments, or work fields, the same story will often differ greatly and give rise to different texts. The role of the performer is, within limits, to shape its mutations. Written literature appeals to the eyes and oral literature to the ears, with voice as the carrier. It is true that a written work can be recited and an oral work transcribed; but in essence, oral literature is listened to by an audience and written literature read by readers. In a society with a developed written culture, some narratives that originally had oral origins were eventually written down; and after being compiled, rewritten, and polished, they may even become literary classics. Then, too, some civilizations have entered various stages of oral and written transmission, sometimes in parallel and often with a great deal of interpenetration between the two media. Nevertheless, in a reading society, vocal poetry has gradually faded or weakened; its effects have less appeal, and its aesthetic norms have become of less concern to people. Using only the eyes in silent reading, the brain will not usually arouse the sounds of specific words; but a reader may enjoy reading certain passages again and again. In oral cultures, though the audience participates in the creation of a work, the actual state as listeners is much more passive. As the brain follows the voice signals of the narrator step by step, it cannot fast forward or skip uninteresting, excessively elaborate, or redundant passages; and listeners cannot rewind a live performance to dwell on deeply expressive verses. People tend to draw a clear dividing line between literary authors and folk artists. On one side are writers, the “engineers of the human soul,” elites in society who have long been widely praised and admired. Many have a distinct literary persona and can be recognized by their artistic originality. Quite differently, the performers of literary folk art have their grassroots in the ordinary folk and are often recognized as the representatives of a particular people. They are not particularly respected for being good at their art; and though they are often masters of vivid folk language, few appreciate their original ability. Quite the opposite, many criticize them if they deviate from the traditional rules. For literary authors, originality is their lifeblood; for folk performers, what is important is conforming to traditional artistic canons. Different though writers and folk artists may be, just as there are various paths by which one can become a literary writer, it takes long

“Returning to the voice”  107 effort to become a well-noted folk artist. Literary authors are classified as novelists, essayists, poets, dramatists, etc.; and folk artists can be divided into eulogists, epic singers, and storytellers. Some writers write in different genres; and some folk bards can perform in different modes, famous epic singers commonly being adept at both eulogy and storytelling. When a writer writes, there is a large amount of material fermenting in his mind; when a folk singer creates a story, he or she must not only have a vast horde of stories in his/her mental arsenal but also be quick to manipulate the storyline freely in line with the needs and likings of the on-site audience.

The poetics of oral texts An important mark of oral texts that demands study in terms of oral poetics is their inter-correlation. “In variable ways, a song in a tradition is independent, but it cannot be separated from other songs” (Lord 1960:123). This is clear in oral epic, where no poetic sequence is really original and where a particular poetic sequence has a relationship with other text sequences not only in the epic itself but also within the whole prior literary tradition. The meaning of such an oral epic sequence is thus always beyond the scope of the sequence itself. In a series of epic episodes, as in the Kyrkyz epic Manas, a poetic canto can be regarded as a relatively independent text, but it is also an integral part of a larger textual web that extends beyond the weave of the epic itself. The text of the canto is constantly changing and shifting in the process of creation, performance, and diffusion; and as it does so, its relationship with other texts forms a series of multiple continuations and differences. Without its role in the series, an oral text cannot survive. Texts usually exist in a synchronic symbiotic relationship; and as they change in the course of multiple repetition, there arise contradictions in detail. This is obviously different from the relationship of a series of chapters in a written novel. An experienced audience understands any particular oral narrative text in the context of the network of meaning composed of numerous other texts; and that meaning takes different shapes in the repeated performance of the story. In sum, the mode of existence and circulation of oral texts is not independent but depends on their specific inter-relationships. As the Parry–Lord Theory stresses, an oral text is also characterized by stylized formulaic expressions and scenes. Parry found that Homer’s poetic syntax is nothing if not a matter of formulae. He selected expressions in the first twenty-five lines of Iliad which are solidly underlined as being found unchanged elsewhere in Homer count up to 29, those in the passage from the Odyssey to 34. More than one out of every four of these is found again in eight or more places, whereas in all Euripides there was only one which went so far as to appear seven times. This highlights the distinction between oral and written literature (Lord 1986:478). The formulae density of Beowulf led Lord to conclude that it, like the Homeric epics, was originally an oral work. Though Claes Schaar and Kemp Malone have denied it, Jean-Marie Maguin has also maintained that Beowulf was originally an

108  Theories and methods of oral poetics impromptu minstrel’s song; and after analyzing the stylized formulae of the entire text, Maguin’s student Robert P. Creed found that the text is inextricably and without exception related to oral tradition (Foley 2000b:162–167). As regards the concept of text itself, folk literature and folk art have their own understanding based on specific paradigms of the discipline. In The Folklore Text: From Performance to Print, Elizabeth C. Fine traces the discussion of textual issues in the history of American folklore and the origins and development of folklore text theory. She finds four stages in that development: (1) the textual paradigms of ethnolinguistics; (2) literary paradigms; (3) various textual definitions by the forerunners of performance theory, including comparative literature approaches of the Prague School, narrative ethnography, and the theories of Parry and Lord; (4) performance-centered text experiments (Lord 2004:178). With regard specifically to oral epic texts, scholars like Foley and Honko have made various theoretical explorations of the categories and definitions of text types. In terms of epic characteristics and their contexts of composition and dissemination, Foley and Honko have tabled epic texts in the three categories or composition, performance, and reception noted in Chapter 1 (see Table 11.1). To grasp the diversity and meaning of oral poetry, we also need to explore its course as oral and written poetry (Chao Gejin, Yin Hubin, and Bamo Qubumo 2001). In How to Read an Oral Poem, Foley summarizes that course in Table 11.2 (Foley 2002:50). In recent years, with advances in digital technology, many singers in local communities have begun spontaneously to record their epic performances and audiences. These recordings range from those of early cassette tapes to current miniature cameras. Some of these are for self-appreciation, some for posterity, some to take the place of night-long chanting sessions, some just for money. How Table 11.1 Text types of oral poems From composition Composition Performance Reception to reception

Example

Text type Oral text Oral-derived text Tradition-oriented text

Oral Oral/written Written

Oral Oral/written Written

Aural King Gesar Oral/Written Homeric poetry Visual Kalevala

Table 11.2 Classification of oral poems Medium

Composition Performance Reception

Example

Oral performance Voiced texts Voices from the past Written oral poems

Oral Written Oral/written Written

Tibetan paper singer Slam poetry Homer’s Odyssey Bishop Niegoš

Oral Oral Oral/written Written

Aural Aural Oral/written Written

“Returning to the voice”  109 to treat such electronic texts has become another dimension that needs to be considered by the academic community, in particular how much this self-recording has been influenced by the media, journalists, and scholars using digital technology and has thus become a fashion among the public. In recent years in China, moreover, the Delvin clan in the Golog Prefecture of Qinghai Province has begun formally or informally to transcribe oral performances at singers’ requests and to store for posterity texts written by singers themselves. There are also epic narratives and narrative fragments recorded in traditional ways, such as the Dongba pictographic script of the Naxi people, the hand-painted pictures and woodblock prints of deities of the Yi, Tibetan Gesar carvings and thangkas, the epic motifs from Song of the Butterfly or Song of the Sweetgum embroidered on Miao costumes, and the ancestral pictures of the She people. All these linkings of poetry and painting deserve to be included within the scope of academic research. Chinese scholars have begun to participate in the analytical thinking about the poetics of oral texts, providing fresh views on some dimensions based on the rich texts and empirical field data collected in China. For example, Bamo Qubumo’s classification and definition of “male” and “female” texts and Black Yi and White Yi in the Hnewo narrative tradition of the Yi people provides a multi-dimensional interpretation of oral text in the social context (Bamo Qubumo 2004b). Similarly, Gao Hehong’s analysis of the receiver of the inheritance of Manchu traditional storytelling as “writing inheritor” (Gao Hehong 2011) and Wu Gang’s research on Daur uchin (wuchin) folk art (Wu Gang 2011) are also valuable attempts to parse and summarize the differences in text types between oral and writing traditions. The present author has discussed the causes and laws governing the phenomenon of objectification of oral texts.1

The poetics of mental texts As a theoretical discussion of the generation of oral texts, Honko’s Textualizing the Siri Epic (1998) gives a masterly explanation of the concept of text from the perspective of oral poetics. To show how an oral text is born and stored in a singer’s mind, Honko discusses the concept of mental text. As he explains it, the mental text is a pre-text that is in existence before a singer performs an epic. It is composed mainly of four elements: (1) plot; (2) structural textual elements, such as formulae, typical scenes, and motifs; (3) the poetic laws by which the singer converts the mental text into a specific performance event; and (4) the contextual framework, such as the memory of past performing experiences (Honko 1998:94). These elements are not independent units in the mind but are associated with one another and combined according to certain norms for adaptation to the needs of each performance in repeated use. Through listening, learning, memorization, imitation, storage, and repeated creative use, a singer gradually constructs his or her mental text. It is the life-spring of every performance and far larger than any specific narrative. A singer’s entire performance lifetime may not be enough to drain all its resources. Each singer, moreover, has his or her own distinctive

110  Theories and methods of oral poetics mental text; but it has a traditional thrust. Singers learn from one another, sharing and inheriting the main elements of the mental text, such as specific formulae, typical scenes, story-patterns, etc. To some extent, the notion of mental text can explain why singers have different variants when they perform the same story on different occasions. According to the information obtained from a large collection of field investigations, we can roughly judge that in the singer’s brain, the material of a narrative is not stored in different categories as concisely as in the drawers of a Chinese medicine shop but in a more random manner, perhaps with the singer’s voice pattern playing a role in the indexing and guiding of subsequent improvisations. The mental text has strong absorptive and formative properties. We have seen that a field experiment in Yugoslavia demonstrated that an experienced singer can retell a new story immediately after he has just heard it and expand it with more detail than in the original narrative (Lord 2004:111). On the other hand, in different narrative traditions, singers often integrate eulogies, ballads, proverbs, myths, and other modes of folk creativity into the storyline when performing a large work, which means that the mental text itself is often a hypertext and hyperlinked. Using the concept of mental text, Honko explicated the process of compilation of Lönnrot’s Kalevala as follows: As Lönnrot collected numerous Finnish oral folk poems, he gradually formed the text of Kalevala in his mind. The written version then became the embodiment of that text that serves as a gobetween between folk song and literary poetry. In his mind and at his hand, the oral poetic material Lönnrot obtained became much more than a collection of historical poems. Facing more complex conditions than the original folk bards, he compiled and organized a large-scale poetic work fashioned according to the canons of folk narrative. The different versions of Kalevala that he constructed are like the various narrative versions of different folk singers with their varying lengths. Based on empirical observations of the Tulu singer Gupala Naika’s performing activities in southwestern India, Honko deduced the working pattern of that singer’s mental text. It took Naika fifteen hours to sing the 7,000 verses of Kooti Cennaya; but he did so on Indian radio in just twenty minutes. When Honko asked him to chant it for a third time, again on radio, it took twenty-seven minutes. Naika himself believed that he could recount the epic completely in each of those three times because its skeleton and veins are all embodied in each version (Honko 1998:30). Obviously, the basic bone structure of the story is generally fixed in the singer’s mind as the bearer of the core narrative; the rest is variable. This is similar to what one finds among the Miao people’s ancient songs in southern China. The core of a song, the gegu, is its stable backbone, while variable gehua flesh out each singer’s improvisational performance. In all of this, the particular performance of his or her story is limited while the mental text is infinite. Of course, some epic traditions emphasize the sacred source and unpredictability of a text in a way that provides a counter-balance to Honko’s theory. In the Tibetan epic tradition, the cognition of bards engaged in dream-taught storytelling,

“Returning to the voice”  111 treasure storytelling, or prasenā divination storytelling forms a text that does not accord with his concept of mental text. As is shown in the phrase “dream-taught storytelling bard,” the epic text is believed to be infused in the singer’s mind from a sacred source and then objectified in performance. In prasenā storytelling, moreover, the bard needs special props as the bearers of the story’s contents to reinforce memory.

“Poetics of voice” From Honko’s discussion of the intangibility of mental text to the tangible objectification of an oral text, we have signs of major progress in oral poetics. Oral texts, themselves, are very much alive, and their core form is voice. Once voice is textualized, the objectified text is just the solidification of the mental and voice texts in one way or another. However, it is precisely this embodiment of oral communication that expands the scope of the voice text on another level, setting it beyond time and space, where it can be permanently preserved. Babylon’s Gilgamesh is an excellent example of this phenomenon, as are Homer’s epics and the myriad other epics. The European scholars Paul Zumthor and Marilyn C. Engelhardt once proposed that we lack a term of universal reference for this, but that it may be called “poetics of the voice” (Zumthor and Engelhardt 1989:73). In the first chapter of Oral Poetry: An Introduction, Zumthor gives an in-depth explanation of the presence of voice in the process of oral re-creation (Zumthor 1990). With the rapid expansion of our writing civilization, the number of societies in which oral traditions operate as the main channel of information dissemination and wellsprings of oral poetics and poetics of voice seems to be gradually shrinking. Nevertheless, as Foley has emphasized, orality is both an ancient and often a new way of disseminating wisdom. It is actually gaining fresh vitality in the era of new technologies with the links among daily life, thinking, and cyberspace. This chapter was originally published in Northwestern Journal of Ethnology. Vol. 4. 2014.

Note 1 Chao Gejin: “Oral Epic Traditions in China,” Online lectures: www. oraltradition​.org​/ articles​/web​cast

12 “How long is long” Epic length

When informed of the establishment of the London Seminar on Epic, the British sinologist Arthur Waley once commented: “Whenever I think of epics, I think how different they are.” Though this may seem an off-the-cuff statement, most scholars engaged in epic research agree that epic is a genre with huge internal differences and is not easy to define. The influential The Oxford Dictionary of Literary Terms has this to say about the nature of epic: Epic is a long narrative poem celebrating the great deeds of one or more legendary heroes,in a grand ceremonious style. The hero,usually protected by or even are descended from gods performs super-human exploists in battle or in marvellous voyages,often saving or founding a nation—as in Virgil’s Aeneid (30–20 BCE)—or the human race itself, in Milton’s Paradise Lost (1667). Virgil and Milton wrote what are called “secondary” or “literary” epics in imitations of the earlier ‘primary’ or traditional epics of Homer.Whose Iliad and Odyssey (c.8th century BCE) are derived from an oral tradition of recitation. They adopted many of the convention of Homer’s work, including the invocation of a muse, the use of epithets, the listing of heroes and combatants, and the beginning in medias res (for other epic conventions, see epic simile, formulaic, machinery). The Anglo-Saxon poem Beowulf (8th century CE), a primary epic, as is the oldest surviving epic poem,the Babylonian Gilgamesh (c.3000 BCE). In the Renaissance, epic poetry (also known as “heroic poetry”) was regarded as the highest form of literature, and was attempted in Italian by Tassothe Gerusalemme Liberata (1575), and in Portuguese by Camoëns in Os Lusíadas (1572). Other important national epics are the Indian Mahābhārata of (3rd or 4th century CE) and German Nibelungenlied (c.1200 CE). The action of epics takes place on a grand scale.and in these sense the term has sometimes been extended to long romancrs, to ambitious historical novels like Tolstoy’s War and Peace (1863–1869), and to some large-scale film productions on heroic or historical subjects.For a fuller account,consult Paul Merchant. The Epic (1971). (2004:81–82). In The New Princeton Encyclopedia of Poetry and Poetics (1993), a more professional reference book, the entry of epic takes up thirteen pages and is divided DOI: 10.4324/9781003258001-14

“How long is long”  113 into two parts: history, reviewing the epic phenomenon from ancient times to the present, and theory, introducing the main theoretical achievements of three stages: classical and Alexandrian Greece, the classic Latin and Medieval periods, and the Renaissance to the Modern. The core definition given in the entry is as follows: An epic is a long narrative poem that treats a single heroic figure or a group of such figures and concerns an historical event, such as a war or conquest, or an heroic quest or some other significant mythic or legendary achievement. In the fourth (2012) edition, the core definition was changed to: An epic is a long narrative poem about the heroic action:“narrative.”in that it tells a story, “poem,” in that it is written in verse rather than prose. “Heroic action” while reinterpreted by each major epic poet, in that, brordly defined, it recounts deeds of great valor that bear consequence for the community to which the hero belongs. These definitions state that epic is a long narrative, but as Lauri Honko has rightly asked, “How Long Is Long” (Honko 1998:35). Changes in wording between the different editions of The New Princeton Encyclopedia of Poetry and Poetics show that scholars no longer think that epic is a genre with clear boundaries, stable connotations, and definite spatial limits. This change in perception is not unrelated to the fact that in recent years, new epics have been discovered, recorded, and researched. The voluminous Vietnamese epics of the Central Highlands published from 2002 to 2007 are prominent examples; and the numerous fresh examples that have streamed into the epic realm, with their different forms, contents, and functions, continue to challenge the old epic concept based on European epics. Today, there are disagreements and controversies over almost every important issue related to epics; but length is still one of the standards that most Analysts adhere to, even though no one is clear about how long (its length). This chapter discusses the definition of epic, beginning with its length, based on a wide range of epic works, mainly from primary, that is, oral epic traditions.

Epic definition Scholars’ emphasis on length distinguishes epic from myth, legend, and prose tales. The world’s countless myths and countless legends are defined as such whether they be long or short; and most folk tales are short, with no special requirement in terms of length. The definition of narrative poems, folk songs, proverbs, elegies, and songs of praise refers mainly to their content and function. Length is not a standard. Epic as a genre can be juxtaposed with other narrative genres, but it has special features, being both an independent genre and one that often contains other genres.

114  Theories and methods of oral poetics The American epic scholar Richard P. Martin has argued that as a “super-genre,” epic is different from most verbal art forms, which use words in accordance with a relatively fixed style. It has what can be called “pervasiveness” (Martin 2005:17). John Miles Foley called the epic the “master genre” of the ancient world. No matter the era or area, it has played an important role in ancient society, from history to politics, from culture to education. As a mark of identity, ancient epics were always at the center of things (Foley 2005:1). The present author strongly agrees with Martin and Foley. The epic genre, especially oral epic, is not a static or highly self-consistent phenomenon. Length is an issue of external form; but from the perspective of dialectics, the accumulation of quantity will to a certain degree give rise to qualitative transformation. From a philosophical point of view, in general, the content of things determines the form. The content of epics often involves major events related to the life and death of a nation. Its protagonists are often extraordinary figures—gods or demigods—and its world is often magnificent. Any epic narrative that contains these marks cannot be too short in form. As a super-genre, epic must have sub-genres. Guida M. Jackson-Laufer’s rich and finely organized guide to world epic, Traditional Epics: A Literary Companion, includes about 1,500 epic works, geographically covering Africa, Asia, Australia and Oceania, the Caribbean, Central America and Mexico, Europe, India and the Middle East, North America, South America, and Southeast Asia. It lists the following sub-genres: ballad cycles, chanted fables,1 collections,2 creation epics, migration epics, rituals, prophecy cycles, sagas, epic sagas,3 fertility epics, folk cycles, folk epics, heroic cycles, heroic epics, heroic poems or ballads, heroic epic narratives, legends, myths, narrative poems, prose romances, verse romances, women’s epics, etc. (Jackson-Laufer 1996:651–654). The list suggests that epic is conceptualized as an aggregate genre, its contents full of variety. It also shows that epics and adjacent genres overlap, infiltrate, and merge with one another. However, the boundaries that Jackson-Laufer sets for epic are too broad, from myths to legends, from literary drafts to religious classics, from historical documents to memoirs of national migration. Her approach ignores, moreover, the core attributes and features of epic; and the system of sub-genres that she uses lack fundamental divisions and logically conceived levels, such as regional divisions for the sub-genre of chanted fables, gender-oriented divisions for women’s epics, and content-divided sub-categories like fertility epic and myth. As the most influential epic scholar of the 20th century, Honko’s insights on epics are of great value. For him, as we saw in Chapter 4, Epics are great narratives about exemplars, originally performed by specialized singers as super-stories which excel in length, power of expression and significance of content over other narratives and function as a source of identity representations in the traditional community or group receiving the epic. (Honko 1998:28) They are, moreover, both a complex genre and a multi-generic tradition (Honko 1998:29). Whether as Martin’s super-genre, Foley’s master genre, Honko’s

“How long is long”  115 superstory, or complex genre, the point is that the epic cannot be understood and interpreted simply as an univocal general genre.

Epic length Among many literary genres, length and volume are decisive for classification. With novels, there are full-length novels, novelettes, and short stories. Epic is generally considered a long narrative, but what should be the lower limit? Honko is one of the few scholars who have thought about this issue. He recalled the view of Edward R. Haymes, the most generous scholar on the length of oral epic, who thought it should generally exceed 200–300 lines (Honko 1998:35–36); but he disagreed with Haymes. He thought that this criterion was much too low, because it would place a whole epic at the level of a single epic episode. Honko proposed 1,000 lines as the usual standard but recognized that oral singers might have a lower standard (Honko 1998:35–36). How can the length of the super-genre of epic be determined? That is a difficult question. We have noted that Western academics previously believed that Mahābhārata was the longest epic in the world, with about 100,000 to 200,000 śloka. But the Kirgiz Manas and Mongolian Jangar are about that long; and the Tibetan Gesar (Mongolian Geser) is said to contain about 500,000 lines, excluding variants (Encyclopedia of China 2009), which makes it much longer. By comparison, the Iliad and the Odyssey, with about 16,000 and 12,000 lines, respectively, are medium-sized epics; and so, too, are the Germanic Song of the Nibelungs, the Anglo-Saxon Beowulf and King Arthur series, France’s Song of Roland, Spain’s Poem of the Cid, Iceland’s Edda, etc. Relatively short epics are also common throughout the world. According to the Mongolian Heroic Epic Series, the majority of Mongolian epics have fewer than 3,000 lines; and some have only a few hundred lines. Among well-known short epics today, the ancient Turkic Legend of Oghuz Khagan has fewer than 400 lines in the various manuscripts that have been handed down. Still, however, its content is serious; and the protagonist’s heroic deeds fully conform to the criteria of epic. On the whole, the inclusion of large and medium-sized epics in the epic category is fully in line with the requirements of the epic genre for length. Disagreement arises about the size of small epics; and the length of oral epic is never easy to ascertain, because it is relative. On the whole, the epic genre has its own pedigree, some works being pure-bred, others not; and there is a constellation of epic forms among which some, like that of the Horqin grasslands of Inner Mongolia, can only broadly be classed as epic.4 Honko has given pertinent examples of the relative length of oral epics. To expand on what we saw in the previous chapter, in January 1995, All India Radio invited the famous epic singer Gupala Naika to chant the famous long Tulu epic Kooti Cennaya on radio; and the broadcast version lasted twenty minutes. Three years later, Honko’s team recorded Kooti Gennaya as performed by the same singer. They spent three days recording nineteen epic segments; and the length of each ranged from twenty-eight to 120 minutes, or fifteen hours for a total of 7,000 verses. Honko was surprised when he then heard the twenty-minute broadcast

116  Theories and methods of oral poetics and asked Naika to tell the story once more in twenty minutes. He agreed and completed the task in twenty-seven minutes. Honko found Naika to be the most outstanding master of epic flexible compression he had known in the course of his epic research (Honko 1998:30). Nevertheless, according to Honko’s criterion, a work of 7,000 lines can readily be included in the epic genre, but if a compressed version is far less than 1,000 lines, it should be rejected. According to Honko, too, Silke Herrmann reported at a seminar on Kalevala and the World’s Epic held in Turku, Finland in 1985 that Ladakh Radio once invited singers to perform Gesar throughout a winter series of programs. The performing time was limited to thirty minutes for each session, so singers needed to compress each canto (ling) to a half hour. In one highly compressed version, one singer performed the entire Gesar epic in three hours, and another in sixteen hours (Honko 1998:30). Not surprisingly, a singer readily adjusts the length of an epic story to fit occasions and circumstances. After surveying the epic tradition of western Mongolia, Vladimirtsov noted that for an experienced bard, the same epic can be sung in either one night or three nights, using direct or indirect methods to extend the length. The idea that some epics are actually too long to be sung at one sitting can be found in many parts of the world; and it is said among the Oirat Mongols that no bard needs to learn to sing the seventy cantos of Jangar in full, as this would be hurtful for him (A Collection of Papers on Jangar 1988). It is actually considered disgraceful and even sinful for an Oirat singer not to shorten the narration or delete a certain passage when needed, but the theme and content should not be changed. Everything depends on the singer’s inspiration and his or her flexible control of the material (Chao Gejin 2000:39).

Epic elasticity Of course, length depends to a great extent on the complexity or simplicity of the poetic embellishment at the hands of different singers. We have seen how Parry invited the famous singer Avdo Medjedović to attend another bard’s performance of a song Avdo had never heard and then challenged him to sing it better. When he did so, his version reached 6,313 lines, almost three times the length of the other bard’s (Foley 2000b: 94).5 In Lord’s view, the reason was that Avdo was a much more talented bard and had mastered many more ready-made poetic formulae and typical scenes. Whereas a scene might be passed over by the other singer, Avdo would add numerous embellishments and transform it into a tapestry rich in colorful details. Citing similar instances, Lord observed that the learned version of an epic does not follow the original step by step. Instead, the singer will make many changes with the aid of familiar formulas and even revise it, so that the same story will undergo major changes and achieve different length in the process. As was noted in Chapter 9, to clarify an epic’s variability, Lord proposed three field experiments: (1) changes in a sung narrative from one singer to another; (2) differences in the same story when told by the same singer after a short time interval; (3) differences in the story by the same singer after a long interval (ten years or more). In any case, the performance of an epic narrative story recorded in a field

“How long is long”  117 survey can be either a compressed or an extended version; and the marks and classification of an oral poem cannot be reliably determined by length. The present author has found that in Mongolian epic, a singer’s flexible use of embellishments can vary considerably. In Jangar, for example, the core epithet asar ulagan Honggur (giant red Honggor) appears thirty-three times, often proceeded by Aguu yehe hüchütei (very great power) to form a two-verse epithet. Similarly, the epithet hündü gartai Sabar (mighty-armed Sabar) appears twentythree times, sixteen of which are proceeded by Hümün nu nacin (eagle among the mass): eagle among the masses / mighty-armed Sabar. In four instances, moreover, a linked four-verse embellishment precedes these two verses, creating a sixline epithet: ama tai humiin amalaju bolusi ugei heleei yagum_a ni helejil bolusi ugei htimiln-ii nacin hundil gartai sabar

People who have mouths Dare not gossip [about him], Creatures that have tongues Dare not talk about [him]— Eagle among the masses, Mighty-armed Sabar

The schema in Table 12.1 summarizes the variety of epithet embellishments in Mongolian epic. In Mongolian epic, this poetic technique can be seen everywhere in scenes of the hero’s dress, his steed, expeditions, feasts, battles, etc. It is quite complex, has a wide range of uses, and affects length to the degree that it is increased or omitted Depending on how many lines are thought to meet the requirements of the epic genre, a work can be classified as epic or just some type of narrative poem. In general, an epic is a kind of super-genre; but its length is only of relative significance. Moreover, some of the transcripts that we have may be only a kind of prompt book (Nagy 2008:42). Some such prompts that we find in India and southern

Table 12.1 The variety of epithet embellishments in Mongolian epic

118  Theories and methods of oral poetics China are actually pictures, not dictated text. It is clearly a mistake to measure a performance by the length of a prompt. The 19th-century German scholars who developed the Liedertheorie believed that long epics are mostly formed by the gradual aggregation of shorter songs. The Analysts followed that view, maintaining that the Homeric epics were composed from smaller narrative units and scattered songs. Sanskrit experts believe that the Mahābhārata gradually converged into a large epic from relatively short sequences over what the Austrian scholar Moriz Winternitz has calculated to be about 800 years. The Mongolian epic expert Rinchindorji has argued that Mongolian epics confirm the convergence process of oral epics, expanding from relatively short single-plot epics to medium-length compound-plot works and then to large, parallel compound-plot epics like Jangar and Geser (Rinchindorji 2013). Many oral epic traditions still have works in an active stage of transmission in what can be regarded as their growing period, including the Tibetan Gesar, with new cantos emerging continuously. For an epic still in circulation, it is not helpful to grade it by length because its size is far from fixed. As regards large epic cycles, the situation is even more complex. As with many grand epic traditions, the Mahābhārata is composed of many sub-genres such as legends, myths, stories, royal lineages, etc. The parallel compound-plot epic Jangar consists of dozens of poetic sequences, each of which, like Hongor’s Wedding and Mightyarmed Sabar, is both an independent poem and at the same time an integral part of the entire epic; and this poem itself may also contain several smaller genres and units of expression. As we have seen, moreover, in many cultural traditions, epics do not exist in isolation but are closely related to activities such as rituals. The Miao people’s epic King Yalu, recently discovered in the Mashan area of Guizhou Province, is a combination of creation epic, migration epic, and heroic epic that is sung by a donglang (priest) at night in a ritual to guide the soul of a deceased loved one to the other world. Embedded as it is in the ritual, its length is regulated by the length of the rite itself. The singing of the Yi people’s epic Hnewo takes the form of rap-like exchanges among storytellers on ritual occasions, with no one recitation being a complete version (Liao Mingjun and Bamo Qubumo 2004).

Notes 1 Such as the narratives of the French Middle Ages in alternating sung verse and recited prose. The only such work that has been handed down to this day is Aucassin et Nicolette. 2 Such as the Chinese Chuci, Songs of Chu. 3 Such as the various prose forms found in Medieval Iceland. 4 Regarding atypicality, see Wu Xiaodong, “Epic Category and Atypicality of Southern Epic,” Folk Culture Forum, No. 6, 2014. 5 For a comparative analysis of the two performances, see A. B. Lord, The Singer of Tales, Harvard University Press, 1960, pp. 102–105.

Part III

Indigenous research on Mongolian oral poetics



13 Mongolian oral epic poetry

Mongolian tuuli, or epic poetry, the most important genre in Mongolian literary history, consists of a vast tradition of oral works. Accompanied by musical instruments such as the tobshuur and the choor, tuuli tell of these nomadic peoples’ glorious past: their ideal heroes—the bravest hunters and herdsmen—and their ideal world—rich pastures, open steppes, decorated yurts and palaces, beautiful maidens, and swift steeds. The epic heroes guard these riches, defend them with mighty deeds, and acquire new herds and new nomadic territories. As we have seen, investigations of Mongolian epic can be traced back to the beginning of the 19th century with the German traveler Bergmann; and we have German and Russian texts from that time gathered from among the Volga Kalmyks. Around the beginning of the 20th century, researchers such as G. J. Ramstedt, C. Zhamcarano, and B. Vladimirtsov traveled to many Mongol areas and recorded a number of valuable epics (Nekljudov 1991:25). At that time, scholars spoke of three Mongolian epic regions. In Vladimirtsov’s words: At the present time in the Mongolian world, as far as we know, there are three areas, three regions, where the heroic epic cycles live or still exist, where professional singers of tales are found, … distinguished one from the other by many individual features. The bearers of these three types and forms of Mongolian epic are the following Mongolian tribes: the Buryats of the Irkutsk government and the Trans-Baikal district; the Volga Kalmyks (Oirats), together with those who in the second half of the 18th century nomadized out of Russia and now live in Dzungaria and in the Tianshan Mountains; and finally the Oirats of North-West Mongolia, together with some Mongolized Turkic tribes. In each of these regions populated by one of these tribes, we find the heroic epic, organically mature, having its own definite tradition and being preserved or living at present in one or another characteristic form. (Vladimirtsov 1923:11) Vladimirtsov goes on to note that “there still exist areas, Mongolian tribes about which our information in this regard is quite insufficient, and we can say almost nothing as to the position of the heroic epic there” (Vladimirtsov 1923:11–12). DOI: 10.4324/9781003258001-16

122  Indigenous research on Mongolian oral poetics Apparently little was known at that time about epic centers and the epic tradition in China, but scholarship on Mongolian epic poetry has developed considerably. The work of accumulating, publishing, and studying the tradition has been carried out for more than forty years, with a pause from the second half of the 1960s through the 1970s due to the Cultural Revolution; and new discoveries have been reported in regions that were never before studied. Several relevant research institutions have been founded; quite a number of important epics have been published in Mongolian, Chinese, and Oirat; and there is a constellation of scholars researching Mongolian epic in China. An example is The Geser Office of the Inner Mongolia Autonomous Region, where scholars collect, print, and study Geser with the aim of publishing twenty-three versions of this epic in thirtyfour volumes. According to reliable statistics (B.·Bürinbeki and Buyanheshig 1988:1), in addition to the two most famous works Jangar and Geser, seventy other epic poems were published in the Mongolian language in China by the 1980s. As for Jangar and Geser, we have ten versions of the Mongolian Geser in thirteen volumes1 and more than seventy cantos (exceeding 200,000 verse lines) of Jangar, not including variations. These epics were found in Mongol areas in China—Barag, Buryat, Jarud, Horqin, Qahar, Ordos, and Ulanqab—as well as in other regions with Mongolian inhabitants, such as Xinjiang, Qinghai, Gansu, and Liaoning. The majority were published in Mongolian, Todo Oirat, and Chinese in the 1980s and 1990s. Over the years, scholars have reached a consensus that there are three areas of China where Mongolian epic has flourished: Barag and Horqin in northeastern Inner Mongolia and the Oirat area in Xinjiang, Dzungaria. A few epics have also been found in Inner Mongolia in Bayannur and Qahar, as well as in other locations; but these belong to one of the three main epic centers and do not represent independent areas.

The three epic centers: Barag, Oirat, Horqin An epic center is an area characterized by a strong epic tradition in which a group of epics is linked together by common historical, geographical, and tribal features that differentiate it from other such groups. The Barag people, with their long history, inhabit the three Barag Banners (counties) of Hulun Buir Aimak in northeastern Inner Mongolia. Nearly twenty epic poems have been collected from this relatively small region, along with several related epic poems from other Mongolian areas of the province, such as Ulanqab and Ordos.2 Barag epic retains the ancient themes of hunters’ and herdsmen’s primeval ideas and customs, such as interclan fighting, revenge expeditions, and the hero’s marriage by bride-capture. Compared with the Mongolian epic tradition as a whole, the epics are quite short, none exceeding 2,000 verses, with only ten to twenty motifs in each work. Typical characters include the main hero—always a hunter or herdsman—and sometimes one or two comrades. His opponent is a monster, usually the many-headed manggus. In form, the Barag epic are semi-verse, semi-prose, which points to a degeneration of the oral art. There is no report of professional epic bards among the Barag; performers are amateurs—common herdsmen and

Mongolian oral epic poetry  123 their wives. This indicates that epic creativity and development ceased long ago in the area. Performances witnessed by the author amount to just an echo of this tradition’s glorious past, when there existed a great many talented bards. Though Barag epics have in some respects maintained the original form of the early stages of epic poetry, the Barag have lost their tradition of epic singing. It seems that even the ability to play accompanying musical instruments during a performance has been lost. Nevertheless, Barag performers have tried to keep oral performance in its original form to preserve these living fossils of folk art. As an independent tradition, the Oirat epics have spread among the Dörben Oirat, or Alliance of the Four Oirats—Chorus, Torghut, Dörbet, and Khoid— and the Qahar in Xinjiang; and it is reported that they can also be found among Mongol tribes in Qinghai and Gansu. Around the end of the 19th and the beginning of the 20th century, a few performances were collected in Xinjiang and Qinghai and printed in Russian (Nekljudov 1991:1–4). The majority, however, have been found only during the last twenty years of the 20th century. The most influential is undoubtedly the Jangar Cycle, which has several Mongolian versions, a Todo Oirat script version, and two Chinese translations. Since this cycle is certainly a series of narrative poems, some scholars do not regard it as a single epic but rather, a compilation of independent epic poems connected only by the fact that Jangar Khan appears in all of them. Twenty other Oirat epics have been collected more recently by scholars like Rinchindorji, Badma, and Jamcha (Rinchindorji 1987:23–25). Because of these modern discoveries, Vladimirtsov’s conjecture that the Oirats have only one epic needs to be revised. Oirat epic has its own characteristics. These include the same primitive motifs found in the Barag tradition; but we can also find later historical layers, such as Buddhist notions and new relations between the khan, his heroic comrades, and the people, which reflect actual social aspects of the Dzungaria Khanate. Along with the personal fights and clan revenge found in Barag epic, a completely new type of contest is found: battles between kingdoms. Oirat epic boasts a well-developed tradition, both historically and poetically. Generally longer than Barag epics, the majority of the works exceed 2,000 verses and include the much longer Jangar Cycle. They have a great many characters, including a large group of heroes, each with a distinct personality and different role in the khan’s court. The storylines have complicated plots molded from traditional formulae, epithetic phrases, typical scenes, and common motif series. Quite naturally, some similarities exist between Oirat epics and those of their Turkic neighbors. For example, the main hero is most often a khan, and it is common that when the old hero is defeated, his descendants continue fighting against the enemy. Finally, in the Oirat Epic Center, there is a long tradition of welltrained and skillful professional bards called tuulch(i) and for singers of Jangar, jangarch(i), with Hoboksar, Wenquan, Nilka, Hejing, and Hoxud Counties being famous for their Jangar singers. It may be, however, that a jangarch can sing only a few cantos of the whole Jangar Cycle. The Horqin represent a unique brand of Mongolian regional culture. Located in an agricultural area of eastern Inner Mongolia, they slowly changed their way of

124  Indigenous research on Mongolian oral poetics life from nomadic herding to farming and for hundreds of years have maintained a lifestyle rooted in both agriculture and animal husbandry. They possess their own dialect and have stronger shamanistic traditions than other areas. The region is the birthplace and core of the bensen üliger (booklet tales), or text story, a folk art in which a usually illiterate singer holds a text in hand to symbolize that the performance is not fictional but historically true to its sources—commonly, Chinese historical legends. The area is well known, too, for its holboo (folk odes), which usually tell a tragic tale about a hero’s revolt or a sad love story. Horqin singers make use of the narrative skills native to the booklet tales and folk odes but follow the traditional practices freely, drawing upon their own inspirations to compose their work creatively. Sometimes, they draw upon other epics for raw material. For example, in the Horqin epic Asar Chagan Haiching sung by Chuluu and Nasuntemur, characters and plots are borrowed from the popular epic Silin Galjuu Bagatur. In content, Horqin epics have many unique features. The main hero is sometimes the ruler and so does not personally participate in the fighting; instead, he sends out his champions to fight for him. As in other traditions, the hero’s opponent is the manggus; but it is more powerful and dangerous than those found in the other two epic centers. Each has its own name, whereas in other traditions, the manggus has many heads but no name. On the other hand, Horqin epic heroes are quite weak and dull. They cannot defeat the adversary but must rely on a moomai manggus (surrendered manggus) to do so. Another distinguishing feature is that marriage, one of the two basic themes of Mongolian epic, is absent. Horqin singers find it difficult to think that their hero should perform heroic deeds and conduct dangerous exploits just to win a wife. They see this as shameful. In their epics, the beautiful maiden must come to marry the hero on her own initiative. Women’s chastity is held in high esteem; so the hero’s wife must do all she can to protect her chastity if she is captured by the manggus and always manages to keep herself pure until rescued. Buddhist elements are also found in the epics. The Buddhist worldview has influenced many formulaic descriptions, such as the opening motif, which tells how the sun, moon, and stars were formed under the power of the Buddha. The Horqin epic hero may burn incense and present his steed with a hadag, the Buddhist silk ritual scarf. Moreover, sometimes, the hero’s weapon is some sort of Buddhist instrument.

Structure and theme Mongolian epics tell the story of a hero, his uncommon birth, his fertile fields and innumerable livestock, courtship of and marriage to a maiden with beautiful cheeks, and heroic fights with an adversary—usually the many-headed manggus, but in some cases, an entire enemy kingdom. With the assistance of his steed and a heroic comrade, he successfully defends his lands against all intruders. Heroes differ from epic to epic in the various traditions; but all Mongol epics follow two basic themes: fighting and courtship. The main focus of the audience is not how the poem will end but how the story develops. According to the eminent German Mongolist Walther Heissig, it develops with the aid of more than 300 motifs,

Mongolian oral epic poetry  125 which break down into fourteen motif series: time, the hero’s birth, his locality, the hero himself, his steed, expedition, sworn brothers and aides, threats, the enemy, fighting against the enemy, the hero’s stratagem, courtship, wedding, and return to his homeland. This series has been widely applied by epic scholars in China ever since Heissig’s article was translated into Chinese in 1983 (Burchinova 1983– 84:352–375). Various interconnected motifs form each of these motif series. For instance, under the hero’s birth, we have the submotifs: normal birth from parents, birth from a stone, coming into life by himself, destined to be born from a deity, description of his parents. Under the last submotif, we have two further submotifs: childless old couple and couple that seeks a child (B.·Bürinbeki 1997:32–33; B.·Bürinbeki and Buyanheshig 1988:987–1021). The constituent units of Mongolian epics can be classified according to three main levels: theme, subtheme, and motif, which we take to be the smallest formative unit. Within this classification, we have two epic types: single-plot epics with one unifying theme and multi-sequence epics with two or more themes. The latter can be categorized in turn into two types by structure: serial and parallel. The single-plot, single-theme epic Aguula Khan collected by the Finnish linguist G. J. Ramstedt in the early 20th century goes as follows: In the glorious days of old, there lived the ruler of the northeastern land Agula Khan. His wife once had an ominous dream and woke up her husband, but he paid no attention to her fears. Once he had lain down again to sleep, his wife saw in a dream that the fifteen-headed yellow manggus was coming to attack them. She again woke up the khan; and by divination, he realized that her dream was true. He girded himself for an expedition and went out to meet the manggus. After asking who he was and where he was headed, he engaged in a duel with him and slew him. He then trampled to death the manggus’ wife, who was just the size of a spider, gathered up the manggus’ possessions, returned home, and lived peacefully and happily. (Poppe 1979:106) As for subthemes, the fighting theme has two: fights for revenge and fights for possessions. The marriage theme has three: marriage by capture, marriage by contest, and marriage arranged by parents: A Fighting A1 fighting for revenge A2 fighting for possessions B Marriage B1 marriage by capture B2 marriage by contest B3 marriage arranged by parents Multi-round epics with more than one theme follow patterns such as A+B, A+B+B, B+B (B.·Bürinbeki and Buyanheshig 1988:987–1021; Rinchindorji 1989). There

126  Indigenous research on Mongolian oral poetics is generally an introduction, followed by the opening sequence portraying the battle with a manggus or other foe and the hero’s defeat. The second sequence might depict the hero’s suit for a heavenly maiden whom he needs to win in manly contests. The third sequence might tell of the reanimation of the hero who was laid low in the first sequence or his victory over a manggus monster and his return to his wife and family (Hangin, Krueger and Service 1989:8). In terms of structure, in multi-sequence parallel-type epics, the various cantos progress as parallel, independent units without a serial build-up and development. In Jangar, they are connected only by the fact that Jangar Khan and the baghatur (heroes) who serve him appear in every sequence with no clear serial order of events. Nonetheless, they constitute a single epic work. Each canto describes the same group of heroes with the same narrative style and conventional scheme, including a brief introduction of the central hero, Jangar Khan, and the other heroes. Each canto unfolds in accord with the same story-pattern, with a large feast opening and closing it as a kind of ring-frame. This repeated frame gives the impression that the story as a whole constantly moves forward without a series of separate beginnings and endings. As Vladimirtsov puts it: Jangar has far more internal similarity of action than it has distinct features of separate song-poems. All are linked not only by having one and the same khan; each consists of a natural continuation, a development of the preceding. Contradictions almost never arise; and the singer performing any given song calls others to mind and sketches them against a distant background. A real jangarch is one who knows and clearly presents the whole poem, the entire cycle, but can perform any one song of the Jangar Cycle in such a way that it, in itself, produces a complete impression. The Jangar Cycle differs from a Russian byliny [epic cycle], moreover, by the fact that Jangar is himself the hero, the major functioning figure of the poem. Though he turns up in some songs only in the background, yielding place to this or that hero-knight, he is everywhere a functioning hero. Nonetheless, the various songs of the cantos of the Jangar Cycle already existed as discrete poems. They were joined together into the Cycle because they all focus on heroes who are knights close to Jangar. (Vladimirtsov 1983–84:5–58) Certain features of Mongolian epic have their roots in ancient real-world historical practices and customs. The plot of battles for revenge surely reflects the social realities of actual life; and marriage by capture is undoubtedly an echo of an actual custom that existed in nomadic clan society when exogamy (interclan matrimony) persisted for many years. In later periods, marriage by contest became more common, stemming from developing tribal alliances. A tribal chieftain had to become ever more powerful, and inter-tribal marriage functioned as a bond to achieve tribal union and enhance his power. The contest always involved the “Three Manly Games”: archery, wrestling, and horse racing, which were thought to be the best for determining the bravest and strongest herdsman or

Mongolian oral epic poetry  127 hunter (Poppe 1979:119–120; Rinchindorji 1989:18). These games still survive as the most common form of organized competition among Mongolians. It is well to keep in mind, however, that although epics include historical elements, their main function is far from recording history. We should rather think of them as a reservoir of historical echoes.

Epic figures The most important figure is, of course, the hero. Some scholars argue that Genghis Khan, as the “chief and leader of the Steppe Aristocracy of nomadic feudalism,” has been the main hero of all Mongolian epics (Poppe 1979:7; Chimud-dorji Liang Yiru and Zhao Yongxian 1981). As Poppe points out, legendary accounts of Genghis Khan became so interwoven with genuine historical fact that Mongolian chroniclers soon ceased to distinguish between the two. Legends about him and his exploits occupy a leading position in Mongol written epic literature, where he appears not only as hero and leader of mighty warriors but also as a wise teacher. This is understandable, since these works to a large degree go back to the period when a unified Mongolia existed. They feature Genghis Khan, the emperor of a considerable part of the Mongol people, as the chief hero. Nevertheless, though Oirat Mongolian epics reflect actual aspects of the later Dzungaria Khanate, one of the most characteristic features of Mongolian epic poetry is its complete neglect of any authentic historical events and personalities. There certainly were national heroes and great military campaigns suitable for heroic epic treatment, but we cannot find any evidence of historical fact in the works. In some epics that may have been composed in earlier periods, the hero is apparently the representative of the Mongol people’s ideal hero—an outstanding hunter or herdsman, strong and brave. In some later epics, the hero takes on the identity of a feudal lord; but even here, one finds no evidence to support the hypothesis that there are actual correspondences between the hero in an epic and the historical Genghis Khan. Mongolian epic poets, moreover, carefully avoid using geographical names in their works. The “hero’s locality” is one of the fifteen motif series, and bards have developed many formulae describing the open steppes, the lofty mountains, and the clear rivers; but no one has any idea of where or during what period the stories took place. Some local geographical names occasionally appear (Jamcha 1983), but no Mongolian epic speaks of a specific historical period. All takes place in a “glorious time of old.” The hero’s adversary is quite often the mythical monster manggus. It always has many heads—fifteen, twenty-five, or even as many as ninety-five; and its activity always sparks the action of the epic as it strives to seize treasure, livestock, and beautiful maidens. As the chief villain, it exists everywhere, from the Horqin in eastern Mongolia to the Oirat of the Tianshan Mountains and the Kalmyk on the Volga. Some scholars argue that it stands for natural forces that the Mongols of ancient times could not control, while others believe it represents the feudal lord. But Mongolian epics do not belong to any one historical period or social class, and neither does the manggus.

128  Indigenous research on Mongolian oral poetics The word manggus itself emerges rather early, appearing in the most important document of Mongolian history, The Secret History of the Mongols, compiled by nomadic Mongol historians during the first half of the 13th century; and it is said to have the meaning of “great serpent” (Poppe 1979:134). The image of the manggus is connected with the Mongol people’s age-old shamanic beliefs, which were widespread and constituted their dominant worldview. According to the tenets of Mongolian shamanism, the world is divided into two domains—white and black, west and east, good and evil, etc.; and there are ninety-nine tenger, or heavenly deities, divided between east and west. The thirty-three eastern deities are evil, while their western counterparts stand in the opposite camp. This worldview is reflected in the structure of Mongolian epic poetry, in which characters are clearly grouped into two opposite camps, one aligned with the hero and the other with the manggus, and bipolar features pervade the atmosphere: Hero versus manggus beautiful versus ugly female relative swift steed versus sluggish donkey towering palace versus gloomy cave tasty food versus decaying corpses rich steppes versus deserted wilds domestic animals versus wild beasts deity as patron versus devil as patron This binary opposition is a conventional structural principle that reveals the people’s archaic outlook on nature, society, and the spiritual world (B.·Bürinbeki 1997:87–138). Mongolian epic is not a historical record, but it embodies their historical spirit. The manggus is not just a particular fictional villain or embodiment of a particular social evil; it represents everything of a negative nature in Mongol experience. The hero is not a historical hero per se but rather, the embodiment of the dreams, ideals, and aspirations of the Mongolian nation. Among animals, the horse plays a tremendously important role in Mongolian epic poetry. “The horse is the hero’s main, often his only, ally. It understands human speech and answers in human words, warns the hero of impending danger and saves him from death, travels over forests, water and mountains at fabulous speed” (Chadwick and Zhirmunsky 1969:313–314). We have not yet found any epic in this nomadic tradition that is without the warrior’s steed and the assistance it gives him. After all, how could it be possible that a steppe hero should set out on an expedition to pursue a maiden or fight a multi-headed manggus without his mount? A number of features illustrate the horse’s importance. An exceptional hero must possess an extraordinary steed, and the frequent motif of the simultaneous birth of the hero and his horse reveals the notion that there is a predestined correlation between them. In some cases, the horse is born prior to its master’s birth, thus heralding the hero’s coming into the world. Epic steeds possess not only supernatural strength and speed but also wisdom and magical power. Poppe points out that the horse usually serves as faithful comrade

Mongolian oral epic poetry  129 and counselor to the hero and plays a crucial role in events: “For the KhalkhaMongolian epics, and not only for them, but also in Oirat and Buryat-Mongolian epics, the horse typically assumes the role of advisor to the hero, foreseeing events and forewarning his master” (Poppe 1979:128). The hero regularly turns to his steed as a friend, companion, and coequal comrade; and in some epics, the horse plays a decisive role in the story. When the hero Bomerdeni cannot defeat his opponent, he must turn to his steed, bow three times, and beg for help; only then does he gain the horse’s needed advice. In another case, the hero Silin Galjuu fights unsuccessfully for three years against the manggus’ thirty copper followers until his horse tells his master the way to kill the foe. We frequently find the theme of the hero who does not listen to his steed’s advice; but it invariably turns out that the horse was right. The horse’s role is not always limited to battle, moreover, but sometimes also to aid in the hero’s courtship. It may transform itself into a bee or mosquito to secretly help its master win the games in competition for the bride. Basically, the Mongolian epic steed has a threefold nature: it is an animal with the form and function of a horse; it is the hero’s companion and friend, vested with human speech and wisdom; and it is a supernatural being who can foresee events and forewarn of danger. Toward these ends, it can transform itself into various shapes and help the hero to achieve ultimate success. It is not surprising that the horse, which plays such an important role in Mongolian daily life, assumes a divine nature in the epics. It is a combination of animal, human, and god (B.·Bürinbeki 1997). The material of this chapter was originally published as “Mongolian Oral Epic Poetry: An Overview,” Oral Tradition, 12/2 (1997): 322–336.

Notes 1 The ten are Ordos Geser, Oirat Geser, Zaya-in Geser, Usutu Zuu-in Geser, Ling Geser, Nomchi Hatun Geser, Beijing Modon Bar Geser, Long Fu Si Geser, Pajai Geser Un Tooji, and Abay Geser Hübegün. 2 Representative epics include Gurban Nasutai Gunagan Ulagan Bagatur, Altan Galagu Hüü, Aburaltu Khan, Bayan Bold Ebügen, Ajig Teneg Bagatur, Sireetü Mergen, Batu Uljei Bagatur, and Silin Galjuu Bagatur.

14 The Oirat epic cycle of Jangar

It is difficult, if not impossible, to specify the exact moment of the emergence of Mongolian epic. As far as we know, no convincing clues have ever been found about epic singing in surviving documents composed by historians, missionaries, or travelers over the past centuries. The Russian Mongolist Boris Vladimirtsov points out that a predisposition toward epic and perhaps even epic narrative patterns existed among North Asian hunters and herdsmen in earlier eras and were then developed during Genghis Khan’s time, the period of the rise of Mongolian nationality. Through the end of the 12th and the beginning of the 13th century, large epic songs known as byliny were created, as well as epic cycles. Martial concerns, military achievements, and most importantly, the steppe aristocratic class provided a supportive framework for the evolution of epic singing. One piece of evidence is The Secret History of the Mongols (1240), which is filled with epic motifs and features even though it is a history of Genghis Khan and his “golden family” (Vladimirtsov 1983–84:6–7). The earliest printed epic text is the Beijing Woodblock Geser, which appeared in 1716. As for Jangar, scholars agree that this prominent Oirat epic cycle matured in the 15th to 17th centuries (Rinchindorji 1999:195), the period during which the Oirat peoples moved to the Tianshan Mountains, formed the Alliance of the Four Oirats, and eventually settled on the banks of the Volga in 1630. Oirat (Oyrat) designates any of the peoples speaking western dialects of the Mongolian language group. Before entering the Mongol steppes, they were called “Forest Folks” and lived in Siberia. In the 13th century, the western Mongols were enemies of the eastern Mongols of Genghis Khan’s empire. During the following centuries, the western Mongols maintained a separate existence under a confederation known as the Dörben Oirat, or Alliance of the Four Oirats. At times they were allies, at times enemies of the descendants of Genghis Khan. A part of the western Mongol population remained in their homeland, northern Xinjiang, or Dzungaria, and western Mongolia. Another part of the Oirat Confederation, including all or some of the Torghut, Khoshut, Dörbet, and other groups, moved across southern Siberia to the southern Urals at the beginning of the 17th century. From there, they moved to the lower Volga; and for a century and a half, until 1771, they lived as nomads to the east and west of the lower Volga, where they acquired the name Kalmyk. During the course of the 18th century, the Oirat were DOI: 10.4324/9781003258001-17

The Oirat epic cycle of Jangar  131 absorbed by the Russian Empire, which was then expanding to the south and east. In 1771, those on the left bank, east of the Volga, returned to China. The right-bank Kalmyks, comprising the contemporary Torghut, Dörbet, and Buzawa, remained in Russia. We call Jangar an epic cycle because it is composed of many cantos with close mutual connections. The story as a whole concerns the Khan Jangar and his twelve warriors’ heroic deeds, how they build his palace, defeat threatening invaders, conquer others’ territories, and woo and marry beautiful maidens according to the dictates of destiny. Each canto is a somewhat independent story about one or more of the warriors’ adventures; but as we noted in the previous chapter, each also shares the overall framework and basic elements. The opening canto, Jangar un ehin bölög, provides essential information about the kingdom, Jangar Khan’s palace, his incomparable deeds, his distinguished warriors, and his worthy lady. Besides the Jangar Cycle, the Oirat people possess other, comparatively short epic songs that bear a close relation to Jangar in story-patterns and motifs. Some of them may have arisen earlier and thus had an influence on Jangar.

Jangar texts Among the Kalmyks in Russia, twenty-five cantos of Jangar have been collected. The same number have been discovered and printed in the Mongolian Republic, but about half of these are only provisionally identified as belonging to the Jangar Cycle. The Oirat Mongols in the Xinjiang area of northwest China have maintained the Jangar singing tradition up to the present time. During the late 1970s and early 1980s, the Xinjiang Jangar Office combed the Mongol area to collect and record epics. They recorded songs of 106 jangarch (Sodnomrapten and Batu 1996) and issued the twelve-volume Jangar Material, with 124 cantos, in successive installments from 1985 to 1996. These 124 stories are not all distinct from one another, some being highly homologous; so the total number of independent cantos found in Xinjiang is considerably less than 124. As was noted in Chapter 10, the author suggests classifying the various Jangar texts into five types: retold or paraphrased texts, dictated texts, handwritten texts or manuscripts, transcriptions of audio recordings, and lithographs and modern printings. The following descriptions build briefly upon what was said in that chapter.1 The Jangar epic was first introduced to the outside world at the beginning of the 19th century by the German traveler Bergmann, who encountered the nomadic Kalmyks in Astrakhan in 1802 and 1803. In his publication a few years later, he retold two stories about the hero Jangar. Much later, when the Han Chinese Bian Yuan went to Xinjiang in 1935 for revolutionary activities and was captured by Governor Sheng Shicai, he heard while in jail a Mongolian prisoner named Man Jin sing the story of Hongor. Deeply impressed, he kept it in memory; and when he was released in 1942, he began to write down a paraphrase of it in Chinese. Through Bergmann’s notes, the outside world first came to know that an epic called Jangar was prevalent among the Kalmyks in the Lower Reaches of the Volga. Through Bian Yuan’s retold story, people first realized that an epic story

132  Indigenous research on Mongolian oral poetics about a hero named Hongor circulated among the Oirat Mongols in Xinjiang. These retold texts reveal important information about the epic singing tradition. With regard to dictated texts, the Russian Mongolist A. Bobrovnikov published one of his two Jangar texts in Russian translation in 1857. The two texts had belonged to O. Kovalevskij and G. I. Mikhailov; and scholars believe that at least Mikhailov’s text was a genuine dictated text, taken down with the help of a local Kalmyk amanuensis(Burchinova 1983–84:i, 235). Another Russian Mongolist, Golstunskij, recorded a text in 1862, again with the aid of a Kalmyk assistant (Burchinova1983–84: ii, 146–147). Later, in December 1908, Nomto Ochirov, a student of Golstunskij’s student WI.L. Kotwicz, visited the great jangarch Eela Ovlaa (1857–1920) and in two days, managed to take down the singer’s entire Jangar repertoire of nine cantos by dictation. For his part, the Finnish philologist Ramstedt acquired a great many dictated texts, including some of Jangar, during his several journeys to Mongol areas. He described his field methods in this way: I got girls, boys and oldsters to relate epics. Like the Russian merchants who buy up different kinds of goods in the Mongolian countryside. I played the part of a merchant. I bought songs and epics, proverbs, riddles and similar things. I let it be known that I paid five co-pecks per page for transcriptions in a black-covered notebook, but if the text in my opinion was free of errors I would pay even up to ten co-pecks. (Ramstedt and Krueger 1978:78) Obviously, dictated texts require two participants. The person who takes down the dictation must have adequate knowledge of the language, and the person who narrates the story must have the patience to stay the course. Golstunskij apparently tried to acquire more dictated texts, but failed, not because of a lack of singers, but because those he worked with were unaccustomed to being interrupted during performance to allow time for someone to copy the libretto into a notebook (Minzu Wenxue Yicong 1983–84: ii, 150). Scholars have known that handwritten manuscripts of Jangar were discovered in Russia and the Mongolian Republic, but we do not have concrete information about these discoveries. From the mid-1950s to the 1980s, however, ten such manuscripts were found in Xinjiang, China. As far as we know, a few of them are almost the same as the printed Kalmyk versions in circulation in some areas of Xinjiang from the late 1940s onward; the rest accord with some local singers’ songs. Some scholars believe that the tradition of making Jangar manuscripts began shortly after the Oirat “clear script” (todo bichig) was invented in 1648 and that the tradition lasted until the 1950s (Sodnomrapten and Batu 1996:5). Rinchindorji contends that at least the manuscript Hara Hinis in the library of the Inner Mongolia Academy of Social Sciences is a copy of a Kalmyk version printed in Russian. According to Batnasan’s report in 1984, lords and aristocrats in the Oirat area maintained the tradition of preserving Jangar manuscripts. However, the Cultural Revolution destroyed the majority of these precious artifacts (Jamcha 1988:26).

The Oirat epic cycle of Jangar  133 In general, Oirat manuscripts were created when an aristocrat hired an amanuensis to record a singer’s libretto; and in rare cases, a literate singer would write down his own libretto. Manuscripts are quite similar to a dictated text; but we suggest distinguishing between the two in terms of the transcription’s sponsor and the purpose of transcribing it. If a text is taken down by an outsider and the purpose is scientific research or introduction to the outside world, then we call it a dictated text. If locals write out the text for their own use, we refer to it as manuscript. It is hard to slow down a singer’s pace in order to make dictation feasible. The modern practical and convenient way to do so is to use an audio tape recorder, although we still lack a precise evaluation of how modern audio technology influences a singer’s performance. Experienced fieldworkers readily recognize that an outsider with strange machines always makes a singer nervous and feel that performing before a microphone is quite different from singing for his folk audience. In such artificial circumstances, the interaction between singer and audience is lost. Moreover, performance is more than singing a libretto text. Gestures, voice modulation, and musical instrumentation all converge to convey an oral work’s meaning; an audio recording cannot preserve all these elements. Still, transcription encodes a singer’s text with reasonable precision even if it does present special challenges. It needs to be pointed out, however, that in China, quite a few transcriptions are unreliable. Some transcribers and editors consider their own knowledge of the folk culture more authentic than that of the singer, and some editors from the same native culture believe they have the right and duty to emend the text. Then, too, some compilations are confused. In Jangar Material (1998:I, 739), we read the note: This canto (dogsin hara hinis un bölög) is compiled and emended by A. Taibai, based on Arimpil’s singing, with reference to an early manuscript and two variants from the singer Jawa of Hüriye in Mongolia and the singer Binba of Tekes. More reliable is the twelve-volume “clear script” Jangar Material, which is not heavily edited; and an ideal transcription has been made by the young native scholar Dr. Dawataya and printed in Japan in 1999. Entitled Singer Arimpil of the Jangar: Heroic Epic of the Oirat Mongols in Xinjiang, this is the first emendationfree corpus of a jangarch’s libretto from China. Lithographs and modern printings started with Bergmann’s notes, and Jangar has seen numerous printed versions over the past 200 years. The Kalmyk script lithograph appeared in 1864 in St. Petersburg, and the famous Eela Ovlaa’s ten-canto Jangar was published in 1910 in Kalmyk as well. In China, the most important texts include “the fifteen-canto version,” “the seventy-canto version,” “the 124-canto version,” and a photocopy of a Jangar manuscript. However, the avenue from oral text to printed publication is not a one-way street. In some cases, the printed story has influenced oral epic singing. For example, the 1716

134  Indigenous research on Mongolian oral poetics Beijing Woodblock Geser has had a distinct influence on the singing of Jangar (Vladimirtsov 1983–84).

Singers According to widespread legends, the earliest jangarch can be traced back to the end of the 16th or the beginning of the 17th century. The story goes that Jangar contained a total of seventy-two cantos and that no bardsinger had been able to master all of them. There was, however, an old couple, Tur Bayar and his wife Tübsinjirgal, who lived in Hoboksar, Xinjiang. Tur Bayar was in the habit of putting a small stone under his Mongolian gown every time he had mastered one canto. The stones kept increasing in number until they reached seventy. The local lord was very glad to hear of his attainments and conferred upon him the title of “seventy-canto pouch” (dalan tobchi) and announced this to the forty- nine banners of the four allied Oirat tribes. This all took place before the Oirats moved to the Volga (Batnasan 1984:42). Hoboksar is a region that has maintained a strong Jangar tradition. The prominent singers Sisina Buural (mid-19th to early 20th century) and Hulbar Bayar (?–1943) came from this region; and the two best recent jangarch, Junai (1926–) and Arimpil (1923–1994), are also from the area. The single most famous jangarch is Eela Ovlaa from Astrakhan. His family was well known for its association with Jangar singing. According to the Kalmyk scholar A. Kichikov’s research, the singing genealogy of Ovlaa’s family is as follows, with the dates of the time span of their singing of Jangar: 1st generation: Jintemür 1690–1720 2nd generation: Jinceg 1720–60 3rd generation: Chagan Emegen 1760–1800 4th generation: Khusmu 1800–1880 6th generation: Ovlaa 1880–1920

(Rinchindorji 1999:29)

Ovlaa’s father Oela was not an epic singer (thus the lack of a fifth generation), so Ovlaa learned the epic from his two uncles, Delter and Margasi. Ovlaa’s stammer meant that he had to expend great effort to become a singer, but he at last succeeded, began his performing career in the 1880s, and won great fame in later years. Junai was from Hoboksar, where his grandfather Erhetü and father Jaba were excellent singers. His father was the lord’s toastmaster and was thus in a position to foster his son’s education. As a literate singer, Junai learned eight cantos from manuscripts of the local jangarch Sira Nasun’s libretto, which had been written down by that bard himself and presented to the lord’s scribe Uljitu as a gift. Junai became a pupil of Uljitu from the age of seven and so had ready access to the manuscripts and an opportunity to master them. Junai also learned quite a few cantos orally from his father and from Sira Nasun himself and the famous jangarch Hulbar Bayar. With twenty-six cantos to his credit, Junai has

The Oirat epic cycle of Jangar  135 been identified as the jangarch who can sing the most units of the Jangar Cycle. He wrote his librettos down himself. For his part, Arimpil (1923–1994), the focus of “Arimpil: From oral folk narrative to scholarly research” of Chapter 1, was born to a family of the Torghut clan three years earlier than his fellow Hoboksar countryman Junai. With twenty-one cantos of Jangar in his repertoire by the early 1990s, he was the most prominent illiterate jangarch ever known. His father Perlei was Lord Orlogjab’s courier and a close friend of his neighbor Hulbar Bayar. Arimpil enjoyed the privilege of listening to Bayar’s singing of Jangar from the time he was seven or eight years old, and the majority of his repertoire comes from this early experience. Arimpil’s father was a devout Buddhist; and he sent his son to a lamasery, hoping to add one more lama to his family. But Arimpil was more fond of singing heroic tales than reciting Buddhist Scriptures, and the only thing he achieved during his lamasery experience was the acquisition of a modest amount of Tibetan. To force him to continue learning the Scriptures, however, his father then sent him to his uncle Dambi, the lord’s augurer, when the young man was seventeen years old. But once again, the plan backfired. Arimpil was attracted by the Jangar singing of Höhegünjen, an elderly neighbor of his uncle; and, in addition, he had the chance to learn from another jangarch, Ikir Aliya, a poor singer who made a living by epic performances and used to drop in occasionally at Höhegünjen’s. As part of Arimpil’s repertoire, he learned the canto Hündü Gartai Sabar in Bölög from Bayar and Hongor’s Wedding from Aliya. His uncle chided him time and again for falling into epic singing; and his father warned him that since he was a Buddhist, he was not allowed to perform epics like Jangar and Geser, which are full of killing and violence. To perform those stories was to “commit a sin.” But his father’s warning went unheeded; and when Arimpil was eighteen, he began singing Jangar around his neighborhood and at children’s gatherings. However, he did not learn a musical instrument to accompany his singing. With the advent of Communism in China, Arimpil became a People’s Commune member, working as a coal miner and then a farmer; but he continued to sing Jangar for local audiences in his spare time through the 1950s and early 1960s. During the Cultural Revolution of 1966–1976, however, he was criticized for promoting old ideology through Jangar singing and was even imprisoned for about two months. Other accusations centered around one of his uncles, who had been a chamberlain of a Living Buddha, and another uncle who used to perform Cham, the Buddhist exorcism ritual performed by masked lamas. By his own actions and those of his relatives, his accusers argued, Arimpil was surely a promoter of the old traditions and “feudal trash.” He was charged as a “reactionary and silent resister” of “the revolutionary new ideology.” His defense against these charges was that his singing was beneficial for the Revolution, since he entertained commune members in their spare time, helping them to recover after a day of demanding labor. Good fortune came to Arimpil after 1980. This began with a visit from Choijinjab, a professor of linguistics at Inner Mongolia University, who made

136  Indigenous research on Mongolian oral poetics a recording of his epic singing, which resulted in his gaining entry to more and more social activities. He was invited to Jangar singing shows in various places, such as those organized by the Xinjiang Jangar Office, which would gather jangarch from neighboring areas to perform and have their texts recorded. He was nominated to the Chinese Folk Arts Society, Xinjiang Branch and was elected as his county’s committeeman for the Political Consultant Congress in 1984. He several times visited Urumqi to perform Jangar for domestic and foreign scholars; and in 1989, he was invited to Beijing to participate in the Jangar Exhibition. There, he performed the epic for an enthusiastic audience; and a brief biography, together with photographs, was exhibited during the show. In the same year, he won the First Place Award of the State Ministry of Culture and in 1991, a similar award by the Xinjiang Jangar Society. He died in Hoboksar County on May 20, 1994 (Chao Gejin 2000:120–124). The lives and experiences of Junai and Arimpil epitomize the lives of jangarch during the past century. What of the status of other singers? Long-term investigations of Jangar and jangarch in Xinjiang began in 1979; and by the end of the century, at least 106 singers had been identified and their performances recorded. Compared with the entire Mongolian population of 120,000, they represent only a tiny group, of which 103 were male and only three female, sixty-two illiterate and forty-four literate. Though no taboo prevents it, it is much harder and rarer for a woman to become a singer than for a man; and statistically, unlettered Torghut males are most likely to take up the art. The basic qualification for being considered a jangarch has been that a singer performs at least one complete canto of Jangar. Some Hoboksar singers have recalled that their lord’s jangarch Sisina Buural boasted a repertoire of about thirty cantos, but investigations have found the following: 26 cantos: Junai 21 cantos: Arimpil 6 cantos: T. Badma, Jawa 4 cantos: Badibazar, Harz’ha 2–3 cantos each: thirty-three singers 1 canto each: sixty-seven singers The tribal backgrounds of the singers are as follows: 62 from the Torghut 21 from the Ogeled 15 from the Qahar 7 from the Hoshod 1 Han Chinese More than half of the singers were from the Torghut tribe, indicating that the Torghut have served as the major tradition-bearers among the Dörben Oirat (Alliance of the Four Oirats). They are rumored to regard Jangar as their own

The Oirat epic cycle of Jangar  137 tribal poetry (Vladimirtsov 1983–84); and it has been said by a well-informed scholar, “It is quite possible that Jangar first emerged among the Torghut, one of the Alliance of the Four Oirats, and then spread to other Oirat tribes and became their common heritage” (Jamcha 1988:88). According to field notes made by the Xinjiang Jangar Office and the author’s own experience, jangarch who come from a family tradition have traditionally been called yasun-u jangarch, meaning “singer from a family line” and pointing to their singing authenticity. The training of singers varies quite a lot, with no single routine predominant. They normally start learning the craft at an early age; and an excellent memory, passion, and inspiration are highly valued qualities for aspirants. As a special group in the community, they enjoy respect from audiences of various social classes. This differentiates them from epic singers in Chinese society, where folk artists (yi ren) have traditionally been looked down upon. At least until the early decades of the 20th century, there were still professional singers. The top local bard would be awarded a title such as “the lord’s jangarch”; and some were retained in the lord’s palace or yurt for his pleasure, earning their livelihood entirely by performing for him. Nobles held competitions to verify their singers’ talents, and these events gave them an opportunity to gain honor from their bards’ performance. The ages of the singers investigated breaks down by date of birth as follows: 11 born between 1900 and 1910 38 born between1911 and 1920 33 born between1921 and 1930 10 born between1931 and 1940 10 born between1941 and 1950 4 born between 1951 and 1958 During the course of the investigations, a significant percentage of the elderly bards passed away. Their demise and our analysis of the surviving singers’ ages show that this folk tradition has approached its very last stage. Generally speaking, it no longer exists in Oirat areas. The author’s fieldwork reveals that some aged herdsmen—not to mention young people—have never even heard a Jangar performance! When asked about Jangar and jangarch in the Melchig summer camp in Wenquan County in 1999, three elderly herdsmen said they had some knowledge of the story and of persons who sang it, but had personally never attended a performance. Once, a singer named Jongarab was sitting in his summer yurt on the bank of the Jambiin River rolling a cigarette with a scrap of newspaper when we dropped by. The Bayanbulag area where he lived was said to be rich in Jangar performances, but what Jongarab said took us by surprise. He calmly told us that his last performance had taken place over twenty years ago when a university professor visited him in 1979 to record his Jangar. Jongarab had not had a chance to tell the old story to anyone until we arrived that afternoon. Four grandchildren lived with him, but none had learned a single line of the epic.

138  Indigenous research on Mongolian oral poetics

Performance Thanks to field notes from the latter part of the 20th century, we know that in days gone by Jangar was performed in military camps, on the long journeys of merchant caravans, in herdsmen’s yurts, and probably most often in a steppe lord’s palace (A Collection of Papers on Jangar 1988 and Sodnomrapten and Batu 1996). It is hard to imagine that any large feast or important occasion went without such a performance in the old days. The Oirats were so fond of Jangar that they could spend whole days and nights listening to a performance. The poem could be sung in any season, but most especially on a cold winter night. Sitting around a fire with tobacco and hot milk-tea, people would be immersed in the stories of the hero’s glorious deeds. A nomadic campsite was normally rather small, consisting of only a few families; so it was natural that most of the performances took place before such small groups. On the other hand, during the Jangar competitions organized by nobles and more recently by the Jangar Office, large groups of people would gather, some from far away. Such performances are termed Jangar un danggar, meaning a large show or jangarch competition. Some singers believe that the epic amounts to genuine history and that the central hero Jangar still has magical power, even though he lived long ago. Thus, one cannot shorten or modify the epic plots. Despite what some think, however, Jangar, like other Mongol and Turkic epics, is non-historical (Chadwick and Zhirmunsky 1969:142–159). Its motifs and storylines do not correlate with historical events or particular persons’ biographies; instead, they follow traditionally molded fictional patterns. As for the practice of modifying the storylines of the work, Arimpil once stated: I think it is not proper. Firstly, Jangar is real history; secondly, Jangar himself has magical power. So to add irrelevant things is blasphemous. What’s more, it would be ridiculous if Jangar could survive the attack of a fifteenheaded monster but could not escape the poetic machinations of a jangarch. If nothing can be changed, are the songs that jangarch have performed for generations always the same? The answer is surely in the negative. The songs keep changing, and the different historical layers—on both lexical and plot levels— point to the evolution of the Jangar Cycle. Still, most singers insist that they follow the traditional way. One example of the discrepancy between singers’ beliefs and actuality was Arimpil’s frustration over the differences between his published libretto and his sense of how the song should go. Five cantos of his stories had been printed under both his name and the names of one or two other bards (Dawataya 1999a:453–458). When someone then read these versions aloud to him as they had been subjected to editors’ alterations and printed together with other bards’ songs, he complained, “It seems these are not my Jangar songs. They are mixed up with those of other jangarch. These are not my Jangar” (Chao Gejin 2000:130).

The Oirat epic cycle of Jangar  139 There are taboos connected with the singing of Jangar, differing from region to region, as the author has learned from field trips. In Wenquan County of northwestern Xinjiang, people have firmly believed that to sing Jangar during the daytime would offend the deities and thus cause disasters. In Bayan Bold, southern Xinjiang, Jongarab claimed that performing at an hour other than nighttime would ensure that the singer would fall into poverty. Most are convinced that the only proper time is after dark, and some maintain that the yurt door and roof window should be tightly shut. Yet, some singers have dared to break prohibitions. When a Jangar extravaganza was held in the Salihintai Commune of Usu County in 1988 as an extension of an International Jangar Symposium held in Urumqi, most of the famous jangarch were gathered there; and after a grand opening ceremony and ritual conducted by a group of lamas, these singers performed Jangar to large audiences during the daytime, albeit with bursts of rifle fire to ensure that their activity be protected. A more common injunction prohibits learning all of the Jangar cantos. It is believed that if a person ever were to accomplish this impossible feat, he would be in dire danger of having his life shortened. Conversely, there is also a strong prohibition against singing an incomplete canto—such an action would also lead to one’s doom. Thus, in any one performance, a singer would sing neither all the cantos nor any incomplete canto. There is no record of anyone declaring mastery of all the cantos in the cycle; and answers vary about how many cantos the epic actually has: In Hoboksar, legend says seventy-two, in other regions just twelve or thirty-six, stemming perhaps from the tradition that Jangar has twelve warriors. In any case, Jangar is thought to be sacred and possess magical power. The bard Shokai, a forty-nine-year-old Dörbet Mongol herdsman, told us that in his experience, singing it too frequently is harmful. He said that his elder brother Badma lived only to the age of forty-two because he performed the work too often. He added that singing the epic too frequently would be harmful not only to the offending jangarch but also to his offspring. Another possible negative effect of too frequent performances is a long, severe winter. To perform something sacred without incurring harm, a singer needs to do so after conducting a sacred ritual, often including animal sacrifice. Muta, the wife of a late jangarch from Wenquan County, told us that a live white sheep with a yellow head must be tied in front of the gate if all twelve cantos of Jangar are to be sung. As soon as the performance is over, the sheep will then die of a stomach malfunction; and if it does not die, the singer’s own stomach will become distended, and he will die. Another family said that the animal should be a white horse. Both agreed, however, that such a sacrifice is not necessary if one plans to sing only a few cantos. We also heard in Wenquan that a simpler form of ritual is to burn incense and have a lama recite the Scriptures. On the positive side, it is said that performing Jangar exerts a strong influence on nature and human life in that its chanting nourishes mountains, meadows, and rivers, raises people’s spirit before a hunt, and brings good fortune. In any case, the main function of Jangar singing is entertainment, as is manifested most clearly in the performances by a noble’s jangarch to give pleasure to him, his

140  Indigenous research on Mongolian oral poetics family, and honored guests. Such a jangarch is basically one of two types: one who depends primarily on his master for his livelihood, or a wandering singer who is conferred the title by the lord and makes a living by singing publicly. To be a lord’s jangarch is a great honor and was highly treasured in singers’ circles. The usual arrangement was that a few families would invite a jangarch to assume such a position jointly and then share the expenses of sacrificial animals and financial awards. The reward given for a first-rate show was quite handsome: a horse, a camel, silk, tea, or other valuable goods. The melodies sung can be grouped by region, with certain communities typically possessing their own stock of melodic chants marked by discernible regional styles (Jamcha 1988:26). Certain cantos have specific musical patterns, and Arimpil observed that only renowned jangarch could create new melodies and have their own personal strains. He said that his own airs came from his teacher Hulbar Bayar; and his nephew Dr. Dawataya found six types of melodies in Arimpil’s singing (Dawataya 1999a). Musical instruments like the hugur (huur), horsehead fiddle (morin huur), tobsigur (tobshuur), and lute (pipa) have commonly been used to accompany Jangar performances (Vladimirtsov 1983-84, Jamcha 1996). Most common is the tobsigur, played by the singer himself.2

Story-patterns It quickly becomes apparent that the preponderant story pattern of Jangar is fighting. Next comes marriage, an independent, relatively rare type that is simple in plot and structure, but includes lesser sequences centered on heroes or their kingdoms, probably as a reflex of the fighting theme. The two story-patterns both embody the core ideology of Mongolian epic: conflict, fighting between good and evil, the hero versus his adversary. This is obvious in the fighting story-type; but fighting also lies at the core plot of the marriage-type epic. To win the bride’s hand, the hero either has to pass successfully through one or more duels against other suitors or take part in the Three Manly Games—archery, wrestling, and horse racing. To make a marriage epic more warlike, moreover, there is usually added a sequence of life-challenging tests set by the bride’s father, involving subduing dangerous beasts or monsters or winning precious objects. Bravery is necessary for a hero-suitor, and the competitions reveal his fighting spirit. Even in a minor story-pattern, such as facing down a warrior and winning him over as a sworn brother, fighting constitutes the core plot. In most Jangar cantos, heroes are idealized while monsters are demonized, creating a stark contrast between the two forces. As we have seen, the majority of the cantos share the same beginning and ending: Jangar and his twelve warriors and his 8,000 soldiers (in Ovlaa, 6,000) feasting in the palace. This structurally aligns the cantos. What happens between these structural “bookends” remains flexible; but to lay the groundwork and cement internal coherence, Jangar includes a lengthy introduction telling of Jangar’s heroic deeds from an early age, his lady, his palace, kingdom, warriors, steed, etc. In Arimpil’s seventeen-canto libretto, three initial cantos recount Jangar’s early

The Oirat epic cycle of Jangar  141 deeds and the establishment of his Bumba Khanate. The ten other cantos open with the same scene of Jangar and his warriors feasting in his fifteen-story palace, seven beginning with exactly the same phrase: Arban tabun dabhur / altan charlig bambalai dotora (inside the fifteen-storied golden shining palace).

Language Until recently, Jangar was still a living tradition; but is the language of the epic the same as the singers’ and their audience’s everyday language today, or different? When the present author posed this question to Jongarab in 1999, he answered: Jongarab: Different. Chao: What are the differences? J: [laugh] In the old days Mongols used to haila Jangar. C: With melody? J: Yes, they would sing it with a melody; now they just narrate it. Hailahu is an Oirat word meaning “to perform in melodic verse”; and haila means to perform in traditional epic language—that is, in verse enhanced by certain melodic strains, fixed phrases, and poetic embellishments. Jongarab clearly knew that the register of traditional epic language differs from his usual, everyday speech; but most contemporary singers do not use that specialized language. They tell the story in a register close to their everyday language. The singing tradition has undoubtedly declined, and the loss of the traditional epic language is one of the symptoms of that decline. Mongolian epic language can be identified by at least the following features: archaic and foreign words and phrases; a rhythmic dynamic, with expressions created to match the rhythm; and a high density of formulaic diction. These work together to create the idiomatic context of epic register. In Jangar and the entire Mongolian epic tradition, listeners and readers encounter archaic and foreign expressions not common in everyday use. Two archaisms, for example, are alta, a unit of measurement equal to an adult’s outstretched arms, and bere, a distance of about two kilometers. The derivation of the name of Jangar’s kingdom Bumba in archaic Mongolian is unclear; but in the Oirat dialect, bumba means “spring water container.” The Mongolian Dictionary (C. Norjin 1999:1038) defines it as “a rotund bottle with a high neck, used in the third part of the Buddhist eight-part sacrifice.” However, the original meaning of expressions that are no longer used in Oirat people’s daily life have become blurred over time. As for foreign expressions, to adorn his version of Romance of Enfeoffment to Deities (Fengshen Yanyi), the eastern Mongol huurch (singer) Jana used many Chinese idioms, such as yishou zheyian (to cover the sky with one’s palm), an expression signifying hiding the truth from the masses. Jana would then add an explanatory phrase in Mongolian. In western Mongolian epics, one finds Turkic, Tibetan, and Sanskrit words, along with expressions in line with the geographical proximity of these language groups. Jangar’s flag is “the red flag galbar” (galbar

142  Indigenous research on Mongolian oral poetics ulagan tug), the word galbar coming from the Sanskrit, galbarwasun or galbaragcha being the name of a tree deity in charge of satisfying desires. Unlike editions of other folk genres, current epic editions always gloss rare and difficult words. Many epic expressions are picturesque, meaningful, and rhythmically melodic, and constitute poetic formulae. The central feature of epic language is the use of such ready-made formulae adopted and adjusted by singers from generation to generation. The British scholar Charles Bawden has discussed this long tradition (Bawden 1980:268–299). He cites a couplet “with embers in the eyes, / with fire on the cheeks,” which appears in epics like Han Harangui, Egel Mergen Haan, and Gants Modon Honogtoi in almost exactly the same form: nüdendee tsogtoi, nüürendee galtai. or nüürendee galtai, nüdendee tsogtoi. The couplet appears with, the same meaning and almost the same wording in Arimpil’s Jangar as sung in 1991: nüür tal-an galtai, nüden tal-an tsogtoi. (Dawataya 1999)3 Similarly, the couplet formula meaning to wipe out the enemy, leaving not even an “orphan-boy of a bitch-dog,” in the 13th-century The Secret History of the Mongols is also found in Arimpil’s Introduction of Jangar: önöchin hübegün ölögchin nohai üledegel ügei. (Dawataya 1999) Experienced listeners and readers are aware that some expressions need to be understood not for literal meaning but in terms of poetic usage, such as rhyming counterparts. For example: emnig gool unmodobar eligehigsen hanggal gool unmodobar habcha higsen

[With] the Emnig River’s wood Making the elige [side parts of a saddle], [With] the Hanggal River’s wood Making the habcha [parts of a saddle].

The names of the rivers Emnig and Hanggal form head rhymes with elige and habcha. Such expressions with the function of filling rhyming slots exist throughout the Mongolian epic tradition.

The Oirat epic cycle of Jangar  143 Table 14.1 The differences between Oirat hero-tale and epic Epic

Hero-tale

Sung often with a melody and accompanied by a musical instrument Professional and amateur singers Mostly rhyme, highly formulaic, epic register Molded framework, concerns the whole world

Narrated without melody or musical instrument Amateur narrators Prosaic, everyday language Narrative fragments, concerns specific event

Oirat hero-tales are similar to epic in story-patterns, plots, and motifs but with the following important differences shown in Table 14.1. An experienced audience can easily distinguish epic from hero-tale.

Formulae In the canto Hündü gartai Sabar-un Bölög (Mighty-armed Sabar) of Jangar as sung by Arimpil in 1991, we find multitudinous repeated formulaic phrases, verse lines, and stanzas, which we will analyze at length in the next chapter. These repeated units are easily identified ready-made blocks used by Arimpil and other Oirat singers for composing an epic narrative in performance, some coming from the Mongol oral tradition as a whole. There are at least two types of formulae connected to characters—epithets and character embellishments. Epithets are used primarily or only with certain personages, while descriptive embellishments can be used as independent units, normally several lines long, to fit various compositional situations. Epithets are also comparatively more stable in form and wording. The following are found in our canto: aldar bogda noyan Jangar dogsin sira manggus hagan asa rulagan Hongor hündü Gartai Sabar dogsin Hara Sanal altan chegeji Babai Abaga agai Shabdala Gerel hatun

Great-famed Lord Jangar Atrocious yellow manggus khan Giant red Hongor Mighty-armed Sabar Atrocious black Sanal Gold-chested Babai Abaga Lady Agai Shabdal Gerel

Singers’ rhetorical skills are highly regarded, and more is expected in performance than single-stich epithets. To demonstrate their verbal proficiency, they thus often bring into service more elaborate couplet-epithets, such as:

144  Indigenous research on Mongolian oral poetics aguu yehe hüchütei asar ulagan Honggur hümüm nü nachin hündü gartai Sabar bolingga run hübegün dogsin hara Sanal

Very great power, Giant red Hongor. Eagle among the people, Mighty-armed Sabar. Son of Burin Gar, Atrocious black Sanal.

Arimpil regularly resorts to an epithet every time he introduces a character, sometimes joined with the set phrase for an assembly of warriors: aldar noyan Jangar ni tologailagad araja in naiman minggan bagatur ud mini asar ulagan Hongo rmini araja-yin naiman minggan bagatur-udmini aldar noyan Jangar ni tologailagad araja in naiman minggan bagatur ud mini asar ulagan Hongo mini araja-yin naiman minggan bagatur-udmini

Greatly famed Lord Jangar at the head 8,000 Arajal warriors [My] giant red Hongor [My] 8,000 Arajal warriors Greatly famed Lord Jangar at the head, 8,000 Arajal warriors [My] giant red Hongor [My] 8,000 Arajal warriors

Further embellishments normally consist of several parallel couplets: irehü yeren yisün jil-i ailadchu mededeg önggeregsen yeren yisünjil-i tagaju mededeg ama tai hümün amalaju bolosi ügei hele tei yaguma helejü bolosi ügei

[Things] of the future 99 years Known by surmise, [Things] of the past 99 years Known by retrospect. People who have mouths Dare not gossip [about him]. Creatures that have tongues Dare not talk [about him].

This common formulaic increment has been used by almost all jangarch, mainly for Alatancheeji Babai Abaga, first among Khan Jangar’s twelve-warrior band. He serves as their brain trust, a man of incomparable intelligence who knows both past and future. Eela Ovlaa used it in Astraxan in 1908 and Arimpil in Xinjiang in 1991 eighty-three years later. Fixed formulaic phrases also celebrate the warrior’s horse as his powerful animal companion. Arimpil sings of Sabar’s steed with such a fixed phrase seven times in the present canto and in other cantos as well: Hüreng haljan hölög [The] strong chestnut steed with a white spot on its forehead Other examples are: tonjir ud un üre tonggag gegüü ni unagan naiman minggan aranjal jegerde ugchi gi ergime hurdun

Hawk’s offspring Mare’s first colt 8,000 Aranjal sorrel steeds Swiftly round a hill, its eyes flashing

The Oirat epic cycle of Jangar  145 The following four-line embellishment shows how sprucely the hero’s steed eludes a pursuing horde while carefully protecting its master from being wounded: bum agta hügebe gejü burugu jöb ügei daldiradag ejen dege nmese hürgejü üjegedügei eb tei bugurul haljan hölög tei

[As]100,000 stallions chase from behind, [It] dodges right and left. Keeping its master untouched by [the enemy’s] sharp blades, Dashing grey haljan steed.

Formulaic expressions for horses cover a wide variety of aspects and situations: the horse’s mystical birth, its great size (like a mountain), rites for summoning it (e.g., burning incense), the detailed process of saddling it, its great speed, its magical ability to fly in the sky and travel underground. While horses are generically called mori, agta, adugu, or hölög, important steeds, like Jangar’s Aranjal, customarily possess their own distinct names; and generic horses are referred to as sagta hülüg, a poetic designation typical of the epic register not used in everyday speech. The common formulaic designation for Jangar Khan’s palace is: arban tabun dabhur altan carlig bambalai dotora

Inside the 15-storied Golden shining palace.

This couplet appears at the beginning and end of virtually every Jangar canto to introduce a typical scene in which the Khan and his warriors are feasting, or threatening messages from dangerous enemies arrive, vital decisions about the kingdom are made, grand expeditions set out, or clashes among the heroes occur. In the idiomatic epic register, the palace is thus not simply a grand building but the arena in which storylines begin and conclusions are reached. Numerals and directions are highly formulaic. In most cases, numbers do not imply any specific quantity but rather, a generic amount like “some” or “many”; and very often used are three, seven, ten, thirteen, fifteen, twenty-one, thirty-three, eighty-one, and ninety-nine. When a bard sings out gurba dahin (three times), it always means “a few times”; fifteen and twenty-one connote “quite a lot” and thirty-three, eighty-one, and ninety-nine “uncountable.” In the epic register, when the singer portrays a hero as having crossed thirty-three rivers and ninety-nine mountains, he is employing traditional language to indicate an extremely long way while avoiding abstract terms like “many” or “a large amount.” As for directional terms, statistically, left or east and right or west are used unequally. In Mongolian shamanism, the ninety-nine heavenly deities (tenger) are divided between east and west, representing realms of evil and good, respectively. It is thus understandable that Arimpil mentions the west more frequently. Finally, Mongolian epic formulae deal not only with character traits, and scenes, but also with dynamic actions: For instance, dagulaldun nairlahu (“singing at a get-together”) and hüngginen helen baiba (“say in a loud voice”). When a hero gets angry, he will conventionally “gnash his thirteen sharp teeth, / his large

146  Indigenous research on Mongolian oral poetics black eyes flashing with fury”; and when heroes start out on a journey, “their horses’ mouths are freed [from their halters] to set out.” Some such formulae simply describe a single action, others a series of actions that help build up a scene escape or a duel. In short, Oirat epic formulae cover every aspect of narration and exhibit a high degree of recurrence.

The Oirat Jangar and related epic traditions The Jangar Cycle is the unique heritage of the Oirat people. Jangar cantos have been found in other Mongol regions and also have affinities with Turkic epic; but they are fundamentally Oirat epic. Non-Oirat Mongol cantos bear a close connection to the Oirat singing tradition in one of two ways: Either the performers are themselves actually Oirat, or they learned the songs directly from Oirat bards. The few other Jangar cantos or fragments here and there can hardly be called an independent tradition. Though unique, however, the Oirat Jangar Cycle shares many characteristics with other Mongolian epic traditions: 1. Same story-patterns: fighting, marriage, etc.; 2. Same motif-series and sequences, starting with the hero’s birth and ending with his return home and feasting; 3. Same prosody, with head rhymes and parallel patterns; 4. Special epic register, dense with formulae; 5. Non-historical; 6. The monster manggus as the hero’s major opponent; 7. Important role of the horse; 8. Both professional and amateur singers; 9. Multiple practical functions of epic singing; 10. Performances accompanied by certain rituals; 11. Mainly oral transmission, but sometimes involving manuscripts and other forms; 12. Gives rise to folk legends about epic heroes and bards; 13. Draws upon foreign cultures and religions, such as Tibetan Buddhism. As neighboring peoples throughout many centuries, Mongols and Turks share similar cultural elements. As Zhirmunsky has said about affinities between the two narrative traditions, Kirghiz and Kazakh folklore provide many examples of heroic folktales (Heldenmärchen), with the plot limited by the family and tribal relations of a patriarchal society and the action laid against a fabulous background; but the origin, ideological and artistic content, and motifs and imagery can be more clearly traced in the more archaic folktales of the Turkic and Mongol peoples of southern and eastern Siberia. (Chadwick and Zhirmunsky 1969:312)

The Oirat epic cycle of Jangar  147 Mongol and Turkic epics share similar story-patterns. Turkic epic scholars have shown that the Turkic tradition has numerous marriage-type epics, the quest for a far-away bride who is destined for the hero (Reichl 1992: 143–170; Chadwick and Zhirmunsky 1969:271–339). He must overcome various obstacles, including battles against monsters or hostile heroes who also have a claim to his desired bride; and he wins her by the three contests of wrestling, archery, and horse racing. The future father-in-law may send him out, too, to conquer monsters or perform deeds of bravery. Turkic battle-type epics also have similar structural and traditional features to those of Mongol battle epics. The hero in both Turkic and Mongolian epic is often described as having a legendary birth and childhood, with predictions of future heroism or kingship. He grows up as a poor shepherd but eventually attracts the attention of a nobleman or the khan himself. In the Oirat Jangar, Jangar has an unusual birth and becomes an orphan when a monster attacks his homeland and kills his parents; but he is adopted by a nobleman, who discovers that the young boy is destined to become a khan. In the Kirgiz Turkic Manas Cycle, the warriors are often portrayed in full, each with his own life story. This is reminiscent of Jangar’s twelve warriors, each boasting his own title and epithet, playing his hard-won role, and taking his assigned position under Khan Jangar’s “left hand” or “right hand.” Some have their own canto, though this is not common in other Mongol traditions. In Central Asian versions of the Turkic epic Gorogli, the hero has forty warriors at his command; and in the Uzbek Gorogli, he is portrayed as a wise and powerful sovereign, the protector of his people against alien invaders. He is the embodiment of the popular ideal of patriarchal authority as one looking after the good of the people and having care for the oppressed and unfortunate. The legendary Age of Gorogli and his kingdom Chandybil, like Jangar’s Bumba, have the aura of a utopia where the eternal dream of social justice comes true under a wise ruler. In the Turkic tradition, the epic cycles sometimes tell not only the hero’s story but also the tales of a second or even a third generation. Though rarely seen in Mongolian epics from other regions, this motif can be seen in eastern Mongolian bensen üliger (text-tales) and the Jangar Cycle. Not only Jangar’s second generation but also Hongor’s and other heroes’ offspring appear in the epic. Turkic epic places the same kind of importance, moreover, on the hero’s steed that we find in Mongolian epic. It is a noticeable feature of Turkic poems, as of Russian byliny [oral epic], that while the hero may, and often does forget his heroic quest in the pleasures of eating and drinking, the horse is never at fault and invariably recalls him to his senses. (Chadwick and Zhirmunsky 1969:15) Another folk narrative element shared by Turkic and Mongolian epic is the young hero’s fights against dragons or many-headed giants (devs), old witches, etc., in Uzbek dastans. We cannot know whether the image of the many-headed monster entered the Turkic tradition from the Mongol or vice versa.

148  Indigenous research on Mongolian oral poetics Comparing different Altaic epic traditions is a challenging project. The Mongols and the Turkic-speaking peoples of Central Asia—Uzbeks, Kazakhs, Kirgiz, Karakalpaks, and others—have been linked ethnically and culturally for centuries and share a similar historical development. The so-called “barbarian Kalmyks” conducted predatory raids in Central Asia during the 15th and 18th centuries that gave them the reputation of being the most threatening enemy of Central Asian Muslims. It is thus understandable that in some Turkic epics the enemy is a Kalmyk khan or warrior, and that in a few Jangar cantos a Turkic khan and his property are the objects of enemy attacks and plunder.

Jangar research Jangar studies in China began in the 1950s with a few papers published in journals and newspapers. Academic scholarship stems from the 1980s; and by the end of the century, hundreds of papers were appearing each year in Mongolian, Oirat Todo clear script, and Chinese all over the country, along with a handful of editions and translations. A brief review of selected publications in the 1990s can give foreign readers a sense of the growing research in China. Rinchindorji 1994, On the Epic Jangar (Chinese, Hohhot: Inner Mongolia University Press); 1999, The Heroic Epic Jangar (Chinese, Hangzhou: Zhejiang Education Press), revised from the 1990 edition. Rinchindorji has studied Jangar for many years. He has conducted extensive fieldwork in Mongolian regions and assembled a large archive of tape recordings. His writings concentrate on epic themes and structure as well as the epic’s distribution and artistic accomplishments. Jagar 1993, Studies on Epic Jangar (Mongolian, Hohhot: Inner Mongolia Education Press). This is an introductory work consisting of five chapters. The first two cover publications and research concerning Jangar, while the following three deal with such topics as plot, theme, structure, rhetoric, characters, and folk life. Gerel 1995, The Aesthetics of the 13-Canto Jangar (Mongolian, Hohhot: Inner Mongolia Education Press). Proceeding from an aesthetic perspective, Gerel tries to probe the Mongol people’s idea of beauty through an analysis of their epic creativity; but deficiencies in knowledge of international aesthetics and academic rigor weaken the book’s theoretical proposals. Jamcha 1996 and 1997, The Source of the Epic Jangar (the author’s native clear-script Mongolian and Chinese, Urumqi: Xinjiang People’s Publishing House). As a diligent native scholar’s work based on personal observations, this work gives a vivid and detailed account of the epic tradition. However, Jamcha raises some now outdated concerns, such as the attempt to relate the epic hero to Genghis Khan; and his exclusive focus on his own people’s epic accomplishments constrains his judgment and renders some of his hypotheses somewhat simplistic. Jin Feng 1996, Jangar: Four Yellow States (Mongolian, Hailar: Inner Mongolia Culture and Arts Press). This is a questionable work. A professor of Mongolian history, especially of the western Mongols, the author tries to draw

The Oirat epic cycle of Jangar  149 a direct connection between Jangar and real history; but some of the linguistic explanations are unconvincing. Sarangere 1998, Epic Jangar and the Mongolian Culture (Mongolian, Hohhot: Inner Mongolia People’s Publishing House). The author sets out to demonstrate how well the epic cycle reflects Mongolian culture, but his argument is ultimately unpersuasive. Sechinbatu 1999, Jangar and the Mongolian Religious Culture (Chinese, Hohhor: Inner Mongolia University Press). This work treats religious elements and meaning in Jangar. The conclusion that shaman chants were the predecessor of Mongolian epic singing bears further examination. Chao Gejin 2000, Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing (Chinese, Nanning: Guangxi People’s Publishing House). This is the first work to concentrate on a particular canto as sung by one singer. Relying on Oral Formulaic Theory, the work aims to explain the core feature of traditional Mongolian traditional oral poetics, formulaic diction. This chapter was originally published in Oral Tradition, 16/2 (2001): 402–435.

Notes 1 Scholars use different methods to transcribe the Mongolian language. To avoid confusion over different dialects, I follow the written spelling. I do not differentiate masculine from feminine, and I spell “hand” as gar and “yurt” as ger. I use the written form hüreng haljan hölög (a sorrel horse with a white spot on its forehead), which alliterates perfectly and causes no confusion, whereas spelling the same phrase as küreng qaljan külüg would cause confusion. I follow the common usage for Jangar and Hongor and write tegri (heaven) as tenger. 2 The tobsigur (from the verb tobsihu, “to pluck”) is a two-stringed plucked lute found among the Mongols and Manchu. Marco Polo described how it was played before battle. The tobsigur of the Baits, Dörbets, and Hotons may be small and round; and that of some Altai, Uriankhais, and Torguts has a necked bowl. The skin covering is often taken from the groin of a horse, sheep, or goat and should be as thin as possible. Among Baits and Dörbets, the strings are made from sheep intestines that are washed several times, stretched and twisted clockwise and counterclockwise, and then dried. They are tuned to one-fourth intervals. 3 Arimpil uses different prepositions, and we follow the colloquial pronunciation of Bawden’s transcription, which differs from our usual literal spelling.

15 Analysis of Mongolian epic formulae

Oral epic is composed of spoken or sung sequences, in which the creator and communicator are one, and creation and performance are different facets of the same activity. The communication takes place in the context of two kinds of interaction: visual and aural, and performer with audience. Creative performers make full use of this situation, drawing not only on words to shape their particular style but also on their facial expressions and bodily gestures. If we read the recorded text of an oral epic as a work of written literature, we miss numerous important aspects generated in the oral context. Nonetheless, careful study of the recorded text itself can be quite fruitful. In this chapter, we analyze poetic formulae in the text of Mighty-armed Sabar, a rather short epic sequence of 652 verses in the grand epic Jangar as sung by the eminent jangarch Arimpil (1923–1994). Mighty-armed Sabar is a traditional story in which there are a great many recurring fixed phrases and verses. Though there are slight changes and some added syntactic elements, their core form can be easily identified. Mongolian epic research has not yet begun to grasp the systematic use of formulae, and no systematic analysis of such formulae has been previously carried out or yielded any theoretical formulation. Based on Arimpil, their core form can be easily i we seek to decompose Mongolian sung epic to its most basic structural units. We can then analyze the rulesare slight changes and some added syntactic elements, their core form can bgol oral poetry and establish a theoretical understanding of the basic structural laws of Mongolian oral epic.

Research method The kind of textual analysis that we pursue could limit itself to this one text of Arimpil; but in order to comb the material for what it reveals about laws of Mongol oral epic in general, it is helpful to refer also to other versions of the same text or other stories sung by Arimpil or other jangarch on different occasions. Given the myriad Kalmyk, Buryat, and Oirat texts of Jangar, it is not feasible to do statistical or other methods of calculation on the frequency of formulaic features in the whole body of Jangar material. Our research principle is that as long as we can show that a particular formula appears in some other texts DOI: 10.4324/9781003258001-18

Analysis of Mongolian epic formulae  151 by other bards, this shows that it is a fixed expression accepted in the sung epic tradition. Our study is based on textual analysis, but this does not mean that we take no other factors into account. Field experience and careful reading of the text, coupled with oral poetic and folklore theory, allow us to base our research on a combination of theoretical thought and the reality of Mongolian epic performance. We also make reference to previous research, such as Vladimirtsov’s discussion of the form and function of formulae in Mongolian epic based on his fieldwork observations. His views are in line with the general rules of oral epic creation; but Oral Formulaic Theory was then still in its early stages and could not help him to make a more in-depth and systematic analysis. At that time, formulae were understood as just one of many rhetorical devices, not as a basic factor in oral epic creation. Anyone acquainted with Jangar or other Mongolian epics readily recognizes their many repetitive elements—phrases, sentences, and combinations of sentences. Whether called “formulae” or by other names, they are recognized as a feature of Mongolian epics deserving of study for their aesthetic function. There is a clear difference between formulae and simple repetitions, and it is important to distinguish the two. Some scholars have made a preliminary classification of them, such as warrior’s oaths and pledges of allegiance, descriptions of heroes and the beauty of women, banquets, etc. (Bitkeev 1984:52–57). Our study does not refer, however, to observations based mostly on intuitive understanding, with no particular theoretical ramifications. The poetic formulae that we use to draw our conclusions consist of words, phrases, and sentences, both simple and compound, based on the Oirat dialect. The words have a high artistic level and are more complex and beautiful than those which the Oirat use in their daily lives today. They are interwoven with ancient and uncommon words and have many rhetorical defects, including tautology, repetition, and obscurities. However, these defects have little to do with our research purpose. What we study is how the language forms the poetic text by following structural rules. We deconstruct the poetic text to identify its smallest lexical units and see how their combined power produces an oral epic text. From Arimpil’s text, we learn that the smallest unit in constructing a verse of poetry is not a single word but a number of fixed and often inseparable phrases, the basic phrase formulae that constitute a poetic system. Statistical analysis of the formulaic density shows that such formulae form the linguistic basis of the entire singing tradition of Jangar and of the whole Mongolian oral epic tradition. In line with what we saw at the end of the previous chapter, the formulae used for our examples can be divided into five categories in terms of content: epithet, horse, object or place, number or direction, and action. In the entire tradition of Mongol sung epic, the list could go on almost indefinitely—including weapons, expeditions, banquets, etc. As the Russian scholar Poppe and the British Mongolist C. R. Bawden have shown, they appear on all occasions and express various meanings.

152  Indigenous research on Mongolian oral poetics

Epithet formulae The most common type of epithet formulae consists of phrases closely associated with the characters appearing in episodes of a work to highlight certain personal traits and commonly, to introduce the character: aldar noyan janggar dogsin sir _a manggus hagan asar ulagan honggur hündü gartai sabar dogsin har_a sanal altan cegeji babai abag_a gai šabdal_a gerel hatun

greatly famed Lord Jangar atrocious yellow manggus khan giant red Hongor mighty-armed Sabar atrocious black Sanal Uncle Alatancheeji Lady Agai Sabdal Gerel Hatun

Usually composed of several words, such formulaic phrases form part of a line verse, sometimes combined with other syntactic elements to create a rhythm. Not only the main characters but also some less important, rarely appearing figures are also introduced in this way. Uncle Alatancheeji and Lady Agai Sabdal Gerel Hatun are simply names, not epithets; but these appellations have a fixed form that does not change with the ups and downs of the story and so can be treated as having formulaic traits. Characters’ names themselves, moreover, tend to have a figurative sense in line with the tendency in oral societies to concretize and visualize everything. For example, Shira manggus means “yellow devil,” Ulaan Hongor “endearing red man,” Hara Sanal “black heart,” Sabdal Gerel Hatun “radiant lady,” and Alatancheeji “gilded chest.” Some such formula can also be used to describe not a particular named character but a certain kind of general character. Such is “cingg_a cihirag bagatur/höbegün” (strong and powerful warrior/hero), which Arimpil uses four times in our text, not as a complete verse but supported by other elements: cingg acihirag bagatur cag-tu bagatur cingg _acihirag höbegün bolun_a cingg acihirag höbegün bolun_a cingg acihirag höbegün bolun_a gejü

[becomes] a strong hero hero [becomes] strong and powerful

53 93

becomes a strong and powerful hero [will] become a strong and powerful hero

127 270

Some formulae serve only to help form a larger formula. For example, the phrase “gurban hurugu dutagu” (three fingers) appears in numerous instances to indicate a small gap. In Arimpil’s Prelude to Jangar, the skylight of Jangar’s palace is only “Three fingers away / From the white clouds in the sky.” In our present text, we find “The fatal lifeline viscera / Just three fingers away to cut”; and in Eela Ovlaa, Alatancheeji’s loud cry, “This palace is solemn and majestic, / Lower than the blue sky by three fingers.” Whenever a character is mentioned in Jangar, the name tends to have a specific pattern that constitutes a line verse unit or the main part of a verse, even though the character may have different positions or roles as the story unfolds. In Oirat epic,

Analysis of Mongolian epic formulae  153 and as almost a rule in Jangar, names tend to be four or five metrical feet long and serve, like epithets, to lengthen a verse and give it the same structure every time the character appears. This rule is in evidence in the twelve volumes of Jangar Material published in classical Todo Mongolian script in Xinjiang from 1985 to 1996; but it is not typical of Arimpil, especially in his performances of this poem. The examples of formulaic names in the cantos of Jangar listed in Table 15.1 have been randomly selected from Volumes 1 and 4 of Jangar Material, but they appear in the same way in cantos in the other volumes. The names listed in Table 15.1, like Jangar and Hongor, occupy only one rhythmic foot; but when they are introduced in Arimpil’s actual sung performance, almost none of them appears only in this way. When Arimpil introduces Hongor, he always tends to treat the name as a single verse and most frequently, simply as the subject or object of an action. Only in rare cases does he combine it with a verbal predicate to make a statement of complete meaning. In Table 15.2, we list the thirty-three appearances in Arimpil’s sung performance of the formulaic epithet “giant red Hongor,” the most frequently appearing name in Table 15.1, adding the numbers of the verses in which the particular formula appears. Typical of the entire Mongolian epic tradition is that in some instances, the verse in which a name appears is combined with another verse to form a couplet, Table 15.1 Selected formulaic names in Jangar Jangar Canto and Names

Jangarch

Meeting of Mungen Sigshireg and the Orphan Jangar Mungen Sigshireg Hands Shireg Khan’s Territory to Jangar ErgmpilSigshireg Hands SInvades Ujung Aldar Khan HaHometown, and the Wrestler Mungen Sigshireg Names Ujung Aldar Khan an Orphan Like Jangar The Giant Altan Soyo Khan on Mount Bortolgoin Arbas Har, Son of Jangar, Captures Ergüü Mungen Tebeg Khan Alive Bogda Jangar Holds the Jade Seal of Dogsin Shireg Khan to Summon the Lion-like Heroes Bormangnai, Son of Jayatu Aldar Khan, Conquers Dush Mangnai Khan Hongor Looks for His Uncle Jangar Grants the Jade Seal to Hoshun, Son of Hongor Hongor and His Son Hoshun Fight Uhurch Khan and Sharaigolin Gurban Khan Hoshun, Son of Hongor, Annihilates Gunan Har Sönöghi, Son of Narandalai Khan Hongor’s Wedding Hoshun Ulaan Annihilates Mal Habha Khan Hongor Annihilates the Fifty-headed manggus Andulam Hara Hongor Annihilates the Fifty-headed manggus Hesuhor Hara Hoshun Molom Harha, Son of Huanggor Wedding of Hoshun, Son of Hongor

Arimpil Arimpil Arimpil Arimpil Arimpil Junai Junai Purbujab Badamjab No. Chünjün Sa. Jongarab Basang Basang Sha·Java Sha·Java Do. Purbai Do. Purbai

154  Indigenous research on Mongolian oral poetics Table 15.2 Instances of the epithet “giant red Hongor” Transliteration of Mongolian

Translation

Verse Number

asar ulagan honggur höbegün

giant red [son] Hongor 8 times

asar ulagan honggur

giant red Hongor 6 times

asar ulagan honggur mini

[my] giant red Hongor 4 times

asar ulagan honggur höbegün tei

[with son] giant red Hongor 3 times

asar ulagan honggur-un

giant red Hongor [’s] 2 times

asar ulagan honggur-i

[to make] giant red Hongor 2 times [with] giant red Hongor [facing] giant red Hongor [to make you] giant red Hongor giant red Hongor [with two other persons] [my son] giant red Hongor [my son] giant red Hongor [goes out] [my son] giant red Hongor [starts off] [gives my son] giant red Hongor [to me]

196 202 317 343 353 572 600 614 39 303 365 388 412 649 16 104 136 278 41 91 586 478 496 95 130 280 305 337 618

asar ulagan honggur tai asar ulagan honggur-tu asar ulagan honggur cim_a-yi asar ulagan honggur hoyar asar ulagan honggur höbegün ci asar ulagan honggur höbegün garugad asar ulagan honggur höbegün bosugad asar ulagan honggur höbegün-du öggün_ebi

602 376 607 595

and the couplet often adopts the form of head rhyme, that is, rhyme at the beginning of a verse. aldar noyan janggar ni tolugailagagd araja-yin naiman minggan bagatur-ud asar ulagan honggur mini araja-yin naiman minggan bagatur-ud mini

The greatly famed Lord Jangar at the head 8,000 Araja warriors [My]giant red Hongor [My] 8,000 Araja warriors

Analysis of Mongolian epic formulae  155 The order is relatively fixed, the general coming first and the warriors—the focus of the couplet—in the second verse, which sometimes stands alone as an independent expression. The following fixed couplet focused on Hongor appears sixteen times, with the first verse ending with the rhyme ei: aguu yehe hücütei asar ulagan honggur

very great power giant red Hongor

In some cases, several verses precede the couplet to form an extended, multi-verse group. Whether in one, two, or multiple verses, the modifiers center on the core component to form an extended formulaic epithet: agcim-un jagur_a-du arban gurba hubildag amin bey_e düni ügei aguu yehe hücütei asar ulagan honggur

In the blink of an eye He has thirteen transformations, His spirit leaves his body. Very great power, Giant Red Hongor.

verse 35 36 37 38 39

This extended form has the nature of a formula rather than an off-the-cuff improvisation. In our sample, it appears only once; but it is also found in Arimpil’s Ode to Jangar and Hongor (Dawataya 1999:390), and a similar instance exists in a Xinjiang manuscript of Annihilation of Hinas’ Entire Army: hongor’s spirit leaves his body. in the blink of an eye he has thirteen transfigurations. The Jangar warrior Sabar has the following fixed epithet couplet: hümünü nacin hündü gartai sabar

Eagle among the masses, Mighty-armed Sabar.

From this and the couplet identifying Hongor, we see that in Arimpil’s way of singing, the basic structural form consists of an initial phrase plus the core epithet along with syntactic components. In the vast majority of cases, the core epithet comes at the end, so that there is no problem with the head rhyme. This is very common in Mongolian epic and can be regarded as one of the syntactic features of the genre. Another warrior who plays an important role in the Jangar Cycle is Black Sanal: ebteibugurul halian hülüg tei bolinggar-un höbegün dogsin har a sanal

Riding a handsome, strong grey horse, Son of Burin Gar, Atrocious black Sanal.

156  Indigenous research on Mongolian oral poetics In the final two core epithetic verses, we find no rhyme, which is relatively rare in the Mongolian epic. Of more interest is the fact that Eela Ovlaa uses a more concise introductory verse: bugurul haljan moritai bolinggir-un höbegün dogsin har_a sanal

Riding a grey horse, Son of Burin Gar, Atrocious black Sanal.

Arimpil uses lug for “horse,” and Ovlaa mori; and Arimpil uses the pronunciation bolinggar, whereas Ovlaa uses bolinggir. Of more interest, however, is why Arimpil fills out introductory verse with the added modifier hülüg tei (handsome, strong), whereas Ovlaa’s more concise expression must have been the traditional formula. The answer must lie in the sound that ends Arimpil’s verse, “ei,” which forms a neat rhyme (here underlined) with its paired verse in the extended format: bum agta högebe gejü burugu jöb ügei daldiradag ejen degen mese hürgejü üjeged ügei ebtei bugurul halian hülüg tei bolinggar-un höbegün dogsin har_a sanal mini

Driving 100,000 horses, Galloping right ahead, Never getting injured by knife or sword, Riding a handsome, strong grey horse, Son of Burin Gar, [My] atrocious black Sanal.

What are the rules for epithet formulae? Let’s look again at the extended format for the core epithet of the warrior Hongor: agcim-un jagur_a-du arban gurba hubildag amin bey_e düni ügei aguu yehe hücütei asar ulagan honggur

In the blink of an eye He has thirteen transformations, His spirit leaves his body. Very great power, Giant Red Hongor.

One rule is that an epithet is not necessarily preceded by an introductory verse. In this case, “Giant red Hongor” is not always preceded by “Very great power.” Of the thirty-three instances in which this core epithet appears in Arimpil’s text,1 only sixteen, about half, are preceded by such an embellishing verse to form a couplet. Only once, moreover, does Arimpil extend the embellishment further with the couplet “[He] has thirteen transfigurations / [His] spirit leaves his body.” In the case of the epithet of the warrior Sabar seen earlier, the usage is similar: hümünü nacin hündü gartai sabar

Eagle among the people, Mighty-armed Sabar.

Analysis of Mongolian epic formulae  157 The core epithet “mighty-armed Sabar” appears twenty-three times in our text, preceded only sixteen times by the verse “eagle among the people.” These two examples suggest that as a rule, a couplet consisting of the core epithet and a subordinate embellishment is a basic pattern in Jangar singing, but that the preceding subordinate verse is not always present. As with Hongor and Sanal, moreover, an extended group of verses further fills out the epic’s depiction of Sabar: ama tai hümün amalaju bolusi ügei hele tei yagum_ a helejü bolusi ügei

People who have mouths Dare not gossip [about him]. Creatures that have tongues Dare not talk about [him].

Such a unit, too, is not always present when a character is mentioned, but its frequency is quite high. For Sabar, it appears four times in this short episode, but not always with exactly the same structure or wording. Of the two rhyming couplets that constitute the unit, sometimes the one beginning with “people who have mouths” occurs first, sometimes the one beginning “creatures that have tongues.” In some instances, moreover, the negative expression bolusi iigei is replaced with boldag iigei. All this exemplifies the rule that the formula of such a four-line poetic unit is fixed in form and rhythm, but as long as each individual word conforms to the basic meaning and rhythm, there can be variants. The formula is thus composed in terms of specific meaning and rhythm rather than a simple repetition of words, and adjustments are allowed. Such four-line groupings appear often in Jangar. Another introduces Jangar Khan’s military commander Alatancheeji, who is characterized by his wisdom: irehü yeren yisün jil-i ailadcu mededeg önggeregsen yeren yisün jil-i tagaju mededeg

[Things] of the future 99 years Known by surmise. [Things] of the past 99 years Known by retrospect.

The name Alatancheeji appears a total of six times in Mighty-armed Sabar. These verses illustrating his extraordinary strategic wisdom are used only the first time the name appears, but audiences readily recall Alatancheeji’s cleverness and wisdom each time his name is later mentioned. In other stories chanted by Arimpil and other jangarch, the same four-verse unit may refer to other counselors who stand as an image of wisdom, and sometimes even to depict a noblewoman. As a relatively fixed phrase depicting a character, the verses are thus not specific to Alatancheeji. In any case, their relationship with the name of a character is relatively loose; and in many cases, they are not used at all.

158  Indigenous research on Mongolian oral poetics A set formulaic group of verses also describes the hero Alia Shonhor: utulhul_a ulagan cilagun boldag cabcihul_a cagan cilagun boldag aliy_a šonghur gedeg bagatur ni

When they cut him, He turns into red rock. When they cleave him, He turns into white rock. Hero Alia Shonhor.

In our text, this particular group is used only once; but according to our reading and field investigation, this kind of descriptive epithetic formula with its neat parallelism must have a long tradition. Like the many other formulae, it is a regular phrase in an impromptu performance to be skillfully shaped to make the singing flow smoothly. In all these examples from Mighty-armed Sabar as sung by Arimpil, we see that the basic components of an epithet formula remain unchanged. The formula itself, however, is available for linkage with other formulae, usually in the form of a parallel structure in which two independent verses form a rhymed antithetical couplet. As we have seen, too, despite their basically fixed nature, the formulae are open to change within limits. In accord with the rhythm and pace of the adjacent verse, “holy saint” may be added to “greatly famed Lord Jangar” to lengthen a verse; and the commonly used term noyan (lord) can be removed to shorten a phrase or, as in the following two examples, be replaced by another term: aldar janggar tolugailagad aldar janggar tolugailagsan

greatly famed Jangar at the head great Jangar is the head

Arimpil himself rarely omits an epithet. Though he has just mentioned the name of a hero, he still adds one or two verses of the embellishing formula when he sings of him again. The extensive use of epithet formulae found in his singing is similar to what we find in folk epics elsewhere in the world, including Homer’s poems (Table 15.3). Nonetheless, the kind of lengthy Mongol formulaic epithets that we find in Arimpil’s singing are rare in other traditions because they take up a rather long time.

Table 15.3 Epithet formulae in Arimpil’s singing and Homer’s poems Arimpil

Homer

mighty-armed Sabar descendant of Takil Jula Han grandson of Tangsug Bumba Han son of Ujung Aldar Khan

fleet-footed Achilles Laertes’ son offspring of Zeus resourceful Odysseus

Analysis of Mongolian epic formulae  159

Formulaic density The role of formulae in oral epic performance can be studied from two aspects. One is what we have just seen, their regular use and diversity. Another is the frequency of their recurrence in a text, that is, formulaic density, the approximate proportion of formulaic expressions in the whole text. The analysis of formulaic density is one of the basic methods developed by the Oral Formulaic Theory. It has been applied to re-examine famous works like The Song of Roland, The Poem of the Cid, and a large number of chanson de geste, which, like epics, use formulaic expressions and aurally appealing sequences to praise the achievements of a hero, but often with a more ideological orientation. Joseph Duggan, Professor of French and Comparative Literature at the University of California, Berkeley, has proposed the following about the proportion of repeated formulae in oral and written texts: “If the pure repetition is less than 20%, then it may be derived from written or written creation; and when the frequency of the formula exceeds 20%, it can be proved that it is a verbal creation” (Duggan 1973). Though his argument inevitably met with criticism, it has been seriously tested by scholars and gained some success. The work of calculating formulaic density is mechanical and tedious, but not meaningless. It is closely related to the singing style and traditional skills of oral epic, the relationship between quantity and quality being dialectic, each fostering the other. We have seen that the most frequently used formula in Mighty-armed Sabar is the epithetic phrase “giant red Hongor,” which appears thirty-three times in this poem of 652 line verses. In lines 303 and 305, it appears as a single verse, but in most cases, it forms the core of a formulaic unit supported by other elements, including syntax markers like “together,” “facing,” and “make you.” In sixteen of the thirty-three appearances, the phrase is preceded by “Very great power” to form a couplet, adding up to forty-nine verses used to introduce Hongor; and the one three-verse elaboration “In the blink of an eye / He has thirteen transformations, / His spirit leaves his body” brings the total to fifty-two verses, or about 8% of the whole poem. As one of the two main characters in the poem, it is not surprising that Hongor’s name appears frequently. Mighty-armed Sabar also appears often, twenty-three times, sixteen of which are preceded by “eagle among the people” in a couplet, adding up to thirty-nine instances. In some of these, after the fixed phrase, there is added “together” (170), “to make” (180), “hero” (251), or “facing” (529); and in one case, “ene htindii gartai sabar i” (569) comes before the fixed formula. This suggests that in Mongolian epics, beginning as verses often do with head rhymes, additional components are usually placed at the end of the verse. In addition, in four instances, there is a fourline embellishment—“People who have mouths / Dare not gossip [about him] / Creatures that have tongues / Dare not talk about [him]”—making the overall total fifty-five verses referring to Hongor, 8.4% of all verses in the poem. Variants of the formulaic epithet “greatly famed Lord Jangar” appear thirty-six times (listed alphabetically):

160  Indigenous research on Mongolian oral poetics aldar bogda noyan janggar aldar bogda noyan janggar mini - 2 (appearing two times) aldar janggar aldar janggar cini tolugailagad aldar janggar gedeg hoyar bain_a aldar janggar haliyagsan cag-tu aldar janggar helegsen-di - 2 aldar janggar ni helebe aldar janggar tolugailagad aldar janggar tolugailagsan aldar janggar-un aldar janggar-un bagatur bolhul_a cini aldar janggar-un haihirugsan dagu ni aldar janggar-un hin aldar janggar-un inggiged heleged aldar janggar ni tolugailagad aldar noyan janggar aldar noyan janggar bolugad aldar noyan janggar-iyen haliyahul_a aldar noyan janggar ni - 7 aldar noyan janggar tolugailagad aldar noyan janggar tolugailagsan aldar noyan janggar-tu - 2 aldar noyan Janggar-un aldar noyan janggar-un emün_e bey_e-dii aldar noyan janggar-un hin aldar noyan janggar-un hin bagugad The formula “Uncle Alatancheeji” appears in six verses: altan cegeji babai altan cegeji babai abag_a ni - 2 altan cegeji babai mini altan cegeji babai ni altan cegeji hoyagula ban “8,000 Araja warriors” appears seventeen times, which is not a small number: araja-yin naiman mmggan - 5 araja-yin naiman minggan bagatur aca talbigad araja-yin naiman minggan bagatur-ud - 6 araja-yin naiman minggan bagatur-ud-i araja-yin naiman minggan bagatur-ud mini - 2 araja-yin naiman minggan bagatur-ud ni araja-yin naiman minggan bagatur-ud tagan

Analysis of Mongolian epic formulae  161 In a word, in his singing of Mighty-armed Sabar, Arimpil uses 173 verses containing formulaic epithet phrases, or 26.5% of the total number of verses, not including set phrases for figures like craftsmen, etc. This high number is a distinctive feature of Mongolian epic, far higher than in some other epic traditions. There are also other types of fixed expressions that recur multiple times and so can be considered formulae. In Arimpil’s text, the shortest is helen baiba (talk about), which appears five times, each as a separate verse with no changes in verb, voice, or grammatical case. Even in such a simple expression, the singer relies on a traditional ready-made expression, usually without bothering to change it. Other fixed phrases are hogsin darhan (old craftsman), which appears twice, each time constituting a complete verse, and endegtiii sonusugad (misunderstand), three times. Using verses as formulaic units involves no major methodological research difficulties, but length can create a problem. Arimpil’s complete text of 652 verses is too large to analyze in its entirety. We have, therefore, randomly selected just one segment of 100 verses to estimate the percentage of repeated formulaic phrases in the whole poem; but the percentage for the whole work may not be exactly the same as for this sample. For clarity, the 100 verses are given in alphabetical order: abai sir_a balda ban abai sir_a balda ban abhulju orhihul_a bolun_a gejü agta hülüg-ün ama-yi aguu yehe hücütei aguu yehe hücütei aldar bogda noyan janggar mini aldar janggar aldar janggar helegsen-dü aldar janggar tolugailagad aldar janggar tolugailagsan aldar noyan janggar bolugad aldar noyan janggar ni aldar noyan janggar ni aldar noyan janggar ni aldar noyan janggar-un hin bagugad altan cegeji babai altan cegeji babai mini altan cegeji babai ni altan cegeji hoyagula ban ama-yi tatagad ama tai hümün ama tai hümün amalaju boldag ügei amalaju boldag ügei araja-yim naiman minggan

162  Indigenous research on Mongolian oral poetics araja-yim naiman minggan araja-yin naiman minggan bagatur aca talbigad araja-yin naiman minggan bagatur-ud araja-yin naiman minggan bagatur-ud mini araja-yin naiman minggan bagatur-ud tagan aranjal jegerde-yin arg_a ügei bolba asar ulagan honggur höbegün asar ulagan honggur-i asar ulagan honggur mini asar ulagan honggur mini bagatur-ud-i dagagulugad bagatur-ud ni magtagad baragun bey_e degen bodudag hereg bisi boldag höbegün sanjai bolhu ügei yaguman gejü cingga cihirag höbegün bolun_a cohigsan cag-tu ejegui cagan hüürüg elesün sir_a tohui bar endegürel ügei heledeg jöbtei bile ene ayul gürüm tei ene hoyar-i ermen cagan hödege ese odbal güjegen gümbü güjegen gümbü hoyar gurban dolug_a harguldugad jöbtei bolba gejü heger_e bagugad hejiy_e cü ühüdeg yaguman hele tei hümün hele tei hümün helejü boldag ügei helejü bolusi ügei helen baiba horin nige honug-un gajar yabugad hoyagula ban helen baiba hüiten hara araja-yi hüiten hara arihi ban uugugad hümün amitan haldadag ügei hümün-ü nacin hümün-ü nacin hündü gartai sabar ni

Analysis of Mongolian epic formulae  163 hündü gartai sabar tai hüngginen helegsen-dü hüngginen helejü bain_a hürüged iregsen cag-tu ir mörgün ügei nairlaju baital_a nayan nigen esitei nayan nigen esitei nige hücün cidal-i ni nige jirühe barin nige sanagan odo öggüged baital_a orugad iregsen cag-tu sogtugad heisbe solugai baragun ügei athugad sonuscu baigad suladhagad garugsan cag-tu tere heb tegen tere önücin gagcahan höbegün gejü tere önücin höbegün-i tere tuhai-du tögerigsen hümün tüsirgürden haliyan baiba üjegülcegülügsen bolhul_a uuguju abugad yagahidag bile ta gejü yagahihu ban cu medejü cidagad bain_a yehe hücütei birman bolju garba zandan modun-u onggi-du In this sample, the percentage of repeated formulae is surprisingly high, 44%. In selecting a segment that can reasonably be expected to reflect the percentage of formulae contained in the whole poem, we have avoided passages from the beginning and end of the poem, where Mongolian epics tend to have a relatively high proportion of formulaic expressions. In our calculations, moreover, we group all subtle variations in an expression under one formula. For example, “greatly famed Jangar” and “greatly famed Lord Jangar” are regarded as one and the same formula; and the same goes for “when [he] spoke sonorously” and “[he] is speaking sonorously.” A little common sense indicates, however, that the percentage of 44% is surely lower than what can be found in Arimpil’s Mighty-armed Sabar as a whole. This is because a formula that is not counted in our calculations because it appears only once may very likely appear a second or third time in the text as a whole. Moreover, some formulae that appear only once in this poem appear in other poems sung by Arimpil.2

164  Indigenous research on Mongolian oral poetics Table 15.4 The analysis of formulaic phrases in the Classic of Poetry Category

Number of verses

Repeated phrases

%

Guofeng (Airs of the States) Xiaoya (Lesser Court Hymns) Daya (Major Court Hymns) Song (Eulogies)

2,608 2,326 1,616 734

694 532 209 96

26.6 22.8 12.9 13.1

People raised on the appreciation of written literature will surely be turned off by all this verbal patterning and repetition; but it conforms to the basic laws of oral poetry, in which the repetition of fixed patterned units does not produce a sense of redundancy or awkwardness. Appreciation is, as a matter of fact, facilitated by their indicative and guiding functions. Listeners molded by this stylized tradition will never denounce a singer because the text is stereotyped and old-fashioned. On the contrary, in the judgment of the folk, an outstanding performer must be a master who can skillfully use set programs in many ways. The use of poetic formulae to construct oral narrative verse is not limited to epic. There are examples from other folk genres throughout the world. Nearer to home, many can be found in the Huur-un üliger storytelling popular in Eastern Mongolia. In the booklet Mongolian Idioms recently compiled by scholars, pieces performed by famous hulchi-hul (Huur-un üliger singers) are listed by categories such as “descriptions of war,” “descriptions of women,” etc. to illustrate how singers create artistic patterns in their storytelling. Some scholars have studied China’s Classic of Poetry in terms of Oral Formulaic Theory and have come to the conclusion that an average of about 21% of its poems have been constructed with the aid of highly modeled poetic formulae (Table 15.4). The total number of verses in the Classic of Poetry is 7,284, while those composed of formulaic phrases total 1,531. The composition of the work is characterized not only by the large number of such formulae but also by their systematic use. Traditional Peking Opera is also a highly stylized performing art. Its educated audiences clearly know what the patterns of a performance are before the opening, including story clues, melodic strains, and staged movements. It is not curiosity about the unknown or suspense created by the story but other factors that attract the audience.

Arimpil’s use of formulae An epic like Jangar has a long history of transmission, its language polished and refined by generations of singers to a very high level of oral folk art. Interpretation of such a text demands more than piecemeal analysis; it requires a comprehensive interpretation and an overall grasp of the flow of the narrative. When the words pour out of Arimpil’s lips like flowing water, appearing one by one and then disappearing into thin air, what do they convey to the audience? What kind of picture

Analysis of Mongolian epic formulae  165 unfolds? We can get a glimpse of this by seeing how he weaves the “long words” of traditional formulae into a poetic sequence in this forty-verse segment from the beginning of Mighty-armed Sabar: arban tabun dabhur altan čarlig bambalai dotur-a aldar noyan janggar ni tolugailagad araja-yin naiman minggan bagatur-udni dagulaldun nairlaju baital_a baragun bey_e-yin bagatur-ud-un ehin-dii sagugsan irehii yeren yisun jil-i ailadcu mededeg onggeregsen yiren yisun jil-i tagaju mededeg altan cegeji babai abag_ani hüngginen helen baib-a aldar bogda noyan janggar araja-yin naiman minggan bagatur-ud asar ulagan honggur mini bagatu rdalai abu tai bayan dalai eji tei ama tai humiin amalaju bolusi ügei helti yagumü ni hülejil bolusi ügei hümiln-ii nacin hündil gartai sabar gedeg gagca höheglün tei bile gun_a orudag nasun-du ni eji ni önggeregsen dun_e orudag nasun-du ni abu ni önggeregsen hümün-u nacin hündü gartai sabar hobegun degen eji abu hoyar ni helegsen bile ene naratu yirtincü-du agcim-un jagur_a-du arban gurba hubildag amin bey_e duni ugei aguu yehe hücüte asar ulagan honggur aldar janggar gedeg hoyar bain_a

Inside the fifteen-storied Golden shining palace, Greatly famed Jangar leads 8,000 Araja warriors. Singing and drinking. Right-wing Heroes are seated in positions of honor. [Things] of the future 99 years, Known by surmise. [Things] of the past 99 years, Known by retrospect. Uncle Alatancheeji Speaks loudly. Greatly famed Lord Jangar, 8,000 Araja warriors, Giant Red Hongor’s Father Batar Dalai, And [his] mother Bayin Dalai. People who have mouths Dare not gossip [about him], Creatures that have tongues Dare not talk about [him]. Eagle among the masses, Mighty-armed Sabar, A lonely hero. When he was three years old His mother died. When he was four years old His father died. Eagle among the masses, Mighty-armed Sabar. His father and mother spoke of him In this way. In this world of sunshine, In the blink of an eye, [He] has thirteen transformations, [His] spirit leaves his body, Very great power, Giant Red Hongor. And greatly famed Jangar.

Reading this sequence, we are impressed by the slow pace of narration. The first seven verses introduce the site and the characters—Jangar and his men “singing and drinking.” In a reference to time, the next six verses introduce Alatancheeji, who “speaks loudly,” continuing for a total of ninety-three verses up to verse 106. The forty-verse sequence is composed of the formulaic “long words” used by illiterate singers, often encapsulated in a single line, which Foley sees as the smallest

166  Indigenous research on Mongolian oral poetics unit of expression, but sometimes extending to a typical scene or story-pattern (Foley 1997). Most are fixed expressions. Unlike our everyday use of sentences of just one or two clauses, the passages of the sequence are structured in multiple layers. The first two verses, “On the fifteenth floor / Of the golden palace,” constitute a “word” that Arimpil uses in several stanzas (different from Eela Ovlaa’s formula, “ten-floored and nine-colored palace”). The next two antithetical verses introduce Jangar and his 8,000 warriors (6,012 in Eela Ovlaa), filled out with the fixed verb formula “singing and drinking.” The six lines describing Uncle Alatancheeji consist of two common fixed patterns. Then come three epithets. The two verses “Father Batar Dalai / And mother Bayin Dalai” are not actually formulaic expressions and can be set aside. The next six verses introducing Sabar combine general and particular formulae, followed by the verse “A lonely hero.” The four-verse unit “When he was three. … When he was four” constitutes a group of progressive, parallel, fixed expressions that are also used to speak of Jangar’s childhood in several versions of Jangar. After two more lines, a group of general and particular formulae are used to modify Hongor; and the whole sequence then ends with “greatly famed Jangar.” Of these forty verses, five deserve special attention. The parallel pattern of the couplet “Father Batar Dalai / And mother Bayin Dalai” exemplifies neatly the traditional Mongol mode of expression; and the phrase “A lonely hero” is a typical formula. However, we are reluctant to regard as a formula the sentence “His father and mother spoke of him / In this way” (verses 32–33), even though it follows a traditional pattern. In the Alphabetical Table of Verses in Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing (Chao Gejin 2000), “father” and “mother” appear together four times. Only once does there appear the expression “abu eji-yin helegsen iige-yi” (his father and mother said), probably to rhyme with “aldar bogda noyan janggar mini” (my greatly famed holy saint Lord Jangar) and with abu added at the beginning to fit the rhythm. Finally, the Alphabetical Table shows that the phrase “In this way” appears twice; so it is not a random expression here. It conforms to Arimpil’s practice of stating expressions of speaking in a single verse that creates a pause between a character’s words and the storyline. All in all, this forty-verse sequence exemplifies how Arimpil shapes his story through the systemic use of formulaic expressions. Each verse is based on the traditional method of composition, with syntactical elements added before or after the large words to connect the various formula units so as to make the story flow smoothly. (The bracketed words are not part of the formula itself but are often combined with it.) Clearly, he does not create a large hoard of specialized poetic expressions out of the blue but draws upon the store of fixed formulae in his memory as ready-made units that are available for his use at any time. Due to his fame, Arimpil’s singing of Jangar has been recorded many times; and the recordings clearly show alterations in his use of formulae in different episodes of the same poem. Dawataya found, moreover, that the text of Chiba University, Japan that he recorded of Arimpil’s singing of Mighty-armed Sabar differs from Canto 15 of the work as recorded by Badma and Buyanheshig on several counts: Not found in his Chiba text is the sequence in which Sabar becomes

Analysis of Mongolian epic formulae  167 an orphan and loses his way due to the “atrocious yellow manggus” and that in which Alatancheeji and Güjeen Gümbe propose that all soldiers and horses be dispatched together when conquering Sabar. In Badma and Buyanheshig’s Canto 15, Jangar’s injured soldiers flee to the snow-covered mountain, while in the Chiba version, they flee to the silver-white mountain of Orhe Rotoru; and Hongor asks his horseman to lasso the horse, whereas he lassoes it himself in the Chiba text. In Canto 15, Sabar voluntarily swears allegiance to Jangar, whereas in the Chiba version, Jangar urges Hongor and Sabar to become his brothers; and Sabar sits below the handsome Mingyan, whereas in the Chiba version, he sits in the third seat from the left-handed hero (Dawataya 1999). The present author’s recording of Arimpil’s text also differs from that of Badma and Buyanheshig on several counts. Their text has 680 verses, whereas the text we have been analyzing is a little shorter, with 652 verses. There are also two important differences in the plotline. In Canto 15, Sabdal Gerel Hatun lies about Jangar being taken away by the manggus; and the Canto describes horses, weapons, and equipment in greater detail. In terms of specific expressions, Hatun is described as “arban naiman nasutai / agai sabdal_a gerel ha-tun” (eighteen-year-old / Sabdal Gerel Hatun), whereas in our text, the first verse is lacking. The material of this chapter is taken from Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing, chapters 4 and 6 (Nanning: Guangxi People’s Publishing House,2000).

Notes 1 The thirty-three lines are 16, 39, 41, 91, 95, 104, 130, 136, 196, 202, 278, 280, 303, 305, 317, 337, 343, 353, 365, 376, 388, 412, 478, 496, 572, 584, 593, 598, 600, 605, 612, 616, and 647. 2 In the Alphabetical Table of Verses of Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing (Chao Gejin 2000, we list all verses of the entire poem Mighty-armed Sabar to understand the total repetition rate.

16 Analysis of Mongolian epic prosody

This chapter discusses issues of the poetic rhythm, rhyme, alliteration, assonance, and parallelism in The Mighty-armed Sabar in order to attain a degree of theoretical understanding of the prosody of Arimpil’s verse and by extension. that of the whole of Jangar.

Poetic meter It is a challenge to try to schematize the versification of Mongolian poetry. V. N. Pozdneev (1851–1920) attempted to do so but in the end did not reach any definite conclusions, claiming: “Laws of meter for Mongolian poetry, if they do exist, remain for me indefinite” (Poppe 1979:171). Poppe is also one of the few scholars to get involved in this issue. He argued that Pozdneev failed to understand Mongolian stress correctly. Based on the incorrect assumption that it is placed on the last syllable, Pozdneev believed that there can be only two types of meter in Mongolian poetry: iambic, and anapestic or paean (four syllables, one long), along with a mixture of them (Poppe 1979:172). Pozdneev concluded that accentual rhythm does not fits Mongolian poetry, nor does metrical syllabic rhythm or any of the other usual methods of verse analysis (Pozdneev 1880:323). Scholars gradually reached a consensus that Mongolian verse stress is on the beginning, not the last syllable, and that it is purely accentual, with no relation to vowel length. The rhythm of Mongolian poetry is “equally-accented tonic verse, with a predominance of trochee” (Poppe 1979:172–173). A simple syllabic metrical analysis is, therefore, not enough to deal with the problem; and metrically incomplete catalysis is now recognized as one of the unique characteristics of Mongolian poetry. For his part, Poppe tried to establish the rhythmic pattern of Mongolian Khalkha poetry taking Enkhe Bolot Khan as his model.1 xentei xanig cox’o baixad xerlen tulig salbag baixad dalai lamig bandi baixad dajan xanig balcig baixad xangain modig dzuldzag baicad xatan xarig unag baixad DOI: 10.4324/9781003258001-19

Analysis of Mongolian epic prosody  169 galin tujagulairch baixad gadzrin xöbög sarladz baixad manaigal’in ex’nde maidrin gal’in tuxende dewen - garin ujede dewadzingin orondo xad xadin jixel gen xamag xadin jixel gen edzed edzdin turunix gen enxe bolod xan gedz negen j’xe xan ba’ba gen The rhythm here is almost invariably the accented–unaccented pattern of a trochaic: —~|—~|—~|—~ —~|—~|—~|—~ —~|—~|—~|—~ —~|—~|—~|—~ —~|—~|—~|—~ —~|—~|—~|—~ —~|—~|—~|—~ —~|—~|—~|—~ —~|—~|—~~ —~|—~|—~~ —~|—~|—~~ —~|—~|—~~ —|—~|—~|— —~|—~|—~|— —~|—~|—~~— —~|—~|—~| — ~ | — ~ | —|— ~ | — Another example from Poppe provides a reference for our later discussion of Arimpil’s text. It is from the western Buryat epic Alamji-Mergen: xunei junei huga – gui xunei junei ezle - gui xulgei junei gesxe - gui xuler mongon dubunda garzal baibal - da t’isen jasan xo’no - lo oido - xanai garaza oido baihan modij’ obo sobo unagaza uzurhan’i otolzo

170  Indigenous research on Mongolian oral poetics uzurhen’i xirbebel —~|—~|—~~ —~|—~|—~~ —~|—~|—~~ —~|—~|—~|— —~|—~|— —~|—~|—~~ —~|—~|—~~ —~|—~|—~~ —~|—~|—~|—~ —~|—~|—~~ —~|—~|—~~ Poppe participated in direct field surveys; so when he talks about epic prosody, he has an advantage over scholars who only work from books. As a performance art, its poetry is for chanting and listening, not printing and reading; and Poppe rightly recognized that syllables that seem too short can be made up for by prolonging vowels during actual singing. In any case, from an oral and auditory perspective, shortness of syllables does not mean faulty versification. G. J. Ramstedt (1873–1950) very early saw that Mongolian and Turkic oral poetry have much in common in terms of tetrameter patterns (Nekljudov 1991:25). Subsequently, C. R. Bawden (1924–2016) came to the conclusion that there is no particular verse structure in Mongolian epic. The most important feature is, rather, the lead rhyme of the initial verse of a unit, which is not necessarily followed in all attached verses. In terms of phonetics, moreover, Mongolian epic follows a law of vowel harmony, which is of great help in the formation of parallel patterns (Bawden 1980:285). Bawden’s approach to Mongolian epic prosody is not novel; yet his work attracted the attention of scholars from other fields but not many epic scholars. In general, the versification of Mongol folk poetry is relatively free in length. Allowing for functional and grammatical components, tetrameter is predominant; but the versification is not like that of Homeric epic or Serbo-Croatian narrative, with their relatively complex prosody that requires the singer to insert words in specific verse slots to meet the requirements of the rhythm. Field reports and the author’s own direct observations show that singers often leave out or forget certain cadences, sentence patterns, stanzas, and storylines; but there is no indication that they easily forget the basic performance units (nige ama-yin üge). Arimpil did not forget what he learned from famous jangarch, including their traditional singing patterns. There are 652 verses and 3,194 words in the text of Mighty-armed Sabar as he sang it, on the average four or five words for each verse.

Analysis of Mongolian epic prosody  171 Let us look at two randomly selected sequences, verses 70–79 and 450–459, from our text: ermen cagan hödege ejen ügei cagan büürüg-tü elesün sir_a tohui-du önücin gagca bar urgugsan naidag zandan modun -du odugad bodul bodugad ene modun-u següder ece hüreng haljan hülüg-i uyaju orhigsan nayan nigen esi tei

Endless white wilds, Masterless white desert, Ever-stretching golden dunes. Growing alone and solitary, The thin sandalwood has gone. Pondering Under the shade of the tree, A chestnut steed with a white forehead Is tied up there. 81-tuo long2

70

79

—~|—~|—~ —~|—~|—~|— —~|—~|—~|— — ~~ | — ~ | — ~ | — ~~ —~|—~|—~|—|—~ —~|—~ —| — ~ |~ | — ~~ — ~ | — ~ | — ~~ — ~ | — ~~ —~|—~|—~|— The second verse has five words; but in performance, jen ügei is treated as one word, ejegei. As tetrameter, the verse is thus rhythmically incomplete; it is trimeter plus one long syllable. When sung, the rhythmic pattern of the whole short sequence forms a pleasant cadence of alternating long and short syllables. The repeated incomplete catalectic verse pattern strengthens the rhythmic tension in a way that intensifies the story mood. In the following sequence, too, the basic meter is trochaic: sir_a-yin naimanminggan bagatur-ud ni irügegsen ir ni yaguman-du hürcü üjeged ügei gilagai sir_a bolud üldü-yi abugad baragun gar tagan sigüsün dusun dusutal_a athugad aldar noyan janggar tolugailagsan araja-yin naiman minggan bagatur-ud mini gurban dolug_a

Shila’s 8,000 warriors, 450 The blessed Sharp edge Has not made contact, Glittering buffed steel knife held In the right hand Held so tight as to bleed. Greatly famed Lord Jangar leads [His]8,0000 Araja warriors of three sevens. 459

172  Indigenous research on Mongolian oral poetics —~|—~|—~ —~|—~|—~~ —~|—~|— —~|—~|—~ —~|—~|—~|—~—|—~ — ~ | —| — ~ —~|—~|—~~|—~ —~|—~|—~|—~~ — ~ | —| — ~ | — ~ | — ~ ~ | — —~|—~ Arimpil appears to be wandering about on his horse chanting this ancient story about the grasslands without a care for verse irregularity. As in the previous sequence, however, every verse in his rendition makes use of the basic standard performance unit. Even the shortest, the two-feet final verse “three sevens” — ~ | — ~, is a commonly used fixed formula. So is the previous verse, almost the longest in the entire sequence: — ~ | —| — ~ | — ~ | — ~ ~ | —. One may wonder why the short verse “three sevens” cannot be combined with the similarly short verse that follows in the original text: horin nige honug tu ~~ | — ~ | — ~ | —

Twenty-one days

The reason is that this second formulaic phrase constitutes an incomplete fourfeet verse, and the two combined would be longer than the normal tetrameter. In the five other cases in which the formula “twenty-one days” appears, it also stands alone; for its combination with another verse in those instances, too, would exceed the normal verse length, such as: horin nige honug hasiyalagsan horin nige honug-un gajar yabugad

besieged for twenty-one days walked for twenty-one days

On the whole, the verse length of our sampling of Arimpil’s text is largely determined by the number of feet in the formulaic phrase and is generally tetrameter. This shows that the verse pattern of Jangar is determined by the basic structural unit—a phrase formula—and once again points to the ubiquity and force of oral formulae.

Head rhyme and alliteration The prosody of Mongol heroic epic has its own characteristics, different from the versification patterns, elaborate rhymes, strict canons, and obscure imagery of the occult poetry of Mongolian monks, which was strongly influenced by Indian and Tibetan culture (B.·Bürinbeki 1990:103–122). Of all the features of Mongolian

Analysis of Mongolian epic prosody  173 epic poetry, the head rhyme (tolugai holbulta), rhyme at the beginning of a verse, is the most prominent. In one type, AAA, the head rhyme is repeated in three continuous verses: ermen cagan hödege ejen ügei cagan büürüg-tü elesün sir_a tohui-du

Endless white wilds, Masterless white desert, Ever-stretching golden dunes.

70 71 72

In a similar neat pattern, AAAA, the head rhyme is repeated in four verses: dugtui dotur_a baihul_ dolbing sir_a-yin önggetei dugtui aca ban garhul_a dolugan naran-u gereltei

When the banner is in its holder, 220/243 It flashes with the glow of the moon. 221/244 When the banner is out of the holder, 222/245 It shines like seven suns. 223/246

From what we can see from other texts sung by Arimpil and other jangarch, the type AAAA is used everywhere in the singing tradition of Jangar. AAA appears less, but Arimpil likes to use it. In a more complex type of head rhyme, AABCCB, the initial rhyme of the main verse alternates with other rhymes: sinji önggetei nogtu-yi sili-yin hini mih_a-yi darugulugad notulagad has önggetei hajagar-i halh_a-yin hini mih_a-yi darugulun hajagarlagad

The beautiful reins— Put them over the neck, Fasten them around the neck. The jade-colored bridle— Put it on the face, Tighten it on the face.

377 378 379 380 381 382

Even more complex is the following 24-verse pattern, ABA​ABAA​AAAA​CADE​ AADF​AAAAA: asar ulagan honggur höbegün bosugad yabagan sagugad arban cagan üy_e bulgan-du ni arslang jagan hoyar-un hücün bürildün törügsen arban cagan himusu ban önggiged yabagan sagugsan cag-tu arban luu-yin dagun garugad arban luu-yin cahilgan gilbaljan baiba aguu yehe hücütei asar ulagan honggur höbegün helen baiba:

The hero giant red Hongor Gets up and squats. The joints of his ten fingers Have the strength of a lion or elephant. His ten fingers [He] rubs back and forth with his thumbs. When [he] squats down, [He] lets out the roar of ten dragons, [He] shines with radiance of ten dragons. With great strength, The hero giant Red Hongor Speaks as follows:

174  Indigenous research on Mongolian oral poetics arban üjüg ece minü arban cagan hurugu-du oljalagdaju iregsen arban üjüg-ün šolom-un hagad aldar noyan janggar-tu maguhai dotunu bolugsan bui? nadadu nige aman üge helel ügei yagahigad yabugsan bui? ged arban gurban soyug_a ni tacihinagad asar har_a nidü ni böriyegcireged asar ulagan honggur agurlan saguju baital_a

“From ten directions My ten fingers Captured The ghastly khans from ten directions. Can the greatly famed Lord Jangar Become a close friend with them? But you didn’t say a word to me. Why did you leave?” He said so. He gnashed his thirteen sharp teeth, His large black eyes flashed with fury. Giant red Hongor Is in a fit of anger.

The head rhyme of each verse is mainly A (seventeen times). The overall rhythm of the sequence seems quite irregular; but in terms of oral poetic artistry, it has the charm of a cyclical pattern composed of independent formulaic couplets connected by transitional verses that lack the A rhyme. The result is a relatively loose and natural rhythm commonly seen in Arimpil’s singing. Common in well-framed texts is the pattern AABB, with paired head rhymes, as in the following unit from the long sequence seen in the previous chapter: ama tai hümün amalaju bolusi ügei hele tei yagum_a ni helejü bolusi ügei

People who have mouths Dare not gossip [about him]. Creatures that have tongues Dare not talk about [him].

This succinct, neat kind of head rhyme pattern composed of two couplets is common in Mongolian epic. Another example seen in the previous chapter is: utulhul_a ulagan cilagun boldag cabcihul_a cagan cilagun boldag

When they cut him, He turns into red rock. When they cleave him, He turns into white rock.

In instances like the following, the head rhyme involves alliteration: hündü hüreng haljan hülüg-ün hurdun cüilem_e cag-tu hündü gartai sabar cingga cihirag bagatur cag-tu

[The] strong chestnut steed with a white spot on its forehead Runs as fast as the wind. Mighty-armed Sabar [Is a] strong and vigorous hero.

In this example, the first word hündü is a redundant expression caused by a slip of the tongue; but it initiates a pattern of alliteration with the following three words

Analysis of Mongolian epic prosody  175 in the verse and constitutes a head rhyme. The last verse also contains internal alliteration: ci. In the following example, there is head rhyme between the third and fourth verses; and three verses, including four words in the last verse, contain alliteration of the consonant b: eighty-one tuo long handle Abai’s giant yellow axe. At the right Is a little hero.

nayan nigen esi tei abai sir_a balta-yi baragun bey_e degen tüsigürdegsen bag_a bicihan bagatur bain_a

It is not common to have head rhyme in the last verses of a unit or to find the same alliteration in three consecutive verses or in four feet of the same verse. But the adept use of alliteration here demonstrates Arimpil’s skill at creating a beautiful and sonorous singing tone. The expression “a little hero” is a fixed formula that appears many times in other texts. So is “black armor” in the following verses: dalu-yin hini hogurundu tobcilagsan dalanhoyar hö huyag-un hini sinjirge-yi tasu cohigad

Tied at mid-shoulder, Seventy-two buckles of black armor Cut off.

The following samples from Arimpil’s Mighty-armed Sabar further illustrate his use of alliteration, with the number of times the phrase is used indicated: harbun haliyagsan cag-tu horin nige honug hüngginen helen hüreng haljan hülüg nayan nigen esitei ocin höhe haljan hülüg

2 times 5 times 3 times 7 times 5 times 6 times

Scout and lookout Twenty-one days Say loudly Chestnut steed with white spot on the forehead Eighty-one tuo long Grey horse with white spot on the forehead

End rhyme, internal rhyme, assonance End rhyme in Mongolian poetry is directly related to the syntactic structure of the Mongolian language. Many verses end with the declension endings of nouns or conjugation endings of verbs, so that it is easy to produce concordant end sounds. In the earlier AABCCB head rhyme sequence cited again here, the ending yi of the verb meaning “put” also creates an end rhyme for the verses “put the reins over the neck” (verse 378) and “put the bridle on the face” (381). In the same unit, the adverb gad creates an end rhyme for verses 379 and 382:

176  Indigenous research on Mongolian oral poetics sinji önggetei nogtu- yi sili-yin hini mih_ a-yi darugulugad nogtulagad has önggetei hajagar-i halh_a-yin hini mih_a-yi darugulun hajagarlagad

The beautiful reins— Put them over the neck, Fasten them around the neck. The jade-colored bridle— Put it on the face, Tighten it on the face.

377 378 379 380 381 382

The end rhyme in the following unit seen in the previous chapter is similar. The noun ending i forms a rhyme for verses 1 and 3, and the verb ending eg for verses 2 and 4. Arimpil and Eela Ovlaa use such a pattern often. irehü yeren yisün jil-i ailadcu mededeg önggeregsen yeren yisün jil-i tagaju mededeg

[Things] of the future 99 years Known by surmise. [Thing]s of the past 99 years Known by retrospect.

8 9 10 11

Post positions also sometimes, but rarely, create end rhyme, as here in verses 256 and 258: agta hülüg yamar cinegen hurdun bolba gejü güiceged arslang yamar cinegen manggus bolba gejü

As long as it is a steed, It can run fast. As long as it is a lion, It can be fierce.

256 257 258 259

It is said that The laws of phonetic harmony in Mongolian involve mainly vowels, in particular the relationship between vowels at the beginning or end of words. The laws determine which vowels can appear together in words and which cannot, and how are they arranged. The relationship between vowels is called “vowel harmony.” (Chinggeltei 1991:77) In our text, internal rhyme often fills out this harmony. It adds phonological beauty to a simple, two-word verse like “tegged ireged” (In that case I’ll come). Aurally appealing, too, but much less common is the triple rhyme an and eight instances of assonance of the vowel a in the verse “arban luu-yincahilgan gilbaljan baiba” (The radiance of ten dragons). Uncommon, too, is the internal rhyme pattern ABAB in the “nagaruban cagaruban” (This way or that way). In a more complex pattern, in the initial verse of the formulaic unit about Giant Red Hongor seen earlier, ban constitutes an internal rhyme, and hi and ni assonance; and in the second verse, sar and hara form a rhyme when sung:

Analysis of Mongolian epic prosody  177 dahin arban gurban soyug_a ni hargigad asar har_a nidü-ber-iyen

His thirteen teeth rattled, His jet-black eyes flashed with fury.

Through this preliminary analysis of rhythm and rhyme in Arimpil’s sung text, we see that for folk singers like him, there are various time-honored methods to shape the complex patterns of the great tradition of Mongolian language art. Singers have long been immersed in this cherished tradition and become outstanding inheritors of its superb skills at manipulating the language. Of course, Arimpil may not understand grammatical concepts of internal rhyme or assonance and may be confused by complicated terms of prosody. But through long study and singing practice, he has naturally acquired an experiential grasp of the ins-andouts of epic phonology and is at ease with the basic canons of Mongolian poetry. He uses the methods of Mongol prosody skillfully; and by analyzing his singing, we can make theoretical inductions about his style in doing so. The intricate patterns that he so skillfully uses enhance the basic units of Mongolian oral poetic formulae with extraordinary phonological beauty. It needs to be kept in mind, however, that his head rhymes, end rhymes, internal rhymes, alliteration, and assonance are all borne by the formulae. Even internal rhymes, which have very low requirements in terms of verse length, are all handled in terms of position in a formula. Any analysis of the prosody of Mongolian poetry and scholarly thinking about the oral poetics of Mongolian epic must thus be related to formulae.

Parallelism Poetic rhyme and rhythm are often clearly perceptible in parallel structures. Parallelism is common in Homeric epic, Chinese classic poetry, and the folk poetry of Dravidian Toda, Semitic, Uralic, Slavic, Turkic, and Mongol languages (Preminger, Brogan, and Warnke 1993:361). It is widespread in Mongolian epic, and it is regrettable that it has not attracted much attention from scholars in the past. As we have noted, however, Bawden has identified parallelism as a stylistic means to strengthen vowel harmony in Mongolian epic and other forms of Mongol poetry, not just between adjacent verses but sometimes between two or four-verse units. Inspired by W. Steinitz’ study on parallelism in Finnish Karelian folk tales, moreover, Poppe specifically studied this issue of parallelism in Mongolian narrative poems.3 The parallelism of Mongolian epic poetry differs from the highly developed rules of classical Chinese poetry, but it is extensive and diverse. Based once again on what we find in the relatively short canto from Jangar, as sung by Arimpil, we can make a basic assessment of the use, scope, and methods of parallelism prosody in Mongol oral folk poetry, beginning with his frequent use of antithetical parallel structures. Antithetical parallelism The following four-verse unit that we have seen earlier from Arimpil’s Mightyarmed Sabar presents a good example of the kind of antithetical parallelism that Arimpil likes to use:

178  Indigenous research on Mongolian oral poetics [Things] of the future 99 years He knows by surmise. [Things] of the past 99 years He knows by retrospect.

rehü yeren yisün jil-i ailadcu mededeg önggeregsen yeren yisün jil-i tagaju mededeg

8 9 10 11

In this regular set of parallel verses, verses 8 and 10 refer to time and are identical except for the attributives “future” and “past”; and verses 9 and 11 are identical except for the verbs stating the kinds of knowledge. Interlaced with end and internal rhyme, the set is marked by a harmonious circular rhythm. The following example from the previous chapter has the same pattern but in a stricter format: bagatur dalai abu tai bayan dalai eji tei

Father Batar Dalai, And [his] mother Bayin Dalai

17 18

With head rhyme, coordinated internal rhyme, approximate end rhyme, and exactly the same syntax, this typical antithetical parallel structure serves well to facilitate memory. Another example that we have seen before is similar but somewhat more elaborate and complex: ama tai hümün amalaju bolusi ügei hele tei yagum_a ni helejü bolusi ügei

People who have mouths Dare not gossip [about him]. Creatures that have tongues Dare not talk about [him].

The head rhymes AABB, the end rhyme CDED; and the syntactic components have a one-to-one correspondence in accord with the law of strict alignment. In some units, such as the following, the parallel arrangement is not as strict in terms of metrical pattern; but the meaning remains exactly parallel: solugai baragun ügei adhugad ir mörgün ügei cohigsan cag-tu

Hold it in left hand or right hand. Bash it with the edge or back of the axe.

The same is true for the following: ermen cagan hödege ejen ügei cagan büürüg-tü elesün sir_a tohui-du

Endless white wilds, Masterless white desert, Ever-stretching golden dunes.

70 71 72

254 255

Analysis of Mongolian epic prosody  179 These rhythmically uneven verses are not common in Arimpil’s singing, and the rhyming is also unusual—different head rhyme AAA and end rhyme CDD. However, there is a neat parallel correspondence among the end words—wilds, desert, and dunes—as well as the adjectives—white, white, golden—and the meaning of the initial words—endless, masterless, ever-stretching. The following verses do not appear as parallel when written but have the character of parallel arrangement when sung: malai darhan dabtan higsen högsin darhan güicin cohin higsen šagjamuni lama ni arimiyalagsan sir_a-yin naiman minggan bagatur-ud ni irügegsen

The Malay blacksmith hammers over and over again. The old blacksmith Hammers repeatedly. Śākyamun lama blesses Hila’s 8,000 Warriors blessed.

392 393 394 395 396 397

The first three verses form a set of parallel or mainly parallel meanings, with “Malay blacksmith” corresponding to “old blacksmith” and the rhyming higsen (hammers) in verses 392 and 394, though the adverbial modifiers differ. The structure of the three-verse unit, 395 to 397, is the same as that of the first three verses, with verses 395 and 397 being parallel in meaning. In the entire six verses, the rhymes derive from the verb tense ending in the characteristic feature of the Mongolian language noted earlier. The following four verses from the long sequence seen earlier are parallel in both meaning and the echoing rhyme: hündü hüreng haljan hülüg-ün hurdun cüilm_e cag-tu hündü gartai sabar cingg_a cihirag bagatur cag-tu

[The] strong chestnut steed with a white spot on its forehead Runs as fast as the wind. Mighty-armed Sabar [Is a] strong and vigorous hero.

50 51 52 53

The head and end rhymes are AAAB and CDED, respectively, with the word “strong” appearing twice. The alliterated head rhyme is lacking in the fourth verse; for that verse is a fixed formula that the singer finds no reason to change to meet the needs of rhyme. This is often seen in Mongolian epic singing. It shows that when forming parallel patterns, the singer will consider the balance between the verse forms, but when encountering a traditional formula, he lets rhyme harmony give way to parallel meaning. In short, the fixed formula will not be easily detached or changed. Experienced singers grasp the parallel structure of the following fixed formulae at first sight. The head rhyme is AABB, the end rhyme CDCD. The second and fourth verses are the same in the middle, but they have a slight change at the beginning:

180  Indigenous research on Mongolian oral poetics utulhul_a ulagan cilagun boldag cabcihul_a cagan cilagun boldag

When they cut him, He turns into red rock. Whey they cleave him, He turns into white rock.

In some cases, the antithetical parallelism is only relative. For example: solugai baragun ügei adhugad ir mörgün ügei cohigsan cag-tu

Hold it in left hand or right hand, Bash it with the edge or back of the axe.

254 255

The two verses here lack the symmetry of true parallelism and cannot replace each other. Asymmetrical, too, is the fourth verse of example about the chestnut steed with a white spot on its forehead, which lacks the head rhyme of the other three verses. Progressive parallelism In instances such as the following, parallel couplets constitute the simple kind of progressive movement common in folk song. “Three years old” becomes “four years old”; and “mother” changes to “father,” with other elements repeating cyclicwise and the end rhyme having the traditional pattern ABAB, all quite harmonious: gun_a orudag nasun du ni eji ni önggeregsen dün_e orudag nasun-du ni abu ni önggeregsen

When he was three years old, His mother died. When he was four years old, His father died.

26 27 28 29

The following is similar: dugtui dotur_a baihul-a_ dolbing sir_a-yin önggetei dugtui aca ban garhul-a dolugan naran-u gereltei

When the banner is in its holder, It flashes with the glow of the moon. When the banner is out of its holder, It shines like seven suns.

220/243 221/244 222/245 223/246

With the continuously alliterated head rhyme ABAB and end rhyme CDCD, these verses are parallel in both the rhythm and the progressive meaning. Parallel forms with progressive relationships like this are not uncommon in our examples, and they exemplify one of the important techniques of oral poetics. In the next example, the progressive relationship is not obvious; but the interlaced rhyming verses—AABB and CDCD—have a strong sense of progressive rhythmic circular movement: emnig gool-un modu bar eligehigsen hanggal gool-un modu bar habca higsen

[With] Eming River wood For the Elige [side of a saddle]. [With] Hanggal River wood For the Habcha [a part of a saddle].

383 384 385 386

Analysis of Mongolian epic prosody  181 The basic syntactic structure is stable, with the proper nouns varying in another common pattern of Mongolian epic poetry, the arrangement of names of rivers or mountains according to the needs of rhythm. Similar in rhyme is the following example, which also forms a fixed progressive parallel pattern that appears in other texts sung by Arimpil: nair-un ehin-dü naya honuju jirgalang-un ehin-dü jira honugad

The banquet continued For eighty days. The happiness continued For sixty days.

631 632 633 634

In end rhyme, the verbs honuju and honugad of verses 632 and 634 are the same words with different verb endings, an example of the relative randomness of oral song; but the difference does not involve accented syllables and so does not disrupt the rhythmic balance. This can often be observed in the singing of folk singers, as can the use of a formula like “seventy-day banquet” in the performances of Xinjiang jangarch. Other examples, with slight differences in the order, form, and wording, are: jira honog un jirgal hiju dala honog un danggarai hiju naya honog un nair hiju jirgal yin yehe dü jira honoju dalan honog un danggarai önggeregü jü nair un yehe dü naya honoju

Sixty days to share happiness, Seventy days to enjoy a banquet, Eighty days to celebrate a reunion.4 Sixty days of great happiness, Seventy days of a banquet, Eighty days of a big reunion.5

Different singers deal with the formulaic pattern in a more or less flexible manner, sometimes, like Arimpil, changing or reversing the order. Compound parallelism There is a more complex kind of parallelism consisting of couplets and formulae. We find this in the long sequence seen earlier about Giant Red Hongor: arbancagan üy_e bulgan-du ni arslang jagan hoyar-un hücün bürildün törügsen arbancagan himusu ban önggiged yabagan sagugsan cag-tu arban luu-yin dagun garugad arban luu-yin cahilgan gilbaljan baiba

The joints of his ten fingers Have the strength of a lion or elephant. His ten fingers [He]rubs back and forth with his thumbs. When [he]squats down, [He]makes the roar of ten dragons. [He]shines with the sheen of ten dragons.

This compound parallel set is not rigorous in rhythm, but it is unified by the head rhyme AABACAA and several components that form a parallel relationship with one other: the number ten, which unifies the whole; the images of fierce beasts— lions, elephants, dragons; and the power and majesty that these images highlight in the bearing of the hero. Some of the seemingly ordinary phrases in the set are

182  Indigenous research on Mongolian oral poetics in fact fixed formulae, such as the first three verses, which can be found in almost exactly the same pattern in the text of Eela Ovlaa. The parallel nature of the following instance likewise depends mainly on a combination of formulaic images: hündü gartai sabar bagatur asar ulagan honggur hoyar agta hülüg deger_e ben mordagad hüjügüü hüjügüü ben teberilceged hoyar agta hülüg ni hüjügüü hüjügüü-ber-iyen nige nige deger_e ben toglan nagadun talbigad

Mighty-armed Sabar, Giant Red Hongor, Two heroes riding their steeds. [They] put their arms around each other’s neck. Two fine steeds, [They] rub their necks against each other. One against the other, In playful sport.

The close relationship of the two heroes is reflected in their embracing of each other’s necks, which finds a parallel in the image of the two steeds playing around with their necks. According to Walter J. Ong, the American scholar known for his study of the differences between oral and written modes of communication, parallelism is a phenomenon that frequently appears in oral performance; and it has the basic function of aiding oral improvisation. This is not to deny that parallelism often appears in literary poetry as well; but it is not something that a bard like Arimpil uses to show off artistic dexterity. It is demanded by the very nature of oral poetic creativity. In Mongol oral poetry, it is a traditional technique of syntactic structure gradually refined over time to aid a singer’s on-site improvisation; and in most cases, as in our examples, it has the nature of formulaic creativity in that multiverse combinations of formulae constitute the parallel structures. This chapter originally appeared as Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing, chapter 5 (Nanning: Guangxi People’s Publishing House, 2000).

Notes 1 The transliteration here is based on Poppe’s original text, which is slightly different from that used in this book. 2 Tuo: Mongolian measure of length, two outstretched arms. 3 Der Parallelismus in der epischen Dichtung der mongolen, UAJB. Bd 30 (1958). The author is indebted to the Russian scholar S. J. Nekljudov for this information. 4 See page 249, Canto 15 of the Chiba University text referred to in the previous chapter. 5 See page 1085 of Canto 15.

Part IV

Comparative study of four epic traditions



17 Challenges in comparative oral epic (co-authored with Miles Foley)

In this chapter, we round out the main contents of this book with a comparative examination of fundamental issues in the study of oral epic as exemplified by four epic traditions widely separated in space and time. Two of them, Mongol and South Slavic epic, are or were recently still living and therefore observable by fieldworkers. The other two, ancient Greek and Old English epic traditions, are preserved only in written form. Although no comparative treatment can ever claim to be exhaustive or universal, we feel that these four witnesses represent considerable diversity and collectively offer a platform on which to forge a suitably nuanced model for oral epic. We organize our investigation in five sections, each keyed by a question that reflects an issue of contemporary importance. Looking at macro-structure, we ask in the first section: “What is a poem in oral epic traditions?” Scholars have argued from many different perspectives about the large-scale organization of oral epic: Are the smaller tales integral parts of a single whole, facets of a single gemstone, or simply individual narratives that collectors have assembled into anthologies based on literary models? On a slightly smaller scale, in the second section, we then consider the question “What is a typical scene or theme in oral epic traditions?” The focus here will be on recurrent passages, such as the arming of a hero or the boast before battle, that epic singers use as “large words” in their storytelling vocabulary. How are the large words configured in each tradition, and what kinds of flexibility do they show? From macro-structure, we then move to micro-structure. The third section sets the stage by asking “What is a poetic verse line in oral epic traditions?” This turns out to be a more difficult and open-ended inquiry than it might at first seem. Once we leave the narrow sample of Greco-Roman meter, for example, the verse unit draws its definition from features other than syllabic criteria and word-breaks. Additionally, there is the matter of the performed line, a voiced verse form, as distinguished from the kind of poetic line that we come to know spatially as a feature of a manuscript or printed page. The fourth section deals with the question “What is formula in an oral epic tradition?” It concentrates on the recurrent phrase, the smallest “word” or unit of utterance in the oral epic poet’s compositional lexicon. Basing our conclusions on the poetic verse that is the vehicle for the formula and working over the four quite different poetic traditions, we aim to show how this DOI: 10.4324/9781003258001-21

186  Comparative study of four epic traditions smallest increment of poetic structure can vary both within its own tradition and from one tradition to another. Finally, the fifth section addresses what may be the most important question of all: “What is register in oral epic poetry?” We ask about the two principal aspects of the specialized language of oral poetics: its overall structure—from cycle to story-pattern, typical scenes, and formulae—and the meaning that each of these structures bears for audiences attuned to their indigenous poetic traditions. In this section, we summarize the results of the first four questions and inquire into the traditional implications that each one has: How do we understand one canto of the Mongolian Jangar in the traditional context of the whole cycle? What bearing does a boast before battle have on subsequent events in the connotations that the scene has in Old English epic? Does Homer’s phrase “but come” mean anything in the poetic tradition beyond what can be established by reference to a dictionary? In short, what is the traditional referential significance of each “word”—from the smallest phrase to the whole performance—in an oral epic tradition? For Mongol oral epic, we have drawn from Arimpil’s performances of Jangar as transcribed and edited by his nephew Dr. Dawataya and published in 1999. Dawataya’s collection of his work is unique in at least two ways: it is the first collection of the Jangar Cycle from a single singer in China, and it is also the only anthology to be published without editorial interference. How Dawataya’s edition came about is a story in itself. As we have seen, a young girl had read to Arimpil from a highly edited anthology of Jangar epics, which included the versions of both Arimpil and other singers. He was disappointed that his versions had been altered, saying that they weren’t his. As a result, he asked Dawataya, the only person in the family who had received formal education, to transcribe and publish his performances just as he sang them, without any editorial changes. The South Slavic oral epic tradition is a vast conglomerate of regional and ethnic traditions that can be categorized as Muslim and Christian. The Balkans have seen enormous unrest and social change over the past 700–800 years; and the tradition reaches far beyond the former Yugoslavia to cognate epics in Bulgarian, Russian, and other Slavic languages. We frame our discussion by concentrating on the Muslim and Christian epics performed in the language once called SerboCroatian, which is today spoken by residents of Bosnia, Croatia, and Serbia. We briefly cover both types of South Slavic epic to give a true picture of the whole genre.1 The chief collection of Christian oral epics in Serbia was made mostly by Vuk Stefanović Karadžić, the 19th-century ethnographer, linguist, and reformer of the two alphabets. Via a network of amanuenses around the country, he gathered written transcriptions of singers’ performances, with special emphasis on the heroic deeds of the Serbian hero and mercenary Prince Marko and the events surrounding the Battle of Kosovo in 1389, in which the Serbs were defeated by the Ottoman Empire. Unlike virtually all other 19th-century fieldworkers, Karadžić did not often intervene during the editorial process; for the most part, he published what the singers sang without addition, subtraction, or emendation. We draw our examples of Christian epic from the second of his four volumes of Narodne srpske pjesme (The Oldest Heroic Songs; Karadžić 1841–62; 1988). We take our

Challenges in comparative oral epic  187 examples of Muslim epic from the Parry Collection of Oral Literature at Harvard University, drawing both from the published series Serbo-Croatian Heroic Songs, which so far includes performances from the Novi Pazar, Bijelo Polje, and Bihać areas, and from unpublished performances from the Stolac region in central Hercegovina.2 Turning to our two examples of originally oral but subsequently written epic, ancient Greek and Anglo-Saxon, we analyze poems that have come down to us only in that written form. While there is little question that these poems have strong and vital links to oral traditions, the fact that they have long been written down means that we cannot be confident about the precise nature of those links. The first whole Iliad manuscript, from the 10th century CE, records many variants; and partial texts preserved in fragmentary papyri offer additional witnesses to a variable, complex tradition. We here cite from the Oxford editions of the Iliad, the Odyssey, and related works (Munro and Allen 1969); and we use the Thesaurus Linguae Graecae, the searchable CD-ROM that contains original-language texts of all ancient Greek works, as the key to unlock Homer’s phraseology and narrative patterning. Like Homer’s works, the Old English poems stem from an oral tradition but have come to interact with written texts in unknown ways from no later than the last third of the 10th century and their modern rediscovery and edition in the 19th century. Approximately 32,000 verses of Anglo-Saxon poetry survive to our time; but of that corpus, only Beowulf (3,182 verses) and the fragmentary Waldere are truly epic. However, extensive narrative poems—especially saints’ lives or retellings of the Bible—abound; and the same kind of poetic verse line supports all the surviving poetry. In practice, this means that different genres interact easily with other genres, so that phrases and motifs migrate readily from one type of poem to another. We quote Beowulf and related poems from the standard collective edition, The Anglo-Saxon Poetic Records (Krapp and Dohbie 1931–53); and we employ the Bessinger–Smith concordance to that edition (Bessinger and Smith 1978) to establish the recurrency of phrase and narrative units.

What is a poem in oral epic traditions? Mongolian Jangar has always been the most prominent epic tradition in Mongolia, with Geser (Tibetan Gesar) the next most significant of many dozens of mostly interrelated oral epic forms. To determine the nature and dynamics of a poem within such a large and complex tradition, it is necessary to grasp its fundamental organization. As we have seen, the cantos (bölög) of Jangar are simultaneously two things: stories complete in themselves, and also parts of a larger cycle. One result of this is that most cantos can be combined in numerous different ways and various sequences without suggesting the passage of real time that we expect in novels. In Jangar, the heroes and other characters are always twenty-five years old; and the weather is always spring all the year round. Likewise, almost all cantos have

188  Comparative study of four epic traditions a happy ending, returning to the same state of affairs with which the story began, each beginning and ending with a feast. To speak of a poem in the Mongol oral epic tradition thus involves a complicated set of assumptions. In one sense, each canto is a poem: It tells a story with a beginning, a middle, and an end; it features one or more heroes; and it follows one of a limited number of story-patterns—battle, wedding, blood-brotherhood, life history—or a combination of these. But in another sense, each canto is itself incomplete since it presumes knowledge of other bölög. Even though the singer may not actually perform those other cantos, the actions and characters that they involve are traditionally associated with what he happens to be singing at the moment. This larger, composite network of cantos—both those actually sung and those that remain implicit—is the cycle, equivalent to an unpublished book that contains all the chapters or a constellation that takes shape from all its individual stars. Arimpil has pointed to this underlying network at various places in his repertoire; and Vladimirtsov has long ago observed: Jangar has a great internal similarity of action of its separate song-poems. They are linked not only by internal connections (one and the same Khan), but each of them constitutes a natural continuation and development of the preceding. Contradictions almost never arise; and the singer performing any given song calls others to mind, sketching them against a distant background. (Vladimirtsov 1923, 1983:17–18) In the Mongolian epic tradition, a “poem” is thus a slippery and contingent term; the tradition consists of performed parts of an implied, untextualized whole. A canto is a poem, to be sure, but the cycle to which it always refers is also a poem. Two perspectives—one story-based and the other scholarly—point to this interactive, part-and-whole structure of a poem in the Jangar tradition. First, the various cantos of the cycle seldom feature Jangar himself in the main heroic role; usually, another subsidiary hero, often Hongor, undertakes the actual task or adventure. Jangar is, however, always present; and the singer explains his distance from the main action by citing him as a patron deity to the warrior central to a particular story. Whatever the case, Jangar serves as the uniting focus for all of the disparate characters, the major link between and among their diversity; and the very naming of the entire cycle after him indicates his central prominence in the tradition as a whole. Correspondingly, scholars refer to the cantos and the cycle they study not as one or a series of items, and not by means of separate, carefully calibrated titles. Rather, they name the entire tradition by the hero’s name, Jangar, again placing him at center-stage, locating him at the heart of the network. Citing particular editions is one thing, but reference to the epic is quite another. South Slavic In some respects, the answer to “What is a poem in the South Slavic epic tradition?” sounds quite similar. Although each performance is to an extent complete in itself, each also remains contingent—dependent not only on an implied

Challenges in comparative oral epic  189 constellation of stories and characters but also on the particular circumstances of the individual performance, singer, and audience. That is, every poem is inherently linked to other poems; and what emerges in any single instance will always amount to one of myriad possible instances. We seek a stabilized text in this kind of natural multi-formality at our peril. In the case of the Christian tradition, in which the performance texts are relatively short (seldom exceeding 250 verses), poems behave like possible chapters in an as yet unwritten book. Adventures in different poems are loosely tied together, if at all, and direct reference to prior or parallel events or biographical specifics is rare. Rather, the parts become a whole by simple agglutination in audience experience, with the listener or reader drawing the connections between and among “chapters” that the tradition leaves inexplicit. Do we wonder why Prince Marko loses his temper with the Turkish tsar, defiantly placing his boots on the Muslim leader’s prayer-rug in the poem Marko Drinks Wine at Ramazan? We need to know that it is in Marko’s personality to be hotheaded, to act impulsively when challenged. We need to know that he takes great enjoyment in disobeying the ruler for whom he fights as an unwilling mercenary. None of this information may be contained in the particular poem we are reading, so depending exclusively on that performance text will leave us without evidence for the hero’s motivation. Only by bringing other implied “chapters” into consideration—by consulting the unspoken context attached to Prince Marko as a character who lives beyond any single poem—does his behavior make any sense. The longer Muslim epic tradition, which can run to 10,000 verses or more, also exists in an implied network of traditional narrative. More complete in themselves, the Muslim songs follow particular story-patterns—such as return, rescue, wedding, or siege of a city—but here, too, characterization and events are larger than any single performance text. Once again, our understanding of South Slavic epic will fall short if we focus only on a single instance at the expense of the much larger unspoken context. In the Muslim tradition, however, epics are not so much chapters in a single unpublished book as volumes from a shelf of books that belong to a particular series. We can describe each “book” as belonging to a huge narrative cycle only if we clearly understand that their primary reference and linkage is not to one another in a particular cycle but rather, to the tradition at large. Performances overlap with one another in structure and content contingent on the aims and day-to-day realities of singers and audiences as traditional storylines are brought into play. Mustajbey of the Lika Region will always be treacherous, but exactly what shape his treachery takes depends on which story-pattern or particular story underlies the narration. The protagonist of Tale of Orašac will always seem a lazy and selfish buffoon even as he proves indispensable to any Muslim army’s encounter with the Christian enemy; but we do not know of his unpromising character from any single poem. Ancient Greek With the Iliad and the Odyssey, we enter the realm of oral epics that today survive only in manuscript. Although both historical and internal evidence point toward an origin and currency in oral tradition, the exact relationship between the poems and

190  Comparative study of four epic traditions that tradition will always lie outside our certain knowledge. For that reason, the question “What is a poem in the ancient Greek oral tradition?” must be answered by piecing together different kinds of materials: the two major works themselves and the other poems, fragments, and summaries that reach us from more than two millennia ago. That evidence collectively indicates a thriving epic tradition from at least 1200 BCE, but with the Iliad and Odyssey being recorded in textual form no earlier than the invention of the Greek alphabet in about 775 BCE. Evidence of other poems about the Trojan War and its heroes can be found in references made by ancient authors to such epics as the Cypria, the Aethiopis, and the Nostoi, as well as quoted fragments of these poems. These lost works form the so-called Epic Cycle, which some scholars understand as a well-organized, interlocking series of accounts surrounding the Trojan War and others as a loose constellation of orally performed stories that were later recorded in writing (Davies 1989; Foley 1999a). Together with the poems of Hesiod, like Theogony and Works and Days, the Epic Cycle indicates that a great deal was going on in the ancient Greek epic tradition in addition to performances of the Iliad and Odyssey. What seems to have happened in the decades and centuries following the invention of writing is the gradual establishment of the Iliad and Odyssey as the canon of ancient Greek epic. That fossilization of a living oral tradition into only two texts has obscured our vision of a dynamic traditional process and led to some unfortunate assumptions. It was not until Parry’s initial investigations of the 1930s that Western scholarship began to realize that these were originally oral poems. From his analysis, the comparative field of oral poetics quickly emerged, affecting dozens of traditions around the world; and yet, the full impact of the Homeric epics’ genesis and context in an oral tradition was not to be felt for many years. Only recently have scholars come to grips with the reality that as oral or orally derived works, the Iliad and the Odyssey must be understood differently. They are individual poems, to be sure; but they also share traditional elements and strategies such as formulae, typical scenes, and narrative patterns both with each other and with poems like the Homeric Hymns (Foley 1995:136–180 and 1999b:115–237). Therefore, trying to understand the nature of an individual poem in the ancient Greek tradition is similar to attempting to project the nature of prehistoric life from fossil remains. The best we can do is to be aware of the dynamics and expressive usefulness of the formulaic features and narrative patterns that we can recognize. Analogies from living oral traditions, such as the Mongolian and South Slavic, are crucially important in this effort.

Old English The question “What is a poem?” provokes a series of answers in relation to Old English, or Anglo-Saxon, poetry. Firstly, these poems derive from an oral tradition brought to England by Germanic settlers from about 450 CE; but the exact relationship between the manuscripts that survive and this long-lost oral tradition must remain uncertain. Secondly, much more than in the ancient Greek tradition, the Old

Challenges in comparative oral epic  191 English materials are very different from one another in genre. Within the 32,000 extant verses, we find that approximately 10% are epic, chiefly Beowulf, with the other 90% consisting of elegies, charms, riddles, lives of saints, histories, biblical stories, and other forms. Focusing on epic, we find that Beowulf has no direct parallels in Old English, although it shares a mythology and poetic style with Germanic poems in Medieval languages such as Old Norse, Old Saxon, and Old and Middle High German. From all of this disparate evidence, it seems that Beowulf was part of a pan-Germanic oral tradition of which very little actually survives to us. Even though the epic genre and story of Beowulf are rarely paralleled, the Old English poetic style is well attested. Unlike many other traditions, all AngloSaxon poems, no matter what their genre, share the same meter and to some extent, the same poetic diction. Thus, traditional units such as typical scenes and formulae occur across the spectrum from one narrative form to another. For example, Beowulf and the biblical stories share the theme of Sea Voyage as well as poetic formulae for ships, heroes, and so forth. Even non-narrative genres such as charms and riddles draw on the same phraseology as epic. As a result, Beowulf is suspended in a network of formulaic and narrative patterns that can be observed in other genres, all depending on the same traditional register. As we saw in relation to Mongolian, South Slavic, and ancient Greek works, Old English poems are both complete in themselves and yet idiomatically linked to a much larger network of structure and implications.

What is a typical scene in an oral poetic tradition? Mongolian Lord has defined thematic typical scenes as multiform narrative units, “groups of ideas regularly used in telling a tale in the formulaic style of traditional song” (Lord 1960:68). In the Oral Formulaic model, these are flexible building blocks of narrative that can be molded to suit the immediate narrative context. The same scene serves many singers and many songs as a “large word” or unit of utterance. Indeed, as we have seen, South Slavic epic singers and their counterparts in other traditions think of “words” as just such integral units, not as lexical entries in dictionaries. Simply put, the typical scene is a story-byte, a tale-telling increment. Many cantos in Mongolian epic follow a fairly stable series of events, starting from a stock beginning with citation of the time as a Golden Age and the famous place in which the action is set. The canto then continues with a description of the hero’s wondrous horse and his aide or comrade in arms before a subsequent threat emerges in one of many forms—attacking monsters, invading armies, and the like. Depending on the particular story-pattern, a maiden may be involved; and if she is captured and the homeland destroyed, the hero will be called upon to rescue her and restore the land. Fighting against a monster or another hero is often involved, and victory over a human adversary may bring that adversary over to the hero’s side. A wedding can follow, and the story comes to a close with the hero’s return home.

192  Comparative study of four epic traditions No matter how an individual canto of Jangar proceeds, it regularly begins with a thematic typical scene that we can call the Palace scene. Based on available evidence, the only exception to this rule occurs when the information contained in this unit is assumed already to be part of the audience’s prior knowledge. The Palace scene consists of Jangar together with his 6,012 warriors drinking and singing boisterously, with the protagonist himself on a bench supported by fortyfour legs, his face shining like the full moon. In some versions, this scene may be very brief; in others, the singer may give elaborate details about the grandeur and opulence of the dining hall—its great height, with fifteen stories reaching to the sky, and its decoration with animal skins, gems, gold, and silver. Other variants can include a detailed physical description of Jangar, an account of his early childhood deprivations or other aspects of his life history, or a description of his peerless steed, sometimes with an account of its life as well. The length and elaborateness of the Palace scene, or its brevity and starkness, depend on the singer’s ingenuity, the performance situation, and the audience’s receptivity. In one version of Arimpil, as we have seen: arban tabun dabhur altan čarlig bambalai dotur-a aldar noyan Jangar ni tolugailagad araja-yin naiman minggan bagatur-udni dagulaldun nairlaju baita-a

Inside the fifteen-storied Golden shining palace, Greatly famed Jangar leads 8,000 Araja warriors, Singing and drinking.

(Dawataya 1999:64) In another version: arubey-e-yi ni arslang-unsoyug-a-bar sihan daramalan boshagsan emün-e bey-e-yi ni ölü manghan bugu jaganhoyar-unsoyuga-bar sihan daramalan boshagsan jegünbey-e-yini usunsil-iyer önggelen boshagsan baragun bey-e-yi ni badmarag-a Erdeničilagu-bar sihan daramalan boshagsan gadanah-i dörbenönčüg-ini gal sil uglan barigsan

Behind [the palace], Lion’s canine teeth, Face upward. At the front, [Teeth of] far-off Gobi deer And elephant tusks, Face upward. At the east side, Water-mirrors Are set, radiant with color. At the west side, Precious rubies Are placed facing upward. At the four outer corners, Fire-mirrors were built.

(Dawataya 1999:18–19)

Challenges in comparative oral epic  193 The idiomatic nature of this beginning scene provides evidence of how singer and audience conceive of an individual canto as an entire world in itself—a single unit filled out in their minds with a larger implied background—not as a fragmented, freestanding folktale scene. Within the Jangar Cycle, one and the same typical scene can vary substantially from one singer to another, with differences in characters, description, and minute details that depend on the particular bard’s version of the given canto. For example, the two versions of The Wedding of Hongor as sung by Bosugomji and Li Purbai are roughly similar; but their length and constituent scenes are clearly different. On the macro level, differences from song to song vary as a scene is linked to a long or short series of narrative patterns (Rinchindorji 1999: 238–243). However, the formulaic language in which typical scenes are expressed differs more from one singer to another than from one song to another. In singing the typical scene “Encountering a stranger,” Arimpil uses “nigur tala-ban gal tai / nidün tala-ban čog tai” (with fire on your cheeks / with embers in your eyes). In singing the epic Gants Modon Honogtoi, another bard uses the dialectal variant “nüürendee galtai / nüdendee tsogtoi / shilendee ööhtei / shilbendee chömögtei” (with fire in your face / with embers in your eyes / with fat on the nape of your neck / with marrow bones in your shin). Individual singers may also use slightly different diction as they work through different parts of their repertoire, but different singers depart from one another in many more ways, showing how dialect differences can enliven epic language. Within any singer’s inventory, certain typical scenes are linked to specific story-patterns. For example, the thematic episode of the hero’s transformation into a pauper and his proud steed into a shabby, two-year-old colt is commonly associated with cantos following the overall storyline of winning a maiden. The scene thus helps predict what will happen, providing a kind of map for the whole epic journey. To take another example, a messenger arriving at a feast when all present are singing and drinking indicates that eventually, the story will turn toward battle as its major subject. Or consider the typical scene of the hero or his wife having a nightmare, which betokens some sort of threat, often an invasion and eventually, a battle. In all these cases, the typical scene is more than a cipher, more than an item or tectonic strategy; it foretells future events and reveals the direction of the story. South Slavic The preceding sketch of a typical scene corresponds closely to the realities of South Slavic epic, which of course furnished the original data for the Oral Formulaic model. As with Mongolian narrative, we find the singer, or guslar, thinking and expressing himself in terms of what is called reči, “words,” by which is meant utterance units, rather than typographically defined words. As we have seen earlier in this book, for the South Slavic epic singer, “words” are never as small and partial as printed units, whether Chinese characters or European groups

194  Comparative study of four epic traditions of letters. Instead, reči are storytelling increments, as small as a phrase or as large as an action-centered typical scene, an entire performance, or the story-pattern that underlies it. The guslar composes in “words,” not words (Foley 2002:11–21). In South Slavic epic, typical scenes vary widely in length, detail, and flexibility. As we might expect, the shorter compass of the Christian songs means that fewer such scenes are found within the varied scope of a text-performance. Since the expressive style of these narratives is spare and direct, such scenes, which can exceed hundreds of verses, are not as useful in the composition process. When they do occur, the shorter format makes for a briefer unit and somewhat restricts song-to-song variability. To put it proverbially, Christian South Slavic epic is more the product of verse-length and performance-length “words” than scenes. That is emphatically not the case, however, with Muslim epic, whose style depends on narrative units that belong not to one or a few but to many songs, taking numerous different forms according to the story-pattern, specific story, singer, and individual performance (Foley 1990:278–328). If a guslar is singing a Story of Return, for example, he must know how to use typical-scene “words” such as “Shouting in prison” and “Readying the horse.” The first of these tells how the long-lost hero, separated from his family and his people for years, laments so loudly that he keeps the entire town awake all night. His shrieking prevents his captor’s baby son from sleeping and thereby endangers the boy’s health; if nothing is done to quiet the prisoner, the captor’s wife warns, the infant will die, and the royal line will be extinguished. No matter how the particular story may go, however the prisoner, captor, and others happen to be named, and regardless of exactly how many nights the intolerable lamenting goes on, the general shape of the scene is roughly the same. Beneath superficial variations is the same “word.” Likewise with “Readying the horse,” a description of how a hero prepares his or her horse for the inevitable journey is found in Return Stories and other kinds of Muslim epics as well. Here, the action usually starts with the hero running down to the stable and leading the steed out into the courtyard, where an extensive grooming process takes place. Just how extensively the guslar draws out the scene depends on a host of situational factors, but common elements include washing the horse’s coat and rubbing him down with a goatskin pouch. After these preliminaries, the singer continues with a description of the blanket, saddle, bridle, and reins—usually in that order—before closing the unit with an account of the animal’s ability to prance around the courtyard without a rider to direct its actions. In a performance of The Wedding of Mustajbey’s Son Bećirbey, the singer Halil Bajgorić added a simile, comparing the riderless horse to a carefree young shepherdess.3 In any case, whatever form the typical scene may take at the hands of this or that singer on this or that occasion, as we shall see in the discussion of “What is the register in an oral epic tradition?”, its structure and content have idiomatic connotations in South Slavic epic. Without idiomatic elements of the traditional epic vocabulary, a guslar cannot compose his song. Nor can an audience fully understand it.

Challenges in comparative oral epic  195 Ancient Greek Typical scenes in Homeric epic include such recurrent passages as Feasting, Assembly, Arming the Hero, Lament, and many more. Although we are limited to about 28,000 lines in the surviving corpus of the Iliad and the Odyssey, that sample is sufficient to observe how some of these scenes work. The Feast scene, for example, occurs no fewer than thirty-five times throughout the two poems, thirty-two in the Odyssey (Foley 1999b:171–187). As a “large word” in the poet’s compositional vocabulary, each instance includes four variable elements: Seating the guest(s), serving food and drink, a verse or two marking feasters’ satisfaction and end of the meal, and thoughts about an impending problem. Whatever the location, identity of the host and guest(s), or particular moment in the story, these elements are always present. Like other traditional units, the Feast scene is a nexus for drawing together specific details in an expectable frame of reference focused on the moment at hand. For its part, the typical scene Lament occurs six times in the Iliad (Foley 1991:168–174). In this pattern, a woman is mourning the death of a fallen warrior, either her husband, her son, or someone with whom she has had a deep and longstanding relationship. The basic sequence has three parts: address to the fallen hero, a recounting of their personal history and the consequences of his death for those left behind, and a final, intimate address to the hero. This form underlies passages as different as Andromache’s and Helen’s laments for the fallen Hector and Briseis’ mourning for the slain Patroclus. Since each sorrowing person’s relationship to the hero is unique, this typical scene must leave ample room for variation. At the same time, it must be cohesive and focused enough to be useful compositionally and expressively. It and the Feast scene are part of the traditional epic register, and as such, have idiomatic connotations whenever they appear. Old English In Anglo-Saxon narrative poetry, thematic typical scenes work somewhat differently than in the other three bodies of poetry. Although scenes like Sea Voyage, Exile, and Beasts of Battle reveal a consistent sequence of ideas, their individual instances do not correspond with one another as closely as in other oral epic traditions. What varies is the line-to-line texture of the instances, a phenomenon that results from the differing nature of formulaic structure in Old English as described in the discussion of “What is a poetic line verse in an oral epic tradition?” and “What is a formula in an oral poetic tradition?”. One of the most thoroughly studied scenes in Old English is that of Exile (Foley 1990:330–331). Critics have shown that it consists of four basic elements: status, deprivation, state of mind, and movement in or into exile. The narrative pattern begins with a description of a character’s loss of social and/or familial status. Separation from the network of society and family in Anglo-Saxon culture was a crippling condition, one that led to complete loss of identity. As presented in the poems, nothing worse can happen to a person. The typical scene then continues with a statement of what the character has left behind, often using the

196  Comparative study of four epic traditions Anglo-Saxon formulaic expression “X bidæled” (deprived of X) or an equivalent as a marker for this stage. The third element focuses on the exiled figure’s state of mind—full of sorrow and hopelessness. Finally, the thematic scene closes with a note on the direction of the story’s further movement (always negative): The protagonist’s woes will deepen or at least fail to improve. The scene is commonly applied to a wide range of characters. The protagonists in the two lyric poems The Seafarer and The Wanderer are exiles, as is Grendel in Beowulf. Like so many other traditional structures in Anglo-Saxon oral-derived poetry, the Exile theme crosses generic boundaries, occurring in both epic and non-epic genres. The Sea Voyage offers us another example of a typical scene, one found chiefly in Beowulf (Foley 1990:336–344). We can describe it as a sequence of five elements, again with very flexible line-to-line composition: The hero leads his men to the ship; the moored ship awaits them; the men board the ship, carrying treasure; the ship departs, sails, and arrives at its destination; the ship is moored, and the men encounter a coastal guard in the new land. This theme occurs twice in Beowulf, once when the hero journeys from his home in Geatland to Hrothgar’s territory in Denmark and again on his return trip. Each time, the five-part outline provides the basic structure for the action, tracing the hero’s voyage across the sea via a traditional, recognizable series of actions. A third instance seems to occur at the very beginning of the poem with the ship-burial of Scyld Scefing, a legendary hero long before Beowulf’s era. In narrating the Anglo-Saxon funeral rite of burying the hero in a sailing ship with his treasures, the poet of Beowulf appears to be using the same typical scene with appropriate modifications. In discussing “What is register in an oral epic tradition?”, we shall see that this is more than just a convenient ploy designed to take advantage of a ready-made compositional structure; it marks an astute artistic strategy.

What is a poetic line verse in an oral epic tradition? Mongolian Mongolian versification is a subject that has not been addressed by either native or foreign scholars from the perspective of comparative oral epic prosody. We need to start by observing that performances sometimes include both poetry and prose in a hybrid medium that is found in other traditions as well (Harris and Reichl 1997). As a general rule, however, the more capable and experienced singers compose entirely or almost entirely in poetry, while the more amateurish and less experienced performers depend to a greater degree on prose. As for the texture of individual songs, traditional elements like typical scenes tend to be sung as poetry regardless of who performs them, while prose emerges between these units as the bard moves his narrative forward. Within the tradition itself, poetry is understood as the original medium. As the bard Jonggarab once observed, “In olden days, singers sang their songs; nowadays they speak their stories” (Chao Gejin 2000:312). The poetic verse of Mongolian epic is marked in a number of ways, some optional, some obligatory. Metrical rhythm is a required constituent of each verse;

Challenges in comparative oral epic  197 and so, too, are the sung melodies and accompanying musical instrument of a performance, though these are usually not indicated in the text itself. For Jangar, the most common accompanying instrument is the tobshuur. As in South Slavic oral epic, to remove the musical context is to delete a defining dimension of what a poetic verse is. Nonetheless, given the textual context of our discussion, we here deal only with characteristics that can be presented on the written page. Hand in hand with the music goes another feature that recurs in performed poetic verse, the vocal pause between verses that defines the boundaries of the unit. A Mongolian epic singer makes a clear break between verses by hesitating briefly before continuing with the next verse. As with Native American poetry, an important measure of the integral verse structure is thus the breath-group, the spoken unit delimited by pauses in breathing. The succession of words that the singer isolates and foregrounds in this way coincides with the succession of words as marked by the vocal melody and instrumental accompaniment, so that the breath-group and musical unit are best understood as different but coordinated manifestations of the same reality. Focusing on the text itself, we find that most verses of Jangar exhibit the parallelism common in oral epic traditions, including the others treated in this chapter. The side-by-side, paratactic verse structure leads to a lack of enjambement and more generally, to an overall organization in which a verse is usually complete in itself. It may expand into the next verse unit by apposition, enlargement, coordinate structures, etc.; but there is an integrity to each verse unit that marks it as expressively complete in itself. A verse of Mongolian epic poetry also depends on characteristics that recur regularly, though not necessarily in every verse. Foremost among these secondlevel features is sound-patterning, which has a number of types, most prominent being the head rhyme discussed in the previous chapter. For example, ama-tai hümün amalaju bolusi ügei hele-tei yagum-a helejü bolusi ügei hümün-ü način hündü gar-tai Sabar

People who have mouths Dare not gossip [about him]. Creatures that have tongues Dare not talk [about him]. Eagle among the masses, Mighty-armed Sabar.

The acoustic link of head rhyme connects verses in a series while apparently acting as a mnemonic aid. Verses that are related in this kind of latticework are more stable and less subject to change from one performance to another. End rhyme (tail rhyme) also helps to bind successive verses together. utulhul-a ulagan čilagun boldag čabčihul-a čagan čilagun boldag

When they cut him, He turns into red rock. When they cleave him, He turns into white rock.

198  Comparative study of four epic traditions Sound-patterning can also take the form of alliteration, with as many as four alliterated syllables per verse included in the matrix: “Baga bičihan bagatur bayin-a” (Little young hero to be). Internal rhyme and assonance also bind a verse together: “Arban luu-yin čahilgan gilbaljan bayiba” (Ten dragons’ lighting flash). Assonance builds on the natural language characteristic of vowel harmony in Mongolian. Indeed, it is an obvious, but sometimes overlooked, rule of thumb for all the oral epic traditions we examine here that the structure and phraseology of a poetic verse depend directly on the nature of the given language. For this reason, universal definitions of poetic verse and phrase are of limited usefulness. Some features familiar from Western, especially Greco-Roman, poetry do not apply to Mongol verse. In the first place, there is no syllabic constraint to speak of. In our sample, a verse can range in length from four to eleven syllables, with the different configurations sung according to the same rhythm and melody. Secondly, there is no ictus or metrical stress pattern that underlies Western trochees and iambs. Mongolian epic follows the natural language stress of initially accented words. This produces the impression of trochees, but that is an illusion. Finally, the word order in a Mongolian epic verse line is as a rule no different from that in everyday speech. Since Mongolian is chiefly an analytic language, dependent to a substantial degree on word-sequence rather than inflection, it tends to maintain a regular order. In any case, what matters about the verse prosody of Jangar is parallelism, musicality, breath-grouping, and sound-patterning. South Slavic The epic poems of Slavic guslari depend principally on epski deseterac, their decasyllabic epic verse form (Foley 1990:85–106). As the name implies, a verse consists of ten syllables; but that is not its only or most important dimension. The verse form also has internal organization and performance characteristics that are important for understanding its symbiosis with traditional phraseology. Internally, deseterac consist of two terse utterances, or cola, of four and six syllables each: rano rani Djedjelez Alija i Alija, carevagazija

Djerdjelez Alija arose early, Even Alija, the tsar’s hero.

A hiatus always occurs between the fourth and fifth syllables, and normally, the two cola that are formed by that break are complete grammatical units. The two cola can combine to shape a variety of meaningful utterances, but each is to some extent independent. For instance, rano rani lički Mustajbeže i Alija, više Sarajeva

Mustajbey of the Lika arose early, Even Alija, above the city of Sarajevo.

Challenges in comparative oral epic  199 South Slavic verse has other characteristics as well. Originally, scholars thought that deseterac was made up of five trochees, but this has proved to be a misapplication of Greco-Roman prosody. In South Slavic verse, stress occurs primarily on syllables 3 and 9, and somewhat less regularly on syllables 1 and 5, with syllable 7 usually unstressed. Between syllables 3–4 and 9–10, word-break is prohibited; and a word serves as a bridging zeugma, sometimes with internal rhyme. For example: ubećara nema hizmećara

For a bachelor there is no maidservant.

On syllables 3 and 9, -ćar- bears the stress of the verse and forms an internal rhyme. The stress is a necessary and defining feature of the South Slavic epic verse, as is the lack of a word-break between 3–4 and 9–10. The internal rhyme is an optional feature that occurs approximately once or twice every fifty lines and especially in proverbs. Along with formal characteristics, deseterac also has a musical feature. Most epics in the Muslim tradition and many in the Christian tradition are sung to the accompaniment of the single-stringed, bowed, lute-shaped musical instrument called gusle. Singers use it not just as ornamental accompaniment but to help voice their performance. The verse melody follows the pattern of the instrumental melody as a crucial dimension of the epic. In most instances, a bard begins a ten-syllable verse with I (and) or Tad (then); but at times, he begins with a vocal pause, allowing the instrument alone to mark off the first one or two syllables. He then starts singing only with the second or third syllable. Here is an example from Halil Bajgorić’s The Wedding of Mustajbey’s Son Bećirbey: [X]  zavika Djuliću Nuhane

[Pause] Djuliću Nuhane began to shout.

This is a difficult for persons trained to understand poetic verse as a textual phenomenon in which the words are primary; but as in a Mongolian performance, non-textualized aspects of a verse often play a crucial role that written texts are not well equipped to represent. Ancient Greek The poetic verse of Homeric epic is a very complex instrument, but we can gain a basic understanding of how it works by considering both its external and internal structure (Foley 1990:68–84). From an external perspective, it is dactylic hexameter, that is, six feet of dactyls (–⏑⏑) with occasional substitution of spondees (– –), measured not accentually but quantitatively (long/short). A verse can have a different number of syllables from twelve to seventeen, with the fifth foot usually a dactyl and the final, sixth foot always a spondee. No contemporary metrical or musical notation survives, so we must be content with this limited level of textual analysis.

200  Comparative study of four epic traditions The internal metrics is much better known. A verse of Greek epic is composed of four unequal cola, each of which is of common length in the Homeric poems, forming the metrical basis for verse structure. Whereas the external measure of six feet does not correlate with the system of traditional diction, the internal dynamics of four cola helps to explain how aoidoi (epic singers) constructed their verses. This simplified diagram shows how the four cola work: –⏑⏑/ – /⏑⏑– / ⏑/ ⏑– / ⏑⏑/ –⏑⏑–– A1A2 B1 B2 C1 C2 The slashes (/) mark the three possible word-breaks, each of which must occur at one of two possible positions—A1 or A2 for the first, B1 or B2 for the second, and C1 or C2 for the third—yielding three divisions and so, four cola for every verse. The first colon extends from the beginning of the verse to A1 orA2, the second from A1 orA2 to B1 or B2, and so forth. In its combination of fixed but flexible structure, the Homeric verse acts as a functional partner of ancient Greek epic phraseology, the subject of “What is a formula in an oral poetic tradition?”. Old English The poetic verse of Beowulf is quite unlike these other three epic traditions, and in this difference, we can see how regularity can take different forms in the various canons that underlie oral and oral-derived poetic traditions (Foley 1990:106–119). First and foremost, Anglo-Saxon verse does not depend on syllable count. The verses in Beowulf have between eight and sixteen syllables with no systemic pattern of length. This means that the kind of cola in ancient Greek and South Slavic poetry are irrelevant and that any feature deriving from regular syllabic prosody is incongruous. Also, although the use of an accompanying musical instrument is attested to by references in the poems themselves as well as by archaeological evidence, we know nothing about the role of either vocal or instrumental music in the performance of this poetic tradition. What, then, characterizes Old English verse? The two primary features are alliteration and stress. As with Mongolian prosody, an Anglo-Saxon verse unit is defined in part by a sequence of matched initial sounds, as in the following examples from Beowulf, verses 51–54 (alliterated sounds underlined): -secgan tosoðe, hæleð under heofenum, Ða wæs on burgum -leof leodcyning

selerædende, hwa Þæmhlæste onfeng. Beowulf Scyldinga, longe Þrage.

[s] [h] [b] [I]

Challenges in comparative oral epic  201 Truth to tell, hall-counselors, Heroes under the heavens, Then in the strongholds was Beloved people-king

who received that burden. Beowulf of the Scyldings, for a long time.

Each verse divides into two halves, each containing a varying number of syllables; and the scope includes at least one matching alliterative sound in each half verse, often two in the first half verse. Without such alliteration, a verse is flawed; it is an absolute requirement of the prosody. The alliterative meter functions on two levels, organized in both whole and half verses; and it has an active role in the composition and systematic usage of formulae. The other chief feature of the alliterative meter is the ictus stress. Germanic languages are all accentual, and Old English is no exception, with four major accents in each verse in accord with favored accentual patterns. This regularity contrasts with the irregularity of syllable count, and it provides an organization based on stress rather than sequences of syllables or cola. For example, the first of the verses quoted here features stresses on the alliterated syllables and on –rœd-, the root of –rœdende: /      /       / / secgan to  soðe    selerædende As with typical scenes, alliteration and stress-patterns characterize not only epic but also non-epic Anglo-Saxon poetry. The structure of Old English verse leads to frequent parataxis, apposition, and enjambement. Poets have the opportunity to continue their thought beyond the limits of a single verse line, often adding one phrase to another in multi-verse sequences of independently viable, highly flexible increments. This unique set of rules allows much more variation than Mongol, South Slavic, or ancient Greek verse. This relative freedom from encapsulation has significant implications for the kind of formulaic structure that can arise in Anglo-Saxon poetry.

What is a formula in an oral poetic tradition? Mongolian Perhaps the most direct way to begin discussion of formula and its identity in Mongolian epic is to quote Milman Parry’s definition focused on the “large words” of the South Slavic tradition: “a group of words regularly used under the same metrical conditions to express a given essential idea” (Parry 1971:272). Epithet formulae in Jangar range from one to five line verses, and multiple formulae can

202  Comparative study of four epic traditions be added to produce a traditional sequence of attributes extending to six or more verses with a flexibility that goes beyond that of the other three epic traditions we examine. For a single-verse formula, we have: asar ulagan Honggur

Giant Red Hongor

For a two-verse couplet: aguu yehehüčü tei asa rulagan Honggur

Very great power, Giant Red Hongor.

Five-verse version: agčim-unjagua-r du arban gurba hubildag amin bey-edüni ügei aguu yehehüčü tei asuru ulagan Honggur

In the blink of the eye, He has thirteen transformations. His spirit leaves his body. Very great power, Giant Red Hongor.

These formulae afford the singer great flexibility in performance; and despite their obvious differences, they share some core features. For one thing, the name Hongor occurs at the end of the last verse regardless of how many verses the formula has, in a resolution common in Mongol oral epic. For another, each of the three acts as a unit, no matter what the length, as a traditional adjectival phrase identifying the hero. We have seen that in addition to epithetic formulae, many other traditional phrases are found throughout Mongolian epic, such as the single-verse introduction of an important speech: “hüngginen helen baib-a”ü (He/she says loudly). More than simply marking the onset of direct discourse, this formula alerts the audience to the beginning of a forceful speech that contains a warning or prediction about future events. A formula whose literal meaning is similar, but whose idiomatic usage is quite different, is “čüngginen garb-a” (It comes out loudly). This phrase regularly occurs only when Jangar is in serious danger and is calling out for assistance. The voice always enters a person’s left ear and exits with great loudness through the right ear, sometimes Hongor’s own ears, sometimes those of his wife, who then awakes Hongor and tells him of danger. Other formulae may group around such actions, as when Hongor’s wife is set the task of waking a husband whose sleep was brought on by excessive toasting with his 6,000 comrades. In the epic tradition, such deep sleep is “heroic sleep” from which it is difficult to be roused, commonly expressed with the double simile: suhai metü ulaigad sur metü sunugad

As red as willow, As soft as a leather thong.

Challenges in comparative oral epic  203 To pierce his deep slumber, the wife employs a traditional strategy: Tugging at his braided šaluu: malmagar hara šaluu-yini hoisi-ban tatan gurba dahigad

The soft, dark black šaluu, She pulls back, Three times repeating.

“Tugging the braids” is a local expression, referring to the old Oirat custom of braiding the hair, beginning at age five with another added every year, so that a champion like Hongor will have many such braids. As is common with Mongolian phraseology, the final verse here is optional. How the action then develops depends on each story; but it eventually leads to readying the horse, preparing for battle, and Hongor’s riding off to Jangar’s aid. In short, a Mongolian epic formula acts like a “large word.” It may be as short as a single verse line or as long as five verses, and one formula may follow another to yield a compound series of yet greater length. The compounding is always at the discretion of the singer, who suits his performance to the song, the audience, and the moment. As in this example, formulaic phrases are commonly part of the idiomatic register. What the singer controls by their usage thus includes the traditional implications of the formula’s meaning. South Slavic The smallest of South Slavic epic large words consists of a colon of four or six syllables, but many such formulae extend to ten. Guslari use them as a flexible vocabulary to fit into the particular story they wish to tell. As with Mongolian epic, many are traditional epithet phrases that serve to identify the hero. In the following four examples, the same six-syllable reč (“word”) formula fills out the second colon of a decasyllabic verse in combination with the varying initial cola of four syllables to yield quite different meaning: rano rani Mustajbeželički i besjeda Mustajbeželički: posle toga Mustajbeželički “pobratime Mustajbeželički”

Mustajbey of the Lika arose early, And Mustajbey of the Lika addressed [them]) After him Mustajbey of the Lika. “Oh blood-brother Mustajbey of the Lika.”

These examples demonstrate how wide a variety this process can produce, from simple statements of 1) narrative fact, to 2) the introduction of a speech, to 3) a line from a catalogue of heroes, to 4) direct address by another character. Such whole-verse patterns with constant and variable parts are quite common, and they work in various ways. In the following short sequence, the same syntactic pattern unites verses describing a hero’s pledge to rescue a kidnapped maiden in the guslar Halil Bajgorić’s The Wedding of Mustajbeyequence, the sa (verses 542–544):

204  Comparative study of four epic traditions o tako mi mačaji junaštva, o tako mimojega bjelana, o tako mičeter’es’godina!

O by my sword and by my heroism, O by my white steed, O by my fourteen years [of fighting]!

Here, the Turkish hero Djerdjelez Alija is swearing that he will seek a maiden who has been carried off by the enemy and return her to her bridegroom for their planned marriage. He makes his vow formulaically, pledging “Oh by my …” in the first colon and then filling out the second colon with various objects. In the following four variations of a single-verse phrase, the verse cannot be subdivided without compromising its compositional usefulness and idiomatic meaning: a od tala nanoge skočijo a od tala nanoge skočila i skočijo nanoge lagane i skočila nanoge lagane

He jumped from the ground to his feet. She jumped from the ground to her feet. And he jumped to his light feet. And she jumped to her light feet.

These limited examples give a glimpse of how singers and audiences communicate by means of “large word” formulae, which may be a colon, a verse, or group of verses. We shall discuss the ideological significance of such formulaic phrases in response to “What is register in an oral epic tradition?”. Ancient Greek Homeric quantitative meter and formulae operate as cooperative partners in a system of mutual symbiosis. As shown earlier, Greek epic verse provides a traditional flexible structure for the organization and maintenance of the poetic language. The formulae fit into the metrical pattern, and the verse structure supports the formation of “large words” that match that pattern. The formulae may cover from one to four cola in the hexameter verse. Multiple formula verses can occur in sequence, but scholars ascribe such series to a combination of formulaic units rather than making a single multi-verse statement. Here are some commonly used formulae that may begin, end, or fill out a whole verse: ἀλλ᾽ ἄγε … τὴν δ᾽ ἠμείβετ᾽ ἔπειτα γλαυκῶπις Ἀθήνη ποῖόν σε ἔπος φύγεν ἕρκος ὀδόντων ἧος ὁ ταῦθ᾽ ὥρμαινε κατὰ φρένα καὶ κατὰ θυμόν

(begin) (begin) (end) (end)

But come … Then [he] answered her … … grey-eyed Athena … what word escaped your teeth’s barrier!

(whole)

while he pondered these things in his mind and in his heart.

Challenges in comparative oral epic  205 Each of these phrases satisfies two major criteria for being a formula: It occupies a significant segment of the verse (one or more of the four cola), and it recurs frequently enough in the poems to demonstrate its usefulness for the composing bard. The shorter formulae readily combine to make up a whole verse unit. To cite another example, the formula τὸν/τὴν/τοὺς/τὰς δ᾽ αὖτε προσέειπε (And again addressed him/her/them) is used hundreds of times in the Homeric poems, filled out in the following cola by the character’s name, to introduce an utterance: πολύτλας δῖος Ὀδυσσεύς θεὰ γλαυκῶπις Ἀθήνη μέγας κορυθαίολος Ἕκτωρ ἄναξ ἀνδρῶν Ἀγαμέμνων Γερήνιος ἱππότα Νέστωρ

(end) (end) (end) (end) (end)

… much-enduring divine Odysseus … goddess grey-eyed Athena … Hektor of the shining helm … lord of men Agamemnon … Gerenian horseman Nestor

Using this pattern, the bard can introduce utterances by a wide variety of humans and gods. Old English As has been said, Old English versification is very flexible. Conjoined with traditional diction, it allows a great deal of variation in formulaic structure. Instead of encapsulated phrases, Anglo-Saxon poetry consists largely of formulaic systems, patterns in which one or more constituent words can vary from one use to the next. The stressed and alliterative elements tend to remain stable and constant, while the unstressed and non-alliterating elements vary. Of course, there are formulae that are repeated exactly every time they are used; but in the main, the phraseology is fluid. Consider the following series of half-verse phrases, all of which mean approximately “in the old days”: in geardagum (Beowulf, line1) on fyrndagum (Andreas, line1) Þe git on ærdagum (“The Husband’s Message,” line 16)

in year-days in olden days when you two in earlier days

This flexibility gives scope to suit the ready-made phrase to the immediate context. Since there is no syllabic constraint in versification, the poet of “The Husband’s Message” can shape the half verse into a more complex unit while still maintaining the basic pattern. Another example of adaptability within limits is clear in the following halfverse phrases, drawn from a wide variety of epic and non-epic poems:

206  Comparative study of four epic traditions Þæt wæs god cyning (Beowulf 11) ac Þæt wæs god cyning (Beowulf 863) Þæt wæs an cyning (Beowulf 1885) wæs ða frod cyning (Beowulf 2209) Þæt wæs god cyning (Beowulf 2390) Þæt is soð cyning (Juliana 224) Þæt is wis cyning (Meters 24.34) Þæt is æðele cyning (Andreas 1722) Þæt wæs grim cyning (Deor 23) næs Þæt sæne cyning (Widsith 67)

that was an excellent king but that was an excellent king that was a peerless king he was then a wise king that was an excellent king that is a true king this is a wise king that is a noble king that was a savage king that was not a negligent king

The word “king” (cyning) plus an adjective make up the core of the formula, with the adjective bearing the alliteration that links this half verse with the other (non-cited) in the verse. The traditional syntactic pattern “that was/is” has minor variations; but taken as a whole, this idiomatic formula provides the poets with a ready-made way to celebrate (or criticize) the achievements of a king. The Beowulf poet employs it no fewer than five times as he characterizes Scyld Scefing, Hrothgar (twice), Beowulf, and Hygelac. In the other poems just quoted, the celebrated king is Guthhere, Eormanric, or the Christian God.

What is register in an oral epic tradition? We now embark on the fifth and final question, which in many ways constitutes a summary of our discussions of the other four. We inquire into the idiomatic significance that each of these features has in their epic traditions and that goes beyond the meaning of the words in everyday language. We inquire into role of referential epic register in the composition and reception of each of the four traditions. For the anthropologist Dell Hymes, registers are “major speech styles associated with recurrent types of situations” (Hymes 1989:440). We follow this definition, adding that register refers to both speech signs and their signification. Mongolian In some ways, the singer’s style of speech to evoke the connotations of traditional epic register in Jangar is very much like the everyday language of contemporary Mongolia. Children enjoy the stories in part because they find the language interesting and readily intelligible; and what they do not immediately recognize they can quickly learn through repeated listening to the traditional stories. Indeed, Jangar singers not uncommonly begin to learn the art of performance before age ten. They can do so because the epic word order and linguistic make-up are in many aspects almost the same as those of everyday speech. Nonetheless, there are also significant differences that mark the diction as alerting the audience to the underlying connotations of Mongolian epic language. One of these differences is the inclusion of archaic words, such as the terms for measurement bal and bere. The former has a definite meaning equivalent to degree of temperature; but it

Challenges in comparative oral epic  207 is never used in daily life. The latter, bere, is less exact; it seems to indicate a distance of about two kilometers. For singers, however, the precise measurement suggested by these terms is less important than their almost ritualistic role as terms of the traditional epic register arousing a certain poetic atmosphere. A more telling example of an expression where the idiomatic usage of traditional register points to meaning that goes beyond what can be found in a dictionary is the following three-line formula frequently used by Arimpil: ermen čagan hödege ejegü ičagan bögereg elesün sir-atohui-du

Endless white wilds. Masterless white desert. Ever-stretching golden dunes.

Bound together semantically, metaphorically, and acoustically by head rhyme, this formulaic unit bears much more significance than its literal meaning. In the traditional epic register, it regularly coincides with one of three narrative situations: A rest break during a long journey, a site for a battle soon to be joined, or a lonely place in which a hero contemplates what his next action should be. There is nothing explicit in the three-line unit that in any way suggests this implicit content, but singers and audiences knowing the epic register understand well its encoded connotations and thus share a deeper and fuller, but not explicitly stated, level of communication. One unfamiliar with the register, however, would be unable to grasp the singer’s full meaning, like an uninitiated reader lacking a dictionary. Another example of register referentiality is a four-verse segment that Arimpil and other singers employ numerous times in their performances: dugtui dotur-abayihul-a dolbing sara-unönggetai dugtui-eče-bengarhul-a dolugan naran-u gereltei

When the banner is in its holder, It flashes with the glow of the moon. When the banner is out of its holder, It shines like seven suns.

More than a well-wrought poetic building block cemented by alliterated head rhyme, this formula is resonant with connotative implications. Though the description seems to focus simply on the unfolding of a banner, what is at issue is much more momentous. Without exception, this cluster of verses identifies Jangar and his great army before an imminent battle; and listeners attuned to the epic convention know that the powerful enemy is already present and that a duel or full-scale engagement will very soon take place. Register also operates at the level of typical scenes as descriptive details enrich the narrative with encoded idiomatic connotations. A common example is what we may call the scene of healing the wounded hero. When a champion from the “white side” (Jangar’s troop) is wounded, he is treated with a standard series of three medicinal preparations: magical rain, curative water, and medicinal balm. The first is summoned down by the supernatural power of Jangar’s aide Altan Čegeji (Golden Chest). Next, the wounded hero ingests or has his

208  Comparative study of four epic traditions hands and face cleansed with healing water of unspecified origin. Finally, to the wound are applied the white balm called čagan and an ointment called üyeng (original meaning unknown) (Foley 1999b:23–24, 74–75, 80–83). The scene can be applied to any hero from Jangar’s troop, whether he was injured in a battle against the enemy or in a duel with a rival from within the troop itself. But, its idiomatic implications run deeper and mesh with the most basic conceptions of heroism in the Mongolian epic tradition. Because the three-part treatment never fails, mere invocation of the scene guarantees how the action will proceed in either a single episode or the whole Jangar Cycle. Indeed, in this instance, the traditional register referentiality is instrumental in maintaining the nature of Mongolian heroism with respect to both the characters who fit that role and the ideal as a whole: Heroes never die. As a final example of the idiomatic implications of the traditional register of Mongol oral epic, we have a singer’s customary closure of a canto. As stated earlier, cantos begin and end with palace feasting as a frame for whatever action intervenes. As a canto draws to a close, the following series of verses regularly occurs, signaling closure and completeness to audiences brought up on the connotations of epic register: jiran honug-unjirgalhijü dalan honug-un danggarai hijü nayan honug-un nayir hijü

Sixty days to share happiness. Seventy days to enjoy a banquet. Eighty days to celebrate a reunion.

South Slavic The question “What is register in the South Slavic oral epic tradition” invites us to consider further the questions we have been asking throughout this chapter: What does “word” mean in epic tradition? What implications do a formula, typical scene, and story-pattern have in epic performance? What is an epic singer communicating to an audience in addition to literal, dictionary-based meanings? (Foley 1999b:65–111). To begin, let us consider the final example of a formula discussed earlier in relation to “What is a formula in an oral poetic tradition?”. The whole-verse phrase takes two main forms: A od tala na noge skočijo (He jumped from the ground to his feet) and I skočijo na noge lagane (And he jumped to his light feet). From a grammatical and compositional perspective, this is one of the most adaptable sound-bytes in a guslar’s epic vocabulary. At the same time, as a trope in the South Slavic epic register, the formula traditionally designates “an honorable response to an unexpected or threatening turn of events that demands the principal’s immediate attention” (Foley 1999b:108). No matter who the character may be, the traditional implication is that a heroic mission is about to take place in response to something dire or unforeseen and that he (or she) is about to play a distinguishing role. By convention, the person will soon leave on a life-threatening mission to rescue a maiden, join an army force, perform a secret spy mission, or whatever. The tradition does not specify the exact mission or the

Challenges in comparative oral epic  209 outcome, but it does identify the character and the actions to be performed in accord with an implied verbal map. Commonly used formulae also have an idiomatic force within the poetic register. Epithets like “Mustajbey of the Lika” do not just identify the person of Mustajbey (who actually has no need to be identified) and his homeland Lika, a border area between Turkish and Christian territories (whose name is likely irrelevant). What matters is that by employing this formulaic large word, the bard evokes the entire characterization of Mustajbey that the audience knows well from his adventures in the whole epic tradition: The hero who commands the Lika armies, fights staunchly against Christian enemies, has a son named Bećirbey, and despite his noble status, can regularly prove traitorous to his own comrades. Similarly, the short phrase kukavica crna (black cuckoo) designates not a darkcolored bird but a woman who has lost her husband or is in imminent danger of losing him. By giving her this formulaic name, a guslar brings into play this traditional connotation and adds the character he is describing to the list of other “black cuckoos” in his and the audience’s experience of epic poetry. Likewise, by using an idiomatic transitional large word like “But you should have seen …,” filled out with the name of the chief character in the next scene, a bard can bridge a narrative gap between events. Or, a guslar might turn to one of hundreds of proverbs—all ten-syllable verses—to evoke a traditional context for a specific narrative moment. The larger narrative units of typical scenes also help constitute the epic register, connecting seemingly unique individual scenes to the larger world of South Slavic storytelling. By employing scenes that carry encoded meanings, guslari link immediate descriptions to an implied network of signification. One example is the scene “Readying the horse” examined in “What is a typical scene in an oral poetic tradition?”. In addition to simply getting the job done by providing a ready-made, capsule account of cleaning, grooming, and caparisoning the animal, the scene forecasts further typical scenes of arming of the hero and an expedition that he will undertake, and ultimately points toward a life-and-death adventure far from the hero’s home. The typical scene “Shouting in prison” offers another example of how South Slavic epic register functions. As we saw earlier, the focus is on a captured hero whose loud laments disturb the sleep of his captor’s baby son so severely that it threatens the survival of the captor’s royal lineage until a powerful female figure, usually the captor’s wife, intervenes and negotiates the prisoner’s release. This set of characters and events creates the literal scene, but there is more to its meaning. From the moment it begins, customarily with the verb cmiliti (cry out, scream), the “Shouting in prison” pattern forecasts a hero who, once released from long-time captivity, will successfully make his way back home to test the faithfulness of his wife. He will face severe challenges en route, and his wife (or fiancée) will have to defend herself against the advances of suitors who are trying to win her in marriage. There is no doubt but that at some future point they will meet, and a resolution, positive or negative, will take place. All that embedded story-material is implied when the guslar sings the scene.

210  Comparative study of four epic traditions Ancient Greek The register of ancient Greek epic is also far more than a useful compositional device. With its encoded idiomatic implications, Homer’s way of speaking conveys much more than a literal level of meaning (see Foley 1999b:115–239). Both formulaic phrases and typical scenes make implicit reference to the traditional network of ideas, which is much larger and deeper than can be discovered in a dictionary or lexicon. In the area of formulaic diction, epithet phrases such as “much-enduring divine Odysseus” and “goddess grey-eyed Athena” are much more than fillers. They amount to a coded message to gain access to the fuller characterization of the persons they name. It is often not important in the immediate context that Odysseus is called “much-enduring” and “divine.” These adjectives do not apply specifically to an appearance in a particular situation any more than a greeting like “Hello” or a wish like “Have a pleasant weekend” serves only one moment in a person’s life. The epithets are idiomatic phrases that access the traditional register network and enrich any one episode with tradition-wide implications. Two other formulae mentioned earlier work in similar ways. Brief though it is, the two-word phrase “but come” (ἀλλ᾽ ἄγε), which occurs 149 times in the Homeric poems, regularly serves two idiomatic purposes: 1) It divides one section of a speech from another, preparing the listener for a change of focus; 2) it leads to a command or prayer. Likewise, the exclamation “What word escaped your teeth’s barrier!” (ποῖόν σε ἔπος φύγεν ἕρκος ὀδόντων) brings concisely to awareness a built-in traditional context that reaches beyond the literal meaning. The audience knows from it that the speaker—an older or socially superior figure—is scolding a younger or inferior person for something he or she should have known or done. The same evocative strategy underlies the use of typical scenes in the Iliad and the Odyssey. The standard Feast scene regularly carries with it an expectable set of implications. Whatever the situation, and whoever the host and guest(s) may be, it always leads to thoughtful action; and the well-attuned audience expect that things will take a turn for the better. In the first book of the Odyssey, the Feast hosted by Telemachos for the disguised Athena leads to the young man’s speaking boldly to his mother’s suitors and eventually to the trip to Menelaos’ and Nestor’s homes that prepares him for the reemergence of his father Odysseus later on in the story. According to the same connotative thrust, a Feast precedes Kalypso’s release of Odysseus from captivity, Kirke’s assistance to Odysseus, and many other such remedial episodes. Finally, the three-part Lament scene—address to a fallen hero, account of his life story and the consequences of his death for those left behind, and final intimate address—is also more than a structural convenience (Foley 1998b:188–198). It amounts to a blueprint for later speeches of mourning by Andromache, Helen, Hekabe, and Briseis. While Hektor is visiting his wife Andromache with their infant son in a brief respite from battle, she implores him to remain away from the fighting and preserve himself for their sake; but he denies her sad request and

Challenges in comparative oral epic  211 returns to the battlefield, where he dies at Achilles’ hand. An alert Homeric audience realizes that she is speaking in the format of a typical Lament scene, already mourning the death of her husband, even though he stands alive before her. Old English In the same way, we grasp the expressive power of register in Old English oralderived poetry in terms of the broader idiomatic significance of a work’s typical scenes and formulae in the context of the whole poetic tradition. As has been noted, the Exile scene occurs across a broad spectrum of poems and genres. In every case, it brings with it the idea of a person separated from essential networks of kinship and society and implicitly compares the person in the scene with other exiled personages in the poetic tradition. In the poem Deor (The Lament of Deor), for example, a singer who has lost his position at the royal court laments the fact that he has been displaced by a new singer, Heorrenda, and compares his plight to those of familiar figures from Germanic legends. When Deor says in verses 1 and 4 that he “had knowledge of exile,” of “winter-cold exile,” he places himself in that traditional category. He is not simply estranged; he is a typical Germanic exile, and that helps the audience appreciate his inner pain and desolation to the full (Foley 1999b:263–270). In the case of the three Sea Voyage scenes in Beowulf, the two voyages from Geatland to Denmark and back again are presented straightforwardly as actual expeditions: A hero leads his men on an expedition. But, the third is not a real voyage; it deploys a series of four type tropes to metaphorically depict the ship-burial of the legendary hero Scyld Scefing (Foley 1990:336–344). 4).– 1990l,” the dead hero and his men go to the ship (trope 1), which waits, moored (2). They board the ship and place treasure in the form of grave-goods by the mast (3), and the ship is said to depart and sail (4). But, the poet here brilliantly trips up the audience’s expectations of a typical Sea Voyage scene. Instead of the anticipated landing at the end of a journey and customary encounter with a coastal guard, the poet states that “men don’t know, to say truthfully, hall-counselors, heroes under the heavens, who received that burden” (lines 50b–52). Using the expected Sea Voyage pattern of the epic register as a strategic ploy, he concretely reinforces the idea that no one can be certain about the afterlife of Scyld Scefing. Both his destination and his final reception there are beyond our knowledge. As for the idiomatic thrust of set formulae, these two scenes begin like many Anglo-Saxon narratives with the alliterated phrase “in X days.” The expression is a set part of a larger cluster of phrases that signal the beginning of an episode while invoking traditional myth. Combined with the interjection Hwœt (“Lo!” or “Listen!”), this formula identifies the onset of a heroic tale against the background of other stories in the tradition. For its part, the alliterated half-line formula “that was an X king” also has connotations beyond its literal meaning. Wherever it appears, it certifies the character as a figure whom both his people and succeeding

212  Comparative study of four epic traditions generations will celebrate as an ideal king. This status is automatically conferred by the tradition and has no need for independent verification (Foley 1991:210– 223). Thus, for instance, Hrothgar is certified as an excellent king (line 863), even though under his reign Grendel has been able to ravage his people, killing them at will during the night. Hrothgar’s excellence derives from his earlier effectiveness, and Grendel’s ability to overcome his followers is simply a measure of the monster’s unprecedented strength and fury. The Deor poet upsets hearers’ expectations by turning this expression to a negative purpose when he says of Eormanric, “that was a savage king” (line 23). In this way, the poet briefly, but adroitly, creates a memorable characterization. The four epic traditions we have examined—Mongolian, South Slavic, ancient Greek, and Old English—represent an enormous variety. They cover an extensive geographical area from northern Asia to western Europe as well as a time period of at least 3,000 years. As we have attempted to answer each of the five questions for all the four traditions, we have come to appreciate their diversity even more deeply. To start, we asked: “What is a poem in an epic tradition?” Though this may seem like a simple, straightforward question, we discovered that its answer depends upon particularities of each of the traditions for understanding both structure and meaning. Different poems follow different story-patterns and depend on a variety of implied references in the development of characters, events, and situations. Asking “What is a typical scene?” led to similar conclusions. Each of the poetic traditions uses narrative increments as building blocks, and the patterns change with each performance according to the particular terms of the individual language and tradition. Typical scenes in Mongolian epic are shared from one canto to another while taking slightly different shape from the mouths of different singers. They are in some ways quite unlike corresponding narrative units of South Slavic, ancient Greek, or Old English epic. Our third question—“What is a poetic line verse?”—again addressed the phenomenon of unity and diversity. In some traditions, verses are measured by syllables, some by stress, some by musical melodies, and others by head rhyme, tail rhyme, alliteration, and assonance. Each oral epic language must be understood on its own terms, not by imposing irrelevant scholarly criteria. The same is true of formulae, the subject of our fourth question. A formula exists in conjunction with the poetic verse. Since formulaic phraseology is defined by its metrical component, it also will vary from one singer to the next and even more from one epic tradition to another. These questions and answers led us to our fifth question and our final concern—“What is register in oral epic poetry?” With this focus, we reached beyond the level of structure and utility to the level of idiomatic meaning. Mongolian cantos commonly begin and end with a scene of palace feasting to frame the action in a ring-like structure that signals closure and completeness. Homer uses a Feast scene differently, to signal thoughtful action that will hopefully improve matters. South Slavic guslari refer to a woman as “black cuckoo” to indicate that she either has been or soon may be widowed. Old

Challenges in comparative oral epic  213 English scopes certify a leader’s effectiveness and heroism by saying “That was an excellent king!” All these narrative bytes mean much more than they seem to mean; their traditional, idiomatic sense goes well beyond the literal. In their various ways, each of the four ways of using epic language thus has great connotative resources. As was noted at the beginning of this chapter, this study was composed as a separate article by the late John Miles Foley and the present author many years ago at the University of Missouri. Recognizing the need to respond to the current challenges in epic studies in a way that would illustrate the richness and diversity of particular epic traditions, we started to work on what has now become this chapter throughout the summer of 2001, sitting side by side and drawing upon our respective experiences and backgrounds. When first published in a Chinese version in a 2003 collection of papers on Oriental literature, it provoked reflections on basic dimensions of epic poetry and facilitated multiple ways of understanding epic in the trend to move beyond purely Homeric criteria for appreciating epic poetry. Like much of what has been said in this book, this chapter reflects both our common scholarly interests and our longterm friendship. When John made his first trip to China, he delivered lectures in our institute, conducted field trips to Inner Mongolia, and visited a number of scholars working on related research. His own works then began to include Mongolian examples, such as the singers Choibang and Losor; and from then on, he visited China regularly and came to hold appointments in the Institute of Ethnic Literature. It was a great misfortune that our newly designed program on oral tradition and the internet with partners in Missouri, Helsinki, and Beijing was interrupted by John’s sudden demise. Nevertheless, without a doubt, the blueprint of this international program will be followed and fulfilled. As he intended, electronic materials on oral epic are available via the internet at the website of the Center for Studies in Oral Tradition, University of Missouri (www​.oraltradition​.org). His cutting-edge thoughts will continue to shed light on our explorations. In the Mongolian epic tradition, a true baatar (hero) never passes away. John is, by any measure, just such a baatar. In closing, it is hoped that the comparative analysis of this chapter and the account of contemporary oral epic research presented in this book will prove useful for scholars in various fields. There is an enormous amount of investigation still to be done. We need to understand the structure of performance texts thoroughly enough to hear and read them on their own terms. We need to assemble dictionaries of “words” (formulae, typical scenes, story-patterns) for each of the world’s oral poetic traditions. We need to take full account of the fact that no matter how many similarities we may find among them, significant differences will also remain. In short, we must strive to become better, more fluent audiences for epic singing. The chapter was originally published in Oral Tradition with John Miles Foley. 27/2 (2012) 381–418.

214  Comparative study of four epic traditions

Notes 1 On Muslim versus Christian epics, see Foley 1991:61–134, 1999b:37–111, and 2002:188–218. 2 For published material, see SCHS, Serbo-Croatian Heroic Songs. Collected, ed., and trans by Milman Parry, Albert B. Lord, and David E. Yunm. Mass: Harvard University Press and Belgrade; Serbian Academy of Sciences, 1953; for unpublished materials, we are grateful to Stephen Mitchell, the Curator of the Milman Parry Collection. 3 An edition of this performance in available in Foley, The Wedding of Mustajbey’s Son Bećirbey as Performed by Halil Bajgorić. Folklore Fellows Communications 283 (Helsinki: Suomalainen Tiedeakatemia, 2004). eEdition: at http://oraltradition​.org​/zbm.

Appendix: Basic concepts in the study of oral epic

The main schools of interdisciplinary research on oral traditions—Oral Formulaic Theory, Performance Theory, and Ethnopoetics—have produced a well-established system of terminology that implies a theoretical framework and has to some extent become the common foundation of international epic studies. The systemization of epic poetry is possible, however, only after carefully defining terminology, especially similarities and differences in terms as applied to written and oral literature. This Appendix explains a few of the most common terms.1 Oral epic: An epic is a long narrative poem centered on a hero or heroes in their engagement with historical events such as war or conquest or in a grand mythical or legendary pursuit. Epic has commonly arisen in societies with an oral culture at a time when the nation was forming is religious tradition and historical culture. Most epics are based on historical events, but their role is not to record history. The central hero of an epic is sometimes a demigod engaged in a difficult but just cause, often one that involves conflicts between deities and humans. Epic events often touch on the daily lives of ordinary people but are such as lead to changes in a nation’s historical course. An epic is typically large in size, full of detailed descriptions, and structured sequentially. It makes extensive use of stylized characters, expanded similes, and detailed formulaic descriptions of weapons and rituals. The stories commonly include formal fights between warriors, explanations of contests and their rules, unbelievable adventures that sometimes involve intervention by supernatural powers, and often superhuman feats and schemes. Epic uses methods not only of narrative but also of lyric and dramatic poetry. The diversity of the world’s epics has caused considerable difficulty in defining epic.2 Oral epic composition: This refers to improvised narrative poetry that relies heavily on traditional expressions and poetic principles but enjoys a degree of freedom, placing it between strict creative composition and artistic arrangement. The term “composition” is commonly used for written works but here refers to creative improvisation based on traditional poetic units in a live performance. Epic cycle: This refers to a complete series of works, in particular to a collection of narrative poems or legends centered on an important event or person, usually created over time by several authors. First used for narrative poems created by the late Greek “group poets” to supplement Homer’s depiction of the Trojan

216 Appendix War, the term later came to refer to a series of poetic narratives, such as those centered on Charlemagne and King Arthur. It is also used to refer to a group of verse poems or a series of related poetic cantos centering on a particular heroic figure and related events. The central hero is not necessarily the protagonist of each canto but like Jangar, a structurally important core figure. Ring composition: Different from epic cycle, this is a common structure found in many oral traditions, originally in that of Homer. Ring composition arranges characters, actions, or scenes in a circular or concentric structure, following an A-B-C sequence and then repeating it in reverse. It is widely used in Anglo-Saxon and Slavic poetry, where it has both a mnemonic and an aesthetic function. Canto: Originally referring to a long troubadour’s narrative sung only once, this term generally indicates a stanza or sequence of a long poem. In an epic cycle, a canto constitutes a relatively complete story unit, more or less independent in form and structure. In the study of Chinese epics, it is used along with the terms “chapter” and “section” borrowed from prose narrative. Song and singer: As a literary term, song is closely related to lyricism and associated with music. Under the term, adherents of Oral Formulaic Theory include epic, ballad, lyric poem, and other oral art styles involving rhythm and melody, sometimes accompanied by musical instruments. The lines between these genres are not clear; and folk singers commonly master them all. In Chinese studies, the term “song” is not used to cover these various genres, and epic singers are usually called “epic artists.” Oral tradition: The general concept of tradition points to a collection of beliefs, customs, folk literature, maxims, and fixed ideas handed down from generation to generation, usually orally. In literature, tradition means something inherited from the past rather than the author’s own creation, and also the whole inheritance of literary formulae. More specifically, “oral tradition” sometimes refers to all factors involved in a folk performance, not just the content of a singer’s performance or repertory but also implicit meanings shared by singer and audience that cannot be found in a dictionary. Tradition sometimes refers, too, to a specific research trend or area, such as the Oirat poetic tradition. Text: In the most general sense, text refers to a combination of words and sentences created according to particular language standards. In different academic traditions, it has different meanings. In English and French, it means a discourse. In traditional linguistic and literary studies, it is commonly thought of as something written, whereas discourse is vocal and oral. In epic research and oral poetics, it includes not only transcriptions and printed copies but also what is produced in the course of performance. In the narrow traditional sense, text refers to a writer’s work or a document, etc., and may also be called “corpus”; but for most semioticians, it is any symbolic expression with underlying meaning, whether or not it is composed of linguistic signs. A rite, dance, or facial expression is thus also a text. All these various kinds of text refer to something expressible, and the

Appendix  217 process of text creation itself can also be regarded as a text. A text is something expressed but also something being expressed. Context: This term refers to the ambiance in which the language of a specific text or discourse is used. It includes relations between words, sentences, and paragraphs as well as semantic constraints in a specific form of communication, such as the social background, style, emotions, and customs reflected in the meaning of words. In broad terms, context contains many factors, such as history, geography, nationality, religion, language, and social conditions, all of which can be important in understanding the relationship between a work’s creation and transmission. In the study of oral poetic texts and the on-the-spot improvisation of an epic performance, context refers to the cluster of social relations applicable at a specific time, such as human beings themselves, but also time, place, process, cultural features, and the modes of generation of meaning. Register: This term refers to a mode of language with a specific purpose as opposed to the dialect of a particular group or region. In a speech community, social factors such as occupation, class, age, and gender—called “sociolects”—affect the way one speaks, while each person in the community speaks with slightly different “idiolects.” All this gives rise to language “register” or “style.” The register of ancient epic often differs from people’s daily language in both vocabulary and syntax. In general, register can be subdivided into various jargons (e.g., fishing jargon, gambling jargon); printed material, epistolary letters, and audiotape information; and also formal, casual, intimate discourse. Mathematical and statistical models are often used to analyze the impact of such various ways of speaking on register and style. Modern linguistics regards them as a system, and sociolinguistics studies them in conjunction with various social factors. Orality: This refers to the characteristics and laws of oral communication as opposed to writing. The basic characteristics of oral expression are juxtaposition (rather than progression), aggregation (not separation), redundancy or repetition, and traditional, conservative tendencies. Scholars originally thought that there was an insurmountable gap between orality and writing but have increasingly come to realize that intermediate transitional forms exist between the two. Both orality and writing are found in contemporary folk culture. Many Mongol jangarch in Xinjiang today are literate; and their basically oral ways of thinking and expression have come to be influenced by the written culture. Textuality: This refers to the elements that give a text the quality of language rather than just a collection of sentences. Its structural elements generally include three aspects: cohesion, prominence, and macro-structure. Intertextuality: Proposed by the French structuralist critic J. Kristeva, this refers to any connection between a literary text and other works, any one text being dependent on other texts. Post-structuralist theory thinks that there are no truly innovative texts and that a new text always involves hidden imitation. It just redistributes language symbols to form a new language texture; and its new traits

218 Appendix are limited. Intertextuality plays an especially important role in oral traditions. In the epic cycle Jangar, for example, each canto is linked with every other canto; each singer’s version has relationships with those of other bards; and the whole tradition of Jangar is dependent on other forms of epic singing and folk art. Formula: According to Milman Parry, a formula is a stable, repetitive phrase used to express a basic idea under the same meter. It ensures smooth narration in on-the-spot performance and is recognizable by the audience. It can also be a whole sentence or a larger structural unit. Formulaic diction: This refers to all phonetic factors, forms, and vocables that constitute a specific people’s discourse. They influence the mode of expression of formulae and are employed by authors to express meaning. Formulaic density: This refers to the frequency of the recurrence of a formula in a given unit. Oral Formulaic Theory scholars think there is a significant relationship between a work’s formulaic density and orality and so, often use it to determine whether or not a work has oral origins. Meter: This refers to the rhythm formed by the repetition of similar set phonetic units. There are four basic types: (1) In Classical Greek and Latin poetry, the rhythm is based on the alternation of long and short syllables. (2) In Old English poetry, it is based on the stressed accents of words. (3) In many legendary poems, it is based on a set number of syllables in a line, though the accentuation varies. 4) In English poetry, it is typically based on a set number of syllables and accents in a line. Parallelism: This refers to the structural arrangement of phrases, sentences, paragraphs, or larger units in a text, in which more or less equal ideas, wording, and structure are juxtaposed. It is widely used in Mongolian epic, where it is closely related to the thought characteristics of oral expression, juxtaposed rather than progressive. Enjambement, run-on lines: This refers to a meaningful phrase that does not end with the end of a verse but continues into the following line without a rhythmic pause. It is common in verse couplets and written poetry but not so much in oral folk poetry. Theme, or motif: A narrative unit in scale somewhere between formula and story-pattern. Story-pattern: This refers to narrative paradigms in oral traditions, to an organizational element in the composition and dissemination of oral texts, though the stories built around them vary. The story pattern of the SerboCroatian Return Song, for example, consists of a five-faceted sequence of absence–disaster–return–revenge–wedding. In larger form, this Appendix appeared as an article written together with the late John Miles Foley and was published in Chinese in 2003 as “Five Topics in Oral Poetics: A Comparative Study of Four Traditions” and later as “Challenges in Comparative Oral Epic” in Oral Tradition, 27/2 (2012): 381–418.

Appendix  219

Notes 1 The main reference works are Alex Preminger and T. V. F. Brogan et al 1993, eds. The New Princeton Encyclopedia of Poetry and Poetics. Princeton University Press; M. H. Abrams 1999. A Glossary of Literary Terms. Wu Songjiang trans. Peking University Press. 2 For comparisons among epics of different traditions, see Albert Lord The Singer of Tales.

Bibliography

Auty, Robert and Hatto, Arthur T., Eds. 1980. Traditions of Heroic and Epic Poetry. London: Publications of the Modern Humanities Research Association. Baldick, Chris 2001. The Concise Oxford Dictionary of Literary Terms. Oxford: Oxford University Press. Bamo, Qubumo 巴莫曲布嫫 2003. Shishi chuantong de tianye yanjiu-yi nuosu yizu shishi “le’e” wei ge’an 史诗传统的田野研究——以诺苏彝族史诗“勒俄”为个案 [A Field Study of Epic Tradition: Nuosu Yi Epic Hnewo As a Case]. Doctoral Dissertation. Beijing: Beijing Normal University. Bamo, Qubumo 巴莫曲布嫫 2003/2004. “Minjian xushi chuantong “geshihua” zhi piping-yi yizu shishi le’eteyi de wenben yilu weili 民间叙事传统“格式化”之批评— 以彝族史诗〈勒俄特依〉的 “文本迻录”为例(上、中、下) [A Critique of the “Formatted Approach” to Traditional Oral Narratives: A Case of Problematic Textualization of Yi Epic Hnewo Teryy, A Monograph Serialized as Parts I, II, and III]].” In minzu yishi民族艺术 [Ethnic Arts Quarterly]. 4:39–45; 2004(1):18–26; and 2004(2):32–40. Bamo, Qubumo 巴莫曲布嫫 2004a. “Xushi yujing yu yanshu changyu—yi nuosu yizu de koutou lunbian he shshi chuantong weili 叙事语境与演述场域— 以诺苏彝族的口头论辩和史诗传统为例 [Narrative Context and Performative Field: Nuosu Verbal Dueling and Epic Tradition as a Case Study].” In wenxue pinglun 文学评论 [Literary Review]. 1:150–158. Bamo, Qubumo 巴莫曲布嫫 2004b. “Xushi xinggou wenben jiexian xushi jieyu: chuantong zhishexing de faxian叙事型构·文本界限·叙事界域: 传统指涉性的发现 [Narrative Configuration, Textual Boundary and Narrative Domain: Discovering Traditional Referentiality].” In minsu yanjiu 民俗研究 [Folklore Studies]. 3:113–133. Bascom, William 2006. “Koutou chuancheng de xingshi:santi xushi 口头传承的形式:散体叙事[The Forms of Folklore: Prose Narratives].” In Xifang shenhuaxue duben西方神话学读本 [Sacred Narrative: Reading in the Theory of Myth]. Ed. Alan Dundes. Trans. Chao Gejin et al. 朝戈金. Guilin: Guangxi Normal University Press 广西师范大学出版社. 5–37. Batnasan, J. 1984. “Jiange’er zai hebukesai’er liuchuan qingkuang diaocha 〈江格尔>在和布克赛尔流传情况调查 [Investigation into Jangar’s Transmission in the Hoboksar ].” In minzu wenxue yanjiu 民族文学研究 [Studies of Ethnic Literature]. Trans. Wang Qing 王清. 1:42–54. Bawden, Charles R. 1980. “Mongol: The Contemporary Tradition.” In Traditions of Heroic and Epic Poetry. Ed. Arthur T. Hatto. Vol. 1. London: Publications of the Modern Humanities Research Association. 268–299.

Bibliography  221 Bergmann, Benjamin Von 1804–05. Benjamin Bergmann’s Nomadische Streifereien Unter Den Kalmuken in Den Jahren 1802 Und 1803. Riga: C. J. G. Hartmann. Bessinger, Jess B. Jr. 1978. A Concordance to the Anglo-Saxon Poetic Records. Programmed by Philip H. Smith. Ithaca, NY: Cornell University Press. Bian, Yuan 边垣 1958. Honggu’er 洪古尔 [Hongor]. Beijing: The Writers Publishing House 作家出版社. Bitkeev, N. T. 1984. “Ka’ermeike yingxiong shishi de tedian 卡尔梅克英雄史诗特点 [The Features of Kalmyk Heroic Epic].” Trans. Liu Lunzhen 刘伦振. In minzu wenxue yicong 民族文学译丛 [Ethnic Literature Translation Series]. Vol. 2. Eds and Print. Beijing: Institute of Ethnic Literature, Chinese Academy of Social Sciences中国社会科学院少数民族文学研究所. 52–77. Blackburn, Stuart. H. 1989. “Patterns of Development for Indian Oral Epics.” In Oral Epics in India. Eds. Stuart H. Blackburn, Peter, J. Claus, Joyce B. Fluckiger, and Susan S. Wadley. Berkley and Los Angeles: University of California Press. Bormanshinov, Arash 1983. “Shishi jiangge’er yanjiu xianzhuang 史诗江格尔研究现状[The Status Quo of Epic Jangar Research].” Trans. He Kaige何凯歌. In minzu wenxue yicong 民族文学译丛 [Ethnic Literature Translation Series]. Vol. 1. Eds. and Print. Beijing: Institute of Ethnic Literature, Chinese Academy of Social Sciences 中国社会科学院少数民族文学研究所. 206–246. Bormanshinov, Arash 1984. “Eliyang aofula de yanchang yishu鄂利扬 ·奥夫拉的演唱艺术 [Singing Art of Eela Ovlaa].” Trans. Tang Guangting 唐光庭. In minzu wenxue yicong 民族文学译丛 [Ethnic Literature Translation Series]. Vol. 2. Eds. and Print. Beijing: Institute of Ethnic Literature, Chinese Academy of Social Sciences 中国社会科学院少数民族文学研究所. 146–164. Burchinova, L. S. 1983. “Jiangge’erxue zai eluosi de qiyuan 江格尔学在俄罗斯的起源 [The Origin of Jangar Studies in Russia]. ” Trans. Chen Hongxin 陈洪新. In minzu wenxue yicong 民族文学译丛 [Ethnic Literature Translation Series]. Vol. 1. Eds. and Print. Beijing: Institute of Ethnic Literature, Chinese Academy of Social Sciences 中国社会科学院少数民族文学研究所. 247–270. Bürinbeki, B. 巴·布林贝赫 1990. Menggu shige meixue lungang 蒙古诗歌 美学论纲[Aesthetic Essentials of Mongolian Poetry]. Hohhot: Inner Mongolia People’s Publishing House 内蒙古人民出版社. Bürinbeki, B. 巴·布林贝赫 1997. Monggol Bagaturlig Tuulis-un Silüg Jüi [The Poetics of Mongolian Epic]. Hohhot: Öbör Monggol-un Surgan Hümüjil-un Heblel-un Horiya [Inner Mongolia Education Press 内蒙古教育出版社]. Bürinbeki, B. 巴·布林贝赫 and Buyanheshig 宝音和西格 1988. Monggol Undusten-un Bagaturlig Tuulis un Songgomal 蒙古族英雄史诗选. Hohhot: Öbör Monggol-un Arad-un Heblel-un Horiya [Inner Mongolia People’s Publishing House 内蒙古人民出版社]. C. Norjin,策.诺尔金 1999. Mongolian Dictionary. Hohhot: Inner Mongolian People’s Publishing House. (In Mongolian). Chao Gejin 朝戈金 1999a. “Kouchuan shishi de wudu- chaogejin fangtanlu 口传史诗的误读—朝戈金访谈录 [On the Misreading of Oral Epic: An Interview with Chao Gejin].” In minzu yishu民族艺术 [Ethnic Arts Quarterly]. 1:8–18. Chao Gejin 朝戈金 1999b. “Koutou chengshi lilun yu shishi chuangbian wenti '口头程式理论'与史诗'创编'问题 [Oral Formulaic Theory and Epic Composition]. In zhongguo minsuxue niankan中国民俗学年刊 [Annual Journal of Chinese Folklore]. Shanghai: Shanghai Literature and Art Publishing House 上海文艺出版社. 176–189.

222  Bibliography Chao Gejin 朝戈金 2000. Kouchuan shishi shixue: ranpile jiangge’er chengshi jufa yanjiu 口传史诗诗学:冉皮勒〈江格尔〉程式句法研究 [Oral Poetics: Formulaic Diction of Arimpil’s Jangar Singing]. Nanning: Guangxi People’s Publishing House 广西人民出版社. Chao Gejin 朝戈金 2002. “Guanyu koutou chuanchang shige de yanjiu—koutou shixue wenti 关于口头传唱诗歌的研究-口头诗学问题 [Studies on Verbally-Transmitted Song-Poems: Issues in Oral Poetics ].” In wenyi yanjiu 文艺研究 [Studies of Literature and Art ]. 4:99–101. Chao Gejin 朝戈金 and Foley 2003. “Koutou shixue wuti: sida chuantong de bijiao yanjiu 口头诗学五题: 四大传统的比较研究 [Five Topics in Oral Poetics: Challenges in Comparative Oral Epic].” In dongfang wenxue yanjiu jikan 东方文学研究集刊 [Bulletin of Oriental Literature]. Changsha: Hunan Literature and Art Publishing House 湖南文艺出版社. 33–97. Chao Gejin 朝戈金 and Bamo Qubumo 巴莫曲布嫫 2004. “Minzuzhi shixue 民族志诗学 [Ethnopoetics].” In min jian wenhua luntan 民间文化论坛 [Folk Culture Forum]. 6:90–91. Chao Gejin 朝戈金, Yi Hubin 尹虎彬 and Bamo Qubumo 巴莫曲布嫫 2010. “Zhongguo shishi chuantong: wenhua duoyangxing yu minzu jingshen de bowuguan 中国史诗传统:文化多样性与民族精神的“博物馆” [Living Oral Epic Traditions in China: Celebrating Cultural Diversity and Ethos ].” In guoji bowuguan国际博物馆 [Museum International]. 1:5–27. Chao Gejin 朝戈金 2012a. “Yaluwang: fuhexing shishi de xianhuo anli〈亚鲁王〉: 复合型史诗的鲜活案例 [King Yalu: A living Case of Compound-type Epic].” In zhongguo shehui kexue bao 中国社会科学报 [Chinese Social Sciences Today]]. March 23. A05. Chao Gejin 朝戈金 2012b. ““John Miles Foley (1947–2012).” In Fabula. 53(1–2):111–113. Chadwick, Nora K. and Victor Zhirmunsky 1969. Oral Epics of Central Asia. Cambridge: Cambridge University Press. Chen, Yongchao 陈泳超 2005. Zhongguo minjian wenxue yanjiu de xiandai guizhe 中国民间文学研究的现代轨辙 [Tracing the History of Scholarship in Folk Literature in Modern China]. Beijing: Peking University Press 北京大学出版社. Chimud-dorji 齐木道吉, Liang Yiru 梁一孺 and Zhao Yongxian 赵永铣 1981. Mengguzu wen xue jianshi蒙古族文学简史 [A Concise History of Mongolian Literature]. Hohhot: Inner Mongolia People’s Publishing House 内蒙古人民出版社. Chinggeltei 1991. Menggu yufa 蒙古语语法 [Mongolian Grammar]. Hohhot: Inner Mongolia People’s Publishing House 内蒙古人民出版社 Chinese Folk Literature and Art Association, Ed.中国民间文艺家协会 2011. Yaluwang亚鲁王[The King Yalu]. Beijing: Zhonghua Book Company 中华书局. Connelly, Bridget 1986. Arab Folk Epic and Identity. Berkeley, CA: Unversity of California Press. Davies, Malcolm 1989. The Greek Epic Cycle. Bristol: Bristol Classical Press. Dawataya 塔亚 1999. “Geshou ranpile de jiangge’er: xinjiang weilate-menggu yingxiong shishi 歌手冉皮勒的（江格尔〉: 新疆卫拉特-蒙古英雄史诗 [Jangar of Singer Arimpil: Heroic Epics of Oirat-Mongol in Xinjiang].” In Journal of Chiba University Eurasian Society, Special Issue. 1:1–458. Dong Xiwen 董希文 2006. Wenxue wenben lilun yanjiu 文学文本理论研究 [On Theories of Literary Text ]. Beijing: Social Sciences Academic Press 社会科学文献出版社. Duggan, Joseph J. 1973. The Song of Roland: Formulaic Style and Poetic Craft. Berkely, CA: University of California Press.

Bibliography  223 Fine, Elizabeth C 1984. The Folklore Text: From Performance to Print. Bloomington and Indianapolis, IN: Indiana University Press. Foley, John Miles 1988. The Theory of Oral Composition: History and Methodology. Bloomington and Indianapolis, IN: Indiana University Press. Foley, John Miles 1990. Traditional Oral Epic: The Odyssey, Beowulf, and the SerboCroatian Return Song. Berkeley and Los Angeles, CA: University of California Press. Foley, John Miles 1991. Immanent Art: From Structure to Meaning in Traditional Oral Epic. Bloomington and Indianapolis, IN: Indiana University Press. Foley, John Miles 1995. The Singer of Tales in Performance (Voices in Performance and Text). Bloomington and Indianapolis, IN: Indiana University Press. Foley, John Miles 1997. “Koutou chengshi lilun:koutou chuantong yanjiu gaishu 口头程式理论：口头传统研究概述 [Oral Formulaic Theory and Research: An Introduction and Annotated Bibliography].” In minzu wenxue yanjiu 民族文学研究 [Studies of Ethnic Literature]. Vol. 6. Ed. A. Dundes. Trans. Chao Gejin 朝戈金. New York and London: Garland Publishing Inc. 86–90. Foley, John Miles 1999a. “Epic Cycles and Epic Traditions.” In Euphrosyne: Studies in Ancient Epic and its Legacy in Honor of Dimitris N Maronitis, Eds. John N. Kazazis and Antonios Rengakos. Stuttgart: Franz Steiner Verlag. 99–108. Foley, John Miles 1999b. Homer’s Traditional Art. University Park, PA: Pennsylvania State University Press. Foley, John Miles 2000a. “Dianlv zhi jiegou 典律之解构 [The Impossibility of Canon].” Trans. Zhai Shengde翟胜德. In minzu wenxue yanjiu 民族文学研究 [Studies of Ethnic Literature]. Supplementary Issue. 32–48. Foley, John Miles 2000b. Koutou shixue: pali-luode lilun口头诗学：帕里一洛德理论 [The Theory of Oral Composition: History and Methodology]. Trans. Chao Gejin朝戈金. Beijing: Social Sciences Academic Press社会科学文献出版社. English version published in 1988 by Indiana University Press. Foley, John Miles 2002. How to Read an Oral Poem. Urbana, IL: University of Illinois Press. (With web-page companion at www​.oraltradition​.org). Foley, John Miles, Ed. 2005. “Analogues: Modern Oral Epic.” In A Companion to Ancient Epic. Oxford: Wiley-Blackwell Publishers. 196–212. Foley, John Miles 2012. Oral Tradition and the Internet: Pathways of the Mind. Urbana, IL: University of Illinois Press. Gao, Hehong 高荷红 2011. Manzu shuobu chuancheng yanjiu满族说部传承研究[Studies of the Manchu Storytelling Inheritance]. Beijing: China Social Sciences Press 中国社会科学出版社. Golstunskij, F. K. 1864. “Wubashi hongtaiji zhuan minjian changshi jiangge’er he shenqi de sishi 乌巴什·洪台吉传、民间长诗〈江格尔〉和神奇的死尸 [Ubashi Hongtaiji, Jangar and Magical Corpse].” In Todo. St. Petersburg: Printed in lithography. Hangin, John Gombojab, Krueger，John R. and Service, Robert G. 1998. Mongolian Folklore: A Representative Collection from the Oral Literary Tradition. Indiana: Mongolia Society. Hanson, Victor Davis and John Heath 1998. Who Killed Homer? The Demise of Classical Education and the Recovery of Greek Wisdom. New York: The Free Press. Harris, Joseph and Reichl, Karl, Eds. 1997. Prosimetrum: Crosscultural Perspectives on Narrative in Prose and Verse. London: D. S. Brewer Press. Hegel, Georg Wilhelm Friedrich 1997. Meixue 美学 [Lectures on Aesthetics]. Trans. Zhu Guangqian 朱光潜. Beijing: Commercial Press 商务印书馆.

224  Bibliography Homer 1994. Aodesai奥德赛[The Odyssey]. Trans. Chen Zhongmei 陈中梅. Guangzhou: Huancheng Press 花城出版社. Honko, Lauri 1996. “Epic and Identity: National, Regional, Communal, Individual.” In Oral Tradittion. 11(1):18–36. Honko, Lauri 1998. Textualising the Siri Epic. Helesinki: Academia Scientiarum Fennica. Hu Shi 胡适 and Ouyang Zhesheng 欧阳哲生 1998. “Baihua wenxueshi 白话文学史 [History of Vernacular Literature].” In hushi wenji 胡适文集 [Selected Essays of Hu Shi]. Vol. 8. Beijing: Peking University Press 北京大学出版社. Huang Baosheng 黄宝生 2006. Mohepoluoduo daodu〈摩诃婆罗多〉导读 [An Introduction to Mahābhārata]. Beijing: China Social Sciences Press 中国社会科学 出版社. Hymes, Dell 1989. “Ways of Speaking.” Explorations in the Ethnography of Speaking, 2nd ed. Eds. Richard Bauman and Joel Shezer. Cambridge: Cambridge University Press. Jackson, Guida M. 1994. Encyclopedia of Traditional Epics. Santa Barbara, CA: ABC-CLIO. Jackson, Guida M. 1996. Traditional Epics: A Literary Companion. Oxford: Oxford University Press. Jamcha 贾木查 1983. “Shilun jiangge’er chansheng shijian he didian wenti 试论〈江格尔>产生时间和地点问题 [On the Time and Place of Jangar’s Origin].” In xinjiang shehui kexue 新疆社会科学[Social Sciences in Xinjiang]. 2:46–49. Jamcha 贾木查 1988. “Jiangge’er de liuchuan ji yucang gaikuang〈江格尔〉 的流传及蕴藏概况 [The Transmission and Reserve of Jangar].” In Jiangge’er lun wenji 〈江格尔〉论文集 [The Anthology of Jangar]. Eds. Xinjiang Uygur Autonomous Region Branch of Chinese Folk Literature and Art Association 中国民间文艺家协会新疆维吾尔自治区分会. Urumqi: Xinjiang People’s Publishing House 新疆人民出版社. 22–44. Jamcha 贾木查 1996. Shishi jiangge’er tanyou史诗〈江格尔〉探源 [On the Origin of the Epic Jangar]. Urumqi: Xinjiang People’s Publishing House 新疆人民出版社. Jiangbian, Jiacuo 降边嘉措 1986. Gesa’erwang zhuan yanjiu wenji〈格萨尔王 传〉研究文集 [The Anthology of King Gesar]. Chengdu: Sichuan People’s Publishing House 四川民族出版社. Jiangbian, Jiacuo 降边嘉措 1994. Gesa’er yu zangzu wenhua 〈格萨尔〉与藏族文化 [Gesar and Tibetan Culture]. Hohhot: Inner Mongolia University Press 内蒙古大学出版社. Karadžić, Vuk Stefanović 1841–1862. Srpske narodne pjesme. Vienna. Rep. 1975. Belgrade: Nolit. Krapp, George Philip and Dohbie, Elliott Van Kirk, Eds. 1942. The Anglo-Saxon minor poetics. The Anglo-Saxon Poetic Records: A collective edition. Vol. 6. New York: Columbia University Press. Kress, Gunther 1996. Before Writing: Rethinking the Paths to Literacy. London: Routledge. Liao, Mingjun 廖明君 and Bamo Qubumo 巴莫曲布嫫 2004. “Tianye yanjiu de wuge zaichang 田野研究的‘五个在场’ [Five Presences in Field Research].” In minzu yishi 民族艺术[Ethnic Arts Quarterly]. 3:14–27. Liu, Zongdi 刘宗迪 2004. “Cong shumian fanshi dao koutou fanshi: lunminjian wenxue de fanshi zhuanhuan yu xueku duli 从书面范式到口头范式：论民 间文艺学的范式转换与学科独立 [From Written to Oral: Paradigm Shift and Discipline Independence of Folk Literature and Art]. In minzu wenxue yanjiu 民族文学研究 [Studies of Ethnic Literature]. 2:57–64.

Bibliography  225 Lord, Albert B. 1959. “The Poetics of Oral Creation.” In Comparative Literature: Proceedings of the Second Congress of the International Comparative Literature Association. Ed. Werner P. Friederich, Chapel Hill, NC: University of North Carolina Press. Lord, Albert B. 1960. The Singer of Tales. Cambridge: Harvard University Press. Lord, Albert B. 1968. “Homer as Oral Poet.” In Harvard Studies in Classical Philology. 72:1–46. Lord, Albert B. 1986. “Perspectives on Recent Work on the Oral Traditional Formula.” In Oral Tradition. Ed. John Miles Foley. Slavica Publishers, Inc. Vol. 1(3):467–503. Lord, Albert B. 2004. Gushi de geshou 故事的歌手 [The Singer of Tales]. Trans. Yin Hubin 尹虎彬. Beijing: Zhonghua Book Company 中华书局. Macpherson, James 1974 [1805]. The Poems of Ossian. Manhattan: AMS Press, Inc. Martin, Richard 2005. “Epic as Genre.” In A Companion to Ancient Epic. Ed. John Mile Foley. Oxford: Blackwell Publishing Ltd. 9–19. Munro, David B. and Allen, Thomas W., Eds. 1922. Homeri: Opera. 2nd ed. 4 Vols, Oxford: Clarendon Press. Murray, Gillbert 1988. Guxila wenxueshi 古希腊文学史 [History of Ancient Greek Literature]. Trans. Sun Xizhen 孙席珍, Jiang Bingxian 蒋炳贤 and Guo Zhishi 郭智石. Shanghai: Shanghai Translation Publishing House 上海译文出版社. Nagy, Gregory 1996. Poetry as Performance: Homer and Beyond. Cambridge: Cambridge University Press. Nagy, Gregory 2008. Hema zhuwenti 荷马诸问题 [Homeric Questions]. Trans. Bamo Qubumo 巴莫曲布蟆. Guilin: Guangxi Normal University Press 广西师范大学出版社. Nekljudov, S. J. 1991. Menggu renmin de yingxiong shishi蒙古人民的英雄史诗[The Heroic Epics of the Mongolian People]. Trans. Xu Changhan 徐昌汉et al. Hohhot: Inner Mongolia University Press 内蒙古大学出版社. Nekljudov, S. J. 1996. “The Machanisms of Epic Plot and the Mongolian Geseriad.” In Oral Tradition. 11(1):133–143. Norbu, Wangdan 诺布旺丹 2003. “Fuzang gesa’er chuyi 伏藏〈格萨尔〉刍议 On gter-ma (hidden treasure) Text of Gesar.].” In gesa’er yanjiu jikan格萨尔研究集刊 [Journal of Gesar Research]. Vol. 6. Beijing: Ethnic Publishing House 民族出版社. 92–124. Olrik, Axel 1992. “The Structure of the Narrative: The Epic Laws.” In Principles for Oral Narrative Research. Trans. Kirsten Wolf and Jody Jensen, Bloomington and Indianpolis, IN: Indiana University Press. 41–61. Olrik, Axel 1999. “Epic Laws of Folk Narrative.” In International Folkloristics: Classic Contributions by Founders of Folklore. Ed. Alan Dundes. Lanham, MD: Rowman and Littlefield Publishers, INC. 83–98. Ong, Walter J. 2000. Jiyu kouchuan de siwei he biaoshu tedian 基于口传的思维和表述特点 [Characteristics of Orally Based Thought and Expression]. In minzu wenxue yanjiu 民族文学研究 [Studies of Ethnic Literature]. Supplementary Issue. 18–31. Parry, Adam 1971. The Making of Homeric Verse: The Collected Papers of Milman Parry. Oxford: Clarendon Press. Poppe, Nicholas. 1979. Heroic Epic of the Khalkha Mongols. Bloomington, IN: The Mongolia Society. (与96同) Pozdneev, V. N. 1880. Obrazcy narodnoj literatury mongol ‘skih plemen.’ Narodnyja pesni mongolov, St. Peterburg: Tipografija Imperatorsoj Akademii Nauk. (In Russian) Preminger, Alex and Brogan, Terry V. F 1993. The New Princeton Encyclopedia of Poetry and Poetics. Princeton, NJ: Princeton University Press.

226  Bibliography Radlov, V. V. (alias Friedrich Wilhelm Radloff) 1885. Proben der Volkslitteratur der nö rdlichen türkischen Stämme. Vol. 5. Der Dialect der Kara-kirgisen. St. Petersburg: Commissionare der Kaiserlichen Akademie der Wissenschaften. Ramstedt, Gustav John and J. R. Krueger 1978. Seven Journeys Eastward. 1898–1912: Among the Cheremis, Kalmyks, Mongols and in Turkestan and to Afghanistan. Bloomington, IN: The Mongolia Society. Rao, Velcheru Narayana 1986. “Epics and Ideologies: Six Telugu Folk Epics”. In Another Harmony: New Essays on the Folklore of India. Eds. Stuart H. Blackburn and A. K. Ramanujan. Berkeley, CA: University of California Press. 131–165. Reichl, Karl 1992. Turkic Oral Epic Poetry: Traditions, Forms, Poetic Structure. (Garland Reference Library of the Humanities), New York: Garland Publishing Inc. Ren Naiqiang 任乃强 1945. “Man sanguo de chubu jieshao” “蛮三国”的初步介绍 [An Introduction to the Tibetan Three Kingdoms]” In bianzheng gonglun边政公论 [Frontier Affairs]. 4–6:22–28. Rinchindorji 仁钦道尔吉 1987. “Lun xinjiang de jiangge’er he jiangge’erqi 论新疆的江格尔和江格尔奇[On Xinjiang Jangar and Jangarch ].” In minzu wenxue yanjiu民族文学研究[Studies of Ethnic Literature]. 4:23–36+51. Rinchindorji 仁钦道尔吉 1988. Jiangge’er yanjiu gaikuang〈江格尔〉研究概况[A Survey of the Research on Jangar]. In Jiangge’er lun wenji.〈江格尔〉论文集 [The Anthology of Jangar]. Eds. Xinjiang Uygur Autonomous Region Branch of Chinese Folk Literature and Art Association中国民间文艺家协会新疆维吾尔自治区分会. Urumqi: Xinjiang People’s Publishing House新疆人民出版社. 63–83. Rinchindorji 仁钦道尔吉 1989. “Menggu yingxiong shishi qingjie jiegou de fazhan蒙古英雄史诗情节结构的发展 [The Development of Plot of Mongolian Heroic Epic].” In minzu wenxue yanjiu民族文学研究 [Studies of Ethnic Literature]. 5:11–19. Rinchindorji 仁钦道尔吉 1999. Jiangge’er lun〈江格尔〉论 [A Study of the Mongolian Epic Jangar]. Hohhot: Inner Mongolia University Press内蒙古大学出版社. Rinchindorji 仁钦道尔吉 2001. “Mongolian-Turkic Epics: Typological Formation and Development.” In Oral Tradition. 16(2):381–401. Rinchindorji 仁钦道尔吉仁 2013. Menggu yingxiong shishi fazhanshi蒙古英雄史诗 发展史[The Development History of Mongolian Heroic Epic]. Beijing: China Social Sciences Press中国社会科学出版社. Sodnomrapten 索德那木拉布坦 and Batu 巴图1996. Jiange’er shouchaoben. [Jangar Manuscript]. Chifeng: Inner Mongolia Science and Technology Press 内蒙古科学技 术出版社. Somers-Willett, Susan B. A. 2009. The Cultural Politics of Slam Poetry: Race, Identity, and the Performance of Popular Verse in America. Ann Arbor, MI: University of Michigan Press. Stein, R. A. 1993. Xizang shishi yu shuochang yiren西藏史诗与说唱艺人 [Tibetan Epics and Singers]. Trans. Geng Sheng 耿昇. Lhasa: Xizang People’s Publishing House 西藏人民出版社. Tedlock, Dennis 1977. “Toward an Oral Poetics.” In New Literary History. 8(3):503–519. The Anthology of Jangar〈江格尔〉论文集, Eds. 1988. The Xinjiang Uygur Autonomous Region Folk literature and Art Association新疆维吾尔自治区民间文艺家协会. Urumqi: Xinjiang People’s Publishing House 新疆人民出版社. Thomson, Derick S. 1952. The Gaelic Sources of Macpherson’s Ossian. Edinburgh: University of Aberdeen [by] Oliver & Boyd.

Bibliography  227 Thucydides 2004. Boluobennisa zhanzhengshi 伯罗奔尼撒战争史 [History of the Peloponnesian War]. Trans. Xu Songyan 徐松岩 and Huang Xianquan 黄贤全. Guilin: Guangxi Normal University Press 广西师范大学出版社. Vasenkar, Maria 2002. “A Seminar Commemorating the Bicentennial of Elias Lönnrot’s Birth.” In FFN. 23. Vico, Giovanni Battista 1948. Xinkexue 新科学 [New Science]. Trans. Zhu Guangqian 朱光潜. Beijing: People’s Literature Publishing House 人民文学出版社. Vladimirtsov, B. Y. 1923. Menggu weilate yingxiong shishi 蒙古-卫拉特英雄史诗 [The Oirat-Mongolian Heroic Epic].  Petersburg: Gosudarstvennoe izdatelstvo Vladimirtsov, Boris Ya 1983. “The Oirat-Mongolian Heroic Epic.” In Mongolian Studies. Journal of the Mongolia Society. Trans. John Krueger. 8:5–58. Vladimirtsov, B. Y. 1993. “Menggu weilate yingxiong shishi 蒙古-卫拉特英雄史诗 [The Oirat-Mongolian Heroic Epic].” In minzu wenxue yanjiu 民族文学研究 [Studies of Ethnic Literature]. Trans. Chao Gejin 朝戈金. 1:91–94. Wellek, René and Austin Warren 2005. Wenxue lilun 文学理论 [Theroy of Literature]. Trans. Liu Xiangyu et al. 刘象愚. Nanjing: Jiangsu Education Puplishing House江苏教育出版社. Wen, Yiduo 闻一多 1997. “Ge yu shi 歌与诗 [Song and Poem].” In shenhua yu shi神话与诗 [Myth and Poetry]. Shanghai: East China Normal University Press 华东师范大学出版社. Wu, Gang 吴刚 2011. “Cong sere wuqin kan dawo’erzu koutou yu shumian wenxue guanxi 从色热乌钦看达斡尔族口头与书面文学关系 [The Relationship between Oral and Written Literature of the Daur from the Perspective of Sere Uchin].” In wenxue yu wenhua 文学与文化 [Literature and Culture]. 3:132–133. Xiao, Bing 萧兵 1989. “Fengshen yanyi de nishishixing jiqi shengcheng 〈封神演义>的拟史诗性及其生成 [The Quasi Poetics and Its Formation of Creation of the Gods].” In mingqing xiaoshuo yanjiu 明清小说研究 [Journal of Ming-Qing Fiction Studies]. 2:112–125. Yang, Enhong 杨恩洪 1995. Minjian shishen: gesar yiren yanjiu 民间诗神：格萨尔艺人研究 [The Muses of Folk Poetry: Studies on the Performers of Gesar]. Beijing: China Tibetology Publishing House 中国藏学出版社. Yang, Lihui 杨利慧 2004a. “Biaoyan lilun yu minjian xushi yanjiu 表演理论与民间叙事 研究 [Studies of Performance Theory and Folk Narrative].” In minsu yanjiu 民俗研究 [Folklore Studies]. 1:30–46. Yang, Lihui 杨利慧 2004b. “Minzuzhi shixue de lilun yu shijian民族志诗学的理 论与实践[Ethnopoetics: Theory and Practice].” In beijing shifan daxue xuebao 北京师范大学学报·社会科学版 [Journal of Beijing Normal University]. Social Science. 6:49–54. Ye, Shuxian 叶舒宪 2005. Yingxiong yu taiyang-zhongguo shanggu shishi de yuanxing chonggou 英雄与太阳-中国上古史诗的原型重构 [Heroes and the Sun: Archetypal Reconstruction of Ancient Chinese Epics]. Xi’an: Shaanxi People’s Publishing House陕西人民出版社. Yin, Hubin 尹虎彬 2002. Gudai jingdian yu koutou chuantong古代经典与口头传统 [Ancient Classics and Oral Traditions]. Beijing: China Social Sciences Press 中国社会科学出版社. Zhao, Bingli, Ed. 赵秉理 1990. Gesa’erxue jicheng 格萨尔学集成 [The Collected Papers on Gesar Studies]. Vol. 1. Lanzhou: Gansu People’s Publishing House甘肃人民出版社.

228  Bibliography Zheng, Zhenduo 郑振铎 1998. Zhengzhenduo quanji 郑振铎全集 [The Complete Works of Zheng Zhenduo]. Vol. 1. Shijiazhuang: Huashan Literature and Art Publishing House.花山文艺出版社. Zumthor, Paul and Marilyn C. Engelhardt 1984. “The Text and the Voice.” In New Literary History. 16(1):67–92. Zumwalt, Rosemary L. 1998. “A Historical Glossary of Critical Approaches.” In Teaching Oral Traditions. Ed. John Miles Foley. New York: The Modern Language Association. 75–94.

Index

Note: Page numbers in italics indicate figures, bold indicate tables in the text, and references following “n” refer endnotes. 20th-century, epic studies 38 21st century: Chinese epic poetry in 27–29; Chinese epic research 4 Alcheringa: Ethnopoetics (journal) 10, 103 Alexander the Great 5, 18, 81 Alexandrian Greece 113; see also ancient Greek oral epic tradition Altaic epic traditions 148 American folklore 55, 108 American Slam Poetry 55 analogical etymological approach 62 Analysts vs. Unitarians 6, 8, 10, 40, 81 Antin, David 10 aoidós (bard) 7, 29n8 Apollodorus of Athens 18 Apollonius 18, 82 “Apology for Heroic Poetry and Poetic License” (Dryden) 33 Aporēmata Homērika (Homeric Questions) 33; see also Homeric Question Argonautica (Hellenistic epic) 18 Arimpil 4, 21–27, 134; epithet formulae in 158, 158; formulaic epithet “giant red Hongor” 153, 154; Homer’s poems 158, 158 Aristotle 5, 33 Art of Poetry (Boileau-Despréaux) 33 Art of Poetry (Vida) 33 audiences 46–48, 104; see also epic, audiences Bader, F. 15 The Ballads of Ossian (Ossian) 17 Bamo Qubumo 13 Bascom, William 43

Bayar, Hulbar 134 Bender, Mark 29n7 Beowulf (Anglo-Saxon) 32, 39n1, 40, 75, 107 Bian Yuan 97 Black Yi 37 Bobrovnikov, A. 132 Bowra, Cecil M. 34, 35 Brothers Epics, Dravidian 70 Buural, Sisina 134 Canto 15, 16, 97 Center for Studies in Oral Tradition (CSOT) 36, 50, 51, 213 Charlie (Prince) 20 Child, Francis James 35 China’s illiteracy rate 56 Chinese epic poetry 27–29 Chinese epic research 91–95 Chinese epic studies 36–38 Christian oral epics 186 Christian tradition 18, 189, 199 Chronika 18 comparative epic poetics 44–45 composition-performance-diffusion model 62 The Consequences of Literacy (Goody and Watt) 35 cross-cultural traditional aesthetic issues 10 cross-genre analysis 9 cross-linguistic research 25 cultural functions 46, 57, 65 Cultural Revolution 122, 132, 135 cultural system 26, 66–67, 74 Cumhail, Finn Mac 17

230 Index dēmioergoi 7 Dēmodokos 7 Diamond, Stanley 10 Die Thaten Bogda Gesser Chan 22 Divine Comedy (Dante) 18 Dravidian Tulu- and Kannada-language 68 dream-taught bard (gnyid lam bab) 4 dream-taught singers 37 dream-taught storytelling 110; see also storytelling Elements of Criticism (Lord Kames) 33 enjambement 80, 82, 197 An Enquiry into the Life and Writings of Homer (Blackwell) 33 epic: audiences 46–48; authenticity 48–50; composition 95; definition of 42–44, 65, 113–115; elasticity 116–118, 117; inheritance 24, 26, 37, 63; length 115–116; singers 46–48; study 40–42; transmission 46–48 epic cycle 130–131 epic register 52, 54, 93 epic texts: analysis of 66; complex formation of 12; creative analysis of 83; Mongol 96; traditional connotations of 65 epic traditions 104; Altaic 32, 148; Chinese 24, 28, 37; five presences of 25; Greek 79; Indian 47, 62; Mongolian 141, 146; Oirat 47; reconstruction of 14–17; Tibetan 110; see also oral epic tradition Epithalamion 18 Essay on Epic Poetry (Voltaire) 33 An Essay on Epic Poetry (Hayley) 33 “Essay on Poetry” (Mulgrave) 33 An Essay on the Original Genius of Homer (Wood, Robert) 6, 61 ethnic identity 46–48 ethnopoetics 10, 16, 25, 55 Ethnopoetics and Performance Theory 9–10 Etymology 33 eulogy 107, 110 The Faerie Queene (Spenser, Edmund) 18 Fine, Elizabeth C. 10, 108 Finga (Macpherson, James) 17 Finske-Svenskt Lexikon 11 Foley, John Miles 10, 19–21, 34, 36, 51; internet and electronic media 57–58; Oral Formulaic Theory 51–53; oral poetics 53–54; oral tradition 54–57

folk art 106–109, 121 folk literature 108; Chinese 24; formulae 84; independent status of 26; oral 55 folklore: American 55, 108; theory 10–11 The Folklore Text: From Performance to Print (Fine) 108 folk oral epic literature 91 Forest Folks 130 formulaic density 95, 151, 159–164 formulaic diction 141, 149, 210 formulaic phrases 92, 143, 144, 161; in Classic of Poetry 164, 164 formulaic themes 8 Fragments of Ancient Poetry (Macpherson, James) 17 Franke, A. H. 22 From Virgil to Milton (Bowra, C. M.) 34 Gaelic narrative 17, 19, 73 Gao Hehong 109 Gejin, Chao: Chinese epic research 91–95; interview 91–95; oral poetics 91–95 Genesis 13, 190 Genghis Khan 130 Gesar 22, 36, 37; of Tibetan areas 43 Gesar of Ling (Tibetan Three Kingdoms) 23 Geser 47 Gilgamesh (Babylonian) 39n1 Goethe, Johann Wolfgang von 17, 33 Golden Fleece 18 Golstunskij, F. K. 97, 98, 132 Goody, Jack 103 Greek oral epic tradition, ancient: Foley, John 36; poem in 189–190; poetic line verse in 199–200; register in 210–211 Greek oral poetic tradition, ancient: formula in 204–205; typical scene in 195 group identity 53, 65–67, 69 Guo Songtao 21 The Gutenberg Galaxy (McLuhan, Marshall) 35 handwritten texts 98–100 Hartmann, C. J. G. 37 Hatto, Arthur T. 35 Haymes, Edward R. 115 Heissig, Walther 37, 85 Hermann, Johann 6 heroic epic 22, 23, 43, 85, 96 Heroic Poetry (Bowra, C. M.) 34 Homeric epic tradition 10, 40 Homeric Hymn to Apollo 5 Homeric poetry: and oral tradition 81–82

Index  231 Homeric Questions 5–7; 18th century 5; 19th century 4 Homeridae (Homer’s offspring) 5 Homerische Untersuchungen (WilamowitzMoellendorff, Ulrich von) 6 Homer: The Origins and the Transmission (Allen, Thomas W.) 7 Honko, Lauri 13, 16, 34, 52, 113, 114; epic identity, personal and community 68–71; identity-bearing thrust of epic 65–68; Indian epic 74–76; national identity epic 71–74 Horqin epic 47 How to Read an Oral Poem (Foley) 19 human thinking 57 Humboldt, Wilhelm von 74 Hu Shi 36 Huur-un tiliger storytelling 164 Hymes, Dell 10, 103 Hyvonen, Jouni 11 Iliad (Homer) 5, 6, 39n1 Indian cultural system 74 Indian epic tradition 47, 62 information technology (IT) 51, 57 International Society of Oral Tradition Research (ISSOT) 51 Interpretation of Aristotle’s Poetics (Castelvetro) 33 intertextuality 38, 104 Irish Fenian Cycle 17 Jackson-Laufer, Guida M. 114 Jangar (Mongol epic) 16; formulae 143–146; formulaic names in 153, 153; language 141–143, 143; Oirat epic cycle of 130–131; performance 138–140; and related epic traditions 146–148; research 148–149; singers 134–137; storypatterns 140–141 jangarch (or jangarchi) 16 Jangar Material 133 Jangar texts 131–134 Jerusalem Liberata (Tasso) 18 Johnson, Samuel 17, 73 Josephus, Flavius 5, 81 Kalevala (Lönnrot, Elias) 11, 48; composition of 12 Kalmyks 72, 96, 97, 121, 148 Kantele 11 Kanteletar 11 kernel theory 6 King James Bible 17

King Yalu (Miao) 43, 44, 47, 48 King Zhenu of Esuo 13 King Zhiga Alu 13 Kittredge, George Lyman 35 Kotwicz, W. L. 97, 132 Kovalevskij, O. 97, 132 Lachmann, Karl 6 Laocoon: An Essay on the Limits of Painting and Poetry (Lessing) 102 Laozi 20 La Pensée Sauvage (Lévi-Strauss, Claude) 35 Lectures on Aesthetics (Hegel) 33 Lévi-Strauss, C. 35, 103 Liang Qichao 21 Liedertheorie (Lachmann) 6 Lönnrot, Elias 11–14 Lord, Albert B. 8, 34, 41, 62 Lower Ladakhi Version of the Kesar Saga (Franke, A. H.) 22 Macpherson, James 17, 73 Madame de Staël 17 Mahābhārata (Indian epic) 14–17, 39n1, 42, 47, 74–76 Mamay, Jüsüp 4, 7, 24, 49 Mambrun, Jesuit 33 Manas (Kirgiz epic) 13, 22, 40, 47, 107, 115, 147 Manchu traditional storytelling 52, 109 mangus 128, 129 mental text 13, 109 meter, poetic 168–172 methodological revolution 10 Mikhailov, G. L. 97, 132 modern epic studies 34–36 Mongolian epic formulae: Arimpil’s use of formulae 164–167; epithet formulae 152–158; formulaic density 159–164; research method 150–151 Mongolian epic prosody: alliteration 172–175; antithetical parallelism 177–180; assonance 175–177; compound parallelism 181–182; end rhyme 175–177; head rhyme 172–175; internal rhyme 175–177; parallelism 177–182; poetic meter 168–172; progressive parallelism 180–181 Mongolian epics 37, 117; Choibang (singer) 59; heroic spirit 43; poetry 128; Rinchindorji 118; traditions 41, 59, 146 Mongolian oral epic poetry 121–122; Barag 122–124; epic figures 127–129; Horqin

232 Index 122–124; Oirat 122–124; structure and theme 124–127; structure of 128 Mongolian oral epic tradition: poem in 187–188; poetic line verse in 196–198; register in 206–208 Mongolian oral poetic tradition: formula in 201–203; typical scene in 191–193 Mongol nomadic traditions 43 motif series 85 The Mountain Wreath (Niegoš) 18 Murko, Matija 8, 34, 87 Muse 15 musical context 197 Muslim epic tradition 94 Muslim oral epics 186 Muslim oral traditions 18 Muslim vs. Christian epics 214n1 Nagy, Gregory 7, 10, 34, 61–64 Naika, Gupala 110 narrative ethnography 55 national culture and identity 67 Nekljudov, S. J. 37, 90 Neo-analysts 7 Neo-unitarians 7 New Kalevala 11 The New Princeton Encyclopedia of Poetry and Poetics (Preminger) 42 New York Review of Books (Sokal) 18 Niegoš, Petrović 18–20, 49 Nitzsch, Gregor Wilhelm 6 Novel Monthly (Homer’s epics) 21 nyki (ghosts) 6 Odyssey (poem by Homer) 5, 6, 29n2, 39n1 Oirat epic cycle: of Jangar 130–131 Oirat epic traditions 47 Old English oral epic tradition: poem in 190–191; poetic line verse in 200–201; register in 211–214 Old English oral poetic tradition: formula in 205–206; typical scene in 195–196 Old Kalevala 11 Olrik, Alex 34, 105 On Epic and Dramatic Poetry (Goethe and Schiller) 33 On Epic Poetry (Mambrun, Jesuit) 33 On Poetics (Traité du poème épique) (Bossu) 33 On the Heroic Epic (Tasso) 33 oral composition 56, 63 oral-connected text 13

oral context 150 oral culture 35 oral-derived text 12, 13 oral epic texts, types of: handwritten texts 98–100; live recorded text 100; orally dictated texts 97–98; printed texts 100–101; retold, paraphrased texts 96–97 oral epic tradition, poetic line verse in: Ancient Greek 199–200; Mongolian 196–198; Old English 200–201; South Slavic 198–199 oral epic tradition, register in: Ancient Greek 210–211; Mongolian 206–208; Old English 211–214; South Slavic 208–209 oral epic traditions, poem in: Ancient Greek 189–190; Mongolian 187–188; Old English 190–191; South Slavic 188–189 oral epic transmission 86–87; Croatian oral epic 87–89; Mongol heroic epic 89–90; Serbian oral epic 87–89 Oral Formulaic Theory 4, 8–10, 83, 103; in 20th century 6; Lord’s contributions to 83–84; Milman Parry of 34; significance of 84–85 orality theories: implications of 7 orally dictated texts 97–98 oral nature: of Homer’s poetry 82–83; Parry’s determination of 82–83 oral poems: classification of 19, 19; text types of 12, 12 oral poetics 10, 79–81, 91–95, 102–107 oral poetic tradition, formula in: Ancient Greek 204–205; Mongolian 201–203; Old English 205–206; South Slavic 203–204 oral poetic tradition, typical scene in: Ancient Greek 195; Mongolian 191–193; Old English 195–196; South Slavic 193–194 oral poetry interpretation 17–21 oral-related text 13 Osmanbeg Delibegovic and Pavicevic Luka 8 Ossian 17–21 Ovlaa, Eela 134 Oxford Dictionary of Literary Terms (Baldick) 42 Pabuji, western India’s Rajasthan epic 70 Pallas, P. S. 22, 37 Palnâtiviracaritra, Telugu epic 74

Index  233 Panathenaea, festival of 5 paper singers 46 Paradise Lost (Milton) 18 parallelism 177–182; antithetical 177–180; compound 181–182; progressive 180–181 Parry, Milman 34, 81 Parry–Lord Theory 8, 53, 55, 83, 84, 107; see also Oral Formulaic Theory Peisistratos (Athenian ruler) 81 performance events 104 performance theory 55, 93 pervasiveness 114 Phēmios 7 Philosophy of Art (Schelling) 33 Plato 32 The Poems of Ossian (Macpherson, James) 17, 73 poetic register 79 poetics: of mental 109–111; of oral texts 107–109, 108; of voice 111 Poetics of Oral Creation (Lord) 102 poetic tradition, cultural soil of 80 poiētēs (poetry maker) 7 Poppe, Nicholas 37 Pozdneev, V. N. 97, 98, 168 Preface to Plato (Havelock, Eric A.) 35 printed texts 100–101 Prolegomena ad Homerum (Wolf, F. A.) 6 quasi-epic 47 Radloff, Friedrich Wilhelm 8; see also Radlov, V. V. Radlov, V. V. 8, 34 Rāmāyana 47 Rāmāyana (Indian) 39n1 Ramstedt, G. J. 121, 125 Reflections on Aristotle’s Poetics (Rapin) 33 Reichl, Karl 34, 37 Ren Naiqiang 36 researchers 104 retold, paraphrased texts 96–97 rhapsōidos (singer) 7 Romanticism 4, 11, 17 Rothenberg, Jerome 10 run-on lines 82–84 Sananlaskuja 11 Schmidt, L. J. 22 Schott, W. C. 37 Scott, John A. 6

Scott, Walter 17 scribbled text 46 Serbian literature: greatest poet of 18; magical practices 55; Muslim epic tradition 94 The Serbian Mirror (Niegoš) 18 Serbo-Croatian: Heroic Songs 29n4; Muslim epic tradition 94; oral epic 86 Shepeardes Calendar 18 The Singer of Tales (Lord) 9, 33–35, 58–60 The Singer of Tales in Performance (Foley) 55 singers 46–48, 134–137 Siri Epic of India 44, 46 Sixth Panchen Lama 22, 36 Snyder, Gary 10 social functions 46 Social Text 17 Socrates 32 Sokal, Alan 18 Song of Bagdad (Ugljanin) 9 The Song of the Nibelungs 6, 82, 88, 112, 115 South Slavic oral epic tradition: poem in 188–189; poetic line verse in 198–199; register in 208–209 South Slavic oral poetic tradition: formula in 203–204; typical scene in 193–194 Spenser, Edmund 18 story-patterns 140–141 storytelling 109; by folk singer 106; oral 105; skills 105; South Slavic 209; vocabulary 185 Sumpa Yeshé Penjor 22, 36 sung and written, concepts and classifications of 79 superstories 65 super tradition 67 Tarn, Nathaniel 10 Tedlock, Dennis 10 Temora (Macpherson, James) 17 text-making patterns 10 text technology (TT) 57 text types: of oral poems 12, 12 textual analysis 104 textual evolution: of Homer’s poetry 63 textuality 63 themes: accumulation of 84; contemporary 19; formulaic 8; traditional formulaic 34 Tibetan epic tradition 110 Tibetan Gesar 13, 22

234 Index Tibetan hidden treasure texts 24 traditional classical studies 62 traditional connotation 80 traditional context 186 traditional language and art 54 traditional referentiality 53, 54 tradition-culture-identity 67 tradition-oriented text 12–14 Trissino 33 Trojan War 6 Turkic Oral Epic Poetry: Traditions, Forms, Poetic Structure (Reichl) 37 typical scene 185, 186

Wars between the Black and the White 13 The Wedding of Mustajbeg’s Son Bećirbey 53 The Wedding of Smailagic Meho (Avdo) 8, 87 Western epic studies 32–34 Western written poetry 80 White Yi 37 Wolf, F. A. 6, 61 writing inheritor 109 written epics 65 written poetic texts 103–107 written vs. oral communication 49

UNESCO 54 Unitarians 61; vs. Analysts 6, 8, 10, 40, 81

Yang Enhong 30n14 yasun-u jangarch 137 Yi 4 Yin Hubin 27 Yugoslav oral tradition 36 Yugoslav poetic tradition 54

Vico 5, 6, 81, 82 Vietnamese epics 113 Vladimirtsov, B. 89, 121, 126 Voice of Mountains (Niegoš) 18 vowel harmony 176 Vujnovic, Niko La 8 Vyāsa 14–17 Waley, Arthur 112

Zhamcarano, C. 121 Zhang Taiyan 21, 36 Zhirmunsky, V. 35, 37 Zhong Jingwen 23 Zumwalt, Rosemary L. 103