The Oral Epic: From Performance to Interpretation 2021002065, 9780367761318, 9781003189114

This book focuses on the performance of oral epics and explores the significance of performance features for the interpr

729 104 11MB

English Pages [282] Year 2021

Report DMCA / Copyright


Table of contents :
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
List of Illustrations
A Note on Transcription and Transliteration
Part I: Settings
Chapter 1: How to Identify an Oral Epic
Oral: Shades and Grades
The Challenge of Native Classification
An African Interlude
The Uzbek Dastan
Chapter 2: The Singer
Epic Singers: Types and Terms
Women Singer-Narrators
The ‘Reader of Tales’
How to Become an Epic Singer
South Slavs
The Chain of Transmission
Singers’ Schools
The Singers’ Awareness of Transmission
Creativity and Innovation
Changes in the Plot: Happy Endings
The Composition of New Epics
The Notion of Authorship
Chapter 3: Introducing Performance
The Ethnography of Communication
The Setting
The Message
The Event
Part II: Performance
Chapter 4: Voice
Shamanic Voices
Central Asian Overtone-Singing
Altaian, Khakas, Shor qay/khay
Chapter 5: Gesture
Conventional Gestures: The Karakalpak Jïraw
Stylized Gestures: The Kyrgyz Manaschï
Gesture and Inspiration
Gesture, Miming, Stage Props
The Turkish meddah
Rajasthan: Reading the paṛ
Chapter 6: Oral Epics as Songs
Song as Vehicle, Song as Music
Monotonic Narrative Melodies – Papua New Guinea
Solo Singing
Choral Singing
‘Riding the Song’: The Singing of the Kyrgyz Epic Manas
Music and Metre: Some Examples
Rajasthan: Pābūjī
West Africa
Chapter 7: Voice and Instrument
Gusle, Qobïz, Horse-Head Fiddle
Case Study: Karakalpak
Lute, Dutar, Dombira
Case Study: Uzbek
The Interplay of Song and Instrument
Part III: Interpretation
Chapter 8: Words, Music, Meaning
Meaning and Expression
What’s in a Name?
The Tibetan Gesar
Turkic Terms for ‘Melody’
Köroğlu in Turkish and Azeri
Case Study: Alpamïs
Leitmotifs in Siberian Oral Epics
Expression and Convention
The Fulani Motto
Chapter 9: The Singer and the Tale
Point of View
Mythological Epics, Sacred Time
First-Person Narration, Shamanistic Traces
The Narrator’s Presence in the Narrative
Chapter 10: Performance and Interpretation
Visualization and Imaging
Aria and Recitative
Case Study: Ghärip-Ashïq
From Context to Text
Performance Contexts
Texts without Context
Appendix A: Notes on Oral Epic Traditions
Sub-saharan African Epic Traditions
North African Epic Traditions
South East Asia, the Far East, and the Pacific
Slavic Epic Traditions
Finno-Ugrian Epic Traditions
The Tungus
Mongolian Epic Traditions
Turkic Oral Epics
Appendix B: Audio/Video Examples
Appendix C: Discography
Glossary of Musical Terms and Non-Western Instruments
Musical Terms
Musical Instruments
Recommend Papers

The Oral Epic: From Performance to Interpretation
 2021002065, 9780367761318, 9781003189114

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

The Oral Epic

This book focuses on the performance of oral epics and explores the significance of performance features for the interpretation of epic poetry. The leading question of the book is how the socio-­cultural context of performance and the various performance elements contribute to the meaning of oral epics. This is a question which not only concerns epics collected from living oral tradition, but which is also of importance for the understanding of the epics of antiquity and the Middle Ages which originated and flourished in an oral milieu. The book is based on fieldwork in the still vibrant oral traditions of the Turkic peoples of Central Asia and Siberia. The discussion combines fieldwork with theory; it is not limited to Turkic epics but branches out into other oral traditions. Karl Reichl is Professor Emeritus of the University of Bonn (Institute of English, American and Celtic Studies). He has had visiting professorships at Harvard University, the Hebrew University of Jerusalem, the École Pratique des Hautes Études in Paris, the University of Madison at Wisconsin, and the Karakalpak State University in Nukus. His main research interests lie in medieval oral literature and in contemporary (or near-­contemporary) oral epic poetry, especially in the Turkic-­speaking areas of Central Asia.

Routledge Studies in Medieval Literature and Culture

Zöopedagogies Creatures as Teachers in Middle English Romance Bonnie J. Erwin Before Emotion The Language of Feeling, 400–1800 Juanita Feros Ruys Forging Boethius in Medieval Intellectual Fantasies Brooke Hunter Sanctity and Female Authorship Birgitta of Sweden & Catherine of Siena Edited by Unn Falkeid and Maria H. Oen Tracing the Trails in the Medieval World Epistemological Explorations, Orientation, and Mapping in Medieval Literature Albrecht Classen Vernacular Verse Histories in Early Medieval England and Francia The Bard and the Rag-picker Catalin Taranu Polyphony and the Modern Edited by Jonathan Fruoco The Oral Epic From Performance to Interpretation Karl Reichl Barbarians in the Sagas of Icelanders Homegrown Stereotypes and Foreign Influences William H. Norman For more information about this series, please visit:

The Oral Epic From Performance to Interpretation Karl Reichl

First published 2022 by Routledge 605 Third Avenue, New York, NY 10158 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2022 Taylor & Francis The right of Karl Reichl to be identified as author of this work has been asserted by him 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. Library of Congress Cataloging-­in-­Publication Data Names: Reichl, Karl, author. Title: The oral epic : from performance to interpretation / Karl Reichl. Description: New York : Routledge, 2021. | Series: Routledge studies in medieval literature and culture | Includes bibliographical references and index. | Summary: "This book focuses on the performance of oral epics and explores its significance for interpretation. The discussion is also relevant for the understanding of medieval and earlier oral-derived epics. The study is based on field-work on the oral traditions of the Turkic peoples of Central Asia and Siberia"-- Provided by publisher. Identifiers: LCCN 2021002065 | ISBN 9780367761318 (hardback) | ISBN 9781003189114 (ebook) Subjects: LCSH: Epic poetry--History and criticism. | Oral tradition in literature. | Epic poetry, Turkic--Asia, Central--History and criticism. Classification: LCC PN1303 .R35 2021 | DDC 809.1/32--dc23 LC record available at ISBN: 978-­0-­367-­76131-­8 (hbk) ISBN: 978-­1-­032-­03808-­7 (pbk) ISBN: 978-­1-­003-­18911-­4 (ebk) Typeset in Sabon by SPi Global, India Access the Support Material:


List of Illustrations A Note on Transcription and Transliteration Preface Introduction

viii ix xi 1


Settings 1 How to Identify an Oral Epic Oral: Shades and Grades The Challenge of Native Classification An African Interlude The Uzbek Dastan 2 The Singer Epic Singers: Types and Terms How to Become an Epic Singer The Chain of Transmission Creativity and Innovation 3 Introducing Performance The Ethnography of Communication Textualization

9 11 12 14 16 18 23 24 28 34 39 48 50 59

vi Contents PART II



4 Voice


Speaking Singing Shamanic Voices 5 Gesture Conventional Gestures: The Karakalpak Jïraw Stylized Gestures: The Kyrgyz Manaschï Gesture and Inspiration Gesture, Miming, Stage Props 6 Oral Epics as Songs Song as Vehicle, Song as Music ‘Riding the Song’: The Singing of the Kyrgyz Epic Manas Music and Metre: Some Examples 7 Voice and Instrument Gusle, Qobïz, Horse-­Head Fiddle Lute, Dutar, Dombira The Interplay of Song and Instrument

68 73 75 85 88 90 97 101 107 110 116 124 130 132 136 143




8 Words, Music, Meaning


Meaning and Expression What’s in a Name? Imitation Leitmotifs in Siberian Oral Epics Expression and Convention 9 The Singer and the Tale Point of View Mythological Epics, Sacred Time

152 155 163 167 172 179 180 184

Contents  vii First-Person Narration, Shamanistic Traces The Narrator’s Presence in the Narrative

188 193

10 Performance and Interpretation


Visualization and Imaging Aria and Recitative From Context to Text

203 208 216

Appendices A Notes on Oral Epic Traditions 228 B Audio/Video Examples 234 C Discography 237 Glossary Bibliography Index

239 243 264

Illustrations Illustration 1

Gulum Ilmetov and spike-­fiddle player (1995)


Illustration 2

Jumabay Bazarov (1)


Illustration 3

Jumabay Bazarov (2)


Illustration 4

Jumabay Bazarov (3)


Illustration 5

Döölötqan (1)


Illustration 6

Döölötqan (2)


Illustration 7

Döölötqan (3)


Illustration 8

Döölötqan (4)


Illustration 9

Döölötqan (5)


Illustration 10 Döölötqan (6)


Illustration 11 Yakut olonkhosut 98 Illustration 12 Jüsüp Mamay (c. 2000)


Illustration 13 Chāri Khojamberdi-­oghli (c. 1980)


Illustration 14 Jaqsïlïq Sïrimbetov (1983)


Illustration 15 Genjebay Tilewmuratov (1990)


A Note on Transcription and Transliteration

The Cyrillic letters of Russian are transliterated according to the widely used Anglicized system. Note that Cyrillic is here always transliterated as ; it is, however, pronounced /ye/ at the beginning of a word and after a vowel. Cyrillic < э> is transliterated as . Palatalized consonants are marked by (, etc.). Most of the Turkic languages here exemplified use the Cyrillic alphabet (with individual modifications); Turkish, Azerbaijanian (or Azeri), Uzbek, Turkmen, and Karakalpak use the Latin script. For Turkish and Azeri the Latin writing system is used. All the other Turkic languages, irrespective of whether they use the Cyrillic or the newly introduced Latin script, are written in a unified system, modelled on Turkological practice. The letters , , , , symbolize clear vowels as in Italian; and correspond to the sounds found in French peu and mur, respectively; (in Uzbek) is a long, back a-­sound, as in English ball; is a back unrounded /i/; symbolizes an open /e/. For some Turkic languages (also Mongolian) long vowels are represented as double vowels, such as . Most consonants correspond roughly to their English equivalents; is pronounced as the in English jet; in some languages (Kazakh, Karakalpak) it is pronounced as the in English genre; is always rolled; is a voiceless velar fricative as in Scottish loch, is the voiced variant of ; is a velar /k/; the pronunciation of varies between /v/ as in English vat and /w/ as in English why, is generally pronounced as /au/, as /ou/. In Turkish and Azerbaijanian orthography corresponds to , to , to , and to ; the latter is not pronounced in Standard Turkish, but lengthens the preceding vowel. In some words (of Arabic origin) instead of is found. In Azeri the open /e/ is symbolized by ; in both Turkish and Azeri corresponds to ; in order to distinguish capital from capital , the latter is written . Despite my attempt at synchronizing the various transliteration systems for the Turkic languages, some words shared by several languages

x  A Note on Transcription and Transliteration appear in different spellings. As far as possible, I have tried to keep dialect variations of terms and names to a minimum. For Yakut terms I have followed Russian usage. The Yakut term for the epic is olongkho, but it is customary to refer to the epic by the term olonkho. In Yakut, intervocalic /s/ becomes /h/. The epic singer is called olongkhohut in the original language, but designated by olonkhosut in the literature; the same applies to other terms such as abaasïï (Yakut abaahïï). In my transliteration of Yakut, is transliterated as and as . Persian and Arabic words are transliterated according to the system employed by the International Journal of Middle Eastern Studies. Note that the letters , , , and with a dot underneath denote so-­ called emphatic sounds in Arabic words. Macrons denote long vowels; in Persian stands for a long open /o/. In the transliteration of Indian scripts, letters with a dot underneath (as in paṛ) denote retroflex sounds. For African languages, I have used the spellings found in my sources. They show some variation depending on whether they come from French or English-­speaking scholars, or from Francophone or Anglophone parts of Africa. Some of the variations (such as mâbo vs. maabo) are mostly typographical, others (such as the name of the epic Sunjata) are due to linguistic variation.


Un libro de ciencia tiene que ser de ciencia; pero también tiene que ser un libro. A book of science must be about science, but it must also be a book. José Ortega y Gasset This is one of the guiding principles prefaced to Ernst Robert Curtius’ monumental European Literature and the Latin Middle Ages, originally published in German in 1948. It expresses the dual nature of a scholarly book: it must conform to scientific standards, but it must also be readable. The author of a scholarly book wants to say something to the scientific community, something of interest, something that will advance our knowledge, however minimally. But the author also wants to be read. This implies walking a tight-­rope, balancing between technical jargon for the initiated and comprehensibility for a wider circle of potential readers. Every book is a new venture in its attempt to be ‘de ciencia’ and to be ‘un libro’. I sincerely hope that this study on the oral epic is both – and that it will find readers. Although writing is generally a solitary occupation, it is never without friends and companions. To acknowledge my debt to all the individuals who have not only helped me with this particular book, but have guided me through my scholarly life, leading to this publication, a long list would be necessary. From among the many teachers, colleagues and friends who have encouraged me to pursue the study of oral epic poetry, I would like to thank specifically Professor Joseph Harris of Harvard University. Through his initiative I had the opportunity of teaching Turkic oral epics as visiting professor at Harvard in 1990, a unique experience and a decisive influence on my future research. I also owe a great debt to all the singers and narrators I met and recorded in the various Turkic-­speaking areas where I was able to do fieldwork. I began my fieldwork at the time when the Soviet Union was still in existence and when the Chinese province of Xinjiang was just beginning to open up to Western scholars. In that period it would have been impossible to do fieldwork without

xii Preface the help of local scholars and the universities and academic institutions behind them, in my case in particular the Uzbek Academy of Sciences and the Chinese Academy of Social Sciences. I am grateful to Professors Qabul Maqsetov, Tora Mirzaev, Sarïgül Bahadirova, Mambetturdu Mambetaqun, Yang Ling, Chao Gejin, and many others for their support. For help with specific questions in connection with this book, I would like to thank a number of scholars from various Turkic-­speaking areas: Aitalina Kuz’mina (Yakutsk), Busarem Imin (Ürümchi), Talantaaly Bakchiev (Bishkek), Alymkan Jeenbekova (Osh), Monire Akbarpouran (Iranian Azerbaijan), Säydin Ämirlän (Nukus), and Jabbār Eshānqulov (Tashkent). For critical suggestions, I am very grateful to my colleague and friend Dr Julia Rubanovich (Jerusalem) and to my friend since high school days, Dr Joyce Irwin (Princeton). My special thanks are due to my wife Deborah, who has devoted many an hour that she could have spent more pleasantly to the drudgery of correcting the English of my manuscript. Despite all this help, mistakes and inaccuracies may still be found, for which I alone am responsible. Bad Honnef, December 2020


The era of epics is past: it is now the era of statistics. This is one of the impressions the writer Joseph Roth brought back from his travels in the Soviet Union in 1926. Roth was a shrewd observer and his reports, originally published in the Frankfurter Zeitung, make fascinating reading, also in view of what was still to come after 1926. This pithy saying catches the spirit of the age: the burst of industrialization and the impending collectivization. The ‘era of epics’ might have been past in the young Soviet Union by 1926, but epics, in the literal sense as epic poems, had not yet disappeared, neither in Russia nor in the vast area of Soviet Central Asia. Substantial collections of Russian narrative songs, bylinas, were still recorded in the 1950s, and in some areas of Central Asia and Siberia oral epic traditions are still alive at the beginning of the twenty-first century. It is in Central Asia that my own field-work and research on oral epic poetry is centred and where I was first confronted with the questions that led to the writing of this book. I would therefore like to begin with a personal note. I started my first research trip to Central Asia, supported by a stipend of the German Research Foundation, on 1 June 1981. Coming from Frankfurt with Lufthansa I arrived in Moscow’s Sheremetevo Airport together with a group of German journalists, who were making jokes and having an animated conversation till they, like everybody else, were silenced by the gloom that long and very slowly moving queues in front of the Soviet passport control booths used to spread. I was taken care of by the Soviet Academy of Sciences and put on a flight to Tashkent on the next day. By 1981 I had published two small books of Uzbek and Turkmen folktales in the original language and in German translation, but was otherwise, by my publications and training, a medievalist and linguist, working in an English department. My interest in oral

2 Introduction epics was kindled as a student by my reading, and Wilhelm Radloff’s magnificent multi-volume edition (and German translation) of Turkic oral poetry, housed in the Bavarian State Library, had motivated me to try to travel to Soviet Central Asia. In Tashkent I was met by Tora Mirzaev, Head of the Folklore Department of the Pushkin Institute of the Academy of Sciences of the Uzbek Soviet Socialist Republic. That was in 1981. In the 1990s Tora Mirzaev went on a pilgrimage to Mecca and thereupon called himself Tora-hāji Mirza. Also after independence, the Pushkin Institute was re-named as the Alisher Navā’ī Language and Literature Institute in honour of the fifteenth-century national poet of Uzbekistan. My research project in Uzbekistan was to consult the manuscript (from 1927/28) of an Uzbek oral epic which I had translated (entitled Rawshan) and to clarify textual problems. What I was hoping for was to meet, listen to, and possibly record an Uzbek epic singer or bakhshi. This turned out to be difficult. I was told that there were no epic singers in Tashkent and that in order to travel to the nearest place where an epic singer was to be found I needed special permission, issued by the Ministry of Interior Affairs. As this permission was not obtainable – but my disappointment great – my hosts decided to ignore Soviet protocol. They organized a minibus, sat me in the middle of a group of five or six people and off we drove to a kolkhoz in the district of Boka, about 70 km south of Tashkent, close to the Tajik border. We arrived at around midday; the kolkhoz director was persuaded to let the singer off work; our party was welcomed and waited upon in a pavilion-like open-sided wooden garden house. And then came the singer. He was a man in his fifties, tall and slender. He had a tanned, leathery face, clearly much exposed to the elements, and wore a moustache. He had a young voice, with a warm timbre. He was wearing the usual Uzbek hat (doppi), a square beret, covered in black silk with white embroidery, in a pattern typical of the locality. His name was Chāri Khojamberdi-oghli, but he was generally called Chāri-shāir, Chāri the singer-poet. Chāri-shāir was born in the Qashqadaryā province in 1925. Our party was joined by the local teacher, who had written down some of Chāri-shāir’s repertoire. Chāri-shāir was known to the Tashkent scholars, but had not been recorded by them. He was clearly excited to be asked to perform, and he talked about his art with enthusiasm. Chāri-shāir had, of course, come with his dombira, a two-stringed plucked instrument. He tuned his instrument, began to strum the strings, modulated a melody, played a sequence of chords, fell into a fast and pronounced rhythm, and then began to sing. Chāri-shāir first sang about his dombira, then intoned another song, a song about the Communist Party. I knew that epic singers would introduce their performance with short, topical, and often improvised songs, called terma. Songs about the dombira have been written down from a number of Uzbek bakhshis,

Introduction  3 and songs about the Communist Party naturally figured in collections of modern Uzbek folklore. But then Chāri-shāir began a new song: Arzim eshit, khalāyiqlar! Uzaqtan keleyātibman, Ghamli kunlarga bātibman: Bir gozaldi yoqātibman, Bir gozaldi yoqātibman! Hear my request, o people! I have come from afar, I have fallen into despair: I have lost a beautiful girl, I have lost a beautiful girl! This was the beginning of an episode in Rawshan, the Uzbek epic I was translating. I had originally been attracted to this work in Derzhavin’s Russian translation by the beautiful verses, the mixture of romance and heroic deeds, perhaps also by the Oriental grace of the illustrations. Rawshan existed as a poetic tale in my mind, keeping company with other poetic tales, such as Aucassin and Nicolette, Sir Gawain and the Green Knight, The Faerie Queene, Eugene Onegin … But now something new was born. Here, in the afternoon heat of early summer in Uzbekistan, with the perfume of the orchard, the smells of the food on the table, the singing of birds and the sound of the voices of humans and beasts in the background, half dazed with the dream-like feeling of being in unfamiliar surroundings, in a different cultural world, here poetry became song, the narrating voice of the tale became the voice of the singer, the poetic tale in the mind acquired a new existence: as a tale heard and as poetry sung, absorbed with one’s eyes and ears, in the bodily presence of the teller and singer of tales and his listeners. To me as a medievalist, this experience seemed to be a journey undertaken in a time-machine, transporting me back to the days when the Chanson de Roland or some Middle English popular romance was still performed by the jongleur or the minstrel to an enraptured audience. Here I was, in the last quarter of the twentieth century, witnessing the traditional art of orally performed epic poetry by a skilled singer of tales, an art that must have been practiced for centuries. True, Chāri-shāir’s words and melodies were no continuation of some European medieval oral tradition, and I was listening to an excerpt from an Uzbek dāstān and not to an episode from a chanson de geste or a Middle English romance. But I was listening to the oral performance of a story and of poetry that I had first encountered as a written text and which had now been transformed into something new and exciting. What I had known theoretically was now experienced: that performance is the primary mode of existence of

4 Introduction oral poetry, the written text is, as it were, an afterthought. So far I had experienced the pressed flowers of oral poetry, now I was in the presence of the living flower, fully exposed to it with all the senses of the beholder. For someone coming from a textual culture where epics exist in written form only, two basic questions arise: How is an oral epic performed, and how does performance impinge on interpretation? To ask how an oral epic is performed is to ask whether recitation is in singing or speaking, with or without instrumental accompaniment, by one or more performers, with or without additional implements, with or without the incorporation of dancing or miming. Many more questions of this kind can be asked about the manner, time, occasion, locale, and participants of an epic performance. These questions can in theory be answered by descriptive and comparative analysis. The more detailed the description and the wider the network of compared cases, the more revealing our answers will be. Description and comparison will indeed be the main concern of much in this book. The question about how performance impinges on interpretation is more difficult to answer. If we have an oral epic in textual form – like the edition of Rawshan, from which I made my translation – we know that practically all performance features have disappeared. I am saying ‘practically’, because some texts preserve ‘performance markers’, such as narrator’s comments, deictic particles and other elements much studied in medieval and ancient Greek epic texts as indications of their presumed oral background. For Rawshan it is possible to get somewhat closer to performance than the edited text allows by consulting the manuscript which preserves the original transcript of the singer’s performance.1 Even then, however, everything that is not textualized is missing. How does this affect our understanding and appreciation of the epic? Are we only missing ‘ornamental’ elements or are we missing something significant? It is easier to formulate what is at stake by using an example from art. Let us take Michelangelo’s well-known fresco of the creation of Adam in the Sistine Chapel. Adam is on one side, stretching out his arm toward God, God on the other side, almost touching Adam’s forefinger with his. If we have a black-and-white reproduction of this scene, its composition and the powerful gesture of the two outstretched arms and almost touching forefingers will be preserved. With a black-and-white reproduction we can interpret many aspects of the scene, but will have to leave aside an important aspect of pictorial art, colour. Is this how text and performance relate to each other, when one element is missing, perhaps important, perhaps only marginal? But this is not all. For the Michelangelo fresco, a full understanding presupposes a cognizance of the whole pictorial programme of the Sistine Chapel. A reproduction would have to be embedded in reproductions of the immediate and total context of the scene. So not only one element, colour, is felt to be missing, but the whole context is absent. The context extends also to

Introduction  5 the architectural structure for the pictorial programme. In fact, a full contextualization calls for either the presence of the beholder in loco or at least a video reproduction. Does the text of an oral epic have to be similarly contextualized in order to be fully understandable? In other words, is performance as an event with all its various elements essential for interpretation? With the frescoes in the Sistine Chapel we can even go one step further, if we remember that it is a chapel, a place of worship, which, of course, is furthermore the locale of the conclave, the election of a new pope. Should we keep this in mind as well? C. S. Lewis begins his study of Milton’s Paradise Lost with an amusing argument in favour of considering the use and function of a work of art before interpreting it: The first qualification for judging any piece of workmanship from a corkscrew to a cathedral is to know what it is – what it was intended to do and how it is meant to be used. After that has been discovered the temperance reformer may decide that the corkscrew was made for a bad purpose, and the communist may think the same about the cathedral. But such questions come later. The first thing is to understand the object before you: as long as you think the corkscrew was meant for opening tins or the cathedral for entertaining tourists you can say nothing to the purpose about them. (1942: 1) Is Rawshan only fully understandable if, in addition to the experience of performance in vivo, the cultural framework in which the event takes place is considered and analysed? These are the main questions this book tries to answer. As to the question of how an oral epic is performed, the description and analysis of performance has to be limited to specific cases if the material is to stay manageable. Of course, the more cases are added, the more detailed the picture that emerges will be. By the same token, however, road signs indicating major directions and diversions will lose their visibility when submerged in a forest of signs. The book is organized into three parts, Settings (Chapters 1 to 3), Performance (Chapters 4 to 7), and Interpretation (Chapters 8 to 10). In Chapter 1, I will explain what is to be understood by an oral epic in this book and in Chapter 2, I will introduce the ‘singer of tales’ in various oral traditions, that is, the different types of singer-narrators and their art, characterized by both traditionality and creativity. Chapter 3 provides a sketch of analytic approaches to performance. In the second part, I will focus on embodiment in Chapters 4 and 5, first on the singer’s voice in Chapter 4, then on gesture and gesticulation in Chapter 5. The musical form of epic performance – solo or with the accompaniment of a plethora of musical instruments – will be the subject of Chapters 6 and 7. In the third part, Interpretation, the

6 Introduction problem of the expression of meaning and mood in music will be tackled in Chapter 8. Narratological aspects will be in the foreground in Chapter 9, and in Chapter 10 the topic of discussion will be the dynamic connexion between epic text (or textualized epic) and performance, in particular with reference to visualization and contextualization. As emerges from the personal note at the beginning of this introductory chapter, my primary material is centred on the oral epic traditions of the Turkic peoples, or rather of peoples speaking a Turkic language. Turkish, the national language of Turkey, is only one of many Turkic languages, a family of languages that stretches from the Balkans – via Turkey, Azerbaijan, Turkmenistan, Uzbekistan, Kazakhstan, Kyrgyzstan – to Xinjiang in north-western China, with sizable groups of speakers of Turkic languages in Iran and Afghanistan. A number of Turkic languagespeakers are found in various parts of the Russian Federation: in the area of the Ural Mountains (Tatars, Bashkirs), in the Altay-Sayan region (Altaians, Khakas), and in north-eastern Siberia (Yakuts). All of these ethnic groups had an oral epic tradition, and some of them either still have one or had one until recently. While my primary material comes from Turkic oral traditions, this is not a book about Turkic oral epic poetry. The questions I am asking and the answers I am proposing are more fundamental and more general. For this reason I have included a number of oral epics and traditions that enlarge and diversify the picture presented by Turkic oral traditions. My choice has been guided by familiarity on my part with the languages and cultures in question, by the occasional possibility of witnessing and recording a performance, and by the availability and reliability of sources. In order to provide some background information on these traditions I have compiled an appendix with basic data and bibliographical references (Appendix A). More information on the audio and video examples discussed in detail will be found in Appendix B, and on recordings of the music of epic in Appendix C. There is also a glossary of musical terms and of non-Western musical instruments. While music is considered an important performance element in this book, I should make it clear from the outset that the book is neither written by a musicologist nor destined for a specialized musicological readership. The reader I have in mind is someone who is interested in oral epics and who is wondering, as I am, what all the non-textual, or non-textualized, components of an oral epic performance are and what, if anything, they contribute to the understanding of the epic. My musical notations are meant to give a visual impression rather than an exact notation of sound. It is not necessary for the reader to read music; my comments on the music of epic and the occasional notation are intended to be accessible to everybody. For the transcription (or transliteration) of non-Latin scripts, see the Note on Transcription and Transliteration (pp. ix–x). Abbreviations of

Introduction  7 book series and journals are listed at the beginning of the bibliography. All translations of primary sources are mine unless indicated otherwise; my translations of quotations are marked by ‘trans.’ after the page reference. After these practical considerations I would like to return to the central issue, well illustrated by the following quotation from the edition and translation of a Bambara (West African) version of the epic Silamaka Fara Dikko by Ousmane Bâ: A great deficiency of all textualization is that musical elements, facial expressions and gesticulation of the djali [singer], which often have an explanatory and disambiguating effect, cannot be rendered adequately without severely limiting the readability of the text. The same is true of the reactions of the audience, such as snapping one’s fingers, singing etc., which we were unable to record in the text of this study. To tell the truth, the text transcribed is merely the skeleton of an epic, which takes shape only in actual performance. A textualized epic such as this one can therefore only serve as makeshift to give no more than a superficial impression of the epic, an impression which, on account of the sole medium of writing, is rather one-dimensional. (1988: 6 trans.) Ousmane Bâ is, of course, right. His remarks do, however, whet the reader’s appetite for knowing more about music, facial expressions, gesticulations and snapping one’s fingers. It is hoped that the chapters that follow will ‘flesh out’ the ‘skeleton’ and compensate for the makeshift character of editions like Bâ’s.

Note 1 On the singer (Ergash Jumanbulbul-oghli, 1868–1937) and the manuscript, see Reichl 1985a: 4–12.

Part I


1 How to Identify an Oral Epic

In the disputable and usually futile task of classifying the forms of poetry there is no great quarrel about the epic. (C. M. Bowra, From Virgil to Milton) In Plato’s dialogue Laches, Socrates asks the question: ‘What is courage?’ Laches, a Greek general during the Peloponnesian war, is quick to give his answer: ‘Why really, Socrates, it’s not hard to say: if someone is willing to stay in ranks and ward off the enemy and not flee, rest assured he is courageous.’ Socrates is not pleased with this reply. Instead of defining courage, Laches and also the other participants in the dialogue give examples of what they consider courageous behaviour. Socrates was not asking for examples, but for a definition. His goal was to work out the idea (eidos) of courage. The ‘idea of the epic’ has been elaborated since the beginning of Western literary criticism in Aristotle’s Poetics. With the Homeric poems as models, Aristotle stated that the epic is representational (mimētikē) and narrative (diēgēmatikē), in metre (en metrō) and of a certain length (mēkos); like tragedy it is a representation of heroic action (mimēsis spoudaiōn). With variations and specifications Aristotle’s definition has survived until today. Bowra could confidently say that ‘there is no great quarrel about the epic’ and continue by defining the epic along Aristotelian lines as follows: ‘An epic poem is by common consent a narrative of some length and deals with events which have a certain grandeur and importance and come from a life of action, especially of violent action such as war.’ Other scholars, as for instance J. B. Hainsworth in his book The Idea of Epic, base their studies on fundamentally the same principles.1 This is not to say that Aristotle has held sway in genre theory unchallenged. The epic has received attention in Western thinking from Hegel to György Lukács and Mikhail Bakhtin, and also structuralism has left its imprint on modern theorizing, with studies on genre by Tzvetan Todorov, Gérard Genette, and Jacques Derrida, to mention only some.2 Dissenting voices can also be heard from scholars in the fields of folklore and cultural anthropology, who argue against an understanding of genre entirely based on Western literature.

12 Settings

Oral: Shades and Grades Before I address some of the problems raised by looking beyond the confines of Western literature, I will briefly comment on the qualification ‘oral’. Oral poetry is oral first of all in the etymological sense of the word ‘oral’: it is ‘mouthed’ poetry, poetry spoken, recited, or sung. Its primary reception is aural, that is, that of poetry which is listened to rather than read. Not all orally performed and aurally received poetry, however, is oral in the sense we understand the word when we speak of oral poetry. A recital of Gounod’s ‘Ave Maria’ is oral, but the song is not an oral poem. Oral poetry is not only orally performed, but also orally transmitted. This means that a song or a narrative is learned by word of mouth in a chain of oral transmission, not just from one person to another. Typical examples of oral transmission are the ‘singers’ schools’ discussed in Chapter 2. Oral transmission does not exclude the existence of written texts, which might be diffused parallel to oral transmission or might even be the ultimate source of an orally performed and transmitted poem, but the emphasis is on oral in the transmission process. Typically folksongs in non-urbanized, traditionally oriented societies are transmitted orally, rather than memorized from a printed text. This is true of the ballads collected by Cecil Sharp in the Appalachians at the beginning of the twentieth century. It is axiomatic of the songs collected from societies where literacy is rare or even non-existent, as in the songs discussed in Maurice Bowra’s Primitive Song or the songs collected by Raymond Firth from the Tikopia on the Solomon Islands in the Pacific Ocean.3 Songs of this kind were not only transmitted orally, but also composed orally, without the help of writing. Oral composition can mean the spontaneous improvisation of songs, such as the contest songs popular among the Kazakhs and other Turkic peoples, but the term is not co-extensive with that of improvisation. We know that oral poets have often spent time and energy on composing, polishing, and elaborating their works of oral art.4 Oral composition is not synonymous with what Albert Lord calls ‘composition in performance’, by which he means the ability of singernarrators to perform an oral epic smoothly and fluently, not on account of a great power of memorization, but on the basis of their ability to weave a poem together from various building blocks, in particular formulaic lines and expressions and so-called ‘themes’, that is, motifs and typical passages or scenes. His definition of the oral epic, at least in the case of the South Slavic epic tradition, incorporates the concept of composition in performance: Stated briefly, oral epic song is narrative poetry composed in a manner evolved over many generations by singers of tales who did not know how to write; it consists of the building of metrical lines and half lines by means of formulas and formulaic expressions and of the

How to Identify an Oral Epic  13 building of songs by the use of themes. This is the technical sense in which I shall use the word ‘oral’ and ‘oral epic’ in this book.5 The concept of composition in performance underestimates the importance of verbal recall in the training situation on the one hand and on the other has a tendency to minimize the creative power of singers, reducing their art to a mechanical combination of pre-fabricated elements.6 Furthermore, in an article entitled ‘How Oral is Oral Literature?’, Ruth Finnegan argues that the puristic view of orality as expounded by Lord in his Singer of Tales is not tenable. It does not conform to reality. The illiterate singer of tales who has acquired his epic songs from the oral performances of other singers and has no written version of his song is a marginal rather than a typical case. In reality, oral and written traditions keep crossing paths, and it is therefore perhaps preferable to talk of an oral-literate continuum. She notes that ‘the relation between oral and written forms need not just be one of parallel and independent coexistence, far less of mutual exclusion, but can easily exhibit constant and positive interaction’.7 This is certainly true. As to oral transmission, even in traditional societies, for instance in Central Asia a century ago, literacy was not totally uncommon among epic singers, and today it is, of course, the norm. Some of the works performed by these singers were available in manuscript or print and even used in performance in some local traditions. Moreover, both in Central Asia and in other parts of the world we find the public ‘reader of tales’ as an alternative or supplement to the singer of tales (see Chapter 2). The impingement of written texts on oral transmission is particularly noticeable in the epic subgenre of romance. In some cases writing is, for instance, also used in the learning process. The Iranian storytellers (naqqāl) who perform episodes from the Persian national epic Shāh-nāma (‘The Book of Kings’ by Firdausī, c. 1000 ce) must as students with a master singer not only memorize passages from the epic, ‘but the student must also copy and learn the tumār, which he receives from his teacher. This tumār is a story outline in prose of the episodes making up the stories he will tell.’8 Nigel Phillips, in his study of the oral epic tradition of the Minangkabau of West Sumatra, mentions that in the predominantly oral tuition of singers, some used written sources when learning the epic. (1981: 13–14) There can be no doubt about the interaction of oral and written transmission in many of the oral epic traditions that have stayed alive to this day or came to an end in the course of the twentieth century. It is, however, equally undeniable that the ideal case – oral performance in conjunction with oral transmission and oral composition – did and does exist and that it makes sense, when talking of oral tradition to take this case as the yardstick by which the variations and combinations of orality and literacy in specific situations are measured. When looking at different

14 Settings aspects of oral epic poetry in the following chapters, the precise meaning of ‘oral’ in a given case will have to be discussed. At this point, it suffices to see the type of singer Milman Parry sought to record as the ideal case; it is positioned at the oral end of the oral-literate spectrum.

The Challenge of Native Classification The definition of an epic becomes more complicated when we step outside the Western literary tradition and look at works that have been called epics in the oral traditions of Asia and Africa. René Wellek and Austin Warren have underlined the culture-specific character of genre categories: But we must not narrow ‘genology’ to a single tradition or doctrine. ‘Classicism’ was intolerant of, indeed unwitting of, other aesthetic systems, kinds, forms. Instead of recognizing the Gothic cathedral as a ‘form’, one more complex than the Greek temple, it found in it nothing but formlessness. So with genres. Every ‘culture’ has its genres: the Chinese, the Arabian, the Irish; there are primitive oral ‘kinds’. Medieval literature abounded in kinds. (1963: 234) But does this mean that genre categories do not apply across cultural boundaries? A well-known case is the controversy about the epic among Africanists. In her survey of oral literature in Africa, Ruth Finnegan covers a great number of genres and traditions, but not the epic. She only has a note on epic of two and a half pages, where she maintains that works that have been called ‘epic’ by scholars – as for instance the ‘Lianja epic’ of the Congo – are not epics, partly because they are in prose rather than in verse, partly, and more importantly, because prior to having been written down they were ‘a very loosely related bundle of separate episodes, told on separate occasions and not necessarily thought of as one single work of art’. She ends by affirming that ‘epic seems to be of remarkably little significance in African oral literature, and the a priori assumption that epic is the natural form for many non-literate peoples turns out here to have little support’.9 Her contention has not found universal consent. A number of books on particular epics and epic traditions in Africa, most importantly in West Africa, but also in the area of the Congo, and on epic in Africa in general have been published since. As one Africanist puts it in the title of one of his publications, alluding to Ruth Finnegan’s denial of the existence of epic in Africa: ‘Yes, Virginia, There is an Epic in Africa’.10 The controversy hinges on an understanding of the genre term ‘epic’ in the Aristotelian tradition. Anthropologists like Bronislaw Malinowski and folklorists like Dan Ben-Amos, however, warn us that for the analysis of non-Western traditions, especially those where narrative and poetry are predominantly, if not exclusively, oral, native terms and ethnic categories

How to Identify an Oral Epic  15 have to be employed. They show that criteria quite different from those enumerated above can be used in classifying verbal art. A well-known case is Malinowski’s report on three types of narratives among the natives of the Trobriand Islands in the Solomon Sea: kukwanebu, stories which are believed (but not very seriously, as Malinowski adds) to have a beneficial effect on the newly planted crops, libwogwo, stories thought to be true, and liliu, sacred tales or myths. What distinguishes these types of narratives is not primarily their textual side, that is, the kind of story they tell, but rather their function and place in the life of the society in question. Malinowski makes this point repeatedly and insists that we can neither fully grasp the meaning of the text, nor the sociological nature of the story, nor the natives’ attitude towards it and interest in it, if we study the narrative on paper. These tales live in the memory of man, in the way in which they are told, and even more in the complex interest which keeps them alive. (1926: 27) Similar examples can be found in many regions of the world. According to Paul Radin, the Native American Ho-Chunk (Winnebago) distinguished between two types of narratives: Like most American Indian tribes, the Winnebago divided their prose narratives into two types: those that dealt with a past that was irretrievably gone and which belonged to the realm of things no longer possible or attainable by man or spirits; and those which dealt with the present workaday world. The first is called waikan, what-issacred, and the second worak, what-is-recounted. No waikan could be told in the summertime or, at least, when the snakes were above ground. Waikan could not end tragically. … The worak, in contrast, could be told at any time and had to end tragically. (1972: 118) In a study of sung tales from the Papua New Guinea highlands, the editors state as main distinctive features of these tales, called by different names according to language and ethnic group (pikono, bì té and others), that they are vocal compositions, sung solo, without an accompanying instrument, generally improvised by a specialized singer-narrator, performed for the entertainment and sometimes instruction of the audience, which is expected to respond verbally to the tale; the tale is composed in heightened language; the story is ‘not secret and is not considered “true”, but the characters are placed in a real landscape of known places’. It is interesting to note that in these traditions performance aspects (solo singing, audience participation) play as important a role as narrative structure and poetic form.11

16 Settings While it is essential to base a discussion of oral epics ‘in the field’ on native classification, any kind of comparative analysis presupposes a recognition of class identity between the items to be compared. And in order to formulate such class identity one needs a ‘meta-terminology’, which applies to more than one native tradition. That these terms might have different weight in different traditions and might have modifications that increase dissimilarity rather than similarity does not disqualify such an undertaking. The epic is a historical genre and, as far as we know, not a universal genre; similarly, the elements of such a meta-terminology need not be universal. They arise from specific speech forms and have to be tested as to their use in comparative analysis. I will discuss the question of how to identify an oral epic in largely divergent traditions by two examples (choosing Laches’ rather than Socrates’ method), one from Africa, the other from Central Asia.

An African Interlude One of the scholars who have been active in collecting and translating African epics, Christiane Seydou, has specifically addressed the question of how to define the epic.12 She takes the Occidental concept of ‘epic’ as her point of departure and then, on the basis of a comparison between two African oral traditions, argues for a set of nine parameters by which the genre is to be defined: (1) the sociological status of the artist, (2) the musical accompaniment, (3) the formal structure of the narrative, (4) its contents, (5) the mainspring of the action, (6) the style, (7) the effects of the style on the audience, (8) the behaviour of the audience, and (9) the function of the epic and of the epic performance. The oral epic traditions selected are those of the Fulbe (Fulɓe, Peul, Fulani) in Mali and of the Fang in Gabon and Equatorial Guinea. The Fulbe do not only live in Mali, but also in other parts of West Africa, which explains why Fulbe epics have also been collected outside Mali. Their genre term for the oral epic is hoddu; this is also the word used for the instrument of epic performance, a plucked lute of three or four strings. The epic tradition of the Fulbe is very similar to that of the Bambara and other speakers of Mande languages in West Africa. The best known Mande epic is no doubt Sunjata, a legendary epic about the thirteenthcentury founder of the empire of Mali, Sunjata Keita; it has been written down in a number of versions and been repeatedly translated and retold.13 Christiane Seydou has highlighted the interpenetration of the oral traditions of the Fulbe and of the speakers of Mande languages, made all the easier through the bilinguality of epic singers: In this area [the Massina, a vast plain between Ségou and Timbuktu in Mali] it is incidentally no rarity that one and the same griot [epic singer], who is bilingual, performs Fulani or Bambara epics without

How to Identify an Oral Epic  17 any difference or, even better, recites the same narratives, using one or the other language according to his momentary audience. (1976: 35 trans.) The Fang epic, with which Seydou compares the Fulani epic, is designated by the native term mvet. This word, like hoddu, denotes also a musical instrument. The musical instrument mvet is a type of harp with three or more strings and three or more calabashes as resonators. The players of this instrument belong to different degrees of competence, depending on their training and initiation. The performance of the epic is reserved for the masters. Pierre Alexandre mentions that the longest recorded mvet lasts for twelve hours.14 The comparison between these two oral traditions reveals a number of dissimilarities. Under the heading ‘style’, for instance, Seydou describes the Peul narratives as showing the ‘simple and sober style of ordinary narration which contrasts with the expressive poetic style of certain devices and fixed formulas’, but the Fang mvet as possessing a ‘style marked by excessiveness: accumulation and effusion of verbal resources, and unbridled fantasy’ (1983: 57). Under the heading ‘behaviour of the audience’ the contrast is between ‘Silence, attentive and contemplative immobility of audience concentrated on listening. They manifest tense but silent excitement. Intense but passive participation’ on the Fulani side, and on the Fang side, ‘Exuberant and extroverted audience which is constantly called upon by the reciter and which thus expresses itself frequently. Active and noisy participation’ (58). There are many more differences: the griot belongs to the caste of professional musicians, the mvet player has no caste affiliation, but he has to undergo initiation, which the griot does not. The narrative structure of the Fulani epic is episodic, without digressions or active audience participation; the mvet is characterized by interspersed digressions and dialogues between reciter and chorus/audience in the interludes. Also in content, choice of protagonists, placement in space and time, the two traditions show significant differences. They do, however, agree on three important points: (1) the singer-narrator uses a specific instrument, which defines his performance; (2) Seydou stresses that despite the differences that exist between the narrative structure and the contents of the epics, action in the narrative is in either case initiated by various forms of transgression; (3) under the heading of ‘function’ Seydou lists various features, one of them being, for the Peul epic: ‘Solidarity regenerated by the renewed feeling of cultural and ideological identity proposed in the text’, and, for the mvet: ‘Reaffirmation of identity and social unity through the solidarity experience in the celebration’ (59). Seydou’s observations tie in with the results of other anthropologists, ethnologists, and folklorists who base their taxonomies on culture-specific categories.

18 Settings

The Uzbek Dastan In my introductory chapter, when describing my first encounter with an Uzbek epic singer, I used the terms ‘oral epic’ and dāstān as synonyms. The Uzbek word dāstān is a loan-word from Persian. In Persian it has the basic meaning ‘story’ or ‘tale’. This word is also found in other Turkic and Iranian languages, pronounced and hence written slightly differently in the various languages. In some, as in Uzbek, it denotes all types of oral narratives performed by professional singer-narrators; in others it has a more restricted meaning. Differences in usage will be indicated where appropriate. For simplicity’s sake I will use the spelling dastan, irrespective of the correct linguistic form of the tradition in question. In Uzbek, dastan denotes two types of narratives, long epic poems of written literature, such as the narrative works of cAlī Shīr Navā’ī (d. 1501), and narrative works of oral literature, such as Rawshan. These long narrative poems of classical literature (in Chaghatay, the literary Turkic language of Central Asia up to the end of the nineteenth century) continue an older tradition in Persian, with the twelfth-century poet Niẓāmī (d. 1209) as its most influential representative. His khamsa (‘group of five [poems]’), which contains two justly famous love stories, Khusrow and Shīrīn and Laylī and Majnūn, was much imitated, not only by cAlī Shīr Navā’ī. The dastan of oral tradition is a narrative with a number of formal characteristics, typically performed by a professional singer-narrator, the dāstānchi or bakhshi (see Chapter 2). The most striking feature of Uzbek dastans is that they are, without exception, composed in a mixture of verse and prose. In this they differ significantly from the concept of epic as a narrative in verse. This alternation between verse and prose – known as prosimetrum – should not be understood as consisting of a prose text with some verse interspersed, in the manner of some of the tales in the Grimms’ collection of German fairy tales. Despite lengthy prose passages in some dastans, verse has a prominent position. This is expressed by the fact that the verse passages are sung, while the prose passages are spoken. Narratives in a mixture of verse and prose are widely spread in world literature, and in the following chapters we will encounter a number of prosimetric narratives in different languages and from different parts of the world. These are generally referred to as epics in their respective traditions, such as, for instance, the Gesar epic of Tibet or the Ḍholā epic of northern India. Clearly, the requirement that an epic has to be in verse has to be dropped.15 It has also to be dropped on account of the tenuous nature of the distinction between verse and prose in a number of oral traditions. Cases in point are the African epics on which Christiane Seydou bases her arguments for the definition of epic. Neither for the mvet nor for the West African oral epics can verse be defined by language-internal characteristics. Nevertheless, the term ‘verse’ is not entirely meaningless in these

How to Identify an Oral Epic  19 traditions as will be discussed in a later chapter (Chapter 6). ‘Prose’ is an equally variable concept. It can be highly patterned in oral traditions and is in many traditions closer to verse than the rigid dichotomy between prose and verse suggests.16 This is also true of the prose passages in the Uzbek dastans, which are often in rhymed prose (called saj, from Arabic), a form that frequently occurs in the tales of The Thousand and One Nights. Uzbek dastans are, as a rule, extended narratives, but there are also shorter versions of dastans. Fāzil Yoldāsh-oghli’s version of the epic Alpāmish contains over 13,000 verse-lines (and sizeable prose portions); by contrast there is a version of this dastan, written down from the bakhshi Saidmurād Panāh-oghli, which comprises a mere 1,655 verse-lines. However, even the performance of such a short dastan lasts for several hours, since the verse passages are sung and the instrumental accompaniment normally ‘embroiders’ sung portions. At an earlier time, when Uzbek oral tradition was still flourishing, the performance of dastans lasted through the night, sometimes through several nights in a row. More important than length as a quantitative measure is the narrative pace and conception. Even short dastans have a certain epic breadth, with monologues and dialogues, descriptive passages and typical scenes or themes. A further deviation from the Aristotelian concept of epic is the fact that only a portion of Uzbek dastans are ‘representations of heroic action’ (mimēsis spoudaiōn). The prime example of a heroic epic in Uzbek is the dastan Alpāmish. Very summarily, the plot of this dastan has two parts, a bride-winning tale and a return story. The latter has a number of close parallels to the home-coming of Odysseus. Maurice Bowra makes frequent mention of this dastan in his study of heroic poetry, rightly so, as the prevailing spirit of the dastan is heroic. The dastans performed by Uzbek bakhshis embrace, however, all kinds of narratives. Leaving aside narratological ramifications, the clearest division of Uzbek dastans into subgenres on account of their plot – and related to this, their narrative development and style – is into heroic dastans and lyrical or love dastans. The lyrical or love dastans are, as a rule, narratives about two lovers who have been cruelly separated and who have to undergo a number of trials and adventures before they are either happily reunited or find a tragic end. Many of the verse passages express the feelings of the protagonists in monologues and dialogues, hence the characterization as lyrical dastans. Some of the plots come from classical Uzbek literature. From the point of view of Western literature the term ‘romance’ comes to mind for this type of dastan. For medieval literature, W. Ker has made a very good case for distinguishing between epic and romance, and it might be useful to keep the term ‘romance’ for this type of dastan. Rawshan, mentioned in the introductory chapter, has incidentally a mixture of heroic and romantic elements. It belongs to a cycle of dastans about an outlaw and

20 Settings military leader with his band of warriors, in which martial and adventure episodes alternate with romantic and sometimes also comic plots. So one can see that the division of the Uzbek dastans into ‘epic’ and ‘romance’ is in many cases problematic.17 What has been said about the Uzbek dastan can be said about other genre terms for ‘epic’ and ‘romance’ in the Turkic oral epic traditions. I will only mention one, the Turkish term hikâye (from Arabic hikāya, ‘tale’), a tale in a mixture of verse and prose performed by the Turkish minstrel. Many of these hikâyes are romances in the sense of ‘lyrical dastans’.18 ***** In summary, the discussion of native genre terms such as hoddu, mvet or dastan shows that among widely distant and different oral traditions similarities exist that allow a comparative approach. All these genre terms denote narratives that are performed by singer-narrators, generally professionals, who recite, chant or sing them in what one might call ‘heightened speech’. In other words, their clearest shared trait is their performance. As will be argued in this book, it is indeed performance – written large – that is at the core of the understanding of these native genres with the same ‘family resemblance’. Of secondary importance is content. Depending on the tradition, it seems appropriate to distinguish between ‘heroic epic’ and ‘romance’. As the Uzbek example has shown, it would distort the native perception of genre to exclude the latter from our study. Also in other traditions such as the Arabic folk epics (siyar) ‘the distinction between “heroic epic” and “romance” can be drawn only with great difficulty’19 (Reynolds 2002: 340). M. Bowra has excluded shamanistic epics from his study of heroic poetry; he has given good arguments for this, but as we will see, shamanism and epics with shamanistic traits are widely disseminated in oral epic traditions and cannot be ignored in a discussion of epic performance. Most of the oral epics discussed are extended narratives, have epic breadth. Occasionally, however, a reference to shorter sung narratives is in place. This concerns primarily the South Slavic junačke pjesme (‘heroic songs’) and the Albanian kënge kreshnikësh (‘songs of the warriors’), which, as their native names indicate, are heroic songs rather than heroic epics. Some of these South Slavic heroic songs do actually have epic proportions, such as The Wedding of Smailagić Aga, performed by Avdo Međedović, the longest junačka pjesma in the Milman Parry collection. Other shorter forms referred to are the Mande hunters’ epics, the Russian bylinas and the Finnish rune-songs (runo). It seems to me that from a comparative perspective these heroic songs have to be included in a study of oral epics. It might therefore be more appropriate to speak of ‘oral epic poetry’ instead of ‘oral epic’, meaning by ‘poetry’ not only verse, as in English, but poiēsis in its wider sense of ‘verbal art’. This would include dastans in a mixture of verse and prose as well as heroic songs. As this

How to Identify an Oral Epic  21 is not the common use of ‘poetry’ in English, I will continue to speak of epics – and, depending on context, of romances and heroic or simply narrative songs. Purists might object that this catholic understanding of the oral epic is not sufficiently stringent. In reply I would like to refer to the title of this chapter, ‘How to identify an oral epic’. The previous discussion is meant as no more than a rough guide to the kind of work of oral verbal art designated as ‘oral epic’ in this book. The ‘view from within’, the native view of narrative genres, plays the leading role for identification. For such a position not only folklorists and cultural anthropologists have argued, but also literary scholars interested in comparative approaches. An issue of the 2007 volume of PMLA was devoted to genre theory, with an introductory essay by the Americanist Wei Chee Dimock, in which she argues for a concept of genres as fields of knowledge and pleads for a global view of genre in the spirit of world literature studies. Among genres she singles out the epic and decidedly advocates an ‘open’ understanding of this genre: ‘There is no reason to make a closed set out of the Iliad, the Odyssey, the Aeneid, and the Divine Comedy, when there are vibrant epic traditions in the Near East, South Asia, Africa, eastern Europe, and the Americas.’20 She could have added Central Asia and Siberia with their Turkic and other oral epic traditions – traditions still too little known to Western readers.

Notes 1 See Hainsworth 1991: 1–10; Bowra 1945: 1; Laches quotation from Allen 1996: 73. 2 For a useful anthology, see Duff 2000, which includes also an English translation of Bakhtin’s influential work on speech genres as well as translations of some of Todorov’s, Genette’s, and Derrida’s works on genre; for a linguistically orientated study, see Fowler 1982. 3 See Sharp 1932; Bowra 1962; Firth with McLean 1990. 4 On Turkish contest songs, see Erdener 1995. See Chapter 2 on the Aceh singer Dôkarim, who spent five years composing and elaborating his epic about the wars of the Aceh with the Dutch. 5 Lord 1960: 4. Immediately following the passage quoted, Lord gives a definition of ‘formula’, ‘formulaic expression’ and ‘theme’: ‘By formula I mean “a group of words which is regularly employed under the same metrical conditions to express a given essential idea.” This definition is Parry’s. By formulaic expression I denote a line or half line constructed on the pattern of the formulas. By theme I refer to the repeated incidents and descriptive passages in the songs’ (4). 6 See Chapter 2 on the Uzbek singer Fāzil Yoldāsh-oghli. 7 Finnegan 1974: 57. On the oral-literate continuum, see also Reichl 2015b. 8 Page 1979: 198; tumār means ‘book, roll’; on orality and literacy in Persian epic poetry, see Rubanovich 2012. 9 Finnegan 1970: 108–10; quotations from pp. 109 and 110. 10 Johnson 1980. The title is a pun on the phrase ‘Yes, Virgina, there is a Santa Claus’ in a well-known editorial of the New York Sun of 1897 (entitled ‘Is There a Santa Claus?’).

22 Settings 11 Rumsey & Niles 2011: 1–38, at 6; see also Chapter 6. For more examples of ethnic classification, see Ben-Amos 1976: 227–37. See also Propp 1976, who argues that a folklore genre is defined by four features, its poetic form, its place in life, the form of its performance, and its relation to music. 12 Seydou 1982; for an English translation, see Seydou 1983. 13 Depending on the original dialect and the transcription system employed, the name is spelled differently: Sunjata, Sundiata, Soundiata, Son-Jara etc. 14 On these degrees or ranks, see Alexandre 1974: 1–3; for an edition and translation of the longest recorded mvet, see Pepper 1972. 15 On the mixture of verse and prose in narrative in a wide range of literatures, see Harris & Reichl 1997; on prosimetrum, see ibid. 1–16. 16 This has been demonstrated with regard to Native American tales by Hymes 1981; compare also Tedlock’s remarks on Mayan verse (1983: 216–30). 17 For the distinction between epic and romance in medieval literature, see Ker 1908; on the romance genre in the Turkic oral traditions, see also Chapter 10. On the various subgenres of the Uzbek dastan, see Mirzaev & Sarimsāqov 1981 and Mirzaev 2008: 204–65. 18 On the Turkish hikâye, see the study by Başgöz 2008. 19 Reynolds 2002: 340; see further Chapter 2, pp. 27, 30–31. 20 Dimock 2007: 1383. Compare also Martin 2009, and for performanceoriented approaches, Hanks 1987 and Seitel 2003.

2 The Singer

Once, in summer, when I was returning home from my work, I hobbled my horse in the mountain pass of Qïzïl-Qïya, lay down in the grass and fell asleep. In my dream I seemed to be riding in the pass of Qïzïl-Qïya, when an old man came towards me and led me with him. … This old man took me into forty yurts and in every one of them I drank a bowl of kumiss [fermented mare’s milk]. When I woke up it was already evening. Now, from this time onward I began to narrate the epic. (Kydyrbaeva 1996: 387 trans.) This is the initiation dream of a Kyrgyz epic singer, more specifically of a manaschï, a singer of the epic Manas. This singer, Choyuke Ömür-uulu (1863–1925), claims like many other Kyrgyz manaschïs to have had a dream or vision, in which either Manas himself or some other protagonist of the epic appears to him and commands him to sing the epic. These initiation dreams are also known from contemporary Kyrgyz manaschïs. Talantaaly Bakchiev (b. 1971), who is well known for his performances of several episodes from the Manas cycle, reports that he had a number of dreams in which Manas as well as other heroes and heroines of the epic appeared and told him to become a singer. When I asked him in 2010 in Bishkek whether he had memorized any texts, he answered: ‘If had memorized a text I could perform for a maximum of only twenty minutes.’ Initiation by a dream and hence the gift of inspired recitation was for him the prerequisite for an extended performance.1 Classicists will be reminded of Hesiod’s inspiration by the Muses as told at the beginning of his Theogony and medievalists of the illiterate Cædmon’s divine initiation into the art of oral poetry in seventh-century Northumbria. These various experiences and their reflections in initiation tales are wide-spread and have been the subject of interpretation from many different angles.2 In some Siberian epic traditions the singer’s and the shaman’s initiation visions or dreams are very similar and attest to an ancient connection

24 Settings between shamanic possession and the gift of epic performance. I will come back to the traces of shamanism in oral epic performance and also in the epics themselves.3 Whatever the interpretation of these dreams and visions, to my mind the initiation tales can be best understood as an expression of the deep existential commitment that the decision to become a singer entails and of the singers’ conviction of the exceptional character of their gifts. Whatever the role of a dream or vision, the overwhelming majority of singer biographies show that the path to bardic art is through dedicated training, generally from an early age and lasting for several years.

Epic Singers: Types and Terms Before discussing some of the evidence we have for the training of epic singers, a few general remarks on the singer-narrator are called for.4 The performers of oral epics are known by a multiplicity of names in the different cultural areas where they are to be found. Every name entails a number of specific characteristics which distinguish this particular singer type from another singer type. Sometimes various names for the same type of singer-narrator are used; sometimes different singer-types flourish in the same cultural space. Every name and every type has to be discussed individually in order to avoid ambiguities and misunderstandings. As an example I will briefly comment on the terms for singers from three Turkic traditions; other terms will be clarified as they occur. The Uzbek singer Chāri-shāir, who was mentioned in the Introduction, is called shāir, ‘poet’, an Arabic word (shācir) which is used as a term of honour. The usual Uzbek word for the epic singer is bakhshi, but other names such as dāstānchi (‘dastan performer’), sāzanda (‘musician’), and sannāwchi (literally ‘prattler’) are also found. The word bakhshi is already recorded in Old Turkic (c. eighth to thirteenth c.), where it means ‘teacher, master’ and is thought to be derived from Chinese. It has the meaning ‘epic singer’ in Uzbek, Turkmen, and Karakalpak, while it has the meaning ‘shaman, faith healer’ in Kazakh, Kyrgyz, and Uyghur. In Uzbek bakhshi can also mean ‘faith healer’; by the middle of the twentieth century no Uzbek singers were shamans, but it is reported that older singers did remember cases of the union of singer and shaman in one person. No doubt, bakhshi is a word rich in meaning and associations, suggestive of the closeness between singing and trance.5 Among the Kyrgyz the performer of the Manas epic cycle is called manaschï, while the performer of other Kyrgyz epics (of which a great number exist) is called aqïn. Both types of singers differ in their manner of performance: the manaschï sings the epic without instrumental accompaniment, the aqïn sings to the accompaniment of the qomuz, a plucked string instrument. Many aqïns, however, perform only lyrical songs and contest songs and are celebrated for their ability to improvise.

The Singer  25 Both terms have further ramifications. The term aqïn is also the Kyrgyz term for the poet in general, not only the oral poet and performer of traditional poetry. As a name for the poet and oral singer, aqïn is also found in Kazakh. As to manaschïs, singers of the epic who specialize in the second part of the cycle (on Manas’s son Semetey) are called semeteychi; this term is sometimes also used for singers only performing the third part of the cycle (on Semetey’s son Seytek). Also several degrees of manaschïs are differentiated, depending on how comprehensive and developed a singer’s competence is, of which the highest grade is chong manaschï ‘great manaschï’, designating a singer who can sing the whole cycle. Earlier terms for the singer of Manas were jomoqchu (from jomoq ‘tale’) and ïrchï (from ïr ‘song’); ïrchï was also an alternative term for the aqïn. A similar division into two types of singers, characterized by their mode of performance and by their epic repertoire, is found among the Karakalpaks. The Karakalpaks (‘black hats’) speak a Turkic language close to Kazakh and the majority live on the lower reaches of the Amu Darya (the Oxus of Antiquity) and the southern shores of the Aral Sea. The jïraw (a word derived like Kyrgyz ïrchï from jïr/ïr ‘song’) performs to the accompaniment of the qobïz, a bowed chordophone, while the baqsï (the Karakalpak form of Uzbek bakhshi) accompanies himself (or herself) on the dutar, a plucked lute. The jïraw sings predominantly heroic epics, the baqsï romances, with some overlap, however. Although both kinds of singer perform within the same ethnic group and cultural milieu, they differ significantly from each other. The jïraw’s epic repertoire harks back to the Noghay Horde of the fifteenth century, while the baqsï’s shows close connections to the oral traditions of the Turkmens and Khorezmian Uzbeks. In Turkish and Azeri (Azerbaijanian) the traditional oral singer is called âs¸ık (Turkish) or ashıq (Azeri). The word âs¸ık (from Arabic) means literally ‘a person in love’. In a religious context, this can mean ‘in love with God’, and there is indeed a strong religious element in much of âs¸ık poetry. P. N. Boratav provides a detailed explanation of the word with reference to the dreams some âs¸ıks aver to have had, dreams in which their pir or patron saint, sometimes also a beautiful girl, their future bride, offers them the âs¸k bâdesi, the ‘cup of love’. Among the ashıqs in Azerbaijan various types can be distinguished: (1) the ashıqustad is the singer of tales, the transmitter and performer of dastans, who is both a creative traditional poet and a musician; he is also a master and teacher of his art, hence the title ustad or usta ‘master’ (from Persian); (2) the ashıq-performer is also an accomplished narrator and musician, but performs only literary and musical works that have been composed by others; (3) finally, the ashıq-poet or el shairi (‘folk poet’) composes poems and music, is often more of a poet than a musician, but differs from the poet of written literature in that his productions belong to the

26 Settings realm of folk poetry.6 In Turkey the âs¸ıks are musicians and folk poets; at an earlier time âs¸ıks also performed romances in different parts of the Turkish-speaking area; by the second half of the twentieth century, however, the tradition of tale-telling was only alive in north-eastern Anatolia, especially in the province of Kars. In many traditions epic singers are professionals, but in many areas singers cannot earn a living by their profession any more. T. Bakchiev is a university lecturer and Chāri-shāir was a tractor driver and the leader of a workforce (brigade) on a kolkhoz. In some traditions the professional singer has disappeared, at least the professional with no other job than that of performing before an audience. This is true of the bylina singers who have been recorded right up to the last century in northern Russia, but it is also true of the South Slavic singers of tales. A. B. Lord states that the singers Milman Parry recorded in former Yugoslavia generally had some other occupation than that of performing heroic songs and that ‘professionalism was limited to beggars’. A ‘kind of semiprofessionalism’ was found among the Muslims in the evenings of the Muslim fasting month, when junačke pjesme (‘heroic songs’) were performed in the coffee houses. The Slovene scholar Mathias Murko, who studied South Slavic oral epic songs at the beginning of the twentieth century, mentions an exceptionally vigorous singer, 85 years of age, who was in former days employed by a beg or Muslim nobleman.7 Women Singer-Narrators The singing of epics is generally the domain of men, but this is not universally so. Among Uzbek bakhshis we find women early on, with Tillakampir in the eighteenth century and Sultān-kampir in the nineteenth century (kampir meaning ‘older woman’). In Azerbaijan a sizable group of women ashıqs are found, most of whom are musicians and perform folk poetry; the same is true of Karakalpak women baqsïs. Among Kyrgyz epic singers, one of the most famous semeteychi was a woman, Seydene Moldoke-qïzï (1922–2006). She refused to follow the traditional calling to become a singer at first on the grounds that she was a woman and manaschïs were men. In the end she followed her calling and became a highly respected epic performer. In Chapter 5, I will illustrate the gesticulation of Kyrgyz manaschïs by the recording of a woman singer. More than forty epics formed the repertoire of the last traditional Yakut woman singer, Dariya Tomskaya (1913–2008). While Dariya Tomskaya performed the Yakut oral epic (olonkho) in the traditional style and to a wide audience, woman singers in the South Siberian traditions perform the epic in a somewhat restricted manner, by reciting rather than singing. The same has been reported of Muslim women in Bosnia and Herzegovina at the beginning of the twentieth century. These women did not perform in public.8 In Turkic epic traditions, women are also found

The Singer  27 among amateurs. A great number of Tatar oral narratives and epics, for instance, have been written down from women. A special case are women performing exclusively for a female audience, as for instance in the oasis of Khorezm. Khorezm as a geographical and historical area is wider than the modern province of Khorezm in north-western Uzbekistan on the lower reaches of the Amu Darya. It comprises parts of north-eastern Turkmenistan, where one of the former capitals, Köne-Urganj (Gurganj), is situated, and the south-eastern part of Karakalpakstan. In the seventeenth century Khiva became the capital; in 1873 the khanate of Chiva came under Russian rule. These Khorezmian women performers are called khalfa (‘apprentice’). They are of various types; the so-called ‘single khalfas’ recite dastans from memory or read them from books or manuscripts. They do not sing or accompany themselves on a musical instrument. They recite mostly religious dastans and are also involved in various religious ceremonies (such as circumcision).9 The ‘Reader of Tales’ In some cases the ‘singer of tales’ is actually a ‘reader of tales’. This is true of the performance of the Arabic sīra (plural siyar), popular pseudohistorical prosimetric narratives, among which the Sīrat cAntar, the Sīrat Baybars, and the Sīrat Banī Hilāl have wide currency. While in the latter case an oral tradition has continued into the late twentieth century, the other siyar used to be performed from manuscripts or printed sources in the coffee houses of the Levant and Northern Africa. E. W. Lane, who lived in Egypt from 1833 to 1835, left a vivid account of the public entertainers in the coffee houses. He made a distinction between the performers of the three siyar mentioned. The performer of the Sīrat Banī Hilāl was called shācir ‘poet’; this singer, like his modern descendant in Egypt, ‘always commits his subject to memory, and recites without book’, sings the verse passages, and accompanies himself on the rabāb. The Sīrat Baybars, named after the thirteenth-century Mamluk sultan, was recited without a book and without instrumental accompaniment, while the Sīrat cAntar was read from a written source. The coffee house entertainers, generally called rāwī, cultivate a narrative art between orality and literacy. On the one hand they use written sources, on the other hand these sources are themselves a reflection of an at least partially oral tradition and become, in the hands of a skilful narrator, more of a prop than a text to read out aloud.10 The professional reader of popular narratives and romances is called qissakhān in a number of Turkic and Iranian traditions. The word is derived from Arabic qiṣṣa ‘story’ and Persian khwāndan ‘to read, to recite’. It is characteristic of these public entertainers that they did not sing or accompany themselves on a musical instrument. Although they did not always keep to the book or manuscript in front of them and felt free to

28 Settings embroider the tale to be told, they were, nevertheless, one of the main channels for the transmission of written poetry to a popular audience (on the related figure of the Turkish meddah, see Chapter 5). With reference to the Uzbek tradition, V. Zhirmunsky and H. Zarifov stress this role in the dissemination of romances from written sources: In the towns and the villages around towns there were also professional readers (the so-called qissakhan), who read popular books aloud to an illiterate audience, performing in the bazaars or, by invitation, in private homes. On these occasions an experienced reader could retell a text from memory, with corresponding individual deviations. The folk singers (ashulachi) had in their musical and poetic repertoire works of Classical Uzbek poetry and music. Through these routes the influence of written literature has long since penetrated into Uzbek oral epic poetry. (1947: 28–29 trans.) They also point out that the bakhshis of Khorezm have a particularly close relationship to written sources: It is well known that in Khorezm the influence of the feudal urban centres, with which the life of the rural population of the small oasis was intimately connected, was particularly strong. Here singers who can read and write are not rare. They do not improvise the verse parts of their dastans, but learn them by heart, holding sometimes the manuscript of the poem in their hands as a help for their recitation. (1947: 55 trans.) As with other aspects of the oral epic and epic performance, the figure of the singer-narrator shows much variety across oral traditions, and yet we can also locate a number of similarities that allow us to establish some general characteristics. One concerns the way a singer learns the art of oral performance and acquires his or her epic repertoire. In many cases an emphasis on active learning and training can be observed. A better grasp of the learning process helps to throw light on the singer’s ‘performative competence’, the ability to perform often sizable epics without the help of a written text.

How to Become an Epic Singer When Mathias Murko studied South Slavic heroic songs in Bosnia and Herzegovina in the years preceding World War I, he heard that the youngest singer performing in public was fifteen years old, while the oldest was 104. He added: ‘The singers begin to “acquire” their songs at a very early age; when they are still sitting on the knees of their father or some

The Singer  29 relative, they try their skill at playing the gusle and in singing’ (1919: 281 trans.). What Murko said about the early start of learning the art of epic singing in Bosnia and Herzegovina at the beginning of the twentieth century is true of many epic traditions even today. Quite naturally children become first attracted to the epic art in a family of epic singers and hence often take their first steps as performers under the tutelage of a close relative, often their father. Sometimes, however, apprentice singers begin their training only in their teens. In Kyrgyzstan I have several times heard young boys recite extracts from the Manas epic; some of these children appear also on YouTube.11 These early steps are based on imitation. Imitation, however, is not enough. Many singers have stressed the role of training, that is, the acquisition of skills and of a repertoire, under the guidance of master singers. In the acquisition of musical competence, the learning process of the epic singer parallels that of the traditional musician.12 South Slavs In his Singer of Tales A. B. Lord addresses the question of how to become an epic singer in the South Slavic oral tradition. The starting point of his analysis is the observation that in an oral tradition there is no text for the singer to memorize: ‘An oral poem is not composed for but in tradition.’ From this follows that ‘singing, performing, composing are facets of the same act’. Then Lord distinguishes three stages in the training process: In the first stage the future singer’s interest is kindled by listening to the performance of others, often a relative, and, when still a young boy, he then tries to recall the words of the epic poem and begins to learn to play the accompanying instrument. In the second stage the singer-to-be improves his skills by imitating a master singer. ‘Learning in this second stage’, writes Lord, ‘is a process of imitation, both in regard to playing the instrument and to learning the formulas and themes of the tradition. It may truthfully be said that the singer imitates the techniques of composition of his master or masters rather than particular songs.’ The third stage begins with the singer’s public performance; it is characterized by the singer’s continual widening of his skills and his repertoire.13 Mande What Lord writes about the South Slavic guslar (‘player of the gusle’) has been said in similar words about a number of oral epic singers, also in fairly different and distant traditions. I will bring a few examples, the first coming from West Africa. Charles Bird gives a succinct description of the apprenticeship of a singer of the Mande hunters in Mali, whose language belongs with Bambara, Mandinka, Mende, and others to the large Mande

30 Settings language group. Two types of oral epics are current among the Mande peoples, the epics of the griots and those of the hunters’ bards. The latter are variously called heroic songs and epics; a version of Kambili edited and translated by Bird, for instance, comprises over 2,700 lines.14 Bird underlines the long and demanding process of training, which comprises several stages: The position of the hunters’ bard is the result of a long apprenticeship spanning upward of ten to fifteen years. The principal bard with whom we worked, Seyidou Camara, has been singing for twenty years and has enjoyed the status of a master singer, or ŋara, for about the past eight or nine. His ensemble was composed of his wife and three apprentices, the youngest of whom was about twelve years old and the oldest about twenty-five. The oldest apprentice has been with the group for about six years. The first stage in the apprenticeship is devoted principally to mastering the ŋarinya, a ridged metal pipe that is scraped with a metal bar to provide the basic rhythmic background to the hunters’ songs. The next stage of his development is devoted to mastering the donso-nkoni, a six-string harp-lute with a calabash resonator. In most ceremonies, the master will be accompanied by at least one apprentice playing a harp-lute. The master himself will play a harp-lute, weaving extremely complex rhythmic patterns in between the base rhythm and the harp-lute of his apprentice. All this time the apprentices are listening to the prosodic style of their master, analyzing his techniques, and gradually perfecting their own oral art. As we will see, the complexity of this art is such that the apprenticeship is necessarily long and difficult. (Bird 1972: 279) Arab Studies of the performance of the Arabic Sīrat Banī Hilāl, a tale of the tribe of the Banī Hilāl, also known under the name of one of its main heroes, Abū Zaid, which has survived as a sung oral epic in Egypt, show a similar picture: First, a boy heard the stories told and retold by family members in performances. When he reached a certain age, usually between five and seven, family members would begin to quiz him on details of the story. Great emphasis was placed on mastering the complex plots; this process, however, focused only on the narrative elements, not on mastery of the poetry as such. Next, the boy was taught to play drone notes and short melodic phrases as ‘responses’ on the rabāb [a spike-fiddle] so that, though not yet singing portions of the epic,

The Singer  31 he could perform alongside his father or an uncle or grandfather. The time spent traveling between performances was often devoted to more and more detailed questioning about the epic. Since the child had always heard the story in sung verse, it was only a matter of time before he began to respond to questions in verse rather than prose. Quite simply, at some point it would become easier to respond to specific questions in poetry (as heard) rather than recast an idea into prose.15 Mongol As a parallel and complement to the traditions already mentioned, I will now turn to the oral traditions of Asia. S. A. Kondrat’ev remarks about the training of the west-Mongolian or Oirat epic singer, called tul’chi (from tul’ ‘narrative, epic’)16: The heroic epics are usually learned at a young age among the Oirat of north-western Mongolia. Most often it happens that someone who feels an interest and love for these works of oral literature begins to perform extracts of the epic in a playful manner, imitating known epic singers, tul’chis, who have often been listened to. If the young Oirat afterwards discovers that he has the talent, the capability and the desire to become a singer of heroic narratives, he will turn to some well-known singer and go to him for instruction. … When he begins his training, the young man tries first of all to familiarize himself with the subject matter of an epic he likes and with the development of the plot. … In order to orient himself, he learns to divide the epics into their component parts – into beginning, main part and sub-plots; he learns to distinguish various descriptions, for instance of a place, a horse, or the beauty of some princess, contrasting them with the description of the deeds of the hero, the main subject. After he has mastered these and has firmly fixed the development of the plot in his mind, the young apprentice proceeds to learn about ‘commonplaces’ and ‘ornamentation’, i.e. figurative expressions. … The apprentice learns and internalizes various repeated ‘commonplaces’, for instance, attacks, praises of the hero’s native land, the description of his horse and the description of the fight of two heroes. Then he learns a great number of expressions, rhetorical figures, and epithets and tries to apply them to a familiar plot. At this stage the young singer-apprentice normally retires to some secluded place, in the steppe and the mountains, and there, with the tobshur [plucked string instrument] in his hands, he practices singing the epic he had learned in the way befitting a true singer, that is, with no omissions, not even of one episode, and with skilful variation of the ‘commonplaces’ of the descriptions, ornamenting them with favourite epithets

32 Settings and other figures. With a good memory, love, and especially inspiration, the young man, following the teaching of his teacher, is quickly able to learn to sing the long heroic epic, comprising four to five thousand verse lines, and to sing it quite well. But this is only the beginning: when the young singer has learned to sing an epic, he is still far from the goal of receiving the title of a true singer – a tul’chi. (Kondrat’ev 1970: 9–10 trans.) Uzbek A fairly detailed analysis of the learning process is given by V. Zhirmunsky and H. Zarifov in their still definitive study of Uzbek oral epics of 1947: The pupil learned above all by listening to the singing-narrating of his teacher in his native village and by accompanying him during his travels to other villages. In addition to this the teacher also gave his pupil specific tasks, which comprised committing to memory typical passages of the dastans, as well as epic stereotypes, such as the saddling of the horse, the hero’s setting-out, the description of a battle, specific fully developed episodes and so on. In other respects, the beginner was only able to keep to the basic thread of the epic narrative, like to a script, varying the teacher’s performance as best as he could. In this way the young singer’s ability for creative improvisation was formed, a quality that distinguishes the best and most talented narrator-singers (shairs). … When the teacher heard his pupils’ first attempts, he would correct deviations regarding the content as well as mistakes of style and of word usage. He would give critical comments on these and offer advice, urging the pupil to remake unsuccessful passages or showing by his own example models of a correct performance. … A dastan was not learned at once, but in separate portions …. Later on the pupil ‘put together’ a whole poem from the separate passages he had learned. For this the connecting passages were not learned by the young singer but improvised in the traditional style. … When the teacher thought that the training was complete, he would arrange a kind of public examination for his pupil. Such an examination was organized either at the house of the singer himself, who invited guests and hosted them, or in the pupil’s village or in one of the neighbouring villages, where the teacher himself often performed and where there were old men, true connoisseurs of singing, who had heard many good dastanchis. The teacher would introduce his pupil to the audience assembled and either ask them to choose any of the poems in the repertoire that the pupil had

The Singer  33 mastered or propose himself a certain dastan to the pupil. The pupil would then perform the poem in its entirety for the first time, without the participation of his teacher. At the end of the performance the teacher would ask the listeners for their opinion. In the case of a positive response, the young singer received the title of bakhshi; he was considered to have completed his training and had the right to sing independently. (36–37 trans.) It emerges from these descriptions that learning to perform an oral epic requires a number of different skills which are acquired in somewhat different ways. After the first interest is kindled, generally at a young age, the learning process is largely imitative. The young singer learns to play the accompanying instrument(s); in the Mande case this instrumental apprenticeship is particularly stressed. The apprentice singer listens to the master singer’s performance of an epic and first of all tries to get a firm grasp of the plot. The next step is, as it were, analytic, leading to the realization of the various narrative and stylistic elements from which the epic is composed. As Bird summarily puts it: ‘All this time the apprentices are listening to the prosodic style of their master, analyzing his techniques, and gradually perfecting their own oral art.’ An interesting point in the description of the making of a bakhshi is that the pupil’s performance is examined and corrected by the teacher. Such criticism implies that epic passages can be right or wrong, in other words that there is a certain normative element implied in the learning/teaching process. Zarifov and Zhirmunsky’s presentation of the art of the Uzbek epic singers is similar to what Wilhelm Radloff had said about the Kyrgyz manaschïs when he explained their performance as a skilful combining of themes (called Bildteilchen or Vortragsteilchen by Radloff): ‘The art of the singer consists mainly in stringing together all these ready-made narrative units as the course of the narrative demands and linking them with newly composed verse-lines.’17 Finally, somewhat at variance with the idea that a particular passage has a correct form is the observation that a pupil should vary his teacher’s performance in order to develop his ‘ability for creative improvisation (tvorcheskaya improvizatsiya)’. I will come back to this idea below with an example from Uzbek epic poetry. The picture which has been established of a singer’s training can be further elaborated and modified by looking at other traditions.18 Some of these traditions, and their respective singers of tales, will be discussed in the following chapters.

34 Settings

The Chain of Transmission One of the voices sceptical of the purity of oral transmission as postulated by Milman Parry and Albert Lord is that of Ruth Finnegan, who adduces the existence of ‘singers’ schools’ in a number of cultural areas as an argument for a mixed oral-written tradition: In such cultures there may be a conscious learned tradition in which literary specialists deliberately train new recruits into their acquired skills – the poetic training in Ruanda, for instance, the Maori ‘schools of learning’, the Uzbek singer-teachers with their various pupils, and perhaps the medieval Irish bardic schools. These often involve careful control over recruitment and, sometimes, monopoly over particular types of literary productions, and seem far from the unconscious and undifferentiated kind of culture sometimes assumed as the characteristic context for oral literature. Again, this point is not just about the exceptional case, for a large proportion of recorded oral literature – for instance the vast corpus of oral epic poetry from Central Asia – in fact comes from this kind of context. (1974: 54–55) Singers’ Schools On closer analysis, the examples Finnegan lists are quite different from one another and would all merit a more thorough discussion. The concept of a singers’ school is also found in the Slavic world. Mathias Murko, for instance, used the term ‘school’ for different performance styles in an article from which I have already quoted (1919: 285). The term ‘school’ was, however, used much earlier by the collector of Russian bylinas Alexander Hilferding (Gil’ferding) in 1871 with reference to the bylinas performed in the Lake Onega area (1873: xxx). These singer schools (shkoly skaziteley) were later studied in great detail by the Russian bylina scholar Vladimir Chicherov (1982). The term ‘school’ means a special style cultivated by a group of singernarrators who trace the acquisition of their repertoire to a particular singer. The formality of training varies. The crucial element in the training process is imitation. In order for the training to be successful a singer must start early, have a good memory, the ability to play a musical instrument (in most, but not all traditions), and more generally the gift of performing with confidence and skill in public. Different local styles (‘schools’) are found in all Turkic oral traditions, and singers from all traditions have stressed the decisive role of a model or teacher in their training. Here I will focus on the Uzbeks. By dāstānchilik maktabi ‘school of dastan art’ both an institution and a place is meant. The place is the master singer’s home, where the pupil bakhshis live. Traditionally

The Singer  35 the apprentices did farm and house work in exchange for the teaching they received. The Karakalpak epic singer Jumabay Bazarov, for instance, stayed three years with a master singer and in that period learned three dastans. The young singers acquired their teacher’s repertoire and style. This repertoire and style their teacher had learned from his teacher and the latter from his teacher and so on. In this way ‘singer genealogies’ have been compiled, where singers ultimately ‘descending’ from the same ‘ancestor singer’ are grouped together. A number of such schools have been identified, named according to their geographical position. The best studied are: (1) the school of Bulunghur, in the Samarkand province, (2) the school of Qorghān, also in the Samarkand province, at the foothills of the Nurata mountains, (3) the school of Shahrisabz, in the Qashqadaryā province, (4) the school of Qamay, also in the Qashqadaryā province, (5) the school of Sherābād, in the Surkhandaryā province, (6) the school of southern Tajikistan (of the Uzbek-Laqay), and (7) the school of Khorezm.19 The Uzbek singers’ schools differ from one another in various respects. One of them is the repertoire of the master singers, and hence their pupils. In the ‘school of Bulunghur’, the bakhshi Fāzil Yoldāsh-oghli was trained (see below). As it turned out, there were three brothers in his native village Lāyqa (in the Bulunghur district), all of them singers: Yoldāshshāir, Qoldāsh-shāir, and Suyar-shāir. Fāzil learned the art of a bakhshi from Yoldāsh-shāir, the oldest brother, who was the pupil of a singer by the name of Yoldāshbulbul, whose teacher was in turn a singer of the second half of the eighteenth century, Muhammad-shāir. The school of Bulunghur was famous as a centre for the cultivation of heroic dastans. Representatives of the second school, connected to the name of the village Qorghān, but whose bakhshis are found in a much wider area south and north of the Aqtau mountains, are, for instance, Ergash Jumanbulbuloghli (1868–1937) and Muhammadqul Jānmurād-oghli Polkan (1874– 1941). In the repertoire of these singers love romances and adventure dastans predominate. ‘The most salient stylistic traits characterizing the dastans of the singers of the school of Qorghān are their pronounced lyricism, beautiful descriptions, poetic ornaments, the careful elaboration of details, gracefulness and decorative splendour’ (Imāmov et al. 1990: 237 trans.). Similarly, in other schools particular dastans are preferred – such as dastans from the cycle of Goroghli by the Uzbek-Laqay in Tajikistan or love romances by the bakhshis of Khorezm.20 One of the most widely known contemporary bakhshis of the ‘school of Sherābād’ is Shāhberdi Bāltaev (b. in 1944), from whom I was able to record a branch of the Goroghli cycle (Nurali) in 2002.21 Finally, the dastans of the bakhshis of the ‘school of Shahrisabz’ are characterized by a lively, often humorous narrative style. The most distinguished singer of this school is Islām Nazar-oghli (1874–1953), who was also influenced by the ‘school of

36 Settings Qorghān’. Other aspects differentiating the singer schools concern the musical side of performance, dialectal and regional traits, influences from other oral traditions, and the effects of literacy. Similar ‘schools’ have been posited for other traditions, among them Turkmen and Kyrgyz.22 The Singers’ Awareness of Transmission Although the learning of epic performance is described as a structured process involving imitation, analysis, repetition, memorization of fairly fixed passages, correction, and also variation, it is still conducted by word of mouth and is unlike written instruction. The institution of the dāstānchilik maktabi is not a school in the normal sense of the word and does not counteract oral transmission. Also without this institution, there is a strong sense of standing in a line of singers in many traditions. Often singers make explicit references to this chain of transmission at the beginning or end of their performances. A Mandinka version of the epic of Sunjata begins with the words: ‘It is I, Bamba Suso, who am talking.’ Here is the singer, squarely putting himself into the foreground. The griot continues by also introducing the kora player Amadu Jebate, who accompanies him.23 He then indicates the home of both singer and musician and specifies the origin of his tune and of his tale: The tune that I am now playing, I learned it from my father, And he learned it from my grandfather. Our grandfather’s name – Koriyang Musa. That Koriyang Musa Went to Sanimentereng and spent a week there; He met the jinns [benevolent or malevolent spirits], and brought back a kora. Tale and tune, it is told, come from the singer’s grandfather, transmitted to him by his father. Both the singer’s and his kora player’s fathers were griots and cousins. Then, after twenty-three lines of introduction, the singer is ready to tell his tale, with the customary appeal to listen carefully: All right,24 I’m going to tell you the story of Sunjata, And you must pay attention. The narrator – as a person facing his audience – is making it clear that he is telling a tale that he has been told, or rather taught, by another teller of tales. As listeners (and later readers) we assume on the basis of the introductory words that we are not hearing (or reading) a story that

The Singer  37 originated with the narrator. He did not compose the story himself. But it is he, Bamba Suso, who is telling the story. There seems to be a certain tension here between narrator and tale, between putting the spot-light on the performer, or rather performers in this case, and yet receding into the background when it comes to claiming authorship of the story – and the tune. This version of Sunjata is not the only West African epic where we find an explicit reference to performers and the chain of transmission, and the occurrence of such lines in the griot tradition is no singular phenomenon in the wider field of oral epic poetry. The singer’s reference to himself or herself and to the source is, however, generally part of the information given outside the performance, to an interviewer or also to a curious audience. We know, for instance, from Milman Parry’s and Albert Lord’s recordings of South Slavic guslars and the interviews conducted with the singers by their helper Nikola Vujnović, that a number of them had heard and learned their epic songs from a blind singer of Kolašin, Ćor Huso Husović. In Lord’s words, he had become a legendary singer in the Sandzak of Novi Pazar, where we first heard of him, and in Montenegro. … to Parry, I believe, he symbolized the Yugoslav traditional singer in much the same way in which Homer was the Greek singer of tales par excellence. (1987: 473) By the time Parry did his recordings, Ćor Huso had passed away, but he was, despite his legendary fame, a singer who lived only one generation before the singers that have been immortalized in the Parry collection of South Slavic oral epics. In some traditions genealogies of singers transmitting an epic are given by the singers and on the basis of this information are later compiled by scholars. In a Karakalpak version of the epic of Edige we find a singer genealogy reaching back further than one or two generations. This version was written down from the Karakalpak singer Qïyas-jïraw Qayratdinov (1903–1976). At the end of his version he says: There is no untruth in my words. The dastan of Edige, Which is spread among the people, Has come from Soppaslï-jïraw. When he answered the questions posed, He was honoured with gold from the khan. As his pupil he took The Karakalpak Düysenbay-jïraw. … Taken as his pupil,

38 Settings Very eloquent in the old style, Wandering constantly among the people, Was Düysenbay’s apprentice The singer-jïraw Seydulla. … Seydulla-jïraw gathered pupils round him, Travelled to well-known places for pasture, Committed the words to his memory And spread them widely, Spread the dastan of Edige Through the whole Karakalpak people.25 In this epic, a wise jïraw by the name of Soppaslï Sïpïra-jïraw, or simply Sïpïra-jïraw, makes his appearance. He prophesies at the court of Tokhtamysh, khan of the Golden Horde, that Edige will one day become a serious threat to him. What Soppaslï Sïpïra-jïraw prophesies in the epic agrees with what we know from history. Tokhtamysh was khan of the Golden Horde and Edige his emir (commander), who rebelled against him. In the winter of 1406/1407 Tokhtamysh was killed; Edige, who had practically ruled the Golden Horde since 1395, died in 1419. According to Qïyas-jïraw the epic was first composed by Soppaslï Sïpïra-jïraw, who passed it on to Düysenbay-jïraw, who in turn passed it on to Seydullajïraw, from whom Qïyas-jïraw learned the epic. If Sïpïra lived at the time of Tokhtamysh and Edige and Düysenbay-jïraw in the early nineteenth century, there is a gap of at least four hundred years between the legendary singer Sïpïra-jïraw and his pupil. This gap does not, however, enter into the singers’ reflections. This is illustrated by a dialogue with the Kazakh singer Murïn-jïraw (1860–1952). Not only the Karakalpaks, but also the Kazakhs claim Sïpïra-jïraw as their first epic singer. When Murïnjïraw was asked by which transmission chain his epic cycle ‘The Forty Noghay Heroes’, to which Edige belongs, came to him, he answered: I cannot go into depth, my friend. Qashaghan-jïraw said: ‘I got these epics from Nurïm-jïraw.’ And Nurïm-jïraw said: ‘I got these epics from Abïl-jïraw.’ Abïl-jïraw said: ‘I got these epics from Sïpïra-jïraw.’ Esekeng [Murïn’s interlocutor] thought somewhat, shook his head and said: ‘This can’t be right. When you roughly determine Sïpïra’s life-time, then he was a man of the fourteenth century. There is therefore a great interval. But,’ added Esekeng, ‘it is possible to say that the origin of the epic goes back to Sïpïra-jïraw.’ (Khakimjanova 1990: 256 trans.) Clearly, three or four hundred years are nothing to the singers, in this case even less so, as in the epic Edige, Sïpïra-jïraw is said to have lived for 360 years!

The Singer  39 Despite seeing himself as a tradition bearer, as a singer who performs an epic transmitted to him through a chain of singers, Qïyas asserts his own role in the life of the epic, his ownership, as it were, of the version he has just performed. These are the final lines: I have travelled among the Karakalpaks And have sung for forty years for the people. Qïyas-jïraw has spread the dastan of Edige Far and wide. If you want to know from where it has come: This dastan has come from the mouth of Qïyas. It has to be added that the various versions of this epic recorded from Karakalpak singers are surprisingly close to one another, in plot, motifs, typical scenes, and even wording. In fact, the epics and tales about Edige written down from other Turkic traditions (Kazakh, Noghay, Tatar) also conform to the same overall plot and structural pattern.26 This shows that the epic tradition of the Karakalpaks is conservative, that is, that the singers keep fairly closely to the version they learned from their teachers. Variations do, of course, occur, due to the variable skills of individual singers, to their exposure to different versions and to their own creative energy. Even when highly divergent versions of an epic are performed, they all share a common narrative tradition regulating the legendary background of the tale and the poetic and performative forms of its expression. The singers perform their epic, but it is an epic which they have heard and learned from others and which they share with other singers. It belongs to them, but it is also, as it were, common property.

Creativity and Innovation Singers and narrators assume multiple roles in the shaping of an epic: they are the transmitters of tradition as well as creative forces in the transmission process, they are the guardians of tradition as well as innovators, and they are the voice of tradition as well as the bearers of their own individual voice. Depending on their leaning either towards fidelity to the transmitted poem or to creative variation, different kinds of singers can be distinguished. In the Russian bylina tradition, Alois Schmaus has posited an ‘intensive’ and an ‘extensive’ type of singer. The former strictly adheres to the tradition, to what has been passed on to him by his masters, and concentrates his efforts on strengthening the effects of the narration and on elaborating the poetry in the traditional epic style. The latter, on the other hand, is an innovator, with a gift for improvisation, a tendency to re-compose and re-arrange narrative units and an interest in

40 Settings extending the narrative by cyclization.27 The ‘intensive’ or conservative singer can become a memorizing performer when oral tradition is fed or at least supplemented by written texts. At the other end of the scale stands the singer-poet, the ‘extensive’ and creative singer who feels free to alter tradition and to compose new epics. I will comment on the creative singer, by using the Uzbek bakhshi-shāir Fāzil Yoldāsh-oghli (1872–1955) as my example. Fāzil-shāir has already been introduced above as the most prominent representative of the singers’ school of Bulunghur. From this singer more than thirty dastans have been written down, some of them twice. Fāzil’s version of Alpāmish is justly celebrated and generally considered the most accomplished Uzbek version of the epic. Comparing Fāzil’s version of Alpāmish with other Uzbek versions, it is clear that he is telling the same story, drawing from the store-house of typical passages and epic stereotypes as well as using the conventions of epic metre and formulaic diction. As Zhirmunsky has shown, we can go even further and see the various Uzbek versions of Alpāmish as only one of several branches of a widely disseminated epic and hero tale. Fāzil’s version is more elaborated than other Uzbek versions, also longer, although not the longest. In 2018 a version was published (by the singer Abdunazar Pāyānov) which runs to 14,230 verse-lines, five hundred more than Fāzil’s version.28 Every version has its idiosyncrasies despite a great number of shared scenes, motifs, and elements of style and diction. Fāzil’s Alpāmish is impressive in its details, clear narrative structure, polished monologues and dialogues and pervasive poetic style, all, however, within the confines of traditional Uzbek poetics. The fact that Uzbek dastans are composed in a mixture of verse and prose gives the singer more freedom for the prose narration of the story than if the epics were entirely in verse. Also the verse passages, however, differ from version to version, although the singers express themselves in highly traditional and formulaic style. Changes in the Plot: Happy Endings It has often been affirmed by folklorists and students of epic that no two performances of an epic by the same singer are alike. This is also true of Fāzil-shāir. His creative impulse has, however, carried him further. In his version of the romance of Farhād and Shīrīn he actually changes the plot. This is a dastan with a venerable written tradition. The literary source of this tragic love-story is the verse-epic Khusrow and Shīrīn by the twelfthcentury Persian poet Niẓāmī. Niẓāmī was much imitated, both in Persian and in other languages. The story of Farhād and Shīrīn has also inspired poets writing in various Turkic languages. The most brilliant of these verse narratives is no doubt cAlī Shīr Navā’ī’s Farhād and Shīrīn. This poem was composed in 1484 and comprises about 5,780 couplets. It

The Singer  41 is from cAlī Shīr Navā’ī’s poem that the numerous popular romances in several Turkic languages are derived. The story of Farhād and Shīrīn celebrates the love of the stonemason Farhād for the beautiful Shīrīn, whose hand is demanded by King Khusrow. In Niẓāmī’s poem this is only an episode in a tale that is focused on Shīrīn and Khusrow, two historical figures, the Sassanian King Khusrow II (r. 590–628) and his Armenian spouse Shīrīn. In the story of Farhād and Shīrīn, Khusrow engineers his rival Farhād’s death by sending an old woman to Farhād, who announces Shīrīn’s death. In despair Farhād kills himself. Fāzil, however, ends his version with an unexpected twist of the story. When the old woman tells Farhād that Shīrīn has died, Farhād immediately sees through the crone and realizes that he is being tricked. He accuses the old woman of lying and kills her. The dastan has an unusual happy end with the marriage of Farhād and Shīrīn. It seems fairly certain that this change in the ending of the dastan was the singer’s invention and not provoked by some source he might have used. Navā’ī’s narrative poem circulated in popular retellings in the form of printed ‘folk books’ or chapbooks. The editor of Fāzil’s Farhād and Shīrīn identifies the author of this chapbook as the Uzbek poet Makhsun from the first half of the nineteenth century. This popular book had, however, the traditional tragic ending. The same is true of the Uzbek folktale of Farhād and Shīrīn.29 The change of a tragic ending to a happy one is not unique. It is, however, not always initiated by the singer. The Turkish folklorist İlhan Bas¸göz mentions such an occurrence during the performance of the Turkish hikâye Kerem and Aslı: We possess examples of the audience’s deep concern for a favorite hero. Everybody knows that in the story Kerem dies and that he is not united with his beloved. Nevertheless, one day a minstrel who performed the story had the following experience when he was telling the story of Kerem at a party given by a bey. It was towards the end of the story, Kerem was about to burn up. Suddenly, one of the audience, a person of high rank, pulled out his pistol and threatened the minstrel with ‘Either you don’t kill Kerem or I kill you’; whereupon the minstrel found a way out and did not kill Kerem, who, then was united with his maid. In south Anatolia, this minstrel is known as ‘the minstrel who does not kill Kerem’ (Şükrü Elçin: Kerem ile Aslı Hikâyesi, published by the Ministry of Education, Ankara, 1949). The story of Tufarkanlı Abbas, which today ends happily, formerly ended with death. The two lovers are separated because the hero dies from a snake bite. But later, the minstrels changed this ending to unite the lovers. (1952: 333, n. 11)

42 Settings A similar incident is mentioned by M. C. Lyons in his survey of Arab siyar: At whatever cost to verisimilitude, the hero cannot be allowed to die prematurely, if at all, but at the same time at the end of each evening the popular narrator must leave his audience anxious and eager for the sequel. A constant diet of last-minute stays of execution and executioners killed in the act of striking may not suit sophisticated palates, but its effectiveness can be seen in an anecdote in the Fulklūr Baghdād, which shows ‘the poet’, who had ended his evening’s recitation by leaving the Hilālī, Abū Zaid, in prison, followed home by one of his audience, who then forced him at knife-point to continue the story until the hero was released. (Lyons 1995: I, 74) The Composition of New Epics Fāzil-shāir’s creative talent is also shown by his composition of new dastans about comparatively recent historical events. I will briefly look at one of them, a dastan that treats of the uprising in Jizzakh, a town in southern Uzbekistan, north-east of Samarkand, in 1916. When the local population of Russian Turkestan was conscripted as a working force in Russia during the First World War, riots broke out that led to the storming of the fortress of Jizzakh and the murder of the hākim (governor). Troops were, however, brought in and the uprising was quelled. Fāzil’s dastan is composed in a mixture of verse and prose, the verse comprising c. 1,300 lines.30 The plot begins by placing the events in the First World War and describing the plight of the peasants and simple people, as well as their anger about the high sum necessary to become exempt from conscription, a sum only the rich (and corrupt) can muster. Due to the topic, a number of passages introduce the concepts and hence vocabulary of the modern world, such as telegrams, trains and Bolsheviks. The dastan has a clear political agenda, that is, to castigate the oppression of the peasants and the poor in Turkestan under the Tsars’ rule and to exalt the courage of the leaders of the uprising. There is also a strong anti-clerical element in the dastan, brought out in an episode when an imām admonishes the insurgents not to use violence, but is shouted down as the devil’s advocate. The dastan ends with the promise of a rosy future, expressed by young men from Jizzakh, who as labourers in Russia had heard of the Bolsheviks: When the Padishah [i.e. tsar] is thrown from his throne, The day we have been hoping for will come, Freedom will be proclaimed. Hear all of you gathered here,

The Singer  43 We will have a happy prosperity, We will see flourishing The country which was in tears and grief-stricken. The heavy days will have passed, The toil of the labourer will have finished. The tyrant will fall, The worker will have reached his goal. The dastan is composed in traditional language and in the style of traditional oral epic poetry. The plot unfolds in descriptions and passages in direct speech, both in prose and verse. The fiery speech of the insurgents’ leader Mawlān, the lament of the oppressed population, the description of the fighting in the fortress are specific to the topic of this dastan, but couched in a language that is steeped in the Uzbek epic idiom. Fāzil has here clearly poured ‘new wine into old skins’. Fāzil-shāir is not the only bard who has composed new narrative poems of epic breadth. A famous case is Dôkarim, an Aceh (Achehnese) of Sumatra, who at the end of the nineteenth century composed an oral epic on contemporary affairs, appropriately named Hikayat Pram Gômpeuni, ‘The song of the wars with the Dutch’. The Arabist Snouck Hurgronje, who was among the Aceh at that time, says about the poet: Writer we may not call him, for he can neither read nor write. He went on, as he tells us for five years gradually composing this poem in celebration of the heroic deeds of the Achehnese in their conflict against the Dutch, adding fresh matter from time to time as he gained enlightenment from eye-witnesses. … We can here witness for ourselves one of the methods by which an Achehnese heroic poem is brought into the world. Some one man, who like most of his fellow countrymen knows by heart the classic descriptions of certain events and situations as expressed in verse by the people of the olden time, but whose knowledge, owing to his training and environment, is somewhat greater than that of others; one who is endowed, besides, with a good memory and enthusiasm for the poesy of his country, puts his powers to the test by celebrating in verse the great events of more recent years.31 The Notion of Authorship Should we call singers like Dôkarim or Fāzil, the composer of Jizzakh qozghālāni, authors or is the concept of author out of place when it comes to oral poetry? The inappropriateness of searching for the author

44 Settings of traditional oral epic poetry is stressed by A. B. Lord, who maintains that the search for the original of an oral epic is misguided: Indeed, we should be fully aware that even had we this ‘original’, let us say, of the wedding of Smailagić Meho, we would not have the original of the basic story, that is, the song of the young man who goes forth into the world to win his spurs. We would have only the application of this story to the hero Meho. Each performance is the specific song, and at the same time it is the generic song. The song we are listening to is ‘the song’; for each performance is more than a performance, it is a re-creation. Following this line of thinking, we might term a singer’s first singing of a song as a creation of the song in his experience. Both synchronically and historically there would be numerous creations and re-creations of the song. This concept of the relationship between ‘songs’ (performances of the same specific or generic song) is closer to the truth than the concept of an ‘original’ and ‘variant’. In a sense each performance is ‘an’ original, if not ‘the’ original. (1960: 100–1) From this, Lord infers ‘that the words “author” and “original” have either no meaning at all in oral tradition or a meaning quite different from the one usually assigned to them’ and ‘that the author of an oral epic, that is, the text of a performance, is the performer, the singer before us’.32 Lord’s remarks are convincing, but one has to make some exceptions. One case is when an oral narrative is ultimately based on a written source. One could argue that the first bakhshi who turned Navā’ī’s Farhād and Shirin into a popular dastan was its author. On closer analysis, however, we realize that the written ‘original’ is no more than the source of the plot and perhaps some metaphors and poetic lines, but otherwise the dastan conforms to conventions of popular epic poetry, including, incidentally, the metrical form of the verse parts. Navā’ī and the oral bakhshi use different metrical systems, which makes wholesale borrowings difficult, although some bakhshis occasionally have poems in the quantitative metre of Classical poetry in their dastans. The second possible exception is the composition of new oral epics, for instance by adding another branch to a cycle of epics. The Kyrgyz bard Jüsüp Mamay (1918–2014) has done this with the Manas cycle, enlarging the canonical three-generation cycle to an eight-generation cycle. At my first meeting with Jüsüp Mamay in Ürümchi in 1985, I asked him whether he is able to make an epic out of a story he is told. After some reflection he said: ‘You mean, turn prose into verse. Yes, I can do that!’ And this is what he did with the later branches of Manas cycle, turned prose tales into verse. The results are traditional epics; plot-wise these are new epic tales, but the characters and events are variants and elaborations of the common stock of motifs and narrative units of the Kyrgyz epic. There is, however, some

The Singer  45 oddity about the later additions to the Manas cycle by Jüsüp Mamay and about new epics like Fāzil’s Jizzakh qozghālāni. They have not (or at least not yet) entered tradition. What makes newly composed oral epics untraditional is not that they are of recent date or treat of recent events, but that they are not accepted into the stream of living tradition. Pëtr Bogatyrev and Roman Jakobson have convincingly argued for a clear distinction between written and oral literature, which hinges precisely on the different concepts of author and transmission. They mention the French poet Lautréamont (1846–1870), who in his time was little read and appreciated, but who was re-discovered later by the Surrealists. This could not happen to an oral work. If an oral poet had composed a new work, but the audience, the community of listeners, did not like his work, it would be irretrievably lost. ‘Only the accidental recording of a collector can save it, by transferring it from the sphere of oral poetry into that of written literature.’ From this follows the important role of the audience in folklore. Bogatyrev and Jakobson coined the expression ‘Präventivzensur der Gemeinschaft’, the preventive censorship of the community, meaning that by the tastes and expectations that regulate the favourable reception of a work of folklore, the community shapes its birth and survival. Given this ‘preventive censorship’, there is a radical difference between the author of a literary work and an oral author: while the former can try to change ‘the demands of the milieu’, the latter is unable, and actually also unwilling, to do so. Bogatyrev and Jakobson support the idea of the individual rather than the ‘folk’ as the creator of a work of folklore, but oppose at the same time the equation of the ‘folklore creator’ with a literary author.33 Nevertheless, the individuality and creativity of outstanding bards does give them authority over the works of verbal art they perform, even if we reserve the term ‘author’ for the works of written literature. ‘The exclusion of the author can go too far. Even the author, poor devil, has a right to exist.’ Graham Greene wrote these lines long before Michel Foucault speculated about the disappearance of the author in modern fiction (1970: 93). Greene was thinking of the modern novel in the wake of Flaubert or James, and he was certainly not thinking of anonymous oral poetry. His words, although coming from a different context, are a reminder that it is easier in theory than in practice to exclude the notion of ‘authorship’ from oral tradition. It is not without a good reason that talented Uzbek bakhshis are awarded the title shāir ‘poet’ as a sign of respect.

Notes 1 Bakchiev has also written about the intiation dreams of Kyrgyz bards (2015: 69–139); compare also van der Heide’s dissertation (2008: 111–23). 2 For an interpretation of Hesiod, see Walsh 1984: 22–36; on Cædmon, see Magoun 1955 and Lord 1993; on comparative studies of initiation dreams, see Zhirmunsky 1979; Putilov 1997: 45–67; Tolley 2019.

46 Settings 3 See further Chapters 4 (‘Shamanic Voices’), 5 (‘Gesture and Inspiration’) and 9 (‘First-person Narration, Shamanic Traces’). 4 For a general survey, see the chapter on the bard in Bowra 1952: 404–42. 5 On the Uzbek bakhshi, see Zhirmunsky & Zarifov 1947: 23–58; on various singer types of the Turkic peoples, see Chadwick & Zhirmunsky 1969: 319–39; Reichl 1992: 57–91. 6 On the âs¸ık, see Boratav 1964: 129–31; for the Azerbaijanian ashıq, see Èl’darova 1984: 11. 7 See Lord 1960: 13–21; Murko 1919; for the bylina singers, see the discussion in Chadwick & Chadwick 1932–40: II, 238–69. 8 Murko notes that ‘Muslim women are often able to recite heroic songs (without instrument) in order to educate their sons in a martial spirit or to keep them at home so that they don’t seek other amusement’ (1919: 280–81 trans.). On the performance of Altaian women narrators, see Golubeva & Surazakov 1965: 3. 9 On the Khorezmian women singers, see Kleinmichel 2000; see also Sultanova 2016. 10 See Lane 1860: 391–425; Herzog 2012; on the contents and structure of the siyar, see Lyons 1995. 11 See 12 On the folk musician’s training, see Merriam 1964: 145–63. 13 Lord 1960: 13–29; quotations from pp. 13 and 23–24. 14 Excerpts of this edition and translation are published in Johnson, Hale & Belcher 1997: 100–13. On the hunters’ bard as opposed to the ‘casted bard’ or griot (jeli), see Bird 1972: 278–79; Bird prefers the term ‘song’, because these heroic songs can be newly composed on ‘the fairly recent deeds of a local hunter’, while the Mande epics ‘are invariably products of a long tradition of singers’ (290). 15 Reynolds 2002: 341–42; compare also Slyomovics 1987 and Reynolds 1995; see also Reynolds’ website with introduction, texts, translations and audio files. 16 On Mongolian oral epic traditions, see Appendix A. – Kondrat’ev uses the term ‘commonplace’ (obshchee mesto) in the sense of both motif and typical scene (the ‘theme’ of oral-formulaic theory). 17 1885: xvii trans.; for a critical discussion of Radloff’s ideas, see Reichl 2020; for an English translation of Radloff’s introduction to the Kyrgyz volume, see Radloff 1990. 18 On the apprenticeship, training and initiation of Ukrainian minstrels, for instance, see Kononenko 1998: 86–132; on the training of the Minangkabau singers of West Sumatra, see Phillips 1981: 10–17. 19 The Uzbek-Laqay are an Uzbek tribe found in Uzbekistan in the province of Surkhandaryā and in some districts of Tajikistan; see Karmysheva 1954. On the Uzbek ‘singers’ schools’, see Mirzaev 1979: 38–60; Imāmov et al. 1990: 235–38; Mirzaev 2008: 99–139; see also the dissertation by Walter Feldman 1980. 20 On the Goroghli cycle, see Chapter 3 (p. 56). 21 On this bakhshi, see also Levin 1996: 146–49. 22 On the related concept of yol ‘way’ in Turkmen oral epic tradition, see Żerańska-Kominek 1998; on Kyrgyz singer’s schools, see Kydyrbaeva 1996: 289–341. 23 Quoted from Innes 1974: 41. The kora is a plucked harp-like instrument. For a book-length study of griots and griottes, see Hale 1998; on their training, see pp. 172–92; see also Okpewho 1992: 20–41.

The Singer  47 4 The Mande text has bisimilai, from Arabic, literally ‘in the name of God’. 2 25 See Reichl 2000a: 136–40, from where the lines are quoted. Examples from Edige will be given in Chapters 4, 5 and 7. 26 On the historical background and development of the epic tale, see Schmitz 1996; for an edition and translation of a Karakalpak version of Edige, see Reichl 2007; on Sïpïra-jïraw, see ibid.: 106–11. 27 Schmaus 1958: 119–20; for a discussion of two somewhat similarly defined types of bylina singers, see Putilov 1997: 230–76. 28 On the Alpamish epics in general, see Zhirmunsky 1960; Chadwick & Zhirmunsky 1969: 292–96; Reichl 1992: 333–51. Fāzil’s version was first written down in 1922 and published in short extracts in 1923; the manuscript has, however, disappeared. The epic was written down again in 1928 and was critically edited, with a Russian translation, in Mirzaev & Abdurakhimov 1999. The manuscript is kept in the archive of the Folklore section of the Language and Literature Institute of the Uzbek Academy of Sciences. It has the shelf-mark 18, is written in Arabic (Chaghatay) script and comprises 947 pages. Abdunazar Pāyānov’s version, edited by Eshānqulov & Chāriev 2018, differs significantly from other Uzbek versions of the epic. On the Uzbek versions, see Mirzaev 1968; for a summary of this book, see Reichl 2001: 77–107. 29 On various Turkic versions of this romance, see Özarslan 2006, where, however, Fāzil’s version is not discussed. 30 Fāzil’s Jizzakh qozghālāni (The insurrection of Jizzakh) is edited in Zarif 1965: 59–106; it has been discussed together with other dastans of the Soviet era in Zhirmunsky & Zarifov 1947: 458–84, at 467–70. 31 Snouck Hurgronje 1906: II, 100–1; on this epic, see further Wieringa 1998. 32 1960: 101. Lord concludes his discussion of authorship by stressing that ‘a song has no “author” but a multiplicity of authors, each singing being a creation, each singing having its own single “author”’ (102). 33 Bogatyrev & Jakobson 1929: 901 and 906 trans.

3 Introducing Performance

But, while that work and all the funeral rites Were in performance, guards were held at all parts, days and nights, For fear of false surprise before they had imposed the crown To these solemnities. This quotation comes from George Chapman’s translation of the Iliad, from the last lines of the epic, where the funeral rites for Hector are described.1 Chapman’s translation of the complete Iliad was published in 1610 and 1611. His translations of the Homeric epics are little read today and mostly known only as the inspiration for John Keats’ famous poem ‘On First Looking into Chapman’s Homer’. Chapman talks of rites in performance. According to the Oxford English Dictionary (OED) this is the first occurrence of the word performance in the meaning intended in this book: ‘The action of performing a play, piece of music, ceremony, etc.’2 As the OED explains, performance has a number of meanings, ranging from the most basic ‘the doing of an action or operation’ to specific meanings such as the use of performance in linguistics, as in Noam Chomsky’s distinction between competence and performance. By the end of the twentieth century, performance had become a wide-ranging concept, applied in theatre studies, art criticism, anthropology, linguistics, and cultural studies. In fact, this concept is so comprehensive that it is difficult to keep a discussion of performance focused on specific problems and phenomena. The study of the performance of oral epic poetry is only a small segment of performance studies, but it is nevertheless a part of this larger network and has multiple connections to other strands of the subject. Richard Schechner, who has been instrumental in developing performance studies, takes a very broad view of the field, with close connections to the theatre, the social sciences, ethnography, and cultural studies. Theatre theorists and practitioners like Schechner himself, sociologists like Erving Goffman, cultural anthropologists like Victor Turner, and performance and cultural studies theorists like Barbara KirshenblattGimblett have shaped a discipline that includes the study of a vast array of phenomena, among which the acting of plays and the reciting of oral

Introducing Performance  49 epics occupy only a limited space.3 Different definitions and delimitations of the concept of ‘performance’ are given by different scholars. Schechner proposes a general view of the subject at the beginning of his introductory book, which is worth quoting: Performance must be construed as a ‘broad spectrum’ or ‘continuum’ of human actions ranging from ritual, play, sports, popular entertainments, the performing arts (theatre, dance, music), and everyday life performances to the enactment of social, professional, gender, race, and class roles, and on to healing (from shamanism to surgery), the media, and the internet. … The underlying notion is that any action that is framed, presented, high-lighted, or displayed is a performance. (2002: 2) The idea of a frame comes from social and cultural anthropology (Erving Goffman, Gregory Bateson) and denotes the characterization of an action as an action of a specific kind with the help of various ‘clues’ so that it can be understood and interpreted as the kind of action it is meant to be. Two men insult each other and are threatening to come to blows. The fact that they are standing on a stage frames the action and helps one to interpret it as a play. Other clues (or keys) are, however, necessary to disambiguate the action; the two men could, for instance, be workers having an argument over some repair job to be done on the stage. ‘Framing’ and ‘keying’ are also important concepts for defining and understanding the performance of verbal arts as Richard Bauman has underlined (see below). Among the innumerable cases of performance, it will probably seem more natural to class the performance of an epic singer with that of an actor than, for instance, with that of a businessman. Special skills are required to be active as an epic singer, as an actor, and also as a businessman, and their performance can be either successful or unsuccessful. The performance of businessmen is judged by the company for which they work; it is presumably measurable by the profit or loss it brings. The performance of an actor and an epic singer is judged by the audience. The performers will either be applauded or booed, depending on whether the audience likes or dislikes their performance. What distinguishes the performance of an epic singer, an actor, or a musician from that of a businessman is that it is done for and with an audience. Their performance is characterized by a dialogic situation, with the performers on one side and the audience on the other. Audience participation can, however, blur this distinction. Modern dramatists have explored ways of incorporating the audience into the performance, with actors distributed among the audience or the use of various stratagems to break the dramatic illusion. Similarly with music, audiences might be encouraged to sing or play along in folk music and jam sessions. In such cases performers and audience

50 Settings merge, are perhaps still distinguishable, unless the event is in fact communal and everybody present is meant to participate. There is a continuum ranging from an audience sitting back, as it were, to an audience actively participating to a degree that they have become co-performers.4 The view that musical and dramatic performance presupposes an audience calls for some further comments. Objectively, the Bach partita played by a violinist in her study and in a concert hall might be identical, as far as one can speak of identity in music. Nevertheless, there is a difference. In the first case she is simply playing, in the second she is performing. ‘Simply playing’ might mean practicing, but can also mean that she is enjoying playing the piece. ‘Performing’, on the other hand, might also mean that she carries her programme through, even if no one has turned up for the concert. Here, however, the border between playing and performing becomes so blurred that this point should not be pressed, unless one wants to become entangled in the old philosophical conundrum of whether sound is sound when nobody hears it. A further complication is introduced by the possibility of ‘virtual performance’, that is, acting before a camera to make a film or a video and playing in front of a microphone for a recording. Ranging the performance of oral epics alongside the performance of music or drama does not exclude looking at epic performance from other points of view, but it does give a preference to approaches that focus on verbal art. Different disciplines view the study of performance differently, and within each discipline various theoretical positions have been taken. Richard Bauman proposes a three-fold distinction between performance as practice, cultural performances, and the poetics of oral performance.5 This distinction has been taken up by other folklorists and anthropologists and was restated by F. J. Korom: The first draws on Marxist notions of praxis, life as situated, ordinary practice – a stone mason building a wall, for example …; the second emphasizes cultural display or enactment, when a community presents itself publicly in spectacular events such as the many forms of carnivals celebrated publicly throughout the world …; while the third focuses on verbal art or oral poetics …. (2013: 2) Although the interpretation of performance will have to incorporate elements of the first and second type of performance, I will here limit my observations to the third type.

The Ethnography of Communication Several disciplines have contributed to the development of the study of verbal performance, especially folklore, linguistics, anthropology, literary theory, and theatre studies. In the 1960s a methodology for the

Introducing Performance  51 description and interpretation of ‘communicative events’ was developed by both anthropological linguists and sociolinguists; this ‘ethnography of communication’ or ‘ethnography of speaking’ has proved to be a powerful tool for the analysis of the performance of oral epics. The various scholars working in the theoretical framework of the ethnography of communication have underlined the event-character of oral performance and stressed the importance of context for the understanding of the object performed. In this they followed the folklorists’ concern for context. These scholars have also proposed sets of descriptive terms by which the various components of a communicative event can be distinguished and analysed. Dell Hymes has been foremost in this and has suggested several lists, comprising between seven and sixteen elements. Muriel Saville-Troike has elaborated these into a grid for analysis with numerous examples from various speech communities; John Foley has argued for a synthesis of the oral-formulaic theory and the ethnography of communication, and a number of folklorists, linguists, and anthropologists have suggested further modifications and additions.6 In the following I will give no more than a thumb-nail sketch of the ethnography of communication; but rather than considering these descriptive and analytic issues in the abstract, I will discuss their relevance for the study of oral epics with an example. In doing so, I will consider the various aspects of a communicative event under three headings: the physical setting, the message, and the event as a whole.

The Setting My example is the performance of a Turkmen dastan in Törtkül, Uzbekistan. I will first give a short account of the wider context of this event. In September 1995, while teaching at the University of Nukus in Karakalpakstan (Uzbekistan), a Turkmen colleague invited me to Törtkül to record a Turkmen epic singer. Törtkül (in Russian maps Turtkul’) is a small town on the Amu Darya of about 40,000 inhabitants, c. 100 miles south-east of Nukus. From 1932 to 1939 Törtkül was the capital of what was then the Karakalpak ASSR (Autonomous Soviet Socialist Republic). In 1995, Törtkül was still a backwater town, to be reached from Nukus on a bumpy road coasting the Kyzyl Kum desert. The Lenin mosaic on a hillside, which I had had to admire in the early 1980s on my first trip to Törtkül, had meanwhile been replaced by an advertisement for the Uzbek telephone company; its traces were, however, still visible. The singer I was to record lived far outside Törtkül. It took several hours to reach his small farm on dirt tracks, and it took a long time to convince him that a recording would be a worthwhile undertaking for the popularization of Turkmen culture. The bakhshï, Gulum-bakhshï Ilmetop (1939–1999), spent most of the day, at least when the weather was mild, in a tree-house.

52 Settings He seemed to be loath to exchange his lofty abode for the house where his sister’s family lived, in a village on the outskirts of Törtkül. When Gulum-bakhshï was finally brought to his sister’s house, it took hours until the accompanying musician arrived and again hours until an important functionary of the local cotton kolkhoz turned up. The singer and his accompanist also needed some time to warm up. Finally, when all was set up, the recording could begin. There was an audience of about eight people, sometimes more, sometimes less: the functionary, neighbours, members of the household, men and women, also children (often coming and going), my colleague Gurban Bazarov, and myself.

Illustration 1  Gulum Ilmetov and spike-fiddle player (1995)

Looking at the participants of the Turkmen dastan performance, it will be noted that the dichotomy between singer on one side and audience on the other has to be qualified. As the performing side consisted not only of a singer, but also of an accompanying musician, so was the listening side composed of a number of different figures: the simple listener, the connoisseur, the ethnographer-recorder, the host, the local dignitary, and a kind of tamada. A tamada is someone who encourages toasts at festive gatherings and generally acts as master of ceremonies. In this case one member of the audience urged the singer on with his exclamations and interested questions, somewhat like a tamada. This person expressed what linguists call backchannels, that is, the various listener responses that keep a speaker going.7 While some of the listeners had witnessed many performances, also of this singer (the ‘connoisseurs’), for others, especially the younger members of the audience, this was an unfamiliar

Introducing Performance  53 experience. This particular event also had an unusual participant, that is, myself, the person manipulating a tape recorder and a video camera. The enumeration of the participants makes it clear that they are not only physical presences, but also defined by their various social roles and the specific functions they fulfil in the event. The tamada acted as the spokesman of the audience as a whole. It is customary for audiences to show their interest and enthusiasm by exclamations and remarks encouraging the singer. This interaction can shape the epic both quantitatively and qualitatively. Singers might expand or contract episodes depending on the interest of their hearers; they might embellish a passage or shorten it to a prose summary. V. Zhirmunsky and H. Zarifov remark about the performance of Uzbek epic singers: A good singer, a master of his art, preserves the ability to listen attentively and sensitively during the performance of a dastan to the reactions of the audience to his playing. Depending on the degree of interest and participation shown by the listeners, he enlarges or shortens the text of the poem. Even the choice of the plot and the more detailed elaboration of single episodes take their cue from the composition of the audience and its taste as it is known to the singer: among old people or elderly listeners he will sing differently than among young people, etc. (1947: 30 trans.) Singers who do not get enough applause and encouragement not only do not do their best, but might even break off their recital. The folklorist İlhan Bas¸göz described two performances of a Turkish romance (hikâye), one before a traditional and interested audience, the other to a group of teachers and local dignitaries, who showed little enthusiasm. Also the venue was different, the first being a coffee house in Poshof in eastern Anatolia, the other the teachers’ union hall. While the first performance was lively, interlarded with the singer’s comments and asides and enriched by a number of songs, the second lacked all of this.8 In a normal performance situation, the listeners are not merely passive recipients of the singer’s presentation, but react to it by evaluating the performer’s skills. The notion of evaluation has been stressed by Richard Bauman, whose definition of performance hinges on it: Fundamentally, performance as a mode of spoken verbal communication consists in the assumption of responsibility to an audience for a display of communicative competence. This competence rests on the knowledge and ability to speak in socially appropriate ways. Performance involves on the part of the performer an assumption of accountability to an audience for the way in which communication is carried out, above and beyond its referential content. From the

54 Settings point of view of the audience, the act of expression on the part of the performer is thus marked as subject to evaluation for the way it is done, for the relative skill and effectiveness of the performer’s display of competence. (Bauman 1977: 11) In our example of the Turkmen bakhshï’s performance, the audience included the ethnographer. The role of the person interviewing ‘native informants’ and recording speech and other events has been discussed by both folklorists and ethnologists. It is addressed in Kenneth Goldstein’s distinction between ‘natural’, ‘artificial’, and ‘induced natural’ settings.9 According to this classification, the recording of the Turkmen dastan was neither artificial (since there was a genuine native audience) nor natural (since there was an outsider in the person of the ethnographer), but induced natural, that is, the event happened at the instigation of an outsider, but then took its course as it would have done also in the absence of an outside observer. Of course, the presence of recording equipment acted in a way like stage lighting. Despite the presumed naturalness of the performance, there is no denying that the purpose of making a recording was clear to both performers and audience. Although the audience clearly enjoyed the performance and was eager to share their oral tradition with me, it is probably justified to see the event as a ‘folklore act’ in J. D. Niles’ definition, that is, ‘a folkloric performance (whether singing, dancing, or chair making) that is commissioned and recorded by outsiders for the primary benefit of their own textual communities’ (1993: 137). The most essential coordinates of the setting are the time and the locale of the event, that is, when and where it took place. The epic was performed at night. Traditionally oral epics are sung and recited through the night among the Turkmens as also among other Turkic peoples. How important the time of day or year for a particular rite, including storytelling, is in some societies was pointed out almost a hundred years ago by B. Malinowski in his study of the Trobriand islanders’ telling of stories and myths. He showed that generic distinctions among their narratives were tied to the period as well as the expected effect of storytelling (1926: 20–26). In a great number of traditions, oral epics are performed at night. This has partly to do with the fact that the day is reserved for work and the night for entertainment, but there are also other reasons. In some Islamic societies epics are (or were) listened to in the month of Ramadan not before iftar, the breaking of the fast after sundown. We have accounts of Bosnian, Turkish, and also Uzbek singers who told epics in the evenings of the month of fasting, sometimes specially hired by coffee-house owners for the whole month. The time when epics are performed might also be a holy time, especially when there is a connection to religious rites.10

Introducing Performance  55 While epic performances do occur in front of a crowd of listeners, there is also a more intimate type of performance in a domestic setting. In our example, everybody was gathered in the living room, sitting on the floor on carpets and cushions. The performers took centre stage, with the audience in a semi-circle in front of them. The tamada was sitting to the singer’s left, watching the singer, who in turn fixed his attention mostly on him. Generally, the most distinguished person in a gathering would be seated on the tör, the seat of honour. In a yurt this is the side opposite the entrance. In a traditional house, the tör is at the head of the table or, if there is no table, at the head of the dasturkhan, the tablecloth on the floor, on which food is laid. With an epic singer present, the tör is offered to him. Other persons will be placed according to rank and esteem. It is clear that in such a group, which is linked by a shared common experience, neither social position nor role is obliterated. Both participants and setting can be captured by a list of persons distributed in a circumscribed space at a particular moment in time; but this is obviously not enough. There is a dynamic relationship between performers and audience, as well as between the members of the audience, and there is a non-random, indeed a meaningful dimension to the location of the event in place and time. As I have mentioned, the performance of the epic did not begin until late, after all the guests had arrived and also the singer-narrator and the instrumentalist had warmed up. When Turkmen, Uzbek, and other Central Asian singers perform the epic, they generally begin with shorter songs. These songs might be historical songs, topical songs, ‘wisdom songs’ (series of maxims), and so on. In other words, a definite act sequence is observed.11 Furthermore, the performance of poetry is regularly, as in our example, placed in the wider frame of a meal and a festive gathering. The Message Message-oriented parameters relate to the genre, the content, the form, and the transmission channels.12 As discussed in Chapter 1, the genre term ‘epic’ has to be taken in a fairly broad sense, if non-Western oral traditions are to be included in our discussion. In the present case, it is best to use the native genre term dastan, which comprises the connotations evoked by the two terms ‘(heroic) epic’ and ‘romance’. The message content can be indicated, at least in outline, by a summary of the plot. The dastan that Gulum-bakhshï performed is called Kharman Däli in Turkmen. This is the name of the main protagonist (‘crazy or reckless Kharman’), a beautiful woman, who possesses great musical skill – and physical dexterity when it comes to wrestling. This dastan belongs to a cycle of dastans about Göroghlï, a cycle widely diffused in Central Asia and the Middle East, both among Turkic and non-Turkic peoples.

56 Settings As epics or dastans from the cycle of Göroghlï are repeatedly referred to or discussed in this book, I will here give some information about this cycle. The main hero is known under the name of Köroğlu and Koroğlu in Turkish and Azeri, under the name of Goroghli in Uzbek and Göroghlï in Turkmen and Uygur. Kör-oğlu means ‘the son of the blind man’, while Gör-oghlï means ‘the son of the grave’. In accordance with the name, the cycle has different beginnings. In the Turkish and Azeri versions of the cycle, the hero’s father is blinded by his cruel employer (the bey of Bolu, Shah Abbas and others), while in the Uzbek and other Central Asian versions the hero is born in the grave, from a mother already dead and buried. In the Turkish and Azeri branches of the cycle the hero becomes an outlaw and gathers a band of companions around him, with whom he performs various heroic and daring deeds. In the Central Asian versions of the cycle, Goroghli/Göroghlï recedes somewhat into the background, living in his fortress and sending his companions out on various tasks and military expeditions. An important part in these adventures is played by his adopted sons Awaz and Hasan and their sons, Rawshan and Nurali. The ‘western’ and ‘eastern’ version of the cycle is not always clearly distinguishable; ‘contamination’, to use a term from written transmission, frequently occurs. The most notable example of the cycle outside the group of speakers of Turkic languages is the Gurughli (or Gurguli) cycle of the Tajiks. Kharman Däli is also known among the Uzbeks and the Karakalpaks, with close resemblances to the Turkmen version.13 Kharman Däli has vowed to give her hand in marriage only to the man who surpasses her in musical skills and defeats her in a wrestling match. Any suitor who is defeated forfeits his life. Despite such forbidding conditions, suitors are keen to try their luck. As is Göroghlï. His wife Agha Yunus warns him and advises him to train first with Pir Ashïq Aydïn. Aydïn is both a pir, a holy man and spiritual mentor, and an ashïq, a folk singer. In the course of the narrative various encounters with the merciless beauty are described, showing her wit and her dexterity. Göroghlï, on the other hand, cuts a bad figure and has more of a comical than a heroic character. In the end, Kharman Däli marries instead of Göroghlï a worthier, and younger, pupil of Ashïq Aydïn. Ashïq Aydïn is considered the patron saint of musicians in northern Turkmenistan and Khorezm in north-western Uzbekistan. As V. N. Basilov has shown, Ashïq Aydïn corresponds to the figures of Baba Gambar (in southern Turkmenistan) and Qorqut Ata (among the Kazakhs), who are both patron saints of music and musicians (1970: 63). The various traditions connected to these figures overlap. In a number of legends they are represented as the creators of musical instruments, the dutar of the Turkmens and the qobïz of the Kazakhs. Basilov has argued that in these figures traces of pre-Islamic beliefs and ideas can be found and that they are closely connected to the shaman. One of the shamanic elements is the role the saints play in the acquisition of musical skills. Just as the

Introducing Performance  57 future shaman and epic singer are called in a vision or dream, the future musician is inspired by the saint’s spirit in the course of a supernatural experience. Other characterizations of the message have to do with the channels used for transmission, the linguistic and paralinguistic codes employed, and also the form of the message, including the different levels of linguistic and poetic structure as well as the musical and theatrical aspects of epic performance. The analysis of the patterning of oral poetry has been the concern of a group of folklorists and anthropological linguists working on the elaboration of an oral poetics, among them Dell Hymes, Dennis Tedlock, John D. Niles, and Thomas Dubois.14 The channels used in our example are, of course, oral (and aural). The message form is vocal, comprising speaking and singing. The dastan is composed in a mixture of verse and prose; the prose is narrated, the verse is sung to the accompaniment of the dutar (played by the bakhshï) and of a bowed fiddle, the gïjak. The linguistic codes are those defined by Standard Spoken Turkmen, with some dialect admixtures and, in the poetry, the employment of a poetic register. The musical codes are those of traditional Turkmen folk music, with the specifications due to the instruments used and to local varieties of the bakhshïs’ singing styles. On the paralinguistic level, the singernarrator’s gestures form an integral part of the performance. Since voice and gesture as well as music will be the topic of separate chapters, I will make no further comments here. As to the poetic structure of oral epic poetry, I will come back to some aspects relevant for an understanding of epic performance in a later chapter (see Chapter 10). Gulum-bakhshï, who told me that his repertoire consisted of six dastans, was a lively narrator, who managed to tell the prose story in a highly dramatic and absorbing fashion, with numerous dialogues and exclamations, and sang the verse passages in a varied and competent style, to the accompaniment of dutar and gïjak.15 The Event While it is helpful to isolate the components of a communicative event, it must be borne in mind that all components are connected and held together by several basic conventions. One is what Richard Bauman (following Goffman) has called framing and John Foley performance arena. Bauman emphasizes the need for establishing a frame that fixes the verbal behaviour in question for the audience and makes it clear that the communicative event is a performance. The listener/spectator is asked to interpret what is heard and seen according to the conventions and rules of the type of communicative event it is: ‘performance sets up, or represents, an interpretative frame within which the messages being communicated are to be understood’.16 From this follows what Bauman termed keying. In order to make sure that the communication is understood as being a

58 Settings communication of a specific type, the performance frame is keyed ‘in culturally conventionalized and culture-specific ways’. Bauman lists various means of keying a performance, a list taken up again in more elaborated form in a later article.17 Some of these devices overlap with what Hymes has enumerated under the headings ‘codes’ and ‘forms’. They concern the patterning of the message. Bauman and Braid also list ‘special settings conventionally associated with performance (the nightclub stage, the lectern, the liars’ bench)’.18 These settings affect the attitude towards the message, which will presumably be interpreted differently whether it is delivered from the lectern or from the liars’ bench. The attitude of the audience is correlated to the mood of the event, or what Hymes called ‘key’: Key is introduced to provide for the tone, manner, or spirit in which the act is done. … Acts otherwise the same as regards setting, participants, message form, and the like may differ in key, as, e.g., between mock: serious or perfunctory: painstaking. (1972: 62) While with the figure of Ashïq Aydïn an undercurrent of religious beliefs and legendary lore is present in the dastan, the narrative as a whole is light-hearted and often humorous. Although most of the branches of the cycle stress the heroic adventures of Göroghlï and his companions, in this dastan Göroghlï’s clumsiness in the wrestling match provokes laughter. The key of the event is jocular, at least in parts, rather than persistently serious; it is light-hearted entertainment as far as the story goes, with, however, the added enjoyment of music. The key or mood of an event is intimately connected to its purpose and meaning. This opens up a wide field of questions and speculations about the function of oral poetry in context. To stay with our example, the event can be characterized as entertainment, in the non-trivial sense that the audience is invited into the fictional world of the story and called to appreciate the narrative, poetic and musical art of the primary performer and his accompanying musician. By the same token, the listeners and performers assert their common cultural heritage: it is their poetry, their music, their hero and heroine, their language, their tradition. While presumably all performances of oral epics in a traditional context will be based on shared cultural suppositions and knowledge, other aspects will vary from case to case. Among a number of Siberian peoples – Turkic, Mongolian, and others – epics were performed during the hunting season. By singing and telling tales the tutelary spirits of the animals were meant to be soothed. The Khakas, a Turkic-speaking ethnic group of the Altay, used to address the lord-spirit of the mountain with the words: ‘Listen and give us more animals, and we will tell you more tales.’19 The performer of the Rajasthani oral epic of Pābūjī is not only the teller of the

Introducing Performance  59 tale, but also the priest of the cult devoted to the divine hero Pābūjī (see Chapter 5). Here again an epic performance acquires different nuances of meaning for the people present. When it comes to interpreting the mood, meaning, and function of an epic performance as a communicative event, every individual case will have to be analysed within its own social and cultural context (see also Chapter 10).

Textualization The conceptual framework for the analysis of a communicative event has a mirror image. This is constituted by the various conventions and symbols used for the textualization of the event, in particular its central element, the message. Once the event is past, all that remains is what is remembered by the participants. Only if the event was recorded is some physical evidence left. This is in most cases a written text, either written down from an audio and/or video recording or from the performer’s dictation. The latter was the usual procedure in the time when recording machines were either unavailable or severely restricted in their capacity. Dictation, of course, lies outside the epic performance. It is a communicative event of its own, comparable to the act of translating the message and of explaining what happened or would usually happen in a performance.20 These texts of oral narratives and poetry, especially when they were written down without the help of recording machines, are often fairly colourless replicas of the performed story, poem or epic. Exclamations and interjections, slips of the tongue and unfinished sentences, repetitions and asides, dialect traits and vulgarisms, as well as other features of the live text were in many cases edited out. Even such venerable giants among the early collectors of verbal folklore as the Grimm Brothers, who insisted on authenticity, have nevertheless incurred later criticism. Modern folklorists adhere in their editions closer to the texts they recorded, but even here the issue of fidelity, for instance to dialect traits, has led to discussion.21 The principles for the edition of Uzbek dastans during the Soviet era laid down by Hādi Zarif are an example of fairly radical editorial intervention. Many Uzbek singers spoke and speak a Kiptchak-Uzbek dialect, which shows many affinities to Kazakh and Karakalpak. Fearing that the ‘Uzbekhood’ of the epics recorded from these singers might be doubted, these dialect features were removed in editions. Even more wide-ranging were cuts and changes with respect to pornography, religious allusions, the denigration of other peoples of the Soviet Union or the praise of drunkenness. Also superfluous repetitions and expressions referring to the performance event itself were to be excised from the edited text. Among the latter Hādi Zarif mentions several phrases, such as ‘Now the food has come. First the meal, then the words!’22 For the first Uzbek dastan that I

60 Settings translated I took an edited text. When I later consulted the manuscript of the dastan in the Folklore Archives of the Literature Institute of the Uzbek Academy of Sciences, I realized that in the edition the original text had been substantially abbreviated and also changed in a number of cases. To give only one example. In the first verse passage, the hero, Rawshan, addresses his foster father Goroghli with the words: ‘You are my Mecca, you are my Medina, Lord of the people!’ (Makkamsan, Madinam – yurtning egasi). This is the manuscript text. In the 1956 edition we read: ‘My dear father, you are the lord of the people!’ (Ātajān, sendirsan yurtning egasi).23 Allusions to Islam were not tolerated at that time. Modern folklorists, on the other hand, advocate an editorial approach that is strictly opposed to any tampering with the words, said or sung. The goal of the textualized communicative event is to be as comprehensive as possible. Elizabeth Fine (1984) has discussed both the theory behind this goal and the possibilities of reaching it in a printed text. Dennis Tedlock (1983) has focused on paralinguistic features of the live narration and their rendering in translation, while Dell Hymes (1994) has sought to elicit such features from a careful analysis of the structural and linguistic properties of texts recorded in writing. If there are no additional audio and video recordings supplementing the printed text, linguistic and paralinguistic features have to be expressed by a limited repertoire of symbols. Theoretically the phonetic shape of a stretch of speech can be transcribed fairly accurately by using the International Phonetic Alphabet. Dialect features, for instance, can be captured to the satisfaction of the linguist. What such a narrow phonetic transcription doesn’t show, however, is loudness (amplitude), pitch (frequency), timbre or tone colour, voice quality (smooth, hoarse, raucous, nasal etc.), and speed. Some of these features can be indicated by symbols, but as a rule only approximately. The speed of an utterance can, of course, be measured, and if segmentable stretches of speech were spoken (or chanted or sung) at a steady speed, this could be symbolized by musical terms (presto, allegro etc.) and more exactly by noting the time their production took. Speed, however, can vary continuously, which puts limits on the exactitude of measurements. Other features related to speed are easier to represent in a printed text. These are sound lengthenings and the pauses performers make. A protracted sound, generally a vowel, can be indicated by a repetition of the same vowel or letter, as, for example, in Tedlock’s translation of the Zuni story ‘The Girl and the Protector’ with Grandma written with twenty-two final a’s (1983: 104), or by the use of hyphens after a syllable as in Fine’s textualized African American performance of the story of ‘Stagolee’.24 As to pauses, they can be used to segment speech into smaller units, with differentiations according to the length of the pause. Sometimes extra-textual markers help to segment the text. In the recording of a Karakalpak epic I noticed that the most clearly marked pauses were the occasions when the singer had a sip of tea. I included

Introducing Performance  61 a symbol for these sips in my edition (2007), not because I thought that drinking tea was relevant for the understanding of the text, but rather because it showed clearly where the singer felt he could interrupt the performance, in other words where a break existed in narrative structure in the mind of the singer. Loudness and emphasis (intensity) have been indicated in various ways. Tedlock uses capitals for loud syllables, words, or larger syntactic units; similarly, Fine uses capitals for strongly stressed and emphasized words. Pitch in speaking depends on the voice of the speaker with gender-specific typical registers. Sometimes a speaker would use a higher register, say for something tiny, sometimes a lower register, for instance for the words of a bear or a giant. In addition, pitch can rise and fall legato (i.e. in a continuous line) and staccato (i.e. with one or more breaks in between). All of this can be indicated; the rising and falling by rising or descending letters, higher sounds by smaller, deeper by larger letters. Fine distinguishes also between four falsettos, marked by differently shaped arrows pointing upward. Furthermore, she characterizes the performance of ‘Stagolee’ with marginal comments like ‘emphatic’, ‘confidential, softer’, ‘1/2 step higher, crescendo’, ‘faster’, and ‘loud, almost sobbing’. Tone colour and voice quality can be described only impressionistically. Tedlock adds comments like the following to his translation: ‘with a sad tone throughout’, ‘low and gravelly’, ‘single puff’, ‘double puff’, and ‘sighing’. Fine marks syllables and larger units with a curly line, signifying ‘a rasp, or harsh, guttural, grating quality’. Other performance features can also be included in the text. Fine records various stances and movements of the performer by a number of short qualifications which are explained more fully in the list of kinesic features. The qualification ‘strut’, for instance, stands for: ‘Leans back throwing his chest and head up; swings arms in a slow exaggerated rhythmic way; bends knees as high heels hit the ground, giving a slight spring to the walk.’ As one can see, textualization blends into commentary. The comparatively full descriptions Tedlock and Fine offer are possible for short texts. A similar procedure for a long epic would become very cumbersome. A further element is that epics are generally sung; in the case of prosimetric epics like the Uzbek dastan, only the verse passages are sung. This means that in addition to all the features discussed so far the notation of the music will have to be added. This can, of course, be done. Stephen Erdely has transcribed the music of three South Slavic epics from Bosnia (1995) and Daniel Prior the melodies of a Kyrgyz epic preserved on phonograph (2006), to give two examples. On the whole, however, music is in text editions relegated to an appendix at best, with a general characterization of the music and perhaps some sample transcriptions. The interaction between performers and audience can also be included in a text. Tedlock’s Zuni story, for instance, brings right at the beginning, after the narrator’s introductory formula ‘NOW WE TAKE IT UP’,

62 Settings the audience’s reaction ‘Ye−−−−s, indeed’. The tamada’s interventions in the Turkmen example above could also be recorded in an edition. Such persons reinforce the dialogic element and are in some traditions a fixed element among the participants. When the Rajasthani epic of Pābūjī is performed, there is in addition to the male singer-narrator (bhopo or bhopa) and his female co-performer, usually his wife (bhopī), a hunkariyya, literally ‘yes-sayer’. As Elizabeth Wickett explains, the appointed ‘spectator-cum-respondent’ known as the hunkariyya speaks out during the performance to ‘converse’ with the bhopa and bhopi, to praise their performance and to comment on events as they unfold. The participation of this interlocutor, as other scholars have observed, is an essential component of the performance event.25 An even more striking case is the performance of the Central African mvet. In Herbert Pepper’s edition of the mvet performed by Zwè Nguéma in Anguia, Gabon, in 1960, we have in addition to the singer-narrator (récitant) two ‘assistants’ (premier assistant, deuxième assistant) and a group of people designated a choir (chœur). These interlocutors and coperformers become particularly active in what the editor terms ‘interludes’. In the fifth interlude, for instance, the singer and his assistants talk about the art of performing the mvet, the assistants praise Zwè Nguéma and then the choir joins in, responding to the singer’s self-praise (‘I sing like the hyrax’, ‘I will die for the love of the xylophone’) with a kind of vocalization (‘O o o koleyo ee koleyo’). Pierre Alexander points out that audience participation is generally high; they answer the artist’s calls for attention, join in the singing, and burst out in awed or mirthful interjections at the most moving or humorous passages. They are expected to ask questions, and make comments at the end, the mbomo mvet [singer of the mvet] answering them in metaphorical or enigmatic fashion.26 Also in European folktale traditions audience reaction is of importance. On the basis of recordings of Slovene folktale performances, Barbara Ivančič Kutin (2007), for instance, has suggested differentiating between no less than six roles of audience respondents: motivator, assistant, inquirer, yea-sayer/nay-sayer, complementor, and commentator. While much of the audience reaction can be described or quoted in a textualized performance, it is an aspect of a live event that shows perhaps best the limits of textualization. A many-voiced reaction is difficult to capture in writing, especially as global descriptions like ‘laughter’ or ‘murmur of approval’ and so on are inappropriate to record differing and perhaps contradictory reactions. Non-verbal reactions pose even more of a problem. In a powerful scene in a documentary about the manaschï

Introducing Performance  63 Sayaqbay Qaralaev, the singer is shown reciting the epic in front of a large audience in an open field. When it begins to rain the singer continues and the listeners are so enthralled by the story that they don’t seem to notice the change of weather.27 The effect of the recital on the listeners is written on their faces, but it is not expressed in sound. ***** It is possible to preserve a great number of performance aspects in meticulous and detailed textualization. With the help of audio and video recordings much of the dynamics of oral performance can be captured, and additional commentaries and analyses by participants and observers can fill some of the lacunae in our documentation. Even the most comprehensive coverage, however, will not be able to bridge the gap between the communicative event and the medially and textually reconstructed event. While it has to be admitted that reconstruction is deficient and incomplete, it is nevertheless necessary for a fuller understanding. The oral epic is a story, no doubt, but it is also much more. It is a performance, with all the facets this term entails. Analytic concepts like those elaborated in the ethnography of communication help us to see the oral epic as performed, rather than simply readable, poetry and narrative, as situated in a culturally and sociologically defined context and realized in a dynamic process of giving voice and listening, of interacting and responding. The next chapters will be devoted to performance proper: to the varieties of giving voice and to various aspects of the singer’s bodily presence as a narratorsinger, gesticulator, musician, and figure invested with special powers.

Notes 1 Quoted from the nineteenth-century edition Chapman 1884: 320. 2 OED, s.v. performance, 3.a. The OED distinguishes the action of performing (3.a) from ‘3.c. An instance of performing a play, piece of music, etc., in front of an audience; an occasion on which such a work is presented; a public appearance by a performing artist or artists of any kind. Also: an individual performer’s or group’s rendering or interpretation of a work, part, role, etc.’ 3 See Schechner’s wide-ranging introductory book (2002); see also the survey by Elizabeth Bell (2008), both with ample references. 4 See Pelias & VanOosting 1987 and their distinction between an audience as receiver, respondent, co-producer and producer; compare Bell 2008: 47–50. 5 See Bauman 1986; see also his influential study of verbal art as performance (1977). 6 See Hufford 1995; Saville-Troike 2002; Hymes 1964: 13, and 1972: 59–65; Foley 1992 and 1995: 1–18. 7 This person was an Uzbek. Törtkül is part of Karakalpakstan, but the population of the town is mostly Uzbek; Turkmens are a small minority. The word tamada is a Russian loanword in Turkmen and Uzbek; it is of Georgian origin. On figures with similar functions in other epic traditions, see below the hunkariyya in the performance of the Pābūjī epic and the assistants in the

64 Settings mvet performance; see also the assistants in the performance of Nenets epics (Chapter 9). On the notion of backchannels, see Tolins & Fox Tree 2014. 8 See Bas¸göz 1975; for other examples from Turkic oral epics, see Reichl 1992: 113–17. 9 Goldstein 1964; see also Finnegan 1992: 75–83. 10 On epic performance during the month of fasting, see Lord 1960: 15; Reichl 1992: 93–97. 11 On the notion of act sequence, see Saville-Troike 2002: 122–23; on the act sequence in the performance of Turkic epics, see Reichl 2000a: 40–43. 12 For the following see also the exposition and discussion of Hymes’ components in Saville-Troike 2002 and 2004 and in Bauman & Braid 1998. 13 On the Köroğlu/Göroghli cycle, see Chadwick & Zhirmunsky 1969: 300–4; Karryev 1968; Reichl 1992: 151–60, 318–33. For an English translation of the Turkmen dastan Kharman Däli, see Żerańska-Kominek & Lebeuf 1997; the original text is found in Karryev 1983: 322–68. 14 See Hymes 1981, 2003; Tedlock 1983; Niles 1999; Dubois 2012. 15 The six dastans are: Yusup and Akhmet, Khüyrlukga and Khemra, Övez getiren (The adoption of Övez), Kharman Däli, Bezirgen, Shasenem and Garïp. Of these, Övez getiren, Kharman Däli and Bezirgen are branches of the Göroghlï cycle; Khüyrlukga & Khemra and Shasenem & Garïp are love and adventure romances (on the latter, see Chapter 10). Yusup and Akhmet is a religiously inspired warlike adventure story; an edition of an Uzbek text with a German translation was published in 1911 by H. Vambéry. 16 Bauman 1977: 9; compare Foley’s discussion of ‘performance arena’, 1995: 7–11. 17 Bauman & Braid 1998: 110–11; quotation from Bauman 1977: 16. 18 Bauman & Braid 1998: 111; the expression liar’s bench denotes in New England ‘the bench in front of a country store where men gathered in good weather to swap news and stories and do a bit of trading’ (Hendrickson 2000: 259). 19 Alekseev 1980: 260. See Reichl 1992: 93–100. 20 Taking dictation and the Homeric poems as his point of departure, Jonathan Ready has presented a detailed discussion of editorial and textualization processes with regard to oral epics (2015); for a comprehensive study of the impact of the textualization of oral epics on our understanding of the Homeric epics as dictated texts, see Jensen 2011. Various approaches to the textualization of oral epics are collected in Honko 2000a. Honko has also discussed problems and strategies for the textualization of the Tamil Siri epic in great detail (Honko 1998). 21 On the Grimm Brothers, see Rölleke 1983; on dialect traits, see Preston 1982 vs. Fine 1983. 22 Zarif 1978: 132. On this editorial practice, see also Reichl 2013. These rules were, however, relaxed in the 1970s, but finally abandoned only at the end of the twentieth century. 23 See further Reichl 1985a: 7–12. 24 Fine 1984: 166–95; on the symbolic representation of paralinguistic and kinesic features, see pp. 182–84; the text itself is found on pp. 184–95. 25 Wickett 2010: 16. She adds that ‘Ann Grodzins Gold observed that the hunkar/hunkariyya is not a spontaneous, casual or extraneous adjunct but “a formalised element of performance-audience interaction” (1993: 21).’ 26 Alexandre 1974: 6. The fifth interlude in the mvet edited by Pepper 1972 is found on pp. 188/89–92/93. Compare also Seydou’s comments on the behaviour of the audience in Chapter 1 (p. 17). 27 In the documentary Manaschï by Bolot Shamshiev of 1965. See https://www.

Part II


4 Voice

In recitante sonent tres linguae: prima sit oris, Altera rhetorici vultus, et tertia gestus. In reciting aloud, let three tongues speak: let the first be that of the mouth, the second that of the speaker’s countenance, and the third that of gesture. (Geoffrey of Vinsauf, Poetria Nova)1 In this and the next chapter the subject will be the ‘three tongues’ which Geoffrey of Vinsauf mentions in his ‘Art of Poetry’ (c. 1210): voice, facial expression, and gesture. With their voice, their facial expressions, and their gestures, singers and narrators assert their bodily presence. As listeners we hear the words of the tale in and through the performers’ voice and as viewers we see their bodies, their faces, their movements. Voice has pride of place in this triad; a singer may not be visible, but must be heard. The importance of voice in the live performance of poetry and narrative can hardly be overstated. The voicing of words can take on a multiplicity of acoustic realizations, from whispering to shouting, from speaking to singing, from bel canto to recitative. In this chapter I will explore some of the varieties of voice production in the performance of oral epics, the way sound is produced by the human voice when the words of the epics are recited and sung. ‘Oh (Goddess), I have need of four things: beat, throat, voice, and wisdom’, exclaims the singer of the Indian epic of Ḍholā at the beginning of his performance. This epic is performed in northern India and consists of several narrative strands, one of which is the love-story of Ḍholā and Maru.2 In addition to wisdom the singer asks for three gifts, all of which have to do with sound or sound production: beat, throat, and voice. The singer’s native word for ‘voice’ is svar. This word goes back to Sanskrit, where, apart from ‘voice’, it also means ‘sound’, ‘noise’, ‘tone’, ‘accent’, ‘musical note’, and ‘vowel’. This array of meanings embraces the whole gamut of the human voice, from the discordant sound of noise to the concordant sound of music. The word carries

68 Performance a rich semantic legacy; it is already found in the Vedas and plays an important role in Indian music theory. Of an equally venerable ancestry is also the English word voice. It is a medieval borrowing from French and ultimately goes back to Latin vox. Latin vox is etymologically related to Sanskrit vākas ‘speech’, ‘word’, and also to Greek épos, ‘word’. Here the range of meanings is closer to narrating than to singing. This is highlighted by the Greek word épos, which not only means ‘word’, but also ‘speech’, ‘tale’, and ‘epic’. In modern English voice has a multiplicity of senses, all stemming from its basic meaning, the sound produced by the human speech organs. We say: ‘he has a high voice’, ‘she speaks in a pleasant voice’, or ‘he has lost his voice’. ‘Voice’ can also refer to musical sound and the capacity to sing. A piece of music can be set for ‘voice and piano’ or a choral piece might be composed for ‘four or more voices’.3 The human capacity to utter sounds lies at the basis of human language. Could humans have developed a symbolic system such as natural language without the ability to utter sounds? It is unlikely, even if there are systems of communication that are not dependent on sound. While our speech organs might not have been the cause of language development, they are generally considered essential in having given it the direction it took. While anthropologists, linguists, and cognitive scientists are divided on many issues related to the origin and development of language in humans, they concur in according a central place in the genesis of language to our ability to produce sounds with our speech organs. Voice is both the primary vehicle and main catalyst of language. This also means that spoken language is the primary form of language, while other kinds of manipulating symbols, be it by using writing in its various forms or by employing gestures and bodily movements as in the different sign languages invented, are derived from speaking and hence secondary. Ferdinand de Saussure, generally considered the father of modern linguistics, was most emphatic in stressing the primary nature of spoken language as against other forms of language, claiming that ‘the linguistic object is not both the written and the spoken forms of words; the spoken forms alone constitute the object’ (1959: 23–24). The centrality of the human voice in the production of linguistic utterances is not put into question by the fact that substitutes for the human voice exist, as in the communication systems used with and among deaf and mute persons.

Speaking When we look at the actual performances of the singers of Ḍholā as transcribed and commented upon by Susan Wadley, we find that two basic ways of ‘voicing’ the words are employed: speaking and singing. The epic of Ḍholā is like the Uzbek dastans composed in a mixture of verse and prose, where the verse portions are sung and the prose portions

Voice  69 are spoken. For speaking, the Hindi word vārtā is used, glossed as ‘dialogue’ and ‘talk’ in the dictionaries and explained by Wadley as ‘narrative prose in everyday storytelling style’.4 Speaking and singing comprise a number of options. Speaking can mean uttering the words as they would be uttered in everyday speech, as for instance in face-to-face communication in an acoustic environment that allows for average loudness and clarity of enunciation. Speaking can also mean declaiming the words, as for instance the words of a prayer by a preacher or a declaration by a politician. The OED captures this sense in the gloss ‘to speak aloud with studied rhetorical force and expression’ (s.v. declaim v., 1.a). Similarly, singing can mean the articulation of the words in a melodic line, but it can also mean a singing style modelled on the intonation of spoken language. These distinctions will be the subject of Chapter 6. In this chapter I will focus on the singers’ use and shaping of their voice in epic performance. I will begin with speaking. In prosimetric epics where there is a clear distinction between verse and prose – such as in the North Indian epic of Ḍholā and in the Uzbeks dastans – speaking is generally associated with the performance of prose, while singing with that of verse. In these traditions verse can be defined by metre, and prose consequently by the lack of metre. Metre is culturedependent and the borders between verse and prose cannot always be easily established. To recognize a text in iambic pentameters as verse is straightforward, but to classify a poem in free verse as verse on purely metrical grounds is problematic. A similar difficulty is encountered in the texts of African epics like the mvet which are not composed on metrical principles derived from language-internal constraints. When talking about verse and prose, it is therefore easiest and clearest to refer to a specific poetic tradition. In Karakalpak, the language of my example, a verse line is defined (in folk poetry) as a sequence of words comprising a set number of syllables. Individual verse lines are linked together by end-rhyme. Some irregularity in the number of syllables and the rhyming patterns is allowed. This definition will sound familiar to many readers. Karakalpak folk and traditional poetry is clearly based on a syllable-counting metre (like that of the Romance languages) and employs common verse-binding techniques such as end-rhyme. It is tempting to define prose then simply as nonverse. Although this works, on the whole, for Karakalpak narratives composed in a mixture of verse and prose, we find passages that are highly patterned and often metrically definable as verse, which are, however, not sung. Such passages are, for instance, found at the beginning of an epic and consist of often parallelistic and formulaic lines that describe the setting of time and place of the action that follows. As singing can be considered a performance marker of verse, this means that there is an area of blurred borders when verse is treated like prose. I will come back to metre and also the relationship between metre and music in Chapter 6.

70 Performance Keeping some of the complications of differentiating verse from prose in Karakalpak oral epics in mind, I will now discuss a short prose extract from an epic I have recorded as an illustration of the ‘speaking mode’ of performance. The extract comes from the Karakalpak epic of Edige, recorded 17 September 1993, in Shomanay (Karakalpakstan) from the singer (jïraw) Jumabay Bazarov (1927–2006). The majority of the Karakalpaks, a Turkic-speaking people, live in Karakalpakstan (Uzbekistan), on the shores of the ever-shrinking Aral Sea and on the lower reaches of the Amu Darya; the population of Karakalpakstan is an estimated 1.5 million, of whom about a third are ethnic Karakalpaks. The epic of Edige treats of the emir Edige of the Golden Horde, his opposition to Khan Tokhtamysh, and the struggle between Tokhtamysh and Timur (d. 1405) (see Chapter 2, p. 38). The extract is located in the first part of the epic. Young Edige attracts the attention of Khan Tokhtamysh by his wise decisions in legal disputes, for example when two women quarrel about a new-born boy. Both claim that they are the child’s mother:5 Sol waqtïnda Edige turïp ayttï: ‘Hä, qaysïng tuwdïng?’ dep qatïnlardan soradï. ‘Men tuwdïm’, dedi. ‘Sen ne tuwdïng?’ dep ülkeninen soradï. ‘Men tuwdïm’, dedi. ‘Haw, ekewingning bir ul ma tuwïp jürgening?’ ‘Awa!’ ‘Mïnaw jaladan jabïsïp turïptï’, dedi. ‘Onda balanï sheshindir!’ dedi. ‘Otïrghïz ortagha!’ dedi. Qïlïshtï qolïna aldï: ‘Bir qolï, bir qulaghï, bir közi, bir ayaghïn qaq ortasïnan ayïraman da beremen’, dedi. ‘Äne usïnï alasang’, dedi. ‘Sen häm ïrza, quday häm ïrza. Bola ma?’ dedi. ‘Boladï’, dedi. Balanï ortagha otïrghïzïp jiberip, tïrday yalangash etip, qïlïshtï qolgha alïp, Edige astïngghï läbin tishlep, hä, urmagha qayïm bolghanda, tuwghan qatïn: ‘Meni de qosïp shawïp jiber!’ dep jïghïldï tap balanïng üstine. Tuwmaghan qatïn: ‘Payïmdï ber!’ dep keyin bäsip turghan qusaydï-aw. [•] Shul waqlarïnda: ‘Tuwghan qatïn mïna qatïn, tuwmaghan mïna qatïn, bughan toghïz tayaq’, dep, yawïrïnïna toqqïz qamshï urïp, ­tuwghan qatïngha balanï qosïp qaytarïp jibere berdi gho. Then Edige got up and spoke: ‘Hä, which one of you gave birth to the child?’ he asked the women. ‘I gave birth to him’, said the one. ‘And what did you give birth to?’ he asked the older woman. ‘I gave birth to him’, she said. ‘Haw, so both of you gave birth to a son?’ ‘Yes!’ ‘This one is speaking a falsehood’, they both said. ‘Now, undress the child!’ he said. ‘Lay him in the middle!’ he said. He took a sword in his hand and said: ‘I will divide one arm, one ear, one eye, one leg exactly in the middle and give it to you and you take it. You will be satisfied and God will also be satisfied. Agreed?’ So he said. ‘Agreed’, she [the older woman] said. He had the child laid in the middle, undressed him completely, and took the sword into his hand; as Edige bit his lower lip and got ready to strike, hä, the woman who

Voice  71 had given birth to the child cried: ‘Strike me instead!’ and threw herself over the child. The woman who had not given birth to the child said: ‘Give me my share!’ and stepped back. [•] Then Edige said: ‘The woman who gave birth to the child is this one; the one who did not give birth to the child is that one. She deserves nine lashes.’ He hit her on the neck with the whip nine times and gave the child to the woman who had given birth to the child, and sent them back. This, of course, is the Solomonic judgment from 1 Kings 3. Solomon’s ruse is found in a great number of folktales, also in one of the birth stories of the Buddha and in a Chinese play, imitated by Brecht in his Caucasian Chalk Circle.6 The transcription of the singer’s words reflects the pronunciation of Karakalpak adequately. It is quite close to the Latin script used for Karakalpak since the fall of the Soviet Union. Karakalpak was written in Cyrillic letters from 1940 to 1996, when the script was Romanized; it has since been slightly adapted. The Cyrillic script is, however, still widely employed. A phonetic transcription of this passage with the symbols of the International Phonetic Association (IPA) would only slightly differ from the text above since Karakalpak orthography, unlike that of English, is on the whole phonological. Most deviations are modifications inferable from the rules of Karakalpak phonology and morphophonemics. In my edition I have left dialect traits and also interjections, expletives, and vocalizing particles. The only paralinguistic features I have marked are (1) emphatically lengthened vowels, (2) chords or short melodic phrases played on the accompanying instrument during the prose recitation, symbolized by [♪] and [♫], and (3) pauses in the recital when the singer took a sip of tea, marked by [•].7 Notice how in the episode of the Solomonic judgment, the singer pauses with a sip of tea at the climax of the story, just before Edige announces the verdict. The narrative pace of this passage is fairly swift; the singer recited it in one minute and fifteen seconds. The tea-sipping pause lasted no more than four seconds. Apart from the ‘tea break’, the delivery pace slows somewhat down at two points only. First, when Edige announces his procedure: ‘I will divide one arm, one ear, one eye, one leg exactly in the middle.’ Here, the word for ‘one’, bir, is stressed, and the four parts of the body enumerated are clearly impressed on the hearers. The second somewhat slower passage is the final verdict: ‘The woman who gave birth to the child is this one; the one who did not give birth to the child is that one. She deserves nine lashes.’ Note that the identification of the true mother rests on deixis (mïna qatïn, ‘this woman here’). The singer does not point himself. When he narrates a similar story, immediately preceding this one, he does, however, use his hands to mimic the movement of scales (which play a role in the story) (see Chapter 5).

72 Performance The narrative is particularly lively on account of the numerous utterances in direct speech. Edige’s words are spoken with more force than the women’s words. Only when the true mother cries ‘Strike me instead!’ are her words as loud as Edige’s. The interjections hä and haw come at three points in this short passage; they are always uttered at a slightly higher pitch. Also Awa! (Yes!) is said at a higher pitch. Otherwise the intonation of the various clause types (statements, questions, exclamations) accords with the intonation patterns of spoken Karakalpak. The transcription of this prose passage with my additional comments provides some idea of what this text sounds (or rather sounded) like, at least for a reader with some experience of spoken Karakalpak. It hardly says anything, however, about the unique timbre of Jumabay’s voice and the ways he modulated it in narration. Analysis is possible. Linguists have proposed various methods of segmenting and characterizing voice features, which go a long way towards specifying voice quality.8 As discussed in Chapter 3, some of these features can be introduced by appropriate symbols into textualizations of oral verbal art. Such somewhat fuller textualizations, as folklorists and students of ethnopoetics have urged, are necessary to make us aware of at least some of the richness of vocal performance.9 It has to be admitted, however, that without an audio and/or video recording we will still remain ignorant of the singer’s voice and voice quality. With a sound recording acoustical information can be extracted with the help of spectrograms and the flow of speech can be analysed in precise acoustical terms. This is for the specialist. For the reader of a textualized oral epic a sound recording will be a welcome means of filling the gap between transcription – however detailed – and the auditive impression of a listener (see Audio/Video Ex. 1). A further point that emerges when watching the video recording of this passage is the singer’s gestures. Facial gestures, of course, always accompany speech. But there is one gesture that gives a comment on the tale. When the singer takes his cup of tea into his hands, he quotes the false mother as saying: ‘Give me my share!’ At this stage, the singer shakes his head and smiles, as if to say (one might interpret): ‘Silly woman! She will soon get her share!’ I will come back to gestures in Chapter 5. Here the comment is gestural, but in many cases the singers’ comments are verbal (see Chapter 9). Jumabay’s narrative and elocutionary style in the prose passages is lively and expressive, but stays within the limits of his ‘everyday voice’. Occasionally, however, passages are recited in a declamatory fashion, that is, in a somewhat monotonous style, which lacks the prosodic variety of ordinary speech. Some of these passages recur and are almost rattled off like well-worn formulas in a religious ritual. One of these passages occurs at the beginning of the prose section from which the example above is taken. It is a list of officials at Khan Tokhtamysh’s court and consists of twelve verse lines. At other times, however, the declaimed verse passages

Voice  73 within prose sections come from verse passages that the singer has decided to shorten and to recite rather than sing. While these verses are recited on one occasion, they might be sung on another. When I recorded the epic of Qoblan from Jumabay, once in 1990 and once in 1994, I noticed when comparing the two versions that one of the differences between the two recordings rests precisely on this principle.10 My remarks about the performance of prose passages by Jumabay tally with the observations of folklorists about the performance of folktales. Not always, however, do the tellers of tales perform in the way one expects. Audibility would appear to be a necessary requirement of live narration, and yet we find many recordings with hurried and indistinct passages, not always due to the imperfect recording conditions. Sometimes the reason might be an intimate setting; there was no need for the singer to lift his voice in order to reach a large audience. Also, the singer will often be able to assume that what he has to tell is familiar to the listeners and can proceed at a swift pace. Some recordings have been made with older singers and amateurs whose memory and delivery skills have deteriorated in the course of time; this might account for hastily spoken passages in these cases. It has to be stressed, however, that the general assumption that in an oral tradition a narrator will make use of the full gamut of delivery skills is not true for all traditions, regional styles, and individual singers. A detached, unhistrionic mode of performance can also be typical of the performance of professional musicians and narrators. Yakut epic singers from Siberia, for instance, hardly move when they perform oral epics; also Minangkabau epic singers from Sumatra adopt a performance stance that strikes one as introvert rather than extrovert (see further Chapter 5).

Singing In a live performance, singing the words of an oral epic is far more common than speaking them. The various musical styles – solo singing, accompanied singing, melodic and musical patterning – as well as questions about the relationship between words and music will be the subject of later chapters. In this chapter the topic is voice quality and sound production in singing. When verse is sung rather than spoken, a number of differences can be noticed; the sound of a sung syllable differs from that of a spoken syllable in its acoustic make-up. In singing, pitch, duration, and intensity (amplitude) of a syllable are not dependent on the rules of spoken language, nor is the contour of a melody a copy of the prosodic shape of an utterance. To give an example, the English-Scottish ballad ‘Lord Randal’ begins every stanza with a question with the same syntactic pattern as found in the first line: ‘O where have you been, Lord Randal my son?’ In speaking, the highest pitch would be expected on been (and the equivalent syllable in the other questions), with a drop of pitch in the

74 Performance following phrase (‘Lord Randal my son’). The melodies, however, while sometimes rising to a high note at been, do often continue on even higher notes. The melodic variety of the tunes compiled by Bertrand Bronson casts doubt on any attempt to find correspondences between intonational and melodic patterns in this and any other ballad.11 To readers familiar with the Western literary tradition, the best known sung narratives will doubtless be ballads, the English-Scottish ballads, the Spanish romances, perhaps also the Rumanian or the Greek ballads. W. J. Entwistle (1969) has given a wide survey of European balladry, including also the South Slavic heroic songs and the Russian bylinas. Many of these traditions have either died out or are moribund by now, but most have been well documented, also in sound. For the bylina, for instance, a rich harvest has been preserved in the Phonogram Archive of the Pushkinsky Dom, the Institute of Russian Literature of the Russian Academy of Sciences in Saint Petersburg. These recordings come from the Soviet period, with early recordings from the 1920s and extensive recordings from the 1950s, 1960s, and 1970s. Some of these have been issued as records; in a comprehensive edition of bylinas, beginning in 2001, a number of volumes are accompanied by CDs.12 These sound recordings show a variety of musical styles; at this point, however, my comments are restricted to a very general characterization of the singing voice. When listening to the CDs of the new Academy edition of bylinas, there can be no doubt that they are sung with a natural voice, that is, with a voice well within the register of the informants’ tonal range and without any straining of the vocal chords. The same impression is conveyed by recordings (or live performances) of ballads in English, Spanish, Rumanian, and other languages. This kind of singing voice is widely diffused, but by no means universal. A quite different picture is presented by the performance of the Karakalpak jïraw. Coming back to the example from Jumabay’s narration of the epic Edige, after the singer has finished the story of Edige’s wise judgment, he continues with a sung passage. In the verses that follow we hear that Khan Tokhtamysh’s wife warns her husband that Edige will one day dethrone him and remarks that it would have been better to have killed the infant. The passage is introduced by a short melody played on the qobïz, then an exclamation (‘Oh, my God!’), followed by another short instrumental melody. The actual singing that follows comprises only twelve lines; then the singer changes into a high recitative for the rest of the verse passage (for a musical analysis, see Chapter 6 and Audio/Video Ex. 4). What is remarkable about the jïraw’s singing is that his voice quality changes noticeably from his everyday voice used in narrating. His singing voice sounds strained, tense, and taut. This is caused by glottalization, that is, by the narrowing of the aperture between the vocal chords, the glottis. By the same token the sound is produced with increased intensity,

Voice  75 which results in a pressed and ‘throaty’ voice.13 The singing voice here is decidedly not the speaking voice. Such a change of the voice quality in singing is also found in other musical traditions.14 A tense and sometimes strained voice is, for instance, found in a number of traditional singing styles of the Muslim world, especially in the Middle East and in Central Asia. Other epic singers in Turkic traditions also use a pressed voice, often strained because of singing in a high register. Singing styles of Turkmen singer-narrators, Turkish folk singers from eastern Anatolia, Uzbek bakhshis from southern Uzbekistan or Kazakh folk singers (aqïns) display this tendency to use a strained, guttural voice. There is, however, a difference between the pressed voice of a Karakalpak jïraw and the strained voice of a Turkish âs¸ık or a Kazakh aqïn. The jïraw’s voice shows a stronger glottalized quality, which increases the contrast with a natural voice. Such a pronounced change of the voice quality seems to point to an underlying concept of the singer that goes beyond that of narrator and musician. In the singer’s voice we can perhaps hear the echoes of another voice, the voice of the Central Asian and Siberian shaman.

Shamanic Voices Apart from his singing voice, there are a number of indications that link the jïraw to the shaman. The Siberian shaman has been much studied and innumerable books and articles have been written about Siberian and Asian shamanism as well as about shamanism elsewhere. Shamanic practices have continued among Siberian and non-Islamized Central Asian peoples into the modern age. Side by side with relics of ancient traditions, there have been efforts at revitalization and there has also been a development that has been labelled ‘neo-shamanism’.15 The shaman is in Eliade’s words both healer and psychopomp. As a ‘companion of the soul’ (Greek psychopompos) he accompanies not only the soul of the deceased but also of the sick when it is seized by malevolent spirits, and he strives to restore it to its body. In Islamized Central Asia shamanic practices have coalesced with elements of Sufism and popular Islam and the shaman has survived in the figure of the faith healer.16 This figure is called baqsï in Kazakh (see Chapter 2). The baqsï does not use the customary shamanic drum for his séances, but a bowed instrument which is built like Jumabay’s instrument and is also called qobïz. J. Castagné has painted a detailed picture of the Kazakh baqsï at the beginning of the twentieth century, with photographs, texts, and musical transcriptions. He characterized the Kazakh baqsï broadly in the following words: Singer, poet, musician, soothsayer, priest, and doctor, he seems to be the guardian of the popular religious traditions, the preserver of legends that are several centuries old. It is also he who serves as intermediary between the people of his race: a link that unites the men of his tribe

76 Performance or his horde with the legendary heroes of the Kazakhs. The oral poems handed down from the ancestors sing the praises of their heroes. (1930: 60 trans.) Jumabay’s qobïz, which he built himself, has a face (eyes and nose) carved on the peg-box. The peg-box is, according to his explanations, a representation of a horse’s head and the whole instrument suggestive of a horse. This is the symbolism also underlying the Kazakh baqsï’s instrument and, even more explicitly, the ‘horse-head fiddle’ (morin khuur) of the Mongols. This understanding of the instrument tallies with the interpretation of the shamanic drum by several Siberian peoples. Uno Harva, in his study of the religious concepts of the Altaic peoples, mentions the Yakut belief that the drum symbolizes an animal that carries the shaman into the world of the spirits and refers to the customs of the shamans of the Tuvans and other South Siberian Turks, who sing when they beat the drum covered by the hide of the maral deer: ‘I am a shaman and travel with a wild maral’ (1938: 536 trans.). Similarly, Castagné states: ‘The qobïz, like the shaman’s drum, is seen as an instrument that is furnished with magic power; it is even capable of coming to life’ (1930: 67 trans.). In addition to the instrument, which clearly harks back to shamanism, the singer’s vocation is also reminiscent of the shaman’s calling and initiation. As discussed in Chapter 2, many bards, like shamans, have a dream or vision in which they are called to become singers, experiences that are sometimes accompanied by an illness not unlike that of the shaman. In 1995, Jumabay told me that as a teenager he once spent the night in a mazar (grave-yard), where in his dream an old man admonished him to become an epic singer. But how, one might ask, are these various pointers toward a shamanic background related to the singer’s voice quality in singing the epic? Does (or did) the shaman sing with a pressed, guttural voice in a séance when he had worked himself into a state of trance? Following Éveline Lot-Falck, Gilbert Rouget, in his study of music and trance, distinguishes between a ‘cataleptic’ and a ‘dramatic trance’ in the course of the shaman’s spiritual journey. During a cataleptic trance the shaman rests immobile; it is only during a dramatic trance that music plays a role: In the course of the dramatic trance, however, the shaman describes what he sees during his voyage in the upper or lower world and narrates his adventures in singing and beating the drum, which leads to a genuine theatrical performance or, more precisely, a one man show, during which the most diverse musical interludes and styles follow one another: melodies, recitatives, spoken passages, dialogues, the imitation of animal shouts and sounds of nature, onomatopoeia, alterations of the voice. (1990: 248 trans.)

Voice  77 It is the alterations of the voice (what Rouget terms travestissements de la voix) that are particularly relevant to the present context. It seems to me that the jïraw’s glottalized voice quality is a survival of the shamanic practice of wrenching his voice from the ordinary. Curt Sachs called this ‘the depersonalization of the human voice’. He explains: Whenever singing is an act of ecstasy and depersonalization, it moves away from ordinary human expression. The voice is often remote from being as ‘natural’ as we believe our own execution to be. It is colored by pulsating, yodeling, ventriloquizing, or bleating. One screams, yells, squeaks, mumbles, and nasals [sic]. (1962: 83–84) Sachs also points out that ‘vocal mannerisms’ are widely spread, independent of ritual or ecstatic singing, and observes that ‘nowhere outside the modern West do people sing with a voice for which we have coined the honorific title of “natural”, that, to the western ear, all oriental and primitive singing is unnatural and seasoned with strange, unwonted mannerisms’ (85). This, of course, should make us wary of reading too much into the jïraw’s voice quality. On the other hand, however, it can be shown that in sung epics in Asia vocal styles are also found which stay within the register of the singer’s voice and are not marked by any ‘travestissements de la voix’. An example is the singing of the Kyrgyz epic Manas (see Chapter 6). Central Asian Overtone-Singing In addition to a strained, glottalized voice, other singing styles also occur in the Turkic and Mongolian world with strong links to shamanism. I will give some examples from the epic singing in the area of and around the Altay and Sayan mountains in southern Siberia. The Altay, lying on the borders of Russia, Kazakhstan, China, and Mongolia, is thought to have been the cradle of the Turkic and Mongolian languages. In the Altay Republic of the Russian Federation, Altaian, a Turkic language comprising several varieties, is spoken; closely related is Tuvan, spoken in the Tuva Republic to the south of the Altay Republic, which is also a part of the Russian Federation. The Tuvans have become famous in the West for their throat-singing or overtone-singing, called khöömey. This word is also found in other Turkic language, where it generally means ‘throat’ and ‘uvula’; it is considered to be a loan-word from Mongolian (Khalkha Mongolian khöömiy ‘throat’ and ‘pharynx’). This technique is also practiced by the Altaians and other Turkic-speaking peoples of the Altay (Khakas and Shor), the Bashkirs (a Turkic-speaking people of the Urals), and various peoples and ethnic groups of the Mongolian language family.17

78 Performance Overtone-singing or throat-singing is the production of two (and sometimes three) sounds simultaneously by adjusting the resonance cavity of the mouth in such a way that a drone-like base tone is joined by one or more harmonics or overtones. Carole Pegg offers a concise description in her study of Mongolian musical traditions: There are many different types of Mongolian overtone-singing (höömii), all of which involve the sounding of a fundamental drone while simultaneously producing flute-like notes in a series of chosen harmonics or partials of the fundamental. In most types, these high notes form a melody. A third note is sometimes distinguished, in the pitch range between drone and melody.18 The Tuvans also distinguish between several throat-singing styles. These styles are, however, not used for the performance of epics. Among available recordings of sung Tuvan epics, the singers use a lax, non-tensed singing voice in a middle or low register (Ch. Ch. Oordzhak in 1959; S. A. Kavaakay in 1982, S.-Sh. K. Bïshtak-ool in 1984); they sing solo and do not play an accompanying instrument.19 While in modern times epics are often recited, the traditional way of performing the epics is by singing either in a melodious style called ïrlap ïdar (‘narrate in singing’) or in a recitative style called alghanïp ïdar. The three epic singers I have mentioned ‘narrate in singing’. As to the recitative style, the Tuvan expression alghanïp ïdar means literally ‘to narrate by calling out incantations’; the verb alghan- denotes the shaman’s uttering incantations during a séance. Zoya K. Kyrgys characterizes this mode as recitation at a high pitch. She also emphasizes that the music of the Tuvan epics differs radically from that of other South Siberian Turkic peoples; while the latter use a singing style related to overtone-singing in the performance of the epic, Tuvan khöömey is practised only outside the epic. Altaian, Khakas, Shor qay/khay How can overtone-singing be used for poetry? It stands to reason that the words of a narrative cannot be sung in overtone melodies. Nevertheless, the throat-singing technique is incorporated regularly into the epic singing of the Altaians and the Khakas. What happens here is that the words are sung in a very deep register, with the oral cavity adapted to overtonesinging. The overtone effect occurs when at certain points in the flow of words the singer dwells on sustained vowels. This happens preferably at the beginning and at the end of a group of lines, in vocalizations of meaningless vowels. With some singers these overtone stretches occur fairly regularly, with others seldom or never. But even in the latter case do the singers use a low register and their singing is characterized by a certain hollow, echoic sound.

Voice  79 In Altaian and Shor this type of singing the epic is called qay, in Khakas khay; the epic singer is called qaychï in Altaian and Shor and khayjï in Khakas. In Khakas and Shor the epic singer is also called nïmakhchï and nïbaqchï, respectively, from nïmakh (Khakas) and nïbaq (Shor), ‘epic, tale’. All of these singers play an instrument when they perform the epic. The Altaian qaychï accompanies himself on a two-stringed plucked lute, the topshuur; the same instrument is also used by the Shor qaychï, who calls it, however, qay qomus; the Khakas khayjï, on the other hand, plays a plucked zither, the chatkhan (see Chapter 7). In the dictionaries, the words qay and khay are glossed as ‘throat-singing’. The etymology of this word is not clear. In Chaghatay, the literary Turkic language of Central Asia, qay means ‘omen, fate’ and qay sal- ‘to cause someone to become ill’. In Yellow Uyghur (a Turkic language spoken in Ganzu and Qinghai in China) qay means ‘bewitchment (causing illness); spirit of a shaman’. Kyrgyz has also a verb derived from the root qay, which, as in Altaian and Khakas, denotes a form of speech production; Kyrgyz qaylais glossed as ‘to mumble, to murmur’. From linguistic evidence it emerges that the original meaning of the word qay (or khay) was in all likelihood related to shamanism and signified the power of the shaman’s words and spiritual energy (see Reichl 1992: 63–64). The music of the qay/khay has been studied only comparatively recently. However, one of the first notations of a song from the Altay dates already from the eighteenth century; it is found in the third volume of Johann Georg Gmelin’s Reise durch Sibirien, von dem Jahre 1738 bis zu Ende 1740 (Journey through Siberia, from the year 1738 to the close of 1740). Gmelin was a botanist, but also a keen observer of the life and customs of the native Siberian peoples, including their shamanistic practices. He recorded the text of a Khakas song, put into the mouth of a widow longing for her husband, with translation and notation (1752: 371–72, plate after p. 370); the melody is clearly a folksong and not an example of overtone-singing. Wilhelm Radloff, in the record of his travels in South Siberia and Central Asia in the 1860s, prints four bars of the music of Altaian and Teleut epic singing and remarks that the epics are recited in a Brummstimme (droning voice), alternating between two notes.20 Altaian and other South Siberian traditional music and overtone-singing was first systematically studied by Andrey V. Anokhin (1869–1931). Writing about Altaian qay on the basis of Anokhin’s works and archive material, B. Shul’gin quotes Anokhin’s impression of throatsinging as suggestive of the sound of a flying beetle and then gives the following description: Throat-singing is not a scale of wheezing sounds, but proper singing, though of a peculiar nature. In this kind of singing the sounds are not uttered with the larynx free and open, but rather somewhat compressed. This kind of singing follows all the laws of music. When

80 Performance one happens to listen to throat-singing for the first time, it makes an unpleasant impression, but in the course of time the ear gets used to the sound, so that it begins to please and to act on the nerves in a tranquilizing manner. ‘For the Asiatic ear such a singing is certainly pleasant.’21 It is no doubt very difficult for our present-day understanding to imagine a heroic narrative which continues for seven nights to have a musical form. The utterance of words sung in a low guttural timbre, the ‘ostinato’ of the melodic phrases of the topshuur, the resounding ‘organ point’ of both the voice and the topshuur, the plagal character of the melody and so on22 – all of this leads the listeners into a peculiar ‘psychological trance’, which helps to distance themselves from the surroundings and gives full freedom to their imagination. And finally, listening to the live sung speech of the qaychï, full of picturesque similes, the listener lives through all the adventures which the hero of the narrative experiences. (1973: 459 trans.) From one of the most renowned Altaian epic singers of the second half of the twentieth century, Aleksey Grigorevich Kalkin (1925–98), a number of epics have been written down; we also have some sound recordings. The best known Altaian epic is probably Maaday Qara, an epic that was included, in Kalkin’s version, in the series ‘Epics of the Peoples of the Soviet Union’ (later ‘of Eurasia’; Surazakov 1973) and has been translated into English (Marazzi 1986). This epic, comprising 7,738 verse lines, tells a complex story with Maaday Qara as the main hero, who after many adventures succeeds (partly with his son’s help) in killing his enemy Qula Qan and the latter’s wife, the daughter of Erlik, the khan of the underworld. In the final episodes Maaday Qara descends into the underworld and overcomes also Erlik himself and his son. We are in the presence here of a mythological epic which differs significantly in plot, narrative style and epic world from the heroic epics of Central Asia (see also Chapter 9). Maurice Bowra characterized epics like Maaday Qara as ‘shamanistic poetry’ and, as I mentioned in Chapter 1, excluded them therefore from his wide-ranging study of heroic poetry. The three-layered cosmological conception of this type of epic is expressed visually by the description of the tethering post for the hero’s horse, which comes fairly soon at the beginning of the epic: Altïï uchï altïn oroon – Aybïstannïng bu chadanï, Üstii uchï üstii oroon – Üch-Qurbustan bu chaqïzï. Tal ortozï Qara qaltar jaqshï attu

Voice  81 Maaday-Qara baatïrdïng Bu chadanï bu boluptïr. (Surazakov 1973: 70–71) The lower part of it in the underworld is Aybïstan’s tethering post, the topmost part of it in the upper world is Üch Qurbustan’s tethering post. Its middle part serves as a tethering post for the hero Maaday Qara with the beautiful dark-bay horse. (Marazzi 1986: 33) As Ugo Marazzi points out in the notes to his translation, Aybïstan is another designation of Erlik, the khan of the underworld, and Qurbustan derives ultimately from Ahura Mazda, the creator god of Zoroastrianism (1986: 33–34). A sample of Kalkin’s singing of Maaday Qara, recorded by Radio Moscow in 1972, is accessible on a CD with Bashkir overtone-singing.23 Two extracts from Kalkin’s version of the epic Ochï-Bala have been included on the record accompanying the edition of this and another epic in 1997.24 The main protagonist of this epic is a heroic woman, who fights against the invading army of Khan Taaji-Biy and his son Aq-Jala. She is successful in the end, but only after overcoming several obstacles and destroying the nine-sided blue stone that contains the khan’s external soul. As Yu I. Sheykin and V. S. Nikiforova showed (1997: 62), Kalkin’s singing style developed over the years. While he used overtone-singing sparingly in his performances in the 1940s and 1950s, he extended the use of overtone-singing from the 1960s to the beginning of the 1980s and then reduced it again. While the recording of Maaday Qara in 1972 comes from Kalkin’s peak period, the recording of Ochï-Bala in 1984 is an example of ‘the cooling down of the lava’ of Kalkin’s qay. On the earlier recording the harmonics can be clearly heard, while on the later the ‘overtone’ is indistinct and notated as ‘noise’ rather than as a specific tone in the transcription. In addition to the recordings of epics, we have also recordings of Altaian folktales, among them by Kalkin. In Altaian folktales interspersed verse passages are sung (Kondrat’eva 2002). Interestingly, Kalkin’s singing style here differs significantly from that of the epic qay: he sings in a relaxed voice and uses a middle to high register.25 A similar situation applies to Khakas and Shor epic singing, with some variations however. The epic singers of South Siberian Khakas sing the epic in what is termed ‘recitative khay’, which means that it is sung on one note, generally in a low register and always to the accompaniment of the chatkhan. Depending on the singer, sustained notes can have the overtone effect. It is common for singers to switch between singing and speaking; in the spoken passages the sung passages are retold.26 This method of alternating between singing and speaking is also typical of Shor singers.27 A. I. Khudoyakov and R. B. Nazarenko prepared the edition of a Shor

82 Performance epic with full musical notation of the sung passages (including the music of the accompanying qay qomus); the accompanying record illustrates the change between singing in the qay style and reciting (1998). The low, drone-like singing tone used by the South Siberian qaychï / khayjï is also found in the performances of the Buryat epic singers. The Buryats, who speak a Mongolian language, live mainly in the Buryat Republic of the Russian Federation around and east of Lake Baikal. Roberte Hamayon likens the low drone-like timbre to certain forms of shamanic invocation and affirms that the epic singer ‘shamanizes’ the epic, this is to say sings in the way the shamans sing (1993: 352). At the end of this section a note of caution is perhaps in order. While I have focused on typical cases, it has to be remembered that not only does every tradition offer a variety of options but also that the singers themselves differ one from the other. However traditional they might be, they have their individual story and their individual way of handling the cultural legacy entrusted to them. As can be seen in the case of the Altaian qaychï Kalkin, singers may also change in the course of their career, and so do traditions. A second point I would like to make concerns the relationship between shamanism and epic singing. Many aspects of epic performance, as well as of the singer’s calling, have parallels in shamanic practices and traditions, as I have tried to show. Here, too, changes and shifts have to be taken into account. It is true that the Karakalpak jïraw’s instrument is the same as that of the Kazakh baqsï. And there is no denying that the assumption of a common past of jïraw and baqsï seems reasonable. It has to be noted, however, that the Karakalpak equivalent of the Kazakh baqsï not only had a different name, namely porkhan, but also a different instrument as ‘shamanic drum’, a whip, not a qobïz. Some of the echoes of shamanism in epic performance can be heard, but seem to come from a fairly distant past. Finally, I would like to stress that none of the musical performance styles discussed is even remotely similar to the shrieks and cries or other sounds shamans are said to utter. The singers’ connection to shamanism is much more tenuous than the various parallels suggest. There is an element of cultural interpretation in linking qay with shamanism. Many writers on overtone-singing in Asia have underlined the spiritual quality of this kind of singing. Also native artists, musicians, and hommes de lettre have developed this view of khöömey and other types of overtone-singing. While there is this spiritual appreciation of overtone-singing and while the experience of listening for hours to the voice of a South Siberian qaychï no doubt casts a magic spell on the listeners, it has to be remembered that an epic performance can only metaphorically be called a shamanic séance. The world of the shaman is visible, also in the plots and characters of the epics, but it is no more than a memory, sometimes vivid, sometimes pale and barely noticeable.

Voice  83

Notes 1 Faral 1971: 259, vv. 2031–32; translation Nims 1967: 90. 2 The quotation is taken from Wadley 1989: 75; for a short summary of the plot, see Blackburn et al. 1989: 219–23; for a detailed study, see Wadley 2004. The epic is found ‘in the western desert of Rajasthan, the plains of western Uttar Pradesh, and the rolling hills of Chhattisgarh’ (Blackburn et al. 1989: 219). 3 On selected synonyms of ‘voice’ in the Indo-European languages, see Buck 1949: 1248. 4 Wadley 2004: 72; compare Wadley 1989: 81. 5 Text and translation are taken from my edition and translation (2007: 187, 294). I have adapted my transcription to the system used in this book. 6 See Motif J1171.1 and ATU no. 926; Ludowyk 1959. 7 For a more detailed account, see Reichl 2007: 143–53; on Karakalpak phonology and morphology, see Baskakov 1952. 8 See especially Trager 1964; for elaborations, see Laver 1968. 9 See in this connection my discussion of a prose and a verse passage from Jumabay’s performance of the epic Qoblan in Reichl 2000c. 10 For the declamatory lines in Edige, see Reichl 2007: 184. 11 For a detailed study of the singing voice, see Sundberg 1987; for various tunes of ‘Lord Randal’, see Bronson 1976: 46–54. – On the relationship between contour of speech and song in tone language, see Chapter 6. 12 On this series, see Putilov et al. 2001: 11–20. 13 The closest parallel to this type of voice production in English is found in the articulation of the so-called glottal stop, especially as found in varieties of British English such as London Cockney (‘wha’ a lo’ of li’le bo’les’); see Gimson 1980: 9. 14 On vocal styles in traditional music, see Bose 1953: 51–60; on tenseness, see Merriam 1964: 105–8; compare Reichl 1992: 112. 15 Basic introductions and studies include Eliade 1964 and Vitebsky 1995; for a comprehensive encyclopedia, with entries on various Turkic ethnic groups, see Walter & Fridman 2004. 16 On Sufism, shamanism, and women in Central Asia, see Sultanova 2011; the Khorezmian women singer-narrators (khalfa) are also active as faith-healers (see Chapter 2). 17 For a general and wide-ranging introduction to overtone-singing, see van Tongeren 2004 (with CD). For a lively travelogue and musicological discussion of overtone-singing and related singing styles among the Tuva, see Levin (with Süzükei) 2006 (with CD); for a study of Tuvan overtone-singing, see Kyrgys 2002. 18 Pegg 2001 (with CD): 66–66, at 60; see also Pegg 1992. 19 These singers are represented on a record accompanying Orus-ool et al. 1997; on the music, see Kyrgys 1997. 20 Radloff 1893: I, 342. Teleut is another Turkic language of the Altay, closely related to Altaian. 21 A quotation from Anokhin (archive material). 22 This is a reference to the system of the Church modes, where the ambitus of the plagal scales lies a fourth below that of their respective authentic scales. The organ point or pedal point is a long-held bass note with changing harmonies above it. 23 Uzlyau: Guttural singing of the peoples of the Sayan, Altai, and Ural Mountains (PAN records 1993, PAN 2019CD), track 14.

84 Performance 24 See tracks 1 and 2 of side A of the record added to the edition and Russian translation of the epics (Kazagacheva 1997). 25 See tracks 10 and 11 on the CD accompanying Sadalova 2002. But see the comment by the woman singer M. S. Sumachakova, who uses the same melody for folktale and epic, ‘because there is no difference in the singing of epic and folktale’ (Kondrat’eva 2002: 45 trans.). 26 See Stoyanov 1988 and Shevtsov 1997; the latter comes with a CD illustrating the style of two epic singers. 27 See Funk 2005 on Shor epics; on the music of Shor epics, see Nazarenko 1998.

5 Gesture

IEsture is a certaine comely moderation of the countenaunce, and al other parts of mans bodie, aptly agreeing to those thinges which are spoken. (Thomas Wilson, The Arte of Rhetorique) Thomas Wilson’s comments on gesture (or iesture) come at the end of his Arte of Rhetorique, a treatise first published in 1553. This work is a key text in the history of the English language on account of Wilson’s witty attacks on his contemporaries’ rage for ‘inkhorn terms’, learned neologisms borrowed mostly from Latin and Greek. Wilson understands by gesture both facial expressions and bodily movements and stipulates that they accord with the meaning of the speaker’s words. When we speak of pleasant things, Wilson observes, we should not frown or knit our brows, puff through the nose or bare our teeth, and we should also avoid jerking our arms. He finishes his chapter on gesture by quoting Cicero as saying: The gesture of man is the speech of his bodie, and therefore reason it is, that like as the speeche must agree to the mat[t]er, so must also the gesture agree to the minde, for the eyes are not giuen to man onely to see, but also to shewe and set forth the meaning of his minde. (Mair 1909: 221) The view that gesture is expressive of what is conveyed by speech is the basic tenet of rhetoric in classical antiquity, bequeathed to medieval authors like Geoffrey of Vinsauf, quoted in the last chapter. Cicero formulated this view in his dialogue on the orator and his skills (De Oratore), which was completed in 55 bce.1 Cicero’s ideas were taken up and further developed in Quintilian’s handbook of rhetoric (Institutio Oratoria, c. 96 ce). In Book XI Quintilian devotes the greater part of the third chapter to voice and gesture, with the addition of detailed instructions. He stresses their importance, since voice and gesture are the two gateways to our understanding of what is said: Cum sit autem omnis actio, ut dixi, in duas divisa partes, vocem gestumque, quorum alter oculos, altera aures movet, per quos duos

86 Performance sensus omnis ad animum penetrat adfectus, prius est de voce dicere, cui etiam gestus accommodatur. All delivery, as I have already said, is concerned with two different things, namely voice and gesture, of which the one appeals to the eye and the other to the ear, the two senses by which all emotion reaches the soul. But the voice has the first claim on our attention, since even our gesture is adapted to suit it. (XI.iii.14, Butler 1920–1922: IV, 248–51, text & tr.) After his treatment of the voice, Quintilian offers a great deal of advice about the use of gestures, both facial expressions (what to do with our lips, nose, or eyebrows) and bodily movements. He recommends Demosthenes’ example, who practiced before a mirror to study the effects of his gestures, and, like Cicero and later Wilson, admonishes the orators to harmonize their voice and gestures with the meaning of the words they are uttering. He is aware of the importance of voice and gesture in drama and also points out that gestures can be used independently of speech as semiotic systems of their own, as in different hand movements and fingerpositions. Quintilian is concerned with the orator, but much of what he says is also of interest for an analysis of the gestures and gesticulations of the singer-narrator and public entertainer. The use of gestures is, of course, not restricted to the orator, actor, or singer-narrator. We use gestures all the time when speaking and also independently of speaking. Gestures have therefore not only been studied by rhetoricians, but also by linguists, psychologists, anthropologists, folklorists, art historians, and also specialists in other disciplines. Scholars like Edward B. Tylor and Wilhelm Wundt from the second half of the nineteenth and early twentieth century studied gesture primarily in connection with the question of the origin of human language. Gesture is seen as a major channel of communication, which played a dominant role in the early phases of language development. According to Tylor this dominant role becomes subsidiary or redundant in the formation of a natural language: By eliminating from speech all effects of gesture, of expression of face, and of emotional tone, we go far toward reducing it to that system of conventional articulate sounds which the grammarian and the comparative philologist habitually consider as language. (1873: I, 167) Wundt treated both facial expressions and gestures extensively in the first volume of his Völkerpsychologie, which is concerned with language. Among other topics, he discussed various sign languages (e.g., of the Native Americans and the Cistercian monks), deaf-and-dumb languages, and also miming and pantomime.2 In modern linguistics and semiotics the study of gesture has become a discipline of its own under the heading

Gesture  87 of kinesics. Adam Kendon, a leading linguist in this field, gives a succinct list of the different uses of the term ‘gesture’: The topic of gesture includes bodily movements that occur in close association with speech (gesticulation); expressive movements that can be used independently of speech and can serve as complete utterances on their own, such as the thumb-up gesture, the shouldershrug, or shaking of the fist (autonomous gesture); gestural codes used in certain occupational circumstances in which speech is difficult or impossible, primary sign languages used in communities of the deaf, alternate sign languages used in some tribal and religious communities during periods when speech is forbidden, and the special elaboration of gesture found in religious ritual, as in the complex systems of mudras developed as part of the ritual of prayer in Tantric Buddhism; and the complex gestural systems found in some dance traditions, especially in India.3 Before turning to the oral epic, several distinctions have to be made. In the following I will use gesture as the general term and gesticulation for movements of the arms and hands such as we generally associate with the delivery of an orator, narrator, or rhetorically experienced speaker. The word gesticulation carries the connotation of lively and possibly exaggerated movement. As will be seen, it seems to be the appropriate term for the gestures of singers such as the Kyrgyz manaschï. In the art of narrating also facial expressions – wrinkling one’s nose, knitting one’s brows, pursing one’s lips, and so on – and miming play a role; these will be included in the discussion of gesture. With reference to the performance of oral epics, I will use the terms ‘conventional’ and ‘stylized gestures’. By conventional gestures I mean gestures that accompany the delivery of speech in a linguistically and culturally defined context. The frequency and quality of these gestures will vary according to speaker and context. A politician in a televised interview will make different gestures according to his or her personality, to the language spoken, possibly to his or her political orientation, to the temporal and geographical setting and so forth. It can be assumed that the gestures occurring in the interview will be understood and correctly interpreted by the viewers who speak the same language and belong to the same cultural environment. This does not necessarily exclude viewers who don’t ‘belong’. It is, for instance, probably universally the case that curling one’s lips upwards into a smile is an expression of happiness. Many gestures, however, which seem natural at first sight turn out to be culture or language-specific on closer analysis. There is often a conventional side to what looks like a natural kinesic sign. Shaking your head might signify dissent in one cultural or linguistic space, but assent in another. Although the gestures discussed in the following section seem to be ‘natural’, I propose

88 Performance to call them ‘conventional’ even if they are also immediately understandable to outsiders. The choice of the qualification ‘conventional’ does not express an opinion on the question of gestural universals. In addition to gesture (and gesticulation) as an accompaniment of speech, Kendon also speaks of gestural codes. In this connection he mentions sign languages as alternatives to speech, but also the gestural codes of ritual and dance. I will call these gestural codes ‘stylized gestures’. They are found in dramatic forms like the Japanese Nō plays or the Chinese Peking Opera, in dances such as the Hindu temple dances with their symbolic hand movements, or in the performance of oral epics, as will be seen below. In the performance of various genres, these gestures and the use of body language are interpreted according to certain conventions and rules regulating the performance of the genre. This does not mean, however, that they cannot occur outside a specific performance or that their interpretation requires special clues. Nevertheless, they are expected when an oral epic or some other genre is performed and they follow a certain pattern. Although both ‘conventional’ and ‘stylized’ gestures presuppose conventions, the former are less rigidly regulated and closer to everyday gestures than the latter. The distinction between them will become clearer when specific cases are studied.

Conventional Gestures: The Karakalpak Jïraw As we have seen in the preceding chapter, the Karakalpak jïraw accompanies himself on a bowed fiddle (qobïz). Quite naturally, singers who play a musical instrument are restricted in their gestures. There is little or no opportunity for raising one’s arm, pointing or shaking one’s fist and the like. The jïraw has to sit in order to play the qobïz. When playing an instrument, singers might sway or move the upper part of their body back and forth with the music, just as musicians in other musical traditions move while playing. These movements are expressive of the intensity with which the singer plays, but they have no reference to the words sung. When, however, the singers narrate stretches of the epic without singing or playing their instrument, as is the case in the performance of prosimetric dastans, the singers have more freedom for gesturing, unless, of course, the singers continue strumming their instruments when reciting the prose parts. Jumabay Bazarov holds his fiddle and his bow in one of his hands (sometimes in both) while narrating in prose; he interrupts his recitation at irregular intervals by playing a short motif or a chord on his qobïz. While narrating, Jumabay’s most common posture is one of comparative stillness, his left hand holding instrument and bow, with the instrument resting on his left thigh, his right hand lying on his right knee. In telling the story, the singer moves his head, most commonly to his right, often in combination with changes of the speakers quoted in a dialogue. Jumabay’s gestures are basically of three kinds. The first kind is a type of gesture that is wide-spread among narrators: a hand is moved upward,

Gesture  89 sometimes pointing the index finger, at moments in the story to which the narrator wants to draw attention:

Illustration 2  Jumabay Bazarov (1)

A second kind is imitative of some object or action. This is what Wundt called graphic gestures: ‘The deaf-mute, as also the Indian and the Australian, represents an absent object by pictures outlined in the air’ (Wundt 1916: 62). Other typologies speak of iconic gestures (Haviland 2004: 201). Such a gesture occurs, for instance, when the singer narrates an episode from the hero’s life, in which Edige solves a dispute between two women who gave birth to a boy and a girl respectively, but both lay claim to the boy. Edige asks each of the women to pour milk into a thimble and then weighs their milk: the heavier milk is that of the boy’s mother.4 The singer’s arm and hand movements suggest the pouring and weighing process:

Illustration 3  Jumabay Bazarov (2)

90 Performance The third kind is extrinsic to the story. These are gestures and movements that break up the flow of the narrative, as when the singer pours himself and sips a cup of tea, strokes his beard or twirls his moustache. In addition there is a vivid play of his facial expressions: the singer looks at his audience, looks to the side, or closes his eyes (especially when playing), while his facial expressions vary from seriousness to laughter:

Illustration 4  Jumabay Bazarov (3)

Summarizing Jumabay’s gestures, we can say that they are mainly those of natural narration, which, despite conventional elements (such as the general narrating posture with the instrument and bow in one hand and the other hand resting on the singer’s knee), are remarkably similar to the repertoire of facial expressions and hand and arm movements found among other narrative traditions. Illustrations and analyses have been published for a number of oral traditions, such as Karl Haiding’s study of the gestures of folktale narrators in Austria (1955), Harold Scheub’s interpretations of the gestures and body movements of a Xhosa narrator from South Africa (1977), and Geneviève Calame-Griaule’s discussion of the gestures of a Tuareg narrator (1977b).

Stylized Gestures: The Kyrgyz Manaschï In some traditions the singer-narrator plays no instrument. In the Turkic world, this is typical of the Yakut olonkhosut and the Kyrgyz manaschï, the singer of the Manas epic. Two basic possibilities exist for narrators who do not need their hands for playing an instrument: they can hold them still or use them to gesticulate. The Yakut singer chooses the first alternative (see below), while the Kyrgyz manaschï chooses the second alternative. In the autumn of 2011, I travelled to the town of Ulughchat in the Kyrgyz Autonomous Prefecture of Kizilsu in the Chinese Province of

Gesture  91 Xinjiang. A Kyrgyz minority of about 180,000 live in China; the majority of the Kyrgyz live in the neighbouring post-Soviet Republic of Kyrgyzstan. I went to Ulughchat to re-record the Kyrgyz heroic epic Qurmanbek from the singer Sart-aqun Qadïr, who had sung this epic to me in 1989. Sartaqun brought some of his pupils along, who sang extracts from the Kyrgyz epic Manas. The singers seemed to be somewhat nervous. Perhaps the most obvious sign of nervousness was that the first to perform sat stockstill in his chair and had to be reminded by his teacher to gesticulate with his hands while singing. They took their teacher’s admonition to heart and performed for me with frequent and energetic movements of their arms. We do not know how old these gesticulations are. The earliest students of Kyrgyz epic poetry in the nineteenth century, Chokan Valikhanov and Wilhelm Radloff, do not say anything about gestures. Photographic and later film documentation of Kyrgyz singers in the twentieth century, however, shows that gestures are an important part of performance. While many gestures are ‘conventional’ in the sense defined above, the majority are stylized, that is, specific to the singing of Manas. Every singer has his or her individual way of using gestures – some use more than others and some gesticulate more often than others – but there is a common stock of gestures shared by all. As will be seen, these gesticulations are not random, but comprise a fairly well-articulated set of movements. As an example, I will take the performance of one of the epic singers I recorded in 2011, an episode from Manas sung by a woman by the name of Döölötqan, who was born in 1971. She performed young Manas’s fight with the Kalmyk khan Köngtöy. This is an episode which is typical of the version recorded from the Xinjiang Kyrgyz singer Jüsüp Mamay (1918–2014). Döölötqan keeps fairly close to Jüsüp Mamay’s text.5 The passage comprises 123 lines; it begins with Manas riding on his horse Aqqula to meet Köngtöy in single combat: Aqqula minip bolqoyup, Alatoodoy zongqoyup, Topton chïqtï bölünüp. (1–3) Looking very handsome on Aqqula, Towering like the (mountain) Alatoo, He (Manas) came forward. The two opponents attack each other with their lances: Nayzalarïn sunushup, Küülöp alïp urushup, Qachïrïshïp sayïshtï. (ll. 21–23)

92 Performance They pointed their lances, Held them in readiness and started to fight, Rushed one on the other and stabbed at each other. In the fierce fight shields and armour are shattered. Köngtöy is angry at having to confront a boy of eleven or twelve and thrusts his spear in fury. Manas, however, avoids the blow and gives Köngtöy a stab in his ribs. Köngtöy is thrown from his horse and dies. The passage ends with a description of the fear and consternation that Manas’s victory spreads among the Kalmyks and ends with the words: Aylasï tügöp qalmaqtïn, Alapayï quuruldu. Höy! (ll. 122–23) The Kalmyks had lost their cunning, They knew no way out of the dilemma. Höy! The performance of this passage takes about five and a half minutes. As is usual, the passage is broken up into sections of different length, of which each ends with a special closing melody (see Chapter 6). There are six sections in this passage, all ending with an exclamatory particle (höy!). For every single line Döölötqan makes a hand movement. Only for some of the last lines of a section does she remain still. There are, all in all, 120 instances of gesticulation in the 123 lines of the passage (see Audio/Video Ex. 2). The movements are of two kinds: either the right arm only is involved in the movement or both arms are. Often the two kinds alternate: a rightarm movement is followed by a two-arm movement and vice versa. There are fifty-seven instances of a one-arm movement and sixty-three instances of a movement of the two arms. Each of these movements can be broken up into features, similar to the way phonology analyses sounds as bundles of distinctive features.6 Looking at Döölötqan’s two-arm movements, we notice the following possibilities: (1) in some cases the upper part of her body is upright, in others bent; (2) she either faces forward or to the right or left; (3) both arms are raised to about the height of the shoulders or higher over the head; but they might also be lowered towards the floor; (4) both arms can remain parallel to each other or be opened wide; (5) the hands are generally spread open, with the palm or the back turned to the audience, or they are closed into a fist; (6) a unique two-arm movement in the recorded passage is the following: the two arms are crossed with the right hands on her throat (see below). All of these features have variants: the bending or the turning to one side might be more or less accentuated, the arms might be raised higher or less high towards the ceiling etc. In the

Gesture  93 following three illustrations the first shows one of the positions with raised arms, the second a downward movement and the third the crossed arms:

Illustration 5  Döölötqan (1) 49 Küchüm tursa boyumda/ If I have strength in my body (Köngtöy is thinking)

Illustration 6  Döölötqan (2) 91 Jash Manas kirdi qachïrïp/ Young Manas galloped into the mêlée

Illustration 7  Döölötqan (3) 27 Qoqoloshup buushtu/ They throttled each other

94 Performance Somewhat more varied are Döölötqan’s one-arm movements. The arm does not just move into a particular position and then return to its initial position, but it generally makes a further movement before coming to rest. The following movements comprise the basic repertoire of Döölötqan’s one-arm gesticulations: (1) the right arm is raised – the hand at about the height of neck or head – and is then moved towards the front of the neck/left shoulder in a circular movement. Variations are that the right arm is not raised quite that high and makes the circular movement somewhat lower and that instead of an upright position the upper part of the body is bent forward; (2) the right arm is raised, stretched and points to the side; the hand opens from a fist position. Variations are that the arm is not stretched, but bent, and that the arm points downward; (3) the right arm is raised and then quickly moved downward and towards the body. This is a movement similar to that of hitting an object or swinging a whip or slapping someone’s face. Variations are that the movement is less vehement and looks as if something was being thrown away and that the hitting movement looks like a beheading gesture; (4) the right arm is extended at the side and then jerked towards the middle of the body as if thrusting a dagger into the singer’s abdomen.7 The following drawings illustrate one of the raisedarm positions with a circular movement (1), the ‘whip movement’ (3), and the ‘dagger-movement’ (4):

Illustration 8  Döölötqan (4) 1 Aqqula minip bolqoyup/ (Manas) Looking very handsome on Aqqula

Gesture  95

Illustration 9  Döölötqan (5) 56 Tashtï chapsam qaytpaghan/ When I hit a stone, it (my sword) will not get indented (Köngtöy)

Illustration 10  Döölötqan (6) 43 Öpkösünö malïshtï/ They were intent on making (their spears) penetrate their lungs

Although the distinctions proposed and the examples given do not exhaust the variety of Döölötqan’s performance, the basic shape of her gesticulation repertoire does become clear. The various movements fall into two categories, stylized gestures, and iconic gestures. The stylized gestures express, as it were, a general narrative pose: spreading out one’s arms, raising the right arm in an unspecified pointing gesture, moving the arms downwards, or making a circular movement with the raised right arm. All these movements seem to say: ‘Here I am! Here is my story! Here is what happens!’ In other words, they underline the narrator’s presence and point, one could say, to the story. There does not seem to be a

96 Performance correlation between a specific movement and the content of a verse-line which the movement accompanies. There is perhaps one more gesture of a slightly different kind, the raising of the two arms at the end of a section and especially at the end of the whole passage. Here the gesture serves as a marker of narrative structure; it is a stylized gesture, but with the specific function of segmenting the flow of verses. Exceptions are the one-arm movements (3) (‘whip movement’) and (4) (‘dagger movement’). Here a correlation between gesture and meaning can be established, especially in the case of gesture (4). The ‘dagger-thrust’ occurs in lines 42, 43, 66, and 101 of the passage: 42 ‘Ölör jering ushul’ dep/ Saying ‘This is the place where you will die’ 43 Öpkösünö malïshtï/ They were intent on making (their spears) penetrate their lungs 66 Toltongo qolum salayïn,/ I will dip my hand into your chest (says Köngtöy) 101 Qïy süböödön bir saydï/ And he (Manas) stabbed him (Köngtöy) once in his ribs As to (3), this gesture occurs more frequently; it cannot always be correlated to the meaning of the line. However, in some instances, where the movement is more rapid and more violent, there is a correspondence between gesture and meaning, as in Illustration 9: 56 Tashtï chapsam qaytpaghan/ When I hit a stone, it (my sword) will not get indented (Köngtöy) 98 Qulanïn bashïn imerip/ (Manas) turning Aqqula’s head round Also one of the two-arm gesticulations, the crossed arms in Illustration 7, is certainly iconic as it refers to throttling. Finally, the often quite forceful two-arm movement downwards might in some cases be related to content. The most noticeable case is line 103, where Köngtöy’s tumbling down from his horse is described: Tomorulup qïyshaydï/ Overcome by this blow, he (Köngtöy) broke down. When Döölötqan’s gestures and gesticulations are compared to those recorded from other manaschïs, we can see a number of similarities, but we can also find additional arm movements.8 Sayaqbay Qaralaev (1894– 1971), whose version of Manas comprises no less than about half a million verse-lines, was a particularly agile and lively manaschï. In the film Manaschy by Bolot Shamshiev (1966), Sayaqbay makes the following gestures (among many others): The singer makes the gesture of shooting with a bow and arrow; he strikes the fist of one hand into the palm of the other; he strikes with the index finger of one hand against the fingers (or the palm) of the other hand; he extends both hands, palm upward, as in praying.

Gesture  97 Here we have the same dichotomy between stylized gestures and iconic gestures. In fact, some can be interpreted as natural-conventional gestures of a deictic (pointing) and emphasizing kind. The words are, as it were, underlined or italicized by these gestures. All of these gestures or gesticulations – natural-conventional, stylized, and iconic (imitative) – serve, it seems to me, the same purpose. By accompanying and punctuating the performance, they add a visual dimension to the sung words and ‘translate’ the tension and drama of the tale into a sequence of movements. They are a gestural accompaniment of the words, which is only sometimes expressive of a specific content (e.g., the dagger gesture).9

Gesture and Inspiration Kyrgyz Kyrgyz manaschïs have their hands free to gesticulate. Exceptions do, however, occur. The manaschï Almabek Toychubekov (1888–1962) is said to have held and waved a piece of white cloth while performing. The contemporary Kyrgyz manaschï (and Manas scholar) Talantaaly Bakchiev used to hold a whip when performing and gives the following account of this: My teacher, the manaschï Shaabay Azizov [1924–2004], used to say to me: ‘Look, Talantaaly, take a whip in your hand when you perform Manas in front of a large audience!’ I asked: ‘Why should I take a whip?’ Old Shaabay’s answer was as follows: ‘Your performance could be affected by the evil eye and be a source of gossip and slander among the people. The whip will protect you from both the evil eye and evil speech. That’s why you should take a whip with you when performing Manas.’ Later, I saw Almambet and Sïrghaq [two heroes from the epic] in my dream. They gave me a span-sized whip made of leather, which had no handle. Of course, it happened in my dream. Then I got a craftsman to make a whip and began to have it with me while performing Manas in front of a large audience.10 This is a very suggestive comment. Clearly the whip is a kind of talisman. In Kyrgyz popular belief a whip and a knife are considered a protection from evil spirits. This role of the whip (and the knife) has its origin in Kyrgyz shamanism. In some traditions shamans used a whip instead of a drum. This was the case with the Karakalpak shaman (called porkhan) and also the Kyrgyz shaman (called baqshï). I am told by a Kyrgyz doctoral student from Osh (Kyrgyzstan) that she has seen Kyrgyz baqshïs several times in the Osh bazaar. ‘They used a whip, a knife, and beads during their shamanic séances. According to some sources and to what I was told, they use the whip in order to chase away illness and the evil eye.’11 This ties in with the observations of other researchers.12 Although Kyrgyz epic singers do not enter a state of trance when performing the epic, there is a close connection between epic singer and shaman in

98 Performance Kyrgyz tradition. The connection is provided by the initiation dreams or visions Kyrgyz manaschïs profess to have had, as discussed in Chapter 2. Yakut The Yakut epic singers, like the Kyrgyz manaschïs, perform the epic without the accompaniment of a musical instrument, but do not use gestures in their performance. The olonkhosut, the singer of the olonkho, sits still and does not use his or her hands for gestures. Various positions are possible: The singers sit with their hands on their knees as if they were supporting themselves; the singers sit with their legs crossed, their hands clasping their knee; the singers cover their right ear with their right hand. As the olonkho-scholar Aitalina Kuz’mina comments: ‘There is the belief that the olonkhosuts cover their ear so that the evil spirits (abaasï) cannot whisper to them and disturb the performance of the olonkho. It is possible that in this way the singers can better concentrate.’ She adds: ‘Contemporary performers, artists, sometimes gesticulate, but the audience clearly does not like this very much. In the northern districts of Yakutia the epic is sometimes performed with the singer lying on a bed.’13 The position with the right hand cupping the right ear is illustrated in a drawing by the Yakut artist Elley Semyonovich Sivtsev (1928–94):

Illustration 11  Yakut olonkhosut

The epic scholar I. V. Pukhov has written a detailed study of the olon­ khosut’s performance. He describes the performance of Aleksey Trofimov

Gesture  99 which he witnessed as a boy and compares it to other performances. The singer performed all night from evening to sunrise and interrupted his performance only a few times to drink some tea. ‘He sat with his back to the little chimney, with his legs crossed, the hands on his legs, slightly swaying backwards and forwards with the upper part of his body’ (Pukhov 1951: 137 trans.). Although the singer did not gesticulate, he did not sit completely still: With the unfolding of the gripping events of the olonkho, he became excited and became more and more ‘ecstatic’. And while he was sitting relatively quietly on the little stool at the beginning, he now began to move, changing the position of his body and his hands and legs and becoming livelier and livelier. But after a while he became calm and once again sat quietly, motionless, with his eyes fixed on some point in front of him. (138 trans.) Pukhov also mentions that many singers cup their right ear, sometimes with the whole hand, sometimes only with their fingers. Asked why he does this, a singer (U. G. Nokhsorov) answered: ‘In this way, the sound reverberates more strongly in the head’ (141). Another singer (A. E. Kulakovsky) used to close his eyes and said that ‘this happens in ecstasy, in order to renounce once and for all the sinful world with its daily petty troubles and humdrum tasks’ (141). Other singers simply said that they put their hands or fingers to their ear or on their cheek out of habit. Clearly, these poses are traditional, and while some singers attribute some meaning to them, for others they are either an aid for hearing their own voice better or simply a habit.14 Minangkabau The pose of the Yakut epic singers with their hand on one ear is also found in other oral traditions. In West Sumatra (Minangkabau) the story about the legendary hero Anggun Nan Tunga is performed as a sung narrative by the tukang sijobang, the epic singers. Like the Yakut epic singers, the West Sumatran tukang sijobang sing through the night, and like the olonkhosuts they cup one of their ears: He sits with one knee drawn up, leaning an arm on it and cupping his ear in his hand. Most of the time the tukang sijobang keeps his eyes shut, the better to concentrate on his performance. From time to time he switches to a different tune, but the flow of words continues unbroken. Only after an hour or so does the singer break off the narrative with an appropriate pantun [quatrain] and pause for a few minutes’ rest and refreshment. (Phillips 1981: 9)

100 Performance On the dust cover of Phillips’ study a photo shows the singer Mumin with his left hand cupping his ear and his eyes closed. He seems to be listening inwardly; clearly the reverberation of the singer’s voice in his head helps him to concentrate. Tibet Like the Yakut olonkhosuts and the Kyrgyz manaschïs the epic singers of Tibet sing the epic of Gesar unaccompanied. There are exceptions, but the solo performance is the most common case. As in other traditions, it is difficult to generalize when a great number of variations exist. In the Tibetan case, the epic of Gesar has been transmitted both orally and in written form and is hence sung with and without a written text. R.-A. Stein has written a detailed study of the Tibetan bard. Since Stein’s research a great deal has been written on the Tibetan versions of Gesar and also on the musical performance of the epic (see Chapter 8). The general picture painted by Stein is, however, still valid. While the link between shaman and epic singer is tenuous in the traditions illustrated so far, this link is emphatically stressed by Stein in the Tibetan case. Interestingly, we find here the same pose as in the Yakut and Minangkabau traditions. As Stein says, ‘both Gesar and the inspired bards of Tibet are characterized by the gesture of the hand held to the ear in order to hear the voices’ (Stein 1959: 322 trans.). In other words, the singer is listening to his inner self, where he hears not his voice, but the voices of inspiration, that is, of the gods. There are various kinds of epic singers among the Tibetans, some of them entertainers similar to the medieval minstrel, but the true bard is, according to Stein, a singer who, like the Kyrgyz manaschï, received his gift of singing through supernatural inspiration and who is able to recite the epic in a state of trance: Without referring to a special case, my informant Champasangta has also told me that among the Hor [a Tibetan tribe] a herdsman, asleep in the mountains, might learn in his dream to sing the epic of Gesar. … Both informants insisted on the fact that true bards (sgruṅ-mkhan) only recite when they are in trance. The others, who can only read and recite without trance, are no true bards and less esteemed. (1959, 332 trans.) The singer shares with the shaman supernatural inspiration, but also a special piece of clothing, a peculiar hat:

Gesture  101 As is the case with certain shamans, it is the bard’s hat that allows him to sing in trance and recite correctly. Without it, without being inspired, one can learn to read or to recite the epic, but one will never be able to produce anything but fragments. (333 trans.) Stein analyses the bard’s hat and argues that it parallels the hat Gesar is wearing as a young man (then called Joru) and the hat of Padmasambhava, the eighth-century Indian missionary who founded the Nyingma school of Tibetan Buddhism. As he points out, the hat’s various components refer to specific elements of the epic.15 A similar symbolism of the bard’s clothes is found in the West African Fang mvet tradition. As Pierre Alexandre points out: ‘The performer wears a special costume, headgear, and make-up, every detail of which has a special symbolic meaning’ (1974: 6). Perhaps more interesting in the present context is the fact that the Tibetan bard carries a stick and that this stick is interpreted as a whip. Stein writes that a whip is also carried by the Tibetan shaman and that, as in Central Asian shamanism, it symbolizes the horse’s whip used in the shaman’s spiritual journey (358–59). Stein shows that there is also a connection between whip, horse, and melody (see also Chapter 8).

Gesture, Miming, Stage Props Professional narrators and entertainers have to captivate their audience. If the audience shows no interest and walks away, a performer loses his or her job. This is true at any rate in societies where there are plenty of possibilities for entertainment for those who seek it and also where the art of the professional narrator has become marginalized. This is the case in modern, urban societies. To attract an audience to the performance of traditional narratives and epics more immediately appealing forms have been developed in some societies. In Minangkabau the story of the legendary hero Anggun Nan Tunga is not only sung by the tukang sijobang but is also performed in dramatic form (called randai): In its local form, this is a drama in which the story is narrated by a singer, and the actions danced and mimed by 15–20 male actors, with musical accompaniment on pipes (pupuik) and gongs (calém­ pong). … Each troupe performs one story, or episode, and at present eleven of the 44 troupes use episodes from the story of Anggun Nan Tungga. The story is sufficiently long, and contains enough fighting and romance (which, together with comedy, are considered the essential elements of a randai plot), for eight of the eleven troupes to use a different episode each. (Phillips 1981: 5–6)

102 Performance Among the Kyrgyz, stage performances of Manas presented like musical shows have become popular, not to mention more serious adaptations such as operas. Drama has, of course, been a medial transformation of epic for many centuries, also in traditional societies. We only have to think of the Southeast Asian shadow-play (wayang kulit) based on episodes of the Mahābhārata and the Rāmāyaṇa or the theatrical adaptations of the Indian epics in various forms such as the Tamil street theatre terukkūttu or the wayang wong (human theatre) of Java and Bali.16 A first step towards drama is a narrator’s use of ‘stage props’. In some Chinese traditions of oral narrative (e.g., that of Yangzhou), the narrators have a large repertoire of gestures and many of them use a stick, a fan, or another object. Some of the gestures used by these narrators are conventional and belong to the family of gestures which were exemplified by the Karakalpak jïraw, others are stylized and of the kind illustrated with the Kyrgyz manaschï. These narrators do not perform epics, but tales, both in prose and in a mixture of prose and verse (or speaking and singing), of which many are based on classical Chinese novels such as Outlaws of the Marsh (fourteenth century), The Romance of the Three Kingdoms (fourteenth century) or The Journey to the West (sixteenth century).17 The Turkish meddah A similar figure is the Turkish meddah. In 2008 the art of the Turkish meddah was inscribed on the UNESCO Representative List of the Intangible Cultural Heritage of Humanity. As the speaker of the accompanying video on UNESCO’s website comments: ‘Meddahlık is a type of Turkish theatre performed by a single storyteller called a meddah.’ It is further stated that ‘this cherished oral literary tradition survived throughout the Turkish world during Ottoman rule and is practised nowadays across Turkey and in other Turkish-speaking lands’. On the video, illustrations from books and manuscripts are shown as well as short snippets of the performance of contemporary Turkish meddahs.18 The meddah, however, is not the familiar figure in contemporary society that the promoters of the inscription on UNESCO’s list would like him to be. This type of entertainer belongs to earlier centuries and had its heyday in the Ottoman period. The word meddah is an Arabic loanword in Turkish; maddāḥ in Arabic means ‘praise poet’ and is derived from the verb madaḥa ‘to praise’. J. Th. Zenker, in his Ottoman-Turkish dictionary of 1866, explains why the word meddah, originally meaning ‘encomiast’, is used for the storyteller by glossing meddah as ‘the public narrator of folktales and stories, so called, because he normally glorifies a king, a hero etc.’ (829). These public storytellers have repeatedly attracted the attention of European travellers in the Ottoman Empire. John Auldjo, on his visit to

Gesture  103 Constantinople in 1833, has a long diary entry in which he describes a visit to the coffee-houses in the Armenian quarter, where an improvisatore [sic] exhibits his talents every holyday. Immense crowds of respectable Turks assemble there to listen to the narrations of this accomplished story-teller; and it is even said that the Grand Signior [sic] himself is often present as an auditor in disguise. (Auldjo 1835: 121) The improvvisatore whom Auldjo mentions is no doubt a meddah, in this case apparently one of the most famous of his time. Auldjo is full of praise for the ingenious mimic’s versatility. ‘The changes, too, of voice, manner, look, gesture, suitable to the various characters he assumed, were infinitely ludicrous and entertaining’ (125). As is shown on the UNESCO video, meddahs use a kerchief and a stick, as well as different types of headgear, changing their appearance and their voice when impersonating the characters in a sketch. The med­ dah Lâ’lin Kaba (d. 1601) is represented in an illumination in just such a way, with a stick in his hand and a kerchief over his shoulder. In other pictorial sources the meddah is shown with a wooden sword and, instead of a turban, with a tasselled conical hat.19 Auldjo’s improvvisatore had a stick in his hand and was dressed in a frock coat fashionable at the time. When performing, the meddah often sat on a raised platform, as is shown in an illustration by Thomas Allom for Robert Walsh’s description of the meddah in his travelogue of 1836.20 Apart from performing sketches of a humorous, satirical, and often vulgar nature, the meddah also told traditional folktales. Ignaz Kúnos, who wrote down one of these tales from the mouth of a meddah, remarks: While in a folktale commonly fantastic events are told, the meddah always narrates possible and probable stories, mostly set in Istanbul, which as it were mark the transition from the popular to the literary tale. … Almost without exception these tales contain references to local conditions, partly in Istanbul, so that many of them are of importance also for Turkish cultural history. (1899: xvii–xviii trans.) In a footnote G. Jacob, who has edited and translated some nineteenthcentury meddah stories and sketches, notes that ‘occasionally narrators degrade themselves in front of a plain audience to become readers of chapbooks’ (1904: 6, n. 2 trans.). The meddah’s role as a reader rather than as a teller of tales is, as M. Fuad Köprülü and Pertev Naili Boratav have shown, not an occasional degradation but the original function of the meddah. From early on the term meddah was synonymous with the

104 Performance term kıssahân. The same term is also found among the Turkic peoples of Central Asia (Uzbeks and Karakalpaks) (see Chapter 1). A similar term is s¸ehnâmehân, ‘reader of the Shāh-nāma’, that is, of a Turkish translation of the Persian poet Firdausī’s Book of Kings. Readings from the Shāhnāma could still be heard in a coffee-house in Üsküdar in the second half of the nineteenth century (Boratav 1973: 75). Here then, in the kıssahân and the meddah, literacy and orality meet, as do folktale, romance and epic. By the same token, in the meddah’s oral performance of narrative, the telling of the tale is enriched by theatrical elements: miming, acting, impersonating different characters, and making use of stage props. Rajasthan: Reading the paṛ While in the oral tradition of Minangkabau there is a generic divide between sijobang, the singing of the story of Anggun Nan Tunga, and ran­ dai, its dramatic representation, we saw a certain fusion of the epic and the dramatic genres in the performance of the meddah. A further variation of the mix of narrative and dramatic forms is found in Rajasthan. When the Rajasthani bards perform the epic of Pābūjī, they not only combine song, recitation, dance, music, and gesticulation, but also use an additional pictorial narrative. The main protagonist of the epic, Pābūjī, is thought to go back to a medieval Rāt ḥ oṛ hero of western Rajasthan, probably of the early fourteenth century, the Rāt ḥ oṛ being a Rajput clan (Smith 1991: 71–88). In the story Pābūjī is the son of his father Dā̃dal Rāt ḥ oṛ and a nymph, who returns to the boy as a black mare when he is twelve years old. Pābūjī takes the mare from Lady Deval and promises to protect her from the attacks of a neighbouring clan and their leader, Khīncī (Khī̃cī), who desires the mare. The antagonism of Pābūjī and his companions towards the Khīncī clan as well as the hero’s endeavour to protect Deval form the main part of the epic’s intrigue. In the end Pābūjī loses his life – he generously gives his sword to Khīncī in a duel, because only his own weapon can harm him. His death is, however, avenged by his son Rūpnāth.21 Pābūjī is worshipped as a god; the bard is hence not only the story’s narrator but also the god’s priest. The recital of the epic is both an act of devotion and an all-night entertainment. It begins with a sacrificial offering to the deity (puja), which is followed by a hymn dedicated to Pābūjī (arthi).22 Only then is the epic itself performed. The most outstanding element of this performance is what might at first sight be described as a stage prop, that is, a painted scroll on which protagonists and events of the epic are represented. This scroll is called paṛ in the Rajasthani language (a language closely related to Hindi)23 and the reciting and

Gesture  105 narrating of the epic is referred to as ‘reading the paṛ’. This scroll is rolled up during daytime and only unrolled at night during a performance. It is not only a visual aid for the audience, but is seen as the god Pābūjī’s temple. The singer-priest, called bhopo (or bhopa), performs together with a female partner, generally his wife, the bhopī. John Smith, who has transcribed and translated a version of the epic recorded from Parbū Bhopo, describes the performance situation as follows: The singing of Pābūjī’s epic may constitute a religious liturgy, but it is a fairly informal and cheerful event. The audience does not sit still or maintain a devout silence. In the course of performance the bhopo may crack jokes or make deliberate mistakes, and in return he may be chaffed by his listeners. He sings with his wife (bhopī) to the accompaniment of a spike-fiddle (rāvaṇhattho); his wife holds an oil-lamp to illuminate details of the painted paṛ in the darkness. In between songs he declaims the narrative in a vigorous chant; during some songs he will dance, the bells round his ankles jingling and his red robe swirling about him. There is no doubt that the p ­ erformance – the ‘reading of the paṛ’, as it is called – is an entertainment as well as a religious observance. (1991: 8) Singing, narrating, playing, dancing, and visualizing by showing all combine in this performance. The role of images in the performance of oral epics will be the subject of a later chapter (see Chapter 10), when I will say more about the pictorial programme of the paṛ. As will be seen, the performance of the epic of Pābūjī is not the only case of the incorporation of pictorial material into the performance of narrative. ***** Gesture and gesticulation are only possible if a singer-narrator is present. They are an aspect of live performance and absent in a written text. It is the singer’s voice that materializes the words and it is the singer’s gestures that give them visual accompaniment. This can be of various kinds: underlining the meaning of the words by iconic and imitative gestures; dramatizing the action by pantomimic gestures; rhythmically supporting the recitation by ritualized gestures and others. We find a great variety in different oral traditions and often also among different singers of the same tradition, with frequent and diverse gestures at one end of the scale and an almost total absence of gesture and gesticulation at the other end. Gesturing can involve the use of stage props, and the extension of stage props to images and various kinds of figures can lead to an increased emphasis on visualization, but also to a genre change from epic to drama.

106 Performance

Notes 1 Wilson’s quote refers to Cicero’s De oratore, Book III, Chapter lix. On gesture in classical antiquity, see Graf 1991. 2 For an English translation of Wundt’s discussion of gesture, see Wundt 1973; for an English translation of Wundt’s shortened version of his Völkerpsychologie, see Wundt 1916. 3 Kendon 1992: 179. For a general introduction, see Kendon 2004; for a survey of the literature on the subject up to the mid-1990s, see Kendon 1997. See also Haviland 2004 (anthropological linguistics); Bremmer & Roodenburg 1991 (cultural history of gesture). John Benjamins (Amsterdam) publishes the journal Gesture (first issue in 2001) and the series ‘Gesture Studies’, of which volume 8 appeared in 2020. 4 This story (Motif J1142.1) is often found together with the Solomonic judgment illustrated in Chapter 4; see Bin-Gorion 1920: 153–55. 5 Her text is based on the first edition of Jüsüp Mamay’s version of Manas of 1984, with omissions and changes; I have translated the definitive 2004 edition of Jüsüp Mamay’s version, of which (due to the present situation in Xinjiang) only two volumes have been published in China (Reichl 2014– 2015); this passage (ll. 6935–7045) is found in volume 1 on pp. 200–3 (translation) and 483–86 (text). For the translation, see Appendix B. 6 In the following, I have given a fairly simple description, geared to the case analysed. For a more sophisticated classificatory system and a more detailed analysis of gesture, see the comprehensive study by Geneviève Calbris (2011). 7 More gestural variants can be detected such as: (5) The hitting or throwing movement is not directed toward the body but away from it. (6) The right arm makes a similar hitting movement, but from left to right, starting from the left shoulder. 8 To study the gesticulation of Kyrgyz manaschïs, a number of videos are available online. The Aigine Cultural Research Center has filmed a set of 124 episodes of the three canonical parts of the Manas, which can be accessed via the centre’s website They have also been uploaded on YouTube. 9 For illustrations of Qaralaev’s gestures, see Qarïpqulov et al. 1995: I, 148–49. 10 Personal communication, 27 April 2018 (trans.). Talantaaly Bakchiev enlarged on this story somewhat in a later communication (3 May 2018). 11 I am grateful to Alymkan Jeenbekova for this comment (21 May 2018). On the Karakalpak shaman, see Somfai Kara & Ämirlan 2020. 12 On the Kyrgyz shaman, see Moldobaev 2001; on neo-shamanism among the Kyrgyz, see Penkala-Gawęcka 2014. 13 Personal communication, 5 May 2018. 14 The Yakut epic singers’ musical performance will be discussed below; see Chapter 8. 15 Stein 1959: 342–58; for a symbolic interpretation of the bard’s hat, see pp. 356–57. 16 On these forms, see, inter alia, Osnes 2010 and Frasca 1990. 17 See Børdahl 1999 and Børdahl & Ross 2002. 18 See, click ‘Arts of the Mehhah, public storytellers’ under 2008. 19 For illustrations, see the plates in Nutku 1997, following p. 360. 20 See Walsh 1836–38, plate opposite II, 71. 21 For a summary, see Blackburn et al. 1989: 240–43; Smith 1991: 9–12. 22 For the translation of such a hymn, see Wickett 2010: 13. 23 Or phad according to Wickett 2010 (and others).

6 Oral Epics as Songs

Μοῦσ’ ἄρ’ ἀοιδὸν ἀνῆκεν ἀειδέμεναι κλέα ἀνδρῶν The Muse impelled the singer to sing the glorious deeds of men. (Odyssey VIII.73) After Odysseus, shipwrecked at the coast of the Phaeacians, was hospitably received by Nausicaa and her father, King Alcinous, a feast is held in his honour. The blind bard Demodocus is summoned to perform at the gathering. When Demodocus is impelled by the Muse to sing, he begins with an episode from the Trojan War and later, at Odysseus’ suggestion, recites the story of the Wooden Horse. Odysseus, who had not yet revealed his identity, sheds tears at each performance and finally discloses who he is. In the Odyssey Demodocus is called an aoidós (ἀοιδός), literally ‘singer’, a word etymologically derived from the verb aeídō (ἀείδω) ‘to sing’. That Demodocus does indeed sing is implied by the repeated use of the verb aeídō and by his playing a musical instrument, the phórminx (φόρμιγξ), a type of lyre (VIII.266). Demodocus is not the only aoidós in the Odyssey. In book I we read that a singer by the name of Phemius sings to Penelope’s suitors in Ithaca: at their demand, the herald gives Phemius ‘the most beautiful lyre’ and the singer begins to sing (I.154–56). What Phemius performs is also an episode from the Trojan War (I.325–27). We can assume that these scenes give a reasonably faithful picture of epic performance in the archaic Greek world. In Greek terminology there is a contrast between the aoidós as represented by Demodocus and Phemius in the Odyssey and the rhapsode (rhapsōdós, ῥαψῳδός), the professional reciter of the Homeric and other epics of a later time. This distinction has been much discussed by Classicists, a discussion that concerns also the relationship between orality and literacy in ancient Greece and the formation and transmission of the Homeric epics. These are

108 Performance wide-ranging and controversial questions, which need not be broached here. In the present context it must suffice to remind us that the Greek bard as he is depicted in the Odyssey is a singer and that he sings narrative poetry.1 This tallies with the performance mode of oral epics world-wide. By far the most common way in which an epic is narrated is by singing it. While we have no detailed musical evidence for reconstructing the sound of the Greek bard’s song with a sufficient measure of reliability, there is a rich harvest of epic songs that has been collected from many oral traditions. Surprisingly, this treasure trove is only rarely tapped by epic scholars. Even ethnomusicologists often find more rewarding areas of research than the music of oral epics. In this book only a few gems from this treasure trove, many uncut, can be singled out for closer inspection.2 The music of epic performance can be studied from different points of view. I propose to distinguish between three aspects and corresponding sets of questions. The first concerns the description and analysis of the music, the second the relationship between words and music, and the third the problem of interpretation, that is, the question of what the contribution of music to the overall meaning of an epic is. The first will be the subject of this and the following chapter, the second the subject of Chapter 8 and the third will be discussed in Chapter 10. What first strikes the student of epic music is its diversity. We find bards singing without an instrument and accompanied by a musical instrument; we find bards performing on their own and those accompanied by other musicians. We also find a large quantity of musical styles, with regard to the quality and range of the voice, the rhythmic and melodic patterning of the sung words and the interplay of singing and playing a musical instrument. To do justice to the music of epic, more than one volume would be necessary, as well as the collaboration of a group of specialists. In this and the following chapters I will have to limit my discussion to what are in my view some of the most basic issues, which I will illustrate with a small selection of representative examples. The core of my examples comes from my own research in the Turkic-speaking areas of Central Asia and Siberia; I will, however, also discuss parallels and variants in other traditions. In Chapter 4, the distinction between speaking and singing was briefly addressed. While speaking in an everyday voice on the one hand and singing melodiously on the other can be clearly set apart, there is an intermediary zone where these two forms of utterance cannot always be clearly distinguished. George List, in a paper on the boundary of speech and song, based on ethnomusicological material, defined song as ‘a form exhibiting relatively stable pitches, possessing a scalar structure at least as elaborate as the heptatonic, and showing little, if any, influence melodically of speech intonation’ (1963: 3). The latter is true of languages like English, but might not be true of tone languages, languages where syllables have different tones (rising, falling, high, low etc.), features that

Oral Epics as Songs  109 are generally phonemic, that is, differentiate meaning. George Herzog has looked at the relationship between the contours of tone languages such as Jabo in Liberia and Navaho in North America when spoken and when sung and has come to the conclusion that ‘a slavish following of speechmelody by musical melody is not implied. Rather, the songs illustrate a constant conflict and accommodation between musical tendencies and the curves traced by the speech-tones of the song-text’ (1934: 466). H.-H. Wängler (1958) arrived at similar conclusions in his study of singing and speaking in Hausa, an African tone language. He noted that a singing mode close to speaking follows the sequence of tones, while a more melodious singing style is independent of the tones. As to the intermediary zone between speaking and singing, it can be argued that however close ‘spoken music’ is to declaiming, it follows nevertheless the dictates of music, is embedded in musical sound, and is an approximation of music to speech, an imitation of speech in music. As will be seen, it is a form of singing rather than speaking. When it comes to the analysis of epic music, a number of further distinctions will be helpful. Most of these distinctions are the two endpoints of a spectrum, with several positions in-between. There is (1) the distinction between speech-like and song-like melodies; (2) some epic melodies are stichic, some are strophic. ‘Stichic’ means that every line (or sometimes a group of two or three lines) of the poetry has the same melody, often with small variations. ‘Strophic’ means that a poetic unit such as a stanza has a different melody for every verse-line; (3) a melody can virtually be sung on one note (monotonic) or range over a wide ambitus (with various intervallic steps); (4) the epic can be sung solo or with musical accompaniment. In the latter case a range of instruments can be used, either played by the singer or, in addition to the singer, by one or more accompanying musicians. In this chapter I discuss several cases of unaccompanied epic performance, focusing on the singing of the Kyrgyz epic Manas. I look at the role of monotonic melodies, at melodic variation and alternation, and at the relationship between music and metre. I would like to repeat at this point what I said in the introduction: My discussion is primarily directed at readers who have a certain interest in music, without being specialists. As far as musicological terms are used, they are explained in the glossary at the end of the book. Occasionally a more detailed musical analysis is called for. I will try to keep the analysis clear and will summarize the gist in non-technical language. In the following, some notated musical examples will be presented. The notation of ethnic music has been much discussed by m ­ usicologists. Many factors have to be considered: pitch, duration, accentuation, and the ­ various musical performance characteristics indicated in Western ­notation by expression marks, such as tempo, degrees of ­loudness, p ­ hrasing, and many others. Also the voice quality of some

110 Performance epic performance styles would have to be part of an adequate musical transcription. Musicologists have developed a set of symbols to capture some of the characteristics of ­ethnic music, such as variations in pitch, uncertainties in assigning a definite pitch to a note, fluctuations in the duration of notes, dynamically differing notes and so on. For notation, scholars have chosen different solutions. Béla Bartók, for instance, has provided detailed notational representations of some samples of South Slavic oral epics (1954). Many ethnomusicologists, however, use what in phonetics is called ‘broad transcription’, that is, the representation of phonemes rather than allophones. This is to say, for instance, that a melody is notated as a sequence of pitches, each with a specific duration, without indicating minimal fluctuations in pitch and duration precisely. Signs are added, but these indicate only approximate changes in pitch and duration. Stephen Erdely’s transcription of the melodies of South Slavic epics is of this kind (1995). Another alternative is to dispense with transcriptions altogether and to describe and characterize the quality of a piece of music, possibly supported by spectrograms and audio files. With the development of modern forms of music, such as for instance electronic music, the limitations of the notational system developed for Western (classical) music has become an issue not only for non-Western, but also for avant-garde Western music.3

Song as Vehicle, Song as Music Vendors of old-style European markets often advertise their wares by chanting rather than shouting out their offers. The English Renaissance composer Orlando Gibbons (1583–1625) made use of the musical potential of these chanted words in his ‘The cries of London’, a consort song for five voices and five viols. ‘New mussels, new lilywhite mussels! New cockles, new great cockles! New great sprats, new!’ call out the first voices in a long list of vendors, advertising their wares. Gibbons wasn’t the first to compose music on the sing-song of vendors and hawkers. A hundred years before him the French composer Clément Janequin (c. 1485–1558) had the same idea, in a vocal work appropriately called ‘Les cris de Paris’, in which among other delicacies prunes of Saint Julien and sweet chestnuts from Lyon tempt the buyers. In these compositions, music is in the foreground, on the market, however, the singing is simply used as a vehicle for carrying the words. In this use of song, the vendor of wares is similar to the teller of tales. Also the bard’s words have to be audible and have to reach the audience. Curt Sachs speaks in this connection of the domination of the words over the music. He sees this domination in ‘the unbelievably simple recitation of national epics’ and points out that these simple ‘melodic scraps can be repeated for hours and hours without annoying a listener exclusively interested in

Oral Epics as Songs  111 the narrative’ (1962: 78). Other musicologists have made similar observations. Kurt Bose, a specialist for the music of Finnish oral poetry, sees the length of oral epics as the reason for the comparative simplicity of their melodies: The singer’s intellectual energy is largely engaged in moulding the text; he has little strength left for the deployment of a copious musical scheme. The tunes comprise simple, short strophes constructed of few, but pithy motives. Their musical performance, however, is very lively and complex. The copiousness of the text, spontaneously composed in constantly recurring short strophes, must hold the listener’s attention more by the art of performance than by the charm of the melodies. (1958: 29–30 trans.) Sachs lists among his examples of text-dominated music ‘the archaic cantillation of Persian and Yemenite Jews’. The chanting of sacred texts is indeed a close parallel to the musically simplest types of epic performance. Various religious traditions have developed elaborate musical styles, but especially the recitation of non-poetic narrative texts, at least in Judaism and Christianity (Gregorian chant), has retained a simpler, speech-oriented musical style. For this type of singing the term ‘chant’ is commonly used. The distinction between ‘chanting’ and ‘singing’ is terminologically expressed not only in English, but also in distant cultures like that of the Maori of New Zealand, who posit ‘(ritual) chant’ as a separate category (karakia) besides ‘speech’ (koorero) and ‘song’ (waiata) (List 1963: 5–6). In English the verb to chant has, according to the OED, a wide array of meanings, of which the sense ‘to sing (a song, esp. a repetitive one) in a monotone, or with a prolonged intonation’ is the one intended here. Appropriately the OED lists among the records for this meaning a quote from a book on world music.4 An extreme form of melodic simplicity is a melody sung on one note, a literally monotonous tune. Such tunes, with minimal variations above or below this single tone are typical of speech-oriented chanting. Carl Stumpf, one of the pioneers of ethnomusicology, saw a first step in musical evolution in what he called ‘speech-song’ (German Sprachgesang). He defined speech-song as ‘reciting and declaiming in such a way that the voice dwells longer than usual on specific notes, reciting whole sentences on an invariable pitch and making use of musical intervals at specific points’ (1911: 49–50 trans.). Stumpf underlines that speech-melodies are melodies, however close they might be to the intonational patterns of speech. Such melodies are also called ‘recitative’, characterized in the Harvard Dictionary of Music as follows: ‘In the recitative, the purely musical principles of vocal melody, phrase, and rhythm are largely

112 Performance disregarded, being replaced by speechlike reiteration of the same note, slight inflections, irregular rhythms, purely syllabic treatment of the text, etc.’ (HDM: 718). A further term employed in this context is parlando, generally used as an expression mark in notated music and defined as ‘an indication that the voice must approximate speech; in a sense, “spoken music” …’ (HDM: 643). Monotonic Narrative Melodies – Papua New Guinea Curt Sachs gives as an example of ‘the unbelievably simple recitation of national epics’ the monotonic melody of a narrative performed by a Georgian mestvire, a professional bagpipe-player, who among other genres also performed heroic narratives. His musical example consists of an indefinite sequence of the same note (B) with a final descent (to G) (1962: 78):

Example 6.1  Monotonic melody

Fritz Bose, in his introduction to ethnomusicology, gives a shamanistic recitative from western New Guinea as an example of the simplest type of melody, a monotonic, one-motive recitative tune (1953: 165). In Papua New Guinea sung narratives are also found. They have been studied by both linguists and ethnomusicologists and offer a glimpse of an archaic world of narrative music (see Rumsey & Niles 2011). These sung narratives are found in the eastern part of the island of New Guinea, more precisely in the central highlands. Several language communities cultivate these sung narratives; this means that not only the tales themselves but also the native genre terms are in different languages. The common denominator of these works of oral tradition is that they are narratives and that they are sung. It can be argued that these sung tales share a number of characteristics with oral epics and can be meaningfully compared to the epic poetry of other oral traditions. In the present context, however, I will leave questions of genre aside and confine my remarks to a small sample of melodies. One of the ethnic and linguistic groups of the Papua New Guinea highlands are the Huli. Their sung narratives are called bì té: ‘Bì té (bì = words, talk, talking; té = story) are long fireside stories, told in the houses of both men and women to entertain children and other relatives during the night hours’ (Pugh-Kitingan 2011: 111). The stories are performed in singing or chanting by a recognized storyteller. They are ‘rambling solo prose forms that sometimes include poetic episodes’ (113). The melody sounds

Oral Epics as Songs  113 monotonic, but moves in fact within the interval of a third. It comprises three pitches, a central pitch, a tone above the middle note and a half-tone (or tone) below. Pugh-Kitingan’s notation of the beginning of a bì té gives a melody that swings between A (central pitch), B and G sharp:

Example 6.2  Papua New Guinea

This melodic patterning continues for the rest of the transcription. A similar melodic shape (with a different pitch) is found in Pugh-Kitingan’s second example.5 Just as Papua New Guinea can boast a profusion of languages (c. 850), so do the sung tales of the highlands comprise a wide spectrum of narrative genres and performance styles. While according to Jacqueline Pugh-Kitingan among the Huli ‘melody is generally dominated by speechtone’, Don Niles shows that the sung tales of the Hagen, called kang rom, have a metric structure that can be captured by regular bars. Some of the melodies transcribed have an ambitus of a sixth and differ from other melodies discussed in this book, which often only comprise a third (Niles 2011). It turns out on closer analysis that the characterization of the music of oral epics and narratives as monotonic, and hence monotonous, cannot be upheld for the sung tales of the Papua New Guinea highlands. This is also true of other epic traditions. When oral epics are performed without the accompaniment of a musical instrument, this seems at first sight to stress the importance of the word over the music. It will be seen, however, that the role of music is far weightier than that of a mere vehicle for the audible utterance of the words. Solo Singing Unaccompanied singing of epics and narrative poetry as practised in Papua New Guinea is found in a large group of oral traditions. Sometimes solo singing is the only mode of musical performance, sometimes it exists side by side with accompanied singing; sometimes it is

114 Performance found only in regional traditions or with regard to a particular type of epic. Every tradition with solo performance has its own character, with historical, regional, and personal variations. A few examples will suffice to illustrate this variety. The Russian bylinas are performed without instrumental accompaniment. Solo performance is typical of the northern bylina tradition, that is, of the bylinas collected in a belt stretching roughly from the Lake Onega region along the shores of the White Sea to the basin of the river Pechora. Typically, the melodies are stichic; they vary according to the length of the verse-line, have a small ambitus, and are composed of formulaic motifs. A simple example comes from one of the great nineteenth-century bylina collectors, P. N. Rybnikov6:

Example 6.3  Bylina. [Ay, there lives Sadko the merchant, the rich guest/ He was sleeping in the dark night.]

The melody spans two lines and consists of two parts (musical phrases). It is syllabic and adapted to the length of the line, including an upbeat on the exclamatory Ay at the beginning. The tonal centres are C and B, that is, most syllables are sung on these notes. On account of the descending steps of the final notes of each musical phrase, the ambitus is somewhat wider than that of strictly monotonic tunes; it comprises a fifth. The Finnish-Karelian epic rune-songs are as a rule sung solo; sometimes, however, the singing is accompanied by the kantele, a plucked zither, traditionally with five strings. Solo singing is also characteristic of other Finno-Ugric oral traditions, such as the performance of the oral epics of the Komi and of the Nenets of arctic Russia.7 In most of the oral traditions of the Mongolian peoples the oral epic is performed to the accompaniment of a musical instrument, typically the horse-head fiddle (morin khuur). In some traditions, however, the epic is sung solo, as, for instance, among the western Buryats or EkhiritBulagat.8 In several of the traditions discussed in Chapters 4 and 5 the oral epics are performed without musical accompaniment. This includes the performance of the Tibetan Gesar, of the oral epics of the Tuvans in Central Asia and of the Minangkabau of Western Sumatra. Nigel Phillips remarks that the tukang sijobang, the epic singer of the Minangkabau, generally performs without musical accompaniment and that only a small

Oral Epics as Songs  115 minority of singers accompany themselves on a kucapi, a board-zither, played with a plectrum. Although the typical Minangkabau epic singer sings unaccompanied, ‘he taps out the rhythm on the floor-mat with a half-full box of matches’ (Phillips 1981: 9). Clearly there is a need for the singer to mark the rhythm, that is, to perform one of the functions of a musical accompaniment (see Chapter 7, p. 145). A similar performance mode is reported from the Ainu of the Japanese island of Hokkaido. The oral epics of the Ainu, whose language is now extinct, were also chanted without musical accompaniment. The singer-narrator did, however, mark time with a block of wood, with which he or she tapped the hearth frame (Philippi 1979: 25). In a wide group of Siberian traditions, including the Yakuts, epics are sung unaccompanied. They have a number of interesting musical traits, which will be discussed in Chapter 8. Among the peoples speaking a Turkic language the Bashkirs and the Tatars also sing the verse parts of their prosimetric tales unaccompanied. Many of these narratives are, however, fairly short and close to the folktale. Choral Singing The oldest form of musical performance of the bylina is found in the north. There is, however, also a younger, southern or Cossack tradition in the Don region, where a choral singing of bylinas is usual. ‘Choral singing’ can mean the presence of a group of singers who sing a passage or a whole verse narrative together with, or in alternation with, the epic singer and it can mean that the audience participates by singing specific passages. The Cossack tradition is of the former kind. The bylinas and bylina-songs (a term used for narrative songs combining epic and lyric motifs) are sung by a group of singers in a harmonic polyphonic style. The songs generally begin with one singer singing an opening musical phrase or verse-line, before the other voices join in; this alternation of lead singer and group singing continues for every verse.9 Choral singing of this kind presupposes an essentially fixed text, known by all the members of the choir. Similar kinds of choral performance of epic poetry are found in the Caucasus. In various oral traditions of the northern Caucasus the Nart legends are found. These legends are a cycle of tales (often in prose, in some traditions in verse, in others in a mixture of prose and verse) that originated among the Ossets, the descendants of the Iranians of the steppe (Scyths, Sarmates, Alans), who today live in northern and southern Ossetia in the Caucasus. These legends have spread to a number of peoples speaking North-Caucasian languages such as Abkhaz, Adyge, Kabard-Cherkes, Chechen, and Ingush, as well as to the Turkic-speaking Karachay-Balkars. The Karachays and the Balkars share the same literary language. The older traditional style of performing the Nart epic tales among the Ossets is singing to the accompaniment of a fiddle (see Chapter 7). Under the influence of other Caucasian musical traditions,

116 Performance the Ossetian Nart cycle is also performed chorally. The solo-singer sings the words of the song in a recitative style, while the choir provides a harmonic background by singing a sequence of vowels in the various voices. A similar situation characterizes the epic performance of the Karachays and Balkars. We find different performance modes: solo performance, with or without rhythmical clapping, singing with instrumental accompaniment, with the refrains played on the sïbïzghï, a type of reed-pipe, and choral performance, where the poem is sung antiphonally with solo singing and choral response.10

Example 6.4  Karachay. [Shaway came riding (lit. ambling), Shaway arrived in the morning]

In Africa, the Fang mvet epic recorded by H. Pepper (1972) is performed by a singer who enters in the course of the performance into verbal exchanges with two assistants and also with a choir. As text and translation of this edition show and the extracts on the accompanying LP illustrate, the choir sings short interjections (e.g., Éééeeeéééeee) and repetitions of the singer’s phrases. Choir and/or audience participation in the singing is also found in other African traditions. According to Daniel Biebuyck the Banyanga epic of Mwindo is performed by the singer-narrator, who shakes a calabash rattle and whose recital is accompanied by three young men with percussion sticks. In addition, ‘the percussionists and members of the audience sing the refrains of the songs or repeat a whole sentence during each short pause made by the bard. … Members of the audience also encourage the reciter with short exclamations (including onomatopoeia) and handclapping or whooping’ (Biebuyck & Mateene 1971: 13). Similarly, the performance of a Kpelle epic from West Africa includes the narrator, two instrumentalists and a choir, composed of members of the audience, which sings continuously in the background (Stone 1988: 2).

‘Riding the Song’: The Singing of the Kyrgyz Epic Manas Unlike the dastans of the Uzbeks, the epics of the Manas-cycle contain no prose passages; they are verse epics. They are also very long; from some

Oral Epics as Songs  117 singers we have versions of the cycle comprising over 200,000 verse-lines (Jüsüp Mamay), and even in one case over 500,000 verse-lines (Sayaqbay Qaralaev). The Kyrgyz manaschï uses principally two melodies, or rather melody types, for the performance of the epic. Wilhelm Radloff commented on these as follows: In performance the singer always uses two melodies, one, sung in a faster tempo, for the narration of events, and the other, in a slow tempo and sung as a solemn recitative, for dialogues. I was able to observe this change of melodies in the performance of all tolerably competent singers. (1885: xvi trans.) Radloff’s comments hold true also for performances of the epic of which we have audio recordings or which can still be heard today. His observations, however, have to be somewhat expanded and qualified. Radloff’s first melody, ‘Melody A’, has a recitative-like melodic line, sung for the most part on one note with some occasional small intervallic steps up or down. His second melody, ‘Melody B’, is fairly melodious, with thirds and fourths as the most common intervals. When verses are sung in a fast tempo to Melody A, the Kyrgyz call this jeldirme. This word designates in both Kyrgyz and Kazakh a recitative-like fast melody. It is derived from the verb jeldir- ‘to put a horse into a trot’, which in turn is derived from jel ‘wind’. When verses are sung in a slow tempo to Melody B, this is called jorgho söz, literally ‘ambling speech’. A jorgho or jorgho at is an ambling horse. The word is found in a great number of Turkic languages; the word for ‘ambling’ in Mongolian (Khalkha) is also etymologically related. Roberte Hamayon writes about this word: ‘The Mongols’ concept of rhythm does not seem inherent in that of music; the rhythm reflects the pace of the “mount instrument”, being frequently designated by the term zhoroo (“to amble”)’ (1980: 483). She also points out that Mongolian instruments are generally conceived of as ‘mounts’ (ibid.). This ties in with my earlier remarks about the horse symbolism of qobïz and morin khuur, the horse-head fiddle, and their functional equivalence to the shaman’s drum, at least historically. Interestingly the singer Jumabay-jïraw conceives of his instrument as an animal with a horse’s head and a wolf’s snout (the bottom part of his qobïz). Similar ideas underlie the term rta ‘horse’ for ‘melody’ in Tibetan (see Chapter 8). The horse symbolism applied to musical instruments and their parts, often also expressed in the decorations of the instrument, is wide-spread and probably a legacy from the past.11 I will begin with Radloff’s second melody, which will be illustrated with a transcription of the first ten lines of the epic in a recording I made of Jüsüp Mamay in Ürümchi, 1985.12

118 Performance

Illustration 12  Jüsüp Mamay (c. 2000)

The text is the following:





Ey … ! Aytayïn bir az qanqordu, Anïn arbaqtarï qoldoso. Menim aytqanïm jalghan bolboso. Jarïmï tögün, jarïmï chïn, Jarandardïn köönü üchün. Janïnda turghan kishi joq, Munun jalghanï menen ishi joq. Kök jalïngday kishi joq, Közüng ötüp ketkinche Dele qorqush menen ishi joq. Ey … ! I will tell a little about the brave hero! May his spirit give help! May what I am telling be no lie! Half of it is fabled, half of it is true: May it please the hearts of my friends! No living man was at his side; There is no place for truthlessness. There is no man like the hero! Until your eyes close in death, There is absolutely no dealing with fear.

The singer wanted to hold the edition of his version in his hands while singing, but actually sang a slightly different text!13 The melody the singer sang has the following form in ‘broad transcription’ (see Audio/Video Ex. ):

Oral Epics as Songs  119

Example 6.5  Manas (1)

The epic begins with a long drawn-out exclamation (Ey!), sung on a single note (B flat), which ends in a glissando down to the first note of the melody proper (G). The tempo of the melody is fairly fast. The melody is notated in bars of six quavers (6/8).14 Outside the bar structures lie the initial note (with a fermata) and the beginnings of lines 2, 3, 7, and 10. I will comment on the latter presently. What is important to

120 Performance recognize is that we have a ‘measured’ melody, that is, a melody consisting of equal bars, in other words, a sequence of tonal-rhythmic units of equal duration. As noted in the case of Karakalpak epics, the metre of Turkic oral poetry is syllabic. In Kyrgyz, the epic metre is a 7 or 8-syllable verse-line. A metrically regular line is composed of two feet, the first comprising four or five syllables, the second three syllables. This is mirrored in the music. The melody is syllabic, that is, we have one note per syllable. Every regular verse-line corresponds to a musical line that comprises two bars with four or five notes in the first bar and three notes in the second bar. In singing the syllables are weighted, that is, we have an alternation of crotchets and quavers, which are arranged in figures of three beats. For a metrically regular 7-syllable line (l. 8), we have the following pattern:

Example 6.6  Manas (2)

And for a metrically regular 8-syllable line (l. 6), we have the following pattern:

Example 6.7  Manas (3)

Extra syllables can be incorporated by singing a crotchet as two quavers, as in the second bar of line 4:

Example 6.8  Manas (4)

In lines 2, 3, 7 and 10 we have two extra syllables at the beginning of the verse-line, metrically an anacrusis, which is rendered musically by an

Oral Epics as Songs  121 upbeat. The singer introduced the extra syllables in lines 3, 7 and 10 in performance; in the edited text they are missing. When we look at the melody more closely, we notice that all the lines have a similar shape. The most striking feature of the melody is the last three notes of each line. We have a descent, either in two thirds (lines 1, 2, 4, 5, 8, 9) or in a sequence of third and fourth (lines 3, 6, 7). Only the last line is exceptional in that the descent stops just one tone down from G. The melodic line before the final cadence has different starting points (G in most cases, also B flat and C) and goes either one or more steps up (lines 2, 3, 6) or down (lines 1, 4, 5, 7, 8, 9); the last line, while staying within the same ambitus, has a slightly varied melodic shape. The characteristics of the musical performance of Manas described so far can be summarized in five points. (1) The melody consists of musical measures (bars), which have a regular sequence of beats. The measured character of this melody is by no means a universal characteristic of the music of epic. Many epic melodies have either no regular sequence of bars or no bars at all. Their melodies are free flowing, patterned into various rhythmic groups, without the regularity of segmentation into specific measures (see below). (2) Every note corresponds to a syllable of the text. The melody is hence syllabic and not melismatic. In a melisma a syllable is sung to several notes; free flowing melodies are generally melismatic. (3) The melody is stichic. Stichic melodies are typical of a wide range of epic music and are generally thought to be the epic melody par excellence.15 Many epic melodies are, however, strophic. Not surprisingly, this is most often the case when the verse is arranged in stanzas (see Chapters 7 and 8 for examples). (4) The melody has a sinuous contour. Various methods of describing and categorizing tunes by the contour have been proposed by musicologists. Some melodies are characterized by a step-by-step movement, up or down, others by a wider, pendulum-like rise and fall of the tune; other labels proposed are ‘arched’, ‘undulating’, ‘terraced’, and ‘descending’. Also, some melodies are composed of only one musical motif, which is repeated over and over, while other melodies have several motifs.16 (5) The hallmark of the ‘Manas melody’ is its final cadence. Although the recordings of different singers reveal a variety of melodic shapes, they can all be recognized as the ‘Manas melody’ on account of the melodic contour of the endings of most lines. In actual performance the elements of this melodic pattern are combined with other patterns of a similar physiognomy, without, however, losing the distinctive character of this melodic pattern. The ‘Manas melody’ is considered the signature tune of the epic. As the Kyrgyz musicologist K. Dyushaliev writes:

122 Performance This very simple little tune can be called metaphorically the ‘leitmotif’ of the epic, a tune that has a truly symbolic meaning for the Kyrgyz. It is indeed so popular and so well known that almost all Kyrgyz can recognize it correctly on hearing it and can sing it by heart. (1993: 180 trans.) In Manas there are no regular stanzas with regular rhyme patterns. Nevertheless the verses are segmented into sections and do not form a continuous, uninterrupted sequence of lines. In performance an irregular number of lines are sung to the ‘Manas melody’ and then concluded with a slightly different melody. These sections of irregular length, which are only rarely indicated in editions of the text, are generally held together by one or more rhyme patterns and also by verse-initial alliterative patterns. Neither rhyme nor alliteration is regular, which allows for a certain flexibility in the segmentation of the text into sections. In performance the end of a section is melodically marked, as in our example, where in line 10 the sixth syllable has a fermata. There is some similarity between these sections in Manas and the laisses (‘tirades’) in the Old French chansons de geste.17 Radloff’s first melody, the ‘narrative melody’ (Melody A or jeldirme), takes the form of a fast sequence of syllables mostly sung on one note, with a stress on the fifth syllable and a frequent rise in the middle or at the end of the line. Every syllable/note has the same length; there is no rhythmic patterning as in the ‘Manas melody’; the tempo is flexible and flowing, with no division into bars. V. Krivonosov calls it a melody of recitative character, sung rubato (1939), that is, in ‘an elastic, flexible tempo involving slight accelerandos and ritardandos that alternate according to the requirements of musical expression’ (HDM: 742). With its prestissimo tempo, it has a pressing, insistent character. It is closest to what has been called a monotonic melody in other traditions. In our example, Jüsüp Mamay changes after twenty-five lines from Melody B to Melody A, which he sings for fourteen lines, before he switches back to Melody B. Melody A begins:

Example 6.9  Manas (5). [The words inherited from our forefathers,/ The heroic words, bringing defeat to all]

It has to be stressed that Melody A and Melody B come in a number variants. Furthermore, Radloff’s parallelism between melody and content

Oral Epics as Songs  123 is not as neat as he stated. The narration of events is not only performed by using the narrative melody, Melody A, but can be performed also by using Melody B, the ‘Manas melody’. By the same token dialogues are not only sung to Melody B, but can also be sung to Melody A. Some singers use Melody A for long stretches, sometimes sung on just one note with a stress and a slight rise at the end of the line, sometimes also merging into declaiming for a short space. The ‘Manas melody’ is generally recognizable by its final cadence; it does, however, come in many variants and not all have a clear cadence. To give just one example: The Kyrgyz woman manaschï Döölötqan, whose gestures were discussed in the previous chapter, performed the whole passage of 123 lines in her variant of the ‘Manas melody’. This melody spans two lines and is repeated throughout the passage; the second line is occasionally repeated:

Example 6.10  Manas (6). [He [Manas] had mounted Aqqula and looked very handsome./ He was as huge as the Ala-Too. (lines 1–2)]

The passage consists of six sections, which means that the last line of each section (and at the very end of the passage the last two lines) is sung to a rhythmically freer tune, ending with a final exclamation (höy):

Example 6.11  Manas (7). [The noise of the fight resounded. Höy! (line 14)]

Every singer, Radloff wrote, has his own melody. This is true in the sense that every singer has his or her own variant. There is, however, a certain family likeness between these melodies which characterizes

124 Performance them as belonging to the performance of Manas. They are typical of the whole cycle, but here again variations and individual preferences occur. Döölötqan, for instance, sang an excerpt from the second part of the cycle, Semetey, with a different melody. On the other hand, the extract from Semetey I recorded from the Kyrgyz singer Abdurahman Düney in the Chinese Pamirs in 1985 was sung to the ‘Manas melody’ as exemplified above. A number of scholars have analysed the musical side of the manaschï’s performance. The Russian musicologist Viktor Vinogradov, who has studied the performance style of Sayaqbay Qaralaev, distinguishes four melodic patterns, which he calls recitatives. The first corresponds to a freely flowing pattern as found at the end of laisses; the second and the third can be interpreted as variants of what Radloff called ‘narrative melody’ (Melody A), and the fourth is what has been called the ‘Manas melody’ (Melody B). According to Vinogradov about 60 per cent of Qaralaev’s performance is sung to Melody B. He characterizes the ‘Manas melody’ as follows: Sometimes, during the performance of Manas, there occurs a peculiar transformation of the melody here studied. The rhythm becomes more measured and symmetrical, the verse-lines become more regular, the melody consists clearly of bars. The cadences begin to measure the regular sequence of verse-lines, as time markers like the swings of a pendulum. Then the melody becomes more stable in its contour and the recitative assumes the character of a sung verseline. At this point we are in the presence of the recitative of the fourth type.18

Music and Metre: Some Examples As emerged from my discussion of the ‘Manas melody’, there is a close association between music and metre. The melody parallels the metre, with every note corresponding to a syllable. Extra syllables are incorporated into the melodic line, by dividing a note into smaller units or by adding an upbeat. Also metrically irregular lines which are short of one or more syllables are fitted to the tune by lengthening syllables in order to maintain the measured melodic pattern. Similar observations can be made in other epic singing traditions where the metre is syllable-counting and the melody syllabic. The sung portions of Kazakh, Karakalpak, and Uzbek oral epics, for instance, show the same principle, despite individual differences. Clearly the metrical mould – the production of 7- or 8-syllable lines – goes hand in hand with the musical mould, a melodic sequence of notes arranged in rhythmically defined groups. Such a close correspondence between metre and melody is not found in cases of highly melismatic tunes and in tunes of rhythmically free melodic development, such as the Turkish uzun hava (‘long melody’), exemplified in Chapter 7 (p. 147).

Oral Epics as Songs  125 A fuller discussion of the relationship between metre and music in oral epics would have to take into account not only the types of melody found, but also the different metrical systems underlying the poetic traditions in question. This would first of all require an excursion into metrical theory. Anybody familiar with the subject will be cognisant of the complex nature of metrical systems and also of the controversies raging over metrical theory. There is no space here for metrical details in different oral traditions. The literature on metrical theory is large; in the present context it must suffice to refer the reader to Nigel Fabb’s extensive discussion of metre and metrical theory, which includes a section on the relationship between metre and music.19 In the following I will elaborate on the remarks made about the metre and melody of Manas by looking at two other examples: the singing of the Rajasthani epic of Pābūjī, also briefly discussed by Fabb, and the three performance modes of the Mande oral epics. Rajasthan: Pābūjī The performance of the Pābūjī epic consists of a number of components, of which the alternation of singing and declaiming is only one.20 The epic is mainly sung (gāv ‘song’), but there are also declamatory passages, designated as arthāv ‘explanation’. A characteristic feature of the sung verse is that the tune determines the metrical form of the sung words. As John Smith explains: A given couplet can assume any one of eight or nine – maybe more – quite different strophic forms according to the tune which the bhopo chooses to use: every narrative song-tune has its own individual form and imposes its own distinctive pattern on the way in which the words of each couplet are expanded with extra verbiage. (1991: 30) Studying the performance of a particular singer (Parbū Bhopo), Smith was able to distinguish ten melody types, belonging to four groups: (1) long, slow tunes, sung rubato, (2) fast measured melodies, (3) tunes with fast repetitive sections, and (4) a tune in 7/8 metre. The musical examples provided show predominantly syllabic melodies with some melismas in 2/4 or 7/8 measures.21 The fact that the same line can be sung to one melody and by the addition of extra syllables (‘extra verbiage’) can be adapted to another melody means according to Fabb that ‘the metricality of the text is concealed in the singing of the text, because the musical structure requires the text to be altered in ways which make it unmetrical’ (1997: 103). A similar adaptation of the words to the music as found in the performance of Pābūjī can be seen in the performance of the epic Edige by

126 Performance the Karakalpak singer Jumabay Bazarov. Traditionally, the verse parts of Karakalpak oral epics are composed in lines of 7 or 8 syllables or in lines of 11 or 12 syllables. In the performance of Edige which I recorded from Jumabay, only 7/8-syllable lines occur, while in some other versions of the epic also 11/12-syllable lines are found. Jumabay sings, however, some passages with a melody that converts a seven-syllable line into an elevensyllable line, either with the help of expletives (äy) and additional notes played on the accompanying instrument or by lengthening syllables and adding expletives at the end of the line (Reichl 2007: 172–73). While syllabic melodies, such as the ones illustrated above from the performance of Manas, are coordinated with the metre of the line, musical renderings of epic verse can also overlay the metric structure of the words and impose on them their own texture. Here metre and music form a dynamic partnership. It is a common occurrence in dastans with stanzaic verse passages that the textual shape defined by metre (number of syllables, rhyme scheme) is obscured by the musical shape, where repetitions and insertions yield a different pattern (see Chapter 10). West Africa In a paper originally published in 1960 the linguist Joseph H. Greenberg maintained that ‘the vast majority of African peoples south of the Sahara, including here the non-Moslem peoples of West Africa and all the Bantu peoples except the Islamicized Swahili, do not possess prosodic systems’ (1990: 556). This statement has been disputed by Africanists. Charles Bird, in his analysis of Mande hunters’ songs, maintains that ‘the essential metrical requirement is that the singer keep in rhythm with his instrumental accompaniment. He may therefore form lines of one syllable or fifty syllables’.22 He argues that verse is not only definable by languageinternal criteria or constraints (such as the number of syllables or configurations of long and short, or accented and unaccented syllables), but also by language-external constraints such as the bouncing of a ball, the slapping of a jumprope on the sidewalk, the beating of a heart, the crashing of waves upon the beach, or, perhaps most commonly of all, the rhythm of musical instruments. By this definition, all song falls under the rubric of poetry and well it should. Every song is, in effect, a poem with the additional set of constraints governing melody. (1976: 94–95) Criticism of Greenberg’s position on African prosody (or the lack thereof) is also voiced by John Johnson, who reached conclusions similar to Bird’s in his analysis of the poetic traits of Mande epics, in particular Sunjata

Oral Epics as Songs  127 (1986: 30–38). The recordings and editions of West African epics, whether in Mande (Bambara, Mandinka etc.) or Fulani, suggest that this situation is similar: the strumming of the accompanying instruments provides a rhythmic and melodic frame into which the sung or declaimed words are fitted (see also Chapter 8, p. 174). A further characteristic of the Mande epics, which ties in with the alternation between speaking and singing in prosimetric epics, is the use of different performance modes. For the Mande oral epic tradition, the alternation between singing and speaking has been interpreted as constitutive for the performance. In fact, as Charles Bird and Gordon Innes have shown, one has to distinguish not only between two, but between three performance modes.23 The first performance mode is called ‘speech mode’ by Gordon Innes and ‘proverb-praise mode’ by Charles Bird. Both agree in their characterization: ‘In this mode the singer illustrates his virtuosity and his claim to truth by emitting, usually at a high rate of speed, numerous proverbs, sayings, and praise lines’ (Bird). ‘It differs from speech in using a generally higher level of pitch, though as in speech the lower vocal register predominates’ (Innes). Both authors put this mode on the side of speech, but clearly of a declamatory, heightened kind. The second performance mode is called ‘recitation mode’ by Innes and ‘narrative mode’ by Bird. Both authors affirm that in this mode the singer’s rhythm is regularized. Innes notes the recitation of praise and genealogies in this mode and specifies that ‘the voice is used at a much higher pitch level than in the Speech mode’ and that ‘recitations of ancestry are often marked by long descending melodic phrases of balanced length’. Here we are obviously dealing with music, hovering somewhere between chanting and singing. The third mode is called ‘song mode’ by both authors and is characterized as just that. Anthony King (in Innes) specifies: As the title implies, vocalisation in this mode is in the form of song. The pitch, intensity and resonance of the voice are at levels similar to those of the Recitation mode. The Song mode is however distinguished from the Recitation mode by its much slower syllabic rate, by the rhythmic variety brought to the realisation of its text, by the occurrence of long held syllables especially in line final position, and by the use of contrasted melodic phrases and motion, ascending and descending. (19) There is still another facet of the musical shape of the Mande oral epics: they are accompanied by instruments, which add an extra melodic and rhythmic layer to the musical texture. King explains and notates the accompaniment by the kora, a plucked harp-like instrument, and the balo, the xylophone. The use of instruments in the performance of oral

128 Performance epics, their various kinds, and the interplay of voice and instrument will be the topic of the next chapter. ***** There is no denying the fact that the music of oral epics can be of great simplicity, verging on the monotone, and that prolonged repetition might give the impression of monotony. There is also no reason to doubt the ‘vehicular’ function of melodies, especially of those called ‘speech-song’ by C. Stumpf: they help to carry the singer’s voice and help increase audibility. On closer analysis, however, it can be seen that even the simplest musical performance modes, such as the solo singing of the narrators of the Papua New Guinea highlands or of the Kyrgyz manaschï, show a surprising variety and subtlety of melodic and rhythmic texture. A further point that emerges from the examples discussed is the interdependence of metre and melody. While metre and melody run parallel and support each other in cases of syllable-counting metrical lines sung to syllabic melodies, the two can also have an asymmetric relationship when the melody imposes a formal structure on a differently patterned text. Furthermore, as C. Bird and others have argued, music can even become the defining formal principle of verse in cases where there is no language-inherent metrical system. It is therefore not without good reason that Smith speaks of ‘the utter impossibility of intelligently discussing sung “oral literature” without giving thorough consideration to the music to which it is sung’ (1979: 350).

Notes 1 On the Greek rhapsode, see Graziosi 2002: 18–40. On the singing of the Homeric epics, see West 1981; for introductory surveys of ancient Greek music, see West 1992 and Neubecker 1994; on the singer’s instrument, referred to as phorminx and also as kithara, see West 1992: 51–56; Sarti 2003. 2 General surveys of the music of epic are found in the well-known musicological reference works Die Musik in Geschichte und Gegenwart and The New Grove Dictionary of Music and Musicians; see Suppan & Helgason 1979 and Pegg & Porter 2001. For introductory comparative approaches, see also Hoerburger 1952 and Reichl 2000b: 1–40; for collections of articles on the music of epic, see Zemtsovsky 1989 and Reichl 2000b. 3 For a survey of notational problems, see Rouget (with J. Schwarz) 1970; Ellingson 1992a, 1992b; Nettl 2005: 74–91. 4 ‘Folk songs are still chanted in the fields, unaccompanied and sung during work.’ See OED s.v. chant v. 5 The audio files are available from Click Additional Resources. 6 Dobrovol’sky & Korguzalov 1981: 71. For an English translation of this bylina about Sadko (in a slightly different version), see Bailey & Ivanova 1998: 296–306. 7 On the Finno-Ugric oral epics, see Appendix A; on the oral epics of the Nenets, see Chapter 9.

Oral Epics as Songs  129 8 The Ekhirit-Bulaghat speak one of the dialects of Buryat, a northern Mongolic language. The majority of the Buryats live in the Buryat Republic of the Russian Federation, situated around the northern, eastern and southern shore of Lake Baikal; see Appendix A. 9 For notated examples, see Dobrovol’sky & Korguzalov 1981: 372–422. 10 On Nart legends, see Petrosyan et al. 1969; for an English translation of some tales of the cycle in various Caucasian languages, see Colarusso 2002. The example of the choral performance of the Karachay-Balkar Nart narratives is taken from Rakhaev 1994: 614. Shaway is one of the Nart heroes. 11 See Grame & Tsuge 1972; see also Chapter 8. 12 For the following, see also Reichl 2014; for notations and interpretations of the Manas melodies, see, among others, Krivonosov 1939; Vinogradov 1958: 116–39, 275–82; Vinogradov 1984; Belyaev 1975: 16–18; Dyushaliev 1993: 159–97, 271–79; Prior 2006 (a complete transcription of an extract from the Semetey branch of the cycle). For general information, see Köchümkulova 2016; on the situation today, see Reichl 2016a. 13 Jüsüp Mamay’s text has some dialect features, not expressed in my transcription; qorqush in l. 10 is probably an allegro form of qorqunuch ‘fear’; the published text has qorqqon. 14 For note values the British terms are used; see also the Glossary. Alternatively, the melody can be notated in a 3/8 (or 3/4) measure; as most transcriptions of the ‘Manas melody’ use a 6/8 measure, I have followed this convention. 15 For examples of stichic melodies in European folk songs and oral epics, see Wiora 1966: 18–20. 16 For a typology of ethnic melodies with illustrations, see Bose 1953: 73–81, 164–92; on the terms ‘arched’ etc. see Lomax 1968: 184. For a discussion of various approaches to melody in ethnomusicological thinking, see Nettl 2005: 113–30. 17 On the music of the chanson de geste, see Stevens 1986: 199–234. On metre, line-initial alliteration, rhyme and assonance and formulaic diction in Manas, see Reichl 2020. 18 Vinogradov 1984: 494–95 trans.; somewhat more detailed in Vinogradov 1958: 129–30. 19 Fabb 1997: 25–136 (Chapters 2 to 5), esp. pp. 96–106. For specialized literature, see his further reading sections. 20 See Smith 1991: 14–43, esp. 30–40; notated examples, 46–53; compare also Smith 1979 and Fabb 1997: 102–3. 21 There is some variation in a performance where a pair of drums (māṭā) is used; Smith 1991: 17; 47–48. 22 1972: 283; in a later paper he defines a poetic line as a sequence of three stressed syllables with one to two unaccented syllables between them (1976: 93). 23 Bird 1972: 284–90; Innes 1974: 15–24, musical analysis and examples on pp. 17–24 by Anthony King.

7 Voice and Instrument

Þǣr wæs hearpan swēg,/ swutol sang scopes. There was the sound of the harp,/ the clear song of the scop. (Beowulf ll. 89–90) The scop, the professional singer of the Anglo-Saxons, is repeatedly mentioned in the Old English epic of Beowulf, as is the hearpe, his instrument. Old English hearpe is the ancestor of Modern English harp, but the instrument it denotes is a lyre rather than a harp. The AngloSaxon singer’s instrument is therefore not very different from that of Demodocus in the Odyssey. The lyre is only one of a great number of instruments played by epic singer-narrators. The most common instruments used in epic performance are string instruments, both bowed and plucked. Most chordophones come from the lute family and are historically related. They are widely spread in Europe and Asia; their migrations through space and time are also reflected in etymologically related instrument names, sometimes found in places thousands of miles apart.1 The rabāb played by the Egyptian singer to recite the epic bears the same name as the instrument played by the Uyghur epic singer, the rawap. The Ukrainian minstrels used to accompany themselves on the kobza; this word has the same origin as qomuz, the name of the instrument the Kyrgyz singer (aqïn) plays when performing the so-called ‘little epics’, and as qobïz, the name of the instrument the Karakalpak jïraw uses. Confusingly, however, the same etymology of different terms does not imply that the same instrument is designated. The Arabic rabāb is bowed and the Uyghur rawap is plucked; the former is a spike-fiddle, the latter a lute with a small pear-shaped membrane belly (also called rubāb by Uzbeks and Tajiks). Similarly, the Ukrainian kobza and the Kyrgyz qomuz are plucked, while the Karakalpak qobïz is bowed. The kobza is classified as a lute-zither, that is, it has characteristics of both lute (plucked strings on pegs) and zither (open strings over a resonance

Voice and Instrument  131 corpus). These instruments have evolved over centuries and have in a long process acquired their present-day shapes and associated playing techniques. In some epic traditions the accompanying instrument is, as it were, canonical, in others there is a choice of instruments. When the Egyptian singers perform the Sīrat Banī Hilāl, the story of the Bedouin tribe of the Banī Hilāl, some accompany themselves on the spike-fiddle rabāb, while others prefer the ṭār, a large frame drum, or the kamanja, another type of spike-fiddle. Among Uyghur epic singers (dastanchis), not only the rawāb is found as an accompanying instrument, but also the dutar and the tanbur, two members of the lute family (long-necked lutes with frets and two or three strings, respectively), as well as the satar, a long-necked bowed lute. The Ukrainian minstrels also played the bandura, an instrument closely related to the kobza, and the lira, a hurdy-gurdy.2 The instruments provide a musical addition to the sung words; this addition can take various forms. In its simplest form the accompanying instrument plays the melody the singer sings, with perhaps small variations, supporting in this way the melodic shape of the song. The concordance of voice and instrument is sometimes in unison, often, however, homophonic and also heterophonic. In homophony ‘one voice leads melodically, being supported by an accompaniment in chordal or a slightly more elaborate style’, while in heterophony we have ‘the simultaneous use of slightly or elaborately modified versions of the same melody by two (or more) performers, e.g., a singer and an instrumentalist adding a few extra tones or ornaments to the singer’s melody’ (HDM: 390 and 383). We can also observe cases where the instrumental accompaniment not only complements the sung melody, but takes on its own musical role. In addition to enriching the sung melody melodically and harmonically, the instrumental accompaniment often leads to the insertion of instrumental interludes, in which music takes first place, with the singers demonstrating their skills as instrumentalists. Here the music becomes foregrounded and the epic recital, described by Curt Sachs as ‘text-dominated’, turns into a performance where music is at least on a par with the text, if not even in a more prominent position. All epic traditions and regional styles, as well as all singers and musicians have their own individual musical performance modes. There is no way the variety and diversity of the interplay of voice and instrument, of singing and playing, can be treated fully here. I propose therefore to focus on two epic singers from my own fieldwork, one accompanying himself with a bowed, the other with a plucked string instrument, and to place them in a comparative context.

132 Performance

Gusle, Qobïz, Horse-Head Fiddle Perhaps best known to students of the oral epic is the musical style of the South Slavic singer-narrator. The accompanying instrument of the South Slavic pevač (pjevač) or guslar was the gusle. The instrument is familiar from Albert Lord’s Singer of Tales and, of course, from the earlier writings of Milman Parry, the even earlier works of Mathias Murko of the late nineteenth and early twentieth century, and from eighteenth-century travellers’ accounts. Alberto Fortis, whose Travels into Dalmatia were published in Italian in 1774, describes the singers’ instrument as having one string, plaited from horse-hair, and remarks that the singers’ heroic songs are lachrymose and monotonous and that the nasal sound of their singing accords excellently with the sound of the instrument. This is still true of the gusle today. On top of the instrument’s neck there is a scroll, often representing an animal’s head (horse or goat); the body (sound box) is carved from one piece of wood and has a belly of animal skin. The instrument is played in a sitting position, held between the guslar’s knees or on his lap with the legs crossed. The gusle comes in various regional types, which have different shapes and sound qualities.3 This instrument is also used by the Albanian epic singers; in Albanian it is called lahuta.4 Other bowed string instruments used in epic performance are the rabāb of the Egyptian singer of tales, the satar of the Uyghur dastanchi, and the ghijjak of the musicians accompanying the Turkmen bakhshï and the Karakalpak baqsï.5 The performer (bhopo) of the Rajastani epic Pābūjī plays the two-stringed fiddle rāvaṇhattho, while the Pardhan epic singers of the Gonds in Madhya Pradesh play a three-stringed fiddle, called bana. The rāvaṇhattho is pressed against the chest, the bana is played on the shoulder like a violin.6 Another fiddle played by epic singers to accompany their singing is the khissyn fændyr. This bowed lute with two or three strings is used by the Ossets for the singing of the Caucasian Nart legends (in northern and southern Ossetia). The instrument’s name is composed of khissyn ‘horse-hair’ and fændyr, the latter ultimately derived from Greek pandoura, a word that has spawned a large family of instrument names. The Ossets have two types of singers, the narrator and the narrator-singer. It is the latter that plays the khissyn fændyr.7 The growing interest in the overtone-singing of Tuvans and Mongols in the world music scene since the 1990s has also brought the Mongolian horse-head fiddle to the attention of a wider public in the West. The morin khuur has two strings made of horse-hair, which are tuned in fourths or fifths. It is played as a solo instrument and to accompany the singing of songs and epics. Morin khuur music, especially when performed solo by professional musicians, is often virtuoso in character, with a predilection for imitative elements, suggesting the galloping of horses by rhythmic urgency or, by glissandi over a wide ambitus, the neighing of horses. The instrument is also employed in various rituals, such as a coaxing ritual

Voice and Instrument  133 for camels, inscribed on UNESCO’s list of Intangible Cultural Heritage. In this ritual a camel mother who has rejected her offspring is coaxed to accept the newborn. Some singers also use the khuuchir, a similar instrument with a small, hexagonal body.8 The Karakalpak qobïz is both similar to the morin khuur and differs from it in a number of points. It has two horse-hair strings and is held on the player’s lap. It has, however, a shorter fingerboard and is a smaller instrument than the Mongolian horse-head fiddle; it also has a pear-shaped body, which is hollowed out and has no top, either of wood or skin, while the morin khuur has as its body a rectangular wooden box with two sound-holes. The qobïz is also played by the Kazakhs as a solo instrument (called qïl qobïz, ‘horse-hair qobïz’). In the past it was the instrument of the Kazakh baqsï or shaman and fulfilled the same function as the shamanic drum. As pointed out above (Chapter 4), Jumabay’s qobïz is, like the morin khuur, meant to symbolize a horse. The shamanic associations are certainly there, but are maybe stronger on account of his singing style than the symbolism of his instrument.

Case Study: Karakalpak Jumabay Bazarov acquired his epic repertoire from his teacher, the Karakalpak jïraw Esemurat Nurabullaev (1893–1979). Esemurat-jïraw’s repertoire was not only more comprehensive than his pupil’s, it was also more varied musically. While singers of Esemurat’s generation mastered thirty to forty melodies, Jumabay uses a bare handful. A musical transcription of the Karakalpak epic Alpamïs, for instance, contains about thirty melodies (see Chapter 8, p. 160). The version of Edige I recorded in 1993 from Jumabay has thirty-seven sung verse passages, which are fairly equally distributed over only four basic melodies. These melodies have variations, and it could be argued that perhaps a fifth or possibly even a sixth melody type is recognizable. Jumabay used the same group of melodies for the performance of the other epics in his repertoire (Qoblan and Sharyar) as well as for shorter poems such as the historical-didactic song (tolghaw) Ormanbet-biy.9 The melodies have a small ambitus, generally staying within an interval of a third or a fourth. They are stichic and syllabic. As mentioned earlier, Jumabay occasionally sings a 7/8-syllable line as an 11/12-syllable line (Chapter 6, pp. 125–26). The qobïz is played in unison with the sung melody, but sometimes slight ornamentations are added. To illustrate this, I will give a ‘broad transcription’ of the first lines of a sung passage in Edige (see Audio/Video Ex. 4). This passage comes immediately after the Solomonic judgement discussed in Chapter 4. Khan Tokhtamysh’s wife reproaches her husband that he had ignored her warnings and neglected to get Edige out of the way (Reichl 2007: 188).

134 Performance

Example 7.1 Edige (1). [The wife of Tokhtamysh,/ She who looks after the household,/ Came to Tokhtamysh/ And complained, äy.]

All musically performed passages have a constant structure. They begin with an instrumental prelude, sometimes comprising only a few notes. The verse passage generally consists of a sung and a declaimed portion. The singer begins by singing, but changes toward the end of the passage into declamation. The sung portion is interrupted by short instrumental interludes, sometimes after five, sometimes after around ten lines. A typical interlude melody is the following:

Example 7.2  Edige (2).

Instead of playing an interlude, the singer might make one or more down-strokes with the bow, in this way punctuating the singing or the recitation. These bow-strokes as well as the short interludes occur also in the prose passages, when the singer speaks rather than sings. When the end of a verse passage is reached, the last line is generally sung in a somewhat protracted manner, with one or more additional melismatic melodies on the syllable aw (‘oh’):

Example 7.3  Edige (3).

Voice and Instrument  135 The passage then closes with a short instrumental postlude. This musical structure is remarkably similar to what George Herzog wrote about the music of the South Slavic heroic songs, where voice and instrument interact in much the same way, with instrumental prelude, interludes, and postlude: The instrument begins with a short prelude. After the entry of the voice, the accompaniment follows the latter in a so-called heterophonic technique: the two voices together render somewhat divergent versions or forms of the same melodic line. The pause at the end of each sung line is filled out by the instrument with short stereotyped figures; the interludes and the postlude or coda (which is at times not represented) are also purely instrumental. Prelude, interludes, and postlude bring always related material and are often quite similar. While they and the brief line-final figures are rather ornate, the instrumental part becomes simpler when it accompanies the voice; it follows the latter rather closely so that the two voices are essentially in unison. Simultaneous intervals between the two, usually seconds, occur chiefly in text-final positions.10 The potential of unaccompanied song is significantly enlarged when a musical instrument is added. Despite the simplicity of the sung melodies, the musical texture is enriched by the instrument and the musical aspect of performance becomes an important element in the acoustic impression of the listener. For the singer, the instrumental interludes give him space for rest and the musical accompaniment reinforces the rhythmic and metrical mould of the verses and carries him along in his performance. This role of the instrument is not only typical of the performance of the Karakalpak jïraw and the South Slavic guslar, but also of the Mongolian khuurchi and the Uzbek bakhshi. In a description of three Mongolian epic singers, one of whom accompanied himself on the morin khuur, C. R. Bawden compared the khuurchi with a singer named Bataa, who played the tovshur, a plucked lute, and noted: The instrumental line still seemed to have as its main duty to act as rhythmic support, but it was somewhat more elaborate and more varied, both in melody and in tone, than Bataa’s. … Every now and again, too, the voice would pause, and the fiddle would engage in a sort of musical interlude or cadenza. The purpose may have been to give the voice a rest, but the interludes may have had an independent entertainment value. (1979: 42)

136 Performance

Lute, Dutar, Dombira According to the systematic classification of musical instruments, which is based on structural features, the South Slavic gusle, the Albanian lahuta and the Karakalapak qobïz are lutes: one or more strings run from the neck (fingerboard) over a body as resonator (with or without a top or belly); on the neck they are fastened with pegs for tuning and at the other end they are either fixed to the bottom end of the instrument or to the bridge on top of the lower part of the belly. Whether the strings are bowed or plucked makes no difference to organology. In everyday English, however, a lute is understood to be a plucked instrument. The word lute is first attested in the English language in the fourteenth century; in the fourteenth and fifteenth centuries the word generally appears in lists of musical instruments played by minstrels (Carter 1961: 251–52). The word is derived from Arabic al-cūd, literally ‘piece of wood’, and has kept the Arabic definite article al as l. With the word the instrument was imported to medieval Europe. The cultural symbiosis of Orient and Occident in the Golden Age on the Iberian Peninsula is illustrated by a well-known thirteenth-century illumination in the Escorial manuscript of the Cantigas de Santa Maria by Alfonso X, featuring two lutanists from their respective cultural areas.11 The lutes represented on this illumination have long necks (‘long-necked lutes’); this is the basic form of the plucked lutes of Central Asia. The more typical Arab lute does, however, have a short neck with its peg-box bent backwards. The plucked lute had a long history before its appearance in medieval Europe. The first representations of a lute date to c. 2300 bce and are found on two Mesopotamian cylinder seals. A number of reliefs and other pictorial evidence testify to the wide currency of this instrument among the civilizations of the ancient Near East.12 Probably from there it spread also eastward to Central Asia, where it is found in numerous variations. The plucked lute is the most common accompanying instrument of both the Turkic traditions from Anatolia to Xinjiang and the Iranian traditions in Iran and beyond. Of wide currency is the name dutār for the plucked long-necked lute. This word, appearing in varying spellings and pronunciations in different languages, is of Persian origin and means literally ‘two strings’. The Turkmen, Karakalpak, and Uyghur epic singers call their instrument dutar, as do the Tajik and the Khorasanian (north-western province of Iran) bards. Among the Karakalpaks only the baqsï plays the dutar; the jïraw plays the qobïz. In addition to the dutar, the Uyghur and Tajik singers also use other plucked instruments for their accompaniment. Various types of dutar are current, differing in size, in the material from which the instrument is made, and in the absence or presence of frets and their number. They all have two strings, generally tuned in fourths or fifths, which are plucked either with the fingers or with a plectrum. Uzbek and Kazakh epic singers typically accompany themselves on the dombira/dombïra. The word domb(i)ra is derived from Persian tanbur.

Voice and Instrument  137 Curt Sachs suggests a derivation by metathesis from pandura, the name of an ancient Greek chordophone, which in turn has been connected to a Sumerian lute.13 The tanbur is found as an accompanying instrument of the Uyghur dastanchis. Also the Bosnian epic singers used the tanbur, calling the instrument by the diminutive tanburica, lit. ‘little tanbur’. While the Uyghur tanbur has five strings and the South Slavic tanburica three or more strings, the Kazakh and the Uzbek domb(i)ra has only two strings. Both differ, however, in a number of ways from each other. One is their form: leaving aside regional variants, the Uzbek dombira is as a rule pear-shaped, while the Kazakh dombïra has a more elongated body, often with pronounced shoulders. Furthermore, the Uzbek dombira has no frets, while the Kazakh dombïra does. The top string of the dombira is generally tuned a fourth or a fifth higher than the bottom string. Uzbek musicologists divide the traditional music of Uzbekistan into four regions: (1) the Surkhandaryā and Qashqadaryā provinces in south-western Uzbekistan, (2) Bukhara and Samarkand and their provinces, (3) Khorezm in the north-west, and (4) the Ferghana valley, including also Tashkent.14 In the first half of the twentieth century some of the best known Uzbek epic singers came from the Qorghān and Bulunghur districts of the Samarkand province (Fāzil Yoldāsh-oghli, Ergash Jumanbulbul-oghli and others). Today epic singers are mostly found in the Surkhandaryā and Qashqadaryā provinces (in the districts of Shahrisabz and Sherābād). There are, however, also bakhshis in other regions of Uzbekistan or among the Uzbeks in neighbouring countries, especially in Tajikistan. In Khorezm the famous ‘Bāla-bakhshi’ (‘Child-bakhshi’, because he began performing at a young age), Qurbānnazar Abdullaev, lived almost until the end of the twentieth century (1899–1994). Given that Uzbek dastans are generally prosimetric, two basic modes of performance can be distinguished, one typical of the prose portions, the other of the verse portions. As in similar epic traditions, the prose portions are spoken; while reciting the bakhshis generally strum their dombira at irregular intervals. The verse parts are sung. In some regions, however, singers also recite the verse parts.15 Typical of the bakhshis’ singing in the Surkhandaryā and Qashqadaryā provinces is a pressed, tensed guttural voice production, called ichki āvāz ‘inner voice’ as opposed to tashkari āvāz ‘outer voice’, the natural voice (see Chapter 4). Case Study: Uzbek As an illustration of the Qashqadaryā style I will discuss an excerpt from the epic of Alpāmish, which Chāri-shāir Khojamberdi-oghli sang in 1981 at the encounter described in the Introduction.16 Chāri-shāir’s father Khojamberdi-bakhshi was a well-known folk epic singer, from whom Chāri-shāir learned the art of epic singing. In 1981 he sang, in addition to a passage from Rawshan, extracts from the epics Alpāmish, Rustamkhān, Awazkhān, and Nurali.17

138 Performance

Illustration 13  Chāri Khojamberdi-oghli (c. 1980)

The extract from Alpāmish was introduced by a short contextualization in prose and then sung to the accompaniment of the dombira. Alpāmish is the main heroic epic of the Uzbeks, an epic that is also found in other oral traditions, especially that of the Karakalpaks and the Kazakhs. In Chapter 1 (p. 19) I mentioned the two parts of the epic, a bridal quest and a return story. Two rich men (in many versions brothers), Bāybori and Bāysari, promise to wed their children with each other if their wives bear a son and a daughter. Bāybori has a son, Alpāmish, and Bāysari a daughter, Barchin. Due to a quarrel, Bāysari leaves with his clan and fellow tribesmen and migrates to the land of the Kalmyks. Barchin, meanwhile grown into a beautiful girl of marriageable age, is wooed by Kalmyk brothers. She refuses them and sends a message to Alpāmish to come and win her hand in a suitors’ contest. Alpāmish defeats his co-competitors and brings Barchin back to his homeland as his wife. Later Alpāmish departs again to rescue his father-in-law from Kalmyk oppression, but is made drunk on the way and thrown into an underground dungeon. After an imprisonment of seven years he manages to get free with the help of the Kalmyk khan’s daughter and returns home. He arrives just in the nick of time before Barchin is forced to marry the usurper of Alpāmish’s position as tribal leader, who has been oppressing and harassing Alpāmish’s people. On his return, Alpāmish has a number of encounters with faithful and faithless companions, members of his family, and also with trusty animals, and he is recognized by one of his servants by a scar (more precisely the imprint of a saint’s hand) on his shoulder. The close parallels

Voice and Instrument  139 between Alpāmish’s and Odysseus’s homecoming suggest cultural contacts between Greece and Central Asia, direct or indirect, in the course of a long history of cultural transfers along the Silk Road.18 The passage Chāri-shāir sang comes from the first part of the epic. Barchin was approached by the Kalmyk brothers Kokaman and Qārajān as suitors and answers their advances in anger (see Audio/ Video Ex. 5)19: Āt bāshini saqlab gapir, Kokaman! Qahrim kelsa, qāra jerga tiqaman. Yā hanjar deya ozimga yā senga suqaman. Āt bāshini saqlab gapir, Kokaman, ey Kokaman! 5






Āt boshini saqlab gapir, Qārajān! Qiz bolsam ham, mening achchighim yāmān. Qahrlansam, yā senga yā menga dur ākhir zamān. Āt bāshini saqlab gapir, Qārajān! Qirq ming yātgan mening Qongrat elim bār. Qongrat elda Hakimbekday yārim bār, Sevgili, sevgili, ey, dildārim bār. Agar kelsa sorab shaharlaringga, qalmāqlar, Bolasanmi tani menen baravar? Bāshim qilma ming bālaga giriftār! Āt bāshini saqlab gapir, ey qalmāqlar! Tārtip ālib ketkaningga iya tiymayman, Kallam ketsa ham, qalmāq seni suymayman. Bir shartim bār, qulāq sāling, qalmāqlar! Anda nima bolsa, endi bolaman, Songra taghdirimnan bir kun korarman. Toqqiz kunlik yoldan bedav jiydiraman, Uch kunlik yollardan payga qoydiraman, Chilbir choli aylab, ko(h)na Bābākhān tāghiga, Besh yuz jüyrik ātti ālib bāraman. Shul Bābākhān tāghiga kimning āti ozib kelsa, Mening bakhmal, ey, chātirimga, Shu kākilni shunasidan qiyaman. Āti ozāghān chavandāzga tiyaman, Tā olguncha shul bātirni suyaman. Agar tārtip ālib ketaman desang, Bir kun, qalmāqlar, Kozing oyib kozlaringga, ey, qorghāshin quyaman.

140 Performance Hold firm the horse’s head and speak, Kokaman! If I get angry, I will plant you on the black earth. I will thrust the dagger either into me or into you. Hold firm the horse’s head and speak, Kokaman, oh Kokaman! 5






Hold firm the horse’s head and speak, Qārajān! Although I am only a girl, my anger is deadly. If I get angry, life will have come to an end either for me or for you. Hold firm the horse’s head and speak, Qārajān! I have my tribe Qongrat of forty-thousand peoples. In the tribe Qongrat I have my beloved Hakimbek [Alpāmish], My darling, my darling sweetheart. If he seeks out your town, Kalmyks, Will you then be his equal? Don’t bring a thousand sorrows on my head! Hold firm the horse’s head and speak, oh Kalmyks! Even if you take me by force, I will not marry you, Even if my head is cut off, I will not love you, Kalmyks. I will set a condition, pay attention, Kalmyks! Whatever the outcome, I will be (in agreement), I will then live according to my fate. I will bring together race horses fit for a race over the distance of a nine days’ journey, I will arrange a race over the distance of a three days’ journey. For the race across the Chilbir desert up to ancient Mount Bābākhān, I will have five hundred race horses. Whose horse will first reach Mount Bābākhān, Him (I will take) to my velvet tent, ey, And (for him) I will cut off this plait.20 I will marry the rider whose horse is fastest, I will love this hero until I die. If you want to take me away, One day, oh Kalmyks, I will rip out your eyes by pouring lead into them.

This passage comprises 32 lines. They are divided into four stanzas of varying length: 1–4, 5–8, 9–15, 16–32. The first two stanzas repeat their respective first line as their fourth line. As these lines are identical in both stanzas with the exception of the name at the end (Kokaman/ Qārajān)

Voice and Instrument  141 we have a refrain-like effect. The first three stanzas are mono-rhyme stanzas, rhyming in -an/-ān in stanzas 1 and 2 and in -ar/-ār in stanza 3. The fourth stanza has twelve lines rhyming in -an, two have the final word qalmāqlar (Kalmyks), and two lines don’t rhyme. The lines are basically 11/12-syllable verse lines. By ‘basically’ I mean that this is clearly the intended metre, rather than the other common dastan metre, the 7/8-syllable line. In actual fact, 23 of the 32 lines have 11 or 12 syllables. Seven lines are longer, with 13 (one line), 14 (four lines), and 15 (two lines) syllables. One line has 9 syllables and the last but one line has 5 syllables. The division of this verse passage is based on the rhyme-scheme, but also on the segmentation provided by the dombira interludes. We have here a mixture of two strophic principles: the arrangement of the verse lines as four- or five-line stanzas and the grouping of the verse lines into laisses, longer stanzas of variable length. The musical performance begins with a dombira prelude. After every stanza there is an instrumental interlude. The first two stanzas receive special treatment: there is a short dombira solo after the first line and in the second stanza after the second line. In addition to these dombira interludes, there is a vocalization on ayii between the third and fourth stanza and a short exclamatory flourish on heyii after the last stanza. Symbolizing the dombira solos by D, we have the following structure: D line 1 D 2-4 D 5-6 D 7-8 D 9-15 D ayii … D 16-32 heyii … The melody of the song is fairly simple and close to a monotonic tune. The singer sings in a somewhat tensed, guttural voice in a low register (ichki āwāz). His singing is forceful and forte. Here is the outline of the melody of the first stanza:

Example 7.4  Alpamish (1)

142 Performance The melody moves within a fourth (D–G); G and F are occasionally sung in a row, not unlike the reciting note of psalmody, that is, the singing of a stretch of text on the same note. In their basic melodic shape, lines 1 and 4 and lines 2 and 3, respectively, follow the same pattern (lines 1 and 4 centred on F, lines 2 and 3 on G). Because of the repetition of Kokaman in line 4, the melody has an additional four quavers at the end. The melody is syllabic; in the cases where two notes are song to one syllable, we have more of a glide from a higher to a lower note than two clearly separated notes. Despite its recitative character, the melody is strophic rather than stichic. The stanza consists of two melodic phrases, which form the sequence ABBA; a stichic melody pattern would be AAAA. In the course of the performance, the melody is, however, repeated for every line with only small variations and becomes stichic. The singer deviates from this only in the vocalization after line 15:

Example 7.5  Alpamish (2).

The melody of this vocalization is built from the same tonal material as the rest of the passage, but has a different rhythm, which creates a more insistent, urging impression. The melody of the vocalization is taken up by the dombira in the interludes. The dombira, tuned in a fourth (Bflat–F), plays an active part in the performance, not only as an accompanying ‘drone’, but as a creator of musical sound in its own right. The music of the dombira provides chordal accompaniment, an additional musical texture, particularly rich in the interludes, and a marked rhythm. It is this rhythmic forward movement, this energetic strumming of the instrument, which characterizes the overall acoustic impression of the bakhshi’s performance.21 A striking contrast to the singing and playing style of the bakhshis from southern Uzbekistan is offered by the musical performance of the

Voice and Instrument  143 dastan in Khorezm. As mentioned earlier (Chapter 2, p. 28), the oral tradition of dastans is paralleled and influenced by their written transmission, which entails a marked emphasis on memorization rather than on composition in performance. The repertoire of Khorezmian singernarrators comprises in the main love and adventure romances, such as dastans about star-crossed lovers (e.g., Gharib and Sanam; see Chapter 10) and branches from the Goroghli-cycle. The verse passages are generally in stanzas rather than in laisses. A characteristic feature of the performance of the Khorezmian bakhshi is that his singing and playing is accompanied by further musicians. They consist of a dutār and a dāira (frame drum) player, often with additional players of plucked (rubāb, tār), bowed (ghijjak) and wind instruments (bulamān); also the accordion is used. Instead of the dombira, the Khorezmian bakhshis strum the dutār or the tār, the latter a pear-shaped, two-stringed lute. They sing with an ‘open voice’ and perform more melodious tunes than the bakhshis from southern Uzbekistan. In a dastan over thirty melodies might occur. Some bakhshis, such as Bāla-bakhshi, are said to have mastered more than seventy melodies. A further characteristic of the Khorezmian style is a melodic progression to higher registers in the course of a sung passage; after the melody reaches its ‘zenith’ (auj) it descends to its final position.

The Interplay of Song and Instrument A number of points emerging from the discussion of the musical performance of the Karakalpak jïraw and the Uzbek bakhshi merit further attention. The first concerns the choice of accompanying instrument or instruments. Lutes, in particular plucked lutes, come in many forms and sizes in the Turkic and Iranian world and are the epic instruments par excellence. Apart from lutes other string instruments are also found. Ossetian singers of the Nart epic not only play the fiddle khissyn fændyr, but also the dywwalæstænon fændyr; dywwalæstænon means ‘twelvestringed’ and fændyr here denotes a small harp. The Turkic-language Khakas accompany their epic singing on the chatkhan, an instrument from the zither family with seven and sometimes also nine or eleven strings. In Africa we find among the Fang in Central Africa the mvet and among the Mande of West Africa the kora, classified as a harp lute or as a bridge-harp, with up to about twenty strings; it has a calabash as resonator and is played with both hands.22 This list of chordophones accompanying epic singing is far from complete, nor are chordophones the only instruments played by epic singers and musicians. I have mentioned the presence of the oboe-type bulamān, the frame drum dāira, and the accordion in Khorezmian ensembles. The North Indian Ḍholā epic (see Chapter 3) is performed by a main singer and additional musicians, some also singing. Susan Wadley describes two troupes, one consisting of the singer who plays the cikāṛā, a type of fiddle,

144 Performance a drummer who plays the ḍholak, a barrel-shaped drum, and a cimṭaplayer, the cimṭā being long steel tongs. In the second troupe the main singer plays the harmonium and the other members play the cikāṛā, the cymbals and the drum, respectively (Wadley 2004: 69–71, 79–83). West African epic singers are not only accompanied on the kora, but also on the balo, the xylophone. A curious form of musical accompaniment is described by Ruth Stone for the Wọi epic of the Kpelle of Liberia: In the center of the gathering, the epic singer kneels to position himself on the mat. Near him he appoints two instrumentalists, each to hit a beer bottle with the back edge of a penknife and make the sound, if not the appearance, of the more ancient struck iron idiophone. (1988: 2) Clearly, singers and musicians are never at a loss when it comes to instrumental accompaniment. A second point concerns the structuring of the verse passages in prosimetric epics. The instrument divides the verse passage by interludes and forms it into a unified poetico-musical piece by adding a prelude and a coda. This is also typical of other traditions, as for instance that of the Uyghurs and the Turkmens.23 In musically modelling these verse passages, changes of register are found in a number of traditions (Uzbek, Karakalpak, Turkmen, Khorasanian bakhshis and others). They consist first in a rise from a lower to a higher register and then again a descent to the initial register. These changes are also called perde changes, that is, changes of frets on the plucked instrument and consequently of tonal (and modal) ranges. The musically performed passages in Gulumbakhshi’s version of Kharman Däli, for instance, also show such perde changes. For Turkmen epic music, Dzhamilya Kurbanova gives the following explanation: Any popular melody can have ten to fourteen degrees of development, the so-called oba (lit. ‘encampment’), on each of which several songs can be grouped together. The songs of the initial degree (birindzhi oba) are the lowest in tessitura, with very restrained dynamics. The higher the degree, the more dynamic the singing and the wider the vocal range. The tessitura level of a song is defined by the base note on the finger-board of the dutar, a two-stringed lute with frets. Apart from a melodic rise by chords on the instrument (perde, ‘fret’), during the performance the bagshy periodically tunes the strings of the dutar up, each time raising the level of the base by a fourth. This is a traditional method, which is popularly called chekim (‘a pull’, also ‘a puff’, from

Voice and Instrument  145 chekmek ‘to pull’). Such tunings can be carried out up to seven times, but this is already the sign of a very talented singer. In this way one melody can serve as the basis for more than 30 songs, with an increase in dynamic intensity and tessitura. On this principle a melody can form various patterns, which can then be tabulated according to the respective oba. Such a classification allows us to understand how the folk bagshy arranges a fairly comprehensive song repertoire in his mind. Apart from this, the classification also mirrors the performance of the bagshy, who, regardless of the melodies he selects, always starts his performance with a song in a low register and then rises by stages to songs in a high register. (2000: 117) A third point that needs stressing is the rhythmic function of the accompanying instrument(s). The presence of drums among accompanying instruments of epic performance underlines one of the functions of instrumental music. The drum beats mark the rhythm of the singer’s recital. This rhythmic emphasis is also found in cases of solo singing or chanting, as for instance in the Ainu epic tradition, where the narrator marked time by taps on the hearth frame with a block of wood, or in the West Sumatran tradition of the Minangkabau, where the singer uses a halffilled box of matches for the same purpose. While a steady beat might function as a support for the singing, this is not everywhere the case. In musical cultures with a predilection for complex rhythmic structures (as found, for instance, in India and the Middle East, also in Africa), the drumming might provide an at least partially contrasting rhythmic background to the metre-dependent rhythm of the verse lines. Even the comparatively simple accompaniment on two beer bottles at the Kpelle epic performance features two slightly different rhythms, one syncopated, the other with even beats (Stone 1988: 2):

Example 7.6  Kpelle

A final point concerns the comparatively close correlation between song and instrumental melody in the examples discussed above. The

146 Performance melodic ‘support function’ of instrumental music seems obvious, but is not quite as straightforward as one might think. It is clearest in syllabic and stichic melodies, melodies with one note per syllable and which are repeated for every line (or two lines), as in the extracts from Edige and Alpāmish. The accompanying melody might be in unison with the sung melody, but it might also be heterophonic, that is, paralleling the sung melody in different intervals. The instrumental melody can also supply a bourdon-like tone on which the sung melody is based. In the musical notations of two lines from three different heroic songs in Lord’s Singer of Tales, we find that Avdo Međedović often played the gusle in unison with the singing voice, while Salih Ugljanin repeatedly used sustained notes on the gusle, which the sung syllables melodically embroider. We can also see from these notations that the instrument takes up the sung melody, with its ornamentation, in the intervals between one line and the next (Lord 1960: 39–41). Depending on the instrument, the accompaniment might also add a harmonic dimension to the melody, by playing stops on bowed instruments and strumming more than one string on plucked chordophones. Not all epic melodies are, however, syllabic and not all epic music has a regularly repeated rhythm. An example is the music of Turkish minstrels, who like the Azerbaijanian aşıqs play the saz, a long-necked lute, with cura, bağlama, and meydan sazı as main types (ranging from small to large). The eleven-syllable verse passages inserted in Turkish minstrel romances (hikâye) and the Turkish bozlak, a song-type closely connected to the hikâye genre, are sung in a melodic and rhythmic style called uzun hava (‘long melody’). Typical for the uzun hava is a rhythmically free melodic flow, with a wide ambitus and a gradually descending movement in the course of the song. Melismas and various ornaments such as trills and mordents are also typical of this style. Apart from the bozlak, the lament (ağıt) is also sung in the uzun hava style. Kurt and Ursula Reinhard, in their study of Turkish music, give the following example of a bozlak melody from the romance Ali Paşa, sung by the âşık Kır İsmail, which illustrates the basic characteristics of the uzun hava. This singer, from the Çukurova in south-eastern Turkey, was also recorded by Béla Bartók on his trip to Turkey in 1936. In the edition of his musical transcriptions two songs by this âşık are included. While the melodic development of the song has this rhythmically free, parlando character, the instrument (a three-stringed cura ırızva of the saz family) plays a regularly repeated rhythm, generally in the intervals between sung lines, but also in accompaniment to sustained sung notes.24

Voice and Instrument  147

Example 7.7  Ali Pasha

Irrespective of whether the music is parlando or tempo giusto, the instrumental accompaniment brings a richer musical texture to the epic performance. In preludes, interludes, and postludes the instrumentalist can take on the role of a virtuoso. The music also becomes foregrounded by the extension of the group of musicians. Even by just adding one more musician the epic performance takes on a different character. The Turkmen epic singer of the Tashauz (Dashoguz) style, for instance, plays the dutar and is further accompanied by a ghijjak player; the same is true of the Karakalpak baqsï. And here the rhythmical strumming of the dutar in a fast tempo presses on and gives an impression of haste, while the ghijjak plays a melody that unfolds in wide arches and conveys a certain calm and restfulness. These separate aural impressions, caused by differences in melodic and harmonic shapes, by different sound production (plucking vs. bowing), and, of course, by the different acoustic qualities of the respective instrument, complement each other and produce a musically nuanced accompaniment for the singer’s words and song.

Notes 1 For general information sources on musical instruments and a glossary of non-Western musical instruments, see the Glossary. 2 Satar is the Uyghur form of Persian setār (in Hindi sitar), literally ‘three strings’. The satar is also played by Uzbeks and Tajiks, who use it as a plucked chordophone. On Ukrainian minstrels and their instruments, see Kononenko 1998 (see the plates following p. 198). 3 See Fortis 1774: I, 88; on the playing technique, see Wünsch 1934. 4 On the lahuta and the symbolism of its scroll, see Ahmedaja 2013. On the musical performance of the Albanian lahutar, see Dietrich 2000 (with illustrations on pp. 91 and 93). 5 On the ghijjak (Uzbek; Turkmen gïjak, Karakalpak ghïrjek), see Karomatov 1972a: 110–14. 6 On the bana fiddle, see Knight 2000–2001.

148 Performance 7 On the Ossetian Nart legends, see the translation by Georges Dumézil (1965); on the music of the Ossetian epic, see Tskhurbaeva 1969 (with illustrations of the two instruments). 8 See These instruments are also discussed and illustrated in Pegg 2001: 57–90. On the role of the morin khuur in Mongolian national music, see the study by Marsh 2009. 9 For interpretation, text and English translation of this tolghaw, see Reichl 2000a: 104–15, 189–94. 10 Herzog 1951: 63; compare also Erdely 2000 and Bartók’s notation of ‘The Captivity of Ðulić Ibrahim’ (1954). 11 This illumination has been often reproduced. See, for instance, Harris & Reichl 2012: 169. 12 See Kilmer & Collon 1983; Turnbull 1972. 13 See Sachs 1913: 375; West 1992: 80. 14 See Karomatov 1972b; Abdullaev 1989; Levin 1996. 15 Abdullaev (1989: 114) mentions Uzbek bakhshis from the province of Osh in Kyrgyzstan. Also the women bakhshis (khalfa) of Uzbekistan do not sing; see Chapter 2. 16 Chāri-shāir (born in 1925) still performed to a group of students of the Mirzā Ulughbek National University of Uzbekistan in 2000; in 2020, I was told that he had died. At the time of recording, Chāri-shāir’s family name was given to me as Egamberdi, a name I used in earlier publications. It was only in 2018, when I saw Chāri-shāir’s portrait in the Alisher Navā’ī State Museum of Literature in Tashkent (reproduced here), that I was able to clear up this mistake. 17 For a discussion of Chāri-shāir’s singing of the extract from Rawshan, see Reichl 1985b. 18 For a German translation of an Uzbek version, see Reichl 2001. On parallels between the epic and Homer, see Zhirmunsky 1967; see also Grossardt 2006; West 2012; Ready 2014; Jensen 2017. 19 I have left the most salient dialect traits. 20 Unmarried girls wear their hair in plaits. 21 For discussions of the Uzbek dastan music, also for musical transcriptions, see Belyaev 1975: 288–92; Dzhabbarov 1971; Abdullaev 1989; Karomatov 1999. 22 On the kora (and other instruments of the Mande), see Hale 1998: 146–71. 23 For an analysis of the music of Uyghur epic music, see Dawut & Anderson 2016. 24 Reinhard 1984: II, 33; for a synopsis of Ali Pas¸a, see Eberhard 1955: 27–29; on the bozlak genre, see Mirzaoğlu 2003. See also the photo of Kır İsmail in Bartók 1976: 57; see the nos. 18 and 23 (music 99–100, 108–10, notes 185, texts: 228 and 231). Bartók’s recordings are issued on a CD-set issued by Hungaroton in 1996 (HCD 1818–19); see disc 2, tracks 8 and 9.

Part III


8 Words, Music, Meaning

It followeth to shew you how to dispose your musicke, according to the nature of the words which you are therein to expresse: as whatsoever matter it bee which you have in hand, such a kinde of musicke must you frame to it. (Thomas Morley, A Plaine and Easie Introduction to Practicall Musicke. London, 1597) The late Renaissance English composer Thomas Morley, who is mostly remembered for his madrigals and ayres, wrote his introduction to music in the form of a dialogue between a music master and two brothers, Polymathes and Philomathes. The master instructs them among other things on how to compose ditties, in which the nature of the words must be expressed in the music. ‘You must then’, the master exemplifies, ‘when you would expresse any word signifying hardnesse, cruelty, bitternesse, and other such like, make the harmonie like unto it, that is, somewhat harsh and hard, but yet so that it offend not. Likewise, when any of your words shall expresse complaint, dolor, repentance, sighs, tears, and such like, let your harmonie be sad and doleful’ (1597: 202). This is the time when composers like Luca Marenzio or Claudio Monteverdi searched for musical equivalents and correspondences to the words of their texts. In one of Marenzio’s madrigals, for instance, the line ‘Descend from paradise, Venus’ (Scendi dal paradiso, Venere) is sung to a downward leap of a fifth for scendi and a descending scale of five notes for dal paradiso, a musical figure repeated and taken up by the other voices. Examples of such word-painting in music are also found in medieval music, but there was not yet this new desire to express the meaning of the text in music.1 Similar ideas, affirming a relationship between word and music, the mood of a poem and the tune to which it is sung, the meaning of a passage and its musical performance in song and instrumental accompaniment, are also found in oral epic traditions. An example will illustrate

152 Interpretation this. Amineh Youssefzadeh, asked which melody he chooses for his performance, quotes the answer of a bakhshi from Iranian Khorasan: The performer has hundreds of maqāms at his disposal. Depending on the mood of the audience and the choice of poems he can choose cheerful (shād), burning (suznāk), martial (razmi), or melancholy (hoznāvar) tunes. In the dastans, however, melancholy melodies are preferred, since they treat of distant love. (2002: 198 trans.)2 The choice of melancholy tunes for melancholy subjects follows Morley’s prescription of a ‘sad and doleful harmony’ for ‘sighs and tears’ almost to the letter. Before discussing this and similar statements, however, we have to address the fundamental question of whether particular melodic phrases or harmonies, specific tonal spaces and musical modes have a meaning analogous to the way words and phrases have meaning. Given the complexity of the question, the following remarks can only serve as a rough orientation.

Meaning and Expression Posing the question of the relationship between words and music raises a host of further questions and problems, which are traditionally discussed in the fields of musical aesthetics and the philosophy of music. Is music a language, a semiotic system, comparable to a language system? Can music express content? In Western musical thinking, none of these and similar questions has found simple and uncontroversial answers. In his Vom Musikalisch-Schönen (On the musically beautiful), a milestone in musical aesthetics, Eduard Hanslick emphatically denied that music expresses feelings.3 It evokes feelings, but this does not imply that the feelings evoked are inscribed in the music, comparable to the way a linguistic sign carries a meaning. Music, he argued, does not represent semantic content such as grief or joy. This is not to say that music does not have content. He concedes that ‘music can, with its very own resources, represent most amply a certain range of ideas’ (10). These comprise ‘ideas which relate to audible changes in strength, motion, and proportion; and consequently they include our ideas of increasing and diminishing, acceleration and deceleration, clever interweavings, simple progressions, and the like’ (10). In other words, music has its own domain of meaning, which is inherent in sound effects and sound configurations: Der Inhalt der Musik sind tönend bewegte Formen (‘The content of music is forms of sound in motion’ trans.).4 Hanslick saw instrumental music as the central area of music; as to vocal music, he pointed out that we have here an intimate and not always clearly separable union of two arts, poetry and music. While he states

Words, Music, Meaning  153 that ‘in vocal music, the music adds colour to the black-and-white design of the poem’, he maintains that ‘it is not the tones which are represented in a song, but the text. The drawing, not the colouring, determines the represented content’ (16). To underpin his thesis, he brings examples of melodies associated with certain feelings and shows that these feelings are expressed in the words of the song and only loosely, if at all, in the music. He also lists various cases where the same melody and musical shape have been used for texts with very different, if not contrary meaning. Hanslick’s position has not remained unchallenged, partly due to his outspoken criticism of contemporary composers and styles, especially in his reviews. His denial of a musical semantics implied his refusal to see music as a language. Hanslick wrote long before the semiotic and linguistic turn of the later twentieth century. Philosophers like Susanne Langer and Nelson Goodman extended the language paradigm to all forms of art, and musicians like Leonhard Bernstein built their reflections on music on the multi-tiered structure of language, talking about ‘musical phonology’, ‘musical syntax’, and ‘musical semantics’ (in his Harvard lectures of 1973).5 Modern currents in the philosophy of music and theory of aesthetics are primarily concerned with questions of philosophical semantics, cognition, and ontology. The distinctions that are made are therefore generally tied up with a specific approach to larger philosophical questions and not easily transferable. In order to sharpen the question about the relationship between word and music in our examples, I will, however, comment on a few distinctions. In his Languages of Art Nelson Goodman discusses ‘a picture of trees and cliffs by the sea, painted in dull grays, and expressing great sadness’ (1976: 50). The picture represents objects and expresses feelings. Both representation and expression are denotative, but differ in their domain (objects and events in the one case, feelings in the other). On closer analysis, however, the picture does not denote sadness, but rather exemplifies it, for instance by having a predominance of the colour gray. The relationship between the picture and the emotion is therefore metaphoric, established by an element (such as the colour gray) which is taken metaphorically (‘sad’). Moving from a picture to music we are, of course, dealing with a non-representational art form. Here a piece of music exemplifies certain musical configurations. These might include a minor key, a slow movement, and a long, drawn-out tune on the oboe with a pianissimo chordal accompaniment. If in the musical culture in question such a musical configuration is associated with certain feelings, such as sadness, the piece of music metaphorically expresses sadness. The idea of expression as metaphorical exemplification can be complemented by Steven Davies’ notion of ‘expressive appearance’. In his Musical Meaning and Expression Davies basically holds Hanslick’s position that music is not a language or even like a language and that musical

154 Interpretation meaning can therefore not be imputed to the musical sign in the way meaning can be imputed to a linguistic sign. When it comes to the expression of emotion in music, Davies writes: Just as a willow can be sad-looking, or a person’s face happylooking, music can present an expressive appearance in its sound (without regard to anyone’s felt emotion). This is because we experience the dynamic character of music as like the actions of a person; movement is heard in music, and that movement is heard as purposive and as rationally organized. Within musical styles, these natural propensities for expressiveness are structured and refined by musical conventions, so that the expressiveness of a work might be apparent only to someone familiar with the conventions of the relevant style. (1994: 277) The importance of musical conventions and the fact that a work’s expressiveness is only understandable by someone who is familiar with these conventions emerges also from the discussion of the music of oral epics. One of the many conventional aspects of musical expression is the assignation of emotional characteristics and moods to specific modes. In Greek musical thinking parallels were drawn between the modes and traits of character, as will be exemplified below in Pindar’s ‘Horseman’s song’. Similar correlations have been postulated for the medieval and post-medieval so-called Church modes, for the Indian rāgas, but also for the maqāms of the Arabico-Persian musical tradition. Hermannus Contractus (‘the lame’), a monk of the monastery of Reichenau in the first half of the eleventh century, writes in his De Musica about the Church modes that they have different ‘character traits’. For him the Hypodorian mode is ‘sweet, the Hypophrygian reserved or lingering, the Hypolydian mournful, the Hypomixolydian joyful or exultant, the Dorian dignified or noble, the Phrygian boisterous or dancing, the Lydian voluptuous, and the Mixolydian chattering’. He adds, with a sigh: ‘Oh, the marvellous depth of music in this respect! It opens itself to the intellect to a certain extent but cannot be spoken of except in certain superficial expressions.’6 Hermannus both underlines and relativizes the attribution of moods and characters to the modes: the relationship can only be shown (demonstrari) in superficialibus vocabulis. Other music cultures have developed similar ideas. The best known is perhaps the system of the rāgas of classical Indian music. Sir William Jones, whose fame in linguistics rests on a paper given to the East India Company in 1786, in which he established the historical kinship of Sanskrit with Latin, Greek, and the Germanic languages, also wrote on Indian music. He compared Indian music to the Greek system of modes and pointed out some of the most salient traits of the rāgas,

Words, Music, Meaning  155 especially the relationship between music and emotion: ‘Raga, which I translate as mode, properly signifies a passion or affection of the mind, each mode being intended … to move one or another of our simple or mixed affections’ (1799: 71). He also discussed the classification of rāgas according to the time of day or year suitable for their performance and commented on the rāgamālās (‘garland of melodies’), sets of illuminations that illustrate the rāgas and rāginīs, the personified melodies and their wives.7 If a traditionally assigned relationship between a mood and a ‘melody’ (using ‘melody’ as shorthand for mode and other musical configurations) is opaque to the uninitiated, this does not prevent the listener from reacting emotionally to the melody. The question is not whether a melody evokes a reaction and what it is, but rather whether there is a relationship between words and music within the musical tradition in question. In the following section I will look at melody names and ask whether they tell us anything about the character of the music for which they are used.

What’s in a Name? Pindar (518–438 bce) praises in his first Olympian Ode Hieron, the tyrant of Syracuse, as the winner of a horse race. Aptly for the situation, he says that he must crown him with the horseman’s song, adding ‘in the Aeolian mode’. The ‘Horseman’s Song’ or ‘Rider Tune’ (hippios nomos) occurs in his odes also under the name ‘Castorean’ (Kastoreion), with reference to Castor, the renowned tamer of horses. Pindar was not only a poet, but also a professional musician, who played the phormynx and the lyra. Despite these skills, we have, unfortunately, no possibility of knowing what his ‘Rider Tune’ sounded like and what the connection between the melody and its name was.8 In one passage he connects the singing of the Kastoreion with the playing of the seven-toned phormynx, but that does not throw any additional light on this question. We know a little more about the Aeolian mode (also called ‘Hypodorian’) from Greek theoretical writings and can identify it as one of the scales within the Greek system of modes. In Greek thinking, music has an ethical dimension and the modes are the expression of certain moods and qualities of character. As Aristotle puts it in his Politics: ‘We accept the division of melodies proposed by certain philosophers into ethical melodies, melodies of action, and passionate or inspiring melodies, each having, as they say, a mode corresponding to it.’9 According to Heracleides of Pontos (fourth century bce) the Aeolian mode is named after the Aeolians, who are known for their arrogance, but also their courage, their horse-breeding and their hospitality. Pindar’s choice of the Aeolian mode for his ‘Horseman’s Song’ seems to have been well taken.

156 Interpretation Names of melodies have also been collected from living oral epic traditions. Here, compared to the interpretation of Pindar’s ‘Horseman’s Song’, we are standing on somewhat firmer ground, but, as will be seen, many questions remain. Not all epic traditions have melody names and the names recorded are not always easy to understand in their import and relationship to both text and music. As examples I have chosen melody names used by Tibetan bards in the performance of the Gesar epic and by Turkish and Azerbaijanian minstrels for the sung passages of the Köroğlu epic. The Tibetan Gesar As in other prosimetric epics, in the Tibetan epic of Gesar the prose passages are recited and the verse passages sung to various melodies. This alternation of speaking and singing applies both to professional oral singers (sgruṅ-mkhan) and to persons reciting and singing the epic from a book. In Chapter 6 it was mentioned that for ‘melody’ the Tibetans use the word rta ‘horse’. This expresses the idea that the melody is the carrier of words and ties in with the horse-symbolism connected to musical instruments in many traditions, such as the Mongolian morin khuur or the Kazakh and Karakalpak qobïz. Mireille Helffer, who has studied Tibetan Gesar melodies in detail (1977, 1979, 1982), states that the various songs are often preceded by a formula that announces the name of the melody to which the verse passage is to be sung. Several dozens of melodies are known; in a manuscript of one of the early episodes of the epic, ‘The Horse Race’ (Rta rgyug), Helffer counts twenty-seven melody names in fifty-six songs. In 1979 Mr. Gyurme Agyitsang, a Tibetan living in Switzerland, sang fifteen melodies from the first three chapters of a block-print of the epic in the Department of Central Asian and Tibetan Studies of the University of Bonn. Some of his melody names occur also in Helffer’s lists, as, for example, ‘Submission of the Multitude’, ‘Appeal to the Gods’, or ‘The Short Song of the Hero’.10 Helffer’s analysis of the performances of two verse passages A and B by two different singers led to the following results: (1) the first singer sang the two passages to two different melodies, as did the second singer; but the latter’s two melodies were different from those of the first singer; (2) the melody of passage B sung by the first singer (‘Submission of the Multitude’) to verses put in the mouth of Gesar, was sung by yet another singer to verses put in the mouth of ’Brug-mo, the young woman Gesar was courting (1979: 105). Helffer concluded that it seems ‘that oral tradition transmits to a given singer (and perhaps to others in a certain geographic region) a certain number of melodies or tune-types which for him are at least theoretically linked to certain protagonists of the epic’. And she added: ‘We have seen that in practice these melodies are used fairly freely’

Words, Music, Meaning  157 (109, trans.). There might be a conventional relationship between tune and text, but this relationship seems to be loose and flexible. Turkic Terms for ‘Melody’ Melody names are also found in various Turkic traditions. For ‘melody’ different terms are used, among them hava (Turkish, Azeri), nama (Karakalpak), kuy (Uzbek), and makam (Turkish). Sometimes several terms are used and sometimes the same term can have different meanings. The only word of Turkic origin in this list is kuy, which simply means ‘melody’, but can also denote a specific genre of ‘programme music’ (see below). All the others have an Arabic etymon: hava comes from Arabic hawā’, ‘air’, nama comes from Arabic naghma, ‘melody, sound’, and makam from the Arabic word for the musical mode, maqām. The maqām system of Arabian, Near Eastern and Central Asian (Uzbek, Tajik, Uyghur) music is a system of different musical modes, which have an essential similarity to the so-called Church modes of Gregorian chant and the Indian rāgas mentioned above. This means that a mode is not only defined by its tonality as it is expressed by a scale with a specific sequence of intervals, but also by its tonal range and melodic formulas. The maqām system forms the basis of the traditional art music of the various musical cultures where it is found. In a number of these musical traditions maqām suites are composed, as for instance the nawba of Northern Africa, the shash maqām [six maqāms] of the Uzbeks, or the on-ikki muqam [twelve maqāms] of the Uyghurs.11 Art music and folk music exist side by side in these musical cultures, with mutual influences, but generally as separate types of music. A case in point is the music of the Turkish minstrel. Many âs¸ıks use the term makam instead of hava. Their use of this term has been variously interpreted. Ursula Reinhard and Tiago de Oliveira Pinto leave no doubt about the difference between Turkish folk and art music: On no account can the makams of the minstrels be equated with those of art music, to which they do not belong at all, or at most only in traces. In some âs¸ık songs no more than the scales of art music makams can be identified, but only in fragments, since the ambitus of âs¸ık songs is often less than an octave. The construction that is obligatory for an art music makam can practically never be seen.12 Köroğlu in Turkish and Azeri What is of interest in the present context is the fact that the minstrels give names to the different melodies they sing. The Turkish folklorist

158 Interpretation Ferruh Arsunar, in his edition of a Turkish version of Köroğlu (1963), provides about eighty musical notations for the various verse passages. Some of these melodies are called Köroğlu makamı, but others have more telling names, for example: Öğüt havası (Advice Melody), Hasret havası (Yearning Melody), Nikâh havası (Marriage Melody), Cengi havası (Battle Melody), Meydan havası (Battlefield Melody), Yiğit güzelleme (Praise of the Young Hero), and Yiğit ağıdı (Lament for the Young Hero). All of these melodies are related to specific incidents in the story, some with clear emotional overtones (as hasret ‘yearning’) and some with generic (poetic and musical) affiliations (e.g., güzelleme ‘praise’ or ağıt ‘lament’). Reinhard and de Oliveira Pinto have classified the makams of the Kars region, where over 150 are said to be used by minstrels, into various groups and commented on them on the basis of musical analyses and interviews with âşıks.13 With regard to the relationship between melody and words they observed that makams in the style of uzun hava (‘long melody’), that is, rhythmically and metrically free melodies, are sung with a firm, often pressed voice and end in a deeper register. These makams are only used for sad or yangın (‘burning’) poems. Examples are the Derbeder makamı, the makam of the (homeless and sad) vagabond, and Kerem makamı, the makam of the tragic hero Kerem (see Chapter 2). A recording of the Derbeder makamı in the performance of Âşık Şeref is available on a Smithsonian Folkways CD.14 As the accompanying booklet puts it, ‘the singer begins with a loud voice on one of the top notes. Then he carries the melody in long, descending sequences with many more or less expressive melismas and ornaments to the concluding tone at the bottom. The melodic phrases are expanded widely and ornamented with trill, sobbing and slurring’ (43). What can be heard on the recording, can also be seen in the transcription of the melody of another poem in this makam in Reinhard and de Oliveira Pinto’s study (134–35; recording No. 40 on the accompanying cassettes). They quote Âşık Şeref’s insistence that both makams must be recognizable, despite the variations effected by different âşıks. This tells us that in the thinking of Turkish minstrels specific tunes not only bear specific names, but are also associated with specific content. Despite the meaningful names of some of the melodies (e.g., ‘Battlefield Melody’), it still remains to be seen whether the music actually ‘mirrors’ (exemplifies and metaphorically expresses) the semantic content of the words. A look at the music of the Azerbaijanian minstrels (as¸ıq) will be helpful. In the Azeri tradition we also find names for the various melodies used for the verse passages in the epics or romances. Èmina Èl’darova stresses that these names provide a semantic indication of the songs and often characterize not only the melodies, but also their poetic form and

Words, Music, Meaning  159 the voice register. The latter is explicitly indicated by specifying the frets on the accompanying saz, called perde, in the full melody name (1984: 59). She notes that with one exception every verse form of as¸ıq poetry has its specific tune, linked to the metre and content of the verse, ‘which is incontestable evidence of the unity of poetry and music in the art of the as¸ıq’ (62 trans.). The exception is the qos¸ma, a poem in quatrains, which is of wide currency and hence found with different melodies (see also below). A detailed discussion and analysis of melodies from the Azeri dastan Koroğlu (Turkish Köroğlu), based on recordings collected between 1978 and 1981, is provided by Tariel Mamedov.15 Among others he exemplifies the melody named ‘Misri’ (or ‘Koroğlu Misri’/‘Misri Koroğlu’), a melody also analysed by Èldarova. It occurs in different contexts, generally of a war-like character. One of the dastans with the Misri Melody is Koroğlu’s Journey to Derbent. Koroğlu is invited by Möminə, the beautiful daughter of the pasha of Derbent on the western shore of the Caspian Sea, to visit her. Koroğlu follows her invitation, marries her, but then returns alone to his mountain fortress, leaving her pregnant. If she gives birth to a son, when grown he is to search for his father, proving his identity with the bracelet given by Koroğlu to Möminə. A boy is born and at the age of eighteen rides out to seek his father. Near Koroğlu’s abode the boy meets three of Koroğlu’s companions, whom he defeats and binds in fetters. When Koroğlu hears of this, he rushes to the scene and draws his Misri qılınc, his sword with an Egyptian blade. Unlike the Rustam-Sohrab episode in Firdausī’s Shāh-nāma, however, the father-son fight ends without bloodshed and concludes with a merry feast. The Misri Melody, taking its name from the blade of the hero’s sword, is used for the dialogue between father and son, which comes in the middle of their fight. The verses are in stanzas of four lines. They are in the form of qos¸ma, which means that each stanza has the same rhymeword in the fourth line; sometimes the whole fourth line is repeated throughout the poem as a refrain. The rhyme-pattern is aBaB cccB dddB and so on. The lines are twelve-syllable lines. Koroğlu and his son sing alternate stanzas. The dialogue belongs to the genre of hәrbә-zorba (‘threat’), which is used in minstrel contests when two antagonists face each other. The Misri melody is a recitative-like melody, with one note per syllable, some melismatic ornamentation on the syllable ey, and sung rubato in a fast tempo. The melody is also defined by the music of the accompanying saz, with its specific rhythmic and melodic structure.16 There is, therefore, clearly a connection between melody and content, even if the Misri Melody is also used for other passages and the hərbə-zorba genre is sung and played to more tunes than the Misri Melody.

160 Interpretation Case Study: Alpamïs In 1999, when Uzbekistan celebrated 1000 years of Alpāmish, a number of versions of this epic were edited and re-edited. One of them is a Karakalpak version which was recorded on tape in 1956 from the singer Qïyas-jïraw Qayratdinov (1903–1974) (Mirzaev & Karāmatov 1999). In Karakalpak (and Kazakh) the eponymous hero is called ‘Alpamïs’. The edition of this recording contains twenty-nine verse passages with their music. For every verse passage the title of the melody or nama is given. These titles are of different kinds. Some refer to the protagonist speaking (Alpamïs Melody, Gülbarshïn Melody), some to a singer from whom the tune comes (Shangqay Melody), some to the situation in the story (Homecoming Melody), some to a melody type or genre (Long Speech Melody, Tolghaw Melody), and some come from other epics (Jump-to-the-Moon Melody).17 In order to illustrate the shape of these melodies and some of the problems connected to their occurrence in the epic, I will discuss the Jump-tothe-Moon Melody. ‘Jump to the moon!’ (Aygha shap!) comes from a passage in the epic Qoblan. Before Qoblan was born, the hero’s mother had the desire to eat a tiger’s heart. Her husband went out hunting and when he encountered a tiger, he coaxed the tiger to rise up and, in order to be able to take aim, called out: ‘Jump to the moon!’ I recorded Qoblan in its entirety twice from Jumabay-jïraw (November/December 1990 and June 1994). In the second version Jumabay left the episode of the tiger hunt out, in the first version the episode is present; in the long verse passage the lines with ‘Jump to the moon!’ come at the end and are performed in a declamatory recitative. The music of the verse passage as a whole is part of Jumabay’s repertoire of melodies; it is melody type A according to my analysis of his melodies used in Edige (see Chapter 7). In Alpamïs the ‘Aygha shap!’ melody makes its appearance in two passages. In Qïyas-jïraw’s Alpamïs it comes in the second part of the dastan, when a witch-like old woman tricks the hero and sets a trap for him. We have two musical notations for this verse passage performed by Qïyas-jïraw, one in the 1999 ‘Alpāmish 1000’ edition and the other in a collection of jïraw melodies, edited by T. Adambaeva.18 In the version of the epic written down from other singers, this melody occurs in the first part of the epic. The Kalmyk Qarajan gets on his horse, but the horse refuses to advance. This incident comes shortly before Qarajan and Alpamïs meet for the first time, an auspicious meeting that will inaugurate a life-long friendship. In the verse passage sung to the Jump-to-the-Moon Melody Qarajan asks his horse Qara-at why it is standing still. Two musical transcriptions of this passage from two different singers are found in Adambaeva, one from Dawlet-jïraw Shamuratov (1882–1965) and one from Jaqsïlïq-jïraw Sïrïmbetov (1947– 2003).19 In 1983, I recorded this second verse passage of the Alpamïs epic from Jaqsïlïq-jïraw.

Words, Music, Meaning  161

Illustration 14  Jaqsïlïq Sïrïmbetov (1983)

At the time of the recording, he told me, however, that the name of the melody was ullï ziban havasï, ‘Long Speech Melody’ (a name reminiscent of the Turkish uzun hava). To recapitulate: We have two texts sung to the Jump-to-the-Moon Melody, Text 1 the words of a deceitful crone and Text 2 Qarajan’s words addressed to his horse. Text 1 was twice notated from Qïyas-jïraw’s performance; Text 2 was also twice notated, once from Dawlet-jïraw and once from Jaqsïlïq-jïraw. A third version of Text 2, performed by Jaqsïlïq-jïraw, was recorded by me. These two verse passages are also found in other versions of Alpamïs, for which, however, no audio files or musical transcriptions are available. I will begin with Text 2 in Jaqsïlïq-jïraw’s version (see Audio/Video Ex.6)20: Alïp edim al menen Jüyrik degen dang menen, äy, Alpïs tuwar mal menen, äy. Ne körding, qara at, ne körding? Omïrawdan köbik shashbaysang, Kernip qädem baspaysang, Ne körding qara at, ne körding, äy? Aldïngda jatqan el me edi? Körgening bizden zor ma edi? Omïrawda qalqan bar ma edi? Ne körding qara at, ne körding, äy? Maqpaldan saylap dorba ildim. Kishmishten saylap men berdim.

162 Interpretation Haywan bolsangda, enaghar, Öz basïma tenggerdim. I took with strength, A race horse decked with fame, äy, Among sixty female animals, äy. What have you seen, black horse (Qara-at), what have you seen? As you don’t drip foam from your chest, As you don’t progress, making your chest wide, What have you seen, black horse, what have you seen, äy? Are there people in front of you? Have you experienced sorrow from us? Is there a shield on your chest? What have you seen, black horse, what have you seen, äy? I chose a bag of velvet and fastened it. I chose kishmish (seedless grapes) and gave them to you. Even if you are an animal, by God, I have made you equal with me. The notation edited by Adambaeva agrees essentially with what I recorded:

Example 8.1  Alpamïs

Jaqsïlïq Sïrïmbetov’s melody is predominantly syllabic, but has also long sustained notes and trill-like ornaments. It stays within a fifth, ending in D, also in the other stanzas. His teacher was Qïyas-jïraw, but to

Words, Music, Meaning  163 what extent Qïyas-jïraw’s singing and playing style influenced him is difficult to assess on the basis of the available material. When looking at the other melodies, we notice that Jaqsïlïq-jïraw’s and Dawlet-jïraw’s melodies differ for the same text (Text 2). This is perhaps to be expected; what is, however, surprising is that the two melodies notated of Qïyas-jïraw’s performance of Text 1 not only differ from Jaqsïlïq-jïraw’s and Dawlet-jïraw’s melodies for Text 2, but also differ from one another. In other words, the ‘Aygha shap!’ Melody is sung differently by different singers and can even be sung differently by the same singer for the same passage. The differences between the two notations of Qïyas-jïraw’s performance of Text 1 are certainly striking. Of course, one might argue that ‘sameness’ is a concept in need of interpretation and also that the transcribers might have used different methods of capturing sound in notes. The latter is undoubtedly the case, and as Qïyas-jïraw’s two performances took place at different times, variations are to be expected. Even if the two melodies are re-written so that that they are in the same key and have the same ‘pulse’, it is still difficult to see an agreement in their melodic contours.21 In the absence of available audio recordings, the question of sameness has to remain pending. This is also true when one compares Jaqsïlïq-jïraw’s ‘Aygha shap!’ Melody with the melodies notated from Qïyas-jïraw’s and from Dawlet Shamuratov’s performances: they certainly look different, and despite allowing for a wide margin of variation and notational approaches it is difficult to reconcile the melodies as we have them in writing with one another. In conclusion, it can be seen that the use of melodies by the jïraws parallels that of the Tibetan, Turkish and Azeri singers mentioned earlier: a traditional connection between melody/melody name and verse passage is implied, but free rein is given to individual variation and to adaptation to the performance situation.

Imitation The examples discussed so far have shown that in some cases a certain melody name is associated with a certain melody, which is used for a specific passage. In other cases, however, a melody name designates different melodies; furthermore, a text can be sung to different melodies, or, conversely, the same melody can be used for different texts. The relationship between these three units, melody name, melody, and text, turns out to be tenuous. Nevertheless, it can be observed that despite these fluctuations there is a traditional link between them, affirmed by both singers and audience. The question whether a particular melody (with its name) is apposite to a specific text has not yet been posed. In this section I will therefore ask whether there is a correspondence between the meaning of the words and the music to which they are performed. The clearest connection between music and extra-musical content is established by imitation. Musical systems are like languages regulated

164 Interpretation by conventions. A linguistic sign is as a rule a conventional sign. It is by convention that speakers of English use the word table when they talk about a particular kind of object, for which speakers of Turkish use the word masa or of Hebrew the word shulkhan. Some linguistic signs, however, are considered natural, at least up to a point. Ernst Gombrich humorously remarks about onomatopoeic words like cock-a-doodle-doo in his Art and Illusion: To me, at least, the cock says not ‘cock-a-doodle-doo’ as he calls to the English in the morning, nor ‘cocorico’, as he says in French, nor ‘kiao kiao’, as in Chinese, but still ‘kikeriki’, as he says in German. Or – not to fall into the mistake of Socrates – it is not precisely ‘kikeriki’ he says; he still speaks cockish and not Viennese. My percept of the throaty noise of his call is distinctly coloured by habitual interpretation. (2002: 306–7) The imitation of sound in music is equally onomatopoeic, but also equally influenced by convention. Onomatopoeia can be taken to the limit by the incorporation of sounds from nature into music, such as the song of a nightingale produced from a gramophone record in Ottorino Respighi’s third movement of the Pini di Roma (1924). The singing of birds is a constant imitative element in classical music, in Vivaldi’s Quattro Stagioni (1725), for instance, in Ralph Vaughan Williams’ The Lark Ascending (1914) or in Olivier Messiaen’s Catalogue d’oiseaux (1956–58), which is based on the composer’s meticulous transcriptions of birdsong. Vivaldi’s Four Seasons is also known for more sophisticated imitations of the sounds of nature, such as the gentle wafts of the wind or the rumbling of a thunder storm. The sounds of our technical civilization have not been excluded from musical onomatopoeia, such as the chugging along and gathering speed of a locomotive engine in Arthur Honnegger’s Pacific 231 (1923). Onomatopoeia is also found in oral epics. It is found in the music and in the words. In Kyrgyz oral epics, especially in Manas, onomatopoeic rhyme-words are often linked into chains; it can be shown that these rhyme-chains act as generators of verse-lines much in the way formulas function in the oral epics of the South Slavs and other epic traditions (Reichl 2020). These onomatopoeic rhyme-sequences are also found in other Turkic traditions, as for instance in Uzbek dastans. An example from Alpāmish will illustrate this. The following lines come from a scene in which Alpāmish is riding through the desert to the land of the Kalmyks, where he will take part in the suitor contests for Bārchin’s hand. The ride through the desert is a typical scene in Uzbek dastans. It begins in Fāzil Yoldāsh-oghli’s version of the epic with the following lines: Dubulgha bāshda dongullab, Kark qubba qalqān qarqillab,

Words, Music, Meaning  165 Tilla pāyanak urilgan, Uzangilarga shirqillab. Bedāw ātlari dirkillab, Ālghir qushdayin charqillab, Qolda nayzasi solqillab. (Mirzaev & Abdurakhimov 1999: 135) The helmet on his head is ringing, The bulging shield, made of rhinoceros hide, is resounding, The tip of the golden scabbard is beating Against the stirrups and rattling. The courser is racing forward, Flying like a bird of prey, The spear in his hand is shaking. The rhyme-words, printed in bold, are all verbs built with the derivational suffix -illa- (or -ulla-) and put into a gerundival form, marked by the b-ending. This derivational suffix is characterized as onomatopoeic in Uzbek grammars. It expresses sound effects and also, as in our passage, forms of movement. Strictly speaking, only verbs like dongulla- ‘to ring’, qarqilla- ‘to resound’, and shirkilla- ‘to rattle’ are onomatopoeic. Verbs using this suffix to designate motion, as in dirkilla- ‘to race forward’, charqilla- ‘to fly’, and solqilla- ‘to shake’ partly imitate sound (the whizzing of a horse racing forward, the sound of a bird flapping its wings) and partly suggest a trembling, fluttering, or shaking movement. Their meaning is hence based on both onomatopoeia and convention. Words of this kind are found in many languages, especially in African languages where they have caught the attention of linguists. Africanists generally use the term ideophone for them, a term defined by C. M. Doke as ‘a vivid representation of an idea in sound. A word, often onomatopoeic, which describes a predicate, qualificative or adverb in respect to manner, colour, sound, smell, action, state or intensity’ (1935: 118). Sound imitation in passages like the one quoted from Alpāmish is, however, restricted to the words. It is true that the verses are sung in a sprightly, animated manner and that the strumming of the dombira marks a quick rhythm and could be interpreted as imitative of a horse galloping through the desert. It has to be conceded, however, that the melodic and rhythmic patterning of this passage is also found in verses which don’t have onomatopoeic rhymes and don’t describe a ride or some other kind of movement. The same is true of the Kyrgyz onomatopoeic rhyme-chains in Manas. They are generally performed to a recitative melody, ‘Melody A’ as illustrated in Chapter 6; the sound-shape of the rhyme-words is clearly audible; their onomatopoeic power works through language and elocution rather than through a special melody.

166 Interpretation Imitation in music is not only based on onomatopoeia. Marenzio’s imitation of the phrase scendi dal paradiso by descending notes mentioned above is based on the analogy of the downward movement of a body with the downward movement of sound from a higher to a lower pitch. Similar analogies and metaphors underlie works like Monteverdi’s dramatic cantata Il combattimento di Tancredi e Clorinda (The duel between Tancredi and Clorinda) of 1624. The cantata is based on the Renaissance epic Gerusalemme liberata (Jerusalem Delivered) by Torquato Tasso (1574), specifically on Canto XII. The crusader Tancredi fights against a young Saracene woman, Clorinda, who is disguised as a knight and therefore not recognized by Tancredi, who had earlier fallen in love with her. Tragically, she is killed in the duel. The dramatic intensity of Tancredi’s fight with Clorinda finds its musical equivalent in Monteverdi’s stile concitato ‘agitated style’, the repetition of short, single notes (with a tremolo-like effect), a strong rhythmic patterning and other sound modifications. In addition to sound painting (the clatter of hoofs, the clanging of swords), the musical effects in Clorinda’s death scene, where pizzicati and pianissimo suggest the intermittent breath and the increasing stillness of the dying, owe perhaps more to analogy than to sound imitations. Strongly dependent on musical symbolism of this kind is the Kazakh küy, a narrative genre, where, however, the tale is not told in words, but only in music. Like the Uzbek word kuy, Kazakh küy means ‘melody’, but is, like Kyrgyz küü, also used as a genre term, denoting a particular kind of instrumental piece, either for plucked or bowed string instruments. These pieces, more specifically those that musicologists call ‘küy-legends’, tell a story that is assumed to be known to the listeners. The instrumental piece focuses on ‘illustrating’ the main episodes of the story. Sometimes the singer also supplies the narrative thread in words. In the küy ‘Aqquw’ (The Swan), for instance, a boy, seeking food for his mother, his little sister and himself, spots a swan and aims at it. The swan is wounded, and the beating of its wings as the bird is flying over the water is imitated in the music of the accompanying instrument. Other elements of the story expressed in the music are the happiness of the little sister at the return of her brother (in a lively, ‘hopping’ melody), the lament of his mother (in a slow melody in a deeper register), and the fight of the boy with a group of enemies.22 Imitative elements of this kind are also typical of the music of Kazakh oral epics. Boris Erzakovich, who was the leading specialist of the music of Kazakh epics, underlined the presence of küy-like sound effects especially in the instrumental interludes. While the sung melodies are generally of what Erzakovich calls, the ‘recitative-declamatory type’, the instrumental interludes are largely imitative, ‘portraying the din of battle, horse-races, the cackle of birds and the shouting of people’ (2003: 419). They take these content elements up from the narration and render them in musical form in the manner of the küy.23

Words, Music, Meaning  167

Leitmotifs in Siberian Oral Epics Among Turkic oral traditions, the unaccompanied performance of the Kyrgyz manaschï is paralleled by that of the Yakut olonkhosut. A first hearing gives the impression of a singing style quite similar to the Kyrgyz jeldirme, the ‘trotting pace’. On closer analysis, however, we discover a surprising variety of melodic shapes and a technique that bears a strong resemblance to the use of leitmotifs in Western music. The term ‘leitmotif’ is mostly associated with Richard Wagner’s operas, in particular his Ring of the Nibelungs. Here we find recurring melodic-harmonic figures characterizing different persons, objects, or events, such as the Ring-motif, the Sword-motif, or the Valhalla-motif. Wagner didn’t actually use this term, but used instead expressions such as melodisches Moment (melodic moment) or Grundthema (basic theme). The idea of the leitmotif is, however, much older than Wagner’s operas; it made its appearance in Western classical music already in the eighteenth century. Yakut oral epics are peopled with a great number of mythic personages. They move within the three-tiered cosmological framework of Yakut mythology, the Nether World, the Middle World, and the Upper World. The positive heroes and heroines are headed by the ayïï, benevolent divinities who protect the human inhabitants of the Middle World, the negative heroes and heroines by the abaasï, evil spirits and demons. The central action of the olonkho consists in the fights between Good and Evil, which generally end with the defeat of the abaasï and the settlement of the Middle World by the tribes of the Uraangkhay-Sakha (the Yakuts). The various characters of the olonkho comprise heroes and heroines, monsters and demonic female shamans, gods, spirits, animals, and others (see Chapter 9). These protagonists take centre stage in long monologues. It is in these monologues that specific melodies and singing styles are used, and it is here that we find the leitmotif-technique. Yakut epics are in verse, occasionally interspersed with passages in prose. The metre conforms to the metrical system of Turkic popular poetry, that is, it is syllabic. As in the case of Manas, there are lines with seven or eight syllables, but there are frequently also shorter and longer lines, ranging from four to twelve or fourteen syllables. Similar to Kyrgyz epic poetry, line-initial alliteration is a common occurrence. Yakut epics are generally fairly long; while shorter epics have between 2,500 and 3,000 lines, others extend to between 10,000 and 20,000 lines; in the 1960s an epic of more than 52,000 lines was written down.24 While the descriptive passages are declaimed, the protagonists’ monologues are sung. About a third of an olonkho is taken up by sung monologues. The declaimed passages (sometimes spoken, sometimes sung, mostly on one note) are delivered at a fairly quick pace, while the monologues are sung at a slower speed. Their performance takes twice as long as the performance of the recited parts (see Pukhov 1975).

168 Interpretation The poetic and musical character of an olonkho depends on the skill of the olonkhosut. There is also a certain amount of variation due to regional styles. While the most talented singers endeavour to compose new melodies for each olonkho, many olonkhosuts perform the different epics in their repertoire with the same melodies, with only slight changes. Although first notations of olonkho-melodies were already published by nineteenth-century travellers and geographers such as A. Th. von Middendorf, the study of Yakut epic music only began in the twentieth century, with audio recordings by Waldemar Jochelson at the turn of the century, followed by musical transcriptions by Soviet musicologists from the 1930s onward.25 In the course of time the musical transcriptions have become more sophisticated; for simplicity’s sake, however, I will illustrate these ‘leitmotifs’ by the comparatively ‘unadorned’ transcriptions provided by Viktor Belyaev (1937). Belyaev published five melodies, the ‘leitmotif’ of the Hero of the Middle World, that of the Hero of the Underworld (abaasï), that of the Old Witch (Dyägä-Baaba), that of the Old Slave-Woman (Simäkhsin-Ämääkhsin), and that of the Tungus Hero. For the Hero of the Middle World the following melody is given:

Example 8.2  Yakut (1)

The melody is rhythmically clearly structured: the eight syllables of the verse are evenly distributed over two 4-beat bars. The first two syllables of each bar consist each of two notes, of which each is preceded by a short ornamental note, a third above (appoggiatura), which bears the ictus. This creates a specific effect, somewhat similar to a trill or a tremolo. More elaborate musical notations capture this trill-like effect by writing, instead of four notes per syllable, as above, as many as thirty-two notes.26 For the melody of the Hero of the Underworld the following example is given:

Example 8.3  Yakut (2)

This melody is very similar to the previous one, in particular with regard to the appoggiatura. It is, however, in a deeper register (as is expressed by the bass clef) and has a different end to the phrase. The melody jumps

Words, Music, Meaning  169 down to the last note about an octave lower. Belyaev remarks that it is ‘like a groan of indeterminate pitch, as if coming “from below the ground”’. The melodies for the Witch and the Old Slave-Woman are almost identical, but the motif of the Witch is sung with a high, vigorous voice, while the motif of the Slave-Woman is sung with a feeble, delicate voice. Finally, the motif of the Tungus Hero is exemplified by a somewhat livelier and rhythmically different melody:

Example 8.4  Yakut (3)

These examples give the impression that Yakut epic melodies have a fairly simple melodic contour and conform to a basic rhythmic regularity. When one compares different notational renderings of epic melodies (and listens to the recordings of different epic singers), it becomes clear, however, that there is considerable variation. Also, the leitmotifs are anything but static; they change according to the state of the hero (the hero challenging to battle, the defeated hero etc.). Two song styles are differentiated, called in Yakut dägäräng ïrïa and dyärätii ïrïa, respectively. The word ïrïa means ‘song’ and dägäräng ‘walking on tip-toes’; as a descriptive musical term it denotes a melody with a clear rhythmic structure and a recitative-like tune without ornamentation. The word dyärätii is derived from a verb meaning ‘to give out a ringing sound’ and denotes an ornamental, often rhythmically flowing melody.27 As we have seen in Chapter 5, the Yakut epic singers do not gesticulate like the Kyrgyz manaschïs, but sing and recite the epic in a sitting position, holding their hands still. The drama of their tale resides not only in the story, but also in the music. The various figures of the epic are, as in any other tale, characterized by description and by what they say; but in the olonkho they are also characterized by how they speak, or rather how they sing. Singing in a high or a low voice, fortissimo or pianissimo, with a full or a creaky voice is used for mimetic purposes. Negative characters sing and speak in exclamations, in a jerky manner, pronouncing words unclearly and slurring them over, while positive characters sing in a slow and ceremonial manner. In addition, we have the leitmotif-effect, the association of a range of characteristics with a particular melody used for a particular character. Music is here a means of deepening the impression produced by the words of the tale by assigning a musical label, as it were, to the different protagonists. This label also utilizes mimetic elements, in form and performance: high pitch vs. low pitch, loudness vs. quietness, evenness vs. unevenness in the melodic contour, resonant vs. creaky voice. Yakut singers also use a singing technique that is called kïlïhakhtaakh

170 Interpretation ïrïa and glossed as ‘singing with guttural appoggiaturas’; it bears some resemblance to the throat-singing of the Tuvans discussed in Chapter 4. The use of leitmotifs characterizes the music of oral epics also in other Siberian and Asian traditions. It is found in the singing of the Tibetan Gesar epic and in the performance of Mongolian and Tungus epics. To illustrate this, I will briefly discuss the use of leitmotifs in Tungus, more specifically Evenki, oral epics. The Evenki live in a wide area stretching from northern and eastern Siberia to northern China; in Russia, the largest group of Evenki live in Yakutia, in China in Inner Mongolia. Their traditional life style in the tundra is that of reindeer herders.28 The Evenki epic (or nimngakaan) consists like the Yakut olonkho of narrative passages and monologues (sometimes dialogues). The narrative passages are spoken, the monologues are sung. The text of the epic shows a patterning into syntactic-rhythmic units (lines), which comprise a varying number of syllables and are irregularly linked by line-initial alliteration (anaphora), rhyme/assonance, and parallelistic structures. The segmentation into verse-lines is clearest in the sung passages, where the same melody (or some variation of it) is generally repeated for every group of words interpretable as a verse-line. Every song is introduced by a string of meaningless words, sung to a particular tune. Words and melody identify the speaker. My illustration comes from an Evenki epic (Irkismändya), recorded from the singer N. G. Trofimov (1915–71).29 Trofimov also performed, in addition to Evenki nimngakaans, Yakut olonkhos, but used different melodies for the Evenki and the Yakut epics. Nevertheless, looking at Evenki epic melodies, one can see that these two Siberian oral traditions have much in common. The Evenki nimngakaan Irkismändya is a lengthy and involved tale, which features a great number of characters. It is a kind of family saga that comprises six generations, of which about thirty family members take part in the action. In addition, it is peopled by a number of protagonists from the Lower World. All these characters have the gift of transforming themselves into different shapes, mostly birds, ranging freely not only over the Middle World, but also into the Upper and Lower Worlds. The plot consists of a series of challenges for the various heroes and heroines, challenges that are generally linked with bridal quests or the repulsion of an unwanted suitor from the Underworld. The tale is structured as a succession of monologues connected by short narrative passages. In this epic the hero’s ‘signature tune’ is sung to the words Dyngdonidyngdoni, dälägäy!, his wife Kukkumachan’s to Kidu-kidu, kiduyar!, his son’s to Goldyr-goldyr, goldyrmoy!, and so on. In the edition of the epic there are about three dozen different identifying lines. From an analysis of seven sung sections of this epic by A. M. Ayzenshtadt and Yu. I. Sheykin (1990), I will illustrate two. The first is the leitmotif of the hero of the Upper World, Khurkokchon:

Words, Music, Meaning  171

Example 8.5  Tung\us (1)

In this passage, the hero greets the parents of a shamaness called Ayakchan, introduces himself and then asks for their daughter’s hand. The melody of the opening line is characterized by the tremolo effect of most notes. With the exception of the first and the last syllables, every syllable is sung to a group of notes circling around a central note, a halftone below, a half-tone above, and with additional grace notes. The various lines of the extract are sung to the same kind of melody, creating the auditory impression of a repeated melody, with only slight variations. The melody stays within the fairly narrow ambitus of a major third. The second example is the leitmotif of Däge-Baba, the Underworld abakhii (evil spirit):

Example 8.6  Tungus (2)

This melody is in approximately the same tonal range; there is, however, a clearer, better articulated melodic flow. The melody has less of the tremolo effect and the melismatic circling of notes; it is predominantly stichic. The lines following in this passage are sung to the same melody, with fewer variations than in the previous example. As in the case of the Yakut olonkho, there is both variation and diversity in the use of leitmotifs in Evenki nimngakaans. In summary, I would like to quote a passage from Ernst Emsheimer’s study of Siberian epic music, in which he underlines the versatility of the Yakut olonkhosut, a characterization that is also true of the Evenki epic singer: According to Pukhov, we no longer have here simply a primitive imitation of sounds, but a great talent for observation and a skill acquired over many generations. And the sounding characterization of the individual acting persons in the epic events is realized above all through the tone quality of the voice. The voice, its timbre and shades of expression, correspond to their inner qualitative characteristics. Thus, for example, the monologues of the hero, since he as the protector of humans is strong and mighty, are sung in a melodious

172 Interpretation high bass, those of monsters and evil spirits, on the other hand, in a rough and repulsive voice. As for the heroine, the personification of female beauty and grace, the Olongkhosut sing her monologues in a high register; for those of the hero’s wife they sing in the contralto, etc. In this way the Olongkhosut characterize in sound and melody now birds and animals, now heroes, terrifying and fantastic monsters or gods and spirits. (1991: 160)

Expression and Convention Similarities between different non-Western musical traditions, often reinforced by the assimilating effect of world music, should not blind us to the existence of distinct musical worlds, with connotations and aesthetic systems that might differ significantly from one another. This is especially true when it comes to assigning a meaning, or an emotional value, to a melody, a mode, a tonality, or some other configuration of sound. In an Occidental musical context, folksongs in a major or in a minor key often differ in their connotations, where songs in a major key are felt to convey a joyful mood, while those in a minor key are thought to express sadness. This is, of course, a gross generalization, and browsing through a collection of English or German folksongs will bring to light many counterexamples. Nevertheless, the warning Abraham Idelsohn gave in his discussion of the Arabic maqam system is worth remembering. He notes that in their own musical milieu maqams with scales similar to a minor scale are not considered sad, nor are maqams with scales similar to a major scale considered joyful (1992: 27–28). Given this caveat, it is nevertheless true that singers, their audience, and also researchers affirm a connection between music and content or mood. No doubt detailed musical analysis can reveal the musical structures which are interpreted as denoting a certain concept, mood, event, or object. These musical structures, however, operate within the musical system in question and are not immediately intelligible to the outsider. Goodman’s theory of exemplification and expression and Davies’ term ‘expressive appearance’ underline the conventional character of musical expression. Perhaps the closest connection between verbal content and music can be established for onomatopoeic effects. The imitation of the sound of the hoofs of a galloping horse striking the ground by the rhythmical strumming of a plucked chordophone is certainly persuasive. Nevertheless, even onomatopoeia is based on convention. A better case for the assertion of a match between verbal content and music can be made for certain genres incorporated into the epic. In the Uzbek epic of Alpāmish the returning hero is recognized by his wife in the course of the exchange of wedding songs, sung at the feast prepared for her remarriage. These songs, which are also found in other versions of the epic,

Words, Music, Meaning  173 are characterized by their refrain yār-yār ‘beloved, beloved’ at the end of the second and fourth line of every stanza, sometimes also after every line. They are addressed to either bride or groom, also alternately to one and the other; they can also be sung as contest songs between men and women. They are generally sung in a lively tempo; the melodic phrase of the refrain sets the tone of the song. The content of the wedding songs is often humorous, but they can also express praise of the bride or groom and even sorrow on the part of the bride about leaving her home. The music of these passages incorporated into the epic marks them as wedding songs and indicates a festive mood; given the variety of content, however, no close connection between words and music can be established. In Kyrgyz, the lament (qoshoq) and an elegiac song called arman occur in the epic of Manas. These songs are clearly marked as sad by their content and the occasion of their singing. Closely related to funeral laments is a song called qïz uzatuu ‘sending a girl away’, which is sung when the bride is sent away from her home to that of the groom. V. Belyaev writes about the qoshoq: ‘The lament is generally cast in an improvisatory verse form with varied repetition of a single sorrowful tune. Falling glissando turns of a fourth below the tonic are often used as cadences’ (1975: 3). The ‘sorrowful tune’ is, however, in need of explanation. A. V. Zataevich, a collector of Kyrgyz and other Central Asian folk music in the early twentieth century, notes about an arman he wrote down and transcribed: ‘In this “Arman of a young man” we have sorrowful contents, but, as in a number of other specimens, the melody is here light, optimistic and even humorous’ (1971: 405, no. 244 trans.). Zataevich’s impression is corroborated and explained by K. Dyushaliev: It is typical of the music of the Kyrgyz qoshoq (by both men and women) that independent of their mournful contents they always sound as if they were in a major scale. These scales are in Kyrgyz traditional musical thinking not always associated with happy phenomena and emotions. This is in our view accounted for by the fact that major scales have been most characteristic of all Kyrgyz traditional music and that they deeply permeated the whole scalar-intonational system of the Kyrgyz in history. (1993: 103 trans.) From these examples we can deduce that while these genres (yār-yār, qoshoq etc.) have a recognizable musical form, this form is only loosely linked to the meaning of the words and the mood of the song. As, however, the musical form unambiguously points to a particular genre, it acts as a kind of label. There is some similarity to the leitmotifs in Yakut and Siberian epics. They identify the protagonists, even if they don’t always characterize them musically. Some of these leitmotifs, however, also have imitative elements such as the rendering of the wicked laughter

174 Interpretation of a character by a series of interval jumps and tremolos or trills, or the choice of a low register for heroes of the Underworld. What makes a melody into a leitmotif in these epics is, however, not the presence of imitative elements, but the conventional linking of musical sign and epic hero (or villain). The Fulani Motto This identifying or indexing use of epic music is also found in the oral tradition of the Fulbe. Christiane Seydou has written extensively on the music of Fulani oral epics, where she finds imitative and symbolic elements in the music, but also a different kind of relationship between word and music. About the imitative and symbolic elements, she writes in the introduction to her edition of the epic Silâmaka and Poulôri as performed by the singer Tinguidji: For example, every time a march, a move from one place to another, is evoked, there reappears a melody that is appropriate for the representation of a journey, monotonous and colourless like the very paces – while the tune of a battle paints a kind of imperious gallop with richer, more imaginative variations, which ends in the self-willed and jerky rhythm of the excited vultures, tearing apart and ransacking the corpses. (1972: 48 trans.) Her edition, and translation, is accompanied by a record, on which Boûbacar Tinguidji’s performance can be heard. Like other Fulani singers (mâbo or maabo), he accompanies himself on the hoddu, a threestringed lute. The hoddu provides a constant, rhythmical accompaniment of the recited text. The epic is not composed in metrically defined verselines, but rather in ‘breath-units’, that is, in chains of words, syntagmas and phrases, which are spoken, or rather recited at great speed, in ‘one breath’. These ‘lines’ form passages of irregular length. Between the text passages instrumental interludes are played. While the accompaniment of the recited lines comprises only a few melodies and gives the impression of a fairly unvaried musical background, the interlude melodies are more diversified and more numerous (see Chapter 6, pp. 126–27). In Fulani epic music imitative and symbolic elements are found in instrumental music and not in the performer’s singing. The dominant musical role of the hoddu in Fulani musical culture is also underlined by the use of the motto in both music and poetry. According to Seydou a motto (or devise in French and jammore in Fulani) is found both in the epic text and in the music.30 Textually it can be described as the largely formulaic praise name of an exalted person, such as a chief or the hero of an epic. It is recited by the mâbo to the accompaniment of the hoddu.

Words, Music, Meaning  175 Musically it is a melody that identifies the bearer of the praise name. Seydou explains: In traditional Fulani society, one of the main functions of the maabo griot is to play on his lute the musical motto of his ‘master’. This is the head of a family to which, from father to son, he is bound in a very particular relationship of alliance and clientship that is founded on mutual dependence: economic dependence for the griot, social and psychological dependence for the ‘master’. The effect of this musical motto on the latter is made obvious by the verb designating the griot’s performance: yarnude, ‘to water’, literally ‘to make (someone) drink’ (his music). For it is not a matter of hearing or listening to one’s own musical motto, but of being ‘watered’, ‘irrigated’ with it, the verb yarnude evoking well that impression of physical impregnation and this feeling of the soul being flooded. (2000: 214) ***** I began this chapter with the opinion of a Khorasanian bakhshi about the choice of cheerful, burning, martial, or melancholy melodies for poems of the corresponding mood. Like the Turkish minstrels, the Khorasanian bakhshis use the term maqām, which, however, is not identical to the concept of the maqām as a mode in Persian-Arabic art music. It has its own musical physiognomy, comprising scales and their composition, modal characteristics, and the use of the various perde/parde (frets on the dutar/dotār) in the construction of the melodies. The interplay of verse form and musical form, of singing and playing, of melodic and rhythmic patterning are constitutive elements of the various maqāms. Amineh Youssefzadeh’s detailed analysis (2002: 197–260) reveals a number of features that are also found in the oral epic traditions of Central Asia. As the discussion of various examples in this chapter shows, the bakhshi’s affirmation of a relationship between melody and mood is not easy to verify. The presence of imitative elements in the music can establish a certain connection to the meaning of the words in a passage or the mood of a scene in the narrative. Melodies can be used as labels for the characters in a tale, such as the Yakut leitmotifs, some of which also incorporate imitative elements. The music of specific genres (lament, marriage-songs etc.) connotes the mood of the genre and has hence a certain expressive force in the context of an epic. Much of the alleged correspondence between music and content in oral epics depends, however, on the conventions of a musical tradition. In the light of the previous discussion, I would like to return to the Khorasanian bakhshi’s views about music and mood and look at one of the melodies Amineh Youssefzadeh analyses.

176 Interpretation My example comes from a dastan of the Köroğlu cycle and is a melody named Jolan Kuroghli, ‘Köroğlu’s gallop’. As described with reference to Uzbek and Turkmen bakhshis (see Chapter 7, pp. 144–45), the Khorasanian bards also move in the course of a sung verse passage into a higher register, called by Youssefzadeh ‘the second modal part of a maqâm’ and marked as B. This is, as it were, the culmination point of a verse passage. Youssefzadeh remarks that in the case of the Jolan Kuroghli Melody, which is considered a heroic melody (qahremâni) by the bards, the song begins directly in the second part of the mode, which corresponds to the structural component B. Each stanza begins with an exclamation on the expletive yâr on the seventh degree of the mode. One can by the way think that it is this beginning on the culminating point that gives this melody its warrior aspect. (205 trans.) Once again the assignment of a relationship between music and verbal meaning or mood occurs within a musical tradition and is based on the conventions established there. This does not exclude the possibility that, in the case of imitative elements, many of these can be recognized and appreciated not only inside a musical tradition, but also from the outside. It is not unlikely that universals in ‘musical poetics’ exist, certain transcultural associations triggered by sound effects such as loud versus quiet, high versus low, or fast versus slow. When we look at Western music, in particular at lied compositions, we witness a great variety of setting words to music. With some of the Classic and Romantic composers we feel that their lieder have captured the meaning of the words admirably. In some cases we can perhaps think of a poem only in its musical form. And yet these songs, which seem to express the sentiment of the poem so naturally, so effortlessly, are the product of a long historical development and rest on the conventions of Occidental music just as the Khorasanian bakhshis’ choice of a maqām is embedded in their musical culture.31

Notes 1 See Crocker 1986: 214–16; Milner 1963: 138–41. 2 The bakhshis of Khorasan are generally tri-lingual and capable of singing the dastans in Persian, Kurdish, and Khorasan Turkic; the Turkmen bards sing in their own language only. 3 Hanslick 1982 [1854]; the English translations are taken from Hanslick 1986 (G. Payzant). 4 Hanslick 1982: 74. The phrase tönend bewegte Formen has been variously translated. Langer 1976 translates ‘dynamic sound-patterns’ (225), Payzant ‘tonally moving forms’ (Hanslick 1986: 29), Rothfarb and Landerer translate ‘sonically moved forms’ (Hanslick 2018: 41). 5 See Bernstein 1973; Langer 1976; Goodman 1976; on Goodman, see Robinson 2001.

Words, Music, Meaning  177 6 Ellinwood 2015: 139 (Latin text on p. 138). 7 For a concise introduction to the rāgas, see Wade 1980; on North Indian rāgas, see Ruckert and Widdess 2000; for a survey of mode in music, see Powers et al. 2001. 8 See Race 2012: 56 (text), 57 (translation [equestrian tune]); on Pindar as a musician, see West 1992: 344–47. 9 The quote comes from Book 8, Chapter 7 of the Politics; Jowett & Twining 1952: 218; in his discussion Aristotle distances himself from Plato’s restrictive view as expressed in his Republic. 10 I am grateful to Mr. Gyurme Agyitsang for his permission to record his performance; Institute of Central Asian Studies, University of Bonn, 3 July 1979. Other melody names are: ‘Song of Long Life’, ‘Song of the Nine Lions’, ‘Song of the Cuckoo’, ‘Song of the Lark’, ‘Long and Gentle Melody’, ‘Song of the Snail Trumpet’, ‘Braveness of the Tigress’ etc. See also the list in Helffer 1982: 236–37. 11 As in the case of the Indian rāgas, this is a huge and complex area, which has been much researched. For short introductions, see Marcus 2002 (Arab), Signell 2002 (Turkish), Caton 2002 (Iranian), Sumits and Levin 2016 (Uzbek and Tajik), and Harris 2016 (Uyghur). 12 Reinhard and de Oliveira Pinto 1989: 84–85 trans. By ‘construction’ the progression or development (seyir) in a makam is meant. Compare also Öztuna’s statement that folk musicians outside the tradition of Turkish classical music might use makams, but are not familiar with the system, nor with the makam progression (seyir) (1990: I, 323). 13 One of their main informants was Âs¸ık Şeref Tas¸lıova (1938–2014), with whom I took part in the Harvard Symposium of 2010 on ‘Singers & Tales in the 21st Century: The Legacies of Milman Parry and Albert Lord’; see Reichl 2016b. 14 Song Creators in Eastern Turkey. Traditional Music of the World 6. SF 40432 (1993), track 12. 15 Mamedov 1984; for a list of Turkish and Azerbaijanian (including Iranian Azerbaijan) Koroğlu/Köroğlu melodies, see İmamverdiev 2013; for full musical transcriptions of Azeri and Turkish Koroğlu/Köroğlu melodies, see Mamedov 2011: 455–600. 16 The text is found in Abbaslı & Abdulla 2005: 316–27. A full musical transcription of this passage is found in a collection of Mamedov’s works in Azeri (Mamedov 2011: 455–66). The Misri Melody is one of the favourite epic melodies of Azeri as¸ıqs, also in Iranian Azerbaijan. 17 The tolghaw is a historical poem, generally with didactic and moralizing elements; it is sung to a distinctive melody, also called tolghaw. 18 This verse passage is edited in Ayimbetov et al. 2007: 466; the musical transcriptions are found in Mirzaev & Karomatov 1999: 157–68, and Adambaeva 1991: 201–4. 19 For the verse passage, see Ayimbetov et al. 2007: 427–28; for the two musical transcriptions, see Adambaeva 1991: 204–7 (Dawlet-jïraw) and 149–52 (Jaqsïlïq-jïraw). 20 Jaqsïlïq-jïraw’s text in Adambaeva (A) and in my recording (B) have both the same first stanza; stanza II in A is stanza IV in B, stanzas III, IV and V in A are missing in B, while stanzas II and III in B are missing in A. The latter two stanzas are not found in other edited versions, while all of A’s stanzas are also found in other versions. 21 There is no space here for a detailed discussion; the melody in Adambaeva has two flats, is predominantly written in 3/8 time, with syllables sung on quavers

178 Interpretation and semiquavers; the melody in Mirzaev and Karomatov has one sharp and is written in 3/4 time, with syllables mainly sung on minims and crotchets. 22 See Dzhumalieva & Temirbekova 2000: 256–59 (with notated examples); Elemanova 2012: 212–19; Daukeyeva 2016, and, on the Kyrgyz küü, Nyshanov 2016. For audio-examples, see the website of Levin, Daukeyeva & Köchümkulova 2016 and one of the examples on UNESCO’s Intangible Cultural Heritage website; see Appendix C. 23 On the music of Kazakh oral heroic epic and romance, with musical notations, see also Erzakovich et al. 1982 and Erzakovich 2000. 24 See Pukhov 1962; Illarionov 1982: 11. 25 For a survey with musical transcriptions, see Reshetnikova 1993. 26 See Reshetnikova’s example no. 5, 1993: 62; compare also Emsheimer 1991: 160–61. 27 On the music of Yakut oral epics, see also the early paper by Peyko & Shteynman 1940. On the music of Är Soghotokh (an olonkho discussed in Chapter 9), see Alekseev 1996. Aiza Reshetnikova’s comprehensive work on the music of the olonkho (including Tungusic oral narrative poetry) is collected in Reshetnikova 2005. 28 Another Tungusic group is that of the Even or Lamut, who live predominantly on the northern shore of the Sea of Okhotsk. Both the language of the Evenki (in Russian èvenkiyskiy) and that of the Lamut (in Russian èvenkiy) are endangered. 29 This epic, comprising over 2,000 verse-lines, is edited, and translated into Russian, in Varlamova & Mireeva 2008: 108/109-206/207. The two examples can be listened to and downloaded from the website http://­evenkiteka. ru/stellages/folklore/evenkiyskie-geroicheskie-skazaniya-napevy-iz-­ skazaniya-irkismondya-bogatyr/. For the first example, click Сторона 1 [Side 1] and choose the first track 1 Напев Хуркокчона – богатыря Верхнего мира; for the second example, click Сторона 2 [Side 2] and choose the first track 1 Напев Дегэ Бабэ – богатырки Нижнего мира. 30 See Seydou 1972: 31–38, and, mainly from a linguistic point of view, 1977. 31 The relationship between words and music in Western music has been much discussed by musicologists. For the medieval period, see especially Stevens 1986 (with a detailed analysis of narrative music); for a historical survey of the relationship between words and music in Western classical music, see Winn 1981.

9 The Singer and the Tale

I have strange power of speech … (Coleridge, The Rime of the Ancient Mariner) Coleridge begins his Rime of the Ancient Mariner with a short dramatic scene: an ancient mariner stops a wedding guest on his way to the feast, first holding him ‘with his skinny hand’ and then ‘with his glittering eye’. The guest is compelled to listen to the old man’s tale, a terrible and moving story. When the tale is ended, the wedding guest turns away from the place of merry-making, ‘like one that hath been stunned, And is of sense forlorn’. He has become ‘a sadder and a wiser man’. Coleridge intensifies the horror of the story by putting it into the mouth of the sailor who was guilty of causing the catastrophe. The introduction of a narrator as a figure in its own right is a favourite literary device. Frame-stories such as The Thousand and One Nights, Chaucer’s Canterbury Tales, Boccaccio’s Decameron, but also novels like Charlotte Brontë’s Wuthering Heights or Joseph Conrad’s Lord Jim come to mind. There are at least two narrating voices here, that of the frame-tale and that of the inserted tale put into the mouth of a character. The interplay of tales and narrating voices can become fairly complex, in particular when there are various narrating figures and various degrees of impersonation and reliability on the part of the narrators. Writers have exploited these possibilities, and narratology has developed a number of conceptual frameworks for the analysis of these possibilities. In this chapter I will focus on narratological questions as they apply to orally performed, rather than written narratives, such as: What is the relationship of the bards to their tale? Is it something they have experienced or present as having experienced in a first-person narrative, or is it a third-person narrative, relating to events lying outside personal experience? What is the relationship between the time of the narrator and the time of the tale? Is the story placed in the present, the recent past, the distant past, or a time of mythic beginnings? Are first-person narratives incorporated in a tale told in the third person? Do singers dramatize

180 Interpretation themselves in the tale as narrators? Do they comment on the action, perhaps in the style of Dickens or Thackeray?

Point of View It hardly needs stressing that the great variety we find in oral epic traditions implies a certain diversity of narrative modes and interactions between narrator and tale. I will begin with an example which can be considered typical of a wide array of oral epic traditions. In September 1995, I recorded the epic Sharyar in a private home in the town of Shomanay in Karakalpakstan (Uzbekistan) from the singer Jumabay-jïraw Bazarov. Here was the singer, sitting on the carpet in the middle of the room, with his qobïz in his hand, beginning his performance with the spoken words: Äne, äyemgi ötken zamanda, Zamannïng qädim shaghïnda Neshshe ughlï musïlman ötti, Neshshe ughlï qalmak ötti. Musïlman wälayatïnda, Dar islamnïng quwatïnda, Shähäri paytakht yurtïnda, Qang dere boyïnda Khan Därepshah ötti. Now then, in times long past, In the days of yore, There lived many Muslims, There lived many Kalmyks. In a Muslim country, Under the rule of Islam, In his capital, in his country, On the banks of the river Qanglï, There lived Khan Därepshah. The story is presented as a third-person narrative, which is placed in a particular geographical region and set in a time long past. The quotation comes from the transcript of the recording. Both transcript and recording are physical traces of a narration that occurred at a specific point in time and space. They allow us in their different ways to access the narration, but are clearly outside the narrating event. This difference between event and its reflection in transcript or recording is captured by Gérard Genette’s distinction between narration, the act of narrating, and récit, the narrative discourse or, in the case of written literature, the text.1 The singer in the performance situation is an overt outside narrator. Genette calls the outside narrator an extradiegetic narrator (from Greek

The Singer and the Tale  181 diḗgēsis, ‘tale’). At the same time, the tale he tells has a narrating voice. In the performance situation the overt outside narrator and the narrating voice of the tale are conflated. Once the performance is over and it has been textualized, the situation changes. Now we can only infer the outside narrator. We can attempt to reconstruct the outside narrator of the performance by the field-worker’s notes and possible audio and video recordings, but we can also look for clues in the text that has been written down. Here we are in the presence of a narrating voice in much the same way as in any other written narrative. ‘In the days of yore there lived Khan Därapsha’ points to the same type of narrator as, for instance, the beginning of Henry James’ Daisy Miller: ‘At the little town of Vevey, in Switzerland, there is a particularly comfortable hotel.’ On closer analysis, the basic similarity of the third-person narrator of Sharyar with that of Daisy Miller has to be qualified. While the narrating voice is that of a covert narrator, in both cases the narrator does, however, also make an appearance as an overt narrator. In Daisy Miller this is obvious by the occasional intrusion of a narrating ‘I’, in Sharyar by the very first word (äne), which will be commented on below. The point of view also differs. In the epic, the story (Genette’s histoire) is presented from the point of view of a narrator who overlooks the narrated events from a detached vantage point. In Daisy Miller, on the other hand, the story is told from the point of view of one of the actors; the reader is made acquainted with the action through the prism of the narrator’s understanding of persons and events. An authorial narrative situation in Sharyar contrasts with a figural narrative situation in Daisy Miller, where the narrative is presented through the eyes of a character. While Jumabay does consistently stay with a third-person narration and does not use ‘I’, he is nevertheless present also as a voice in the narrative discourse. The very first word äne, ‘now then’, is indicative of the performance situation and of the fact that the narration is addressed to an audience. This particle closely corresponds to the initial interjection of the Old English epic Beowulf, ‘Hwæt!’, etymologically the origin of Modern English what, but here obviously an exclamation to attract the listeners’ attention. Seamus Heaney translates hwæt as ‘so’ and explains: Conventional renderings of hwæt, the first word of the poem, tend towards the archaic literary, with ‘lo’, ‘hark’, ‘behold’, ‘attend’ and – more colloquially – ‘listen’ being some of the solutions offered previously. But in Hiberno-English Scullion-speak, the particle ‘so’ came naturally to the rescue, because in that idiom ‘so’ operates as an expression that obliterates all previous discourse and narrative, and at the same time functions as an exclamation calling for immediate attention. (Heaney 1999: xxvii)

182 Interpretation The call for the audience’s attention arises naturally from the performance situation. It is common in medieval texts that are believed to have flourished in an oral milieu and to have been performed by the medieval popular entertainer, so-called ‘oral-derived’ texts. A great number of Middle English romances, for instance, begin with an appeal to the audience to pay attention. The same is true of Old French chansons de geste and Middle High German epic poems. As these narratives, however, are only preserved as texts, a performance situation with an outside narrator can only be inferred from such textual marks of orality; furthermore it can never be ruled out that these marks are signs of fictive orality, introduced into texts which are composed in writing in the manner of oral romances and epics. Having signalled by äne that the tale is about to begin, the narrator places the action of his story or histoire into the distant past. The expressions the singer uses are formulaic and are found also in the openings of other epics, not only of this singer and not only in Karakalpak epics, but also in other Turkic epics.2 One finds similar beginnings in the oral epic traditions of Asia, the Near East, Africa, and the Indian subcontinent. In 1981, when I first met the singer, Jumabay began the performance with the same two lines, but then continued with Adam and Eve, from whom the Muslims and unbelievers (Kalmyks) have descended. In a version written down in 1951 (published in 2009) he again begins with the same two lines, but then has instead of the following two lines about the Muslims and Kalmyks the line ‘at the time of (religiously) lawful food’. Similar variations are found in the versions that have been written down from other singers. Öteniyaz-jïraw places the action in Yemen among the Arabs, while Qulemet-jïraw, like Jumabay in one of his renderings of the epic, mentions Adam as the ancestor of both ‘irreligious heathens’ (dinsiz käpir) and Muslims, and qualifies Khan Därepshah as ruler of the Noghay.3 The variations in Jumabay’s versions seem to arise in the performance situation as alternatives available to him on the basis of his store of phrases, formulas, and ready-made poetic lines. They reflect his understanding of the story, but can hardly be interpreted as expressions of a shifting point of view. Despite the varying geographical setting of the story in different versions, they all agree that a tale is told that happened a long time ago. There is a clear distance between the time and world of singer and audience and the time and world of the tale. In this epic the time of action remains vague; it all happened long ago, but quite clearly within a timeframe that provides some fixed points: the advent of Islam is presupposed and, in one version, the period of the Noghay Horde is indicated. For both singer and traditional audience these fixed points might, however, have been rather vague and in no way congruent with the historian’s dates. The story itself has no historical anchoring (unlike, for instance, the epic of Edige). Sharyar is the elaboration of various motifs associated with the

The Singer and the Tale  183 folktale of ‘The Three Golden Children’ as it is found in The Thousand and One Nights. A well-known version of this folktale is Pushkin’s poem ‘The tale of Tsar Saltan, of his son, the glorious and mighty knight Prince Guidon Saltonovich, and of the fair swan-princess’.4 The separation of the time of the narrative from that of its narration has long been recognized as a characteristic feature of the epic. Time, and its representation and presence in narrative, has been a central concern of twentieth-century philosophical and narratological thinking. Paul Ricœur’s three-volume Temps et récit is no doubt a landmark in this field, both by its comprehensiveness and its penetration.5 The role of time specifically in the epic has been repeatedly analysed and interpreted.6 An important voice in narratology has been Mikhail Bakhtin, whose distinction between the epic and the novel, based on the relationship between narrated time and narrator-time, has been widely accepted. Bakhtin refers to Schiller’s and Goethe’s qualification of the epic as vollkommen vergangen (completely past), but he is also in line with Hegel’s idea of the epic or, as Hegel called it, the epopeia. Bakhtin’s argumentation in his ‘Epic and Novel’ can be summarized as follows7: (1) the epic is a poem about the past; (2) the narrator of the epic (both the outside narrator and the narrating voice) and the listeners (both outside and implied) are on a different temporal plane from that of the characters and events of the epic; (3) the two planes also comprise different value systems (world-views); (4) the temporal gap is unbridgeable; (5) if both narrator and listener on the one hand and the characters and events of the narrative on the other hand are on the same time and value plane, we have left the realm of the epic and entered the realm of the novel. For Bakhtin the narrated time is the time of a heroic past, ‘it is a world of “beginnings” and “peak times” in the national history, a world of fathers and of founders of families, a world of “firsts” and “bests”’ (1981: 13). Bakhtin has the heroic epic in mind and sees it as a reflection of a period that is seen by the singers and their audience as a crucial period in the genesis of an ethnic group, a tribe, or a nation. He formulates an idea here that is based on the heroic epics of Classical antiquity, but which can also be substantiated by modern oral epics. Sharyar, our example, is in content closer to romance than epic. But even here the events are situated in the period of the Noghay Horde in one version. This period – historically the fifteenth and sixteenth centuries, when, after the fall of the Golden Horde, the Noghay Horde was the major force in the PonticCaspian steppe – is seen as the period of tribal and eventually national origins of a number of Turkic-speaking groups, in particular the Noghay, Kazakhs and Karakalpaks. Edige is the ‘foundation epic’ of a number of these ethnic groups, celebrating the deeds of Edige, a commander of the Golden Horde who was instrumental in establishing Noghay rule.8 When looking at epics from oral traditions, we see that there are various degrees of depth regarding the time of the narrative and also different types

184 Interpretation of ‘past time’. The narrative can be placed in different layers of the past, ranging from a primeval ‘beginning of the world’ to an unspecified ‘time long past’, but also to a period evoked by historical figures or events in the tale. No doubt tabulating these different degrees and types has an intrinsic worth and contributes to a better understanding of the variety of oral epic traditions. Of more interest, it seems to me, is a discussion of whether Bakhtin’s claim that there is a clear caesura between the ‘now’ of singer and audience and the ‘then’ of the tale is borne out by the evidence of oral epics. I will begin with the epic tradition of the Yakuts of northern Siberia.

Mythological Epics, Sacred Time Yakut oral epics were first written down in the middle of the nineteenth century; the bulk of epics were, however, only recorded in the twentieth century. About 130 full versions of olonkho have been preserved, together with over 100 excerpts and about 300 audio and video recordings. The Yakut epics show traces of a number of archaic features of the Turkic world, in particular in their mythological world-view. The three-tiered cosmos according to Yakut mythology, made up of an Upper World, a Middle World and a Lower World, has already been mentioned (see Chapter 8, p. 167). The central theme of Yakut epics is the formation of the Yakuts, of their various tribes, and of the involvement of their ancestral heroes of the Middle Earth with the good and evil spirits of the Upper and Lower World. According to N. V. Emel’yanov three basic plot groups are found in Yakut epic poetry: (1) the settlement of the Middle World; (2) the ancestors of the tribe of the Uraangkhay Sakha (Yakuts); and (3) the heroes protecting the tribe of the ayïï (benevolent spirits) and the Uraangkhay Sakha. In a number of epics, the main protagonist is a heroine, who is either instrumental in settling the Middle World or acts as defender of the land of the Yakuts. These groups, comprising a total of seventy-five plots, have various subgroups; one of these is ‘solitary heroes as ancestors of the tribe of the Uraangkhay Sakha’. The name of the hero is Är Soghotokh, which means ‘solitary hero’.9 This is also the title of the first Yakut epic published. It is found in the Yakut grammar written by the Sanskritist Otto Böhtlingk, which appeared in 1851.10 According to this version the hero of the epic is literally alone in the world. When he asks the spirit of the Divine Tree about his destiny, he is told that he was brought from the Upper World to the Middle World to become the progenitor of the Yakuts. The spirit gives him a bladder filled with the water of life and asks him to bind it under his left arm. Är Soghotokh departs on a journey that takes him to the realm of Kharakhaan Toyon at the far end of the world, where it borders on heaven. Kharakhaan Toyon has a beautiful daughter, Khotuuna. An evil spirit (abaasï) of the Lower World, Buura Dokhsun, threatens death and destruction if Khotuuna isn’t handed over to him. He gives

The Singer and the Tale  185 Kharakhaan Toyon a respite of nine days. In this period Är Soghotokh arrives and fights with his rival for Khotuuna’s hand. The abaasï Buura Dokhsun mortally wounds Är Soghotokh, but as his arrow pierces not only the hero’s heart, but also the bladder filled with the water of life in his armpit, Är Soghotok is brought back to life and his strength is doubled. He kills Buura Dokhsun, marries Khotuuna, returns with his wife to his own land, begets children, and becomes the progenitor of the Yakuts. In this olonkho the plot is fairly static. We hear of Är Soghotokh’s journey to the land of Kharakhaan Toyon and there is a fierce fight between the hero and his abaasï opponent, but most space is given over to descriptions. These descriptions are highly patterned; they are cast in a language rich in formulaic diction, syntactic parallelism, anaphora, and hyperbole. At the beginning Är Soghotokh’s strength is characterized as so great that if he leans against a tree, it becomes uprooted and topples over, that if he walks on frozen ground it cracks, while when he rides on his horse, craters open under his hooves. ‘His word is thunder, his breath is wind, his shout is a storm, his look is lightening.’ Khotuuna’s face is silver, her red cheeks gold, her eyebrows dark silver. She is also transparent: one can see her body through her dress, her bones through her skin and the marrow in her bones. Whoever looks at her will begin to smile, whoever is touched by her will be sated, and her hand will bring luck. Buura Dokhsun, however, is a figure from a horror story: he rides a three-legged iron horse without skin; he is the size of a frozen larch, with a head like a cauldron, an eye in the middle of his forehead, with two teeth like chisels, with only one fingerless hand the size of a shovel and one leg the size of a post. He is clad in iron and has an iron bow, a stonearrow, and a copper spear. Much space is given to the description of the fauna and flora of Är Soghotokh’s dwelling place as well as of his balaghan, the Yakut logcabin with pyramid-like slanting walls. These descriptions of nature are traditionally found at the beginning of an olonkho. They evoke a primeval time, with an abundance of vegetation and wildlife. These highly formulaic opening lines set the time and place of the events which unfold in the epic. I will illustrate this from another version of Är Soghotokh, a version recorded in 1982 from the singer V. O. Karataev (1926–1990). This text comprises well over 6,000 lines and the plot is consequently more involved than that of Böhtlingk’s version.11 Karataev’s version of 1982 begins with the lines:


Bïlïrghï dyïlïm Bïstar mïndaatïn Bïdan ïnaraa öttüger, Urukku dyïlïm Okhsuhuulaakh uorghatïn Otoy annaraa öttüger,

186 Interpretation






Aaspït dyïlïm Anïskhanaakh aydaannaakh künün Adyas anaraa tahaatïgar, Kuopput dyïlïm Kudulghannaakh kulan ölüü uorghatïn Kuoharakaakh khonnoghor Kihi aymaghïm Käpsätän bilsä iliginä, Sakha azmaghïm Sangarsan dyaahïya iliginä … (Illarionov et al. 1996: 76) In the remote past Of the distant crests Of my former years, In the days of yore Of the stormy peaks Of my early years, In the irretrievably past time Of the windy and noisy days Of my bygone years, In the long chain Of the deadly crags Of my vanished years, When my tribe of humans Did not yet know how to speak, When my tribe of Sakha Had not yet mastered speech …

The chronological placement continues for a few more lines, after which the singer gives a long description of ‘mother earth’, a paradisiac setting with lush flora and fauna. Many parallels from other Yakut epics can be quoted, but one example will suffice. The olonkho Kuruubay Khaannaakh Kulun Kullustuur (‘Obstinate Kulun Kullustuur’) opens with the similarly worded lines: Bïlïrghï dyïl bïdan mïndaatïgar, Urukku dyïl kulan uorghatïgar Tüört sakha törüü iliginä. Üs sakha üösküü iliginä, Ikki sakha iitillä iliginä … (Pekarsky et al. 1985: 8) In the distant crests of past years, In the remote crags of bygone years,

The Singer and the Tale  187 When no four Sakha (Yakuts) had yet been born, When no three Sakha had yet grown up, When no two Sakha had yet been raised … In such formulas the singer-narrator distances himself or herself from the time of the tale. It is set in a mythic past, separated from the present by immeasurable stretches of time. While it is clear that there is a break between the primordial past and the present in these epics, the wording of Karataev’s first lines is somewhat puzzling. Why does he speak of ‘my years’, of ‘my tribe of humans’, and of ‘my tribe of Sakha’? The use of ‘my’ in this passage seems to be idiosyncratic, but it nevertheless raises some questions. For instance, does the ‘my’ entail a change of point of view? Is this the trace of a first-person narration as can be found in epic poems in traditions where shamanism is current? Given the shamanistic background to the Yakut singer of tales (see Chapter 5), this is not impossible. We might also ask whether a mythological past has a different quality in a performance context where singers and listeners still identify with a world-view as ‘theirs’, not only as that of their forefathers, a world-view considered valuable for themselves as their cultural and spiritual heritage. In modern Yakutia, the appropriation of a mythical past in cultic ceremonies such as the annual spring festival (ïsïakh, the feast of fermented mare’s milk) is still fairly conspicuous, with shamanic blessings and rituals and the performance of orally transmitted poems and epic tales. The separation of the narrator’s time from that of the tale becomes problematic as soon as the tales have a religious content (in the widest sense of the word) and their performance is, or is close to, the enactment of a ritual. Borders become blurred, even if it is accepted that the events of the narrative are in the past. They become blurred when some of the protagonists are divine beings in whose continued existence and activity singers and listeners believe. The performer of the Rajasthani epic of Pābūjī, the bhopo, is viewed as both singer and priest. The eponymous hero is not only placed in history, but also venerated as a divine being. The borders become blurred also when the time of a mythological past becomes a ‘sacred time’, the cyclic time of a ritual performance. Mircea Eliade has underlined the difference between the cyclic time of enacted myth, the ‘sacred time’, and the progressive time of history, underpinning his concept of ‘sacred time’ with a wide spectrum of anthropological data. Many of Eliade’s observations also fit the Yakut case. It has to be remembered, however, that in present-day Yakutia a revival of traditions, including epic performance, has become part of state policy in a period when the need for the assertion of nationhood is strongly felt in all parts of the ex-Soviet Union. It is a moot question how many of today’s rituals owe their existence and shape to modern interpretative efforts rather than to the continuation of an earlier tradition.12

188 Interpretation V. O. Karataev, however, is a representative of traditional Yakut oral epic poetry. As such, he sees the mythic past of his tale as relevant to him (‘my time’) and the ancestors of the Yakuts as his ancestors. The possessive pronoun ‘my’ violates the conventions of a third-person narrative and moves these opening lines towards a first-person narrative. As a brief look at epics told from the perspective of the first person will show, Karataev’s Är Soghotokh is not a typical first-person narrative. Oral epics in which the events of the tale are consistently presented by a first-person narrator form a minority in the wide terrain of oral poetry. This narrative perspective has to be distinguished from the intrusion of the narrator in the narrative as commentator or interpreter, which will be discussed below.

First-Person Narration, Shamanistic Traces In fiction, first-person narration can take many forms. In Samuel Butler’s Erewhon (1872) we have an essentially Utopian tale presented as the adventures of the narrator, while in Goethe’s Dichtung und Wahrheit (1808–31) we have an autobiography, which is tellingly entitled ‘Fiction and Truth’, underlining by its title the subjective stance of personal recollection and presentation. We have the narrator or narrators within a frame-tale and we have the multi-perspective novel with different narrators. Variety is also found in oral epics. Genette gives as an example of his terminological distinctions Odysseus’ tale of his adventures at the Phaeacian court in books IX to XII of the Odyssey. The tale as we have it in the Odyssey is the narrative discourse (récit); when Odysseus tells his story he performs the act of narrating (narration); what he tells about his encounter with the Cyclopes, his stay with Circe, his descent into Hades, and so on, is the story (histoire) or narrative content. Such an imbedding of an extended narrative – spread over several books – told by one of the characters of the story, as in the Odyssey, we find also in oral epics of a more recent period. In Sayaqbay Qaralaev’s version of Manas, for instance, the history of one of the major characters, Almambet, is told by himself in the first person in c. 4,000 verse-lines. Almambet flees from his homeland first to the Kazakhs and later to the Kyrgyz and becomes, in the sequel, Manas’s bravest and most faithful companion. While in versions of the epic by famous singers, such as Saghïmbay Orozbaqov and Jüsüp Mamay, Almambet’s story – his origin and flight from home – is told in third-person narration, in Sayaqbay’s version Almambet’s story is told by himself retrospectively at a later stage in the chain of events. The most prominent oral epic tradition with first-person narration is that of the Ainu of Hokkaido, who were originally also found on Sakhalin and the Kuril Islands. When the British missionary John Batchelor reached Hokkaido in 1877, the Ainu still had a rich oral epic

The Singer and the Tale  189 tradition, which was, however, seriously threatened on account of the harsh Japanese measures against this ethnic group. Today not only the epic tradition of the Ainu has died out, but also their language, a language unrelated to Japanese or any other known language. Much of their epic heritage was written down by John Batchelor and later by a number of Japanese scholars in the first half of the twentieth century. Their epics first became known to non-specialist readers in the West through Arthur Waley’s 1951 translation of some passages of Kutune Shirka, an epic tale about the conflict between two tribes, triggered when the hero steals a golden otter. The tale begins with the words: My foster-brother and foster-sister – They it was who brought me up, And so we lived. The epic continues in consistent first-person narration, on which Waley comments: You will have noticed that the epic is told in the first person. This is true also of Ainu folk-tales and almost all their narrative literature. It has been suggested that this form is derived from that of shamanistic communications – the hero of the story speaks through the narrator’s mouth just as the possessing deity speaks through the mouth of the shaman. But so far as I know this form of narration does not exist in other parts of the world where shamanism is even more extensively cultivated. (1951: 235) Waley’s doubts were unfounded. Although there are a number of epic traditions where shamanism plays an important role and yet the narration is in the third person, there are also shamanistically grounded oral epic traditions with first-person narratives. There can be little doubt that the explanation for narrating the epic in the first person is to be sought in the practice of shamanism. This is the opinion of Japanese scholars who have recorded the Ainu epic heritage, but also of Donald L. Philippi, who has translated a fair sample of the Ainu epic tradition into English and who underlines the similarity between the shaman in trance and the narrator assuming the personality of the hero or heroine of the tale (1979: 28). In his ‘Towards an Anatomy of Heroic/Epic Poetry’, Arthur T. Hatto has drawn attention to first-person narratives among a number of oral traditions. Like other researchers he maintains a connection between shamanism and first-person narration in these traditions.13 Apart from Ainu oral epics, Hatto specifically mentions the epic traditions of the Ob-Ugrians (Mansi and Khanty) living to the east of the northern Urals

190 Interpretation and of the Nenets of northern arctic Russia. These different traditions all have their own physiognomy, with numerous nuances and modulations. All of these traditions are interesting not only from a narratological, but also from a performance-oriented and musical point of view. As a complement (and contrast) to the Ainu oral traditions, a brief look at the oral epics of the Nenets is of interest. The Nenets speak a Samoyedic language, a branch of the Finno-Ugric family of languages to which, among many others, Finnish and Hungarian belong. The Nenets live on and beyond the Arctic Circle in the Russian Federation; most of the less than 50,000 Nenets live in two autonomous regions, one along the Barents Sea, the other on the estuary of the River Ob on the Kara Sea and on the Yamal Peninsula north of the Urals, where their traditional life as reindeer herders is today endangered both by climate change and the exploitation of natural gas in their territory. Their folklore comprises a wide spectrum of genres, among them oral epics. Two basic types of epics are generally distinguished, syudbabts and yarabts. The genre term syudbabts, literally meaning ‘song about giants’, is derived from the word meaning ‘giant’, while the term yarabts means ‘lament’. Both of these terms are misleading; giants or monsters are by no means the main protagonists of the syudbabts, nor are they confined to syudbabts; furthermore, there is no close connection between yarabts and laments. Both subgenres of epic reflect in their stories the world of the reindeer herders and hunters of the tundra and the taiga. The epic tales are packed with adventures involving fights and rescues, attacks and pursuits, bridal quests and feasts, with superhumanly brave heroes and heroines as well as gigantic aggressors and warriors. The syudbabts and yarabts differ neither according to content nor to style, but rather by their different point of view. While syudbabts are third-person narratives, yarabts are first-person narratives. On the basis of content, three groups of yarabts can be distinguished, songs about the winning of a bride, songs about conflicts between clans and tribes, and songs about the misfortunes and sufferings of the hero. It is the last category which might have led to the adoption of the name ‘lament’ for these first-person epic songs.14 I will illustrate the yarabts by an epic of the second group, entitled Tri Yabtoné, ‘The Three Yabtoné’.15 The yarabts begins with the words: ‘I have three older brothers, the three Yabtoné’ (Nyar’ ninekav, nyar’ Yabtoné). Yabtoné is the name of a Nenets clan. The speaker, as it turns out, is a woman, the three brothers’ only sister. In her tale we hear that her brothers have a herd of ten thousand reindeer and possess seventy chums, the cone-shaped Nenets dwellings, which in winter are covered with reindeer skin, in summer with birch bark. As enemies are approaching from three territories, the brothers decide to migrate to another place. The brothers and their sister begin a long flight, pursued by their enemies, the Nganasan Seysi, Young Tasiniy, and a Tungus shaman. The Nganasan

The Singer and the Tale  191 are another group of Samoyeds with their own language, the Tasiniy are a clan living at the mouth of the River Ob. The story reaches its climax with a fight between the Younger Yabtoné and Seysi, continuing over a period of years. The Younger Yabtoné asks his sister to get help from the giant Pyryati and from Old Yamal’s son. Pyryati refuses, Yamal’s son offers his help. Seysi is killed when the Younger Yabtoné’s sister manages to destroy Seysi’s external soul, hidden as an egg in a duckling, locked in a small casket in Seysi’s sister’s possession. The Younger Yabtoné, the middle brother, who has no legs, and the sister survive; the enemies are defeated; the giant Pyryati is punished and the yarabts ends with the Younger Yabtoné’s marriage to Seysi’s sister and the first-person narrator’s marriage to Old Yamal’s son. This epic song was written down in 1959 from a male singer, Sofron Andreevich Taleev, who at the time was in his sixties. The performer, a man, tells a tale with a woman as first-person narrator. The reverse is also common, that is, that a yarabts with a male first-person narrator is performed by a woman. The tale is presented as an autobiography – ‘this is what happened to me!’ – but the actual narrator is clearly not identical with the narrating voice of the tale. This is most blatant when the sex of extradiegetic singer and intradiegetic first-person narrator differs. Such a distinction might lead one to think that the first person is nothing but a convention, a stylistic trait that doesn’t influence the way the story unfolds. This, however, is incorrect. It is noticeable how closely the telling of the tale follows the perspective of the ‘I’ that is speaking. In the yarabts here summarized, the thoughts and vision are that of the speaking heroine. On their flight, the sister loses sight of her brothers for a while, but seeing her legless middle brother (És’da Yabtoné), she finally catches up with him: I am travelling [on a reindeer sledge] for a whole day through these woods. At the end of the day I passed through the woods and reached a clearing. When I got out into the clearing, it seemed to me that I saw him and that he also noticed me, but then he vanished from my sight. I didn’t set eyes on him for a week. I thought: ‘Although he hasn’t got legs, he is strong, and if he had legs, it wouldn’t be possible to keep pace with him.’ After a week I noticed him [again]. Legless Yabtoné had apparently got weak and was sitting on a tussock, where generally an owl sits. Presently I drove up to him. After I had reached him, I stopped. Legless Yabtoné raised his hand for me to shoot at him. He said: ‘Nganasan Seysi, shoot at me.’ I tell him again and again: ‘My older brother!’

192 Interpretation He doesn’t hear my words. ‘But I’m not Nganasan Seysi.’ Will I have to talk to him for a long time? I moved my reindeer sledge to him, took him under his arms and lifted him onto the sledge. Sitting him down on the sledge, I drove off following my nose. (Kupriyanova 1965: 402/416 trans.) The listener is not told what had happened to the middle brother. In fact, all we know of the plot is either what the first-person narrator has experienced herself and tells us or what she is told, generally in direct speech, by others. Anything the first-person narrator hasn’t experienced or been told is either left unsaid or communicated as the narrator’s thoughts and conjectures. The Nenets genre of yarabts is fairly consequential in this. Another aspect that will strike the reader is the realism of the scene, also the sister’s practicality. The realism of the yarabts has been stressed by a number of students of the Nenets epic. It permeates the story, despite the hyperbolism inherent in the genre of the epic. While the younger brother is the hero of the crucial fight, it is only through the help of his sister that he manages to defeat his opponent. The sister’s braveness and power of endurance is also praised by her brothers. ‘Younger sister, you are a fine girl!’ exclaims her brother when she has destroyed Seysi’s external soul. She is indeed the heroine of the epic. When studying the Nenets oral tradition, the question arises whether Philippi’s assessment of Ainu oral epics applies to the yarabts as well. Does the epic reciter assume the personality of the hero or heroine, as in a shamanic séance? First, it has to be stressed that Nenets epics are sung. The prose-like edition and translation of these epics in Russian sources obscures their segmentation into lines which correspond to melodic lines (‘verse-lines’), as transcriptions of the music of Nenets epics show.16 Second, a characteristic of epic performance is the presence of a teltanggoda or ‘repetitor’. This person retells sung passages in prose (in the first person) in order to make sure that everything is understood by the audience. Jarkko Niemi points out that this role is related to that of the assistant (teltanggoda) in a shamanic séance, where, however, ‘this kind of mediator is needed, because the words of the spirits are considered dangerous to human ears’ (1998–1999: 95). It seems that there is a certain overlap between the figures of the shaman and the singer. In a similar way, the Finnish linguist and ethnologist Matthias Alexander Castrén, who visited the Samoyeds of the Tomsk region in 1846, drew a parallel between shamanic séance and epic performance: With almost religious devotion the audience listens raptly to every word coming from the lips of the singer. Like the shaman, the singer sits

The Singer and the Tale  193 on a stool or a chest in the middle of the room, while the listeners take their seats round him. In the province of Tomsk, I have noticed that the singer endeavours by his gestures to show a kind of empathy for his protagonists. His body shakes, his voice trembles, with his left hand he always covers his tearful eyes, while his right hand grasps the shaft of an arrow, whose tip is pointed towards the ground. The listeners generally sit there silently, but when a hero is either killed or rises on a swift eagle into the sky, they chime in with hee, equivalent to our hurrah. (Schiefner 1856: 182 trans.) Castrén is not specifically referring to yarabts. In fact, he is talking about the southern Samoyeds or Selkups and their epics. His remarks do, however, confirm what a study of different Siberian and other cultures, where shamanism plays, or rather played, an important role, suggests: namely that the singer and the shaman are often brothers or at least distant cousins.

The Narrator’s Presence in the Narrative The quotation above from the beginning of the Karakalpak epic of Sharyar began with the particle äne ‘now then’, a particle, it was argued, which is meant to formally mark the beginning of the narration and to attract the audience’s attention. Although the tale is presented as a thirdperson narrative, the singer as extradiegetic narrator makes his presence felt in the course of the narration by lifting his voice. Singers do this when they appeal directly to the audience and when they comment on the tale. Singers appeal to their listeners to solicit their attention, generally at the beginning of their narration. They might use ‘structural formulas’ throughout the narrative to guide the listeners to different scenes, before they will finish by addressing the audience directly, sometimes even the patron of the performance event in particular. Singers also address the audience with boasts of their skills and the declaration of their traditionally acquired knowledge, as illustrated in Chapter 2. As commentators, singers offer explanations and evaluations, sometimes explicitly asking the listeners to take note of some incident or some feature of the tale. For the medievalist, appeals to the audience to pay attention and be quiet are a familiar feature of epics, chansons de geste, or romances with a presumed oral background. More explicit addresses to the audience than äne are also found in orally performed epics and romances. A few examples will have to suffice. At the beginning of one of the sijobang performances transcribed and translated by Nigel Phillips, we have as the singer’s opening words: Worn is the thread, the silk is tangled, trailed over the heddles, worn in the room of the Malaccan; although, my friends, you have not asked me to,

194 Interpretation like a workman performing his task, this is just what I want to do. Hear one, friends … (Phillips 1981: 43) The Aceh epic Hikajat Potjut Muhamat (The Story of Prince Muhamat) from northern Sumatra begins in the same vein: Wonderful! Astonishing! I am going to tell a story. Listen! I am going to tell a new tale. (Drewes 1979: 39) In addition to appeals to the audience at the beginning, singers insert also various structural remarks into their tales. Some of these signal a change of plot-line, a change of place and actors. For instance, the formula used for this in the Turkish hikâye is: ‘Let so-and-so remain there and let us see what so-and-so is doing.’ In a hikâye of the Turkish âs¸ık Şeref Tas¸lıova one realization of this formula is: Let Bağdat [the tale’s heroine] grow up there. Now I want to give you news from where? From Hafız [the tale’s hero].17 Similar phrases are found in other Turkish hikâyes and also in the oral epics of other Turkic traditions. A formula pervasive in the Turkmen oral epics of the Göroghlï cycle, used both as the introductory formula of a new branch and as a marker of a scene change is Habarï kimden al? ‘About whom shall we speak?’18 In the Aceh epic about Prince Muhamat the singer frequently explicitly formulates that he is beginning a new episode and about to effect a change of place and characters19: Now I change the subject. Prince Muhamat is about to leave. (Drewes 1979: 77) Further, after this, I am going to tell another episode. Momentarily I drop the story of the Rawa war… (123) Some of these remarks do not only segment the narrative but also the performance. Occasionally the singer needs a break, as the narrator of the Hikajat Potjut Muhamat stresses: I can hardly keep my eyes open. Let me take a rest for two hours or so. … Let me interrupt here the account of the young prince for a while. (71)

The Singer and the Tale  195 As with appeals to the audience, the medievalist will find these structural formulas familiar, also the minstrel’s request for a rest and refreshments: ‘Fill a glass and give us something to drink,/ I want to begin another episode!’ exclaims, for instance, the narrator of the medieval Low German romance Valentin und Namelos.20 Just as the narrator makes an appearance at the beginning of his tale, so he or she does at the end. The Aceh epic ends with the words: Here we end our story; the tale is finished; the song of the young prince is complete. Grant me ample forgiveness for my mistakes, for I am not in the habit of composing poetry. Grant me ample forgiveness for my mistakes; that is what I hope of you, honourable gentlemen. God willing, it is finished now. The story was about Prince Muhamat. (263) The narrators/singers might also ask the listeners to remunerate them and then conclude their performance with wise words and good wishes, as is described by Alexander Chodzko at the end of his translation of an Azeri version of the epic cycle of Koroğlu: Thus ends the tale of Kurroglou, which, when an aushik [aşıq] has recounted, he never forgets to sing a poetical eulogy of his own or of anybody else’s composition, praising the person who is to pay him for his trouble. He complains against the persecutions of coquetish [sic] fate; recommends wisdom and abstinence; and, at last wishes you to live one thousand nine hundred and ninety-nine years, that is to say, as long as lived the patriarch Noah, – of course, on condition that the cloud of your liberality pour a golden shower upon the parched lips of the aushik’s empty and yawning pocket. (1842: 344) In these appeals to the audience and remarks about the organization of their tale, the singer-narrators step out of their story. This is also the case when they comment on their tale. In his article ‘Metafolklore and Oral Literary Criticism’ (1966), the folklorist Alan Dundes underlines the importance of the narrator’s comments, the raconteur’s asides. By remarks addressed to the audience, the narrator can evaluate the tale and also the various events or characters of the story. Dundes sees the importance of such remarks and asides in their enabling us to locate the narratives in the world of the narrators and their audience. They reveal to us the narrators’ and listeners’ world-view and provide a context for the tales. İlhan Başgöz, with regard to Turkish minstrel romances, talks in a similar vein of singers’ digressions and proposed in 1986 a division into

196 Interpretation (1) explanatory and instructional digressions; (2) opinions, comments, and criticism; and (3) self-reproach and confession. I will in the following discuss some examples of explanatory and evaluative comments. A basic explanatory comment is provided when customs, objects or words are explained to the audience. In the hikâye ‘Bağdat and Hafız’, a popular version of the story of ‘Hero and Leander’ of Greek and Roman antiquity by the âşık Şeref Taşlıova, a crucial role is played by the candle or lamp that the heroine puts up on a tower to help her lover to swim across the water (the Hellespont in antiquity, the eastern Anatolian lake Çıldır in the Turkish tale) to a tryst with her. The singer mentions an oil lamp and explains: Eskiden s¸imdiki gibi değildi. Elektrik yok, gaz lambası yok. O zaman fitilli kandiller, çirağlar yakarlarmıs¸. İçinde yağ. ‘Bezir’ deriz biz ona. Keten tohumunu sıkıp çıkarırlar beziri. O sıralar, ondan vardı orda.21 In the olden days it wasn’t like it is today. There was no electricity, there were no gas lamps. In those days oil lamps were lit. They contained oil. We call it linseed oil. Linseed oil is pressed from cotton seeds. So, this is what they had there. In a Turkish hikâye from the Köroğlu cycle, the narrator Behçet Mahir interrupts the story at the point where Köroğlu falls in love with Han Nigâr when he sees her portrait, handed to him by a dervish: O zaman kalemdar devrişler, bir insanın gittiği yerden, yahut geldiği yerden, tasvirini alırdı, ki o adamın haberi hiç olmazdı. Neden? Öyle kalemdar dervişler var idi eskiden. Bir adamın da tasviri kalem ile, ya camekan üstüne, ya da kağıt üstüne alınırdı. Şimdi de tasvirler makineler ile alınıyor. Her bir şey kolaylık olmus¸. (Kaplan, Akalın & Bali 1973: 78) In those days dervishes, able to wield a pen, painted people wherever there were any, without their noticing it. Why? There just were such painter-dervishes in the old times. A person’s portrait was drawn with a pen on glass or paper. Today portraits are taken with cameras. Everything has become easy. As to evaluative comments, they are of various kinds. The evaluation of the protagonists of the tale is, of course, implied by the way their actions are presented, by descriptions either of their beauty or their ugliness and by the use of epithets with positive or negative associations. The description and presentation of the characters are part of the tale and not of the singer’s commentary. In performance, however, such depictions can be coloured by the prism of the singer’s delivery and gestures. When the

The Singer and the Tale  197 Karakalpak singer Jumabay Bazarov smiles before announcing Edige’s Solomonic judgment, for instance, he gives a clear comment on this episode, a thumbs-up sign for the result of the quarrel between the two women (see Chapter 5, p. 90). A singer can also voice outright praise or condemnation. In Karataev’s version of Är Soghotokh, for instance, when the hero has successfully freed himself from his fetters and has hacked down an abaasï, the singer exclaims: ‘Noo! Togho bärdäy!’ ‘Very good! Bravo!’ (Illarionov et al. 1996: 132/133). Another form of evaluation is to call on the audience to witness some deed or character, as if to take part in a particular scene, and to share with the narrator admiration or disgust. In the Uzbek epic Rawshan, for example, the narrator exclaims repeatedly when describing a number of battles: Endi koring Aynāqni … Endi koring mard Hasan … (Zarif 1971: I, 211, 215) ‘Now look at Aynāq! … Now look at hero Hasan!’ … Aynāq’s and Hasan’s prowess in fighting with their enemies is stunning and praiseworthy. This appeal to the listeners to visualize a scene is neatly paralleled in the Old French Chanson de Roland, when the narrating voice calls for empathy: As vus Rollant sur sun cheval pasmét, E Oliver ki est a mort naffrét. (V. 1989–1990) (Segre 1989: I, 192) Now look at Roland, fainted on his horse, And Olivier, who is mortally wounded! It is often difficult to distinguish between a narrator’s comments that are traditional and part of the tale and those that are specific of a certain singer or even of a certain performance situation. Appeals to the audience, structural information (‘Now we come to so-and-so’) and evaluations of heroes and villains are often formulaic and inner-textual. Although in actual performance a singer-narrator will say such lines and hence make the narrating voice of the tale his own, he or she does not step out of the narrative. The ‘raconteur’s asides’ are clearest when they are indeed asides, that is, extra-textual remarks. This is the case with the explanations of words and customs quoted above from Turkish hikâyes. It is only when a performance is carefully recorded and faithfully textualized that an aside can be identified as the performing singer’s rather than the implied narrator’s comment. A trustworthy textualization is the edition of the mvet performed by Zwè Nguéma, to which I would like to return. As mentioned already, the singer

198 Interpretation is accompanied by two assistants and a choir. He begins the epic as a thirdperson narrative. After he has given an exposition of the tale, he switches into the first-person, lists what he has recited up to now and explains the origin of the Fang to Herbert Pepper, who recorded the performance: I have just told you the beginnings of the people of Éngong Zok Mebegue Me Mba …. About our separation from the people of Éngong Zok Mebegue Me Mba I told you the day before yesterday, right? – Yes – This is fairly difficult to explain. We are not the only ones to have separated from them, You also, you Europeans, have separated. I have talked of Nkpa Mebegue and Yo Mebegue. This is the Yo Mebegue who begot Zame Yo Mebegue. The man Zame begot the Blacks. He also begot the Whites, this Zame. You are also a White. Afterwards this Zame begot the Pygmies; Zame begot the gorilla, Zame begot the chimpanzee, the fish, he begot the rainbow and Ondo Zame [the dog]. Zame said: ‘I will give the name Fang to an entire people’. This is us who are of this people, which is called Fang. (Pepper 1972: 37 trans.) Zwè Nguéma continues his explanations and then begins the ‘First Interlude’. In this interlude the singer and the choir interact, with the choir generally uttering wailing exclamations (Eeee ayayaya …) somewhat in the manner of the iō of the choir in Greek tragedy. The epic consists of fifteen such interludes, of which the last one is entitled ‘Initiation to the mvet’ and contains an initiation song; the epic comes to an end with a finale where singer and choir interact as in the other interludes, with a final ‘It is finished!’ In addition, the assistants interact with the singer. To give just one example: When in a scene of slap-stick comedy one of the protagonists, Nsoure Afane Obame, lifts his big foot and plants it on a plate of two kid legs, the assistant exclaims: ‘Hahaha!! Akié!’ and when he then decapitates his adversary, the assistant applauds: ‘Go ahead! Go! This is a real man, this is a proper man!’ (215) Similar interactions between singer, assistants, choir, and audience are found in other African traditions. The Kpelle Wọi epic of Liberia, translated by Ruth Stone, is structured as a continual dialogue between narrator and questioner, with some interjections from the audience and the occasional voice of a supporting singer. Interestingly, the narrator relates his encounter with the hero of the tale, Wọi, and engages in a conversation with him. Ruth Stone explains: ‘Such a device provides veracity for the teller’s assertion that he, in fact, witnessed those things of which he tells’ (1988: 14). Bakhtin’s ‘epic distance’ tends here towards zero, when both narrator and protagonist become figures of the tale. This conflation of two levels differs, however, from first-person narration as here the

The Singer and the Tale  199 narrator identifies with the hero (or one of the protagonists), while in the Wọi epic the distinction between He, the hero, and I, the narrator, is maintained. In his edition and translation of the Bambara epic Silamaka Fara Dikko in the performance of the singer Mamadou Kida, Ousmane Bâ sets the singer’s asides typographically off from the narrative. Many of these are explanatory, others evaluative. As Bâ points out: ‘The commentaries are the medium with which the narrator exerts a direct influence on the audience, showing how a narrated passage is to be understood and which aspects are relevant beyond the tale itself’ (1988: 173–74 trans.). Various facets of the narrator’s presence in the narrative come together in improvised additions to the sung verses interspersed in the Banyanga Mwindo epic, called rori. Daniel Biebuyck illustrates this with an extract of a passage from one of the versions of the epic: 1. 2. 3. 4. 5.

Sky changes to dawn and we are speaking like the sound of the mukuki-initiation. I am sitting, humming Like a bee That passes on the itondo-plant. I am telling the story That was told by Mutia and Irumbo, Mutia, child of twins. I have sung (with) the drums, I am tired. I am squeezed, my father; I am in pursuit of Mwindo Mboru; I am in pursuit of Nanga.22

Biebuyck isolates five points in these lines and summarizes their content: ‘The bard (1) praises the length of his song, (2) expresses his enjoyment, (3) refers to his predecessors, (4) speaks about his fatigue, and (5) simply indicates that he has lost the thread of his story’ (1978: 25). What all these illustrations demonstrate is the presence of the narrator in the narration and also in the narrative, in narration and récit. While the Now of the performance and the Then of the tale are separate, the two levels of time and experience also interact, not only in first-person narratives and in performance contexts of rite and ritual. The singer’s relationship to the tale is characterized by closeness and intimacy. This might be expressed by the identification with the hero as in first-person narratives, or by dramatic and mimetic means as in the performance of the Mwindo epic, as described by Biebuyck: ‘While singing and narrating, the bard dances, mimes, and dramatically represents the main peripeties

200 Interpretation of the story. In this dramatic representation, the bard takes the role of the hero’ (Biebuyck & Mateene 1971: 13). This closeness is not disavowed by a certain detached attitude sometimes expressed by epic singers. On the one hand they are convinced that on the basis of the tradition which they have inherited they are authentic and trustworthy representatives of this tradition; on the other hand they show that they are aware of the discrepancy between story and event. While they affirm the truth of their tale, they also concede that this truth is not absolute. This attitude is perhaps only found in epics that purport to relate stories thought to belong, in Bakhtin’s words, to ‘a world of “beginnings” and “peak times” in the national history’. In the quotation from Manas in Chapter 6, line 4 reads Jarïmï tögün, jarïmï chïn, ‘Half of it is fabulous, half of it is true’. Interestingly, almost the same lines occur in an Uzbek epic, in Fāzil Yoldāsh-oghli’s version of Alpāmish. They come at the end of the dastan:


Fāzil-shāir speaks all that he knows. Some of these words are lies, some are the truth. May the days of our lives pass in happiness! These are the words I have spoken with difficulty. If you give a shout, you show your approval of my words. The singer is a poor man, you know! This then brings my story to an end. (Mirzaev & Abdurakhimov 1999: 422)

The word for ‘truth’ is chïn in Kyrgyz and chin in Uzbek. The word is glossed as ‘truth’ and ‘true’ in the dictionaries; it is found in a great number of Turkic languages, including Old Turkic, and also in Mongolian. The etymological dictionaries derive it from Chinese zhēn ‘true’. Clearly, this is a weighty word that is meant to be taken seriously. In such remarks there is a tension between the individuality of the singer, who is in control of his narrative and feels free to tell the tale in his fashion, and the traditionality of the tale and its ultimate rootedness in a world beyond the narrative. This world is made visible in the tale and comes to life in the singer’s words, songs, gestures – and thoughts and comments.

Notes 1 See Genette 1972: 71–76, and in English Genette 1980 (where narration is translated as ‘narrating’). In the following discussion narratological terminology will be used sparingly; on narratological concepts, see Herman, Jahn & Ryan 2005. 2 See Laude-Cirtautas 1983. 3 Sharyar is edited in Ayimbetov, Bahadïrova & Al’niyazov 2009: 225–96 (Jumabay-jïraw, 1951 version) and Maqsetov 1981 (Öteniyaz-jïraw and Qulamet-jïraw).

The Singer and the Tale  201 4 See ATU no. 707; Marzolph & van Leeuwen 2004: I, 425; on the relationship between Pushkin and the Central Asian tales, see Nurmukhamedov 1983. 5 Ricœur 1983–85; see Fludernik 2003 and her article in Herman, Jahn & Ryan 2005: 606–12. 6 See Putilov 1988: 33–44 (‘Time in the epic’) and Hatto 1989: 196–214 (‘Time and the Times’). 7 See Bakhtin 1981: 13–14; ‘Epic and Novel’ was originally written in 1941. 8 See also Reichl 2000a: 101–34 (‘Heroic Epic and Tribal Roots’). 9 See Emel’yanov 1980 and 1983: 94–209 (on heroines). For a translation of three Yakut olonkhos into French, see Karro & Sabaraikina 1994. For an English translation of the epic Nyurgun Bootur (Hero Nyurgun) in the version collected by the Yakut writer Platon Oyunsky, see Oyunsky 2014. 10 See Böhtlingk 1964: 79–95 (Yakut text and German translation). 11 Three versions of this epic were recorded from Karataev (in 1975, 1982, and 1986); for a comparison of the plots, see Illarionov et al. 1996: 401–12. The version edited in Illarionov et al. 1996 is that of 1982. For help with the Yakut text, I am grateful to Dr. Aitalina Kuz’mina (Yakutsk). 12 See Harris 2017. On Eliade’s concept of sacred time, see inter alia Eliade 1959: 68–113. 13 Hatto 1989: 153–59; see also Hatto 1980. 14 See Kupriyanova 1965. On the Nenets and their musical culture, see Abramovich-Gomon 1999; on the genres of Nenets songs, see Niemi 1998– 1999; on syudbabts and yarabts specifically (with musical examples), ibid. 82–96. 15 Kupriyanova 1965: 397–410 (Nenets), 410–25 (Russian). 16 For discussions and illustrations of the music, see Dobrovol’sky 1965; Niemi 1998; Skvortsova 2001. 17 Bağdat, burda büyümekte olsun, ben size nerden haber vereyim? Hafïz’dan. Türkmen, Tas¸lıova & Tan 2008: 247. See below on this hikâye. 18 Karryev 1983: 322 and passim. 19 This epic can be characterized as ‘oral-derived’; it has been edited and translated from a manuscript source. On its form and diffusion, see Drewes 1979: 1–37. Snouck Hurgronje wrote: ‘We do not go too far in saying that this heroic poem is a gem of Achehnese, nay of Oriental literature’ (1906: II, 99). 20 Me schenke unde geve uns drinken dan,/ ik wil ein ander heven an! Seelmann 1884: 18 (ll. 667–68). 21 Türkmen, Tas¸lıova, & Tan 2008: 266. See also Reichl 2015a. 22 Biebuyck 1978: 25; the full passage is found pp. 210–11.

10 Performance and Interpretation

In even the most perfect reproduction, one thing is lacking: the here and now of the work of art – its unique existence in a particular place. (Walter Benjamin) Walter Benjamin, whose The Work of Art in the Age of its Technological Reproducibility was first published in 1936, acknowledged the potential of photography, film, and gramophone for our reception of art and music, but saw also the limitations of these media.1 One shortcoming is formulated in the quotation above: the work of art is taken out of its temporal and spatial context. This applies also to the oral epic. As has been argued in the previous chapters, an oral epic is a composite of verbal, musical, gestural, and also often dramatic art that comes to life in performance. Performance presupposes a narrator-singer, with possibly additional musicians, an audience, a place and time of performance, an occasion and a purpose. Epic performance is, in short, a specific type of communicative event. When it is finished, the event is no more than a memory for the singer, the musicians and co-actors, and for the audience. If other traces are left, they are written words and symbols, possibly recorded sounds and perhaps also photographs and films. Much is lost, most noticeably when words alone remain. The loss is accompanied by a resetting of the performance traces in a new context: as texts to be read, as sounds to be heard, and as images to be seen – read, heard, and seen wherever and whenever convenient. Various questions can be asked at this point. Does what we lose ‘matter’? Do sounds and images provide a significant addition to our understanding of the epic? Isn’t the text enough? And: Does the de-contextualization of the oral epic and its transformation into a reading event fundamentally alter our interpretation? In this chapter I will try to give an answer to these questions in three steps. I will first look at the role of images in performance; I will then turn to music, focusing on forms of epic performance in which music is in the foreground; in the third part I will discuss problems of

Performance and Interpretation  203 de-contextualization and interpretation. The combination of poetry with images and with music is, of course, not restricted to oral epic poetry. Examples from English literature are Blake’s Poems of Innocence and of Experience and Byron’s Hebrew Melodies. Taking these two works as comparanda, we can re-phrase the questions about the role of sound and image for the understanding of oral epics as follows: Is reading the text of an epic which was performed with the use of visual elements, as in the ‘reading of the paṛ’, comparable to reading Blake’s Poems of Innocence and of Experience without taking notice of their colour etchings? Does it make any difference to see Blake’s portrait-view of a tiger staring to the margin of the picture for a reading of his ‘Tyger! tyger! burning bright’? Similarly with music: Is reading the text of a musically performed epic comparable to reading Byron’s Hebrew Melodies without listening to Isaac Nathan’s setting? As Byron informs the reader in the edition of 1815, the poems were written ‘for a Selection of Hebrew Melodies, and have been published, with the music, arranged by Mr. Braham and Mr. Nathan’. Apparently Byron felt strongly about the music.2 The question whether Blake’s etchings or Nathan’s music are essential for our full appreciation of the poems has probably been answered in practice, as it can be assumed that most readers have read these poems in purely textual editions and are able to appreciate their meaning, disputing any implication that their interpretation was incomplete. Music and art seem to be extras, certainly not negligible, but also not central. Blake’s Songs of Innocence and of Experience and Byron’s Hebrew Melodies are not isolated examples of the complementation of poetry by art or music. The permeability of generic borders and the symbiosis of different art forms is a theme running through the history of literature, art, and music. In the given context it must suffice to point out that the extra-textual performance dimensions of an oral epic must be seen as part of just such a wider issue. The two examples lean on non-textual elements in different ways. The music of Byron’s Hebrew Melodies concerns the performance and reception of the poems, as poems either read or sung, while Blake’s illustrations provide a presentation frame for reading the poems, adding a kind of pictorial commentary. In both cases, however, there is a certain ambiguity about the role of music and art in the appreciation of the poems and this role is hence interpretable in different ways. This is also our experience with oral epics.

Visualization and Imaging Narrative and image have formed multiple combinations and symbioses from antiquity to the comic strip and the graphic novel of our time. From antiquity onwards the hero and other protagonists of epics and myths have been represented in reliefs, sculptures, drawings, paintings, etchings, and other forms of depiction, applied to, and made of, various materials.

204 Interpretation Such images add a visual component to the tale; they are visualizations in the two senses of the word. On the one hand ‘visualization’ is defined as the forming of a mental image of something not present to sight. This is the visualization that takes place when someone hears or reads a story. There is, however, a second sense of ‘visualization’. The OED gives as the earliest record of this sense a quotation, dated 1912, from a film journal, where it is argued that for the study of history, films should be used as supplements to the printed story, because they ‘would visualize the actual occurrences so that all may readily understand’. This is what images do: they make visible what in a tale is neither visible nor tangible. In Chapter 5 it was pointed out that in the performance of the Rajasthani epic of Pābūjī the singer and his helpmate use a picture scroll. In India, this is not a singular case. Before coming back to ‘reading the paṛ’ I would like to briefly introduce some other narrative traditions in which the telling of tales is accompanied by pictures on panels or scrolls. Picture showmen are still found among the narrators and singers in various areas of India, in Rajasthan, Gujarat, Bengal, and Bihar, also in South India and formerly also in the Deccan. The use of pictorial representations varies significantly, as do the narratives and the contexts in which they are performed. The Garodas, a community of traditional storytellers of Gujarat, call their scroll paintings tipanu, meaning ‘recording’ or ‘remark’. They are over four metres long and about 35 cm wide. These scrolls, which are unrolled from top to bottom, comprise nineteen sections or registers. They begin with the picture of a temple, followed by scenes from various narratives and ending with a depiction of the seven hells of Hinduism, showing some of the torments suffered by sinners. The narratives represented on the panels comprise a wide selection from Indian mythology, folk religion and the great epics, Mahābhārata and Ramayaṇa. The various stories all reflect the form in which they are popularly known in Gujarat, including local legends, gods, and goddesses. The recitation begins with the first register and an explanation of the meaning of the temple and the ethical significance of the tales. The temple symbolizes the sacred nature of the scroll, which is seen as a mobile shrine, and the narratives are basically divine stories. The final scene clinches the religious import of the performance. The Garoda storyteller tells the stories in a mixture of verse and prose, the form of deliverance also typical of other Indian traditions. The pictures, painted in the style of folk art, often comprise several scenes or figures on one panel. ‘This depiction of several incidents within a single panel was convenient for audio-visual narration of folk stories, as the narrator could point out the key scenes to his audience. Using these lively paintings he could penetrate the hearts of his listeners’ (Jain 1998b: 87). According to Jyotindra Jain these performances seem to have died out. A totally different use of images for the recitation of stories characterizes the performance of the Mandaheccu storytellers. Mandaheccu is the

Performance and Interpretation  205 name of a ritual performed in honour of a recently deceased person. It is performed among the Golla of the Telangana region in Andhra Pradesh (south-east India). In the course of this ritual, which lasts for four to six hours, instead of pictures, figurines made from wood, cloth, and papiermâché accompany the words of the narrator, who ‘uses a wide array of brightly painted images to animate his formulaic, invocatory narrative’ (Thangavelu 1998: 90). The British Museum has one of these figurines on its website, an almost 50-cm-high figure of the goddess Gangamma.3 Gangamma is the goddess that presides over the ceremony and guarantees the efficacy of the ritual. Conforming to the religious framework of the performance, the main storyteller is both narrator and priest. He performs together with a troupe of celebrants and musicians and shows the various figures to the audience in the course of the narration. At the end all the figures are standing in front of the performers. A complete set has fifty-three figures, comprised of some animals, but mostly of the heroes of the narratives and their wives. As Kirtana Thangavelu points out: ‘The stories that are sung are part of the long and complex oral epic called the Katamaraju-katha, which traces the lineage of the Gollas or the Yadava caste. The epic is specifically associated with this community, and thus their participation in its ritual re-enactments takes on an added significance and investment’ (90). Katamaraju is the name of the main hero, a cattle herder like the members of the Golla, who, after being taunted by his wife for cowardice, dies a heroic death, joined on the funeral pyre by his wife, dutifully observing suttee (sati). Although the incorporation of pictorial elements into the performance of epic and legendary narrative is fairly different in the Gujarati and Golla tradition, they both share the embeddedness of the narrative in a religious-ritual framework. Such a framework is also typical of the performance of the Pābūjī epic and of the Devnārāyaṇ epic.4 The scroll (paṛ or phad) used for the telling of the Pābūjī epic serves also as a portable temple and has to be ritually deconsecrated when it has become threadbare and has no further use. The bhopo (bhopa) is not only the singernarrator of the epic but also the priest of the main hero, who has become deified. Witnessing an epic performance therefore also brings the experience of seeing (darśan) the deity with it (Blackburn et al. 1989: 241). The scroll shows a number of peculiarities. It is not the depiction of a sequence of events, to be read or shown from one end of the scroll to the other, somewhat in the manner in which the Norman Conquest is pictured on the Bayeux Tapestry, beginning with King Edward sitting on his throne and ending with the death of Harold and the defeat of the English in 1066. Rather than following a chronological order, the scenes on the paṛ are arranged according to where they took place. The scroll has a number of areas reserved for different localities and a particular scene is depicted in the appropriate area, independent of where it stands in the chronological sequence. The central position is taken by Pābūjī’s

206 Interpretation court, with the hero as the largest figure of the scroll, followed by his four companions. J. Smith and other commentators on the art of the paṛ have stressed that these scrolls exhibit a sophisticated relationship between the characters and events of the epic and their pictorial representation.5 They serve as a visual guide for the audience and are at the same time material signs of the epic’s narrative cosmos, for which they provide a road-map. The audience’s attention is directed not only to the words and music of the epic tale, but also to the pictorial representation of both characters and episodes. The Indian storytellers and bards are by no means unique in their use of scrolls and pictures. Some of the Tibetan bards, for instance, also use pictures: An itinerant rhapsodist of the Kesar Epic always carries with him a painted image or thaṅ-ka representing the life-story of king Kesar, and an arrow adorned with multicoloured (blue, green, yellow and red) ceremonial scarfs or kha-btags. With the help of this arrow or dā-tar … the rhapsodist points out the various episodes of the Kesar Epic depicted on the painting. (Roerich 1942: 285) Also in Chinese, Japanese, Persian, Indonesian, and other Asian traditions stories are told with pictures or picture scrolls as visual aids. From a European perspective, the street singers or Bänkelsänger (saltimbanchi, mountebanks) come to mind, with their traditional sung narratives, an art that lingered on into modern times.6 Visualization entails tangibility and confers a certain exterior existence to the imagined characters or scenes of a tale. The depictions might not agree with what listeners imagine when hearing the story, but they direct their imagination and fasten it on a visual representation. In the case of religious connotations the image might even be accorded a status of authority and become an object of veneration. There is a great variety of roles that images play in oral performance. These range from testimonies affirming the truth of a tale to providing a guide through the maze of the plot. Panels like those used by the itinerant ballad-mongers, where the gory details of a murder or the stages of a sentimental love story are depicted, tell the audience: ‘This is how it happened! This is the person! It’s true!’ This message seems to underlie most pictorial accompaniments of storytelling. By the same token these depictions, painted or modelled in a traditional and as a rule highly patterned, ‘formulaic’ style, give the impression of a certain generality and typicality. The faces of the protagonists on the paṛ, men and women in portrait-view, are the same for each sex. They are distinguished by size and place on the scroll and can hence be identified. However, as John Smith points out, due to the ‘geographical’ arrangement of figures and scenes, one and the same figure

Performance and Interpretation  207 can represent different protagonists if the scene in question occurs in the same place. These visual representations of scenes and protagonists act like pictorial labels stuck on the narrative. They identify, characterize, and help smooth the flow of the narrative. This is not to say that they are irrelevant or that they don’t have an intensifying effect on the beholder and listener. But they are an accompaniment of storytelling and not stories in their own right. They should not be confused with pictorial narrative, that is, pictures telling a story or rather retelling a (generally known) story, as studied by art historians for various periods and cultural areas. They seem to have more in common with a group of pictorial remains from antiquity. These are tablets that have been inspired by the story of Troy and combine text and picture, the so-called Tabulae Iliacae the ‘Trojan tablets’ or ‘Iliac tablets’, of which the Tabula Iliaca Capitolina is the best known and best preserved. These tablets are interesting because they mark a transition from visualization and imaging to labelling and tagging. The general layout of these carved plaques is exemplified in the fragmentary Capitoline Tabula Iliaca, measuring 25 × 30 cm, where we can see a number of small reliefs of scenes from the Iliad and the Epic Cycle, identified by subscriptions in Greek, mostly giving the names of the characters depicted, then a larger relief showing the capture of Troy and a text in Greek filling a column running between the top and the bottom bands of the tablet.7 These tablets have been much interpreted, in particular the texts and their relationship to the textual transmission of the Troy legend. In recent years the focus of interpretation has moved to an analysis of the tablets in the context of their Roman audience and of the relationship between text and image.8 Many questions remain unanswered, one of them concerning the use of the tablets. Otto Jahn and Adolf Michaelis came to the conclusion that these tablets were used in a school situation to illustrate the contents of the epics which the pupils had already learned and to help their understanding of the depicted scenes by the inscriptions (1873: 89). This explanation has found both adherents and critics. Whatever the purpose of these tablets and however intense the text-image dynamics, the tabulating, conspectus-oriented, and image-supported disposition of information on these tablets strikes the eye. Plots, scenes, and characters are organized and catalogued, and the images seem to be added not only for illustrative, but also for labelling purposes. Like the scrolls of the Indian picture showmen, the Tabulae Iliacae use illustration not for its own sake as pictorial representations of events and characters, but rather as a memory guide through the intricacies of the narrative. No doubt, the illustrations have a primarily pedagogical rather than interpretative function. One important difference between the Tabulae Iliacae and the street singer’s picture tables on the one hand and the Rajasthani scrolls and the figurines of the Mandaheccu storytellers on the other hand, is the

208 Interpretation sacred character of scrolls and figurines, resulting from the ritual nature and purpose of the performance, an aspect of the communicative event that is totally lost in the reduction of the epic performance to a text. It should be noted that the pictorial elements need not disappear in textualization, as is shown, for instance, by editions that include images such as John Smith’s of the Pābūjī epic. It has to be conceded, however, that the dynamics of the use of the images in performance is lost on a reading audience. Their effect on the reader is perhaps similar to that of Blake’s etchings: they allow us a glimpse at the pictorial world of the cultural milieu from which they come, add colour – in the literal and metaphoric sense of the word – to the poetry, stimulate visual impressions, but are predominantly appreciated as ‘extras’. Their contribution to the understanding of poetry and narrative is primarily that of sensitizing the reader to the cultural context in which the performance of oral poetry occurs.

Aria and Recitative At the beginning of this chapter the question was asked whether reading the text of an oral epic without any knowledge of its music is like reading Byron’s Hebrew Melodies without listening to Nathan’s settings. Readers of the Hebrew Melodies will probably agree that Byron’s poems stand on their own and can be appreciated on their own, even if hearing them sung and accompanied on the piano might increase enjoyment. As it is, there are other settings of the Hebrew Melodies, such as Robert Schumann’s, which might even be preferred by music lovers to Nathan’s.9 What, however, about other forms of poetry combined with music? Is reading the text of an oral epic like reading the libretto of an opera? Admittedly, there is a difference between librettos. Some librettos are mediocre texts, but provide the words for much admired music. Some librettos, like that of Mozart’s Marriage of Figaro, are based on the text of dramas. The drama texts have been adapted to their use as librettos, but can still be read on their own as dramas. Independent of the quality and readability of a libretto, it is, however, only the text of an opera and not the opera itself. Reading and possibly also seeing on stage Beaumarchais’ Le Mariage de Figaro and reading Lorenzo da Ponte’s libretto of Le Nozze di Figaro, based on Beaumarchais, are not equal to experiencing Mozart’s opera. His opera, as the subtitle says, is a commedia per musica, with musica writ large. This, it can be argued, is the situation with oral epics in whose performance music plays a prominent role. As discussed earlier, the music of oral epics covers a wide spectrum of realizations. We find styles of chanting with monotone melodies, which appear to support primarily the vehicular function of music as a carrier of words, and we find musical performance styles in which music is in the foreground. Curt Sachs was certainly right when he talked of ‘text-dominated’ performance in the former case;

Performance and Interpretation  209 even here, however, as we have seen for instance in the case of narrative melodies from the highlands of Papua New Guinea, musical variation and modulation occurs. It can be argued that what we lose in a text-only edition in these cases is mainly the performance milieu – the singer’s voice and gestures, the effect of a repetitive melody, and the sometimes trancelike submersion into the soundscape of the performance setting. The further we move in the direction of ‘music-dominated’ performance, the more deficient text-only editions become. In his discussion of the various performance modes of Mande hunters’ song, C. Bird remarks that most Mande epics can be characterized ‘as consisting of a series of songs tied together by narrative links’ (1972: 289). Without the music, it seems, we experience no more than a libretto. Music is especially then in the foreground when lyrical passages predominate. This is the case in the love and adventure romances of Central Asia and the Near East. These prosimetric tales have a long tradition, reaching back at least to the Middle Ages. In European literature the best known love romance of this kind is the Old French tale of Aucassin and Nicolette.10 This is the story of the love between Aucassin, the son of the count of Beaucaire in Southern France, and Nicolette, a Saracen girl, who was captured and then baptized. This tale is found in a thirteenth-century manuscript (Bibliothèque nationale de France, fr. 2168) and written in the Picard dialect of Old French. It refers to itself as chantefable (or rather, in Picard, cantefable), i.e. ‘sing-tale’. And this is precisely what it is. It is composed of prose and verse passages; the prose passages are introduced by the words: Or dïent et content et fablent (Now one speaks and tells and narrates), and the verse passages are introduced by the words: Or se cante (Now one sings). By a happy circumstance, the music of the verse passages has also been transmitted and we can therefore interpret the tale as it is meant to be interpreted, as a ‘sing-tale’, a combination of spoken and sung words. Case Study: Ghärip-Ashïq As an example of a ‘sing-tale’ from oral tradition, I will discuss one of the romances of widest dissemination among Turkic-speaking peoples, variously called Âs¸ık Garip (Turkish), Āshiq Gharib (Uzbek), GhäripAshïq (Karakalpak), and so on. This romance is found in Turkish, Azeri, Turkmen, Crimean Tatar, Karakalpak, Uzbek, and Uyghur. In addition, there are also versions of the tale in Armenian and Georgian. Âs¸ık Garip is the name of the main hero; in some versions the title also comprises the heroine’s name, as in Uyghur Ghärip wa Sänäm, ‘Ghärip and Sänäm’. The hero’s name (varying somewhat according to language) is composed of two words of ultimately Arabic origin: cashīq and gharīb. The former means ‘lover’, the latter ‘stranger’; the heroine is generally also called by a name of Arabic origin, Sänäm or Sanam, from ṣanam, meaning ‘idol’, and in Persian and Uzbek also ‘a beauty’.

210 Interpretation During his banishment to the Caucasus in 1837, the Russian poet Nikolai Lermontov heard from an Azerbaijanian in Tiflis (Tbilissi) a popular narrative which caught his fancy and which he retold in Russian. He entitled it ‘Ashik-Kerib’ and called it a Turkish tale. His retelling – probably the earliest record we have of the romance – captures the main motifs of the story. The basic plot consists of the separation of two lovers and the wanderings and sufferings of the hero, who is a lover (cāshiq) and becomes a minstrel (ashïq), singing his songs of love and longing. When his beloved Sanam is about to be married to a rival, he manages, generally with the help of a saint or by some other miraculous means, to arrive at the scene of the wedding just in the nick of time before the bride is married. In Lermontov’s version Kerib has to get to Tiflis from Aleppo in three days. He gets as far as Erzincan on the first day, but despairs of getting to Tiflis in time. His prayer for help, however, is heard: Hizir-Ilyas, the supernatural helper of Islamic folk religion (explained as St. George by Lermontov), appears and takes him to Erzurum on the second and to Tiflis on the third day. The plot ends with the happy reunion of the lovers. The timely return of the hero before his bride is married to someone else is a motif that has many parallels in tales and romances in a number of oral epic traditions (for instance, in the Russian bylina of ‘Dobrynya and Alyosha’ or in the various versions of the epic of Alpamish) and also in medieval literature (for instance, in the Middle English Horn romance).11 Although the various romances show a number of variations in their plots, there is no doubt that they belong together. The versions can be divided into two groups, the Turkish-Azeri group and the TurkmenUzbek-Uyghur-Karakalpak group. Written texts have also entered into the transmission process of this romance, which explains some of the textual similarities between different versions, especially in the songs. The poems in this and similar dastans are generally the words of one of the protagonists, sometimes also dialogues in alternating stanzas. They are not always poetical effusions of love, longing, or grief over separation or death; but love lyrics and love laments nevertheless form an important part of the poetry. In many versions poetry takes up far more room than prose. A Turkmen edition of this dastan, for instance, comprises 147 verse passages on 156 printed pages; the prose narration is visibly shorter than the verse.12 Similarly, an Uzbek version sung and narrated by the conservatory-trained bakhshi Rozimbek Murādov and published as a set of three 12-inch records, has 34 songs plus a final piece of music; their performance takes up over two-thirds of a total of about 150 minutes.13 These romances are true ‘garlands of songs’, set in a narrative framework. It is therefore perhaps not too far-fetched to compare them to the Singspiel, an operatic work with arias linked by spoken dialogues, or even more generally to the classic opera with its alternation between aria and recitative. This comparison is also supported by the fact that considerable melodic diversity is found in these romances. The Karakalpak

Performance and Interpretation  211 baqsï Genjebay Tilewmuratov (1929–1997), from whom I recorded an extract from this dastan, stated in an interview in 1990 that he knew about one hundred melodies. The importance of music in these romances is also underlined by the enlargement of the instrumental accompaniment, especially in the performance of the Uzbek dastans in Khorezm, where we find, apart from plucked chordophones, the frame-drum dāira, and sometimes also the accordion.

Illustration 15  Genjebay Tilewmuratov (1990)

To illustrate the style of Āshiq Gharib/ Ghärip-Ashïq and similar romances, I will briefly look at a verse passage which I recorded from two Karakalpak baqsïs. The context is the following: Ghärip has returned home at the time of Sänäm’s wedding. She is standing on the roof of her köshki (garden palace), when he addresses her in a poem, in which he reveals his identity and asks her to show her face. Sänäm expresses her joy and bids him welcome in a song, but only shows her veiled face, covered by nine layers of cloth. In the song that follows, Ghärip reproaches her for not showing her face openly and coming to him. At this point Sänäm sings the poem: ‘Hear, Ghärip, my sighs and groans!/ I could not know that you, my love, have come back!’ After this song, she throws herself from the roof, but is caught by Ghärip. Despite more complications, the story soon after comes to a happy end. This scene is very popular in Karakalpakstan. When I visited a regional singers’ contest organized in honour of Esjan-baqsï Qospolatov (1901–1952) in Iyshan Qala in 1993, one of the backdrop paintings on the stage depicted this scene, with Sänäm falling from her köshki and Ghärip, his dutar in his right hand, spreading wide his arms to catch her! The verse passage ‘Hear, Ghärip, my sighs and groans!’ is sung to a melody named ‘Arïwkhan’ (‘beauty’). This is the name of a woman

212 Interpretation singer, who is said to have composed the melody. Women baqsïs are not unknown in Karakalpak folk tradition. Famous early singers are the qïz baqsï (‘girl baqsï’) Qänigül from the first quarter of the twentieth century and Hürliman from the second half of the nineteenth century, the daughter of the Karakalpak national poet Berdaq (1827–1900). Also in contemporary Karakalpakstan women baqsïs are popular; one of the best known is Gülnara Allambergenova. It has to be remembered, however, that like the Turkish as¸ık, the Karakalpak baqsï is primarily a musician rather than a singer of tales, who performs instrumental pieces and different types of songs, among them, however, also songs from dastans. My recordings of this verse passage are from Genjebay Tilewmuratov, accompanied by the ghijjak-player Karimbay Tïnïbaev (1983, Nukus) (see Audio/Video Ex. 7), and from Turghanbay Qurbanov (1946–2007), accompanied by the ghijjak-player Jarïlqaghan Esjanov (1994, Almaty) (see Audio/Video Ex. 8). These two recordings agree in principle in both text and music. Genjebay-baqsï sings four stanzas and Turghanbay-baqsï five; in two printed editions of the romance one has ten stanzas, while in the other this passage is missing. The composer Viktor Shafrannikov wrote down this melody from the Karakalpak singer and actress Ayïmkhan Shamuratova (d. 1993) and printed two stanzas in his edition of the melody.14 It is clear that all singers – Genjebay-baqsï (A), Turghanbay-baqsï (B), and Shafrannikov’s singer (C) – sing the same poem, which in turn is the poem edited from the performance of Esjan-baqsï (D). As is often the case with orally transmitted poems, however, the number of stanzas and their sequence vary in different versions, and there is some verbal instability within stanzas, with rhyme-words and refrains as the most stable elements. All four texts share the first stanza, but diverge in the sequel. A has four, B five, C two, and D ten stanzas. So, yes, it is the same poem, but in a protean shape. This is the poem in Genjebay-baqsï’s version15: Esit, Ghärip, ahïw menen zarïmdï, Yarïm, kelgeningdi bile bilmedim. Tärk äyleyin namïs penen arïmdï, Yarïm, kelgeningdi bile bilmedim. 5


Sen häm kelerme dep külmedim oynap, Häsiret qazanïnda pispedim qaynap, Qanjar urma tänge janïmdï qïynap, Yarïm kelgeningdi bile bilmedim. Döhmet äylep nahaq töqpe qanïmdï, Sen men dep tärk ettim dünya malïmdï. Sen yandïrma burïn janghan janïmdï, Yarïm kelgeningdi bile bilmedim.

Performance and Interpretation  213


Toy ötken song keldim köshkim astïna, Giyne äylep jan almaqtïng qastïna, Sänem özin taslar Ghärip üstine. Köshki üstinde taqat ete bilmedim. Hear, Ghärip, my sighs and groans! My love, I could not know that you had come. I will throw my honour and my shame away. My love, I could not know that you had come.




Hoping that you might come, I did not laugh when playing, I did not get well done while cooking in the cauldron of longing. Don’t stab me with a dagger and oppress my soul! My love, I could not know that you had come. Don’t speak bad words about me and shed my blood! Thinking that we are one, I renounced all my earthly possessions. Don’t set my heart on fire, which is burning already! My love, I could not know that you had come. After the end of the feast, I came to my garden palace. Intending to take my life after my offence, Sänem will throw herself on Ghärip! I cannot wait any longer on the roof of the köshki!

This is a moment of high drama. Sänäm threatens to commit suicide. The poem gives voice to Sänäm’s feelings of grief and despair. It is composed in four-line stanzas, with 11-syllable lines, rhyming aBaB cccB dddB and so on. This is the form of the qos¸ma discussed earlier (see Chapter 8). The mood of the poem is encapsulated in the words ahïw menen zar of the first lines, a Turkicized Persian loan-expression, from āh u zar, ‘sighing and grief’. Sänäm’s grief is caused both by her longing for Ghärip (l. 6 häsiret ‘grief, longing’) and by the accusation of her guilt, although guiltless. The latter is expressed by a group of lexical items of Persian and Arabic origin: namïs and ar ‘honour, self-respect’ (l. 3), döhmet ‘calumny’ (l. 9), and giyne ‘resentment’ (l. 14). Love, the topic of the whole dastan, is not only expressed by the sentiments behind Sänäm’s words, but also by addressing Ghärip as ‘my love’ (yarïm) in the refrain and by the metaphor of the fire of love in l. 11. This poem is a typical example of the lyrics found in love dastans. They are simple folk songs, but in their imagery and lexicon they also share the poetic idiom of classical Persian and Uzbek lyric poetry. The same

214 Interpretation is true of the love romances of other Central Asian and Near Eastern Turkic traditions. These dastan poems hover between folk songs and classical lyrics. Their interpretation has to take account of the conventions and stylistic traits of classical love poetry and yet see the poems in the context of their orality, in their rootedness in folk songs and popular lyrics. Annemarie Schimmel has discussed the images and topoi by which a favourite theme of Persian poetry, the ideal loving couple, is expressed.16 The main difference to classical love lyrics is, apart from the level of sophistication, the metre. Folk poetry is composed in syllabic metres and not in the quantitative metres of classical Arabic, Persian, Ottoman Turkish, or Chaghatay Turkic poetry. Although many of the images are part of the common stock of Persian love poetry, the dastan poems are not without some local colour, as in l. 6, where for Sänäm’s love and longing the image of cooking (but not to the point of getting well done!) in the cauldron of grief is used. The combination of basic agreement and individual variation between the poems as recorded from different singers is also true of the melody. Shafrannikov gives the following sung melody for the first stanza:

Example 10.1  Gharip-Ashiq

Performance and Interpretation  215 This is a predominantly syllabic melody with a tremolo-style melisma on the last but one syllable of each line (on me of bilmedim ‘I didn’t know’). It has a steady beat and is, according to Shafrannikov, sung in a slow tempo. The melody moves within the interval of a fifth and has as final note F at the end of the first three lines (‘F major’), but then, in the fourth line, modulates to D as finalis in the repetition of bilmedim. When one compares Genjebay-baqsï’s and Turghanbay-baqsï’s performance with the melody above, however, a quite different impression is conveyed to the listener. The tempo is fairly fast and the beat somewhat fluctuating, especially in Genjebay’s version, which creates the auditive impression of a freely flowing melody. The strumming of the dutar, however, keeps a fairly regular rhythm present, while the embroidering of the melody by the ghijjak strengthens the impression of an uzun hava melody. The same effect is created by long, sustained tones (sostentuo) and melismas. The tidy transcription offered by Shafrannikov can only be taken as an indication of the melodic line. This line is actually broken up by short instrumental interludes between verse-lines. The slightly pressed voice quality of Karakalpak baqsïs and their specific coloratura-like intonation are also left unexpressed in ‘broad’ musical transcriptions like the one above.17 Furthermore, the melodic contour varies in the following stanzas, most significantly when the singers change into a higher register. As will be noticed when comparing the stanza in its textual form with the stanza in its sung form, the last word of the refrain (bilmedim) is repeated, with the insertion of an extra syllable (yar). This is not surprising. It is common practice in many musical traditions when poems are sung. Azerbaijanian as¸ıqs appropriately call this process of musical ornamentation sırğa ‘ear-ring’. It also found in Nathan’s music. In the first poem of Byron’s Hebrew Melodies, ‘She walks in beauty’, for instance, the first two lines are repeated as a refrain after every stanza in Nathan’s setting, contrary to Byron’s text. This addition to Byron’s poem is introduced by a coloratura on the syllable preceding the refrain. Changes like these alter the shape of the poem when it is sung. These changes can be minimal as in the present example, but they can also significantly affect the metrical pattern of the poem as discussed in Chapter 6 (pp. 125–26). In the present case the sung form does not obscure the metric shape of the word-poem, but it does elaborate the verse-patterning. What can be drawn then from this illustration is that the text can be interpreted on its own, against the backdrop of the conventions of Uzbek (oriental) popular love poetry. As a song, with instrumental accompaniment, it acquires, however, a richer texture and presents itself as a multi-layered work of art comprising poetry and music. In other areas of folksong and sung narrative poetry, as for instance in that of European and North American balladry, scholars have stressed the importance of also considering the music for textual interpretation. Bertrand Bronson,

216 Interpretation who edited the definitive collection of English-language ballad tunes, maintained already long ago that ballads can only be studied profitably when their music is also taken into account: ‘Yet, I insist, if the student of the ballad is not prepared to give equal attention to the musical, as to the verbal, side of his subject, his knowledge of it will in the end be only half-knowledge’ (1969: 38). Vladimir Propp used almost the same words with regard to Russian folk poetry: ‘Among Russians all folklore in verse is sung. To study it without connection to the music means understanding only half of the matter’ (1976: 38 trans.). Propp also expressed the same ideas about the Russian oral epic, the bylina: ‘One of the main characteristics of the epic is that it always consists of songs, which are sung. … The musical performance of the bylinas and their contents cannot be separated’ (1955: 6 trans.). Propp’s comment about the importance of music applies to all oral epics, not just the Russian bylinas; it is particularly pertinent to the interpretation of epics where, as in our example, music is clearly on a par with the poetry. Scholars such as V. Propp and B. Bronson maintain that both the words and the music have to be taken into consideration in order to arrive at a full understanding of a ballad or an oral epic. Music, however, assumes a great variety of forms and shapes and enters into many different relationships with the sung or chanted words. As shown in Chapter 8, the relationship between tune and poetry can be loose, in so far as the same tune can be used for poems or verse passages of differing meaning and mood. On the other hand, there are musical conventions at work that assign certain qualities to certain ‘musical shapes’, as expressed in the words of a Khorasanian bakhshi, quoted earlier (see p. 152). In the case of imitative, ‘onomatopoeic’ music, some of these conventions can be appreciated not only by insiders of the tradition in question, but also by outsiders. What is important is that for the audience these associations exist and that they must be explored in a full description and interpretation of a musically performed passage. Interpretation must not be content with tabulating the rhyme-schemes or the metaphors in a sung passage, but must include musical analysis; it must not be content with an elucidation of the meaning of the words, but must also focus on the conventions and associations underlying the music. The more elaborate the musical side of performance is, and the less text-dominated the epic is, the more urgent it becomes to recognize and analyse such epics and romances for what they are: garlands of songs.

From Context to Text ‘Heard melodies are sweet, but those unheard/ Are sweeter …’ This paradoxical statement is found in Keats’ Ode on a Grecian Urn. It refers to the players of pipes and timbrels, who will remain forever in the position in which they are depicted on the urn and will never end the melodies

Performance and Interpretation  217 they are playing. Unlike music heard, whose very essence is the flow and impermanence of sound, these unheard melodies have escaped the inexorable limits of time passing. The ‘happy melodist’ will be ‘piping songs forever new’. As with every other image on the urn, a moment in time has become fixed, with no further development, except in the mind of the beholder. As an image gives permanence to a scene of moving figures, so the written word freezes the fleeting sounds of speech, halts them and preserves them for future generations. The invention of writing, the passage from the oral to the written, has had an enormous impact on humanity. It has enabled the preservation of myths and holy books, letters and historical documents, hymns and poems, wise words and legal regulations, in short, speech of all kinds, and among them also epics. What have reached us, however, are texts not sounds, and what we interpret are written signs on clay, papyrus, parchment, or whatever other material was used for writing. ‘Text’ and ‘interpretation’ are terms with wide ramifications, which have been much discussed in the philosophical and critical discourse of the late twentieth century. The common meaning of ‘text’ is ‘written text’. In the case of a ‘textual trace’ of an orally performed narrative, the text denotes more precisely the textualized verbal part of the ‘message’ of the communicative event, with possibly additional written symbols indicating non-verbal aspects of performance. As with other texts, the text of a textualized oral epic is both a physical object and a concept. It is the text in front of me, either the original transcript of the performance or a duplicated and possibly edited version of it. But it is also the work of verbal art which the singer performed, in a textualized form. When we talk about the use of formulas in epic A as recorded from singer B at a particular point in time C, we refer to the text (or textualized form) of A in this second sense of ‘text’ as concept and not to the particular text we might use to underline formulas. In orality studies the term ‘mental text’ is also used. This term refers to the singer’s competence to produce poetry and narrative in performance. It is a kind of blueprint for songs and stories, based on story patterns, formulaic diction, memorized (or remembered) passages, metrical and musical moulds, and so forth. In this chapter the term ‘text’ refers to a written text.18 ‘Interpretation’ can be understood as the interpretation of a work of literature, as it is practised in a multiplicity of theoretical orientations, from New Criticism, explication de texte and Structuralism to the ‘postisms’ of a more recent period. ‘Interpretation’ is also a wider term, as used in hermeneutics, a methodological approach that, while also concerned with the interpretation of texts, is centred on the act of interpretation itself. It needs no stressing that a recapitulation of the critical and philosophical debate on interpretation lies outside the bounds of this study. In the case of oral epics, interpretation can mean text interpretation when it is applied to a textualized epic, but it can also mean the

218 Interpretation interpretation of a communicative event when it is directed toward the performance of an epic. Interpretation – or understanding, to use a term originally introduced into hermeneutics by Wilhelm Dilthey – is here directed to two different objects, a text and an event. Part of the event is what becomes textualized and can be interpreted as text; part of the event is also what can be interpreted as music, miming, dance or whatever else can be isolated as an interpretable aspect of the ‘message’. The event as event, however, is a type of communicative practice and not a text, and it calls for an interpretation from the point of view of social, cultural, and linguistic anthropology. While the interpretation of the event is based on the performance context, the interpretation of the text lacks this context; the text is, as it were, de-contextualized. Performance Contexts With preservation comes loss. As Walter Benjamin has rightly observed, no replica of an event, however technically perfect, can bring back the event itself. The ‘here and now of the work of art’ is inaccessible and what Benjamin calls the work’s ‘aura’ has disappeared (2008: 22). In the case of ‘frozen speech’ not only the event is lost but also all the various extra-textual elements of performance. This means that when we interpret textualized oral epics, the focus shifts from performance, from the communicative event, to the text. This also represents a major shift in the reception mode, a shift that is partially paralleled and prepared for by ongoing changes in performance context. To clarify this, I would like to propose a distinction between a traditional and a derived performance context. By traditional performance context I mean the habitual setting that characterizes the communicative event of an oral epic performance. The setting might quite simply consist of an epic singer, a group of listeners, a place where they can gather and a time at which they are free to congregate. In some traditions there are additional specifications regarding the occasion (such as a feast), the time (such as the month of fasting or the hunting season), the place (such as a tavern or a yurt), or a specific functional frame (such as a religious ritual). It is possible to flesh out this sketchy concept by looking at different traditions and describe what the times, places, and settings of oral performances are. In social and cultural anthropology ‘context’ is a much discussed concept and has been approached from different theoretical positions (see Dilley 2002). Although performance seems to impose a limit on the context in which it occurs, such a limit is difficult to establish. William Hanks has illustrated the far-reaching networks of verbal interchange in Yucatán, Mexico, labelling his analysis, conducted along the lines of the ethnography of communication, ‘saturation by context’ (1996: 140–68). Although I am proposing a distinction between contextualized event and de-contextualized text, the event itself, the

Performance and Interpretation  219 anthropologist’s object of interpretation, is in turn available for analysis and interpretation, as Clifford Geertz has argued, only as ‘text’: ‘The ethnographer “inscribes” social discourse; he writes it down. In so doing, he turns it from a passing event, which exists only in its own moment of occurrence, into an account, which exists in its inscriptions and can be reconsulted’ (1973: 19). One of the features that characterize the performance context of oral epics is the audience. In a traditional performance context the audience is also traditional, i.e. it is the kind of audience for whom the singer habitually narrates and sings the epic. No doubt, the definition of a traditional audience is full of pitfalls, especially in a world of social and political change. İlhan Bas¸göz described how the performance of a Turkish âs¸ık to an audience of villagers, and at another occasion to an audience composed mostly of teachers and local dignitaries, led to fairly different audience reactions (see Chapter 3, p. 53). Both audiences were traditional in a general sense, but only the villagers were the habitual addressees of the âs¸ık. By a derived performance context I mean the performance of oral epics to a concert audience, to radio or television (and increasingly internet) listeners and viewers, to tourists and restaurant customers and other listeners and viewers in a non-habitual setting. There are, of course, intermediary zones between traditional and derived performance contexts, cases that are often difficult to place. The as¸ık’s performance to teachers might be one of these. In a derived performance context the audience need not be untraditional, but when performance moves away from its habitual setting, a number of changes occur. One of them is that instead of the whole epic only short extracts or songs from it are performed. This is not the same as the custom, found in many traditions, where only one episode or branch of an epic cycle is performed in one sitting, sometimes from sheer necessity, because the whole epic or epic cycle would simply be too long for one evening or even several evenings in a row. When performance moves to the theatre or the concert hall the passages presented tend in many cases to become concert performances, pieces sung and played without their wider narrative context and often with a certain theatricality and emphasis on virtuosity. This is the impression I have gained from my research on the oral epics of the Turkic-speaking peoples. Native scholars sometimes make a distinction between traditional singers and ‘performers’, that is, musicians and actors who have a limited repertoire, acquired mostly from books, cassettes, and other sources. Wherever we draw the line between traditional and derived performance contexts, the reception mode is in both cases aural and, with the exception of a radio performance, also visual. This is, so to speak, the natural or original reception mode of oral poetry. Once an oral epic is textualized, however, and the text is the only trace we have of an epic performance, the reception mode changes from listening (and seeing) to

220 Interpretation reading, from a multimedial to a unimedial reception mode. Both traditional and derived performance context have disappeared; the text, or textualized oral epic, has become de-contextualized. This shift to de-contextualized reception is reinforced in some of the areas where oral epic traditions are actually still found, by privileging particular versions of an epic and ossifying them as ‘classics of national literature’. This is the case with Fāzil Yoldāsh-oghli’s version of the epic Alpāmish as written down by Mahmud Zarifov in 1928. This version had for a long time been available only in a shortened and editorially smoothed edition, which in turn served as the basis of the literary translation into Russian by Lev Pen’kovsky in 1949. Only in 1999 was a full scholarly edition of the text of this version together with commentary and literal Russian translation published (Mirzaev & Abdurakhimov 1999). Other versions were known, and Tora Mirzaev had written a monograph on them (1968). Some of them were only published in 1999, the year of ‘Alpamish 1000’, the millennial celebrations of the putative origin of the epic. By the beginning of the twenty-first century, Fāzil’s version of the epic was firmly established as a pearl in the treasury of Uzbek literature. It is fair to say, however, that epic scholars in Uzbekistan continue to edit the manuscript versions of the dastans in the Folklore Archives of the Navā’ī Institute and also to record and write down the repertoire of present-day Uzbek bakhshis. Similar instances of privileging particular versions or editions of oral epics and their appraisal as ‘classics’ can be cited from other areas. An example is the olonkho Nyurgun Bootur (‘The glorious hero’) in the semi-literary version of the folklorist and writer Platon A. Oyunsky (1893–1939). It first appeared in 1930–31 and is available on a set of nine records performed by the artist Gavril Kolesov (1968). In the wake of the inscription of the olonkho in UNESCO’s Intangible Cultural Heritage List in 2008, it is this version that was chosen for translation into English (Oyunsky 2014), while a large group of meticulously edited epics, written down from oral recitation, were left aside. Texts without Context The transition from spoken discourse to written text has been one of the concerns of the philosopher Paul Ricœur. In an article from 1973 he made a number of important points about the difference between speech event and written text. He draws a distinction between the act of speaking and what is said, the intended meaning of speech, and stresses that in writing ‘the meaning of the speech event, not the event as event’ is fixed (93). Writing also lacks ‘all the processes by which spoken discourse supports itself in order to be understood – intonation, delivery, mimicry, gestures’ (95). And finally, in the text the face-to-face discourse disappears, there ‘is no longer a visible auditor. An unknown, invisible reader has become the unprivileged addressee of the discourse’ (97). The reader is left with a

Performance and Interpretation  221 text without context. The text has been truncated – lacking ‘intonation, delivery, mimicry, gestures’ – and the speech event has evanesced. What remains are the narrator’s words, their shapes and imagined sounds, their meanings and their structuring into poems and narratives. The narrator’s words remain only, however, if they are put into writing with care, with the goal of preserving them as faithfully as possible. While the original transcription might still respect the singer’s words, editions and translations might not. I have mentioned the ‘canonization’ of Fāzil’s Alpāmish. There is no doubt that Pen’kovsky’s translation, which was republished in 1982 in the prestigious series ‘The Poet’s Library’, founded by Gorky, helped to increase the popularity of the epic among speakers of Russian. It does reflect the narrative thread of the epic, and it also captures its stylistic traits in the verse passages. It does, however, have its shortcomings: it is a literary translation, taking liberties with the original, and it is furthermore based on a shortened and, with respect to Islamic elements, expurgated version. Even further removed from performance are popular editions of oral epics, especially in the form of chapbooks. In the Turkic-speaking world of the Near East and Central Asia lithographic and later printed editions have been circulating since the nineteenth century, issued by printing houses in Istanbul, Kazan, and Tashkent. In Turkey, minstrel tales such as Âs¸ık Garip and Ferhat and Şirin have been repeatedly published as popular books, in which the prose narrative is often simplified. This explains the remark made by the Turkish âs¸ık Müdamî in an interview: ‘A minstrel’s style is one thing and the style of the minstrel tales which appear in books is another. How else can one say it? There are a number of minstrel tales which were printed in book form, but they were published by imbeciles. They are not the sort of thing to which one’s heart can respond’ (Moyle 1990: 161). Âs¸ık Sabit Müdamî (1918–1968) was one of the Turkish minstrels active in north-eastern Turkey in the second half of the twentieth century, where he performed together with Âs¸ık Şeref Tas¸lıova (1938–2014), Âs¸ık Murat Çobanoğlu (1940–2005) and others in the famous Âs¸ıklar Kahvesi (minstrels’ café) in Kars. We can get an impression of the difference between the ‘minstrel style’ and the ‘book style’ by comparing the text of the romance of Âs¸ık Garip as printed in a series of Turkish ‘folk books’ (chapbooks) with his performance of the hikâye in 1946 as translated by İlhan Bas¸göz.19 His tale is a lively, at times racy narrative, unfolding the traditional plot with alternating verse and prose passages. What sets the performance text off from the chapbook text are the frequent narrator’s comments. As discussed in the previous chapter, such comments not only establish a relationship between singer and audience, but illustrate also the performer’s attitude to his or her tale. Careful textualization preserves these comments and hence restores some of the performance context.

222 Interpretation The ‘folklore text’ can in theory incorporate a great deal of the singer’s performance, including gestures and musical notations. Despite its multifaceted character, the ‘folklore text’ is still a text and interpretation focuses on the text, even if the text also consists of non-linguistic symbols, such as, for instance, drawings of gestures or music notations. The difference between performance and text is categorical. This concerns also the purpose of the communicative event as outlined in Chapter 3. When asked about the purpose of the reading event, answers can, of course, be given. But these answers will vary from reader to reader and perhaps also from reading situation to reading situation. Some readers might read the text of an oral epic because they want to learn the language, some because they are interested in the cultural background of the text, some because they like heroic tales, some because they were attracted to the book by its cover – the list of reasons is long and presumably not easily predictable. In a performance situation, on the other hand, the participants might also have different reasons for their presence, but certain predictions can probably be made about the habitual motivations for participating, based on the tradition and event in question. When the Altaians used to listen to epics in the hunting season they were united in the common purpose of soothing the animals’ tutelary spirits (see Chapter 3, p. 58). This, however, does not imply that all listeners shared the same attitudes to the event. As I have argued elsewhere, I see the epic of Edige, from which illustrations have been given in earlier chapters, as a heroic epic that celebrates tribal origins. It confirms, in my understanding, what Bakhtin said about heroic epics as reflecting ‘a world of fathers and of founders of families, a world of “firsts” and “bests”’. 20 This underlying meaning of epic performance does not exclude different reactions and attitudes by different participants. I. A. Belyaev, who edited the earliest Karakalpak version of this epic at the beginning of the twentieth century, gave a vivid description of different responses to the epic: The mullahs and the pious nomads on the shores of the Aral Sea and the banks of the Amu Darya value this poem because it strengthens in many passages the religious knowledge about Islam; the poem teaches: every hero who calls on the name of God with faith will not perish in the unfortunate circumstances of the world’s vale of tears. The mothers – the Karakalpak women – who listen to Edige imagine with emotion the sufferings of Nuraddin’s [Edige’s son] mother, which she experienced because of the painful separation from her son, who was already doomed to die; they are made proud and touched by this poem also because it is dedicated to the description of the tenderness of women’s feelings and their faithfulness in family life. The young Karakalpak men very much like those passages in the poem where battles are described and the fights of heroes, who in

Performance and Interpretation  223 the wars ‘drink the blood of the enemy like water and show him no mercy: only those who run away are saved’. The Karakalpak girls listen to those parts of Edige with enthusiasm where it is described how the valiant Edige frees the khan’s daughter Aqbilek from captivity, from the hands of the heroic div [ogre] Qaratiyin, ‘whom not even an arrow can wound’. The children also like the poem; here the miraculous transformation of Edige’s father into a white swan with a yellow head is described – with the intention of saving innocent people from death, who are constrained under the rays of the scorching sun to roam in the sand deserts. It is hardly necessary to say with what pride the Karakalpak epic singer sings those passages where the time is praised when Genghis Khan and other rulers of the world listened to the songs of the folk singers; they summoned them from distant lands and, regaling them with oriental generosity, ‘fed them with their own hands with the left-overs of the meal’. (Reichl 2007: 127) No doubt the interplay of individual responses and the shared common ground of cultural understanding are both complex and dynamic. What was true for the audiences Belyaev experienced in 1903 is certainly no longer true for the late Communist and early post-Communist period during which the last Karakalpak traditional jïraw, Jumabay-jïraw, performed. Over the years, my own recordings took place in an office of the Karakalpak branch of the Uzbek Academy of Sciences, in the socalled ‘Prophylactorium’ of the University of Nukus, where I was staying at the time, in a private home in Shomanay, the singer’s hometown, in the ‘Culture Centre’ of Shomanay, in a Kazakh yurt in the vicinity of Shomanay, at a conference in Almaty and in my house near Bonn. His audience at these recording events was varied, consisting of epic scholars or university teachers in some cases, and of friends and fellow citizens of his hometown at others. The occasion closest to the traditional setting of a jïraw’s performance that I experienced was at a wedding feast with hundreds of listeners in Shomanay. While Jumabay-jïraw’s contributions to the festivities were appreciated on account of his fame, his performance was kept short, as it was clear that a large number of the audience, mostly of course the younger members of the gathering, had developed different tastes and did not show the interest required for an extended performance. Readers’ responses can be as varied, as naive, and as sophisticated as listeners’ responses. The reception mode has, however, changed, also in cases where we have a performance-oriented text and additional audiovisual material. Scrolling back, stopping and repeating audio or video sequences is possible, just as leafing back, lingering over a passage, or jumping ahead is possible for the reader. The difference between witnessing

224 Interpretation a performance and reading a text has been analysed and highlighted in a number of works, among them Walter J. Ong’s influential study on orality and literacy (1982), Paul Zumthor’s stimulating introduction to oral poetry (1990), Gregory Nagy’s discussion of the importance of performance for Homeric studies (1996), and John Foley’s persuasively argued How to Read an Oral Poem (2002). It is interesting to note that none of these scholars is a folklorist or ethnologist. Homeric and medieval studies are prominently represented among them. It is indeed in these fields that the difference between reading a text composed in writing and reading a text thought to have originated in an oral milieu (‘oral-derived’) has been intensively discussed, especially since the middle of the twentieth century, when the implications of Milman Parry’s and Albert Lord’s research on the oral epics of the South Slavs for both Homeric and medieval studies have received wide recognition.21 At first, scholarship applying the oral-formulaic theory to the epic poetry of antiquity and the Middle Ages focused primarily on the question of whether a particular work had its origin in oral tradition. An origin in oral tradition was generally understood as oral composition. The fact, however, that the basis for the reconstruction of an oral background is a written text – the epic of Beowulf in manuscript Cotton Vitellius A. XV of the British Library, for instance – allows us no direct access to a possible oral background such as folklorists or ethnologists have with their material. We have to look for signs of orality in the texts, in the knowledge that the texts were clearly not written down as ‘folklore texts’ with explicit references to performance aspects. Some of these signs, such as formulas and formulaic diction, prominent in Parry’s and Lord’s analyses, turned out to be of little value for some traditions. It can be shown that all Anglo-Saxon poetry is highly formulaic, including the Old English Metra of Boethius, poetic translations of the verses in Boethius’ Consolation of Philosophy and hence clearly works of written literature. While the written transmission of epics like Beowulf or the Chanson de Roland has made it impossible to prove oral composition, the ‘oral atmosphere’ of such works is undeniable. The very first word of Beowulf, the exclamation Hwæt (traditionally translated as ‘Lo!’), is a signal for the audience that the narration is about to begin (see Chapter 4, p. 181). While oral composition has remained a matter of controversy among medievalists, the oral performance of much of medieval literature has found wide-spread acceptance. The signs of orality in texts (such as formulaic diction or appeals to the audience) have been interpreted as ‘inscribed vocality’ and the concept of voice or vocality has become central in the discussion of medieval poetry. This has led to interpretations of texts that stress their traditionality, their vocality, and the traditional background that a contemporary audience can be expected to bring to the understanding of the performed epics.22

Performance and Interpretation  225 In medieval and other earlier ‘oral-derived’ poetry the voice is inscribed in the text and has to be extrapolated, while in oral poetry the voice is heard in performance. When oral poetry is textualized, whether as ‘folklore text’ or editorially adapted text, the singer’s voice has also become an inscribed voice. The difference between Beowulf and Edige lies in the fact that the singer’s voice is a conjecture in the former case and a reality in the latter. In both cases, however, the interpretation rests on the text (or texts, if more than one version has been preserved in writing). In both cases performance features have to be re-imagined, on the basis of ‘signs of orality’ for Beowulf, and on the basis of an extra-textual commentary (notations, pictures, audio and video files) for Edige. As interpretations of medieval ‘oral-derived’ poetry have shown, a consideration of ‘inscribed vocality’ can adjust our vision, one-dimensionally focused on reading literature composed in writing, and in this way deepen our understanding of poetry composed for performance and possibly in performance. In the fields of folklore and cultural anthropology, ‘oral poetics’ has become a powerful interpretative approach that takes full cognizance of the specificity of oral poetry. Oral epic traditions in various parts of Asia and Africa provide a wealth of material for the study of what it means when narrative and poetry are orally performed. A consideration of performance, of the communicative event which an epic performance is, helps to see the oral epic as more than a text, as well as something different from a text. Epics are stories and epics are poetry, and they can be interpreted as such. But they are also music, gesture, voice, the presence of narrator and listeners, the embeddedness in a temporal and spatial frame. The ‘aura’ of the event, as Benjamin said, is lost, in a written text as well as in the reproduced work of art. But an imaginative awareness of it can be preserved. There is no doubt that oral epics are an endangered species. Despite various attempts at reviving oral tradition, often going hand-in-hand with changes such as stage performances, the oral epic is in many areas becoming more and more marginalized. An example of this is provided by Shweta Kishore’s and Yask Desai’s film, entitled ‘Of bards and beggars’ (2003), which shows how a number of Rajasthani epic singers have nowadays no other means of supporting themselves and their families than by providing musical background in expensive restaurants or folkloric photo material for tourists. With the disappearance of context, the genre is also disappearing. Erik Routley has illustrated this death of a genre with regard to the sea-shanty: When Sir Richard Terry rambled round the docks of Tyneside to look for authentic sea-shanties, he knew that time was short. The seashanty, he tells us, is primarily a ‘labour-song’, and the screw-steamer was, when he was collecting some fifty years ago, abolishing the kind of labour without which the sea-shanty has no meaning. Shanties

226 Interpretation are songs that need a context, and the movement of civilisation was driving the sea-shanty underground, or driving it into an embalmed retirement in Sir Richard’s editions. We shall never again sing shanties in the conditions which gave them birth, because we have carefully and on principle done away with the conditions. The rhythmical discipline of the capstan and the long haul of the great sail produced song as naturally as marching produced song; but no genius has yet arisen to offer us a minstrelsy appropriate to the filling of the tank with diesel-oil, or the replacement of the nuclear power-unit. (Routley 1959: 19) While oral epics like sea-shanties might have been forced into ‘an embalmed retirement’ in some areas, at least they have an afterlife. However much one might regret some of the developments away from performance to text, there can be no doubt that the existence of a text is a gain. Textualization enables oral epics to become part of the musée imaginaire, to use André Malraux’ phrase, of world verbal art, of world literature. We have to be grateful to modern initiatives like UNESCO’s Intangible Heritage programme, but also to all the scribes, ethnologists, folklorists, and amateurs that have written down and preserved for us the words of oral epic performances and sometimes also their music and other performance aspects. Despite the precarious situation in a number of traditions, the oral epic is, however, still alive in some parts of the world. The time for dirges has happily not yet arrived.

Notes 1 The quotation is taken from the English translation, Benjamin 2008: 21. 2 On Byron’s Hebrew Melodies, see Ashton 1972; Braham harmonized the melodies; the composer of the music is Nathan, who in a later edition also harmonized the melodies. 3 See 4 On the use of a scroll in the performance of Devnārāyaṇ, see Malik 2005: 37–54. The splendid paṛ, illustrated in Smith 1991, is kept in the Tropenmuseum Amsterdam; see c87de3fa-aefb-4efb-96e1-c121ea5d749f, search for ‘pabuji phada’. 5 See Smith 1991: 54–70; Singh 1998. 6 For a general survey, see Mair 1988; Mair argues for a diffusion of this type of picture narrative from India. 7 See sala_delle_colombe/frammento_di_tabula_iliaca. 8 See Squire 2011; Petrain 2014. 9 Schumann set four of Byron’s Hebrew Songs to music, in German translation. The singer Dietrich Fischer-Dieskau writes that these lieder ‘have an important position in Schumann’s compositions, on account of their genuineness of expression and beauty of the melodic line’ (1985: 88 trans.). 10 Edited and translated (into French) by Walther 1999; see Butterfield 1997.

Performance and Interpretation  227 11 An outline of the plot of a Turkish version of the romance is found in Bas¸göz 2008: 240–42; for a study of the Turkish versions, see Türkmen 1995. For a comparative study of Turkic versions of this romance, see Korogly 1983: 248– 316. Lermontov’s adaptation is published in his collected works (Lermontov 1962). See ATU no. 974 (‘The Homecoming Husband’). 12 This Turkmen version has been reprinted several times; see Khojaev 1957; on the Turkmen versions, see Korogly 1983: 263–68. 13 Rozimbek Murādov’s version has been partially uploaded on YouTube: see (first part). Bāla-bakhshi’s version is edited in Rozimbāev 2009: 10–30. For the musical transcription of four sung passages from a Khorezmian version of this dastan, see Dzhabbarov 1971: 22–31. On the Uyghur versions of this romance, see Light 2008: 261–92. 14 See Maqsetov & Karimov 1985: 15–187 (from the baqsï Esjan Qospolatov) and Ayimbetov et al. 2011: 243–54 (from Ämet-baqsï Tariykhov); the passage in question is on pages 169–70 (1985 edition). See Shafrannikov 1959: 153–54. 15 The following text has some features of Uzbek pronunciation (e.g., eshit ‘hear’); I have consistently given the Karakalpak forms (e.g., esit). 16 See Schimmel 1992: 129–36; on âs¸ık poetry, see Boratav 1964. 17 Shafrannikov’s song text has minimal mistakes as compared to the text he prints separately; they are corrected above. For a transcription of the dutar accompaniment of ‘Arïwkhan’, based on Genjebay-baqsï’s performance, see Allanazarov 2002: 113–15. 18 On text as object and as concept, see Fish 1980: 303–21; on the concept of the ‘mental text’, see Honko: 1996; on text as textualization, see Honko 2000b. 19 Bas¸göz 2008: 30–68; for a chapbook edition, see Âs¸ık Garip 1975; on these prints in Turkey, see Boratav 1973: 65–66. 20 Reichl 2000a: 101–34; 2002. 21 For an introduction and review of the oral-formulaic theory, see Foley 1988; for a survey of the impact of the theory on medieval studies, see Foley & Ramey 2012. 22 See in particular Zumthor 1987; for medieval narrative, see, among others, the essays in Vitz, Regalado & Lawrence 2005 and, for Middle English romances, see Zaerr 2012. On the contemporary listeners’ understanding of tradition as an interpretative tool, see Renoir 1988; on oral poetics and medieval narrative, see Niles 1999. For discussions and surveys of medieval ‘oral literature’, see the essays in Reichl 2012.

Appendix A Notes on Oral Epic Traditions

The following notes are intended as an aid for readers, giving background information on some of the epic traditions mentioned in the chapters of this book. The bibliographical references are in the main restricted to publications in Western languages, principally English, and provide suggestions for further reading. I have given somewhat more information on traditions that are less familiar, because the texts of the epics are often only available either in the respective native language or in Russian. For general surveys, see: F. J. Oinas, ed., Heroic Epic and Saga: An Introduction to the World’s Great Folk Epics (Bloomington, IN, 1978). A. T. Hatto, gen. ed., Traditions of Heroic and Epic Poetry, 2 vols. (London, 1980–89). Although largely outdated, the volumes of the Chadwicks’ The Growth of Literature (see Chadwick & Chadwick 1932–40) are still worth consulting. See also the informative website ‘World Epics’ (Columbia University): worldepics/.

Sub-saharan African Epic Traditions These traditions have been well researched, both in English and French. For West African epics and epic traditions, see Bâ 1988, Bird 1972, Hale 1998, Innes 1974, Johnson 1986, Seydou 1972, 1976, Stone 1988; for the mvet epic of the Fang in Gabon, see Alexandre 1974, Pepper 1972; for the Mwindo epic of the Banyanga in the Democratic Republic of the Congo, see Biebuyck & Mateene 1971, Biebuyck 1978. A useful collection of translations and summaries is provided in Johnson, Hale, & Belcher 1997; an interpretative survey is given in Okpewho 1992; see also his earlier The Epic in Africa: Toward a Poetics of the Oral Performance (New York, 1979). On African languages (Mande, Bambara, Peul etc.), see G. T. Childs, An Introduction to African Languages (Amsterdam, 2003). On African music, including the music of epic, see R. M. Stone, ed., Africa, The Garland Encyclopedia of World Music 1 (New York, 1998).

Appendix A  229

North African Epic Traditions On oral epic and romance in North Africa, see Herzog 2012, Lyons 1995, Reynolds 1995, and Slyomovics 1987. See also Dwight Reynold’s website On the music of North Africa and the Middle East, see Danielson, Marcus & Reynolds 2002.

India The oral epic traditions of the Indian subcontinent are various and numerous. For a survey, see Blackburn, Claus, Flueckinger & Wadley 1989. In the present book only some North Indian traditions have been considered (see Grodzins Gold 1993; Smith 1991; Wadley 2004). Representatives of South Indian epic traditions are, for example, the Tulu Siri epic and the Telugu Palnāḍu epic: see L. Honko, in collaboration with Ch. Gowda, A. Honko & V. Rai, ed. & tr., The Siri Epic as Performed by Gopala Naika, 2 vols., FF Communications 265, 266 (Helsinki, 1998); G. H. Roghair, The Epic of Palnāḍu: A Study and Translation of Palnāṭi Vīrula Katha, a Telugu Oral Tradition from Andhra Pradesh, India (Oxford, 1982). On Indian music, both classical and regional music traditions, see A. Arnold, ed., South Asia: The Indian Subcontinent, The Garland Encyclopedia of World Music 5 (New York, 2000).

South East Asia, the Far East, and the Pacific The oral epic and narrative traditions of South East Asia, the Far East, and the Pacific are considerably richer than suggested by the examples given in this book. On the epic traditions of the Aceh of Sumatra, see Snouck Hurgronje 1906; Drewes 1979; and Wieringa 1998; on the oral epic traditions of West Sumatra (sijobang), see Phillips 1981. A good selection of Ainu epics (Hokkaido) is translated in Philippi 1979; see also Waley 1951. For a study of the sung narratives of various ethnic groups in Papua New Guinea, see Rumsey & Niles 2011. For the epic traditions of the Philippines, see the translations of oral epics collected on the island of Palawan by N. Revel: N. Revel-MacDonald, ed. & tr., Kudaman. Une épopée palawan chantée par Usuj (Paris, 1983); N. Revel, ed. & tr., La quête en épouse/The Quest for a Wife (Paris, 2000); see also N. Revel, ‘Singing Epics among the Palawan Highlanders (Philippines): Musican and Vocal Styles’, in Reichl 2000b: 191–210. On the p’ansori of Korea, see M. R. Pihl, The Korean Singer of Tales (Cambridge, MA, 1994); Yeonok Jang, Korean P’ansori Singing Tradition: Development, Authenticity, and Performance History (Lanham, ML, 2014).

230  Appendix A

Slavic Epic Traditions The South Slavic oral epic tradition has become well known through the work of Milman Parry and Albert B. Lord (see Parry 1987; Lord 1960). As to the East Slavic epic tradition, a fairly full presentation of the Russian bylinas is found in the Chadwicks’ Growth of Literature (Volume 2, Part 1). Russian bylinas are available in a number of English translations (e.g., Bailey & Ivanova 1998). The text corpus has been edited and studied in a great number of publications in Russian. A recent scholarly edition in 25 volumes, prepared by the Institute of Russian Literature (Pushkinskiy Dom) of the Russian Academy of Sciences, began publication in 2001 (Putilov et al. 2001). The huge tomes present critical texts, full annotations, musical transcriptions and contain CDs. On Ukrainian oral narrative songs (dumy), see Kononenko 1998; for English translations, see G. Tarnawsky & P Kilina, eds. & trs., Ukrainian Dumy (Toronto/Cambridge, MA, 1979).

Finno-Ugrian Epic Traditions The Finno-Ugrian languages are a large family of languages extending from western and central Europe into Siberia. They can be divided in various groups. The best-known European Finno-Ugrian languages are Finnish and Hungarian. Finnish, Karelian, Estonian and Saami, the language of the Saami or Lapps, are Balto-Finnic languages. Hungarian is an Ugrian language, distantly related to the Finnic languages. Not all the peoples speaking Finno-Ugrian languages have an oral epic tradition. The most important with regard to their epic traditions are the Finns and Karelians, the Komi (Finno-Permic), the Nenets (Samoyedic), the Mordvins (Finno-Volgaic) and the (northern) Khanty and the Mansi (both Ob-Ugric). For a survey, see P. Hajdu, Finno-Ugrian Languages and Peoples (London, 1975). It is typical of the epic poetry of most Finno-Ugrian peoples that it tends to consist of narrative songs rather than extended epics. This is especially true of the epic rune-songs or runos, the oral poems from which Elias Lönnrot composed the Kalevala in the nineteenth century. Longer epics (but rarely exceeding a thousand lines) are found in the traditions of the Khanty and the Mansi, and also of the Komi. For a general survey, see: L. Honko, ‘The Epic: Introduction’, in The Great Bear: A Thematic Anthology of Oral Poetry in the Finno-Ugrian Languages, ed. L. Honko, S. Timonen & M. Branch (Helsinki, 1993), 617–31. This volume presents an excellent collection of Finno-Ugrian oral poetry, in the original languages and in English translation. The section on epic contains Khanty, Komi, Mordvin, Estonian and Karelian texts (633–63). A representative collection of Finnish-Karelian runos in the original and with English translations is: M. Kuusi, K. Bosley & M. Branch, eds.

Appendix A  231 & trs., Finnish Folk Poetry: Epic. An Anthology in Finnish and English (Helsinki, 1977). On the performance of runo poetry, see L. Tarkka, Songs of the Border People: Genre, Reflexivity, and Performance in Karelian Oral Poetry, FF Communications 305 (Helsinki, 2013). On the epics of the Khanty, see A. Hatto, The World of the Khanty Epic Hero-Princes: An Exploration of a Siberian Oral Tradition (Cambridge, 2017). On the oral epics of the Nenets, see Chapter 9.

The Tungus The Turkic, Mongolian and Tungusic languages show many structural and lexical similarities and are, by many linguists, grouped together as members of the Altaic language family, with the Altay mountains believed to be the cradle of this linguistic group. The various ethnic groups speaking a Tungusic language are spread over a large area of Siberia and northern China. The Tungusic languages have been variously classified. Prevalent is a division into a northern and southern group. To the latter belongs Manchu, one of the official languages of the Manchu or Qing Dynasty in China (until 1911). For a general ethnological survey, see: W. W. Fitzhugh & A. Crowell, Crossroads of Continents: Cultures of Siberia and Alaska (Washington, 1988). For an interactive atlas of the endangered languages of Siberia, see the UNESCO language map: The epics and tales of the Tungus are part of the wider oral narrative traditions of the peoples of Siberia and northern Asia. A number of parallels exist between the tales of Tungusic-speaking ethnic groups with those of speakers of Palaeo-Siberian languages such as the Chukchi of the Chukotka Peninsula facing Alaska and the Yukagir of north-eastern Siberia. In the oral narrative traditions of the Chukchi, the Yukagir, but also the Even, the raven appears in many stories, as creator of the earth, as cultural hero, but also as trickster. These tales, which have analogues in the narrative traditions of the Native Americans of the North Pacific Coast, have been studied by a number of scholars, foremost among them Eleazar Meletinsky, who saw in the Raven cycle the elements of a PalaeoAsian mythological epic. See E. M. Meletinsky, ‘Typological Analysis of the Palaeo-Asiatic Raven Myths’, Acta Ethnographica Academiae Scientiarum Hungaricae 22 (1973): 107–55. For all their interest for comparative mythology, these Palaeo-Siberian tales cannot be called epics, at least not in the form we have them. Things are, however, different with Even and especially Evenki narratives. Zhanna K. Lebedeva, the leading scholar in the field of Even (Lamut) oral narratives, distinguishes nine main plot types, of which five are centred on women: the shamaness, the woman separated from her husband, the woman looking for a husband and others. These epics are performed in verse and prose; the monologues and dialogues of the various protagonists

232  Appendix A are in verse, while the narrative links are in prose. The five texts Lebedeva has published in Russian translation are comparatively short and fairly compact in their plots. See Zh. K. Lebedeva, Arkhaicheskiy èpos èvenov [The archaic epic of the Even] (Novosibirsk, 1981). By contrast, the Evenki narratives published are extended narratives, which both by their length and episodic structure exhibit the narrative breadth generally associated with the genre of epic. See Chapter 8.

Mongolian Epic Traditions Speakers of Mongolian languages are found today in Mongolia, in China (for the most part in the Inner Mongolia Autonomous Region and in Xinjiang), and in Russia, chiefly in the Republic of Buryatia in the Lake Baikal area and in the Republic of Kalmykia, situated to the west of the lower course of the Volga. The Mongolian languages form a large family of languages, which has been variously subdivided; see J. Janhunen, ed., The Mongolic Languages (London, 2003). Oral epics are widespread among the various Mongolian-speaking groups. Voluminous collections of Mongolian epics have been gathered from the Kalmyks, the Buryats, the Mongols of Mongolia and of China. Much is only accessible in editions in the different original languages; a number of Kalmyk and Buryat epics have been translated into Russian. Thirteen volumes of translations of oral epics in various Mongolian languages were published between 1975 and 1993 in a series edited by the Mongolist Walther Heissig of the University of Bonn (‘Mongolische Epen’); only one of these volumes is in English, all the others are in German: C. R. Bawden, tr., Mongolische Epen X. Eight North Mongolian Epic Poems, Asiatische Forschungen 75 (Wiesbaden, 1982). Compare also C. R. Bawden, tr., ‘Zong Belegt Baatar’, in The Penguin Book of Oral Poetry, ed. R. Finnegan (Harmondsworth, 1978), 430–44. Among the numerous epics of the Mongolian-speaking peoples two epic cycles stand out, the epic of Jangghar and the epic of Geser. Jangghar is primarily known in versions collected from the Kalmyks and the Oirats (in Xinjiang), while Geser is widely diffused among the Buryats, but also among the Khalkha-Mongolians (of Mongolia). The Mongols share the Geser epic with the Tibetans; the two traditions, however, show significant differences, not only in their respective language. For a brief survey of Mongolian epics, see N. Poppe, The Heroic Epic of the Khalkha Mongols, 2nd ed. (Bloomington, IN, 1979). On the Jangghar epic, see Chao Gejin, ‘The Oirat Epic Cycle of Jangar’, OT 16 (2001): 402–35. Excerpts of the Geser epic (in French translation) are found in G. Kara, ed. & tr., Chants d’un barde mongol, Bibliotheca Orientalis Hungarica 12 (Budapest, 1970).

Appendix A  233 On the music of Mongolian epics, see Pegg 2000 and 2001; compare also Bawden 1979, Nekljudov 1982, Marsh 2009, and Yang Enhong, ‘A Comparative Study of the Singing Styles of Mongolian and Tibetan Geser/Gesar Artists’, OT 13 (1998): 422–34.

Tibet The Tibetan epic of Gesar is widely known in the French translation by Alexandra David-Néél, originally published in 1931; for an English translation, see A. David-Néél & Lama Yongden, trs., The Superhuman Life of Gesar of Ling, the Legendary Tibetan Hero, as Sung by the Bards of his Country (London, 1933). On versions of the epic in Ladakh, see Roerich 1943 and S. Herrmann, ‘The Life and History of the Epic King Gesar in Ladakh’, in Religion, Myth, and Folklore in the World’s Epics: The Kalevala and its Predecessors, ed. L. Honko (Berlin, 1990), 485–502. On the music of Gesar, see Helffer 1977, 1979, 1982; compare also Stein 1939.

Turkic Oral Epics The oral epic traditions of the Turkic-speaking peoples are distributed over a wide area, extending from the Turks of Turkey to the Yakuts of north-eastern Siberia, from the Bashkirs and Tatars west of the Urals to the Turkic populations of the Altay and Sayan Mountains and the Uyghurs of the Tarim basin. The oral epics of the Central Asian Turks, especially the Kyrgyz, first became known to Western scholars through the collections and German translations of Wilhelm Radloff. His Proben der Volkslitteratur der türkischen Stämme (Examples of the folk literature of the Turkic tribes) consists of ten volumes (1866–1904), of which the first seven were edited by Radloff and volumes one to six also translated by him. His translations form the basis of the Chadwicks’ study of ‘The Oral Literature of the Tatars’ in the third volume of their Growth of Literature; this part was reprinted with an updating by Viktor Zhirmunsky; see Chadwick & Zhirmunsky 1969. For a survey of Turkic oral epics, see also Reichl 1992. The information in this book has been brought up to date in the articles I wrote for the Encyclopedia of Turkic Languages and Linguistics, edited by L. Johanson et al. and to be published by Brill (Leiden) (Azeri, Bashkir, Karachay-Balkar, Karakalpak, Kazakh, Kirghiz, Noghay, South Siberian, Tatar, Turkish, Turkmen, Uyghur, Uzbek, Yakut Oral Epics; Narrators and Singers; Oral Epics). On Turkish oral romance, see Bas¸göz 2008. On the Turkic languages, see L. Johanson & É. Á. Csató, eds., The Turkic Languages (London, 1998). An excellent survey of the music of Central Asia, including the music of epics, is Levin, Daukeyeva & Köchümkulova 2016.

Appendix B Audio/Video Examples

The following audio/video examples can be accessed on the publisher’s website: Singer, Place and Year of Recording Audio/Video Ex. 1 Audio/Video Ex. 2 Audio/Video Ex. 3 Audio/Video Ex.4 Audio/Video Ex. 5 Audio/Video Ex. 6 Audio/Video Ex. 7 Audio/Video Ex. 8

Jumabay-jïraw Bazarov (Shomanay, 1995) Döölötqan (Ulughchat, 2011) Jüsüp Mamay (Ürümchi, 1985) Jumabay-jïraw Bazarov (Shomanay, 1995) Chāri-shāir Khojamberdi-oghli (Boka, 1981) Jaqsïlïq-jïraw Sïrïmbetov (Nukus, 1983) Genjebay-baqsï Tilewmuratov (Nukus, 1981) Turghanbay-baqsï Qurbanov (Almaty, 1994)

Passage Edige, Prose (iv Ed. Reichl 2007: 147) Manas Manas Edige, Verse (V Ed. Reichl 2007: 188/ ) Alpāmish Alpamïs Ghärip-Ashïq Ghärip-Ashïq

Texts and translations of Examples 1, 3, 5, 6, 7, and 8 are given in the respective chapters. As explained above (p. 106, note 5), the text of Example 2 is slightly different from the definitive 2004 edition of Jüsüp Mamay’s version. In order to give an idea of the content of this passage, I am quoting my translation of the definitive text edition (ll. 6935–7045; Reichl 2014– 2015, vol. 1, 200–3): He [Manas] had mounted Aqqula and looked very handsome./ The lion was of mighty stature,/ Was as gigantic as the Ala-Too./ He rode out from the crowd,/ Looked like Mount Opol. (6940) When he was confronting the hero in battle,/ He roared like a mountain in an earthquake./ When Köngtöy shouted ‘Qanghay!’, the war-cry of the Kalmucks,/ All who were witnesses to the fight were astounded./ When Manas attacked

Appendix B  235 the Kalmuck hero,/ He looked very imposing on his mount, Aqqula./ Aqqula raised with his hoofs/ The dust./ Gripping their lances firmly,/ The warriors did not spare their lives, counted them for nothing,/ (6950) Cared little that their lives were hanging on a thread./ They pointed their spears,/ Held their lances in readiness and started to fight,/ Rushed one on the other and stabbed at each other./ The qalday [Kalmyk official] Köngtöy of the Kalmyks,/ Enraged Manas of the Kyrgyz./ These two fought with each other/ An exceedingly violent duel./ They destroyed their shields,/ Their corselets were torn apart./ (6960) Without attempting to flee, they endured the duel./ This was a grievous enterprise, out of the ordinary./ They recommenced their fight:/ Took their bloody spears/ To stick them straight into the heart./ When they pushed their spears into the adversary’s chest,/ They did not stick fast./ Although the lance was pushed in, no blood was flowing./ They were skilful to avoid being wounded./ Köngtöy thought in his mind:/ (6970) ‘If I put layer after layer of iron/ Round my body,/ His stabs will not pierce through./ If I hold my lance in my hand,/ His blows will not cut into my body./ If the sword is in its sheath and I pull it out,/ It will not get indented, when I hit a stone,/ But will split the stone./ If I beat the mountain with the battle-axe on my belt,/ It will stagger, /(6980) It will tumble over and be destroyed./ If I have my cudgel hanging on the saddle,/ If I continue to gallop and get stronger/ Without getting tired,/ If my horse perseveres under me,/ Then my glorious name Er Köngtöy/ Will endure./ He is so much less than me!/ His years are few, his strength is little,/ Between eleven and twelve years old,/ (6990) He is a small child./ If I can’t kill him with a thrust of my lance,/ Stirring his intestines,/ If I don’t cleave his gullet,/ Fighting against him in this way,/ What’s the use then of being Köngtöy?’/ Burning with anger the pig was getting ready./ Black like a cloud, bursting with rain,/ His face was filled with rage:/ ‘If I don’t squirt my venom/ (7000) Like a tarantula, a scorpion, a snake,/ If I can’t stab him to death,/ Then I don’t want to be Köngtöy any more!’/ He was seized by fury/ And rushed into the fight./ When Manas saw the mighty figure of the Kalmyk,/ Filled with the spirit of unmitigated violence,/ He rode to him, cut off his path./ They circled round each other,/ Were both moving restlessly./ (7010) ‘I won’t let him free, I will stab him to death’, he thought,/ And our lion confronted his enemy./ He gave his horse the spurs, made it gallop,/ So that the dust rose up in clouds./ He made the tulpar [true-bred], the racer Aqqula,/ Jump like a heavy gale,/ So that its mane and its tail fluttered before the wind/ And its chest jutted out like the horns of the ibex./ With the war-cry ‘Manas!’ on his lips/ Young Manas galloped into the mêlée./ (7020) When he heard Manas’s horse,/ Mighty Köngtöy was taken by surprise./ Köngtöy, filled with thirst for revenge,/ Thrust his spear/ And touched our hero’s flank./ Manas turned Aqqula’s mouth round,/ Came sideways to him/ And stabbed Köngtöy once into his ribs./ Overcome by this blow, he broke down./ The young tiger, our hero,/ (7030) Facing the

236  Appendix B enemy in combat,/ Had defeated the mighty warrior, the hero Köngtöy./ Exhausted, Köngtöy’s men shook and staggered,/ Tottered around without direction./ There was no one who talked sensibly;/ They became pale, lost their senses./ Trembling with fear, they dared not approach,/ The men who were Köngtöy’s support in battle./ When they saw Manas’s aweinspiring looks,/ They became dazed and were unable to move,/ (7040) They cowered and lost their manliness./ Many lost their heroism./ The Kalmyks, used to having dominion over others,/ Had not a bit of strength left in their limbs./ The Kalmyks had lost their cunning,/ They knew no way out of the dilemma. The full text and tanslation of Example 4 is found in Reichl 2007: 188 (text) and 294–95 (translation). For convenience’s sake I am quoting the translation (without the music marks); [S] = singing, [R] = recitative (declamation) [S] Oh, my God!/ The wife of Tokhtamysh,/ She who looks after the household,/ Came to Tokhtamysh/ And complained:/ “Hä, Tokhtamysh, Tokhtamysh,/ You are the lord of the land and my khan./ Why don’t you believe what I told you/ And go on living amongst the Noghay?/ Since you did not believe what I told you,/ You did not listen to my words,/ When I said: ‘Now destroy/ This child!’” she said./ [R] “While you are still alive,/ Edige will claim possession of your land./ He will ascend the golden throne/ And Edige will take as his own/ The Noghay of the innumerable yurts,” she said./ “When these two daughters of ours of equal beauty,/ Qanïkey and Tïnïkey,/ Swaying like a cypress,/ Will become marriageable in the near future,/ He will take your daughters,” she said./ “What else should he do but take your throne?/ When he has subjugated you,/ You will lose your head,/ Your body will remain behind as a carcass./ ‘Drive this only son/ Away from the land and kill him!’ I said./ You did not do what I said,/ You did not act according to my words./ His mind and intellect/ Have indeed surpassed you:/ Edige solved the problems that you could not solve:/ You know that, don’t you, oh khan?” she said.

Appendix C Discography

This discography is neither extensive nor exhaustive, but is intended to facilitate access to relevant audio (and video) material. A number of the oral epic traditions discussed in this book are illustrated (with audio/video files) on UNESCO’s website of Intangible Cultural Heritage: On this list are, among others, ‘Singing to the accompaniment of the gusle’, ‘Epic art of Gorogly (Turkmenistan)’, ‘Kazakh traditional art of dombra kuy’, ‘Kyrgyz epic trilogy Manas, Semetey, Seytek’, ‘Mongol tuuli, Mongolian epic’, ‘Art of the Azerbaijani ashiq’, ‘Âs¸ıklık (minstrelsy) tradition (Turkey)’, ‘Gesar epic tradition’, ‘Manas (China)’, ‘Mongolian art of singing khoomei’, ‘Al-sirah al-hilaliyya epic (Egypt)’, ‘Art of the meddah, public storytellers’, ‘Olonkho, Yakut heroic epic’, ‘Pansori epic chant’, ‘Traditional music of the morin khuur’. One should keep in mind that the items have been prepared by the various national commissions of UNESCO; they have a political agenda and do not always reflect current scholarly opinion. For discographies, see the relevant entries in the New Grove Dictionary of Music and Musicians, ed. S. Sadie, 29 vols. (2001); online version: A number of book publications come with CDs (or previously, records). This is true of two series repeatedly mentioned here, the French series ‘Classiques Africains’ (Paris: Armand Colin) and the Russian series ‘Pamyatniki Fol’klora Narodov Sibiri i Dal’nego Vostoka [Monuments of the folkore of the peoples of Siberia and the Far East] (Novosibirsk: Nauka). For representatives of the French series, see Pepper 1972 (Fang mvet), Seydou 1972 (Fulani Silâmaka & Poullôri), and Seydou 1976 (Fulani Ham-Bodêdio). For examples of the Russian series, see Ayzenshtadt & Sheykin 1990 (Evenki); Illarionov et al. 1996 (Yakut Är Soghotokh), Kazagacheva 1997 (Altaian epics), Nazarenko 1998 (Shor epics), Orus-ool

238  Appendix C et al. 1997 (Tuva), Reshetnikova 1993 (Yakut Kïïs Däbiliyä), Sadalova 2002 (Altaian folktales), Shevtsov 1997 (Khakas epic), Skvortsova 2001 (Nenets). The 2000-edition of Lord’s Singer of Tales (1960) includes a CD, with audio and video clips. Audio and video clips are also included on a CD in my edition of Edige (Reichl 2007). CDs are also found in the following books quoted: Danielson, Marcus & Reynolds 2002 (Garland Encyclopedia of Music, 6: Middle East); Levin 1996 (Uzbekistan); Levin with Süzükei 2006 (Tuva); Pegg 2001 (Mongolian); Putilov, Gorelov et al. 2001 (bylina); Ruckert & Widdess 2000 (Garland Encyclopedia of Music, 5: India); van Tongeren 2004 (overtone singing); Youssefzadeh 2002 (Khorasanian bakhshis). A DVD is found in Børdahl & Ross 2002 (Chinese storytellers of the Yanghzhou tradition). Cassettes accompany Reinhard & de Oliveira Pinto 1989 (Turkish âs¸ık). On Rajasthani bards, see the film by Kishore & Desai 2003. Some of the publications referred to have audio files (and partly also video files) available on the publishers’ websites: See for Harris 2017 (Yakut): harris/storytelling; for Levin, Daukeyeva & Köchümkulova 2016 (Central Asia): https:// for Rumsey & Niles 2011 (Papua New Guinea): https://press.anu. Click Additional Resources. As mentioned in Chapter 5, videos of a set of 124 episodes of the Manas cycle can be accessed on the website of the Aigine Cultural Research Center of Kyrgyzstan: These videos have also been uploaded on YouTube. YouTube is a source of information, almost inexhaustable, but of varying quality.

Glossary of Musical Terms and Non-­Western Instruments

Musical Terms


a short ornamental note (also known as Vorschlag)

bel canto

‘beautiful song’, as typical of eighteenth-­century Italian singing styles


‘a melodic or harmonic formula that occurs at the end of a composition’ (HDM)


‘religious chanting in a recitative style’ (HDM)


a type of singing in a monotone

church modes

a medieval system of scales, divided into authentic and plagal modes, the latter marked by hypo-­‘below’; the names are taken from the modes of ancient Greece


‘a rapid passage, run, trill, or similar virtuoso-­like material’ (HDM)

commedia per musica

‘comedy for music’, comic opera


quarter note


prolongation of a note


a sliding transition from one note to another

grace note

ornamental note, printed in small type

Gregorian chant

liturgical chant of the Roman Catholic Church, also known as plainchant


‘the simultaneous use of slightly or elaborately modified versions of the same melody by two (or more) performers’ (HDM)


‘one voice leads melodically, being supported by an accompaniment in chordal or a slightly more elaborate style’ (HDM)


playing or singing a sequence of notes, smoothly joined to one another

240  Glossary of Musical Terms and Non-­Western Instruments leitmotif

recurring melodic-­harmonic figures that characterize different persons, objects or events


‘an expressive vocal passage sung to one syllable’ (HDM)


half note


singing or chanting on the same pitch


‘a musical ornament consisting of the alternation of the written note with the note immediately below it’ (HDM)


‘an indication that the voice must approximate speech’ (HDM)


plucking the strings of a normally bowed chordophone


eighth note


‘speechlike reiteration of the same note, slight inflections, irregular rhythms, purely syllabic treatment of the text, etc.’ (HDM)


the different ranges of the human voice


‘an elastic, flexible tempo involving slight accelerandos and ritardandos that alternate according to the requirements of musical expression’ (HDM)


sixteenth note


an opera with arias and spoken dialogues


sustaining a tone


‘speech-­song’ (C. Stumpf)


playing or singing a sequence of notes in a detached form


using the same melody sung to every verse-­line (or every two verse-­lines)

stile concitato

‘agitated style’, dramatic expression in music as practiced by Claudio Monteverdi


using more than one melody for the lines of a stanza or verse-­passage


singing one note per syllable

tempo giusto

in a strict tempo, in an exact rhythm


rapid iteration of the same note

Non-­Western Musical Instruments For general information on the instruments mentioned in this book, see L. Libin, ed., The Grove Dictionary of Musical Instruments, 2nd ed., 5 vols. (Oxford, 2014). The instruments of the various ethnic groups of the former Soviet Union are described and illustrated in K. Vertkov, G. Blagodatov & È.

Glossary of Musical Terms and Non-­Western Instruments  241 Yazovitskaya, Atlas muzykal’nykh instrumentov narodov SSSR [An atlas of the musical instruments of the USSR] (Moscow, 1963). On the history of musical instruments, see J. Montagu, Origins and Development of Musical Instruments (Lanham, MD, 2007). al-­cūd

lute (Arabs)


a type of saz


xylophone (Mande)


three-­stringed fiddle (Gonds, Madhya Pradesh)


plucked chordophone (Ukrainians)


oboe-­type wind instrument (Uzbeks)


gongs (Minangkabau)


plucked zither (Khakas)


string instrument


two-­stringed bowed chordophone (North Indian)


metal tongs (North Indian)


a type of saz

cura ırızva

a type of saz


frame-­drum (Uzbeks)


barrel-­shaped drum (North Indian)


two-­stringed plucked chordophone without frets (Uzbeks)


two-­stringed plucked chordophone with frets (Kazakhs)


six-­string harp-­lute with a calabash resonator (Mande)

dutar (dutār)

two-­stringed plucked lute (Uzbeks, Karakalpaks, and others)

dywwalæstænon fændyr

twelve-­stringed harp


spike-­fiddle (Uzbeks)


(Karakalpaks) = ghijjak


(Turkmens) = ghijjak


one-­stringed bowed chordophone (South Slavs)


lyre (Anglo-­Saxons)


plucked lute with three or four strings (Fulbe)


‘instruments that consist simply of elastic material (metal, wood) capable of producing sound’ (HDM), e.g. gongs, rattles, xylophone (not to be confused with ideophone, see p. 165)


spike-­fiddle (Arabs and others)

242  Glossary of Musical Terms and Non-­Western Instruments kantele

plucked zither (Finns)

khissyn fændyr

bowed lute with two or three strings (Ossets)


bowed chordophone (Mongols)


a type of lyre (ancient Greeks)


plucked chordophone (Ukrainian)


plucked harp-­like instrument (Mande)


board-­zither (Minangkabau)


one-­stringed bowed chordophone (Albanians)


hurdy-­gurdy (Ukrainians)


lyre (ancient Greeks)

meydan sazı

a type of saz

morin khuur

two-­stringed ‘horse-­head fiddle’ (Mongols)


a type of harp with three or more strings and three or more calabashes as resonators (Fang, Gabon)


metal pipe, scraped with a metal bar (Mande)


long-­necked lute (ancient Greeks)


a type of lyre (ancient Greeks)


pipes (Minangkabau)

qay qomus

two-­stringed plucked lute (Shors)


two-­stringed bowed chordophone (Karakalpaks)


‘horse-hair qobïz’ (Kazakhs)


three-­stringed plucked chordophone (Kyrgyz)


spike-­fiddle (Arabs)


spike-­fiddle (North India)


plucked lute (Uyghurs)


plucked lute (Uzbeks, Tajiks)


long-­necked bowed lute (Uyghurs)


long-­necked lute (Turks, Azerbaijanians)


a type of reed-­pipe (Karachay-­Balkar)


long-­necked lute (Uzbek, Uyghur and others)


small tanbur (Bosnians)


large frame-­drum (Arabs)


two-­stringed plucked lute (Altaians)


plucked lute (Mongols)


Abbreviations ATU:  see Uther 2011 BSOAS:  Bulletin of the School of Oriental and African Studies CSOLC:  Cambridge Studies in Oral and Literate Culture ÈNSSSR/ÈNE:  Èpos narodov SSSR/Èpos narodov Evrazii [The epic of the peoples of the Soviet Union /of Eurasia] FF Communications:  Folklore Fellows Communications HDM:  see Apel 1969 JAF:  Journal of American Folklore Motif:  see Thompson 1975 OED:  Oxford English Dictionary Online edition (3rd ed. Oxford, 2005) OT:  Oral Tradition PFNSiDV:  Pamyatniki Fol’klora Narodov Sibiri i Dal’nego Vostoka [Monuments of the folklore of the peoples of Siberia and the Far East] PMLA:  Publications of the Modern Language Association Abbaslı, İ. & B. Abdulla, eds. 2005. Koroğlu dastanı [The epic of Koroğlu]. Baku. Abdullaev, R. C. 1989. ‘Bytovanie dastanov v Uzbekistane’ [The occurrence of dastans in Uzbekistan]. In Zemtsovsky 1989: 113–17. Abramovich-Gomon, A. 1999. The Nenets’ Song: A Microcosm of a Vanishing Culture. Aldershot. Adambaeva, T., ed. 1991. Jïraw namalarï [Jïraw melodies]. Music transcriptions by M. Jiyemuratov. Nukus. Ahmedaja, A. 2013. ‘The Lahutë between Everyday Practice and Symbolism.’ Music in Art 38: 144–60. Alekseev, È. E. 1996. ‘O muzykal’nom voploshchenii olonkho’ [On the music of the olonkho]. In Illarionov et al. 1996: 42–72. Alekseev, N. A. 1980. Rannie formy religii tyurkoyazychnykh narodov Sibiri [Early forms of religion of the Turkic-speaking peoples of Siberia]. Novosibirsk. Alexandre, P. 1974. ‘Introduction to the Fang Oral Art Genre: Gabon and Cameroon Mvet.’ BSOAS 37: 1–7.

244 Bibliography Allanazarov, D. 2002. Qaraqalpaq khalïq sazlarï [Karakalpak folk melodies]. Nukus. Allen, R. E., tr. 1996. The Dialogues of Plato. III. Ion, Hippias Minor, Laches, Protagoras. New Haven, CT. Apel, W. 1969. Harvard Dictionary of Music. 2nd ed., rev. and enlarged. Cambridge, MA. Arsunar, F., ed. 1963. Köroğlu. Ankara. Ashton, T. L. 1972. ‘Byronic Lyrics for David’s Harp: The Hebrew Melodies.’ Studies in English Literature: 1500–1900 12: 665–81. Âşık Garip ile Şah Senem [Âşık Garip and Şah Senem]. 1975. Saz şairleri hikâye­ leri 2. Istanbul. Auldjo, J. 1835. Journal of a Visit to Constantinople, and Some of the Greek Islands, in the Spring and Summer of 1833. London. Ayimbetov, N. Q., S. Bahadïrova & A. I. Al’niyazov, eds. 2007. Qaraqalpaq fol’klorï [Karakalpak folklore]. Books 1–8 (in one volume). Nukus. Ayimbetov, N. Q., S. Bahadïrova & A. I. Al’niyazov, eds. 2009. Qaraqalpaq fol’klorï [Karakalpak folklore]. Books 9–13 (in one volume). Nukus. Ayimbetov, N. Q., S. Bahadïrova & A. I. Al’niyazov, eds. 2011. Qaraqalpaq fol’klorï [Karakalpak folklore]. Books 27–42 (in one volume). Nukus. Ayzenshtadt, A. M. & Yu. I. Sheykin. 1990. ‘Muzyka èvenkiyskikh skazaniy’ [The music of the Evenki tales]. In Myreeva, A. H., ed. & tr. Èvenkiyskie geroicheskie skazaniya [Evenki heroic tales]. PFNSiDV. Novosibirsk. 89–124. Bâ, Ousmane, ed. & tr. 1988. Silamaka Fara Dikko. Ein westafrikanisches Epos in den Bambara-Versionen von Mamadou Kida und Almami Bah (Mali). Berlin. Bailey, J. & T. Ivanova, trs. 1998. An Anthology of Russian Folk Epics. Armonk, NY. Bakchiev, T. A. 2015. Kyrgyzskie èpicheskie skaziteli [Kyrgyz epic narrators]. Bishkek. Bakhtin, M. M. 1981. ‘Epic and Novel: Toward a Methodology for the Study of the Novel.’ In his The Dialogic Imagination: Four Essays. Ed. & tr. M. Holquist & C. Emerson. Austin, TX. 3–40. Bartók, B. 1954. Musical transcription of No. 4 The Capitivity of Đulić Ibrahim. In Serbocroation Heroic Songs, collected by Milman Parry, edited and translated by Albert Bates Lord. Vol. I. Novi Pazar: English Translations. Cambridge, MA/Belgrade. 437–67. Bartók, B. 1976. Turkish Folk Music from Asia Minor. Ed. B. Suchoff. Princeton, NJ. Başgöz, İ. 1952. ‘Turkish Folk Stories About the Lives of Minstrels.’ JAF 65: 331–39. Başgöz, İ. 1975. ‘The Tale-Singer and His Audience: An Experiment to Determine the Effect of Different Audiences on a Hikaye Performance.’ In Folklore: Performance and Communication. Ed. D. Ben-Amos & K. S. Goldstein. The Hague. 143–203. Başgöz, İ. 1986. ‘Digression in Oral Narrative: A Case Study of Individual Remarks by Turkish Romance Tellers.’ JAF 99: 5–23. Başgöz, İ. 2008. Hikâye: Turkish Folk Romance as Performance Art. Bloomington, IN. Basilov, V. N. 1970. Kul’t svyatykh v islame [The cult of saints in Islam]. Moscow.

Bibliography  245 Baskakov, N. A. 1952. Karakalpakskiy yazyk. II. Fonetika i morfologiya. Chast’ 1 [The Karakalpak language. II. Phonetics and morphology. Part 1]. Moscow. Bauman, R. 1977. Verbal Art as Performance. Prospect Heights, IL. Bauman, R. 1986. Story, Performance, and Event: Contextual Studies of Oral Narrative. Cambridge. Bauman, R. & D. Braid. 1998. ‘The Ethnography of Performance in the Study of Oral Tradition.’ In Teaching Oral Traditions. Ed. J. M. Foley. New York. 106–22. Bawden, C. R. 1979. ‘Remarks on some Contemporary Performances of Epics in the MPR.’ In Die mongolischen Epen. Bezüge, Sinndeutung und Überlieferung (Ein Symposium). Ed. W. Heissig. Asiatische Forschungen 68. Wiesbaden. 37–43. Bell, E. 2008. Theories of Performance. Los Angeles. Belyaev, V. 1937. ‘Yakutskie narodnye pesni’ [Yakut folk songs]. Sovetskaya Muzyka 1937.9: 11–26. Belyaev [Beliaev], V. M. 1975. Central Asian Music: Essays in the History of the Music of the Peoples of the U.S.S.R. Ed. & tr. G. Slobin & M. Slobin. Middletown, CT. Ben-Amos, D. 1976. ‘Analytical Categories and Ethnic Categories.’ In Folklore Genres. Ed. D. Ben-Amos. Austin, TX. 215–42. Benjamin, W. 2008. The Work of Art in the Age of Its Technical Reproducibility, and Other Writings on Media. Ed. & tr. M. W. Jennings, E. Jephcott et al. Cambridge, MA. Bernstein, L. 1973. The Unanswered Question: Six Talks at Harvard. Cambridge, MA. Biebuyck, D. P. 1978. Hero and Chief: Epic Literature from the Banyanga Zaire Republic. Berkeley, CA. Biebuyck, D. & K. C. Mateene, eds. & trs. 1971. The Mwindo Epic from the Banyanga (Congo Republic). Berkeley, CA. Bin-Gorion, M. J. 1920. Der Born Judas. IV. Weisheit und Torheit. Leipzig. Bird, C. 1972. ‘Heroic Songs of the Mande Hunters.’ In African Folklore. Ed. R. M. Dorson. Bloomington, IN. 275–93. Bird, C. 1976. ‘Poetry in the Mande: Its Form and Meaning.’ Poetics 5: 89–100. Blackburn, S. H., P. J. Claus, J. B. Flueckinger & S. S. Wadley, eds. 1989. Oral Epics in India. Berkeley, CA. Bogatyrev, P. & R. Jakobson. 1929. ‘Die Folklore als eine besondere Form des Schaffens.’ In Donum Natalicium Schrijnen. Verzameling van opsteilen door oudleerlingen en bevriende vakgenooten, opgedragen aan mgr. prof. dr. Jos. Schrijnen bij gelegenheid van zijn zestigsten verjaardag, 3 Mei 1929. Nijmegen. 900–13. Böhtlingk, O. 1964. Über die Sprache der Jakuten. The Hague. Boratav, P. N. 1964. ‘La littérature des cāšïq.’ In Philologiae Turcicae Fundamenta. Vol. 2. Ed. P. N. Boratav. Wiesbaden. 129–47. Boratav, P. N. 1973. Türk Halkbilimi. I. 100 Soruda Türk Halkedebiyatı [Turkish folklore. I. Turkish folk literature in one hundred questions]. 2nd ed. Istanbul. Børdahl, V. 1999. ‘Introduction.’ In The Eternal Storyteller: Oral Literature in Modern China. Ed. V. Børdahl. London. 1–14. Børdahl, V. & J. Ross. 2002. Chinese Storytellers: Life and Art in the Yangzhou Tradition. Boston. [With DVD] Bose, F. 1953. Musikalische Völkerkunde. Freiburg i. Br.

246 Bibliography Bose, F. 1958. ‘Law and Freedom in the Interpretation of European Folk Epics.’ Journal of the International Folk Music Council 10: 29–34. Bowra, C. M. 1945. From Virgil to Milton. London. Bowra, C. M. 1952. Heroic Poetry. London. Bowra, C. M. 1962. Primitive Song. London. Bremmer J. & H. Roodenburg, eds. 1991. A Cultural History of Gesture. Ithaca, NY. Bronson, B. H. 1969. ‘The Interdependence of Ballad Tunes and Texts.’ In his The Ballad as Song. Berkeley, CA. 37–63. Bronson, B. H., ed. 1976. The Singing Tradition of Child’s Popular Ballads. Princeton, NJ. Buck, C. D. 1949. A Dictionary of Selected Synonyms in the Principal IndoEuropean Languages: A Contribution to the History of Ideas. Chicago. Butler, H. E., ed. & tr. 1920–1922. The Institutio Oratoria of Quintilian. 4 vols. London. Butterfield, A. 1997. ‘Aucassin et Nicolette and Mixed Forms in Medieval French.’ In Harris & Reichl 1997: 67–98. Calame-Griaule, G., ed. 1977a. Langage et cultures africaines. Essais d’ethnolinguistique. Paris. Calame-Griaule, G. 1977b. ‘Pour une étude des gestes narratifs.’ In CalameGriaule 1977a: 303–59. Calbris, G. 2011. Elements of Meaning in Gesture. Tr. M. M. Copple. Gesture Studies 5. Amsterdam. Carter, H. H. 1961. A Dictionary of Middle English Musical Terms. Bloomington, IN. Castagné, J. 1930. ‘Magie et exorcisme chez les Kazak-Kirghizes et autres peuples turks orientaux.’ Revue des études islamiques 4: 53–151. Caton, M. 2002. ‘The Concept of Mode in Iranian Music: Shūr.’ In Danielson et al. 2002: 59–76. Chadwick, H. M. & N. K. Chadwick. 1932–40. The Growth of Literature. 3 vols. Cambridge. Chadwick, N. K. & V. Zhirmunsky. 1969. Oral Epics of Central Asia. Cambridge. Chapman, G., tr. 1884. Homer’s Iliad. London. [Or.1610-1611 under the title The Illiads] Chicherov, V. I. 1982. Shkoly skzaliteley Zaonezh’ya [Schools of singers in the region of ‘Trans-Onega’]. Moscow. Chodzko, A., tr. 1842. Specimens of the Popular Poetry of Persia, as Found in the Adventures and Improvisations of Kurroglou, the Bandit-Minstrel of Northern Persia, and in the Songs of the People Inhabiting the Shores of the Caspian Sea. London. Colarusso, J., tr. 2002. Nart Sagas from the Caucasus: Myths and Legends from the Circassians, Abazas, Abkhaz, and Ubykhs. Princeton, NJ. Crocker, R. L. 1986. A History of Musical Style. Corr. ed. New York. Danielson, V., S. Marcus & D. Reynolds, eds. 2002. The Middle East. The Garland Encyclopedia of World Music 6. New York. [With CD] Daukeyeva, S. 2016. ‘Kazakh Küi.’ In Levin, Daukeyeva & Köchümkulova 2016: 235–64. Davies, S. 1994. Musical Meaning and Expression. Ithaca, NY. Dawut, R. & E. Anderson. 2016. ‘Dastan Performance among the Uyghurs.’ In Levin, Daukeyeva & Köchümkulova 2016: 406–20.

Bibliography  247 Dietrich, W. 2000. ‘The Singing of Albanian Heroic Poetry.’ In Reichl 2000b: 83–94. Dilley, R. M. 2002. ‘The Problem of Context in Social and Cultural Anthropology.’ Language and Communication 22: 437–56. Dimock, Wai Chee. 2007. ‘Introduction: Genres as Fields of Knowledge.’ PMLA 122: 1377–88. Dobrovol’sky, B. M. 1965. ‘O napevakh nenetskikh èpicheskikh pesen’ [On the melodies of the Nenets epic songs]. In Kupriyanova 1965: 757–80. Dobrovol’sky, B. M. & V. V. Korguzalov, eds. 1981. Byliny. Russkiy muzykal’nyi èpos [Bylinas. The Russian musical epic]. Moscow. Doke, C. M. 1935. Bantu Linguistic Terminology. London. Drewes, G. W. J., ed. & tr. 1979. Hikajat Potjut Muhamat: An Achehnese Epic. Bibliotheca Indonesica 19. The Hague. DuBois, T. A. 2012. ‘Oral Poetics: The Linguistics and Stylistics of Orality.’ In Reichl 2012: 203–24. Duff, D., ed. 2000. Modern Genre Theory. Harlow. Dumézil, G., tr. 1965. Le Livre des Héros. Légendes sur les Nartes. Paris. Dundes, A. 1966. ‘Metafolklore and Oral Literary Criticism.’ Monist 50: 505–16. Dyushaliev, K. 1993. Pesennaya kul’tura kyrgyzskogo naroda [The singing culture of the Kyrgyz people]. Bishkek. Dzhabbarov, A. Kh. 1971. ‘Uzbekskiy dastan (èpos)’ [The Uzbek dastan (epic)]. In Voprosy muzykoznanija [Problems of musicology]. II. Ed. A. Kh. Dzhabbarov. Tashkent. 13–35. Dzhumalieva, T. & A. Temirbekova, eds. 2000. Istoriya kazakhskoy muzyki. I. Traditsionnaya muzyka kazakhskogo naroda: pesennaya i instrumental’naya [History of Kazakh music. I. The traditional music of the Kazakh people: vocal and instrumental]. Almaty. Eberhard, W. 1955. Minstrel Tales from Southeastern Turkey. Berkeley, CA. Èl’darova, È. 1984. Iskusstvo ashygov Azerbaydzhana [The art of the aşıqs of Azerbaijan]. Baku. Elemanova, S. A. 2012. Nasledie tyurkskoy kul’tury. Istoricheskiy obzor kazakhskoy traditsionnoy muzyki [The heritage of Turkic culture. A historical survey of Kazakh traditional music]. Almaty. Eliade, M. 1959. The Sacred and the Profane: The Nature of Religion. Tr. W. R. Trask. New York. Eliade, M. 1964. Shamanism: Archaic Techniques of Ecstasy. Tr. W. R. Trask. Princeton, NJ. Ellingson, T. 1992a. ‘Transcription.’ In Myers 1992: 110–52. Ellingson, T. 1992b. ‘Notation.’ In Myers 1992: 153–64. Ellinwood, L., ed. & tr. 2015. The ‘Musica’ of Hermannus Contractus. Rochester, NY. Emel’yanov, N. V. 1980. Syuzhety yakutskikh olonkho [The plots of the Yakut olonkhos]. Moscow. Emel’yanov, N. V. 1983. Syuzhety rannikh tipov yakutskikh olonkho [The plots of the early types of Yakut olonkhos]. Moscow. Emsheimer, E. 1991. ‘On the Ideology, Function, Musical Structure and Style of Performance of Siberian Epics.’ Tr. R. Carroll. In his Studia Ethnomusicologica Eurasiatica. II. Stockholm. 151–66. Entwistle, W. J. 1969. European Balladry. Oxford.

248 Bibliography Erdely, S. 1995. Music of Southslavic Epics from the Bihać Region of Bosnia. New York. Erdely, S. 2000. ‘Music of South Slavic Epics.’ in Reichl 2000b: 69–81. Erdener, Y. 1995. The Song Contests of Turkish Minstrels: Improvised Poetry Sung to Traditional Music. New York. Erzakovich, B. 2000. ‘Èpos i ego muzykal’noe voploshchenie’ [The epic and its music]. In Dzhumalieva & Temirbekova 2000: 93–108. Erzakovich, B. 2003. ‘Muzykal’noe voploshchenie èposa “Kozy-Korpesh i BayanSulu”’ and ‘Muzykal’noe voploshchenie èposa “Kyz-Zhibek”’ [The music of the epic Kozy-Korpesh and Bayan Sulu/The music of the epic Kyz-Zhibek]. In Qozï Körpesh-Bayan-Suluw. Qïz Jibek. Qazaqtïng ghashïqtïq jïrlarï [Qozï-Körpesh and Bayan Suluw. Qïz Jibek. Kazakh love epics]. Ed. & tr. S. S. Kirabaev & E. A. Potseluevsky. ÈNE. Moscow. 391–418, 419–36. Erzakovich, B., B. Karakulov & Z. Kosbakov, eds. 1982. Kazakhskiy muzykal’nyi fol’klor [Kazakh folk music]. Almaty. Eshānqulov, J. & R. Chāriev, eds. 2018. Alpāmish. Ozbek khalq dāstāni. Aytuwchi: Ozbekistān khalq bakhshisi Abdunazar Pāyānov [Alpāmish. An Uzbek folk dastan. Narrator: Uzbekistan’s folk bakhshi Abdunazar Pāyānov]. Tashkent. Fabb, N. 1997. Linguistics and Literature: Language in the Verbal Arts of the World. Oxford. Faral, E. 1971. Les arts poétiques du XIIe et du XIIIe siècle. Recherches et documents sur la technique littéraire du Moyen Âge. Paris. Feldman, W. R. 1980. ‘The Uzbek Oral Epic: Documentation of Late Nineteenth and Early Twentieth Century Bards.’ Ph.D. Thesis Columbia University. Fine, E. 1983. ‘In Defense of Literary Dialect: A Response to Dennis R. Preston.’ JAF 96: 323–30. Fine, E. 1984. The Folklore Text: From Performance to Print. Bloomington, IN. Finnegan, R. 1970. Oral Literature in Africa. Oxford. Finnegan, R. 1974. ‘How Oral Is Oral Literature?’ BSOAS 37: 52–64. Finnegan, R. 1992. Oral Traditions and the Verbal Arts: A Guide to Research Practices. London. Firth, R., with M. McLean. 1990. Tikopia Songs: Poetic and Musical Art of a Polynesian People of the Solomon Islands. CSOLC 20. Cambridge. Fischer-Dieskau, D. 1985. Robert Schumann. Das Vokalwerk. Munich. Fish, S. 1980. Is There a Text in This Class? The Authority of Interpretive Communities. Cambridge, MA. Fludernik, M. 2003. ‘Chronology, Time, Tense and Experientiality in Narrative.’ Language and Literature 12: 117–34. Foley, J. M. 1988. The Theory of Oral Composition: History and Methodology. Bloomington, IN. Foley, J. M. 1992. ‘Word-Power, Performance, and Tradition.’ JAF 105: 275–301. Foley, J. M. 1995. The Singer of Tales in Performance. Bloomington, IN. Foley, J. M. 2002. How to Read an Oral Poem. Urbana, IL. Foley, J. M. & P. Ramey. 2012. ‘Oral Theory and Medieval Literature.’ In Reichl 2012: 71–102. Fortis, Alberto. 1774. Viaggio in Dalmazia. 2 vols. Venice. Fowler, A. 1982. Kinds of Literature: An Introduction to the Theory of Genres and Modes. Oxford.

Bibliography  249 Frasca, R. A. 1990. The Theatre of the ‘Mahābhārata’: Terukkūttu Performances in South India. Honolulu. Funk, D. A. 2005. Miry shamanov i skaziteley [The worlds of shamans and narrators]. Moscow. Geertz, C. 1973. ‘Thick Description: Towards an Interpretive Theory of Culture.’ In his The Interpretation of Cultures: Selected Essays. New York. 3–30. Genette, G. 1972. ‘Discours et récit: essai de méthode’. In his Figures III. Paris. 65–273. Genette, G. 1980. Narrative Discourse: An Essay in Method. Tr. J. E. Lewin. Ithaca, NY. Gil’ferding, A. F. 1873. Onezhskie byliny [Bylinas from the region of Lake Onega]. Saint Petersburg. Gimson, A. C. 1980. An Introduction to the Pronunciation of English. 3rd ed. London. Gmelin, J. G. 1752. Reise durch Sibirien von dem Jahre 1738 bis zu Ende 1740. Dritter Theil. Göttingen. Goldstein, K. S. 1964. A Guide for Field Workers in Folklore. Hatboro, PA. Golubeva, G. & S. Surazakov, ed. & tr. 1965. Altyn-Bize. Geroicheskie skazanie. Skazitel’nitsa: E. K. Tashtamysheva. [Altïn-Bize. A heroic narrative. Woman singer: E. K. Tashtamysheva]. Barnaul. Gombrich, E. H. 2002. Art and Illusion: A Study in the Psychology of Pictorial Representation. 6th ed. London. Goodman, N. 1976. Languages of Art: An Approach to a Theory of Symbols. 2nd ed. Indianapolis, IN. Graf, F. 1991. ‘Gestures and Conventions: The Gestures of Roman Actors and Orators.’ In Bremmer & Roodenburg 1991: 36–58. Grame, T. C. & Gen’ichi Tsuge. 1972. ‘Steed Symbolism on Eurasian String Instruments.’ The Musical Quarterly 58: 57–66. Graziosi, B. 2002. Inventing Homer: The Early Reception of Epic. Cambridge. 18–140. Greenberg, J. H. 1990. ‘A Survey of African Prosodic Systems.’ In On Language: Selected Writings of Joseph H. Greenberg. Ed. K. Denning & S. Kemmer. Stanford, CA. 553–78. Greene, G. 1970. ‘François Mauriac.’ In his Collected Essays. Harmondsworth. 91–96. Grodzins Gold, A. 1993. A Carnival of Parting: The Tales of King Bharthari and King Gopi Chand as Sung and Told by Madhu Natisar Nath of Ghatiyali, Rajasthan. Berkeley, CA. Grossardt, P. 2006. ‘Die Träume der Penelope im 19. und 20. Buch der Odyssee. Eine ethnographische Parallele aus Usbekistan.’ Wiener Studien 119: 23–37. Haiding, K. 1955. Von der Gebärdensprache der Märchenerzähler. FF Communications 155. Helsinki. Hainsworth, J. B. 1991. The Idea of Epic. Berkeley, CA. Hale, T. A. 1998. Griots and Griottes: Masters of Words and Music. Bloomington, IN. Hamayon, R. 1980. ‘Mongol Music.’ In The New Grove Dictionary of Music and Musicians. Ed. S. Sadie. 20 vols. London. XII, 482–85.

250 Bibliography Hamayon, R. N. 1993. ‘La voix de l’autre, animal ou mort, dans le chamanisme Sibérien.’ In Pour une anthropologie des voix. Ed. N. Revel & D. Rey-Hulman. Paris. 341–54. Hanks, W. F. 1987. ‘Discourse Genres in a Theory of Practice.’ American Ethnologist 14: 668–92. Hanks, W. F. 1996. Language and Communicative Practices. Boulder, CO. Hanslick, E. 1982. Vom Musikalisch-Schönen. Aufsätze. Musikkritiken. Ed. K. Mehner. Leipzig. [Or. 1854] Hanslick, E. 1986. On the Musically Beautiful: A Contribution Towards the Revision of the Aesthetics of Music. Tr. G. Payzant. Indianapolis, IN. Hanslick, E. 2018. Eduard Hanslick’s ‘On the Musically Beautiful’. A New Translation. Tr. L. Rothfarb & C. Landerer. New York. Harris, J. & K. Reichl, eds. 1997. Prosimetrum: Crosscultural Perspectives on Narrative in Verse and Prose. Cambridge. Harris, J. & K. Reichl. 2012. ‘Performance and Performers.’ In Reichl 2012: 141–202. Harris, R. 2016. ‘The Uyghur Muqam.’ In Levin, Daukeyeva & Köchümkulova 2016: 344–53. Harris, R. P. 2017. Storytelling in Siberia: The Olonkho Epic in a Changing World. Urbana, IL. Harva, U. 1938. Die religiösen Vorstellungen der altaischen Völker. FF Communications 125. Helsinki. Hatto, A. T. 1980. ‘Shamanism and Epic Poetry in Northern Asia.’ In his Essays on Medieval German and Other Poetry. Cambridge. 117–38. Hatto, A. T. 1989. ‘Towards an Anatomy of Heroic and Epic Poetry.’ In Traditions of Heroic and Epic Poetry. II. Characteristics and Techniques. Ed. A. T. Hatto & J. B. Hainsworth. London. 145–306. Haviland, J. B. 2004. ‘Gesture.’ In A Companion to Linguistic Anthropology. Ed. A. Duranti. Oxford. 197–221. Heaney, S., tr. 1999. Beowulf. London. Helffer, M. 1977. Les chants dans l’épopée tibétaine de Ge-sar d’après le Livre de la Course de Cheval. Version chantée de Blo-bzaṅ bstan-’jin. Centre de Recherches d’Histoire et de Philologie de la IVe Section de l’École pratique des Hautes Études. II. Hautes Études Orientales, 9. Geneva. Helffer, M. 1979. ‘Réflexions concernant le chant épique tibétain.’ Asian Music 10: 92–111. Helffer, M. 1982. ‘Les airs dans l’épopée de Ge-sar: Tradition orale et tradition écrite.’ In Fragen der mongolischen Heldendichtung. II. Ed. W. Heissig. Asiatische Forschungen 73. Wiesbaden. 231–44. Hendrickson, R. 2000. The Facts on File Dictionary of American Regionalisms. New York. Herman, D., M. Jahn & M.-L. Ryan, eds. 2005. Routledge Encyclopedia of Narrative Theory. London. Herzog, G. 1934. ‘Speech-Melody and Primitive Music.’ The Musical Quarterly 20: 452–66. Herzog, G. 1951. ‘The Music of Yugoslav Heroic Epic Folk Poetry.’ Journal of the International Folk Music Council 3: 62–64. Herzog, T. 2012. ‘Orality and the Tradition of Arabic Epic Storytelling.’ In Reichl 2012: 629–51.

Bibliography  251 Hoerburger, F. 1952. ‘Westöstliche Entsprechungen im Volksepos.’ Die Musikforschung 5: 354–61. Honko, L. 1996. ‘Epics Along the Silk Roads: Mental Texts, Performance, and Written Codification.’ OT 11: 1–17. Honko, L. 1998. Textualising the Siri Epic. FF Communications 264. Helsinki. Honko, L., ed. 2000a. Textualization of Oral Epics. Trends in Linguistics, Studies and Monographs 128. Berlin. Honko, L. 2000b. ‘Text as Process and Practice: The Textualization of Oral Epics.’ In Honko 2000a: 3–54. Hufford, M. 1995. ‘Context.’ JAF 108: 528–49. Hymes, D. 1964. ‘Introduction: Toward Ethnographies of Communication.’ American Anthropologist 66: 1–34. Hymes, D. 1972. ‘Models of the Interaction of Language and Social Life.’ In Directions in Sociolinguistics: The Ethnography of Communication. Ed. J. J. Gumperz & D. Hymes. New York. 35–71. Hymes, D. 1981. ‘In vain I tried to tell you’: Essays in Native American Ethnopoetics. Philadelphia. Hymes, D. 1994. ‘Ethnopoetics, Oral-Formulaic Theory, and Editing Texts.’ OT 9: 330–70. Hymes, D. 2003. Now I Know Only So Far: Essays in Ethnopoetics. Lincoln, NE. Idelsohn, A. Z. 1992. Jewish Music: Its Historical Development. New York. Illarionov, V. V. 1982. Iskusstvo yakutskikh olonkhosutov [The art of the Yakut olonkhosuts]. Yakutsk. Illarionov, V. V., P. E. Efremov, S. P. Oyunskaya & N. V. Emel’anov, eds. & trs. 1996. Yakutskiy geroicheskiy èpos ‘Moguchiy Èr Sogotokh’ [The Yakut heroic epic Är Soghotokh]. PFNSiDV. Novosibirsk. [With LP] Imāmov, K., T. Mirzaev, B. Sarimsāqov & A. Safarov. 1990. Ozbek khalq āghzaki poetik ijādi [Uzbek oral poetic folklore]. Tashkent. İmamverdiev, İ. 2013. ‘Türk Dünyasında Köroğlu Havaları’ [Köroğlu melodies in the Turkic world]. Türk Yurdu 209: 1–16. php?id=741 Innes, G., ed. & tr. 1974. Sunjata: Three Mandinka Versions. London. Ivančič Kutin, B. 2007. ‘The Roles of Participants in a Storytelling Event.’ Folklore 37: 35–41. Jacob, G., ed. & tr. 1904. Vorträge türkischer Meddâh’s (mimischer Erzählungskünstler). Berlin. Jahn, O. & A. Michaelis. 1873. Griechische Bilderchroniken. Bonn. Jain, J. ed. 1998a. Picture Showmen: Insights into the Narrative Tradition in Indian Art. Mumbai. Jain, J. 1998b. ‘The Painted Scrolls of the Garoda Picture Showmen of Gujarat.’ In Jain 1998a: 74–89. Jensen, M. S. 2011. Writing Homer: A Study Based on Results from Modern Fieldwork. Scientia Danica, Series H, Humanistica, 8, vol. 4. Copenhagen. Jensen, M. S. 2017. ‘The Challenge of Oral Epic to Scholarship.’ Humanities 6: 97. Johnson, J. W. 1980. ‘Yes, Virginia, There is an Epic in Africa.’ Research in African Literatures 11: 308–26. Johnson, J. W., ed. & tr. 1986. The Epic of Son-Jara: A West African Tradition. Text by Fa-Digi Sisòkò. Bloomington, IN.

252 Bibliography Johnson, J. W., T. A. Hale & S. Belcher, eds. 1997. Oral Epics from Africa: Vibrant Voices from a Vast Continent. Bloomington, IN. Jones, Sir W. 1799. ‘On the Musical Modes of the Hindus.’ Asiatic Researches 3: 55–87. Jowett, B. & T. Twining, trs. 1952. Aristotle’s Politics and Poetics. New York. Kaplan, M., M. Akalın & M. Bali, eds. 1973. Köroğlu Destanı [The dastan Köroğlu]. Anlatan [narrator]: Behçet Mahir. Ankara. Karmysheva, B. Kh. 1954. Uzbeki-lokaytsi Yuzhnogo Tadzhikistana [The UzbekLaqay of Southern Tajikistan]. Stalinabad. Karomatov, F. 1972a. Uzbekskaya instrumental’naya muzyka (Nasledie) [Uzbek instrumental music. The legacy]. Tashkent. Karomatov, F. 1972b. ‘On the Regional Styles of Uzbek Music.’ Tr. M. Slobin. Asian Music 4: 48–58. Karomatov, F. 1999. ‘O muzyke uzbekskikh dastanov i osobennostyakh ee pro­ yavleniya v “Alpamyshe”’ [On the music of Uzbek dastans and the characterists of its form in Alpamish]. In Mirzaev & Abdurakhimov 1999: 50–63. Karro, Y. & L. Sabaraikina, trs. 1994. Les guerriers célestes du pays yakoute-saxa. Paris. Karryev, B. A. 1968. Èpicheskie skazaniya o Kër-ogly u tyurko-jazychnykh narodov [Epic tales about Kör-oghlu among the Turkic-speaking peoples]. Moscow. Karryev, B. A., ed. 1983. Gjor-Ogly. Turkmenskiy geroicheskiy èpos [Göroghlï. A Turkmen heroic epic]. ÈNSSSR. Moscow. Kazagacheva, Z. S., ed. & tr. 1997. Altayskie geroicheskie skazaniya Ochi-Bala, Kan-Altyn [The Altaian hero tales Ochi-Bala, Qan-Altïn]. PFNSiDV 15. Novosibirsk. [With LP] Kendon, A. 1992. ‘Gesture.’ In Folklore, Cultural Performances, and Popular Entertainments: A Communication-centered Handbook. Ed. R. Bauman. New York. 179–90. Kendon, A. 1997. ‘Gesture.’ Annual Review of Anthropology 26: 109–28. Kendon, A. 2004. Gesture: Visible Action as Utterance. Cambridge. Ker, W. P. 1908. Epic and Romance: Essays on Medieval Literature. 2nd ed. London. Khakimjanova, M. 1990 ‘Murïn jïraw tuwralï esimde qalghandardan’ [Memories about Murïn-jïraw]. In Batïrlar jïrï. VI. [Heroic epics. VI]. Ed. E. Qudaybergenov et al. Almaty. Khojaev, N., ed. 1957. Shasenem-Garïp. Türkmen khalq dessanï [Shasenem and Garïp. A Turkmen folk dastan]. Ashkhabad. Khudoyakov, A. I. & R. V. Nazarenko, eds. & trs. 1998. Shorskie geroicheskie skazaniya. Kan Pergen, Altyn Syryk [Shor hero tales. Qan Pergen, Altïn Sïrïq]. PFNSiDV 17. Novosibirsk. Kilmer, A. D. & D. Collon. 1983. ‘Laute.’ Reallexikon der Assyriologie und vorderasiatischen Archäologie. Ed. D. O. Edzard. Berlin. VI, 512–17. Kishore, Sh. & Y. Desai. 2003. Of Bards and Beggars. Watertown, MA. [Film] Kleinmichel, S. 2000. Halpa in Choresm (Ḫwārazm) und ātin āyi im Ferganatal: Zur Geschichte des Lesens in Usbekistan im 20. Jahrhundert. 2 vols. Berlin. Knight, R. 2000–2001. ‘The Bana, Epic Fiddle of Central India.’ Asian Music 32: 101–40.

Bibliography  253 Köchümkulova, E. 2016. ‘The Kyrgyz Epic Manas.’ In Levin, Daukeyeva & Köchümkulova 2016: 52–68. Kondrat’ev, S. A. 1970. Muzyka mongol’skogo èposa i pesen [The music of Mongolian epic and songs]. Moscow. Kondrat’eva, М. N. 2002. ‘Muzyka altayskikh skazok’ [The music of Altaian folktales]. In Sadalova 2002: 42–69. Kononenko, N. 1998. Ukrainian Minstrels. New York. Korogly, Kh. G. 1983. Vzaimosvyazi èposa narodov Sredney Azii, Irana i Azerbaydzhana [Interrelations of the epic of the peoples of Central Asia, Iran and Azerbaijan]. Moscow. Korom, F. J. 2013. ‘The Anthropology of Performance: An Introduction’. In The Anthropology of Performance: A Reader. Ed. F. J. Korom. Chichester. 1–7. Krivonosov, V. 1939. ‘O napevykh kirgizskogo èposa “Manas”’ [On the melodies of the Kyrgyz epic Manas]. Sovetskaya muzyka 1939.5: 31–35. Kúnos, I., ed. 1899. Mundarten der Osmanen. Vol. VIII of W. Radloff, ed., Proben der Volkslitteratur der türkischen Stämme. Saint Petersburg. Kupriyanova, Z. N., ed. 1965. Èpicheskie pesni nentsev [Epic songs of the Nenets]. Moscow. Kurbanova, Dzh. 2000. ‘The Singing Traditions of Turkmen Epic Poetry.’ In Reichl 2000b: 115–28. Kydyrbaeva, R. Z. 1996. Èpos ‘Manas’. Genezis, poètika, skazitel’stvo [The epic of Manas. Genesis, poetics, art of the narrators]. 2nd ed. Bishkek. Kyrgys, Z. K. 1997. ‘O tuvinskikh èpicheskikh napevakh’ [On Tuvan epic melodies]. In Orus-ool et al. 1997: 39–52. Kyrgys, Z. K. 2002. Tuvinskoe gorlovoe penie: Ètnomuzykovedcheskoe issledovanie [Tuvan throat singing: An ethnomusicological study]. Novosibirsk. Lane, E. W. 1860. An Account of the Manners and Customs of Modern Egyptians. 5th ed. London. Langer, S. K. 1976. Philosophy in a New Key: A Study in the Symbolism of Reason, Rite, and Art. 3rd ed. Cambridge, MA. Laude-Cirtautas, I. 1983. ‘Zu den Einleitungsformeln in den Märchen und Epen der Mongolen und der Türkvölker Zentralasiens.’ Central Asiatic Journal 27: 211–248. Laver, J. D. M. 1968. ‘Voice Quality and Indexical Information.’ British Journal of Disorders of Communication 3: 43–54. Lermontov, M. Yu. 1962. Ashik-Kerib. Turetskaya skazka [Ashik-Kerib. A Turkish tale]. In Sobranie sochineniy v chetyrëkh tomakh [Collected works in four volumes]. Moscow. IV, 264–74. Levin, T. 1996. The Hundred Thousand Fools of God: Musical Travels in Central Asia (and Queens, New York). Bloomington, IN. [With CD] Levin, T., with V. Süzükei. 2006. Where Rivers and Mountains Sing: Sound, Music and Nomadism in Tuva and Beyond. Bloomington, IN. [With CD/DVD] Levin, T., S. Daukeyeva & E. Köchümkulova, eds. 2016. The Music of Central Asia. Bloomington, IN. Lewis, C. S. 1942. A Preface to Paradise Lost. London. Light, N. 2008. Intimate Heritage: Creating Uyghur Muqam Song in Xinjiang. Halle Studies in the Anthropology of Eurasia 19. Berlin. List, G. 1963. ‘The Boundaries of Speech and Song.’ Ethnomusicology 7: 1–16.

254 Bibliography Lomax, A. 1968. Folk Song Style and Culture. American Association for the Advancement of Science, Publ. 88. Washington, DC. Lord, A. B. 1960. The Singer of Tales. Cambridge, MA [2nd ed., with CD, 2000]. Lord, A. B. 1987. ‘Homer, Parry, and Huso.’ In Parry 1987: 465–78. Lord, A. B. 1993. ‘Cædmon Revisited.’ In Heroic Poetry in the Anglo-Saxon Period: Studies in Honor of Jess B. Bessinger, Jr. Ed. H. Damico & J. Leyerle. Kalamazoo, MI. 121–37. Ludowyk, E. F. C. 1959. ‘The Chalk Circle: A Legend in Four Cultures.’ Comparative Literature 1: 249–56. Lyons, M. C. 1995. The Arabian Epic: Heroic and Oral Story-Telling. University of Cambridge Oriental Publications 49. 3 vols. Cambridge. Magoun, F. P., Jr. 1955. ‘Bede’s Story of Cædman: The Case History of an AngloSaxon Oral Singer.’ Speculum 30: 49–63. Mair, G. H., ed. 1909. Wilson’s Arte of Rhetorique 1560. Oxford. Mair, V. H. 1988. Painting and Performance: Picture Recitation and Its Indian Genesis. Honolulu, HI. Malik, A. 2005. Nectar Gaze and Poison Breath: An Analysis and Translation of the Rajasthani Oral Narrative of Devenarayan. Oxford. Malinowski, B. 1926. Myth in Primitive Society. New York. Mamedov [Mәmmәdov], T. A. 1984. Pesni Këroglu [The songs of Koroğlu]. Baku. Mamedov [Mәmmәdov], T. A. 2011. Azәrbaycan aşıq yaradıcılığı [The works of the Azerbajanian aşıq]. Baku. Maqsetov, Q. M., ed. 1981. Shar’yar. Qaraqalpaq khalïq dästanï [Sharyar. A Karakalpak folk dastan]. Qaraqalpaq fol’klorï 13. Nukus. Maqsetov, Q. M. & A. K. Karimov, eds. 1985. Ghärip-Ashïq. Sayatkhan – Hämra [Ashïq Ghärip. Sayatkhan and Hämra]. Qaraqalpaq Fol’klorï 14. Nukus. Marazzi, U., tr. 1986. Māday Qara: An Altay Epic Poem. Naples. Marcus, S. 2002. ‘The Eastern Arab System of Melodic Modes in Theory and Practice: A Case Study of Maqām Bayyātī.’ In Danielson et al. 2002: 33–44. Marsh, P. K. 2009. The Horse-Head Fiddle and the Cosmopolitan Reimagination of Tradition in Mongolia. New York. Martin, R. P. 2009. ‘Epic as Genre.’ In A Companion to Ancient Epic. Ed. J. M. Foley. Chichester. 9–19. Marzolph, U. & R. van Leeuwen. 2004. The Arabian Nights Encyclopedia. 2 vols. Santa Barbara, CA. Merriam, A. P. 1964. The Anthropology of Music. Evanston, IL. Milner, A. 1963. ‘The Late Renaissance.’ In The Pelican History of Music. 2. Renaissance and Baroque. Ed. A. Robertson & D. Stevens. Harmondsworth. 111–98. Mirzaev, T. 1968. Alpāmish dāstānining ozbek variantlari [The Uzbek variants of the epic Alpāmish]. Tashkent. Mirzaev, T. 1979. Khalq bakhshilarining epik repertuari [The epic repertoire of the folk singers]. Tashkent. Mirzaev, T. 2008. Èpos i skazitel’ [Epic and narrator]. Tashkent. Mirzaev, T. & M. Abdurakhimov, eds. & trs. 1999. Alpamysh. Uzbekskiy narodny geroicheskiy èpos [Alpāmish. An Uzbek heroic folk-epic]. Tashkent. Mirzaev [Mirza], T. & Karāmatov [Karāmatli], F. M., eds. 1999. Alpamïs. Dāstāndan parchalar [Alpamïs. Extracts from the dastan]. Music transcriptions by S. Gabrielyan. Tashkent.

Bibliography  255 Mirzaev, T. & B. Sarimsāqov. 1981. ‘Dāstān, uning turlari wa tarikhiy taraqqiyāti’ [Subgenres and historical development of the dāstān]. In Ozbek fol’klorining epik janrlari [The epic genres of Uzbek folklore]. Tashkent. 9–61. Mirzaoğlu, F. G. 2003. Çukurova Bozlağı [The bozlak of Çukurova]. Ankara. Moldobaev, I. 2001. ‘Kyrgyzskiy shaman (bakshy) v 20-e gody XX veka’ [The Kyrgyz shaman (bakhshï) in the twenties of the 20th century]. In Sredneaziatskiy ètnograficheskiy sbornik [Central Asian ethnographic collection] 4: 232–37. Morley, T. 1597. A Plaine and Easie Inroduction to Practicall Musicke. London. Moyle, N. K. 1990. The Turkish Minstrel Tale Tradition. New York. Murko, M. 1919. ‘Neues über südslawische Volksepik.’ Neue Jahrbücher für das klassische Altertum, Geschichte und deutsche Literatur 22: 273–96. Myers, H., ed. 1992. Ethnomusicology: An Introduction. New York. Nagy, G. 1996. Poetry as Performance: Homer and Beyond. Cambridge. Nazarenko, R.V. 1998. ‘O sredstvakh muzykal’noy vyrazitel’nosti v shorskom geroicheskom èpose Kan Pergen’ [On the means of musical expressiveness in the Shor heroic epic Qan Pergen]. In Khudoyakov & Nazarenko 1998: 34–45. Nettl, B. 2005. The Study of Ethnomusicology: Thirty-one Issues and Concepts. New ed. Urbana, IL. Neubecker, A. J. 1994. Altgriechische Musik. Eine Einführung. 2nd ed. Darmstadt. Niemi, J. 1998. The Nenets Songs: A Structural Analysis of Text and Melody. Acta Universitatis Tamperensis 591. Tampere. Niemi, J. 1998–1999. ‘The Genres of the Nenets Songs.’ Asian Music 30: 77–132. Niles, D. 2011. ‘Metric Melodies and the Performance of Sung Tales in the Hagen Area.’ In Rumsey & Niles 2011: 275–301. Niles, J. D. 1993. ‘Understanding Beowulf: Oral Poetry Acts.’ JAF 106: 131–55. Niles, J. D. 1999. Homo Narrans: The Poetics and Anthropology of Oral Literature. Philadelphia. Nims, M. F., tr. 1967. Poetria Nova of Geoffrey of Vinsauf. Toronto. Nurmukhamedov, M. K. 1983. Skazki A. S. Pushkina i fol’klor narodov Sredney Azii [A. S. Pushkin’s tales and the folklore of the peoples of Central Asia]. Tashkent. Nutku, Ö. 1997. Meddahlık ve Meddah Hikâyeleri [The art of the meddah and meddah stories]. Ankara. Nyshanov, N. 2016. ‘Kyrgyz Küü.’ In Levin, Daukeyeva & Köchümkulova 2016: 265–81. Okpewho, I. 1992. African Oral Literature: Backgrounds, Character, and Continuity. Bloomington, IN. Ong, W. J. 1982. Orality and Literacy: The Technologizing of the Word. London. Orus-ool, S. M. et al., eds. & trs. 1997. Tuvinskie geroicheskie skazaniya. Khunan-Kara, Boktug-Kirish, Bora-Shèley [Tuvan hero tales. Khunan-Kara, Boktug-Kirish, Bora-Shèley]. PFNSiDV 12. Novosibirsk. [With LP] Osnes, B. 2010. The Shadow Puppet Theatre of Malaysia: A Study of ‘Wayang Kulit’ with Performance Scripts and Puppet Designs. Jefferson, NC. Oyunsky, P. A. 2014. Olonkho: Nurgun Botur the Swift. Tr. A. Nakhodkina et al. Folkstone. Özarslan, M. 2006. Ferhat ile Şirin. Mukayeseli Bir Araştırma [Ferhat and Şirin. A comparative study]. Istanbul. Öztuna, Y. 1990. Büyük Türk Mûsikîsi Ansiklopedisi [The great Turkish music encyclopedia]. 2 vols. Ankara.

256 Bibliography Page, M. E. 1979. ‘Professional Storytelling in Iran: Transmission and Practice.’ Iranian Studies 12: 195–215. Parry, M. 1987. The Making of Homeric Verse: The Collected Papers of Milman Parry. Ed. A. Parry. Oxford. Pegg, C. 1992. ‘Mongolian Conceptualizations of Overtone Singing (xöömii).’ British Journal of Ethnomusicology 1: 31–54. Pegg, C. 2001. Mongolian Music, Dance, & Oral Narrative: Performing Diverse Identities. Seattle, WA. [With CD] Pegg, C. & J. Porter. 2001. ‘Epics.’ In Sadie 2001: VIII, 265–69. Pekarsky, È. K., G. U. Èrgis, A. A. Popov & I. V. Pukov, eds. & trs. 1985. Stroptivyi Kulun Kullustuur. Yakutskoe olonkho [Obstinate Kulun Kullustuur. A Yakut olonkho]. Skazitel’ [singer]: I. T. Timofeev-Teploukhov. ÈNSSSR. Moscow. Pelias, R. J. & J. VanOosting. 1987. ‘A Paradigm for Performance Studies.’ Quarterly Journal of Speech 73: 219–31. Penkala-Gawęcka, D. 2014. ‘The Way of the Shaman and the Revival of Spiritual Healing in Post-Soviet Kazakhstan and Kyrgyzstan.’ Shaman 22: 35–61. Pepper, H., ed. 1972. Un mvet de Zwè Nguéma. Chant épique fang. Réédité par Paul et Paule de Wolf. Classiques africains 9. Paris. [With LP] Petrain, D. 2014. Homer in Stone: The ‘Tabulae Illiacae’ in their Roman Context. Cambridge. Petrosyan, A. A., A. I. Alieva, V. M. Gatsak & U. B. Dalgat, eds. 1969. Skazaniya o Nartakh —Èpos narodov Kavkaza [Tales about the Narts – The epic of the peoples of the Caucasus]. Moscow. Peyko, N. & I. Shteynman 1940. ‘O muzyke yakutov’ [On the music of the Yakuts]. Sovetskaya Muzyka 1940.2: 84–91. Philippi, D. L. 1979. Songs of Gods, Songs of Humans: The Epic Tradition of the Ainu. Tokyo. Phillips, N. 1981. Sijobang: Sung Narrative Poetry of West Sumatra. CSOLC 1. Cambridge. Powers, H. S. et al. 2001. ‘Mode.’ In Sadie 2001: XVI, 775–860. Preston, D. R. 1982. ‘’Ritin’ Fowklower Daun ’Rong: Folklorists’ Failures in Phonology.’ JAF 95: 304–26. Prior, D., ed. & tr. 2006. The ‘Semetey’ of Kenje Kara: A Kirghiz Epic Performance on Phonograph. Turcologica 59. Wiesbaden. Propp, V. Y. 1955. Russkiy geroicheskiy èpos [The Russian heroic epic]. Leningrad. Propp, V. Y. 1976. ‘Printsipy klassifikatsii fol’klornykh zhanrov’ [Principles of the classification of folklore genres]. In his Fol’klor i deystvitel’nost’. Izbrannye stat’i [Folklore and reality. Selected articles]. Moscow. 34–45. Pugh-Kitingan, J. 2011. ‘An Ethnomusicological Discussion of Bì Té, the Chanted Tales of the Huli.’ In Rumsey & Niles 2011: 109–49. Pukhov, I. V. 1951. ‘Ispolnenie olonkho’ [The performance of the olonkho]. In Doklady na vtoroy nauchnoy sessii. I. Istoriya i filologiya [Papers from the second scientific session. I. History and philology]. Yakutsk. 133–64. Pukhov, I. V. 1962. Yakutskiy geroicheskiy èpos olonkho. Osnovnye obrazy [The Yakut heroic epic olonkho. Basic forms]. Moscow. Pukhov, I. V. 1975. ‘Iz poètiki olonkho (Monologi olonkho)’ [From the poetics of the olonkho: The monologues of the olonkho]. Sovetskaya Tyurkologiya 1975.5: 39–43.

Bibliography  257 Putilov, B. N. 1988. Geroicheskiy èpos i deystvitel’nost’ [Heroic epic and reality]. Leningrad. Putilov, B. N. 1997. Èpicheskoe skazitel’stvo. Tipologiya i ètnicheskaya spetsifika [Epic narration. Typology and ethnic specificity]. Moscow. Putilov, B. N., A. A. Gorelov et al., eds. 2001. Byliny Pechory. Svod Russkogo Fol’klora. Tom 1. [Bylinas from the region of the Pechora river. A collection of Russian Folklore. Volume 1]. Saint Petersburg. [With CD] Qarïpqulov, A. et al. 1995. Manas Entsiklopediya [Manas encyclopedia]. 2 vols. Bishkek. Race, W. H., ed. & tr. 2012. Pindar. Olympian Odes, Pythian Odes. Rev. ed. Loeb Classical Library 56. Cambridge, MA. Radin, P. 1972. The Trickster: A Study in American Indian Mythology. New York. Radloff, W. 1893. Aus Sibirien. Lose Blätter aus meinem Tagebuche. 2nd ed. 2 vols. Leipzig. Radloff, W., ed. & tr. 1885. Proben der Volkslitteratur der nördlichen türkischen Stämme. V. Der Dialect der Kara-Kirgisen. Saint Petersburg. [Text and translation volume] Radloff, W. 1990. ‘Samples of Folk Literature from the North Turkic Tribes, Collected and Translated by Dr. Wilhelm Radloff. Preface to Volume V: The Dialect of the Kara-Kirgiz.’ Tr. G. Böttcher Sherman & A. B. Davis. OT 5: 73–90. Rakhaev, A. I. 1994. ‘O muzyke nartskogo èposa Balkarii i Karachaya’ [On the music of the Nart epic of Balkaria and Karachay]. In Narty: Geroicheskiy èpos Balkartsev i Karachaevtsev [The Narts: The heroic epic of the Balkars and the Karachays]. Ed. & tr. A. I. Alieva, R. A.-K. Ortobaeva, T. M. Khadzhieva & A. Z. Kholaev. ENE. Moscow. 605–14. Ready, J. L. 2014. ‘ATU 974 The Homecoming Husband, the Returns of Odysseus, and the End of Odyssey 21.’ Arethusa 47: 265–85. Ready, J. L. 2015. ‘The Textualization of Homeric Epic by Means of Dictation.’ Transactions of the American Philosophical Association 145: 1–75. Reichl, K., tr. 1985a. Rawšan. Ein usbekisches mündliches Epos. Asiatische Forschungen 93. Wiesbaden. Reichl, K. 1985b. ‘Oral Tradition and Performance of the Uzbek and Karakalpak Epic Singers.’ In Fragen der mongolischen Heldendichtung. III. Ed. W. Heissig. Asiatische Forschungen 91. Wiesbaden: 613–43. Reichl, K. 1992. Turkic Oral Epic Poetry: Traditions, Forms, Poetic Structure. New York. [Rpt. 2018] Reichl, K. 2000a. Singing the Past: Turkic and Medieval Heroic Poetry. Ithaca, NY. Reichl, K., ed. 2000b. The Oral Epic: Performance and Music. Intercultural Music Studies 12. Berlin. Reichl, K. 2000c. ‘Silencing the Voice of the Singer: Problems and Strategies in the Editing of Turkic Oral Epics.’ In Honko 2000a: 103–27. Reichl, K., ed. & tr. 2001. Das usbekische Heldenepos Alpomish. Einführung, Text, Übersetzung. Turcologica 48. Wiesbaden. Reichl, K. 2002. ‘Genealogy and Heroic Conduct: Turkic Oral Epics and the Meaning of Heroic Poetry.’ In The Kalevala and the World’s Traditional Epics. Ed. L. Honko. Studia Fennica Folkloristica 12. Helsinki. 245–66.

258 Bibliography Reichl, K., ed. & tr. 2007. Edige: A Karakalpak Oral Epic as Performed by Jumabay Bazarov. FF Communications 293. Helsinki. [With CD] Reichl, K., ed. 2012. Medieval Oral Literature. De Gruyter Lexikon. Berlin. [Rpt. 2016] Reichl, K. 2013. ‘From Performance to Text: A Medievalist’s Perspective on the Textualization of Turkic Oral Poetry.’ Western Folklore 72: 252–71. Reichl, K. 2014. ‘“Sing, O Muse!”: Reflections on the Singing of Oral Epics.’ In Song and Emergent Poetics. Ed. P. Huttu-Hiltunen, K. L. Frog, & E. Stepanova. Kuhmo. 45–63. Reichl, K., ed. & tr. 2014–2015. Manas in the Version of Jüsüp Mamay. Xinjiang ‘Manas’ Research Centre Publications 4. 2 vols. Beijing. Reichl, K. 2015a. ‘Hero and Leander: Medieval and Folkloristic Variations on a Classical Theme.’ In Dall’Antico al Moderno. Immagini del Classico nelle Litterature Europee, ed. P. Boitani & E. di Rocco. Storia et Litteratura 293. Rome. 87–119. Reichl, K. 2015b. ‘Memory and Textuality in the Orality-Literacy Continuum.’ In Orality and Textuality in the Iranian World: Patterns of Interaction across the Centuries. Ed. J. Rubanovich. Leiden. 29–42. Reichl, K. 2016a. ‘Oral Epics into the Twenty-First Century: The Case of the Kyrgyz Epic Manas.’ JAF 129: 327–44. Reichl, K. 2016b. ‘The Singing of Tales: The Role of Music in the Performance of Oral Epics in Turkey and Central Asia.’ Classics & Issue 14. Washington, DC: Center for Hellenic Studies. display/6702 Reichl, K. 2020. ‘“The True Nature of the aoidos”: The Kirghiz Singer of Tales and the Epic of Manas.’ In John Miles Foley’s World of Oralities: Text, Tradition, and Contemporary Oral Theory. Ed. M. C. Amodio. York. 185–206. Reinhard, K. & U. 1984. Musik der Türkei. 2 vols. Wilhelmshaven. Reinhard, U. & T. de Oliveira Pinto. 1989. Sänger und Poeten mit der Laute. Türkische Âşık und Ozan. Berlin. [With Cassettes] Renoir, A. 1988. A Key to Old Poems: The Oral-Formulaic Approach to the Interpretation of West-Germanic Verse. University Park, PA. Reshetnikova, A. P. 1993. ‘Muzyka yakutskikh olonkho’ [The music of the Yakut olonkho]. In Yakutskiy geroicheskiy èpos ‘Kyys Dèbiliyè’ [The Yakut heroic epic Kïïs Däbiliyä]. PFNSiDV. Novosibirsk. 26–69. [With LP] Reshetnikova, A. P. 2005. Fond syuzhetnykh motivov i muzyka olonkho v ètnograficheskom kontekste [Plot motifs and music of the olonkho in an ethnographic context]. Yakutsk. Reynolds, D. F. 1995. Heroic Poets, Poetic Heroes: The Ethnography of Performance in Arabic Oral Epic Tradition. Ithaca, NY. Reynolds, D. F. 2002. ‘Learning Epic Traditions.’ In Danielson et al. 2002: 339–46. Ricœur, P. 1973. ‘The Model of the Text: Meaningful Action Considered as a Text.’ New Literary History 5: 91–117. Ricœur, P. 1983–1985. Temps et récit. 3 vols. Paris. Robinson, J. 2001. ‘Goodman.’ In The Routledge Companion to Aesthetics. Ed. B. Gaut & D. M. Lopes. 2nd ed. London. 185–97. Roerich, G. N. 1942. ‘The Epic of King Kesar of Ling.’ Journal of the Royal Asiatic Society of Bengal: Letters 8: 277–311.

Bibliography  259 Rölleke, H. 1983. ‘Die Brüder Grimm in Spinnstuben, dämmrigen Küchenwinkeln und an Kohlenmeilern? Oder: Über die Fragwürdigkeit tertiärer Überlieferungen.’ Aurora, Jahrbuch der Eichendorff-Gesellschaft 43: 160–68. Rouget, G. 1990. La musique et la transe. Esquisse d’une théorie générale des relations de la musique et de la possession. Nouvelle édition revue et augmentée. Paris. Rouget, G., with J. Schwarz. 1970. ‘Transcrire ou Décrire? Chant Soudanais et Chant Fuégien.’ In Échanges et communications. Mélanges offerts à Claude Lévi-Strauss. Studies in General Anthropology 6. The Hague. 677–704. Routley, E. 1959. The English Carol. London. Rozimbāev, S. R., ed. 2009. Āshiqnāma [The book of the lover]. Vol. 4. Urganch. Rubanovich, J. 2012. ‘Orality in Medieval Persian Literature.’ In Reichl 2012: 653–79. Ruckert, G. & R. Widdess. 2000. ‘Hindustani Raga.’ In South Asia: The Indian Subcontinent. Ed. A. Arnold. The Garland Encyclopedia of World Music 5. New York. 64–88. [With CD] Rumsey, A. & D. Niles, eds. 2011. Sung Tales from the Papua New Guinea Highlands: Studies in Form, Meaning, and Sociocultural Context. Canberra. Sachs, C. 1913. Reallexikon der Musikinstrumente zugleich ein Polyglossar für das gesamte Instrumentengebiet. Berlin. Sachs, C. 1962. The Wellsprings of Music. Ed. J. Kunst. The Hague. Sadalova, T. M., ed. & tr. 2002. Altayskie narodnye skazki [Altaian folktales]. PFNSiDV 21. Novosibirsk. Sadie, S., ed. 2001. The New Grove Dictionary of Music and Musicians. 2nd ed. 29 vols. London. Online edition: Sarti, S. 2003. ‘La kithara greca nei documenti archeologici.’ Revue belge de philologie et d’histoire 81: 47–68. Saussure, F. de. 1959. Course in General Linguistics. Ed. C. Bally & A. Sechehaye, in collaboration with A. Reidlinger; tr. W. Baskin. London. Saville-Troike, M. 2002. The Ethnography of Communication: An Introduction. 3rd ed. Oxford. Saville-Troike, M. 2004. ‘Anthropological Linguistics and the Ethnography of Speaking.’ In Sociolinguistics/Soziolinguistik: An International Handbook of the Science of Language and Society/Ein internationales Handbuch zur Wissenschaft von Sprache und Gesellschaft. Ed. U. Ammon et al. 3 vols. 2nd ed. Berlin. I, 109–20. Schechner, R. 2002. Performance Studies: An Introduction. 2nd ed. New York. Scheub, H. 1977. ‘Body and Image in Oral Narrative Performance.’ New Literary History 8: 345–67. Schiefner, A., ed. 1856. M. Alexander Castrén’s Reiseberichte und Briefe aus den Jahren 1845–1849. Saint Petersburg. Schimmel, A. 1992. A Two-Colored Brocade: The Imagery of Persian Poetry. Chapel Hill, NC. Schmaus, A. 1958. ‘La byline russe et son état actuel.’ La Table Ronde 132: 114–27. Schmitz, A. Die Erzählung von Edige. Gehalt, Genese und Wirkung einer hero­ ischen Tradition. Turcologica 27. Wiesbaden. Seelmann, W., ed. 1884. Valentin und Namelos. Niederdeutsche Denkmäler 4. Norden.

260 Bibliography Segre, C., ed. 1989. La Chanson de Roland. Tr. M. Thyssens. 2 vols. Textes Littéraires Français. Geneva. Seitel, P. 2003. ‘Theorizing Genres – Interpreting Works.’ New Literary History 34: 275–97. Seydou, C., ed. & tr. 1972. Silâmaka & Poullôri. Récit épique peul. Classiques africains 13. Paris. [With LP] Seydou, C. ed. & tr. 1976. La Geste de Ham-Bodêdio ou Hama le Rouge. Classiques africains 18. Paris. [With LP] Seydou, C. 1977. ‘La devise dans la culture peule: évocation et invocation de la personne.’ In Calme-Griaule 1977a: 187–264. Seydou, C. 1982. ‘Comment définir le genre épique? Un exemple: l’épopée afri­ caine.’ Journal of the Anthropological Society of Oxford 13: 84–98. Seydou, C. 1983. ‘The African Epic: A Means for Defining the Genre.’ Folklore Forum 16: 47–68. [Tr. of Seydou 1982.] Seydou, C. 2000. ‘Word and Music: The Epic Genre of the Fulɓe of Massina (Mali).’ In Reichl 2000b: 211–24. Shafrannikov, V. 1959. Karakalpakskie narodnye pesni [Karakalpak folk songs]. Moscow. Sharp, C. J., ed. 1932. English Folk Songs from the Southern Applachians. Ed. M. Karpeles. London. Shevtsov, V. N. 1997. ‘Muzyka alyptykh nymakha’ [The music of the hero tales]. In Khakasskiy geroicheskiy èpos ‘Ay-Khuuchin’ [The Khakass heroic epic Ay-Khuuchin]. Ed. & tr. V. E. Maynogasheva et al. PFNSiDV 16. Novosibirsk. 48–60. [With LP] Sheykin, Yu. I. & V. S. Nikiforova. 1997. ‘Altayskoe èpicheskoe intonirovanie’ [Altaian epic intoning]. In Kazagacheva 1997: 47–70. Shul’gin, B. 1973. ‘Ob altayskom kae’ [On the Altaian qay]. In Surazakov 1973: 454–60. Signell, K. 2002. ‘Contemporary Turkish Makam Practice.’ In Danielson et al. 2002: 47–58. Singh, K. 1998. ‘To Show, to See, to Tell, to Know: Patuas, Bhopas, and their Audiences.’ In Jain 1998a: 100–15. Skvortsova, N. M. 2001. ‘O muzyke traditsionnykh nenetskikh pesen’ [On the music of traditional Nenets songs]. In Fol’klor Nentsev v zapisyakh 1911, 1937, 1946, 1953, 1965–1987 godov [The folklore of the Nenets in records from the years 1911, 1937, 1946, 1953, 1965–1987]. Ed. & tr. E. T. Pushkareva, L. V. Khomich et al. PFNSiDV 23. Novosibirsk. 50–86. [With CD] Slyomovics, S. 1987. The Merchant of Art: An Egyptian Hilali Oral Epic Poet in Performance. Berkeley, CA. Smith, J. D. 1979. ‘Metre and Text in Western India.’ BSOAS 42: 347–57. Smith, J. D., ed. & tr. 1991. The Epic of Pābūjī: A Study, Transcription and Translation. Cambridge. Snouck Hurgronje, C. 1906. The Achehnese. Tr. A. W. S. O’Sullivan. 2 vols. Leiden. Somfai Kara, D. & S. Ämirlan. 2020. ‘Traditional Spirituality among the Karakalpak of Central Asia.’ Shaman 28: 97–110. Squire, M. 2011. The ‘Iliad’ in a Nutshell: Visualizing Epic on the ‘Tabulae Iliacae’. Oxford. Stein, R.-A. 1959. Recherches sur l’épopée et le barde au Tibet. Bibliothèque de l’Institut des Hautes Études Chinoises 13. Paris.

Bibliography  261 Stevens, J. 1986. Words and Music in the Middle Ages: Song, Narrative, Dance and Drama, 1050–1350. Cambridge. Stone, R. M. 1988. Dried Millet Breaking: Time, Words, and Song in the Wọi Epic of the Kpelle. Bloomington, IN. Stoyanov, A. K. 1988, ‘Iskusstvo khakasskikh khaydzhi’ [The art of the Khakas khayjïs]. In Altyn-Aryg. Khakasskiy geroicheskiy èpos [Altïn-Arïg. A Khakas heroic epic]. Ed. & tr. V. E. Maynogasheva. ÈNSSSR. Moscow. 577–90. Stumpf, C. 1911. Die Anfänge der Musik. Leipzig. Sultanova, R. 2011. From Shamanism to Sufism: Women, Islam and Culture in Central Asia. London. Sultanova, R. 2016. ‘Female Musicians in Uzbekistan: Otin-Oy, Dutarchi, and Maqomchi.’ In Levin, Daukeyeva & Köchümkulova 2016: 421–34. Sumits, W. & T. Levin. 2016. ‘Maqom Traditions of the Tajiks and Uzbeks.’ In Levin, Daukeyeva & Köchümkulova 2016: 320–43. Sundberg, J. 1987. The Science of the Singing Voice. DeKalb, IL. Suppan, W., with H. Helgason. 1979. ‘Epos.’ In Die Musik in Geschichte und Gegenwart. 16. Supplement: Earsden – Zweibrücken. Basel. Cols. 101–14. Surazakov, S. S., ed. & tr. 1973. Maaday-Kara. Altayskiy geroicheskiy èpos [Maaday-Qara. An Altaian heroic epic]. ÈNSSSR. Moscow. Tedlock, D. 1983. The Spoken Word and the Work of Interpretation. Philadelphia. Thangavelu, K. 1998. ‘Itinerant Images: Embodiments of Art and Narrative in Telangana.’ In Jain 1998a: 90–99. Thompson, S. 1975. Motif-Index of Folk-Literature. Rev. and enlarged ed. 6 vols. Bloomington, IN. Tolins, J. & J. E. Fox Tree. 2014. ‘Addressee Backchannels Steer Narrative Development.’ Journal of Pragmatics 70: 152–64. Tolley, C. 2019. ‘Hard it is to Stir My Tongue’: Raiding the Otherworld for the Elixir of Poetry. FF Communications 317. Helsinki. Trager, G. L. 1964. ‘Paralanguage: A First Approximation.’ In Language in Culture and Society: A Reader in Linguistics and Anthropology. Ed. D. Hymes. New York. 274–88. Tskhurbaeva, K. G. 1969. ‘O napevakh osetinskikh nartovskikh skazaniy’ [On the melodies of the Ossetian Nart legends]. In Petrosyan et al. 1969: 469–83. Türkmen, F. 1995. Âşık Garip Hikâyesi. İnceleme – metin [The hikâye of Aşık Garip. Study and text]. 2nd ed. Ankara. Türkmen, F., M. M. Taşlıova & N. Tan, eds. 2008. Âşık Şeref Taşlıova’dan Derlenen Halk Hikâyeleri [The folk narratives collected from Âşık Şeref Taşlıova]. Ankara. Turnbull, H. 1972. ‘The Origin of the Long-Necked Lute.’ The Galpin Society Journal 25: 58–66. Tylor, E. B. 1873. Primitive Culture: Researches into the Development of Mythology, Philosophy, Religion, Language, Art and Custom. 2nd ed. 2 vols. London. Uther, H.-J. 2011. The Types of International Folktales: A Classification and Bibliography, Based on the System of Antti Arne and Stith Thompson. FFC Communications 284, 285, 286. 3 vols. Helsinki. Vambéry, H., ed. & tr. 1911. Jusuf und Ahmed. Ein özbegisches Volksepos im Chiwarer Dialekte. Supplement of the journal Keleti Szemle 11. Budapest.

262 Bibliography van der Heide, N. 2008. Spirited Performance: The Manas Epic and Society in Kyrgyzstan. Ph.D. Diss. University of Tilburg. Amsterdam. van Tongeren, M. C. 2004. Overtone Singing: Physics and Metaphysics of Harmonics in East and West. 2nd ed. Amsterdam. [With CD] Varlamova, G. I. & A. N. Mireeva, eds & trs. 2008. Tipy geroicheskikh skazaniy Èvenkov [Types of heroic tales of the Evenki]. Novosibirsk. Vinogradov, V. S. 1958. Kirgizskaya narodnaya muzyka [Kyrgyz folk music]. Frunze. Vinogradov, V. S. 1984. ‘Napevy “Manasa”’ [The melodies of Manas]. In Manas. Kirgizskiy geroicheskiy èpos [Manas. A Kyrgyz heroic epic]. Ed. & tr. A. S. Mirbadaleva, B. M. Yunusaliev et al. 4 vols. ÈNSSSR. Moscow [1984–1995]. I, 492–509. Vitebsky, P. 1995. The Shaman: Voyages of the Soul, Trance, Ecstasy and Healing, from Siberia to the Amazon. London. Vitz, E. B., N. F. Regalado & M. Lawrence, eds. 2005. Performing Medieval Narrative. Cambridge. Wade, B. C. 1980. ‘Some Principles of Indian Classical Music.’ In Musics of Many Cultures: An Introduction. Ed. E. May. Berkeley, CA. 83–110. Wadley, S. S. 1989. ‘Choosing a Path: Performance Strategies in a North Indian Epic.’ In Blackburn et al. 1989: 75–101. Wadley, S. S. 2004. Raja Nal and the Goddess: The North Indian Epic ‘Dhola’ in Performance. Bloomington, IN. Waley, A., tr. 1951. ‘Kutune Shirka, the Ainu Epic.’ Botteghe Oscure 7: 214–36. Walsh, G. B. 1984. The Varieties of Enchantment: Early Greek Views of the Nature and Function of Poetry. Chapel Hill, NC. Walsh, R. 1836–1838. Constantinople and the Scenery of the Seven Churches of Asia Minor. Illustrated, in a series of drawings from nature by Thomas Allom. 2 vols. London. Walter, M. N. & E. J. N. Fridman, eds. 2004. Shamanism: An Encyclopedia of World Beliefs, Practices, and Culture. Santa Barbara, CA. Walther, Ph., ed. & tr. 1999. Aucassin et Nicolette. Chantefable du XIIIe siècle. Paris. Wängler, H.-H. 1958. ‘Singen und Sprechen in einer Tonsprache (Hausa).’ Zeitschrift für Phonetk und allgemeine Sprachwissenschaft 11: 23–34. Wellek, R. & A. Warren. 1963. Theory of Literature. 3rd ed. Harmondsworth. West, M. L. 1981. ‘The Singing of Homer and the Modes of Early Greek Music.’ The Journal of Hellenic Studies 101: 113–29. West, M. L. 1992. Ancient Greek Music. Oxford. West, S. 2012. ‘Some Reflections on Alpamysh.’ In Homeric Contexts: Neoanalysis and the Interpretation of Oral Poetry. Ed. F. Montanari, A. Rengakos & C. C. Tsagalis. Berlin. 531–41. Wickett, E. 2010. The Epic of ‘Pabuji ki par’ in Performance. World Oral Literature Project, Occasional Paper 3. Cambridge. Wieringa, E. 1998. ‘The Dream of the King and the Holy War against the Dutch: The kôteubah of the Acehnese Epic, Hikayat Prang Gômpeuni.’ BSOAS 61: 298–308. Winn, J. A. 1981. Unsuspected Eloquence: A History of the Relations between Poetry and Music. New Haven, CT.

Bibliography  263 Wiora, W., ed. 1966. European Folk Song: Common Forms in Characteristic Modifications. Cologne. Wundt, W. 1916. Elements of Folk Psychology: Outlines of a Psychological History of the Development of Mankind. Tr. E. L. Schaub. London. Wundt, W. 1973. The Language of Gestures. Tr. J. S. Thayer, C. M. Greenleaf, M. D. Silberman. Approaches to Semiotics 6. The Hague. Wünsch, W. 1934. Die Geigentechnik der südslawischen Guslaren. Veröffentlichungen des Musikwissenschaftlichen Instituts der Deutschen Universität in Prag 5. Brünn. Youssefzadeh, A. 2002. Les bardes du Khorassan iranien. Le ‘bakhshi’ et son répertoire. Louvain. [With CD] Zaerr, L. M. 2012. Performance and the Middle English Romance. Cambridge. Zarif, H., ed. 1965. Mamatkarim, Jizzakh qozghālāni, Āchilāw [Mamatkarim, The insurrection of Jizzakh, Āchilāw]. Tashkent. Zarif, H., ed. 1971. Ergash Jumanbulbul Oghli. Tarjimai khāl. Rawshan. Qunduz bilan Yulduz [Autobiography. Rawshan. Qunduz and Yulduz]. 2 vols. Tashkent. Zarif, H. 1978. ‘“Ozbek khalq ijādi” kop tomligiga qiratigan asarlarni tayārlash printsiplari’ [Principles for the edition of works in the multi-volume series ‘Uzbek folk poetry’]. In Islām shāir va uning khalq poeziyasida tutgan orni [Islām shāir and his place in folk poetry]. Ed. T. Mirzaev. Tashkent. 128–42. Zataevich, A. V. 1971. Kirigizskie instrumental’nye p’esy i napevy [Kyrgyz instrumental pieces and melodies]. Moscow. Zemtsovsky, I. I., ed. 1989. Muzyka èposa. Stat’i i materialy [The music of the epic. Papers and materials]. Yoshkar-Ola. Zenker, J. Th. 1866. Türkisch-arabisch-persisches Handwörterbuch. Leipzig. Żerańska-Kominek, S. 1998. ‘The Concept of Journey (ýol) in Turkmen Musical Tradition.’ Ethnomusicology 42: 265–82. Żerańska-Kominek, S., with A. Lebeuf. 1997. The Tale of Crazy Harman: The Musician and the Concept of Music in the Türkmen Epic, ‘Harman Däli’. Warsaw. Zhirmunsky, V. M. 1960. Skazanie ob Alpamyshe i bogatyrskaya skazka [The tale of Alpamysh and the hero tale]. Moscow. Zhirmunsky, V. M. 1967. ‘The Epic of ‘Alpamysh’ and the Return of Odysseus.’ In Proceedings of the British Academy 52 (1966): 267–86. Zhirmunsky, V. M. 1979. ‘Legenda o prizvanii pevtsa’ [The legend about the singer’s calling]. In his Sravnitel’noe literaturvedenie. Vostok i zapad [Comparative literature. East and West]. Leningrad. 397–407. Zhirmunsky, V. M. & Kh. T. Zarifov. 1947. Uzbekskiy narodny geroicheskiy èpos [The Uzbek heroic folk epic]. Moscow. Zumthor, P. 1987. La lettre et la voix. De la ‘littérature’ médiévale. Paris. Zumthor, P. 1990. Oral Poetry: An Introduction. Tr. K. Murphy-Judy. Theory and History of Literature 70. Minneapolis, MN.


Page numbers followed by ‘n’ refer to endnotes act sequence 55, 64n11 ağıt 146, 158 Ali Paşa 146 c Alī Shīr Navā’ī see Navā’ī Alpāmish 19, 40, 47n28, 137–42, 160, 164–65, 172, 200, 210, 220–1 Alpamïs 133, 160–3 aoidós 107 aqïn 24–25, 75, 130 Aqquw 166 Aristotle 11, 155 arman 173 Är Soghotokh 184–85, 188, 197 ashïq/âşık 25–26, 75, 209 Âşık Garip/Ghärip Ashïq 143, 209–15, 221 Aucassin and Nicolette 3, 209 audience: appeals to audience 182, 193–97, 199; audience reactions 45, 49–50, 52–55, 58, 61–63, 105, 115–16, 206, 219, 222–23 authorship 37, 43–45 Awazkhān 137 Bağdat and Hafız 194, 196 Bakchiev, Talantaaly (Kyrgyz singer) 23, 26, 97 bakhshi: Karakalpak baqsï 25; Kazakh shaman 75–76; Uzbek 24–25 Bakhtin, M. 11, 183–84, 198, 200, 222 Bāla-­bakhshi Qurbānnazar Abdullaev (Uzbek singer) 137, 143 ballad 12, 73–74, 215–16 bänkelsänger see mountebanks

Bartók, Béla 110, 146 Bazarov, Jumabay (Karakalpak singer) 35, 70–76, 88–90, 117, 126, 133–35, 160, 180–182, 196–97, 223 Belyaev, V. 168–69, 173 Benjamin, W. 202, 218, 225 Beowulf 130, 181, 224, 225 bhopo 62, 105, 125, 132, 187, 205 bì té 15, 112–13 Blake, William 203, 208 Bowra, M. C. 11, 12, 19, 20, 80 bozlak 146 bylina 1, 20, 26, 34, 39, 74, 114, 115, 210, 216, 230 Byron, George Gordon, Lord 203, 208, 215 Cædmon 23 Camara, Seyidou (Mande singer) 30 chanson de geste 3, 122, 182, 193 Chanson de Roland 3, 197, 224 chant 111, 157, 208, 239 chantefable 209 chapbooks 41, 103, 221 Chāri-­shāir see Khojamberdi-­oghli, Chāri Cicero, Marcus Tullius 85, 86 Coleridge, Samuel Taylor 179 communicative event 51, 57–59, 63, 202, 208, 217–18, 222, 225 composition in performance 12–13, 143, 224 contest song 12, 24, 159, 173 creativity 39–43

Index  265 dastan 18–20 dastanchi 24 Davies, S. 153–54, 172 de-­contextualization 202, 203, 218–26 Devnārāyaṇ 205 Ḍholā 18, 67, 68, 144 djali 7, 46n14 Dobrynya and Alyosha 210 Dôkarim (Aceh singer) 43 dombira 2, 137–43, 241 Döölötqan (Kyrgyz singer) 91–96, 123 dumy 230 dutar 25, 56, 57, 131, 136, 143, 145, 147, 175, 211, 215, 241 Edige 37–39, 70–74, 89, 125–26, 133–35, 146, 160, 182, 183, 222–25, 236 Eliade, M. 75, 187–88 epic: as genre term 11, 20–21; mythological epics 184–88; in native classification 14–17; see also bì té; bylina; chanson de geste; dastan; dumy; hikâye; hoddu; hunters’ song; junačke pjesme; kang rom; kënge kreshnikësh; mvet; nïmakh; nimngakaan; olonkho; p’ansori; romance; runo; sijobang; sīra; syudbabts; tul’; yarabts epic and drama 101–2; see also randai; terukkūttu; wayang kulit; wayang wong ethnography of communication 50–51, 218 Farhād and Shīrīn 40–41, 44, 221 Fāzil-­shāir see Yoldāsh-­oghli, Fāzil formulas 12, 16, 17, 21n5, 29, 40, 61, 69, 72, 114, 156, 157, 164, 174, 182, 185, 187, 193–95, 197, 205, 206, 217, 224 framing 49, 57, 58 Geertz, C. 219 Genette, G. 11, 180, 181, 188, 200n1 Genjebay-­baqsï see Tilewmuratov, Genjebay Geoffrey of Vinsauf 67, 85 Gerusalemme liberata 166 Gesar 18, 100–1, 114, 156–57, 170, 206, 233

gesture 57, 67, 68, 72, 85–87, 105, 193, 196, 220, 221, 225; autonomous 87; conventional 87–90; extrinsic 90; gesture and emphathy 193; gesture and inspiration 97–101; iconic 89, 95–97, 105; miming 102–3; stylized 87, 88, 90–97 Gibbons, Orlando 110 Goodman, N. 153, 172 Goroghli/Köroğlu 35, 55, 56, 58, 64n13, 143, 156–59, 176, 194–96 griot 16, 17, 30, 36, 37, 175 Gulum-­bakhshï see Ilmetop, Gulum-­bakhshï guslar 29, 37, 132, 135 gusle 29, 132, 136, 146 Hanslick, E. 152–53 hava 157, 158; uzun hava 124, 146, 158, 161, 215 Heaney, Seamus 181 Helffer, M. 156–57 Hermannus Contractus 154 Hero and Leander 196 Hesiod 23 heterophonic 131, 135, 146, 239 Hikajat Potjut Muhamat 194–95 Hikayat Pram Gômpeuni 43 hikâye 20, 41, 53, 146, 194, 196, 197, 221 hoddu 16, 17, 20 homophonic 131, 239 hunters’ bard 30, 46n14 hunters’ song 20, 126, 209 ideophone 165 Iliad 21, 48, 207 Ilmetop, Gulum-­bakhshï (Turkmen singer) 51, 55, 57, 144 initiation dreams 23–24, 76, 98 instrumental interludes 56, 131, 134, 135, 141, 143, 144, 147, 166, 174, 215 ïrchï 25 Irkismändya 170–71 Jakobson, R. 45 James, Henry 181 Janequin, Clément 110 Jangghar 232 Japsïlïq-­jïraw see Sïrïmbetov, Jaqsïlïq jeldirme 117, 122, 167

266 Index jïraw 25, 38, 70, 75; see also Bazarov, Jumabay; Sïrïmbetov, Jaqsïlïq Jizzakh qozghālāni 42–43, 45 jomoqchu 25 Jones, Sir William 154–55 jorgho söz 117 Jumabay-­jïraw see Bazarov, Jumabay Jumanbulbul-­oghli, Ergash (Uzbek singer) 7n1, 35, 137 junačke pjesme 20, 26 Kalevala 230 Kalkin, Aleksey G. (Altaian singer) 80–2 Kambili 30 kang rom 113 Karataev, V. O. (Yakut singer) 185, 187, 188, 197 Katamaraju-­katha 205 Keats, John 48, 216 Kendon, A. 87, 88 kënge kreshnikësh 20 Kerem and Aslı 41, 158 khalfa 27 Kharman Däli 55–56, 144 Khojamberdi-­oghli, Chāri (Uzbek singer) 2–3, 24, 26, 137–43 khöömey 77–78, 82 Khusrow and Shīrīn 18, 40–41 khuurchi 135 Kida, Mamadou (Bambara singer) 199 Kır İsmail (Turkish singer) 147 Koroğlu’s Journey to Derbent 159 Kudaman 229 Kuruubay Khaannaakh Kulun Kullustuur 186 Kutune Shirka 189 küy/kuy 157, 166 Laylī and Majnūn 18 leitmotif 167–75, 240 Lermontov, Nikolai 210 Lewis, C. S. 5 Lord, A. B. 12–13, 21n5, 26, 29, 34, 37, 44, 47n32, 132, 146, 224, 230 lyric genres see ağıt; arman; bozlak; contest song; qïz uzatuu; qoshoq; qoşma; sea-­shanty;tolghaw; yār-­yār Maaday Qara 80–81 mâbo 174 Mahābhārata 102, 204 Mahir, Behçet (Turkish singer) 196 makam 157, 158

Malinowski, B. 14, 15, 54 Mamay, Jüsüp (Kyrgyz singer) 44, 45, 91, 117–22, 188, 234 Manas 23–25, 29, 44, 45, 77, 90–97, 102, 116–24, 164–65, 173, 188, 200, 234–36 manaschï 23–26, 33, 62, 87, 90, 97, 98, 100, 102, 116, 117, 123, 124, 128, 167, 169 maqām 152, 154, 157, 172, 175, 176 Marenzio, Luca 151, 166 meddah 28, 102–4 Međedović, Avdo (South Slavic singer) 20, 146 melody: contour 109, 121, 125; native terms see hava; küy/kuy; makam nama; rta; see also chant; heterophonic; homophonic; jeldirme; jorgho söz; leitmotif; modes; monotonic; overtone-­ singing; recitative metre 11, 44, 69, 120–21, 124–27, 141, 145, 167, 174, 214 miming see epic and drama; meddah Mirzaev, T. 2, 220 modes 152, 172, 177n7; Church modes 154, 157, 239; Greek modes 154, 155; see also maqām; rāga Moldoke-­qïzï, Seydene (Kyrgyz singer) 26 monotonic 109, 111–14, 122, 128, 141, 208, 240 Monteverdi, Claudio 151, 166 morin khuur 76, 114, 117, 132, 133, 135, 156, 242 Morley, Thomas 151, 152 motto 174–75 mountebanks 206 Mozart, Wolfang Amadeus 208 Müdamî, Âşık Sabit (Turkish singer) 221 Murādov, Rozimbek (Uzbek singer) 210 Murïn-­jïraw (Kazakh singer) 38 Murko, M. 26, 28, 29, 34, 132 music: convention 154, 163–64, 172, 174–76, 216; expression 151–55, 172; see also Davies; Goodman; Hanslick; imitation 109, 151, 163–66, 171, 172; küy; leitmotif; onomatopeia; interplay of song and instrument 143–47; musical terms 239–40; instrumental interludes; melody; modes; musical instruments

Index  267 musical instruments: bowed chordophones 132–33; non-­ Western instruments 240–42; plucked chordophones 136–37; see also dombira; dutar; gusle; morin khuur; qobïz mvet 17, 18, 20, 62, 69, 101, 118, 143, 197–98, 228 Mwindo 116, 199–200, 228 nama 157, 160 naqqāl 13 ŋara 30 narrative: first-­person narrative 188–93; frame-­tale 179, 188; histoire 182, 188; narration 180, 188, 199; point of view 180–84, 190; récit 180, 188, 199; sacred time 186–88; time 182–84; third-­person narrative 179–81, 188, 190 narrator: authorial narrator 181; extradiegetic narrator 180–81, 191; figural narrator 181; narrators’ comments 193–200, 221 Nart legends 115–16, 132, 143 Nathan, Isaac 203, 208, 215 Navā’ī , cAlī Shīr 18, 40, 41, 44 nïmakh 79 nïmakhchï 79 nimngakaan 170–71 Niẓāmī 18, 40, 41 Nurali 35, 138 Nyurgun Bootur 220 Ochï-­Bala 81 Odyssey 21, 107–8, 130, 188 olonkho 26, 98–99, 167–71, 184–87, 220 olonkhosut 90, 98–100, 167–72 Ömür-­uulu, Choyuke (Kyrgyz singer) 23 onomatopoeia 76, 116, 164–66, 172 orality: forms of orality 12–14; inscribed orality 224, 225; oral-­ derived 182, 224, 225; orality-­ literacy continuum 13, 27–28, 34, 210; see also meddah; qissakhan; reader of tales; tumār; oral poetics 50, 57, 225 Ormanbet-­biy 133 Orozbaqov, Saghïmbay (Kyrgyz singer) 188

overtone-­singing see khöömey see qay Oyunsky, Platon A. (Yakut poet) 220 Pābūjī 58–59, 62, 104–5, 125, 132, 187, 204–6, 208 Panāh-­oghli, Saidmurād (Uzbek singer) 19 p’ansori 229 paṛ 104–5, 205–6 Parbū Bhopo (Rajasthani singer) 105, 125 Parry, Milman 14, 26, 34, 37, 132, 224, 230 Pāyānov, Abdunazar (Uzbek singer) 40 Palnāḍu 229 perde see register changes performance: concept 48–50; context 218–20; modes (West Africa) 127–8 pevač 132 Phillips, N. 13, 99–101, 114–15, 193–94, 229 picture showmen 204–6 Pindar 154–56 Plato 11 Propp, V. 22n11, 216 prosimetrum 18, 22n15, 27, 61, 69, 88, 115, 127, 137, 144, 156, 209 Pukhov, I. V. 98–99, 167, 171 Pushkin, Alexander 183 Qaralaev, Sayaqbay (Kyrgyz singer) 63, 96, 117, 124, 188–89 qay 78–82 qaychï 79, 80, 82 Qayratdinov, Qïyas-­jïraw (Karakalpak singer) 37–39, 160–63 qissakhān 27–28, 104 qïz uzatuu 173 qobïz 25, 56, 75, 76, 117, 133, 242 Qoblan 73, 133, 160 qoshoq 173 qoşma 159, 213 The Quest for a Wife 229 Qurbanov, Turghanbay (Karakalpak singer) 212 Quintilian 85–86 Qurmanbek 91 Radloff, W. 2, 33, 79, 91, 117, 122–24, 233 rāga 154–55, 157, 177n7 Rāmāyaṇa 102, 204 randai 101, 104

268 Index rāwī 27 Rawshan 2–5, 18, 19, 60, 137, 197 reader of tales 13, 27–28, 103–4; see also meddah; qissakhan recitative 67, 74, 76, 78, 81, 111, 112, 116, 117, 122, 124, 142, 159, 160, 165, 166, 169, 210, 240 register changes 144, 145, 159, 175, 215 rhapsōdós 107 rhyme-­chains 164, 165 Ricœur, P. 183, 220 romance 3, 13, 19–20, 22n17, 26–28, 35, 40, 41, 53, 55, 104, 143, 146, 147, 158, 182, 183, 193, 195, 209–14, 216, 221; see also dastan; hikâye; sīra Rouget, G. 76–77 rta 117, 156 runo 20, 114, 230–31 Rustamkhān 137 Sachs, C. 77, 110–12, 131, 137, 208 Saghïmbay see Orozbaqov; Saghïmbay saltimbanchi see mountebanks sannāwchi 24 Sayaqbay see Qaralaev; Sayaqbay sāzanda 24 Schechner, R. 48–49 scop 130 sea-­shanty 225–26 Semetey 25, 123, 124 semeteychi 25, 26 setting 51–55 Seydou, C. 16–18, 174–75, 228 Seyteḳ 25 sgrun-­mkhan 100, 156 Shāh-­nāma 13, 104, 159 shāir 24, 25, 27, 32, 45 shamanism 20, 23, 24, 56, 57, 75–82, 97, 100–1, 112, 117, 133, 167, 171, 187, 189–93; see also initiation dreams Sharyar 133, 180–3, 193 sign languages 68, 86–88 sijobang 99–100, 104, 114, 193, 229 Silamaka Fara Dikko 7, 199 Silâmaka and Poulôri 174

Singer-­narrators: names and types 24–28; singers’ schools 34–36; training 28–33; see also aoidós; aqïn; ashïq/âşık; bakhshi; bhopo; dastanchi; djali; griot; guslar; hunters’ bard; ïrchï; jïraw; jomoqchu; khalfa; khuurchi; mâbo; manaschï; meddah; mountebank; naqqāl; ŋara; nïmakhchï; olonkhosut; pevač; picture showmen; olonkhosut; qaychï; qissakhan; rāwī; reader of tales; rhapsōdós; sannāwchi; ̣sāzanda; scop; semeteychi; sgrun-­mkhan; shāir; tukang sijobang; tul’chi; women singers; Yangzhou narrators singing 67, 69, 73–75; choral singing 115–16; solo singing 15, 113–24; speech and song 108–9, 111, 169–70; see also chant; overtone-­ singing; recitative Singspiel 210 sīra 20, 27, 42, 229 Sīrat cAntar 27 Sīrat Banī Hilāl 27, 30, 42, 131 Sīrat Baybars 27 Siri 229 Sïrïmbetov, Jaqsïlïq (Karakalpak singer) 160–63 Smith, J. 104, 105, 125, 128, 206, 208, 229 Snouck Hurgronje, C. 43 speaking see voice Stein, R.-A. 100–1, 233 stichic 109, 114, 121, 133, 142, 146, 171, 240 stile concitato 166, 240 strophic 109, 121, 125, 141, 142, 240 Sunjata 16, 36, 37, 126 Suso, Bamba (Mande singer) 36–37 syudbabts 190 Tabulae Iliacae 207 Taleev, Sofron Andreevich (Nenets singer) 191 Taşlıova, Âşık Şeref (Turkish singer) 158, 194, 196, 221 terukkūttu 102

Index  269 text: mental text 217; text vs. context 215–18, 220–26; see also de-­ contextualization; textualization textualization 59–63, 64n20, 172, 197, 208, 221, 226 The Thousand and One Nights 19, 179, 183 Tilewmuratov, Genjebay (Karakalpak singer) 211–13, 215 Tinguidji, Boûbacar (Fulani singer) 174 tolghaw 133, 160, 177n17 Tomskaya, Dariya (Yakut singer) 26 Tri Yabtoné 190–92 Trofimov, Aleksey (Yakut singer) 98–99 Trofimov, N. G. (Evenki singer) 170 tukang sijobang 99, 100, 114 tul’ 31 tul’chi 31, 32 tumār 13 Tylor, E. B. 86 Ugljanin, Salih (South Slavic singer) 146 UNESCO, Intangible Cultural Heritage 102, 103, 133, 220, 226

Valentin und Namelos 195 visualization 105, 203–8 voice 67–84; speaking 68–73; voice quality 72–75 Wagner, Richard 167 Waley, A. 189, 229 wayang kulit 102 wayang wong 102 The Wedding of Smailagić Aga 20, 44 Wilson, Thomas 85, 86 Wọi 144, 198–99 women singers 26–27, 83n16, 212 Wundt, W. 86, 89 Yangzhou narrators 102 yarabts 190–2 yār-­yār 172, 173 Yoldāsh-­oghli, Fāzil (Uzbek singer) 19, 35, 40–43, 45, 47n28, 137, 164–65, 200, 220, 221 Zarifov, H. 28, 32, 33, 53, 59 Zhirmunsky, V. M. 28, 32, 33, 40, 53, 233 Zong Belegt Baatar 232 Zwè Nguéma 62, 197–98