singing the classical,

voicing the modern

$ Singing the Classical,

Voicing the Modern

the postcolonial politics

of music in south india

Amanda J. Weidman

Duke University Press Durham and London 2006

© 2006 duke university press
All rights reserved. Printed in the United
States of America on acid-free paper 
Designed by Amy Ruth Buchanan
Typeset in Bembo by Tseng Information
Systems, Inc. Library of Congress
Cataloging-in-Publication Data appear
on the last printed page of this book.

Duke University Press gratefully

acknowledges the support of Bryn
Mawr College, which provided funds
toward the production of this book.

Portions of chapter 3 were previously

published as ‘‘Gender and the politics of
voice: Colonial modernity and classical
music in South India,’’ in Cultural Anthro-
pology 18, no. 2 (2003): 194–232. © 2003
American Anthropological Association.
$ Portions of chapter 4 were previously

published as ‘‘Can the subaltern sing? Mu-

sic, language, and the politics of voice in
twentieth-century South India,’’ in Indian
Economic and Social History Review 42, no. 4
(2005): 485–511. $ Portions of chapter 6
were previously published as ‘‘Guru and
gramophone: Fantasies of fidelity and
modern technologies of the real,’’ in
Public Culture 15, no. 3 (2003): 453–76.
$ Contents

List of Illustrations vii

Acknowledgments ix

Note on Transliteration and Spelling xvii

Introduction 1

1. Gone Native?: Travels of

the Violin in South India 25

2. From the Palace to the Street:

Staging ‘‘Classical’’ Music 59

3. Gender and the Politics of Voice 111

4. Can the Subaltern Sing?: Music,

Language, and the Politics of Voice 150

5. A Writing Lesson: Musicology

and the Birth of the Composer 192

6. Fantastic Fidelities 245

Afterword: Modernity and the Voice 286

Notes 291

Works Cited 325

Index 343
$ Illustrations

1. Method of holding the fiddle and bow and tuning fork 39

2. Mechanical violin player 44
3. Violin advertisement 48
4. ‘‘Magic violin’’ advertisement 54
5. Mysore Yuvaraja’s Indian and European Musical Series cover page 67
6. ‘‘Music hall silence’’ advertisement 87
7. ‘‘Well done, sabha secretary!’’ cartoon 89
8. ‘‘The nagaswaram player gets flustered’’ cartoon 93
9. ‘‘The everyday excesses of the Bhavnagar Maharaja’’ cartoon 95
10. ‘‘Music vidwans on the world stage’’ cartoon 96
11. ‘‘Politics and Music’’ cartoon 96
12. Idol of Thyagaraja being carried to his house in Tiruvaiyaru 106
13. The excesses of Chief Minister Karunanidhi caricatured 108–9
14. The music season threatens to swamp even Thyagaraja himself 110
15. M. S. Subbalakshmi record jacket 125
16. Binny Silks advertisement 133
17. Sari advertisement 134
18. South India Music Emporium advertisement 137
19. ‘‘Musical angles’’ 162
20. Text explaining that Indian staff notation requires only three lines 204
21. Text explaining notation for ornaments 208
22. South Indian composition in staff notation 210
23. ‘‘Abbreviations and Embellishments’’ 214
24. Composition in raga bilahari in sargam and staff notation 215
25. Notation from Dikshitar’s Sangīta Sampradāya Pradārṣini 222
26. Composition in raga kedaram from Johannes’s Bhāratha Sangīta Svāya
Bōdhini 226
27. Composition in raga senjurutti from Pandithar’s A Practical Course in
South Indian Music 227
28. Title page of Pandithar’s A Practical Course in South Indian Music 228
29. Notation from Ganghadar’s Theory and Practice of Hindu Music 229
30. Page from Sambamoorthy’s Elements of Western Music for Students of
Indian Music 230
31. Composition from Sambamoorthy’s Elements of Western Music for
Students of Indian Music 231
32. Cover page of Mudaliar’s Gramaphon Sangeetha Keerthanamirdam 266
33. Advertisement for sruti boxes, electronic tamburas, and
talometers 283
34. ‘‘Concerto’’ model electronic tambura 284

viii acknowledgments
$ Acknowledgments

My first contact with Karnatic music came unexpectedly in 1990, when

I happened to hear a concert of Karnatic violin. I remember being fasci-
nated by seeing and hearing the violin—an instrument I had been play-
ing for years—as it was played diagonally, its scroll resting on the seated
player’s foot. The performer that evening was Adrian L’Armand, a West-
ern classical violinist who had spent time in Madras in the 1960s studying
Karnatic violin. The Wednesday afternoons I spent at his house in Swarth-
more, Pennsylvania, for the next two years, learning the beginnings of
Karnatic music, opened me to a world of music and musicality that I had
not known existed. Adrian is a supremely generous and gifted teacher.
Over the years, I have learned much from him about music, the violin,
improvisation, and teaching.
In India, I had the opportunity to study with four teachers who each
gave me something different. My main teacher, whose name I withhold
here for reasons of privacy, has a style of violin-playing unique in its ele-
gantly understated virtuosity. She generously took me on as her disciple in
1994 and on my subsequent stays in Madras, allowing me to accompany her
at her concerts and arranging performances for me. In the long afternoons
I spent at her house, eating, napping, talking, and playing, she taught me
to be a musician and, in her own way, forced me to become an anthropolo-
gist. The stamp of her character, both musical and personal, appears often
in these pages. In Madurai, N. S. Saminathan, violinist and nagaswaram
artist, taught me not only a repertoire of Tamil songs and secrets of im-
provisation, but also how to perform; then he arranged concerts for me
in Meenakshi Amman Temple. His humor, iconoclasm, and intellectual
interest in this project have helped me immensely. Also in Madurai, the
midrangist C. S. Palaninathan gave me, with supreme patience, the gift of
rhythm, teaching me how to ‘‘work out’’ rhythms, how to say them, and,
most important, how to remember them. V. S. Narasimhan, a wonder-
ful and original violinist, taught me with exquisite patience and accuracy
in Madras and in the United States. With him, I learned the pleasures of
dwelling on a single raga for hours.
My consultants for this project were many. For sharing their knowl-
edge with me in Madras, Madurai, Tanjavur, Mysore, and Trivandrum,
I thank: Anandi, nagaswaram artist; Subbu Arumugham, villupattu art-
ist; M. Balasubramaniam, violinist; Sethalapathy Balasubramaniam, vocal-
ist; Rajkumar Bharathy, vocalist; Theodore Bhaskaran; film historian;
Srimathi Brahmanandam, violinist; R. Chandrakala, nagaswaram artist;
V. Chandru, violist; P. T. Chelladurai, vocalist and principal of Tamil Icai
Sangam College of Music; Mr. Das of Musee Musical in Madras; R. N.
Doreswamy, veena artist; Dwaram Durgaprasad Rao, principal of Maha-
raja’s College of Music in Vizianagaram; Dwaram Mangathayaru, violin-
ist; Dwaram Shyamala, German teacher; L. R. Easwari, playback singer;
S. Ganapathy, violinist; M. S. Gopalakrishnan, violinist; T. V. Gopalakrish-
nan, vocalist and mridangist; Kadri Gopalnath, saxophonist; M. Gunaseka-
ran, clarinetist and band director; Randor Guy, film historian; S. Janaki,
playback singer; Salem Jeyalakshmi, vocalist; Illayaraja, film music direc-
tor; S. Kalpakam, violinist; A. Kanyakumari, violinist; N. Kesi, flutist;
K. P. Kittappa, dance master; B. L. Kothandaraman, harmonium artist;
T. N. Krishnan, violinist; Sampath Kumar Acharya, musicologist; Lalitha,
violinist; the Maharani of Trivandrum; G. S. Mani, vocalist; R. Mani-
mala, nagaswaram artist; T. Mukhta, vocalist; V. Murugan, instrument re-
pairer; members of the Mysore Police Karnatic Band; Sarada Nambi, his-
torian; Nandini, violinist; Dr. Omanakutty, vocalist; T. A. Thana Pandian,
musicologist; Kovalam Narayana Panikkar, Trivandrum intellectual and
artistic director; Sriram Parasuram, violinist; Ranganayaki Parthasarathy,
veena artist; T. S. Parthasarathy, music historian; D. K. Pattammal, vocalist;
M. Prameela, vocalist and music professor; Puratchidasan, author; H. M. V.
Ragunathan, recordist; A. A. Raja, film musician; R. P. Raja, historian;
N. S. Rajam, mridangist; S. Rajam, vocalist; Rajeswara Rao, film music di-
rector; R. Ramachandran, violinist; T. N. Ramachadran, book collector;
V. A. K. Ranga Rao, record collector and dance critic; T. Rukmini, violin-
ist; Saikumar, flutist; T. Sankaran, flutist; T. Sankaran, vocalist; K. Santana-
krishnan, record collector; Sarada, violinist; B. V. K. Sastry, musicologist;
Savitri Satyamurthy, violinist; V. R. Sekar, cellist; M. P. Sethuraman, naga-
swaram artist; Sarada Shaffter, pianist; V. S. Sharma, historian; Thirupam-

x acknowledgments
baram Shanmughasundaram, vocalist; Sirkali Sivachidambaram, vocalist;
S. Sivakumar, music editor for Dinamani; K. P. Sivanandam, veena artist;
K. Sivatamby, historian; S. Sowmya, vocalist; Anuradha Sriram, vocalist
and playback singer; M. S. Subbulakshmi, vocalist; V. V. Subrahmanyam,
violinist; Surendran, violinist and music teacher; P. Susheela, playback
singer; Dharmapuram Swaminathan, vocalist; Madurai V. Swaminathan,
vocalist; Kala Tawker, literary critic; Aswati Tirunal, vocalist; Kunnakudi
Vaidyanathan, violinist; L. Vaidyanathan, film music director; Valli, naga-
swaram artist; Lata Varma, musicologist; M. B. Vedavalli, musicologist and
historian; R. Vedavalli, vocalist; M. S. Viswanathan, film music director.
I gathered sources for this project at several libraries in Madras. I thank
the staff of the Roja Muthiah Library, V. Geetha of Maraimalai Adigal
Library, the staffs of U. V. Swaminathayyar Library and the Theosophi-
cal Society Library, and the Tamil Nadu State Archives. I am grateful to
the staff of Kalki magazine for allowing me to photocopy music-related
articles from old issues.
Many friends made my time in India enjoyable. The extended family of
N. S. Saminathan—Sunderraman, Vasantha, Meenakshi, Murugan, Parva-
thi, Sivakumar, and Murugeswari—took me in as one of their own during
my time in Madurai. I thank Padma and S. Anantharaman of Mahatma
Gandhi Nagar, Madurai, for their hospitality and friendship and for their
interest in my career as a Karnatic musician. Selvamani and family of
Krishnapuram Colony, Madurai, provided hilarious times and fun ad-
ventures in and around Madurai, as did the ‘‘Misses Madurai,’’ my fel-
low students in the American Institute of Indian Studies Tamil-language
program, Faye Blazar and Susan Schomburg. In Madras, Chellamma and
the late Sanmugham (alias Nilakantan) cooked me delicious meals, and
showed me how to have a sense of humor about things. The members of
the Madras String Quartet allowed me to play along with them and took
me as a guest to the film studios. S. Sanmugham engaged me in many a
long afternoon of talk on matters ranging from Tamil prosody to Derrida.
S. Venkat shared his tremendous knowledge of the Madras music scene
with me on many occasions.
The members of my doctoral committee at Columbia University—
John Pemberton, Nicholas Dirks, Aaron Fox, Valentine Daniel, and Indira
Peterson—all contributed helpful readings of this work when it was in
dissertation form. Although they often had conflicting visions of how the
project should progress, each of those visions has been beneficial. In par-
ticular, I thank John Pemberton for taking an interest at the very beginning

acknowledgments xi
and consistently pushing me to think originally about music, ethnography,
and the voice.
I am grateful to the language teachers I have had over the years: Vasu
Renganathan and Meera Narayanan at the University of Washington;
S. Bharathy, R. Devi, and K. Sangita at the American Institute of Indian
Studies Tamil program in Madurai; E. Somasundaram of the University of
Madras; V. Narayana Rao at the University of Wisconsin; and Gary Tubb
at Columbia University. I thank S. Ravindran and Perundevi Srinivasan
for assistance with Tamil translation.
I have presented portions of this work in many places. The readings,
comments, and suggestions I received on those occasions have helped me
transform this work from a dissertation into a book. I am grateful to
Bernard Bate for inviting me to participate in a conference and work-
shop on language and the historical imagination in South India, and to
the members of that group for helpful comments: A. R. Venkatachala-
pathy, Steve Hughes, Rama Mantena, and Lisa Mitchell. I thank Bonnie
McElhinny for inviting me to participate in a conference on language,
gender, and political economy at the University of Toronto, and Richard
Wolf for inviting me to present my work at a Radcliffe Institute work-
shop on musical culture in South Asia. Ann Anagnost, several anonymous
readers for Cultural Anthropology, Sanjay Subrahmanyam, Beth Povinelli,
Charles Briggs, Penny Eckert, Regula Qureshi, Steven Feld, Hajime Naka-
tani, Joshua Barker, Joel Kuipers, and Dan Neuman all gave me insightful
comments on presented and written versions of parts of this work. I thank
my colleagues in the study of Karnatic music—Matthew Allen, the late
T. Viswanathan, Zoe Sherinian, and Beth Bullard—for arranging and par-
ticipating in conference panels with me. My graduate-school comrades
Dave Novak, Uma Brughubanda, Dard Neuman, Arafaat Valiani, Karin
Zitzewitz, and Peter Cuasay gave me intellectual and social companion-
ship and comments on my work, and Jen Gherardi tracked down photo-
copies of music manuals in Madras and sent them to me.
At Duke University Press, my editor, Ken Wissoker, and assistant edi-
tor Courtney Berger have made publishing my first book a wonderful ex-
perience. Three anonymous readers for Duke University Press commented
thoroughly and insightfully on the various revised versions of this manu-
script, enabling me to make connections among my chapters and clarify
certain arguments that had eluded me for years.
Research funding and writing support for this project came from the
National Science Foundation; Foreign Language Area Studies grants;

xii acknowledgments
the American Institute of Indian Studies; Columbia University; and the
American Association of University Women.
Portions of chapters 3, 4, and 6 were published in Cultural Anthropology
18, no. 2; Indian Economic and Social History Review 42, no. 4; and Public Cul-
ture 15, no. 3. I thank the editors of those journals for making it possible
to publish these sections in their present form.

My family has been with me during every phase of this project, and I am
immensely grateful for their love, support, and involvement. My mother,
Bette, first taught me the pleasures of writing and reading, and how to
take an interest in the poetry of life. My father, Burton, taught me intel-
lectual curiosity, social awareness, and the patience to contemplate things
in the world. My sister, Nadine, whose scholarship has been an inspiration
for my own endeavors, really began to teach me when we were children,
for it was in Tigerland, the imaginary world we created together, that I
first learned—unwittingly—to be an anthropologist.
My husband, Ken Whang, has been with me through the last few years
of this project. His discerning eye and knack for asking the right questions
have helped me immensely. I am grateful for his love and companionship,
and for his willingness to take a ten-day trip to India with me in 2004 so
that we could share the people, places, and music that have been so im-
portant to me. The bright face of Sylvia Irene Whang, whose birth was a
happy interruption to the revisions of this manuscript, has inspired me in
the last phases, and it is to her that I dedicate this book.

acknowledgments xiii
$ Note on Transliteration and Spelling

My transliteration of Tamil words and phrases follows the Madras Univer-

sity Tamil Lexicon Scheme. However, many names and words are more
recognizable in Anglicized forms which do not adhere to these translit-
eration rules. For instance, the word for music is transliterated from Tamil
in the Madras University Tamil Lexicon Scheme as caṅkītam, but generally
spelled sangitam. I have used the conventional, recognizable spellings for
words from Tamil and other Indian languages that appear frequently, as
well as for names of individuals and places, caste names, deities, and in-
stitutions. Where an author writing in Tamil provides a transliteration of
his or her title, I have used the author’s transliteration.
The spelling of the word Karnatic is a matter of some debate. While I
have chosen this spelling throughout, other spellings that appear within
quotes may include Carnatic, Karnatak, Karnataka, and karāṭaka.
$ Introduction

The West is the kingdom of the instrument; the East is the kingdom of the Voice.
—Margaret Cousins, Music of the Orient and the Occident (1935)

That ‘‘East is East and West is West and never the twain shall meet’’ is perhaps most applicable
to music.
—T. V. Subba Rao, Studies in Indian Music (1962)

In the fall of 1994 I found myself in the dubious position of auditioning,

unsuspectingly, before a music teacher who was determined not to take
me as a student. As a student of South Indian classical music, I had sought
out this particular teacher after hearing about the extraordinary qualities
of her violin playing. I had announced my desire to study with her in a
letter, which preceded my arrival in Madras by a few weeks. Since she was
retired from her job as a staff artist at All India Radio, I assumed she would
probably have time to spend with me, and I looked forward to appren-
ticing myself to her in the kind of disciple-guru relationship that is con-
sidered crucial to the education of any student of Indian classical music. I
felt that once I had inserted myself into such a relationship, I would really
have ‘‘arrived.’’
On a September morning I was shown into the front room of her house,
greeted by her younger brother, with whom she lived and who asked me,
in English, numerous questions: who had my previous teachers been, what
part of America was I from, and what university did I attend. After about
ten minutes, he showed me upstairs, where she was waiting. ‘‘Play,’’ she
said, after a wordless greeting. I played something I had learned from a
previous teacher, and her face registered obvious disdain. When I had fin-
ished, she asked me to play it again and this time stopped me to correct
me in a few places. Then, she and her brother and a family friend who was
also present excused themselves and went downstairs. I could hear tearful
voices and hushed negotiations in Telugu, which I did not understand at
the time. Finally, after about twenty minutes, her brother returned and
said that she had accepted me and that I should come the next day at 9 am.
As I was leaving, he mentioned that their elder brother, also a violinist,
had unexpectedly passed away just a few days before my letter had arrived.
He had fallen ill and died while playing a concert with my teacher, and
she was now in a state of depression and had stopped playing her violin.
Their long-awaited tour of the United States, which had been scheduled
for that September and October, had had to be cancelled.
My teacher would subsequently relate the story of my arrival with great
melodrama. At the time I had first come to learn with her, she said, she
had been in the throes of a vow: she would never touch the violin again
in mourning for her brother. She was prepared to reject me from the first.
Relatives and concerned friends prevailed upon her to give me a chance.
Having a student from America, they reasoned, was a great boon: hand-
some payment, prestige, and possible chances to go to America loomed
in the future. Besides, the American tour she was supposed to make with
her brother had just been cancelled, and here was America coming to her.
Could she possibly think nothing of it? Affected by this line of reasoning,
she thought that after all maybe God was sending her a sign that she should
not abandon her violin. She decided she would give me a month, and if I
demonstrated no improvement, she would be convinced that I was just an
empty sign. According to her telling, on the twenty-ninth day, just as she
was losing interest in me, I suddenly showed improvement. She consented
to continue teaching me. On that same day, she decided that she would
reconcile with her main disciple and adopted son, whom she hadn’t seen
in ten years after he had married against her wishes. She thus converted
the prosaic story of beginning violin lessons into the kind of miraculous
story befitting a guru and her disciple.
During the months, and eventually years, I spent studying with her, the
exact nature of our relationship was under constant negotiation. She was
frustrated by the ways I failed to live up to the ideal of discipleship. I did
not live with her; I was in the United States for long periods of time; I
had studied with other teachers; my research prevented me from being a
single-minded and devoted student. I, meanwhile, wondered whether I
was learning enough music and how I could convert the embodied musi-
cal knowledge I was gaining into anthropological scholarship. Just what
kind of ethnographic practice was discipleship?

2 introduction
 Before the rest of the Madras music world, however, we performed the
roles of guru and sisya (disciple) more seamlessly. Perhaps we both sensed
the need to prove our legitimacy. One evening, as we were preparing to
leave for a visit, a friend of the family happened to be at my teacher’s
house. While my teacher was getting ready, I waited, impatiently, check-
ing my watch to see how late we would be. She called out to me to fetch
her ‘‘specs,’’ and as I got up to do so, the visitor exclaimed with a smile,
‘‘You are a typical sisya.’’ At the All India Radio station she made a point
of not handling her paycheck herself, telling the director to ‘‘hand it over
to my sisya.’’ When people asked who I was, she would embark on the
full story of how I had come and saved her from abandoning the violin,
and how she had taught me ‘‘from the beginning,’’ presenting my arrival
as a miracle and herself as some kind of miracle worker who was able to
make a Karnatic violinist out of recalcitrant Western material. Invariably
she would inflate the number of hours we spent together to some extraor-
dinary figure. When I questioned her in private about this and about why
she insisted on saying she had taught me from the beginning when, strictly
speaking, it wasn’t true, she said that the story was more impressive that
way and that it was an unnecessary bother to go into a whole explanation
of my real history.
In the spring of 1998 a reporter for a Madras newspaper interviewed us
for a feature story. When he asked how long she had been teaching me,
instead of answering the question, she said, ‘‘This is not teaching. This is
gurukulavasam’’ (a mode of teaching and learning in which the disciple
lives with the guru for years while the guru imparts musical knowledge).
We posed for a photo that was supposed to capture us in the act of guru-
kulavasam, sitting on a mat with our violins, each facing stiffly outward
toward the camera.
I realize now that at that moment I had arrived, if not at what I had
imagined the authentic role of the disciple to be, then at one of the cen-
tral oppositions through which Indian classical music, at the beginning of
the twenty-first century, has come to be defined. Gurukulavasam (kuru-
kulavācam) refers to a method in which the disciple lives with the guru,
learning music by a process of slow absorption and serving the guru as
a member of his or her household.1 At the turn of the twenty-first cen-
tury, the term has acquired a certain semantic density; not only does it
refer to a specific sense of fidelity, that of the disciple to the guru, but
the enactment of gurukulavasam signifies, at a broader level, a fidelity
to ‘‘tradition,’’ an adherence to the element that makes this music truly

introduction 3
Indian. Gurukulavasam is imagined as the mode through which the essen-
tially oral tradition of Indian classical music is passed on. This is in distinct
opposition to the modes of teaching Western music, which, as the com-
mon stereotype goes, ‘‘can be played just from looking at a written score.’’
Along with the contrast between orality and writing, a host of other oppo-
sitions are commonly drawn to differentiate Indian music from a gen-
eralized notion of Western music. Whereas Western music concentrates
on technical prowess, Indian music is more ‘‘spiritual’’; whereas Western
music has become secular, Indian music retains its ties to its devotional
origins; whereas Western music is primarily instrumental, Indian music is
primarily vocal.
Such pronouncements resonate strongly with colonial and Orientalist
discourses on India.2 Lining up ‘‘written,’’ ‘‘technical,’’ ‘‘secular,’’ and ‘‘in-
strumental’’ on one side and ‘‘oral,’’ ‘‘spiritual,’’ ‘‘devotional,’’ and ‘‘vocal’’ on
the other, they map ‘‘the West’’ and ‘‘India’’ as musical opposites, destined
never to meet, as T. V. Subba Rao so confidently stated.3 More precisely,
such binaries orchestrate the ways in which Western classical music and
Indian classical music, defined by their mutual opposition, are allowed to
meet. Thus, Western students learn to approach Indian music as a great
‘‘other’’ classical tradition founded on diametrically opposite values, and
they become disciples of Indian musicians because they seek the personal
intimacy missing in their Western training; Indian musicians look to the
West to improve their technical skills. These oppositions, and the discourse
in which they are embedded, are central to the institution of classical
music as it is now imagined in India. The emergence of this category of
‘‘classical music’’ and its definition in the context of colonialism, nation-
alism, and regional politics in South India are the subjects of this book.

In the last decades of the nineteenth century and the first decades of the
twentieth, the musical field of South India underwent a series of profound
shifts. Toward the end of the nineteenth century, musicians who had pre-
viously been supported by the patronage of royal courts or village temples
began to move in increasing numbers to the colonial city of Madras (now
Chennai) to take advantage of a new patronage structure: musical organi-
zations and institutions run by a concentration of middle- and upper-class,
high-caste (mostly Brahmin, but also Chettiar and Mudaliar) professionals
working in the fields of law, medicine, civil service, education, and busi-
ness. New kinds of public space came into being with the concert halls
and academies established by this upper-caste elite, who were interested in

4 introduction
what they termed the ‘‘revival of South India’s classical music.’’ 4 This ‘‘revi-
val’’ depended on the selection, from a number of heterogeneous musical
traditions, of particular sounds, performance conventions, and repertoire
that would come to be identified with the indigenous ‘‘classical’’ music
tradition of South India. In controlling the musical institutions of the city
and the music-publishing industry, this elite influenced how classical music
came to be imagined and defined. The new discourse on classical music—
whose assumptions are taken for granted today—combined ideas about
art and the artist with a notion of Indianness formed in opposition to the
West. It suggested that music had a central role in defining what it meant
to be modern while retaining a safely delineated realm called ‘‘tradition.’’
The idea of an indigenous classical music involved a complex maneu-
ver. To be considered ‘‘classical,’’ Karnatic music had to be modeled on the
classical music of the West, with its notation, composers, compositions,
conservatories, and concerts. The use of the English term classical for Indian
music, beginning in the early decades of the twentieth century, placed it in
relation to the art music of Europe. In Europe the term came into use only
after 1800, in reference to a particular ‘‘golden age’’ of composers and as a
marker of high status (Weber 1984, 175; Goehr 1992, 247).5 Indian music,
by contrast, never had a classical period; from its first use, the term was a
marker of cultural status and authenticity whose original referent was not
a historical period or particular style, but the West itself.6 At the same time
as South Indian elites emphasized the commensurability of Western and
Indian classical music, however, they recognized that in order to be truly
Indian, Karnatic music had to be different in reliable and enduring ways.
Much of the discourse on music from the early twentieth century revolves
around finding and articulating that which set South Indian classical music
apart from its counterpart in the West.
The argument of this book is that the voice—the vocal nature of Indian
music and its ties to oral tradition—came to stand for this essential dif-
ference. In twentieth-century South India the voice came to be asso-
ciated with Indianness and not Westernness, originality and not repro-
duction, humanity and not mechanization, tradition and not modernity.
What emerged in the twentieth century as a foundation for the idea of an
indigenous Indian classical music was a particular politics of voice that de-
pended on such associations. It was not just that a certain kind of voice
came to be valued, but that the voice itself came to be privileged as Kar-
natic music’s locus of authenticity, the preserver of its tradition in the
face of modernity. A number of circumstances attended the emergence of

introduction 5
this politics of voice: a current of nationalist thought heavily influenced
by Western metaphysical notions of the self as a speaking subject; a so-
cial reform movement that stressed the virtues of domesticity and female
chastity; the rise of an urban middle class, defined in part by an ideal of
‘‘traditional’’ Indian womanhood and in part by caste politics that privi-
leged a Brahmin vocal tradition over a non-Brahmin instrumental one;
the emergence of language-based identities; and the development of tech-
nologies of recording, amplifying, and disseminating the voice, such as the
gramophone, microphone, and radio.
The valorization of the voice in Karnatic music is part of a distinctly
modern set of ideas about music, the self, and Indianness. At issue here is
the tension between claims to the primordial significance of the voice and
the fact that such claims are made in particular historical and cultural con-
texts. When twentieth- and twenty-first-century musicians, musicolo-
gists, and listeners speak of the ‘‘fundamentally vocal’’ character of Indian
music, they often invoke the authority of a ‘‘civilizational tradition.’’ The
importance of vocalization as a way of knowing, conceptualizing, and ex-
pressing music in present-day Karnatic musical practice is seen as a di-
rect continuation of Sanskrit treatises on music, from the sixteenth cen-
tury and earlier, that emphasize the importance of vocal sound to music
making.7 Imputing an essentially vocal character to Indian music that
stretches into the ancient past, these claims have a distinctly Oriental-
ist ring. Positing ‘‘Indian music’’ as an entity removed from history, they
ignore the entailments of this music in the project of modernity. In doing
so, they exemplify one way in which the voice is ideologized in discourse
about classical music in South India. For it is precisely within modernity
that the voice comes to occupy such a privileged position. In the con-
text of the twentieth-century revival of Karnatic music as one of India’s
classical music traditions, precolonial Sanskrit treatises on vocal sound get
appropriated by a set of discourses about the voice associated with colonial
modernity and the emergence of bourgeois subjectivity.
Rather than approaching modernity either as a European invention or
as simply a time period or a process (‘‘modernization’’) through which all
societies eventually pass—perspectives which essentially depoliticize the
project of modernity—recent scholarship has argued that many of the in-
stitutions and ideas associated with modernity were linked to colonial
modes of knowledge production and structures of power (Chakrabarty
2000; Chatterjee 1993). Modernity is thus not a purely Western or Euro-
pean project; on the contrary, it is constituted in and by the colonial en-

6 introduction
counter. ‘‘Staging the modern,’’ writes Timothy Mitchell, ‘‘has always re-
quired the non-modern, the space of colonial difference’’ (2000, xxvi). In
order to perform its authority, modernity must constantly oppose itself
to what it conceives of as nonmodern. Modernity in this sense is not a
creation of the West but of an interaction between the West and the non-
West. Its staging ‘‘does not occur only in the West . . . to be imitated later
in the non-West. Its authority and presence can only be produced across
the space of geographical distance. It is this very displacement of the West
that enables modernity to be staged as ‘the West’ ’’ (24).
Modernity has been characterized variously as a set of institutions that
constitute a civil society or public sphere; a set of technological develop-
ments that produce certain effects; a shift from feudal to capitalist modes
of production; a set of ideas about the nature of time, history, society, and
the individual. In this last sense, modernity can be seen as a discursive for-
mation which has naturalized particular ways of thinking dependent on a
series of familiar binaries: secular vs. sacred, content vs. form, rational vs.
nonrational, mind vs. body, public vs. private, and, not least, tradition vs.
modernity. Indeed, one of the most powerful ways in which the project
of modernity operates is by defining itself as representative of rationality,
progress, change, and universality, in opposition to ‘‘tradition,’’ a category
which comes to stand for all that is irrational or emotional, stagnant, an-
cient, and local (Bauman and Briggs 2003). Such oppositions gain cur-
rency, of course, by being mapped as the difference between the West and
the non-West.
The modern subject is one who ideally holds these poles—the rational
and the emotional, the universal and the particular—in tension with each
other; this tension is what produces the effect of interiority characteristic
of modern subjectivity. Interiority, as Dipesh Chakrabarty has suggested,
‘‘comes to be constituted by a tension between individual-private experi-
ences and desires (feelings, emotions, sentiments) and a universal-public
reason’’ (Mitchell 2000, 63). As crucial to modern conceptions of the indi-
vidual self as the rational, disengaged stance is the notion of an ‘‘inner
voice’’ that stands for instinct and emotional life; in fact, as Charles Taylor
has suggested, they developed together (1989, 390). The modern concep-
tion of the self, he argues, involves a localization within the subject of what
was previously seen as existing between the knower or agent and the world
(186); the self comes to be seen as a repository of inwardness (284), of inner
emotions, ideas, and desires that need to be ‘‘expressed.’’ Modern subjec-
tivity hinges on the notion of voice as a metaphor for self and authen-

introduction 7
ticity, and on the various techniques—musical, linguistic, and literary—
by which particular voices are made to seem authentic. The modern idea
of art depends on this notion of expression as a creative, original, indi-
vidual act, as opposed to imitation or mimesis, which comes to be seen as
merely reproductive, belonging more in the realm of craft or tradition.
The fact that this inner domain of self, originality, and authenticity is
figured as a voice is not coincidental, for such ideas of the self and art go
hand-in-hand with a certain way of imagining language and the role of
the voice. As Bauman and Briggs have compellingly argued, constructions
of language and tradition played a central role in the modernist project:
one crucial way in which the modern subject of the European enlighten-
ment identified himself was by differentiating his own language—ratio-
nal language, purified of unnecessary associations and suited to expressing
supposedly universal concepts—from the language of lower-class ‘‘folk’’
or ‘‘masses,’’ which was mired in custom and superstition (2003, 10–12).
Purifying language in this sense, making it rational, meant emphasizing
referentiality over all other functions of language; it meant drawing a line
between content and form, and privileging the former. Thus, reason comes
to be defined in opposition to rhetoric (27), just as poetry and ‘‘oral tra-
dition’’ come to be defined in opposition to rationality and ‘‘literature’’
(70–72, 103–4).
These ideas about language combine two ways of imagining the voice.
On the one hand, the association of voice with agency and sincerity is
at the heart of notions of the rational subject; the voice in this sense is
imagined as referring to, or directly expressive of, an individual, interi-
orized self. On the other hand, such a notion of voice is formed in rela-
tion to other voices that come to be labeled in their plurality ‘‘oral tra-
dition’’—those voices which call attention to performance, sound, and
materiality and thus fail to privilege referentiality. The category of oral
tradition and the association of orality with simplified cognitive, social,
and political characteristics are central to the discourse of modernity (Bau-
man and Briggs 2003, 107).8 Within this discourse, voice becomes a power-
ful trope of the modern, rational subject at the same time as its sonic and
material manifestations are seen as standing for authenticity and tradition.
It is precisely this dual imagining of voice that is at work in the twentieth-
century definition of South Indian classical music.
In any work that takes modernity outside of the West as its focus, there
is a tension between finding modernity and finding indigenous resistance
to the culture of modernity, which Joel Robbins has characterized as the

8 introduction
problem of ‘‘tacking back and forth between claims of convergence and
divergence, homogenization and differentiation’’ (2001, 901). One would
not want to view the narrative I present here as simply an imposition of
the culture of modernity onto indigenous ideas and practices. However, it
is equally risky to identify the institution of South Indian classical music
as indigenous resistance to modernity, for such a move would erase the re-
lations of power and regimes of knowledge production that have created
the power and prestige of classical music in South India and enabled its
passage to the West. Rather than identify a ‘‘local’’ or ‘‘alternative’’ moder-
nity, I examine the kinds of claims to modernity that are made in a par-
ticular context. Deferring the question of whether such claims constitute
imposition or resistance, I instead focus on the ways that claims to moder-
nity are made by certain groups of people for specific purposes; claims to
modernity in this sense are claims to power.
To say that the valorization of the voice in Karnatic music is a modern
phenomenon is not to make an argument about the ‘‘invention of tradi-
tion.’’ Such an idea unhelpfully sets up a choice between ‘‘invented’’ and
‘‘real’’ or ‘‘authentic’’ traditions. Indeed, as Marilyn Ivy has suggested, the
idea of the invention of tradition also dangerously posits a world of dis-
course as separable from reality: ‘‘To say that all tradition is invented is
still to rely on a choice between invention and authenticity, between fic-
tion and reality, between discourse and history’’ (1995, 21). It forces one to
separate what is traditional from what is imported, to see things in terms
of an imposition of Western concepts and ideas onto indigenous musi-
cal material. In fact, the institution of classical music in South India—not
only discourse about it but the very sound and practice of the music—has
been produced in and through the colonial encounter. Thus the sounds,
practices, and categories I discuss in this book are neither properly West-
ern nor Indian, but specifically colonial in the sense that they position the
West and India in relation to each other.9

Embedded within the specific arguments of this book is a more gen-

eral claim. In identifying a particular politics of voice that emerged
in twentieth-century South India, I am pointing to the need to treat
the voice as a historical and anthropological object: something locatable
within a particular time and place. I seek to bring the sonic, material
aspects of voice into relation with issues of representation, subjectivity,
and agency. Linguistic anthropologists have for several decades been con-
cerned with ‘‘regimes of language’’ or ‘‘language ideologies,’’ that is, bodies

introduction 9
of culturally constructed knowledge about how language works and how
it should be used. Language ideology ‘‘sets the boundary for what counts
as language and what does not, and the terms, techniques, and modali-
ties of citing and hearing’’ (Inoue 2003, 157).10 Ideas about the voice (what
it should sound like, where it comes from, how it relates to a singer’s or
speaker’s body, its status in relation to writing and recorded sound, etc.)
are undoubtedly part of language ideology, just as they are part of a per-
haps broader regime or politics of voice that includes but is not exhausted
by language. One might say that ideologies of voice, in many cases, set
the boundary for what separates language from music, as well as for what
counts as communication. Most crucial, ideologies of voice determine
how and where we locate subjectivity and agency.
The voice has been most explicitly thematized as an object of study
in psychoanalytic writing. Psychoanalytic theories of voice emphasize its
primordiality, its privileged role in creating a sense of self. These notions
find their clearest expression in Jacques Lacan’s formulation of the voice
as one of several ‘‘partial objects’’ which cannot be considered either part
of the body or fully separate from it, but that are also partial because they
represent only part of the function with which they are associated (1977).
The very partiality of these objects is what gives them a primary role in
creating a sense of subjectivity: ‘‘It is what enables them to be the ‘stuff,’
or rather the lining, . . . of the very object that one takes to be the sub-
ject of consciousness.’’ 11 Responding to such formulations, recent work in
musicology and in literary and film studies on the relationship between
gender, voice, and embodiment has considered the place of the female
voice in the Western cultural imagination and the stakes involved in its
association with or disassociation from the female body.12 Although much
of this work provides a valuable critique of the ways the female voice has
been essentialized either as seductive, dangerous, irrational, or alterna-
tively as a site of women’s potential liberation, its focus remains strictly
within a Euro-American context and in relation to particular psychoana-
lytic paradigm. Building on these critiques, one might ask: how does the
voice assume different significances in different times and places? What
are the technologies by which experiences of voice are constructed and
metaphors of voice made to seem natural?
While psychoanalytic approaches rely heavily on the assumption of
the voice’s primordial status, its prehistoric quality, social science has
generally limited discussions of voice to the metaphorical level of self-
representation. In anthropology especially, the voice, while not always ex-

10 introduction
plicitly thematized, has been identified as a vehicle of empowerment, self-
representation, authentic knowledge, and agency. The assumption that
underlies this metaphorization of voice, a central tenet of Western phi-
losophy, is that the speaking subject is the ground of subjectivity and the
source of agency. As media theorist Jonathan Sterne suggests, ‘‘Face-to-
face conversation, embodied in mutual dialogue between speaking sub-
jects, serves not only as the classic field situation for anthropologists, but
also as the standard for judging the originality and authenticity of utter-
ances and of communication in general’’ (2003, 20, 342). In this meta-
physics the voice is fetishized, made to stand as an authentic source apart
from the social relations that have produced it. In becoming strictly a
metaphor for consciousness, voice and its opposites, silence and deafness,
come to ‘‘inhabit our philosophic and political discourse as nothing more
than clichés’’ (348).
Valuable critiques of fetishization of the voice have been made by post-
colonial theorists and historians. In this regard, Gayatri Spivak’s essay ‘‘Can
the Subaltern Speak?’’ (1988) is programmatic in its critique of the ideo-
logical underpinnings of the project of recovering lost or subaltern voices.
More recently, Mrinhalini Sinha has pointed out the problematic nature
of the project of ‘‘allowing women’s voices to be heard’’ as an antidote
to male-dominated histories and historiography (1996). The desire for a
‘‘pure’’ feminist consciousness, she writes, ‘‘serves, in the end, to remove
the feminist subject from the history of her production within intercon-
nected axes of gender, race, class/caste, nation, or sexuality’’ (498). She sug-
gests that one look instead at the ways women ‘‘come into voice’’ and at
the particular voices that women assume. Such an approach, she maintains,
‘‘would have to take into account both the historical context which made
possible the identity of the Indian woman and the particular strategies by
which women learned to speak in the voice of the Indian woman’’ (479).
Sinha’s approach is useful in that it shows the conscious construction
of voice, the ways women learn to speak so that they will be heard in an
arena already overdetermined by colonialism and nationalist politics. It
shifts the focus away from the notion that the voice of the Indian woman
was simply and naturally there, just waiting to be expressed, toward the
idea that such voices were assumed strategically at a particular historical
moment. Sinha thus illuminates much about the politics of gender, caste,
race, and feminism at the time. At a theoretical level, she suggests that
voices—here, the voice of the Indian woman—are socially produced, that
the link between voice and agency or self-expression is complicated by

introduction 11
questions of genre, audience, and historical location. However, she leaves
the category of voice itself untouched. Indeed, the sense in which she and
her historical interlocutors use the term voice is almost exclusively that of
political self-representation, wherein voice is essentially a metaphor for
representation in writing and in a middle-class, English-speaking public
milieu.
Thus, while critiques such as these begin to challenge the notion of
agency based on the model of the speaking subject, they do little to chal-
lenge the other part of that model, which has always treated the voice itself
as secondary, as a mere vehicle for the expression of referential content.
In order to move beyond the clichéd metaphor of voice as representa-
tion, one needs to focus on the sound and materiality of the voice, and
to consider the way those sounds and material practices of voice are put
into service in the creation of ideologies about the voice. In this regard,
I find my concerns to be parallel to those of the media theorist Fried-
rich Kittler, who in Discourse Networks 1800/1900 (1990) in effect historicizes
the voice by considering the conditions of its production and dissemina-
tion. He traces the notion of an ‘‘inner voice,’’ central to the emergence
of what he calls modern subjectivity in the West, back to certain prac-
tices of reading, writing, and pedagogy. For Kittler, the inner voice, based
on the aural memory of the mother’s voice, which is associated with new
family organization and new reading practices emerging in Germany at
the end of the eighteenth century, is radically historical, not a psycho-
logical universal. Around 1900, when it became possible to record voices,
Kittler argues, the concept of the inner voice and its accompanying forms
of subjectivity were no longer available in the same way. The notion of a
‘‘discourse network’’ as a network that links bodies, sounds, writing and
other technologies, and forms of authority very usefully draws attention
to the fact that philosophical ideas are grounded in the material practices
and technologies through which voices become audible.
Rather than starting with a universal concept of voice and its impor-
tance, then, I look for moments when self-conscious discourse about the
voice arises. Rather than considering the voice as a natural means of self-
expression, I ask how voices are constructed through practice and how
music entails particular ideologies of voice. Musicology and ethnomusi-
cology, as disciplines that focus on sound, have generally treated the voice
as a means or technology for producing music. I propose that we reverse
this idea and examine music as a means or technology for producing the
voice, in both a sonic and an ideological sense. In doing so, we perform the

12 introduction
critical anthropological move of denaturalizing the voice and thus open
up the study of musical sound and practice as productive of particular sub-
jects or subject positions, rather than merely reflective of social structures
or expressive of identities. What new forms of subjectivity and identity are
enabled by changing conceptions and practices of voice?
In addressing this question, I consider the material and sonic aspects
of the voice in relation to the way voice operates as a culturally created
metaphor. Focusing on regimes or ideologies of voice requires an exami-
nation of the relationship between sound or form, on the one hand, and
meaning or function within a discursive realm on the other. It requires
linking ‘‘a phenomenological concern with the voice as the embodiment
of spoken and sung performance . . . [with] a more metaphoric sense of
voice as a key representational trope for social position and power’’ (Feld
and Fox 1994, 26). Recognizing the socially constructed nature of musi-
cal sound is essential to what Steven Feld, Aaron Fox, Thomas Porcello,
and David Samuels call ‘‘vocal anthropology’’ (2004). In many cases, they
suggest, song or musical sound allows for a ‘‘ritualized, explicit consider-
ation . . . of the voice as the material embodiment of social ideology and
experience’’ (332). In other words, it is the capacity of the voice to be both
iconic (able to embody particular qualities) and indexical (able to point
to or index particular subjectivities or identities) that makes it so power-
ful as a metaphor.13 Precisely because voices are embodied and repeatedly
performed, they can serve as deeply felt markers of identity, class, caste,
gender, social position, and nationality. In this sense one can speak of a
‘‘politics of voice’’ as a set of vocal practices as well as a set of ideas about
the voice and its significance.
The notion of vocal anthropology, as elaborated by Feld, Fox, Porcello,
and Samuels, opens up a variety of productive ways to link the study of
musical and linguistic sound with social identity through attention to the
voice. While there is much in common between my project and theirs, my
emphasis here is slightly different. Feld and his coauthors link the voice
to both nature and resistance, assuming that the significance of the voice
arises from universal human experience and identifying the larger pur-
pose of vocal anthropology as the study of voices which express or embody
resistant identities or which ‘‘talk back’’ to hegemonic power structures
(2004, 341–42).14 In part because I examine ideas about the voice asso-
ciated with an art-music tradition that has achieved a kind of hegemonic
status, I am wary of assuming either the link between voice and resistance
or the naturalness of voice as a metaphor. Instead, I use the concept of

introduction 13
‘‘politics of voice’’ or ‘‘ideology of voice’’ to emphasize that practices of
voice, while creative, are also a mode of discipline—embodied and per-
formed—through which subjects are produced. I suggest that ideas about
the voice and its significance are motivated by historically and culturally
locatable practices of voice, rather than by supposedly universal bodily ex-
perience. And since ideologies of voice determine what voices come to be
heard and how, understanding them as particular and changeable is essen-
tial to understanding the kinds of subjects and politics they both enable
and silence.

The chapters that follow trace the emergence of a particular politics of

voice by focusing on a series of moments when voice specifically comes
into question or gets redefined. The first two chapters explore the staging
of the voice in musical sound and performance practice. Chapter 1 con-
cerns the career of the violin, which arrived in South India around 1800
in the hands of British and French colonists and was adapted to Karnatic
music so successfully that by the early twentieth century it was regarded as
an indispensable accompaniment to the voice—indeed, as an instrument
capable of ‘‘reproducing’’ the voice. In the twentieth century, the violin
came to be seen simultaneously as the modernizer of Karnatic music and
as the preserver of Karnatic music’s authenticity, as embodied in the cen-
trality of the voice. What is the significance of having a colonial instru-
ment in this position?
Rather than suggesting a history of Westernization, the story of the
violin points to the emergence of a musical practice and discourse that is
specifically colonial. In both its role as an accompanying instrument and
its central role in experimental fusions of Indian and Western music, the
violin becomes a key instrument for the negotiation of what is Indian and
what is Western and of the relationship between them. I take into account
the instrumentality of musical instruments, that is, the fact that they are not
just vehicles for producing sound but also social instruments, instruments
for thinking about music and its significance. In this respect I suggest the
importance of engaging with performance and sound. Close engagement
with performance, as Carolyn Abbate has noted in her subversive readings
of opera, can disrupt the certainty of an authorial voice or a prevailing in-
terpretation; opera as a text or musical score and opera as a performance
produce very different notions of the authority of male and female voices
(1991). Prevailing explanations and descriptions of Karnatic music empha-
size the vocal nature of Karnatic music and relegate the violin to the status

14 introduction
of mere accompaniment, a colonial add-on to an already existent musical
tradition. In doing so, they disregard the complex mimetic relationship
between violin and voice that emerges in performance, where the violin
effectively functions to stage the voice—not merely to reproduce it but to
produce it.
Chapter 2 moves to actual stagings of Karnatic music that took place in
the newly developing contexts of urban concert halls and music schools in
Madras in the early twentieth century. The emergence of Madras as a cen-
ter of musical culture was a result of the shifts brought about by colonial
rule. The annexation of princely states by the British and the consolida-
tion of British authority over temples deeply affected patterns of musical
patronage in royal courts and temples outside of Madras city in the nine-
teenth century. By the middle of that century, most of the princely states
of South India had been effectively taken over by the British crown; al-
though kings continued to rule in these states, their political power was
diminished by the increasing centralization of the British colonial govern-
ment. Tanjore, taken under British administrative rule in 1779, was offi-
cially annexed in 1856; Mysore was taken over in 1831.15
The consolidation of British rule in South India had direct conse-
quences on the very nature of patronage itself. The old regime, as both
Nicholas Dirks and Arjun Appadurai have noted, operated according to
the principle of the gift: the ability to give gifts was a fundamental sign of
a king’s sovereignty, and gifts were the dynamic medium that constituted
political relations (Dirks 1993, 179–80; Appadurai 1981, 63). Under British
colonial rule, however, the flow of gifts was interrupted even as the British
tried to keep the ‘‘little kings’’ in positions of apparent power as a kind of
‘‘native aristocracy.’’ The little kings, under British administrative control,
were settled permanently on their land as zamindars, or landlords, with the
idea that they would collect revenue and act as middlemen between the
British colonial government and South Indian peasants (Dirks 1993, 181). If
the little kingdoms felt the effects of colonial interference by the late eigh-
teenth century, so did Hindu temples, another center of musical patronage.
According to Appadurai, the formation of the Board of Revenue in 1789
marked the increasing bureaucratic centralization of the colonial state and
the change of the East India Company from a trading power to a political
regime (1981, 109). With this came the decision that the collection and dis-
tribution of all temple revenues should be centralized. Whereas kings and
temple trustees (dharmakartas) had once administered the affairs of indi-
vidual temples as they saw fit, the colonial government was by the early

introduction 15
nineteenth century auditing the use of temple funds, and temple-related
disputes were increasingly brought before the Anglo-Indian judicial sys-
tem. Dharmakartas were reduced to the status of public servants (ibid.,
139, 165).
Madras City served as the center of colonial administration and grew
as the status of princely states and temple cities (such as Madurai) di-
minished. As a result of the Karnatic wars in the late 1700s, the Euro-
pean military presence in Madras increased to several thousand; at the
same time, the nonmilitary colonial population swelled in proportion
(Neild 1979, 223). As a trading power, the East India Company had de-
pended, for much of the seventeenth and eighteenth centuries, on ‘‘com-
pany merchants,’’ who came from the traditional mercantile Chetty caste,
for its supply of goods (Arasaratnam 1979, 20–21). Many of these mer-
chants bought land in Madras city and built temples, acting as patrons of
music (ibid., 23–24). After the mid-1700s, however, as the East India Com-
pany began to consolidate itself as a colonial power, local political influ-
ence and social prestige was transferred to a new class of men, the dubashes,
who were employed by colonial officials to act as go-betweens, transla-
tors, personal servants, and brokers (Neild 1984, 4). Most dubashes were
from the Vellala caste (landholding agriculturalists) and came to Madras
from the surrounding villages to seek employment in the colonial admin-
istration (ibid., 12–13). As dubashes, they enjoyed considerable wealth and
political power, in many cases acting as little kings in their own right. They
built large garden houses in the suburbs of Madras, often serving as the
dharmakartas of city temples (while retaining their ties to the temples in
their native villages) and patronizing musicians and dancers (ibid, 16).16
By the early nineteenth century, as the character of British colonial in-
volvement in South India changed to a more centralized bureaucracy, the
dubashes lost their jobs and their privileged status (Neild 1984, 18). The re-
organization of the East India Company between 1812 and 1832 brought
its profits under the administrative control of the British Crown. During
this period, and especially after the official declaration of the Crown’s au-
thority in 1858, Brahmins, rather than Vellalas, were actively recruited to
fill civil service positions. Many of these Brahmins came from the Tamil
and Telugu subcaste of Smartas, who had long claimed high ritual status
as priests and scholars, and settled in the Madras neighborhood of Myla-
pore (Hancock 1999, 52). Madras High School was founded in 1841 to
train a native elite, mostly Brahmins, in the skills needed for filling civil-
service posts (Suntharalingam 1974, 62–63). Although traditional mercan-

16 introduction
tile castes in Madras, such as the Chettiars, were clearly wealthier, Brah-
mins, working as clerks, lawyers, and publicists by the later decades of
the 1800s, were beginning to claim political and social influence within
the city. Meanwhile, musicians were moving to Madras from the princely
states, particularly Tanjore, in great numbers between 1850 and 1920 in
search of new employment.
Music was used by Brahmin elites to create a kind of public sphere
in Madras and a sense of common outlook. By 1900, Smarta Brahmins
in Mylapore had created a network of voluntary associations, the Radha-
Krishna Bhajanas, which gathered in homes and community halls to sing
bhajans, or devotional songs (Singer 1972, 200–205). Mary Hancock sug-
gests that bhakti, individualized voluntaristic devotion, with its universalist
ethos, was appropriated by Brahmins as a kind of expression of bourgeois
nationalism, a ‘‘hallmark of new, nationalized, and self-consciously mod-
ern elite sensibilities’’ (1999, 57–58).17 Music sabhas, private organizations
that sponsored concerts, were another hallmark of such sensibilities. In
Madras at the beginning of the twentieth century sabha owners became
the new patrons of Karnatic music.
The rise of public concerts and commercial music making in Madras
spawned new standards of taste, embedded in a set of conventions of voice
brought about by the mechanism of the microphone and codified in the
canonization of a group of saint-composers who have come almost ex-
clusively to represent the voice of Karnatic music. Chapter 2 argues that
the establishment of a canon depends not only on the selection and valua-
tion of particular repertoire, but also on the establishment of techniques of
physical staging and modes of listening—disciplines that are as embodied
as they are intellectual. The emergence of concerts halls, music schools, and
university music departments in the early twentieth century was linked
to the emergence of new kinds of performing and listening subjects. Just
as the concert hall entailed a new structure of presentation that demanded
the clear separation of musicians from audience, the new concert format
that developed demanded the separation of musical compositions from
the contexts of their performance. The celebration of the saint-composer
Thyagaraja as the primary voice of Karnatic music entailed a shift from
compositions tied to contexts of performance to compositions that were
imagined to have their origins in a personalized, timeless devotion. At the
same time, the microphone, used in concert halls beginning in the 1930s,
was not only a technology of amplification but also an instrument that
enabled the emergence of new kinds of voices and performing subjects.

introduction 17
 Chapters 3 and 4 discuss the emergence of a particularly modern ide-
ology of voice inflected by both gender and caste politics in the mid-
twentieth century. The discourse of social reform that arose in the late
nineteenth century, associated with elite nationalist thought, was central
to notions of what made music and dance ‘‘classical’’ and on ideas of their
place in a new, urban, bourgeois order of things. Among social reform
projects, those grouped under the ‘‘woman question’’—sati, widow remar-
riage, child marriage, the devadasi issue—loomed large because the status
of women and the status of the nation were linked in nationalist thought.
The definition of the ‘‘devadasi system’’—a general term for a variety of
ways in which women were employed by temples as musicans, dancers,
and ritual performers—as an object of social reform provides a particu-
larly good example of the way social reform and cultural ‘‘revival’’ were
intertwined. The Women’s Indian Association (wia), founded in 1917 by
Annie Besant, Margaret Cousins, and Dorothy Jinarajadasa, was explic-
itly concerned with such issues; the advancement of India as a nation,
they felt, depended on improving the status and situation of its women.
Indeed, it was another wia leader, Muthulakshmi Reddy, who authored
the measures for the Devadasi Dedication Abolition Act. In the discourse
of social reform the devadasi system came to be viewed as synonymous
with prostitution, while the arts of music and dance practiced by devadasis
were seen as needing to be rescued and placed in the hands of ‘‘respectable
family women.’’
Chapter 3 is concerned with the emergence of a set of ideas about
what constituted a ‘‘natural’’ or true voice, the relationship between the
voice and the body, and the relationship between the singing voice and the
speaking voice. This politics of voice was emergent in certain conventions
of performance established in the moment when upper-caste, ‘‘respect-
able,’’ ‘‘family’’ women, who previously did not perform in public, began to
sing on the concert stage and make gramophone recordings. In discourse
about these female musicians, the natural voice and the chaste female body
were linked. Those female voices that represented India’s classical music
traditions were imagined as giving voice at once to an abstract notion of
art and to the body of ‘‘Mother India.’’ I draw a parallel in this chapter
between the literal domestication of music as a sign of bourgeois respect-
ability, its connection to a discourse about family values, and the interi-
orization of music within the body in terms of performance practice. The
modern notion of the artist that underlay the revival of Karnatic music as
South India’s classical music depended on this new sense of interiority.

18 introduction
 In the mid–twentieth century the idea of the ‘‘natural’’ voice was elabo-
rated as well in the context of debates about the relationship between lan-
guage and music and the role of language in music. Such debates emerged
in the context of a broad current of Tamil revivalism, embodied in the
Dravidian and non-Brahmin movements that arose to challenge elite na-
tionalist discourse. Tamil revivalists rallied around the figure of Tamittāy,
or Mother Tamil, whom they imagined as needing protection from the
onslaught of ‘‘Aryan’’ culture and languages.18 Along with language, music
became one of the central fields in which claims to Aryan and Dravidian
difference and Brahmin and non-Brahmin difference were made in the
1930s and 1940s. The Tamil music movement—a demand that the classical
music repertoire include more songs in Tamil rather than Telugu or San-
skrit—although not originally conceived as a non-Brahmin movement,
came to be strongly associated with protest against the Brahmin domina-
tion of classical music.
Chapter 4 argues that the discourse about music’s meaning that emerged
in this context signaled a new set of discourses not only about the rela-
tionship between music and language but ultimately about the singing and
listening subject. In the 1930s and 1940s, as the modern idea of music as
a language began to dominate nationalist discourse, language became the
central metaphor for articulating the relationship between voice and the
singing or listening subject. Once this analogy had taken hold, the cate-
gories ‘‘classical language’’ and ‘‘mother tongue’’ increasingly dominated
debates about how music should be defined and experienced, as did ideas
about the ‘‘meaning’’ of music. Was music akin to a mother tongue or a uni-
versal, aesthetically motivated language? Was the voice best conceived of
as an aestheticized instrument that required cultivation or as a transparent
representation of oneself ? Those on both sides of these debates sought to
endow the singing voice with interiority precisely by comparing music to
language; the concern with meaning relied on an opposition between real
music and mere acrobatics, the threat of purely physical, automatic action,
pure sound, exteriority. This discourse, relying as it did on the opposition
between interiority and exteriority so central to modern concepts of the
self, was essentially occupied with the problem of how to create modern
musical subjects.
Central to modern self-definition is the notion of a rational, enlight-
ened self who can appreciate but is not bound by tradition and is there-
fore an appropriate custodian of tradition. In terms of voice, this means
that the modern subject is one who can quote, refer to, or preserve tra-

introduction 19
dition while standing outside it; the quoting voice marks itself as mod-
ern at the same time as it marks the quoted voice as other than modern
(Inoue 2002; Bakhtin 1981). This logic, in combination with certain tech-
nologies of preservation and recording, makes the notion of an originary
oral tradition possible. The production of music and musical knowledge
in South India in the twentieth century was shaped by the explosion of
printing and technologies of sound recording and broadcasting. Far from
being incidental technologies used to record or preserve oral traditions,
printing and recording were crucial to the creation of a notion of oral tra-
dition as an originary source that needed to be called on or preserved. In
chapters 5 and 6, I examine various claims to voice and oral tradition that
surround the composer and the guru, two figures central to the definition
of Karnatic music as a classical tradition.
The notion of the composer, chapter 5 suggests, came into being in
South Indian music only in the twentieth century, with the widespread
use of notation and the printing of notation and music manuals. With
the idea of the composer came other ideas: that of sole authorship, origi-
nality, and art for its own sake. Most important, the idea of the composer
demanded notation, a way of preserving the original voice of the com-
poser from forces that might change or alter it. Chapter 5 explores these
ideas by looking at a debate about the legitimacy of a particular composer
that erupted in the 1980s and working backward from this moment to ex-
amine conflicting ideas about notation from the early twentieth century.
Notation was seen both as a guarantor of literacy, and therefore of classical
status, and as a transparent and legible representation of orality, and there-
fore of Indianness. While an originary oral tradition came to be privileged
as a sign of the Indianness of the music, the idea of the composer was cru-
cial to proving its legitimacy as one of the classical musics of the world. A
contradictory set of anxieties about the voice is at work here: the fear that
the voice could be lost if not captured by writing coexists with the fear that
the voice could be lost precisely by being completely captured by writing.
Thus, along with the imagination of a pre-existent, somehow pure oral
tradition also come the notions of the sublime and the ineffable, the idea
that the real power of music cannot be expressed by notation or words.
These anxieties about the voice are elaborated as well in the uneasy co-
existence of human embodiment and mechanical mimesis as models of
authenticity in contemporary discourse about Karnatic music. In chap-
ter 6, I explore the relationship between technologies of sound reproduc-
tion and the mode of learning from a guru, focusing on moments when

20 introduction
ideas about voice and music and when practices of listening and perform-
ing seem to change in conjunction with technologies of recording and
broadcasting. For elites interested in the ‘‘revival’’ of South India’s clas-
sical music, gurukulavasam was problematic because it defied standard-
ization and preserved idiosyncrasy, but it was also desirable because it
seemed to embody a particularly Indian mode of disseminating and ac-
quiring musical knowledge, an oral tradition that defied writing. Mean-
while, sound reproduction, correctly channeled, was seen as a potential
tool in the project of developing a standardized body of theory and prac-
tice, of making Karnatic music commensurate with other great musical
systems of the world.
The concept of sound fidelity, as Jonathan Sterne argues in relation to
the history of sound recording in the West, is a social construct, reflect-
ing more the way technologies were conceived than the actual relation of
a sound to its source (2003, 219). In chapter 6, I suggest that the concept
of sound fidelity in the South Indian context is imagined in relation to
and through a distinctly postcolonial sense of fidelity to tradition. To be
sure, the technologies of sound recording and broadcasting created dis-
ruptions in earlier modes of teaching, performing, and listening, but there
is also a way in which desire for the traditional is projected out of the new
technologies themselves. Instead of narrating the takeover of ‘‘tradition’’
by ‘‘modern’’ technologies, I demonstrate how ideas about tradition and
authenticity were themselves formed in the encounter with such tech-
nologies. This is not to argue for technological determinism but rather to
suggest a focus on the ways that technologies both shape and are shaped
by local politics and forms of knowledge.

Finally: what is at stake in this project? The fraught nature of any research
about Karnatic music was made clear to me one evening in Madurai, in
1998, when I was discussing the Tamil music movement with a musician
friend. His father, a distinguished scholar of Tamil, had apparently been
listening from the verandah as we talked. After about fifteen minutes, he
stormed in and asked in anger what the point (nōkkam) of my researching
this topic was. What kind of conclusion ( poruḷ ) was I going to draw from
it? Was I going to tell how Brahmins had taken Tamil music and translated
it into Sanskrit, and how the Brahmins never acknowledged their non-
Brahmin teachers? Was I going to take a side? Or was it all just amusement
(vēṭikkai) for my professors? 19 As I prepared to stammer out an answer, he
drew himself up and uttered the following verse from the Tirukkural:20

introduction 21
 Epporuḷ yār yār vāyk kēṭpinum apporuḷ meypporuḷ kāṇpaṭu aivu.
Epporuḷ ettamait tāyium apporuḷ meypporuḷ kāṇpaṭu aivu.
[Whatever thing from whosever mouth one may have heard that
thing a true thing to see is knowledge.
Whatever thing in whatever state it may be that thing a true thing to
see is knowledge.]

The tension and anger in his voice dissolved as he assumed the voice of
the ancient Tamil sage Tiruvalluvar. The archaic and proverbial sound of
the words, mysteriously different and yet recognizable as Tamil, removed
them from everyday speech, evoking the ancient Tamil past. Their allitera-
tively poetic sound invited repetition and demanded interpretation; their
allure was in the fact that they remained, even after translation, slightly
ambiguous. These words imply that truth is not simply found but created
actively by the knowing subject, who picks out what she hears or sees as
the truth from many possible (and often conflicting) versions, both written
and aural. Finding mey (truth) can thus never be mere vēṭikkai.
Although his anger faded, his message remained clear: any ethnogra-
phy or history of Karnatic music had to come to terms with the exclusions
that the consolidation of classical music had brought about. Many people
with whom I discussed my project in Madras agreed that Karnatic music
is ‘‘more classical’’ now than it ever has been before. By ‘‘classical,’’ they
meant exclusive, dominated by Brahmins and Brahmin ideals, associated
with conservative cultural politics, composition oriented, standardized,
rule bound. They meant that the classicalness of the music was not a given
characteristic but an imposed set of values. As Karnatic music has become
more exclusive, however, it has also become more available to the general
public through new media: cassettes, radio, Tamil cinema. Framed within
a thoroughly culturalist discourse, it operates in a larger public sphere pre-
cisely as a sign of culture, tradition, and conservative values.
If the consolidation of a discourse about classical music served nation-
alist purposes in the early twentieth century, it now serves the needs of
a generation of South Indians who have migrated to the West. Karnatic
music is now transported on a large scale to the United States, Canada,
Britain, and Australia, where it has come to signify tradition and Indian
culture. With the migration of thousands of upper-caste, mostly Brahmin
South Indians to the United States in the last few decades, Karnatic music
has also, through its identification with culture, become a safeguard of dif-
ference; it is there to insure that although East and West may meet, there

22 introduction
will always be a realm in which they can be kept safely separate. Hindu
temples, music schools, and music organizations have sprouted around the
country to feed the growing demand of nonresident Indians for the music
of their motherland. A new pattern of patronage has developed, with soft-
ware engineers and doctors serving as temple trustees and sponsors for
concert tours. Cleveland, Ohio, which sports a world-renowned music
festival in honor of the South Indian composer Thyagaraja, whose atten-
dance far exceeds that of the original festival in Tiruvaiyaru, has earned
the half-joking title, ‘‘the Second Tanjavur.’’ 21
A by-product of this efflorescence is the association of Karnatic music
with a pure, original, Indian Hindu culture. For instance, in 1997 the
right-leaning magazine Hinduism Today featured a didactic centerfold pre-
sumably meant for the children of immigrants growing up in the United
States.22 Through a calculated juxtaposition of images it purports to show
such children what ‘‘their’’ culture is really about and how it is different
from Western culture: a disco scene with boys and girls dancing together
in jeans is side-by-side with a picture of a woman dressed in a sari and
facing her husband with hands together in the ‘‘traditional Indian greet-
ing’’; a picture of a marching band is placed beside a picture of a classical
concert party; below, on the ‘‘Indian’’ side, are the words ‘‘the way Indian
culture prefers it.’’
To answer the question so passionately thrown at me that evening in
Madurai, then: the purpose of the project I have attempted here is to op-
pose culturalist appropriation by revealing the complex political encoun-
ters from which the institution of classical Karnatic music emerged. I focus
on the appearance of particular practices of and attitudes about music that
have become identified with the notion of classical music in South India
to suggest that, far from being natural or purely aesthetically motivated
manifestations of an essentially Indian sensibility, they are the products of
a particular colonial and postcolonial history.23
If one of the goals of this project is to critique the formation of a musical
canon in South India, it is equally concerned with the ways that music—
especially that which is labeled ‘‘art music’’ or ‘‘classical music’’—has been
placed in our own academy. Indian classical music has a canonical place in
ethnomusicology; its status as an art or classical music was what allowed
many ethnomusicology programs in the West to argue in the 1950s and
1960s that their field was legitimate, indeed, commensurate with musi-
cology. Ethnomusicology embraced the idea of a musical tradition that
seemed to be framed in familiar terms—composers, compositions, con-

introduction 23
certs, virtuoso performers, a musicological tradition of its own—and yet
utterly different in its musical sensibility and logic. By uncritically bor-
rowing Indian nationalist and postcolonial discourse, ethnomusicology
elided the colonial history that made the category ‘‘Indian classical music’’
possible.
While ethnomusicology was embracing Indian classical music as part of
India’s ‘‘great tradition,’’ anthropology, at least in regard to South Asia, was
busy developing its own canon around the ‘‘little traditions’’: those prac-
tices and traditions, mostly oral and nonstandardized, associated with the
folk or common people rather than with elites.24 Meanwhile, because of
the way music has been framed in our academy and in our society at large
as something to be studied and ‘‘mastered’’ only by those with specialized
knowledge, anything relating to music or sound generally fell outside of
anthropology’s focus. Indian classical music was thus, like Western classical
music, constructed as an object fit for musicological study but distinctly
unfit for anthropological treatment.
One of the most important aims of this project, then, is to bring music,
generally, and Indian classical music in particular, within the purview of
anthropology. The project is by necessity both historical and ethnographic
in its attempt to dislodge Indian classical music from the musicological dis-
course within which it has persistently been contained. Practices of musi-
cal performance and cultures of listening are crucial as domains in which
modernity is staged and embodied and in which claims to authenticity are
made. In the chapters that follow, I endeavor to show how India became,
in the words of Margaret Cousins, ‘‘the kingdom of the Voice,’’ and what
such a claim might mean.25

24 introduction
1 $ Gone Native?
travels of the violin in south india

No other instrument is so powerful as the Violin.

—T. K. Jayarama Iyer (1965)

Unlike the golden age of India, which Orientalist accounts place in

the precolonial era, the golden age of Karnatic music occurred, accord-
ing to conventional accounts, at the peak of colonialism, the early to
mid-nineteenth century. The so-called trinity of composers—Thyagaraja
(1767–1847), Syama Sastri (1762–1827), and Muthuswamy Dikshitar (1776–
1835)—who are said to have revolutionized the sound and practice of Kar-
natic music were all active during this period. Even more extraordinary is
the fact that while they were composing their masterpieces, a new instru-
ment was changing the sound, practice, and repertoire of Karnatic music:
the European violin. Brought to South India circa 1800 by British and
French colonial officials and visitors to the princely courts, the violin was
taken up and, shortly after its arrival, adapted by South Indian musicians,
who gradually altered its tuning, playing position, and technique.
This is no secret: although Karnatic music is unanimously described as
a vocal music, the violin is one of its most visible and audible elements,
found on almost every concert stage playing solo or doubling the vocal-
ist’s line. In twentieth-century South India, the violin not only became
a vehicle for conveying Karnatic music to modernity but also came to be
seen as essential to preserving Karnatic music’s authenticity. These notions
of modernity and authenticity, so crucially intertwined in the redefini-
tion of Karnatic music as ‘‘classical,’’ are played out in the central role that
the violin has had in constructing the voice of Karnatic classical music.
This chapter traces the career of the violin in South India from its arrival
around 1800 to the present. How did a colonial instrument ‘‘gone native’’
come to be heard as the source of an authentic Karnatic voice?
 In its early days in South India the violin was an instrument for Scottish
jigs and reels, French dancing songs, and English marching tunes, rather
than an instrument of Western classical music. In the beginning of the
twentieth century, however, attitudes toward the violin changed, the old
‘‘fiddle’’ image was rejected in favor of a classicized ‘‘violin’’ whose counter-
part was the classical violin of the West. The newly classicized violin was
central to a discourse about modernity: the violin was seen as the vehicle
for bringing Karnatic music into the modern age since it could repro-
duce the voice with more accuracy than the old-fashioned veena, with
more sensitivity than the ‘‘mechanical’’ harmonium, and with a kind of so-
cial impunity the sarangi had never had. The defining aspect of the violin
was its double character; it was Indian in its ability to reproduce the Kar-
natic voice, yet Western in its origins and form. These double capacities
made the violin the ideal instrument for twentieth-century musical ex-
periments: violin-and-piano, violin-and-Bach, violin-and-computer, and
multi-violin concerts that enact and rearticulate the relationship between
Karnatic music and the West.

Articulations
Descriptions of Indian music by European scholars in the nineteenth cen-
tury and early twentieth century were primarily concerned with com-
paring Indian music with the music of the West. Such studies inevi-
tably placed Indian music in a cultural half-state, as though waiting to be
awakened. Though Indian music was relentlessly compared to the music
of the West, the separation, usually opposition, between the two was
scrupulously maintained. Indian music had to take its place in a historico-
musicological tree of natural connections, but it was also to stand on equal
footing with Western music as one of the great musics of the world. In
contrast to such a vision, the violin and its passage to India represented the
possibility that instead of separation and opposition, there could be mix-
ing and influence. Western scholars of comparative organology in the late
nineteenth century showed little interest in the violin’s increasing popu-
larity in India precisely because it threatened to undermine the neat family
trees they were determined to draw to illustrate the relationship between
Indian music and European music.
In his Researches into the Early History of the Violin Family Carl Engel’s
genealogy of the European violin depended on ‘‘an acquaintance with the
musical instruments of foreign nations in different stages of civilization,

26 gone native?
extant at the present day’’ (1883, 3). With the European violin as the pin-
nacle, or endpoint, of ‘‘civilization,’’ the ‘‘rude fiddles’’ in evidence around
the world were to be taken as stages of development whose natural con-
nections to each other could be traced. Following the evolutionary theory
of the time, such ‘‘natural’’ connections were assumed to be based on func-
tion, not form. Engel hinted that ancient migrations were the cause of
diffusion of fiddle types but considered present-day migrations unimpor-
tant, as admitting of possibly unnatural connections that gave rise to inau-
thentic imitations of the violin. ‘‘Shapes resembling the f-holes are occa-
sionally to be met with on Asiatic instruments. . . . [T]he two sound-holes
are sometimes merely painted upon them, without their being pierced
into the sound board. What can be the origin and object of this fancy?’’
(19). Engel likewise mentioned the use of the European violin in India at
length but then dismissed it as inconsequential.

Nay, it is well known that in some of the seaport towns of Hindu-

stan the European violin has actually been introduced although it does
not appear to have obtained much popular favour. The rajas seem not
to appreciate its really commendable qualities, to judge from the fact
of their having ordered violins to be manufactured of silver instead
of wood. . . . We have seen that the Hindus are not entirely unac-
quainted with the European violin. . . . There is now in Calcutta a
Musical Academy, founded in the year 1871 by the Raja Sourindro Mo-
hun Tagore, in which this instrument is actually taught, so that ere long
we may perhaps expect in our public concerts a Hindu virtuoso aston-
ishing his auditory with the performance of some of our brilliant violin
compositions. However this may be, it can hardly be said that this Euro-
pean introduction has affected in the slightest degree the spirit of the
Hindu national music. (16–17)

Engel thus dismissed the use of the violin in India as a mere superfi-
ciality, just like the f-holes painted onto ‘‘Asiatic’’ instruments. Such a dis-
missal hardly squares with the fact that the violin was, according to Charles
Day in Music and Musical Instruments of Southern India and the Deccan (1891),
rapidly taking the place of the sarangi in South India.1 Strangely, however,
this is the only mention of the violin in Day’s entire work. His large sec-
tion of illustrations and descriptions of South Indian instruments includes
lengthy descriptions of the veena and its tuning but omits the violin en-
tirely. The sarangi is described as the main accompaniment to the voice,
yet the illustration of ‘‘a musical party’’ depicts a violinist, not a sarangi

gone native? 27
player (Day 1891, 98). It seems that for both Engel and Day, the present-
day music of the violin seemed to threaten their ability to hear ‘‘echoes
of an indigenous music . . . remaining in the Indian music of today; but
not yet so clearly heard that we can say we identify here or there a refrain
of an original or pre-historic music, although we may unconsciously be
very near it’’ (ibid., x).
These accounts were pervaded by a longing for a pure Indian music
unaffected by colonialism, a music impervious to its contemporary cir-
cumstances. Such a vision of Indian music continues to dominate West-
ern scholarship even at the end of the twenty-first century. For instance,
Gerry Farrell’s Indian Music and the West (1997), a book about the West’s
encounter with Indian music in the nineteenth and twentieth centuries,
also never mentions the violin. Farrell’s approach seems to invest in an
idea of ‘‘Indian music’’ as an essentially stable field free from the unstable
realm of politics and the whims of ‘‘the West.’’ Accordingly, his analysis
allows only for the coming together of spectacularly different elements:
the use of ‘‘Indian’’ motifs in Western parlor songs for piano; the use of
the sitar in Western pop music in the 1960s. Such examples are certainly
worth discussing, but Farrell’s alignment of them seems to suggest that the
story of Indian music and the West can be told completely in terms of such
spectacular misfits, of misguided Western appropriations of a pre-existent
and independent entity known as Indian music. If, as Farrell states, ‘‘at the
very core of Western attitudes toward Indian music is the idea that it is in
some way deeply unknowable,’’ then his own position unwittingly seems
to fall within such attitudes (9). His approach preserves a pristine place
for Indian music as an entity that exists before and after two centuries of
‘‘misunderstanding’’ on the part of the West, an entity that went on ‘‘de-
veloping and adapting as it would, largely impervious’’ to the debates and
appropriations to which it was subjected (54).
The career of the violin in India surely does not fit into a history that
would prefer to leave Indian music untouched by the West or by colo-
nialism. Following the violin in South India points to a different history,
one that admits the centrality of the colonial encounter in the creation of
what is now called ‘‘Indian classical music.’’ Such a history does not amount
to a narrative of Westernization, a story other ethnomusicologists have
told.2 The Westernization narrative implies that before the arrival of West-
erners, Karnatic music existed as much the same kind of entity we know
now, and that it was then selectively affected by Western musical practices
and ideas. This narrative suggests that definitions of ‘‘Indian’’ and ‘‘West-

28 gone native?
ern’’ are unproblematic, and that the Western influences could be stripped
away, leaving a pure core of Karnatic music. Certainly, the colonial en-
counter changed musical practices and discourses about music in South
India. These practices and discourses are not Western, however, but spe-
cifically colonial. They position Karnatic music as India’s classical music,
definable and audible as such precisely because it is poised in opposition
to the classical music of the West. Pervading this peculiarly colonial dis-
course is a structure of comparison that, as I have suggested, includes both
a claim to commensurability and a claim to essential difference.

The King of Instruments

The first violins in India were probably brought to Calcutta in the 1760s
for the Calcutta Harmonic Society, which played Handel, Corelli, and later
Haydn and Mozart, and for the Catch Club, which played jigs and reels
for dancing (Head 1985, 549). Musical instrument shops opened in Calcutta
in the 1780s, and the Calcutta Band, an orchestra, was established in 1785
(551).3 Yet although there must have been interaction between English and
Indian musicians, the violin did not seem to play a major role. By contrast,
in South India the violin’s influence seems to have been immediate.
Exactly where the violin first came to South India and who first adapted
it to Karnatic music are matters of some contention. One popular story
holds that the violin appeared in Madras in the late eighteenth cen-
tury, probably introduced by colonial officials with musical hobbies. Balu-
swamy Dikshitar (1786–1858), who was from a Brahmin family and the
younger brother of the composer Muthuswamy Dikshitar, moved with
his father to Manali, a village near Madras, around 1800 or slightly be-
fore. Attracted by the English band music at Fort Saint George in Madras,
Baluswamy expressed a desire to learn European music. Manali Chinniah
Mudaliyar, son of the dubash to the governor of Madras, trustee of the
Manali temple, and patron of many musicians at that time, engaged a
European violinist from Madras to give Baluswamy violin lessons (Ragha-
van 1944, 129). He had lessons for three years and also managed in that time
to adapt the violin to Karnatic music, although there is no explicit mention
of this in P. Sambamoorthy’s account (1955a, vol. 1, 37). Baluswamy then
returned to Tiruvarur, near the thriving royal court of Tanjavur, known as
the principal seat of Karnatic music before Madras gained that reputation.
The violin became known in South India through Baluswamy’s perfor-
mances. Indeed, the maharaja of Ettayapuram, a small state south of Madu-

gone native? 29
rai, was so taken with Baluswamy’s violin music that he appointed him
court violinist (samasthana vidwan) in 1824 (ibid.). Baluswamy’s brother,
Muthuswamy, later came from Tanjavur to Ettayapuram, and there had the
chance to hear the violin and the European music his brother played on it.
Another popular story has it that the site of the violin’s first transfor-
mation was the royal court of Tanjavur, where Maharaja Serfoji had ap-
pointed the Tanjore quartet—four brothers from what is now called the
icai veḷḷālar community—as court musicians and dancers (Subrahmanyam
1980, 47).4 Vadivelu (1810–1868), the youngest of the brothers, was a dis-
ciple of Muthuswamy Dikshitar and also studied Western violin with the
missionary Christian Friedrich Schwartz, who had established a Protes-
tant mission in Tanjavur in 1778 and developed a friendly relationship with
the royal court (Seetha 1981, 103). Vadivelu later demonstrated the tunes
he had learned from Schwartz to Muthuswamy Dikshitar, who composed
songs based on them. Vadivelu also introduced the violin to the composer
Thyagaraja, accompanying him while he was singing; sometimes Thyaga-
raja would ask just to hear the violin by itself (Subrahmanyam 1980, 48).
When Maharaja Serfoji and the brothers had a disagreement over their
rights in the temple at Tanjavur, the quartet was dismissed from Serfoji’s
court, but it was not long before they found another post in the court of
the composer-king Swati Tirunal at Trivandrum in 1830. Vadivelu became
an intimate associate of the king and taught violin to several court musi-
cians there. In 1834 Swati Tirunal presented Vadivelu with an ivory violin
inscribed with the eagle, emblem of the Trivandrum royal court, and a
bow made from an elephant’s tusk (ibid., 49).
A third version of the violin’s entry into South India concerns the min-
ister Varahappayyar (1795–1869), who worked for Maharaja Serfoji as the
superintendent of all court musicians (Seetha 1981, 258). He eventually be-
came Serfoji’s trusted minister and when Serfoji wanted to enter negotia-
tions with the governor of the Madras presidency, he sent Varahappayyar,
who was noted for his proficiency in the English language ( Jayarama Iyer
1985, 27). A musical negotiation seems to have resulted: the governor was
hospitable to Varahappayyar and showed him his music room, which had
violins and a piano. Intrigued by these instruments, Varahappayyar played
Indian melodies on them, impressing the Western musicians, who then
agreed to teach him a few violin techniques. ‘‘This news reached the ears
of the governor . . . [who] being very music-minded asked him to play
before him. In order to please him and have his political mission fulfilled,
he agreed to play before the governor (otherwise he would have preferred

30 gone native?
to play it to his Maharaja first). He took the violin and played some Indian
melodies to the pleasure of the governor’’ (ibid.). The governor presented
the violin and a piano to Varahappayyar, who returned to Tanjavur and im-
pressed the maharaja. Varahappayyar later taught violin to Vadivelu, who
was the first to fully realize the potential of the violin, and more violins
were then brought to Tanjavur.
While the three preceding versions of the violin’s history in India could
perhaps be correlated, yet another version places the arrival of the violin
somewhat earlier and offers an alternative narration of its adaptation into
Karnatic music. A musicologist in Mysore, to whom I had mentioned my
interest in the violin’s history in South India, commented that I had prob-
ably only heard the Tanjavur stories but that there was evidence—a woman
pictured playing the violin in a dance party in a mural—to show that the
violin had come to the Mysore area much earlier. The mural is located in
the summer palace of Tipu Sultan, who ruled the royal state of Mysore in
the last quarter of the eighteenth century. The musicologist R. Sathyana-
rayana dates this mural to 1784, presenting linguistic evidence to show the
incorporation of the violin into the Kannada language: ‘‘Fiddle—Fittle—
Pittle—Pitil—Piteel—Piteelu’’ (1993, 11).5 The mural, he states, points to
the fact that the violin was already in accepted usage in Mysore state by
the time Varahappayyar, Vadivelu, and Baluswamy Dikshitar were born
and thus was probably introduced by French army officers. The evolution
of a distinctive school of violin playing in Mysore, states Sathyanarayana,
points to the long tradition of violin playing there (12).
With this the written sources on the violin’s entry into South India end,
so understanding what this early violin sounded like, who played it, and
how it spread requires a certain degree of speculation. From its earliest
use in South India the violin seems to have been played by both Brahmins
and non-Brahmins, traveling through political rather than caste alliances.
In fact, in the case of Varahappayyar, the violin was used as a political bar-
gaining chip: he played Indian melodies on the violin to demonstrate his
diplomacy. Vadivelu’s story shows how the colonial machineries of gov-
ernment, reflected in the policies of princely states toward their musician-
officials, affected the movement of the violin around South India. The
violin moved from Tanjavur to Ettayapuram to Trivandrum in the hands
of court musicians seeking to improve their status, as it helped maharajas
improve theirs; the ivory violin presented by Swati Tirunal was probably
hailed then, as it is now, as a sign of his progressive and forward-looking
stance (Venkitasubramonia Iyer 1975, 3, 159).

gone native? 31
 The violin traveled through other colonial machineries, such as the
military and railroads. The grandfather and father of the well-known
early-twentieth-century violinist Dwaram Venkataswamy Naidu (1893–
1964) were both military men who played violin as a hobby and probably
picked it up from English officials; Naidu’s brother, the elder by approxi-
mately thirteen years, worked for the railroads in Vishakapatnam and used
to acquire violins in his travels. The amateur violinist C. Subrahmanya
Ayyar (1885–1960) used his postings as chief railway accountant in various
parts of India between 1910 and 1940 to buy violins and to learn about the
state of violin playing in India in general (Subrahmanya Ayyar 1944).
And what of the music played by the first Indian players of violin? The
Western music to which Indians were exposed in South India was pre-
dominantly Irish and Scottish fiddling, rather than Western classical violin.
Evidence for the idea that South Indians first heard reels and gigs rather
than concertos and sonatas comes from what could be one of the first
Indian violin experiments: Muthuswamy Dikshitar’s compositions based
on European melodies. In a volume of these ‘‘European Airs’’ culled from
various sources, more than forty of Dikshitar’s compositions are listed
(and notated) with the names of the European songs—such as ‘‘Limerick,’’
‘‘Castilian Maid,’’ ‘‘Lord MacDonald’s Reel,’’ ‘‘Voulez-vous Danser?,’’ and
‘‘God Save the Queen’’—on which they are based (Sankaramurthy 1990,
xiv). It is probable that Muthuswamy Dikshitar heard many of these melo-
dies played on the violin by his brother or other court violinists during his
travels. What is remarkable about Dikshitar’s compositions is that there
is none of the melisma or gamaka (ornamentation) associated with Kar-
natic music, and the lyrics seem to be only a way of translating these West-
ern tunes into a recognizable, palatable form.6 Perhaps more important,
the process of composition implied here was radically different from the
Karnatic ideal of a singer-composer composing words and music simulta-
neously. Here the composer took melodies already formed, pure instru-
mental music, and added the words/voice as the music demanded. These
compositions thus suggest two things: first, the original associations of
the violin in South India were with fiddlers, not with European classical
music; second, the violin was not only a physical sign of the colonial pres-
ence but also the vehicle for a kind of translation of Western music into
Karnatic music.
Other Indian composers show the influence of European tunes in their
work as well. Swati Tirunal (1813–1847), the patron and close friend of the
violinist Vadivelu, also appointed Western musicians to his court. His var-

32 gone native?
nam (a type of composition roughly equivalent to an etude) in the raga
shankarabharanam, whose pitch intervals are equivalent to those of the
Western major scale, includes a passage at the end that sounds distinctly
like a European marching band. Unlike the other sections of the varnam,
the last section breaks all the conventions of Karnatic music by employing
large intervallic leaps and a minimum of gamaka. The principle of the var-
nam as a genre is to exhaust the possibilities of a raga with different permu-
tations of phrases; it is clear from Swati Tirunal’s varnam that in the early
nineteenth century the possibilities of a raga included its transformation
into a piece of semi-Western music. Along these lines, several composi-
tions of Thyagaraja suggest strains of Scottish reels or waltzes, composed
in unique ragas with suggestive names like ‘‘jingla.’’ Pattnam Subramania
Iyer (ca. 1840–1910), a disciple of Thyagaraja, composed his well-known
‘‘Raghuvamsa Sudha’’ in the raga kathanakuthuhoolam, which he had in-
vented using a modified version of the Western major scale and that speci-
fies large intervals and almost no gamaka. This composition and a variety
of other ‘‘English notes’’ or ‘‘Western notes,’’ as they are called, are staple
fare at the end of any instrumental concert of Karnatic music to this day.

A Double to the Voice

In the mid-nineteenth century the violin came to be appreciated for its
ability to mimic the singing voice. Toward the end of his hagiographical
account of the life of the composer Thyagaraja, Wallajapet Ramaswamy
Bhagavatar recalls a scene of ‘‘surprise’’ and ‘‘delight’’ from the mid-1840s.
His grandfather, the composer Wallajapet Venkataramana Bhagavatar, and
his father, Krishnaswamy, had been direct disciples of Thyagaraja. Krish-
naswamy, who from childhood had always had less vocal facility (cārīra
cukam), instead played the ‘‘kinnari-fiddle.’’ 7 According to the account,
Krishnaswamy played the kinnari ‘‘very skillfully’’ (mikavum catūriyamāka)
while others—such as the vidwans (musicians) Govinda Marar, Tillaistanam
Ramayyangar, Krishnaswamy’s father, and even Thyagaraja himself—sang.
Thyagaraja is reputed to have ‘‘attained bliss’’ on hearing the kinnari and
to have said, by way of blessing, ‘‘Child! Krishnaswamy! If you keep on
singing and playing the kinnari at my side, Sri Sita Ramacandra will give
you a divine voice [tivyamāa cārīram].’’ Krishnaswamy immediately made
a namaskaram (bow) at the feet of Thyagaraja and sang a kriti (composi-
tion) with him while playing the kinnari. ‘‘When the pallavi [first section]
was over,’’ recalls Ramaswamy, ‘‘the anupallavi [second section] began and

gone native? 33
the voice and kinnari met with a single sound [orē vitamāa nātattuṭa
kuṭi], so very pleasing that those in the audience were surprised and de-
lighted. . . . When he was singing you couldn’t tell whether it was the
sound of the fiddle or the sound of the voice [ivar pāṭum samayattil piṭil
nātamō cārīra nātamō eu kaṇṭaiya muṭiyātu]’’ (Ramaswamy Bhagavatar
1935, 53–54).
The idea of the violin as the perfect accompaniment to the voice be-
came prevalent in the twentieth century, when the social context of Kar-
natic music shifted to the concert hall.8 Without amplification, the vol-
ume of the voice was no longer adequate for audiences of more than a
handful of people in the large, noisy spaces of Madras. Some kind of ac-
companiment was needed to make the voice audible above the din of the
city without jeopardizing its delicacy; the violin, already widely in use in
South India, seemed a natural solution. The musicologist S. K. Ramachan-
dra Rao states that the first violinist to accompany vocalists in concerts was
Tirukkodikaval Krishna Iyer (1857–1913), followed by a number of other
violinists who lived into the first half of the twentieth century (1994, 5).
Once it became necessary for vocalists to be accompanied, violinists were
in demand and thus had more concert opportunities in what was fast be-
coming a lucrative business. A list of South Indian musicians, compiled in
1917, included more than 100 violinists, many of whom were described as
accompanying artists (Pandithar 1917, 159).
The practice and aesthetic of the instrumentally accompanied voice is
pervasive in South Asian musical traditions. The accompanying instru-
ment, however, does more than simply repeat what the soloist sings; not
only does it impart a particular sound quality, but the relationship, both
musical and social, between soloist and accompanist stages the voice in a
particular way. In much the same way as a quoted voice can potentially
disrupt the authority or primacy of a quoting voice, as Mikhail Bakhtin
suggests in relation to speech and literary genres, the violin, in accom-
panying, can influence the soloist’s voice.9 What, we may therefore ask,
was the effect of putting an instrument with colonial origins in this role?
How did the violin change and/or create the voice it was merely supposed
to double?
Accompaniment in Karnatic music involves a mixture of support and
competition, imitation and creativity, shape-giving and self-effacement.10
P. Sambamoorthy, known as one of the great modernizers of Karnatic
music,11 described the violinist’s duties at length in his 1952 article ‘‘Kacceri
Dharma’’ (Concert etiquette): ‘‘It is his duty to figurate the music of prin-

34 gone native?
cipal performer . . . by giving judicious emphasis on sangatis and gama-
kas. . . . He should not be hasty in deciphering rare ragas and eduppus
[eṭuppu]. He should remember that rare ragas and intricate eduppus are
traps set for him by his (not very friendly) chief to catch him unawares. . . .
His responses to his chief ’s alapana and kalpana swaras should . . . not run
counter to the train of musical thoughts of his chief (1952, 269; italics in
original).12
What is involved in the idea of ‘‘figuration’’? It indicates precisely
the tension between visibility and invisibility, audibility and inaudibility,
sameness and difference.13 At stake is not representation of the voice but a
kind of partial repetition. The violinist plays something that is not identi-
cal to what the vocalist sings but is close enough to give the vocalist’s sound
a greater presence, a recognizable form. In contemporary musical practice,
the role of the violinist as accompanist is to ‘‘double’’ whatever the vocal-
ist sings, either in unison or, during improvisations, with a slight delay.
Although Sambamoorthy insisted on the subordination of the violin to
the vocal line, in reality an accompanying violinist has great power over a
soloist. For instance, at any point in a section of raga alapana, a free-time
improvised elaboration of the raga, the violinist can choose either to dwell
on a single note, reproduce the entire phrase that the vocalist has sung,
highlight only the end of the phrase, or play something that the vocalist
has not sung at all. What the violinist chooses to play will often determine
what the vocalist sings next. Indeed, the best accompanists, one vocalist
told me in 1998, make the soloist sound better by covering up ‘‘imperfec-
tions’’ and by giving the vocalist ideas about what to sing next. Vocalists
often talk of accompanists as inspiring, but also complain that they domi-
nate soloists.
A skillful accompanist is an expert in mimesis and thus also in dissimu-
lation. My violin teacher, who had worked for twenty-five years as a staff
violinist for All India Radio, used English words relating to technologi-
cal reproduction to describe her role as accompanist. One had to be ready
for whatever the vocalist sang and get it ‘‘typed’’ in one’s brain after the
first line, like a ‘‘recording.’’ Often one had to accompany compositions
one had never heard before; in these cases, it was a matter of knowing
the raga and its particular phrases, keeping track of the tala, and being on
the ready—one had to ‘‘adjust’’ (the English term Indian musicians often
use for faking or fudging something) so that one blended in but sounded
confident at the same time. One did not have to know the words of the
composition or even remember it afterward, as long as one could imitate

gone native? 35
realistically enough. Another violinist told me that a certain vocalist pre-
ferred her accompaniment because it blended so well that she forgot there
was anybody accompanying her at all.
The delicacy and fragility of this accompanying operation stands in
contrast to its importance; without proper accompaniment, even the best
singer sounds incomplete. If the accompaniment is off-kilter, a concert lis-
tener once told me, the whole thing is ruined. We were at a violin concert
in which the solo violinist was accompanied by his daughter and disciple.
The daughter, instead of picking up a few phrases here and there, was re-
peating everything the father played in its entirety; the father would stop
playing and wait for her to echo him. My neighbor seemed impatient and
made no attempt to hide the fact that he was looking at my notes, looking
for anything that would relieve his boredom. Although audiences some-
times talk during compositions, they are usually quiet during raga alapana,
so I was surprised to find him telling me, in a stage whisper, how ‘‘child-
ish’’ the performance sounded, as if it had all been staged beforehand. In
accompaniment, one can’t play too much, or the secret of what is going
on becomes obvious and uninteresting. On the other hand, the violinist
is there partly to provide intonation. Many vocalists, particularly when
improvising at fast speed, lose their intonation and simply belt out the
swara (musical note) names instead of singing them on pitch. The audi-
ence gets the impression that those pitches were sung partly because they
hear the swara names, but also because the violinist plays the pitches the
singer names.
The fine line between supporting and overshadowing suggests that ac-
companiment is a social as well as a musical matter. My violin teacher, who
had accompanied the famous flautist Mahalingam many times, prized her-
self on the subtlety of her accompaniment. Once, Mahalingam, known for
both his genius and for his drunkenness on stage, had had another violin-
ist accompany him. On that occasion, Mali, as he was known, started a
complicated rhythmic ending in the wrong place and had to abort it in the
middle. The violinist, however, instead of covering up the mistake, pro-
ceeded to start the ending from the right place and finish the piece, which
incensed Mali. Several months later, in order to get his revenge, Mali ar-
ranged a concert in which that violinist and my teacher both accompanied
him. After a few warm-up pieces, the acid test came: an improvised raga-
malika (combination of ragas) in which the violinists take turns accom-
panying. The first violinist played as usual, demonstrating her skill in that
particular raga. But when, after Mali played some more, my teacher’s turn

36 gone native?
came, she simply laid down her violin, turned her palms upward, and said
into the microphone, ‘‘There is nothing I can play after such divine music,’’
rousing the audience to cheers—and exemplifying how an accompanist
can literally stage, in the sense of making or breaking, a soloist. It is per-
haps for this reason that T. K. Jayarama Iyer wrote, ‘‘The Violin, as it is, is
a wonderful instrument. If we call our Vina the queen of musical instru-
ments, we can call the Violin the king of musical instruments. No other
instrument is so powerful as the Violin’’ (1965, 28).

From ‘‘Fiddle’’ to ‘‘Violin’’

When the violin began to appear regularly on the concert stage in the
early twentieth century, its old ‘‘fiddle’’ image and sound were no longer
adequate. The violinist had to be a classical musician in his own right,
distinguished from those violinists who played dance music or harikatha
(musical storytelling, usually on religious themes). For those who wrote
about the violin in the early twentieth century, an overhaul of violin tech-
nique seemed necessary to rein in the wild gropings of Karnatic violinists
on their fingerboards, to update their antiquated use of the bow. Hardly a
century after the violin had come from Europe, these critics wrote, Kar-
natic violinists seemed to have forgotten that the West had a developed
system of violin technique; violinists in India, ignoring this, seemed to
be starting over again, literally, from scratch. While the ostensible sub-
ject of these writings was the violin, the reformation of violin technique
was seen as the key to reforming and modernizing Karnatic music in
general. Although the suggestions (especially those concerning posture)
made by these writers were never implemented by Karnatic violinists,
they illuminate the contours of debate about music in the early twenti-
eth century, particularly the relationship between ‘‘Indian’’ and ‘‘Western’’
in these debates. The idea that emerges from these writings is that only
Western violin technique can bring forth the true beauty and identity of
Indian music.
In 1912 T. C. R. Johannes, a Christian violinist who had studied under
the violinist ‘‘Photograph’’ Masilamani Mudaliyar (so named because he
was the first musician in those days to have purchased a camera), published
his Tamil treatise Bhārata Sangīta Svāya Bōdhini ( pārata caṅkīta svāya pōtii:
Indian music self-instructor). Johannes devoted a large part of the treatise
to violin technique. In his hortatory introduction of long-winded, spo-
radically Sanskritized Tamil, he railed against the tendencies of Indians

gone native? 37
(whom he addressed with biting sarcasm as ‘‘swadeshis’’ 14) to adopt every-
thing European, to speak in English, and to disdain Indian music for its
lack of harmony (1912, xxii). These so-called swadeshis, he wrote, knew
nothing about the European music they so blindly accepted other than
the word harmony itself, which they repeated as if showing off a new and
expensive possession (‘‘Iṅta ārmoi eum vāttaiyai māttiram īs pīs piṭik-
kumpōtu vilaikkoṭuttu vāṅkiārkaḷ pōlum’’).15 Even North Indians did
not accept everything European as slavishly as did South Indians. Instru-
ments like the guitar, banjo, mandolin, harmonium, and piano were ‘‘use-
less’’ and ‘‘without pride.’’ For Johannes, it was precisely the visibility and
rationality of the piano that rendered it unmusical and presumptuously
vulgar: the piano’s keys were laid out so that you could go up and down
them in five minutes, he claimed, whereas in Karnatic music it ought to
take three days and three nights to finish a raga (xxii).
Surprisingly, the violin escaped Johannes’s wrath; in fact, he argued that
while Indians should reject these other ‘‘foreign’’ instruments, they should
accept the violin. ‘‘It is called the king of instruments; it accommodates
itself naturally to the sound of any other instrument, to the human voice,
to animals [yāvattu karuvikali nātattaiyum, mauṣa, miruka . . . mutaliya
yāvattu iaṇkaḷi tōikaḷaiyum itu iyalpilēyē aucāraai ceyyum]. To say
it is a European instrument is an error. The Europeans did nothing other
than civilize it a bit [koñcam nākarīkamāy ceytirukkiārkaḷ]; it was origi-
nally an Indian instrument brought to the West in Alexander’s time’’ (xix).
In his first chapter, titled ‘‘Some Rare but Essential-to-know Things
about Bharata Sangitam,’’ Johannes devoted a large section to the proper
way of playing the violin. First, he gave advice concerning posture: the
usual Indian way of holding the violin with the scroll against the foot was
wrong and an insult to Saraswati, and it had been condemned by Euro-
peans; the violin should instead be held the European way (5). An illustra-
tion shows what seems to be an English violinist facing an Indian violinist,
each one holding the violin in European fashion. In addition, Johannes
wrote, it was a travesty that Indian violinists only used at best half the
length of the bow. Johannes claimed that, if asked why they were playing
that way, most violinists would say the audience liked it, that the sound
was sweet. Furthermore, according to Johannes, their fingering was so un-
systematic that they didn’t even know what mistakes they were making
(5). They simply played the same ten or fifteen ragas over and over again,
without a clue as to the true notes of the raga or any theoretical knowledge
of music.

38 gone native?
1 Method of holding the fiddle and bow (piṭilaiyum, villaiyum piṭikkavēṇṭiya kramam) and
tuning fork (sruti cērkkuṅ karuvi). Drawing from T. C. R. Johannes, Bhāratha Sangīta Svāya Bō-
dhini (1912).

Johannes related an incident in which he heard another violinist playing

a concert. The audience had been shaking their heads in appreciation like
dolls (‘‘talaiyāṭṭi patumaikaḷai pōl’’). The man of the house where the con-
cert was taking place asked Johannes how he liked the violinist. Johannes
replied, ‘‘He is working very hard and so are his fingers. But he is play-
ing in contradiction [nērvirōtamāy] to the sastras.’’ The house owner was
affronted and said, ‘‘Who are you to insult him like this?’’ Johannes said,
‘‘I will show you where he is wrong.’’ He sidled up to the violinist after-
ward and made a comment about the violinist’s rendition of a particular
raga. The violinist replied, ‘‘Oh, but if I didn’t play it that way, the audi-
ence wouldn’t have appreciated it.’’ ‘‘Very well,’’ said Johannes. ‘‘Then can
you play all the ragas this way?’’ The violinist said, stroking his mustache
as if ready for a fight, ‘‘Of course! What objection could I have to that?’’
Immediately Johannes took a slate board and wrote out the murchana (as-
cending and descending pattern) of a raga. The violinist was dumbstruck.
He asked Johannes if he could play it. So as to put to rest any suspicion
that he might have already practiced that particular raga, Johannes asked

gone native? 39
the violinist to write another, knowing full well he would not be able
to. Finally, Johannes played the raga he himself had written and emerged
victorious (6).
So, besides an appalling lack of knowledge of theory and notation, the
violinist had no knowledge of technique either; indeed, knowledge of
theory, notation, and technique are tellingly conflated in Johannes’s ac-
count. The violinist, resting his violin on his foot and holding the bow in
the middle (already a recipe for disaster), placed his fingers in an acciden-
tal approximation of the raga’s notes. A thorough study of the violin was
necessary to counteract this. Johannes recommended that the beginning
violinist paste paper on the fingerboard so that he might see where the
swaras fell. He even provided the measurement of the distance between
each of the fingers as they should be placed on the fingerboard (11). The
playing of beginning exercises like sarali varisai (different patterns of notes)
and alankarams (different patterns of notes set to different talas to be played
in various ragas), Johannes maintained, was useless unless the teacher told
the student what fingering to use (13). As for the bow, Johannes recom-
mended practicing with the metronome, long bows 128 counts in dura-
tion, and he remarked gleefully on how amusing it would be to watch the
violinists who held their bow in the middle trying to do this (13).
The Art and Technique of Violin Play, written by C. Subrahmanya Ayyar,
appeared in 1941. Having studied the violin in Madras, Subrahmanya Ayyar
gave his first public recital in 1922 at Presidency College, in conjunction
with a lecture on the acoustics of the violin given by his brother, the physi-
cist C. V. Raman (Subrahmanya Ayyar 1944, 1). Although Subrahmanya
Ayyar remained an amateur, he devoted all his spare time to practicing
and lecturing on the violin in various parts of India and worked intermit-
tently on a project concerning the sound waves emitted by the violin. He
took time off in 1933 to make a European tour on which he was primarily
concerned with meeting European violinists, playing Karnatic violin for
them (‘‘I felt . . . [it] might serve as a harbinger of good-will and mutual
understanding between the East and the West’’), and buying violins (28).
The Art and Technique of Violin Play was originally given as a series of
lectures for the students of the Indian music department at Madras Univer-
sity. As one gentleman remarked, ‘‘No violinist of South India today can
satisfy by his technique even a trained European ear, nor make any deep
impression on a Western violinist worth the name’’ (Subrahmanya Ayyar
1941, 2). Subrahmanya Ayyar’s goal was to correct this state of affairs by
helping the young generation of violinists improve their violin tone. His

40 gone native?
first recommendation was to get a European violin, for the Indian factory-
made violins were of much inferior quality. Innovations such as Mysore
Chowdiah’s seven-stringed violin were to be abhorred for their ‘‘metallic
screech’’ (4). The South Indians’ preference for thin strings and their use
of extra-heavy bows contributed to their poor tone (7). Like Johannes,
Subrahmanya Ayyar criticized the posture of South Indian players, imply-
ing that squatting on the floor and playing was unnatural to the violin. ‘‘It
is a pitiable sight to see the cotton towel of the South Indian artist, with
such a mass of it touching the violin to avoid the sweat of his body wetting
it, which further damps the tone. I wonder if even the next generation
will give up this posture’’ (10). The ‘‘cramped posture’’ of South Indian
violinists also prevented them from reaching the higher positions of the
violin and thus from achieving the technological heights of perfection of
Western violinists (10). Even the South Indian violinists’ habit of using
oil (continued to this day) did not escape Subrahmanya Ayyar’s reforming
scrutiny: ‘‘As if all his attempts to damp the full violin tone are not suf-
ficient, the violinist occasionally rubs fine oil on his left finger tips while
playing, to secure easy movement or slipping of fingers. What a mockery
of violin play!’’ (11).
For vocalists, the aim was merely to get the right ‘‘effects.’’ But for in-
strumentalists, there was a visible, and therefore systematic, technique (23).
The veena, which served in Subramanya Ayyar’s scheme as an inspiration
to the voice, was different from the violin, which had succeeded almost
entirely in reproducing the voice. But in their fascination with reproduc-
tion, South Indian violinists had forgotten the violin’s original tone (15).
Violinists must, he urged, strive for a continuity of sound, a technique
of bowing which avoided the ‘‘belching sound’’ even advanced violinists
made when changing the direction of their bow. Likewise, South Indian
violinists must learn to conceptualize their fingering in terms of the posi-
tions in Western violin technique; only then would they get truly accurate
intonation and be able to realize the ‘‘subtle microtonal changes’’ (19). His
advice for young violinists was to play the sarali varisai and alankaram ex-
ercises stripped of all their gamakas and to effect the switch between posi-
tions, like a European violinist, as inaudibly as possible (15). Furthermore,
they should play with all four fingers, avoiding the excessive sliding pro-
duced by the two-finger or single-finger techniques (31). And finally, once
the young violinist had gotten all these elements under control, he had to
work on his facial expression: ‘‘There is no need to look at the strings or
the violin while playing as you become mature, but look at the audience.

gone native? 41
The facial expression of the artist must be one of ease in bearing, with a
graceful look. There should be a general relaxation of the limbs and an ab-
sence of stiffness . . . while there is perfect control over the facial muscles
and a composure in the face of the artist’’ (34).
Getting a full tone out of the violin—putting South Indian violin on a
par with the classical violin playing of the West—thus involved a new and
rigorous discipline of the performer’s body. Sliding and fingering entail
different relationships to the violin as well as the music produced from it.
Sliding along the fingerboard is conducive to understanding a melody as a
continuous phrase, with no sharp distinction made between the notes and
the space ‘‘between’’ the notes. Fingering entails recognizing individual
pitches and their locations on the fingerboard; it means recognizing a stan-
dardized technical discourse on ‘‘positions’’ in Western violin technique.
Knowledge of where to place one’s fingers also includes the knowledge
that the space between the notes is essentially forbidden. While the ‘‘ex-
cessive sliding’’ and ‘‘cramped posture’’ of Karnatic violinists was indica-
tive of laziness, ignorance, and secrecy, the technique of discrete fingering
identified with the West was associated with the advances of modernity:
discipline, clarity, and theoretical knowledge of music.
In the 1940s the Madras Music Academy confirmed the classical status
of the instrument by decreeing that its name be changed from ‘‘fiddle’’
to ‘‘violin’’ in all published programs. Others suggested further refining
the name, replacing the Tamil pronunciation of fiddle, ‘‘piṭil,’’ with lofty-
sounding Sanskritized names like ‘‘bahuleen’’ (Bullard 1998, 255). An essay
in the Madras-based magazine Triveni, self-proclaimed ‘‘Journal of the
Indian Renaissance,’’ suggested the name ‘‘vayuleena,’’ to rhyme with
‘‘veena.’’ The violin was, the essay claimed, even more suited to Karnatic
music than it was to Western music, and its origin lay in the ‘‘genius of the
Aryans’’ (Harinagabhushanam 1929, 202).

The Violin as a Sign of the Modern

In the early to mid-twentieth century the violin came to be explicitly asso-
ciated with modernity. It was hailed as a necessary accompaniment to the
voice, making it both clearer and more audible in the concert hall. There
was something about the violin—its Westernness and newness but also its
uncanny ability to imitate the Karnatic voice—that made it flexible and
resilient enough to withstand various experiments. Like recording tech-
nology, the violin provided a way of reproducing the voice, a new way

42 gone native?
of representing Karnatic music. But the violin reproduced the voice with
a difference, making the human ear into a more precise listening instru-
ment. In 1944 C. Subrahmanya Ayyar wrote,

The advent of the violin to South India from 1800 has made Carnatic
music even more articulate than in the days of Tulajajee of Tanjore. . . .
It has made the South Indian ear perceptive to microtones greater than
twenty-two. . . . The sanity of Tamil genius is also seen in the fact that
the violin was accepted as an accompaniment by the musical elite in
South India . . . instead of the harmonium. The latter entered the por-
tals of the All-India Radio of North India . . . and was later tabooed, as
if by the command of a Dictator. (60)

The violin was thus associated not only with the progress of both music
and its listeners but also with emotional maturity: ‘‘From this wonder-
ful box we get a most ravishing sound, which affects most profoundly the
emotions of the most civilized’’ (Subrahmanya Ayyar 1941, 3). Here the
sign of a modern, ‘‘civilized’’ subject is precisely his or her ability to be
moved by the sound of the violin, to recognize the connection between
his or her voice and the sound emanating from that ‘‘wonderful box.’’
Between 1910 and 1930, several experiments were undertaken to in-
crease the volume and improve the tone of the violin. Beginning in 1918,
C. V. Raman, professor of physics and member of the Indian Association
for the Advancement of Science, engaged in research on the acoustical
principles of the violin family of instruments. His object in ‘‘Experiments
with Mechanically Played Violins’’ was to ‘‘throw light on the modus oper-
andi of the bow . . . [and] to furnish valuable information regarding the
instrument itself, its characteristics as a resonator, and the emission of
energy from it in various circumstances. . . . [It] could also be expected
to furnish illustrations of the physical laws underlying the technique of
the violinist and to put these laws on a precise quantitative basis’’ (1920,
20). The result of the experiment was a sixteen-page report detailing the
construction of the mechanical player and graphs showing the interrela-
tions between bowing speed, bowing pressure, pitch changes, and varia-
tions in the bowed region. Every care was taken to have the mechanical
player, built out of used parts of other machineries, simulate the playing
of a human violinist, even down to a leather-lined clamp to simulate the
violinist’s fingers (28).16 The mechanical player radically changed the ori-
entation of things by keeping the bow stationary and moving the violin
back and forth, for the sake of more reliable data (21). What the experi-

gone native? 43
2 Mechanical violin player designed by C. V. Raman. Photograph from Proceedings of the Indian
Association for the Cultivation of Science, vol. 6, following p. 20. Calcutta, 1920. Courtesy of Widener
Library, Harvard University.

ment provided was the chance for a repeatable, measurable motion to be

observed, without the distraction, as it were, of music. In fact, the report
never mentioned what kind of violinist’s playing was being simulated, al-
though from the horizontal position of the violin it would seem that West-
ern technique (as in the writings of Johannes and Subrahmanya Ayyar) was
taken as the model.
Pudukkottai Narayan Iyer and Marungapuri Gopalakrishna Iyer, violin-
ists active in the first half of the twentieth century, both used horn violins
so that the violin might be heard in the concert hall (Ramachandra Rao
1994, 23); in fact, one of the first violin recordings ever made in South India
was Narayan Iyer on his horn violin, circa 1920.17 The maharajas of Tan-
javur, Mysore, Trivandrum, and Vizianagaram all purchased horn violins
for their personal instrument collections. The famous Mysore violinist
T. Chowdiah, meanwhile, was worried that his violin would not be au-
dible over the strong voices of those whom he was accompanying. Add-
ing a microphone for the violin alone didn’t solve the problem, because it
caused a slight delay; if the violin was amplified, then everything had to
be amplified. Chowdiah instead decided to amplify his violin by making a
nineteen-stringed version: seven playing strings, with the upper three re-
inforced by a string an octave lower in pitch, as well as twelve sympathetic

44 gone native?
strings (45–46). Although this perceptibly increased the violin’s volume,
some objected to the ‘‘metallic screech’’ of this souped-up violin and to the
fact that gamakas were rendered more difficult and laborious by having
to be played on two strings simultaneously. Chowdiah himself eventually
gave up the sympathetic strings because of tuning problems but continued
to play on seven strings until his death in 1967 (75).
Not only did the violin have the ability to produce loud and clear tones;
it was also portable, simple, and flexible enough to meet the demands of
accompaniment. In 1955, P. Sambamoorthy listed the advantages of the
European violin.

1. Its sweet and loud tone.

2. Its handiness.
3. Its plain fingerboard enabling the performer to produce all the deli-
cate gamakas (graces) and subtle srutis (quarter-tones) with ease and
accuracy.
4. Its long bow helping the performer to produce a continuous tone
and to play . . . with artistic finish.
5. Its wide compass of four octaves.
6. The ease with which the instrument could be used to accompany
voices of different pitch, from the high-pitched voice of the lady
singer to the deep voice of the male singer. (1955a, vol. 3, 271)

These characteristics were in contrast to those of the veena, which was

too ungainly to be easily transported to concert halls, too soft to be au-
dible from the concert stage, and, in Sambamoorthy’s words, ‘‘too majestic
and dignified to be used as an accompaniment in vocal concerts’’ (272).
C. Subrahmanya Ayyar wrote that ‘‘fortunately for us, the veena is not a
simple dead curio, but a living instrument, though a drawing room in-
strument relegated to our girls in South Indian homes, and veena play was
all but dead but for the radio coming to save it from oblivion’’ (1944, 61).
The veena might be the original model of Karnatic music, but it could
not bring Karnatic music into the modern age.
Why was the violin, rather than the sarangi or harmonium, the chosen
instrument in South India? The sarangi, an ‘‘indigenous’’ bowed fiddle,
seemed to reproduce the voice just as accurately as the violin. It was (and
is still) widely used in North India and seems to have been used in South
India in the nineteenth century. But the sarangi—long associated in the
North with tawaifs (courtesans) and Muslims, and in the South with nautch
parties (performances by dancing girls) and musicians of non-Brahmin

gone native? 45
caste—did not have the social cachet of the violin.18 As Regula Qureshi
has noted, ‘‘instruments mean’’ (2000, 815); they literally embody mean-
ing. What was important was not that the voice be ‘‘accurately’’ repro-
duced (although this was the rhetoric employed) but that it be reproduced
in a certain way. The sarangi in North India, according to Qureshi, is as-
sociated with an ‘‘excess of meaning’’: its ‘‘embarrassingly human’’ sound
and its associations with courtesans sets it apart from other concert instru-
ments, making it never quite classical enough (815, 820). The violin, on the
other hand, was in South India associated with the West and with moder-
nity. In the context of South Indian musical politics in the late nineteenth
and early twentieth centuries, it seemed more reasonable to adopt a fully
foreign instrument than to bring the sarangi onto the classical stage.
But the instrument could not be too foreign; it could not, for instance,
be the harmonium. Like the sarangi, the harmonium was edged out of
South India by the middle of the twentieth century, although for dif-
ferent reasons. It was (and still is) used in music for dramas and for the
musical storytelling known as katha kalakshepam but made only a brief
appearance on the classical stage. The first to rail against it, perhaps not
surprisingly, were English or highly Anglicized Indians. The art critic and
essayist Ananda Coomaraswamy disparagingly called it a ‘‘blatant instru-
ment’’ (Clements 1913, vii). The British musicologist A. Fox-Strangways
urged Indian musicians to ‘‘prune away’’ such an ‘‘unnatural growth,’’ tes-
tifying that it had already penetrated the remotest corners of India: ‘‘It
dominates the theatre, and desolates the hearth; and before long it will . . .
desecrate the temple. Besides its deadening effect on a living art, it falsifies
it by being out of tune with itself. . . . A worse fault is that it is a borrowed
instrument, constructed originally to minister to the less noble kind of
music of other lands’’ (Fox-Strangways 1914, 164).
The Irish-born music enthusiast Margaret Cousins wrote that the har-
monium had been a boon in some ways, helping to popularize classical
music because it was cheap and easy to learn (1935, 185). But otherwise it
was the bane of Indian music’s existence.

It is only the equipment of Central European beggars. It has no place

as an Indian instrument. It is unworthy of both East and West. It is
the most sinister influence in Eastern music today. It is tuned falsely
and contrary to the natural tuning employed for countless centuries in
India. Its harsh, over-loud tone is quite unsuitable for accompanying
the human voice which is strained in trying to hold its own. . . . It is

46 gone native?
 sapping all musical self-reliance in the voice so that even a good singer
feels helpless unless he is propped up by the harmonium. It is no wonder
it is called the harm-onium, for it works ‘‘harm’’ wherever it goes! (28)

In the same year, C. Subrahmanya Ayyar, lecturing before the seventh

All-India Music Conference on the topic of ‘‘gamakas and srutis in South
Indian music,’’ concluded: ‘‘Burn all the harmoniums in the world and
awake to the delicate intonations on the South Indian vina or the Euro-
pean violin, the queen and king of instruments’’ (Subrahmanya Ayyar
1944, 53).19
The harmonium was banned by All India Radio in the 1930s under
the leadership of John Fouldes, an Englishman who was the head of the
Delhi station’s Western-music section in the 1930s (Neuman 1980, 184).20
In 1938 a story entitled ‘‘Ārmoiya Pahiṣkāram’’ (Harmonium boycott)
appeared in the Tamil journal Bharata Mani (Pārata Maṇi). In the story a
conference of musical instruments (no humans allowed) is held on the
beach at Madras to discuss the ban on harmonium. The tambura, opening
the conference, states that the harmonium is a demon trying to disturb
the peace of Karnatic music with a mixed-up mass of wrong notes (‘‘ellai
illāta pakaikkoṇṭa pēy iicai nāpraavattaik kulaikka apasvarakkalañci-
yamāa ārmoiyam’’) (Kalyanasundaram 1938, 210). The veena stands by
impassively while the violin makes impassioned appeals to the other in-
struments: ‘‘Brothers and sisters! Even though I came from a foreign land,
I immersed myself in the ocean of music and became Indian. The harmo-
nium is also from a foreign land, but when it immersed itself in the ocean
of music it polluted the whole thing!’’ (ibid., 229). The violin goes on to
say that the harmonium has filled its stomach with Indian music and is
letting out belches: it has cheated vocalists, who now sing in a shouting,
bald voice (‘‘kattaik kuraluṭa moṭṭai moṭṭaiyāka apasvaraṇkaḷuṭa pāṭu-
vārkaḷ’’) (ibid).21
To this day the violin retains a privileged status in comparison to other
instruments in South India. Although the mandolin, saxophone, and gui-
tar are now played as solo instruments, none of them has gained the status
of accompanying instrument as the violin has. In accompanying these in-
struments on the concert stage, the violin stages them as ‘‘foreign’’ instru-
ments while presenting itself as a native voice.
The popularity of the violin thus went beyond strictly musical ele-
ments; indeed, its sound and the way that sound was interpreted were
inseparable from its extramusical associations, its social place. Whereas

gone native? 47
3 Violin advertisement. Madras Music Academy Conference Souvenir (ca. 1935–1940).
the harmonium was associated with the ‘‘beggars’’ of Central Europe, the
violin was associated with Western classical music and with the emerg-
ing urban middle class that, in postcolonial India, were expected to be
the custodians of a newly modernizing musical tradition. Unlike other
instruments in the 1930s, the violin crossed caste lines; because it was a
relatively new instrument, buying one was not mired in social or heredi-
tary obligations, as acquiring a more traditional instrument like the veena
might be. Between 1935 and 1940, violins of all sizes and qualities were
imported to Madras from Europe and the factories of Calcutta. Adver-
tisements for violins catered to a rising Brahmin middle class in Madras,
featuring compact, portable outfits complete with case and bow, which,
the ads suggested, would provide music for ‘‘modern’’ households; all one
had to do was purchase a pre-tested outfit and present it to the woman
of the house. Thus, the violin became both the sign of modernity and the
ideal vehicle for bringing modernity into the middle-class home.

The Violin as Voice: Three Revolutions and a Change of Style

Perhaps no figure is more emblematic of Karnatic music’s modernity than
the famous violinist Dwaram Venkataswamy Naidu, who came from an
unknown family in Andhra Pradesh to dominate the Madras music world
from the 1930s to the 1960s.22 In 1962 the musicologist B. V. K. Sastry wrote
that Dwaram’s ‘‘reverence for the classical tradition is tempered with the
spirit of inquiry of the scientific age. . . . Unfettered by gurukula tradi-
tions, he is subject to few inhibitions and freely avails himself of the best
ideas from everywhere, if these harmonise with classical Karnatic tradi-
tions’’ (33). Dwaram was hailed as ‘‘the best violinist of the day,’’ univer-
sally praised for the imaginativeness and violinistic qualities of his music,
called the ‘‘Dwaram touch.’’ Coming not from a family of great musicians
but one of military men and railroad officials, he was largely self-taught.
He was greatly influenced by Western violin technique, especially with
regard to the bow, which he used to give accents, emphasis, and dynam-
ics. He listened to records of the European violinist Fritz Kreisler and is
fabled to have charmed Yehudi Menuhin into letting him play his own
violin after demonstrating the gentleness of his touch and respect for West-
ern technique. At the height of his fame in the 1940s and 1950s, Dwaram
was visited nightly by Indian ministers of parliament and other govern-
ment officers, who would bring him records of Western classical music

gone native? 49
and commission him to incorporate them into his playing. If the record
was brought in the evening, Dwaram sat until the wee hours playing the
record a hundred times, counting each audition with the aid of tamarind
seeds. By the next evening, he would have something new in store for the
visitors.23
A man of eclectic tastes, Dwaram also listened to and incorporated
many of the features of Hindustani music into his playing, commenting
that what he appreciated about Hindustani music was the purity of tone
and intonation, and the slow, meditative pace of much of it.24 In the 1950s
he stopped accompanying vocalists and played solo until his death in 1964.
Dwaram was known for his treatment of rare ragas and his revival of many
of Thyagaraja’s and others’ ‘‘Western notes.’’ The characteristics of his play-
ing were a judicious use of speed, a generally slower unfolding of elabo-
ration in his alapanas, a sparse use of gamakas (‘‘only where necessary’’),
and generally less jāru (sliding).25 A student of Dwaram told me in 1998
that the violin ‘‘did not play in his hands—it spoke.’’ When I asked how it
spoke, she said that through the bowing and the trills and piṭis (from the
Tamil verb piṭi, to catch or grab), he was able to pronounce the words of
the song on his violin.
It is these very characteristics that made Dwaram’s playing seem old-
fashioned to many South Indian listeners at the end of the twentieth cen-
tury. A young violinist told me in 1998 that the Dwaram style was no
longer suitable to Karnatic music: it was too slow and plain and the bowing
too rough, all right for solos but no good for accompanying vocalists. In
the 1960s two more distinct styles had created revolutions in the Karnatic
violin world. One was the Lalgudi style, so named after Lalgudi Jayaraman
(b. 1930). Born to a family of violinists, Jayaraman moved with his family to
Madras in the 1950s, accompanied many vocalists, and began playing solo
from 1957 onward. A violin player in the Lalgudi style once demonstrated
some of the features to me: there has to be a softness and sweetness about
it, as if the voice is singing and pronouncing (‘‘voicepāṭara mātiri, ucca-
rika mātiri’’); there should be no roughness (karakkarappu) in the bow but
a continuity accomplished by keeping the bow on the string; the words
are to be pronounced by means of fewer piṭis and trills and more jāru; the
sweetness of the Lalgudi style is in the way it reproduces the singing voice
without any of the roughness of a human voice.26
A third style, known as the Parur style, also became popular in the 1960s
and remains the preferred style. Its most famous representative, and the
one who popularized it, is M. S. Gopalakrishnan (b. ca. 1935; popularly

50 gone native?
known as ‘‘M.S.G.’’). His father, Parur Sundaram Iyer, though originally
from Kerala, spent time in Bombay in the 1940s and there studied Hindu-
stani music. M.S.G. himself later studied Hindustani music independently
and has given Hindustani concerts. The hallmark of the Parur style is vir-
tuosic speed and range. The technique is completely devoid of piṭis; in-
stead, one or two fingers are used in a constant sliding motion along the
fingerboard to produce a cascading effect. There is relatively little empha-
sis given with the bow. The result is a sound that is very vocal but with no
vestige of speech—the kind of sound that is now recognizable to Western
ears as ‘‘Indian music.’’ M.S.G. gave several public demonstrations in which
he showed off a special bowing technique he used—a kind of spiccato in
which one long bow is divided into discrete, even units—playing in such
a manner for twenty minutes straight. In this style, the use of technique
to enunciate the words of compositions gives way to technique developed
to exploit the possibilities of speed and virtuosity on the violin. A young
violinist in Madras remarked to me in 1998 that the Parur style was the
most suited to vocal accompaniment because it captured the speed and
cascading effect of virtuosic vocal style.
These three revolutions in violin style reveal a more general change in
Karnatic violin playing in the twentieth century: the switch from a piṭi
style full of trills and catches, where the bow also gives emphasis, to a
more florid style (often called ‘‘weepy’’ by its detractors) where the tech-
nique relies on sliding along the fingerboard with only one or two fingers
and the bowing is constant.27 Early recordings of violin playing, from be-
fore Dwaram Venkataswamy Naidu’s time, reveal a general style that used
fingered runs (rather than jāru) for speed, as well as numerous trills and
catches;28 the violinists of the 1990s, by contrast, sound much more florid.
In the Madras of the 1960s, the question of which style to emulate was a
pressing issue for violinists. That in the 1990s the florid style has become
prevalent suggests that something in the notion of the voice has changed.
The reproducible qualities are no longer words and phrases but a gener-
alized, homogenized, virtuosic voice, a voice heavily influenced by the
violin. Even in 1939, C. Subrahmanya Ayyar noted that vocalists in South
India had already begun to increase their tempos in order to match the
capabilities of the violin (115). Several vocalists remarked to me that much
present-day vocal technique—not only the virtuosic speed which many
vocalists command today but also the gamakas that define the sound of
contemporary Karnatic music—are actually attempts to match the sound
and capabilities of this virtuosic violin.

gone native? 51
Magic Violin
In its short career in South India the violin has gone from being a colo-
nial instrument to being, almost literally, the voice of Karnatic music.
Throughout this career, the violin has also been at the center of musical ex-
periments that articulate both its intimate relationship to colonialism and
its seeming ability to reverse colonialism’s effects. In these experiments the
violin acts as a kind of translator or interpreter, a vehicle through which
Karnatic music and Western music can communicate.
In the 1930s and 1940s, two violin-and-piano experiments were re-
leased in Madras. One was the brothers Muthu and Mani playing the
Thyagaraja kriti ‘‘Nāgumōmu,’’ for violin, piano, and mridangam (a double-
headed drum). The other was Dwaram Venkataswamy Naidu playing a set
of raga alapana improvisations with the American pianist and composer
Alan Hovhaness on the piano. In both, the violin played in traditional
Karnatic style, and the piano acted as an accompaniment in the Karnatic
sense, trailing and highlighting the main melody. In Muthu and Mani’s
piece, released on record in the 1930s, the kriti, in the raga abheri, with its
plain notes and fast-fingered passages, already suggested a Scottish flavor.
The piano followed the violin, accompanying the same passage, when it
was repeated, with different harmonies. In the alapana experiment, which
was aired on All India Radio in the 1940s, the piano attempted to capture
the violin’s gamakas, particularly its slides and oscillations. The resulting
impression, and perhaps desired effect, was that the violin seemed to be
pulling the alapana in the direction of a traditional Karnatic elaboration,
while the piano seemed to alight on various harmonies, longing for but
never quite reaching resolution. The magic of the violin was its ability to
lead the piano, without being dominated by the harmonic laws of West-
ern music.
Since the 1980s these kinds of experiments, or ‘‘fusions,’’ have become
more self-conscious in the way they articulate Indian music with Western
music. In his fusion album How to Name It? (1986) the film music director
Illayaraja featured a piece entitled ‘‘I Met Bach in My House and We Had
a Conversation,’’ in which a Karnatic violin ‘‘accompanies’’ Bach’s prelude
to his third partita for violin. The piece starts with the solo violin’s alapana,
beginning in a typical Karnatic way but then building up to a virtuosic
pitch, as if the violin is expecting a Western denouement. The Karnatic
violin, amplified to a rich tone, contrasts with the orchestra of Western
violins in the background, creating a sense of proximity for the listener.

52 gone native?
This violin answers the orchestrated version of Bach’s partita, while in the
second half actual voices (those of the composer Illayaraja and the Kar-
natic violinist V. S. Narasimhan) articulate, or translate, parts of the melody
using sargam syllables (syllables used to name the pitches in a scale, much
like solfege in Western music). The effect is as if one is returning, via the
wonders of multitrack recording, to a scene of first contact, perhaps the ar-
rival of the European violin in India, but from the perspective of the Kar-
natic musician. Illayaraja describes the composition in the liner notes:

‘‘I Met Bach in My House’’ begins with an invocation that at first is

contemplative, introspective, and becomes increasingly importunate;
it comes to a climax, and is interrupted by the first notes of Bach’s Pre-
lude to his violin Partita III. The Prelude is soon played out in full bril-
liant dialogue with Indian instruments, a contrapuntal weaving that
seems completely natural. Nor does it seem strange when voices break
in spontaneously, in rapturous song, and we hear the Prelude articu-
lated at speed in Indian solfeggio, so neatly, so fluently, that a great
light dawns—the two musical cultures, Indian and Western—share a
common ground, far more than is commonly perceived.

The Karnatic violin here literally articulates Karnatic music with Western
music, providing a magical medium of sound through which the voice can
enter and interpret Bach.
In Kunnakudi Vaidyanathan’s Magic Violin (1998), belligerently subtitled
‘‘Everything Personal about It,’’ the object seems to be not so much in-
terpreting as outwitting. Kunnakudi had already used his violin to work
other kinds of magic: in the 1970s he released the album Cauvery (after
the river Kaveri, on whose banks Karnatic music is fabled to have been
born) in which he, with only the aid of his violin, claimed to capture all
the sounds of life along and in the river. Also in the 1970s he started play-
ing with the thavil (the drum usually used to accompany the nagaswaram
and therefore exclusively played by non-Brahmins) raising eyebrows in
the classical world. This, along with his renditions of film songs on violin,
contributed to his popular image as a violinist of the people. In ‘‘Magic
Violin,’’ Kunnakudi’s violin, amplified with echo effects, ‘‘takes on’’ and
‘‘conquers’’ the computer, which is playing flattened, lifeless, synthetic ver-
sions of Western chords. The computer, equated with the West, in turn
is equated with the world; on the cover of his cassette, Kunnakudi sits,
larger than life, astride a cushion floating in space, with the planet earth
in the background. The tracks have appropriately cosmic titles: ‘‘A Thing

gone native? 53
4 ‘‘Magic violin’’ advertisement in SRGMPDN (10 April–10 May 1997).

of Beauty,’’ ‘‘Full Moon,’’ ‘‘Eclipse,’’ ‘‘Creation.’’ Kunnakudi’s violin, with

its bewildering rhythmic effects, dances above the computer, outwitting
it every time.29
If the violin here acts as the instrument that helps Karnatic music out-
wit Western music and technology, elsewhere it helps constitute India’s
status as a nation by invoking Karnatic music’s place in relation to the West.
In August 1998 I was one of a group of violinists gathered by the violin-
ist A. Kanyakumari for a musical celebration of the fiftieth anniversary of
India’s independence: a concert of fifty violins playing exactly in unison.
In Madras concerts with multiple violins are common; concerts of two,
three, and five violins happen frequently during the Madras music season.
But whereas a concert of two, three, or even five violinists was manage-
able because the musicians could accompany one another, a concert of fifty
violins was more difficult to coordinate. In the numerous rehearsals for
the August 1998 event, Kanyakumari pared gamakas down to a bare mini-
mum and revised the program so that it included more ‘‘English notes,’’
which were deemed most suitable to group playing. No one commented

54 gone native?
on the irony of playing a concert of English notes at a celebration of India’s
independence. The concert was performed in a packed hall and got con-
siderable coverage in the press, which described it as an ‘‘aural feast.’’ After
the event, Kanyakumari’s success in having gotten fifty Karnatic violin-
ists to play in unison was hailed as equivalent to the accomplishments of
the great conductors of the West. What was significant was that no other
group of fifty musicians, whether flutists, veena players, or mridangists,
would have been conceivable. Here, perhaps enabled by its colonial ori-
gins, the violin itself became a figure of repetition. But, more important,
the spectacle of unity created by fifty violins against green, orange, and
white saris and veshtis was what people had come for. The violin thus be-
came part of a visual and aural demonstration of national unity. ‘‘It could
well be a one-time wonder,’’ commented a reviewer, ‘‘but nevertheless a
wonder.’’ 30

Voicing/Ventriloquizing
A violin is an instrument, which is to say a tool. . . . The fact is that by using a mechani-
cal contrivance, a violinist . . . can express something poignantly human that cannot
be expressed without the mechanical contrivance. To achieve such expression of course
the violinist . . . has to have interiorized the technology, made the tool or machine a
second nature. . . . This calls for years of ‘‘practice.’’
—Walter Ong, Orality and Literacy

The experiments that use the violin seem to have a common logic: through
the violin they rearticulate the relationship between Karnatic music and
the music of the West. But in so doing, they comment ironically on the
desire—shared by an Indian middle class that looks to it as ‘‘tradition’’ and
by Western consumers—for a pristine Indian music unaffected by colo-
nialism. In fact, they present in each case a music profoundly affected by
colonialism. It is due to the magic of the violin that Karnatic music is not
weakened, but strengthened, by the encounter; each experiment reenacts
and fulfills Karnatic music’s stubborn insistence that it be considered on a
par with Western classical music. If such a claim is the basis of the politics
of classical music in South India, the violin is what makes it audible.
When the violin became the necessary accompaniment for the voice in
Karnatic music, it displayed a seemingly natural ability to reproduce the
voice. But what was elided in this transposition? The Karnatic voice, and
the words it pronounced, were supplemented, or replaced, by the tone of

gone native? 55
the violin and piṭi and jāru and bowing techniques. Thus, the develop-
ment of Karnatic violin technique over the last two centuries has been
the development of a set of conventions by which the voice could be imi-
tated and/or reproduced. That these conventions change (which demon-
strates their conventional, rather than natural character) is apparent from
the shifts in violin style in the twentieth century; as they change, so do
notions of the voice.
In reproducing voice, the violin also becomes a ventriloquizer. The per-
fect violin accompanist is one whose playing is so self-effacing and unob-
trusive that the vocalist forgets there is anything except her own voice. At
the same time, the vocalist cannot be heard without the violin. The violin
thus renders Karnatic music ‘‘more articulate.’’ This same power of ven-
triloquism, which effects a separation between voice and subject, content
and form, is perceived as the very magic that saved Karnatic music from
destruction in the face of colonialism. ‘‘The emergence of the violin in
Karnatic music,’’ writes a violinist from Madras,

is a very significant response—a response comprising co-option and

adaptation of just the musical means, not the content, thought, or style of the
colonial music system to the native/indigenous music system. This re-
sponse, coupled with the high degree of sophistication and complexity
that the indigenous musical system had already accomplished, was one
of the ways by which the classical music system of South India was
able to maintain itself, quite intact, in the face of the cultural onslaught.
(Parasuram 1997, 40; emphasis added)

By the logic of ventriloquism, the site of deepest colonial impact is trans-

formed into the very sign, and sound, of a pure Indian voice. The voice
emerges as that which escapes the colonial impact precisely by allowing
itself to flow through another medium, the hollow body of the violin. The
authenticity of the Karnatic voice depends on articulating the relationship
between violin and voice in terms of two oppositions crucial to modern
thought: the separation between the real, or the original (the voice), and
its representation (the violin, the ‘‘mere’’ accompaniment), and the sepa-
ration between content (what the instrument represents, the true essence
of Karnatic music, the voice) and form (in this case, the mere instrument).
Perhaps we are now in a position to answer the question posed in a lec-
ture in 1935 by C. Subramanya Ayyar, who spent his life trying to fathom
the peculiar power of the violin: ‘‘One may contend that in vocal music, the
meaning of the words affects man. But why should music through stringed

56 gone native?
instruments, reproducing the nuances of the human voice, just like pri-
meval speech akin to that before man ever learned or spoke any language,
affect us? Can we offer any explanation?’’ (1941, 120). Here, the sound of
the instrument is associated with a wordless, primeval voice, a pure voice
that has the power to affect by transcending its colonial instrumental body
and its ties to language. It is precisely the capacity of the instrument to
be heard as a voice, to sound almost human while remaining nonhuman
and to sound Indian while remaining foreign, which makes it powerful.31
This ventriloquism works through a series of displacements which have
occurred at particular historical moments. The voice is supplemented by
the violin, whose sound then becomes a kind of super-voice; this virtu-
osic sound of a colonial instrument ‘‘gone native’’ is in turn imitated by
vocalists. The change in violin styles in the twentieth century from a piṭi
to a more florid sound is indicative not only of a change in vocal style
but of a change in the concept of the voice itself. The history of voice in
a particular time and place, then, is not simply the story of how people
come to express themselves, but rather the story of how people come to
let certain sounds speak and sing for them.32 The late-twentieth-century
preference for a virtuosic vocal sound and the claim that this is the natural
and authentic sound of South Indian music mark a desire to create a dis-
tinctively Indian sound, a representative ‘‘voice’’ not in danger of being
confused with anything remotely Western.
This is a distinctly modern and postcolonial desire. To say that the voice
in Karnatic music is a modern construct is not to say that there was no
vocal music in South India before the violin. Rather, the violin in Karnatic
music stages the voice in a particular way so that it becomes available as a
metaphor for a tradition and a self that have survived colonialism while re-
maining uncolonized. This staging is a repetitive act, borne through gen-
erations of musical practice that have made the violin in Karnatic music
not an unnatural peculiarity but second nature. If one takes seriously Re-
gula Qureshi’s injunction that ‘‘musical meaning and affect need to be con-
sidered as . . . historically and socially situated,’’ then one needs to consider
how the habitual, repetitive patterns of musical practice—indeed, the very
methods of fingering and bowing I have discussed above—articulate with
the perhaps more academically accessible realm of history (2000, 812). One
needs to consider musical practice or custom as ‘‘that obscure crossroads
where the constructed and the habitual coalesce’’ (Taussig 1993, xv). At the
heart of this inquiry is the question posed by Subrahmanya Ayyar of how
sound acquires meaning, as well as the broader question of how things

gone native? 57
that are ‘‘social constructions’’ locatable in particular histories nevertheless
come to be lived and deeply felt as truths (ibid.).
In this regard, one might say that the sound of the violin recalls, or
reenacts, colonial first contact through what Michael Taussig has called
‘‘second contact’’: when the colonial power or personage is refracted in the
images or sounds produced by the colonized. This often occurs through
mimesis, an embodiment of power in which the terms of power are poten-
tially redefined (Taussig 1993). Through mimesis, one is able to grasp and
master that which is strange or other through resemblances or copies of
it. Mimesis, like sympathetic magic, involves some kind of contact, often
embodied in an object taken from its original context and reframed in a
new one. Homi K. Bhabha describes mimicry in this sense as a ‘‘repeti-
tion of partial presence’’ that disrupts the authority of the original (1984,
129). For both Bhabha and Taussig, the force of mimicry/mimesis is in the
power it gives the colonized to disrupt the supremacy of the colonizer.
Indeed, the presence of the violin in Karnatic music, as a modernizing
influence and above all as a classical instrument, was essential to placing
Karnatic music on a par with Western art music and thus displacing the
latter’s supremacy as the classical music of the world.
But the fact that the violin is not native but gone native preserves its un-
canniness, its potential to be disruptive to the postcolonial mythologies of
Indian classical music as well.33 The capacity of the violin to ventriloquize
depends on its capacity to mimic, that is, not to represent the voice but to
repeat it in such a way that the relations between original and copy, voice
and accompaniment are destabilized.34 In this sense, the Karnatic violin
stands at the center of a complex set of mimetic relations between instru-
ment and voice, India and the West, colonizer and colonized. The mimetic
capacity of the violin guarantees the authenticity of the Karnatic voice, as
a colonial instrument is used to ward off—and ultimately redeem Indian
music from—the effects of colonialism.

58 gone native?
2 $ From the Palace to the Street
staging ‘‘classical’’ music

Not long after the advent of the Europeans, the Goddess of Indian Music had to jump from
the palace into the open street.
—M. S. Ramaswamy Aiyar, Lectures on Indian Music

The staging of Karnatic music in public concerts beginning in the early

twentieth century was not just a matter of shifting musical performances
from one kind of venue to another. Rather, it involved a constellation of
developments that profoundly affected the ways in which Karnatic music
came to be performed and heard. In a concrete sense, the concert hall
brought about a particular structure of presentation, one that was based
on a clear separation between the musicians and the audience, and the idea
of a ‘‘repertoire’’ of ‘‘timeless’’ compositions detachable both from their
original context and the contexts of their repeated performances. It in-
volved the microphone, necessary for making live music audible to large
audiences in noisy urban settings but also a technology whose effects went
far beyond mere amplification. In a less concrete but equally as real sense,
the staging of ‘‘classical’’ music involved the creation of an audience, a
group of concertgoers considerably more homogeneous in terms of caste
and class than attendees at earlier temple or street performances.
One incident can serve as an entry point into the kinds of shifts entailed
in this staging of classical music. In the late 1930s the maharaja of Mysore,
Nalwadi Krishnaraja Wodeyar, installed microphones in his durbar (palace
hall) and attached loudspeakers to the outside of the palace.1 The purpose
of these was to make the durbar concerts, which had long been reserved
for an invited few, available to the general public. The implications of such
an act were far-reaching: the private durbar concert was converted into
a public broadcast; the palace became a site of entertainment for a newly
imagined audience that listened from outside. The structure of the dur-
bar hall, with its carefully arranged mirrors affording views of the maha-
raja from all angles while the concert was in progress and its essentially
exclusive, private character, was considerably altered by the attachment
of microphones; the microphones literally turned the palace inside out,
converting the musical events there into concerts. If kingly authority in
South India was built around the physical and geographical concentration
of power in a single source, the microphones and loudspeakers introduced
a very different order of power.2 Theirs was the power not to concen-
trate but to disseminate. Their medium was not authority (in the sense
of a single, authoritative source) but reproduction. Their power literally
came from a reproducible source (electricity) and depended on the repro-
ducibility of another source (music). If kingly authority radiated from a
center, microphones worked by a different logic: that of the circuit. With
the power of electricity, the king and his music literally became wired to
the world outside the palace.
This moment highlights the difference and the conflict between two
economies of music, a conflict that is at the heart of the discourse on clas-
sical music in South India. Musicians at the beginning of the twenty-first
century routinely invoke images of a time before Karnatic music became a
‘‘business,’’ before musicians were at the mercy of sabha organizers, before
the noise, crowds, and ‘‘mechanical life’’ 3 of the city encroached on tra-
ditional methods of teaching and learning. The idea of a historical break,
constructed both temporally (the nineteenth century vs. the twentieth)
and spatially (royal courts and villages vs. the city of Madras), has been
foundational to the politics of classical music in South India. The imagi-
nation of such a break provided the rationale for the ‘‘revival’’ of Karnatic
music in the early twentieth century: the movement to ‘‘rescue’’ the arts
of music and dance from their ‘‘degraded’’ status, which had been caused
by the persistence of obsolete forms of patronage and performance in the
twentieth century, and reframe them in new urban, bourgeois contexts. In
this revivalist discourse Karnatic music was redefined as strictly devotional
and invested with the peculiar power to exist outside politics, outside cir-
cuits of economic exchange or personal motivation.
The rise of public concerts and commercial music-making produced
new standards of taste, articulated on the one hand in the emerging disci-
pline of music criticism and on the other in the curricula of teaching
institutions established to maintain those standards. At the same time,
these standards were being amplified in a new set of conventions of voice
brought about by the mechanism of the microphone and codified in the

60 from the palace to the street

retrospective canonization of the composer Thyagaraja, whose devotional
lyrics have come almost exclusively to represent the voice of Karnatic
music.

The Sound of Royalty: Patronage of Music

in Tanjavur and Travancore
At the beginning of the nineteenth century, the center of the Karnatic
music world was the royal court of Tanjavur.4 Before 1800, Tanjavur had
had several rulers from the Nayak and Maratha dynasties who were inter-
ested in music. For instance, the first Maratha king Shahaji (r. 1684–1712)
composed padams and prabandhas (types of composition characterized by
their prosody and number of verses) as well as a classificatory work in
Telugu entitled Sahaji Rāgalakṣaṇamu (Shahaji’s Raga Lakshanas: Character-
istics of ragas). King Tulaja I (r. 1728–1736) wrote a treatise in Sanskrit,
Saṅgīta Saramrita (Garland of music) (Seetha 1981, 116). By the time of Tu-
laji II (r. 1739–1787), several musicians had been brought from neighboring
villages and towns farther afield to be court musicians. King Amarasimha
(r. 1787–1798) held a miniature court at the town of Tiruvitaimarutur, near
Tanjavur, after he fell out of favor with the British East India company,
which prohibited the composition of songs in his praise.
In the first few decades of the nineteenth century, under the rule of Ser-
foji II (1798–1832), Tanjavur reached its peak as a hub of musical and artistic
activity. The young Serfoji received his training in English language, as
well as other subjects, from the Protestant missionary Christian Friedrich
Schwartz, who had been in Tanjavur since 1778. Through the intervention
of Schwartz, who appealed to the British East India Company, Amara-
simha, a temporary regent by an earlier agreement, was deposed and Ser-
foji was appointed king. During his reign, Serfoji maintained an interest
in scholarly collecting and various projects of improvement.5 He began
the first printing press in South India (with Devanagari script), started a
medical institution, collected European literature, and made preparations
for the building of a port at Nagapattinam. He started the Saraswati Mahal
Library in order to house his collections.
The Saraswati Mahal Modi manuscripts (records written in a version
of Marathi on palm leaves) indicate that a separate division of the palace
administration was given over to musicians patronized by Serfoji II (Ven-
kataramiah 1984, 221). The position of overseer of these musicians was held
by the violinist Varahappa Dikshitar (also known as Varahappayyar, men-

from the palace to the street 61

tioned in chapter 1), who also served a political post as one of Serfoji’s
ministers. The overseer was responsible for hiring musicians and seeing to
their welfare. The most illustrious of the musicians patronized by Serfoji II
were the Tanjore Quartette, four brothers: Ponniah (b. 1804), composer
and vocalist; Vadivelu (b. 1810), composer and violinist (also mentioned in
chapter 1); Chinniah (b. 1802), choreographer; and Sivanandam (b. 1808),
mridangist and nattuvanar (naṭṭuvaṇkam: small cymbals used to keep tala in
a dance performance) (Kittappa 1993, 6). These brothers were responsible
for creating and standardizing a large dance repertoire and developing a
teaching method. All four studied with the composer Muthuswamy Dik-
shitar, and Serfoji’s court provided them with a forum for their own and
their students’ performances. As court musicians, the brothers were re-
quired to sing for Serfoji each morning and regularly composed kritis and
varnams in his name (ibid., 7, 9). Serfoji presented Ponniah with a gift of
5000 rupees on his first performance in the court (ibid., 6). In addition to
attracting Tamil and Telugu musicians from the immediate region, Ser-
foji’s court drew Hindustani musicians from more distant places like Cud-
dapah and Gwalior (Venkataramiah 1984, 230); many also came to Tanjavur
to learn music.
In addition to his interests in European art and literature, Serfoji was
drawn to European music. The Modi manuscripts record the purchases
of instruments like ‘‘fiddle,’’ ‘‘Irish pipes,’’ harp, French horn, flute, trum-
pet, and piano in the 1820s. Serfoji himself studied Western music, and
it was from a European priest employed in Serfoji’s court who played
violin that Vadivelu is said to have learned Western violin (Kittappa 1993,
10). Serfoji also founded the Tanjavur palace band, with forty-two mu-
sicians playing both Indian and Western instruments, and maintained a
separate dance orchestra (Seetha 1981, 116). The band apparently played not
only Western music but also Indian music that had been arranged in staff
notation. Several books in Serfoji’s library contain staff notation under
which he had written the names of the swaras in Devanagari script. Books
filled with staff notation suggest that he employed people to write out
tunes, probably for the band’s use; he himself also composed tunes for
the band. Serfoji’s collections contain a wide variety of printed Western
music (at least twenty volumes), from chamber music for pianoforte and
string instruments by Haydn and Muzio Clemente to ballads from English
comic operas and handwritten books of reels, strathspeys, and waltzes. Ser-
foji’s successor, Shivaji II, also arranged Western music concerts and was a
patron of musicians (ibid., 120). After Shivaji’s death, however, in the ab-

62 from the palace to the street

sence of an heir, the Madras government took over the reign of Tanjavur.
The British, who considered the Tanjavur kings’ expenditures on music
and dance extravagant, discontinued the traditions of patronage that had
developed in Tanjavur. Musicians migrated to the other princely states of
Trivandrum (Travancore), Mysore, Pudukkottai, and Ramanathapuram in
the 1830s (ibid., 22–23).
Among those musicians who migrated was Vadivelu of the Tanjore
Quartette. Because of a disagreement over the nature of their services in
the Tanjavur temple, Serfoji II had removed the brothers from his court.
Although he invited them back shortly afterward, Vadivelu refused the
offer, instead accepting one from the king of Travancore, Swati Tirunal,
in whose court he remained for the rest of his life (Kittappa 1993, 9). Swati
Tirunal (b. 1813) assumed the rule of Travancore state in 1829, at the age
of sixteen. At the suggestion of Colonel John Munro, the British resi-
dent of Travancore state, Swati Tirunal had been tutored in English, San-
skrit, Marathi, political science, and Karnatic music by Subba Rao from
Tanjavur, also known as ‘‘English’’ Subba Rao for his skill in the English
language (Venkitasubramonia Iyer 1975, 4–6). In 1830 Swati Tirunal ap-
pointed Subba Rao as Dewan. Together, they attempted to make the king-
ship an example of enlightened leadership and Travancore a center of
learning.
The most important elements in this improvement were the English
language and Karnatic music. An English school was started in Trivan-
drum as early as 1834, earlier than English education was imparted in the
areas of the Madras Presidency directly under British rule (Venkitasubra-
monia Iyer 1975, 12–13). With the help of Tanjore Subba Rao’s contacts
and superb negotiating ability, Swati Tirunal sought to make Travancore
as much a center of Karnatic music as Tanjavur was. Subba Rao person-
ally negotiated with the durbars of Tanjavur and Pudukkottai to bring
many of Serfoji II’s court musicians to Travancore, including the Tanjore
Quartette and the vocalist Meruswami, who became Swati Tirunal’s music
teacher (156). Under Swati Tirunal’s rule, the forms of music and dance
peculiar to Travancore—the sopana sangitam (literally, ‘‘step music’’) sung
on the steps of temples and the dance form mōhini aṭṭam—were signifi-
cantly altered, brought closer to the forms of music and dance practiced
at Tanjavur. Sopana sangitam, known for its slow renditions and limited
improvisation, was altered not only by the influx of a new style of com-
posing and singing represented by Vadivelu and other Tanjavur musicians
but also by the very structure of music making that Swati Tirunal’s court

from the palace to the street 63

suggested.6 Instead of taking place on the temple steps, music making took
place in Swati Tirunal’s durbar hall as a recital, with the king himself as
the audience; musicians were invited to his court because of their skill.
Swati Tirunal himself is said to have kept a small room in his palace with
a view of Padmanabhaswamy Temple as his composing room. The palace
structure thus introduced the idea of a space for music that was separate
from the temple and from other living spaces. Meanwhile, the economic
structure in which musicians were supported out of the palace budget,
with some holding official positions like chief palace musician, created the
atmosphere in which music became a profession.7
Not content with having musicians from Tanjavur and several disciples
of the composer Thyagaraja, Swati Tirunal desired to have Thyagaraja
himself in his court (Venkitasubramonia Iyer 1975, 161). In 1838 he sent a
party of his court musicians—including Vadivelu and Govinda Marar, as
well as the superintendent of stables who knew the composer personally—
to Thyagaraja’s house in the town of Tiruvaiyaru, near Tanjavur. Thya-
garaja, who had already refused the patronage of Serfoji II, conveyed his
respects to Swati Tirunal but refused to come to Travancore. Otherwise,
however, Swati Tirunal’s handsome payments attracted musicians from
far and wide. He invited Hindustani musicians from Gwalior, Benares,
Pune, and Hyderabad, in addition to maintaining on a permanent basis a
Hindustani dance troupe and several musicians who had learned Hindu-
stani music in Tanjavur at Serfoji’s court. He also employed an Anglo-
Indian to play Western music and purchased Western instruments (165).
What Swati Tirunal’s court introduced, as had Serfoji’s before and the
Mysore and Vizianagaram courts later, was the idea of the court as a show-
piece of culture, a collection of the best musicians from around the world.
The better the musicians were, the more the power that accrued to the
king’s name.8 Swati Tirunal’s desire to patronize the worthy appeared to
be a selfless act of generosity and devotion to the art of music at the same
time as it was also a desire for power, a political act, part of the logic of
the gift that characterized kingship in South India. Swati Tirunal’s lib-
eral patronage of music was eventually curtailed by the administration of
the Madras Presidency. In 1840 a new resident of Travancore state who as-
pired to be Dewan reported to Madras that Swati Tirunal’s spending on his
court musicians was extravagant and unnecessary. The resident’s continu-
ous complaints effectively undermined Swati Tirunal’s authority several
years before his death in 1846 (Venkitasubramonia Iyer 1975, 155). Under his

64 from the palace to the street

successors Vishakam Tirunal and Ayilyam Tirunal, the Travancore court
continued to attract musicians but was no longer at its peak.

Institutionalizing Music at Mysore and Vizianagaram

Between 1850 and 1950, the largest and most renowned court for music
was that of Mysore. Under four rulers—Mummadi Krishnaraja Wodeyar
(r. 1799–1868), Chamaraja Wodeyar IX (r. 1868–1894), Nalvadi Krishnaraja
Wodeyar (r. 1895–1940), and Jayachamaraja Wodeyar (r. 1940–1950)—the
Mysore court was contemporaneous with the rise of Madras as the cen-
ter of the Karnatic music world. The combination of royal patronage of
individual musicians and the founding of institutions to teach music, a
connection with European music publishers and record producers, and an
articulated concern for the musical education of the public made Mysore’s
court, especially after 1900, significantly different from the earlier courts
at Tanjavur and Travancore. But like these other courts, the Mysore court
laid heavy emphasis on preservation. Chamaraja Wodeyar collected books
on music and founded the Oriental Library in 1891 to house them. Also, in
the 1890s, he made phonograph recordings of several musicians and kept
them in a library at the palace (Vedavalli 1992, 29). Under Nalvadi Krishna-
raja Wodeyar, the concept of preservation came to include the printing of
kritis composed by his court musicians and the writing of Karnatic music
in European staff notation.9 Two court musicians—Veena Venkatagiriappa
and Veena Sivaramiah—were employed to write notation for songs on
gramophone records in the collection of the royal family (36).
Nalvadi Krishnaraja Wodeyar also put a premium on teaching. In 1915
he founded the Royal School of Music at the palace; each court musician
was assigned to teach a certain number of students, meeting with them
daily from five to six in the evening.10 Although most of the teaching at
the school was on a one-on-one basis and many court vidwans had been
teachers before becoming affiliated with the court, Krishnaraja Wodeyar
felt it necessary to institutionalize the teaching of music. In 1928 the school
had twenty-five students, most whom were probably hereditary musi-
cians; nevertheless, the idea of such a school may have opened the way for
nonhereditary musicians to learn music. Krishnaraja Wodeyar also sup-
ported the study of European music. In 1913 he paid 1000 rupees—a hefty
sum even now—for an institution in Mysore run by an English woman
to prepare students to appear for the examination of the Trinity College

from the palace to the street 65

of Music, London. The Chamarajendra Ursu Boarding School, which had
been under the administration of the Mysore Palace since 1892, employed
several teachers ‘‘to teach the boys to play drums and pipes and to read
Western music.’’ 11
Although the maharajas of Tanjavur and Travancore were interested in
Western music, the Mysore court seems to have become most involved
with it. In 1920 Krishnaraja Wodeyar arranged for a celebration of Beetho-
ven’s centenary in Bangalore, at which Margaret Cousins played a Beetho-
ven piano concerto with the Palace Orchestra. In her account of the event
she seemed astonished by the good order and tune the piano was in, as well
as the respectfulness of the audience (Cousins 1940, 142). Nalvadi Krishna-
raja Wodeyar, who had himself received training in Western music, em-
ployed the German Otto Schmidt to conduct the Palace Orchestra and to
teach Western staff notation and harmonic theory. Jayachamaraja Wodeyar
extended his patronage to the Russian composer Nikolas Medtner, spon-
soring a series of recordings of his works in the late 1940s. A series of cor-
respondences between Jayachamaraja Wodeyar and the American record
producer Walter Legge demonstrates the maharaja’s interest in recording
‘‘neglected’’ European music (Legge 1998, 187–92).
The Mysore maharajas’ interest in Western music was deep enough to
have produced several varieties of Western ensembles, including a reed
band, a full orchestra, and a small orchestra, whose members, although
conducted by Otto Schmidt, were all Indian. The Palace Orchestra and
Reed Band gave performances for visiting maharajas and British officials,
for the opening ceremonies of medical and industrial conferences, and for
other notable occasions.12 Perhaps even more important, these ensembles
served as the model for the Palace Karnatic Band (also called the Indian
Orchestra), which played Karnatic repertoire on an ensemble of instru-
ments containing violins, veenas, flutes, clarinets, sitars, dilrubas, harmoni-
ums, mridangam, tabla, and dholak. Part of the function of the maharaja’s
durbar was to display the harmonious interweaving of India and Europe,
and the Palace Karnatic Band suggested an ensemble comparable to those
of the West.
The Mysore maharajas were interested in seeing this accomplished on
an even more musicologically minute level, sponsoring various experi-
ments combining Indian and Western music. A clerk named Pattabhiram-
maya was appointed to the palace controller’s office in the late nineteenth
century by Chamaraja Wodeyar because of his ‘‘note swarams,’’ compo-
sitions with English words (Krishna Rao 1917). In Krishnaraja Wodeyar’s

66 from the palace to the street

 5 Mysore Yuvaraja’s
Indian and European
Musical Series cover
page (1940).

time the purpose of having Karnatic melodies written down in European

notation was not only to preserve them but also to render them suitable
for harmonization. These ‘‘harmonized Indian airs’’ were performed by
the Palace Western Band, under the direction of Veena Venkatagiriappa
and Otto Schmidt. A series of airs were composed by Yuvaraja Narasimha-
raja Wodeyar (Krishnaraja’s younger brother) and published in London as
a series in 1940. The title page of these airs, most of which are scored for
small ensembles with tempos like ‘‘fox-trot’’ and ‘‘waltz,’’ bears the motto
‘‘To popularise Eastern music harmonised in Western style.’’
The Mysore maharajas’ orientation to ‘‘tradition’’ seems to have been
different from the orientation prevalent later in the twentieth century: in-
stead of dwelling on a golden past, they concentrated on the future of Kar-
natic music, which for them involved the successful wedding of Karnatic
music not only with Western music but also with technology. Between
1920 and 1930 the bands of Mysore Palace made gramophone recordings
that were sold commercially and competed with the Madras-based Nada-
muni Band, which was then popular at the weddings of the Madras bour-
geoisie. Some of the Mysore palace’s most lavish expenditures were on ac-

from the palace to the street 67

quiring pianos, harmoniums, and organs, as well as more unconventional
instruments like the horn violin, theremin, and calliaphone, a mechanical
music player (Vedavalli 1992, 34, 36).
The maharajas of Mysore were also interested in the technology of
the concert and concert hall itself, from the loudspeakers attached to the
palace to the staging of durbar concerts. The Trivandrum journalist A. Pad-
manabha Iyer offered the following description of the European durbar
in his 1936 account, Modern Mysore: Impressions of a Visitor.

The taking of his seat on the Throne is signalised by the blaze of electric
lights in multi-coloured forms and shape. The huge crowds of spec-
tators in front . . . are naturally surprised at the magnificent sight so
suddenly sprung on them. . . . The set of Indian Bhagavathars come and
go like angels, so suggestive their movements and so magnificent their
surroundings, because the Durbar Hall is Heaven itself on this earth.
These entertainers who are Palace employees are to sit on a seat in the
Electric Lift after tuning their instruments. . . . The Lift goes up and
stops just in front of His Highness. The signal is then given for them
to begin their entertainment. After their allotted time the bell rings
when they bring their entertainment to a close. Immediately, these per-
formers disappear like celestial beings of Puranic fame, as the Lift goes
down. (60–61)

Note here the exquisite staging of the music: if a regular concert stage has
the effect of physically separating the musicians from the audience and
thus presenting the musical performance as detached from the ordinary
activities of tuning or practicing, the electric lift accomplishes this even
more so. A perfectly choreographed performance also requires a certain
kind of audience, one that will sit attentively listening and watching. From
Padmanabha Iyer’s description one can almost feel the discipline that the
electric lights and the sudden appearance of the musicians exerted on the
spectators.
In the princely state of Vizianagaram, at the northern extremity of the
Madras Presidency, similar activities were going on, if on a smaller scale
than those in Mysore. Ananda Gajapathi, who ruled Vizianagaram from
1879 to 1897, had interests in history, philology, literature, and music. He
commissioned translations of Sanskrit treatises into Telugu, patronized
Telugu poets and writers, oversaw the founding of Telugu literary jour-
nals, and founded the Maharaja’s Dramatic Society, which staged new plays

68 from the palace to the street

(Rama Rao 1985, 12–17). In 1894, at the request of the Madras government,
he himself wrote a historiographic treatise on the Vizianagaram treaty of
1758, in which he detailed his family’s involvement in aiding the British
in their war with the French over the Vizianagaram territory (appendix).
An accomplished veena player himself, Ananda Gajapathi patronized both
Karnatic and Hindustani musicians in his court and maintained several
performing ensembles to play Western music (15). These included an Ital-
ian string band (consisting of twelve violins, two violas, and one cello) and
a brass-and-woodwind band.13 Like other maharajas, he purchased a horn
violin for the use of his court violinists.
Most important, Ananda Gajapathi imagined a music college that would
carry on the long-developing artistic traditions of Vizianagaram in a more
modern form.14 The Maharaja’s Music College, however, was not founded
until 1919, and then by Ananda Gajapathi successor, Vijayarama Gajapathi.
The college’s first principal was Adhipatla Narayana Das, a harikatha artist
from Vizianagaram who first received royal patronage from Ananda Gaja-
pathi in the 1890s after returning from successful performances at the My-
sore court. As Narayana Das went from his post as court artist to music col-
lege principal, the Vizianagaram Maharaja’s Music College became the first
of its kind in the Madras Presidency. Some elements of the college, like
free boarding for all students, remained similar to the structure of royal pa-
tronage. However, the relatively unstructured flow of musical knowledge
between musicians in the court was transformed into a six-year syllabus,
drawn up in 1919.15 For each year, the syllabus was divided into a ‘‘theory’’
and a ‘‘practical’’ section, and included lessons on the life and work of fa-
mous composers and the writing of notation. The six-year course was care-
fully graded, officially introducing students to the concept of raga only in
the third year and to improvisatory techniques only in the fifth. The stu-
dents were expected to learn a specific number of compositions. Stories
of court composers and musicians, instead of being handed down through
generations of court musicians, were relegated to lessons on ‘‘music his-
tory.’’ 16 At the end of their studies, the students were tested by six exam-
iners who had been elected with the sanction of the estate collector. An
introduction in the college’s silver jubilee souvenir from 1944 notes ironi-
cally that though Vizianagaram had always been a center for music, there
had never been as much interest in music there as there was after 1919.17

from the palace to the street 69

Narrating the Nineteenth Century
Adhipatla Narayana Das’s transformation from court musician to music-
college principal was repeated by many other musicians in the 1940s.18 As
royal patronage of music began increasingly to be replaced by state pa-
tronage in the form of government music colleges, a new genre began to
appear: anecdotes and biographical sketches of nineteenth-century mu-
sicians in the royal courts. Prominent among these were U. V. Swami-
nathayyar’s Caṅkīta Mummaṇikaḷ (published serially before being collected
in book form in 1936); V. S. Gomathi Shankara Ayyar’s Icai Vallunarkaḷ
(1970), a collection of anecdotes related by his father in the 1930s; numer-
ous anecdotes written by the critic Ellarvi, all in Tamil, throughout the
1940s and 50s; Mysore Vasudevachar’s memoir in Kannada, Na Kaṇḍa Kala-
vidāru (1955); and P. Sambamoorthy’s Great Musicians (1959). These writ-
ings were directed at a new audience of music students and concert-going
music lovers: those who had no direct contact with the royal courts. For
music students at the universities, many of whom did not grow up in musi-
cal families, the stories of these musicians of the nineteenth century were
available only through such books.
Some of the writings seemed to be aimed at a more casual audience of
middle-class Madras music lovers, who might read these stories in their
leisure hours. Ellarvi’s writings appeared in the popular Tamil newspaper
Svadēṣa Mitran before they were collected and published in several vol-
umes in the 1950s. In these writings the generous patronage of kings and
the escapades of musicians in royal courts were remembered as elements
of a disappearing past. Mysore Vasudevachar, for instance, dedicated his
memoir to Chamaraja Wodeyar, the king who had patronized him, but he
wrote it only after the Mysore court had lost its ability to patronize mu-
sicians and he had moved to Madras to become the vice president of the
music-and-dance school Kalakshetra (Vedavalli 1992, 26, 87).
The subject of these writings is a prior time, when music was patronized
by kings and wealthy citizens who themselves took an interest in music,
and before the city of Madras became the center of Karnatic music. The
picture that emerges is of a musical world that thrived on competition
and in which audiences routinely were thrilled to the point of ecstasy by
the performances. In Caṅkīta Mummaṇikaḷ (Three gems of music), Swami-
nathayyar (1855–1942) offers biographies of three nineteenth-century mu-
sicians, portraying this world most vividly. (Swaminathayyar is best known

70 from the palace to the street

for his editing and publishing of Tamil Sangam literature in the 1930s; he
also wrote a number of biographies and memoirs pertaining to the late
nineteenth and early twentieth centuries.) In Caṅkīta Mummaṇikaḷ Swami-
nathayyar uses a florid style to capture the glory of the music of that
period. In his portrait of Ganam Krishnayyar, he describes the atmosphere
of competition and the culture of challenges that thrived in the royal court
of Amarasimha, the Tanjavur king who preceded Serfoji II. When Krish-
nayyar was a young boy, a famous vidwan named Bobbili Kesavayyar came
from Andhra Pradesh to visit the court. Kesavayyar was famous not only
for his ganam (kaam: heavy, grand) style of singing but also for his flam-
boyant manner: he adorned his tambura with a flag and required that
any musician who lost to him in a musical contest surrender his tambura,
shawls, and other gifts from his patron; Kesavayyar would then visit other
courts on horseback, followed by a cartload of these surrendered items
(Sambamoorthy 1939, 429). Swaminathayyar relates that after Kesavayyar’s
performance at the court was over, King Amarasimha asked if any of his
court musicians could equal so difficult a style, with a voice ‘‘like a lion’s
roar’’ (ciṅkakarjanai pōl). The court vidwans answered, one by one, that such
a task was beyond them. Finally, the young Krishnayyar stood up and said,
‘‘If he teaches me the grammar [lakṣaam] of it, I will be able to sing like
that.’’ The king then arranged for the visiting musician to teach Krish-
nayyar, who became so famous for that style that the title ‘‘Ganam’’ was
attached to his name (Swaminathayyar 1936, 8–10).
Similarly, Swaminathayyar relates that the career of Maha Vaidyana-
thayyar (1844–1893) began with a challenge and a competition. Vaidyana-
thayyar was patronized by Subramania Desikar, who held a government
post in Tiruvavaturai, a place noted as a gathering spot for musicians as well
as for experts in Tamil and Sanskrit. Desikar had various designs for boost-
ing the reputation of his young musician. One day, two elder vidwans,
Periya Vaidyanathayyar and Chinna Vaidyanathayyar, came to Tiruvavatu-
rai. Since some people had expressed doubt about Maha Vaidyanathayyar
on account of his young age, Desikar arranged for a contest between him
and the two older musicians, complete with a referee (mattiyastar). After
showing his mettle by correcting the two elder musicians’ rendition of
one raga, picking out a single mistake of a single note, Maha Vaidyana-
thayyar himself sang. He chose a common raga but sang so fast that the
two elder musicians could not identify it and guessed that it was some rare
raga. When Maha Vaidyanathayyar emerged victorious, a long delibera-

from the palace to the street 71

tion ensued over what title he should be given; finally, they came up with
the title ‘‘Mahā’’ (great). Swaminathayyar implies that such titles bestowed
on musicians were not mere honorifics but equivalent to costly presents,
because such a title would forever call up the memory of, in this instance,
Maha Vaidyanathayyar’s victory. Indeed, a song by Tandavarayya Tambiran
in praise of Maha Vaidyanathayyar described Saraswati herself as ‘‘white
with the agony of having lost’’ to him (197).
There was competition among patrons for musicians as well. Ganam
Krishnayyar eventually left Amarasimha’s court to be patronized by the
merchant Ramabhadra Mupanar, and then by the zamindar of Udayar-
palayam, Kacci Kalyanarangadurai. To attract such a well-known vidwan
to a small court was not easy: Kalyanarangadurai had to impress Krish-
nayyar with a fancy palanquin and a handsome monthly salary (31). If a
vidwan employed by a patron showed himself to be poor, it was a loss of
face for the patron. For example, one time Krishnayyar had gotten low on
money and decided to sell a precious belt that Ramabhadra Mupanar had
presented to him. When Mupanar heard of this, he became angry and ex-
claimed, ‘‘If they tell us they need money, will we not give it?!’’ (38). Krish-
nayyar tried to reconcile with Mupanar but could not soften the other’s
anger. Finally, he began to sing a song, addressing his patron informally.

Eaṭā colluṭā eṇ cāmi nīyē

Eaṭā e mītu kōpam
[Hey you, tell me, you are my lord
What for this anger at me?] (38–39)

Hearing his musician’s voice Mupanar’s heart melted. He found it impos-

sible to remain angry and explained that it was ‘‘only a test’’ to see how
Krishnayyar would respond.
In ‘‘Tōṭi Aṭaku’’ (Pawning todi raga), a whimsical story told by Ellarvi,
there is a similar theme of the royal patron having to compete with a local
merchant. In the story, which Ellarvi dates around 1860, a local merchant
decided to cash in on royal patronage when the vidwan Sitaramayyar, fa-
mous for his renditions of todi raga, confided that he was out of cash and
needed a loan. The merchant, who knew how much people would pay to
hear Sitaramayyar sing todi, asked him to mortgage it, so that until the
loan was repaid, he was not allowed to sing todi. Embarrassed by this last
resort, Sitaramayyar did not tell anyone about the deal but secretly went
along with it. The king of Tanjavur was first bewildered, then finally ex-
asperated when Sitaramayyar sang every raga except todi. At last, when

72 from the palace to the street

cornered, Sitaramayyar explained his deal with the merchant and said he
would only be able to sing it again if the merchant was paid off. The exas-
perated king exclaimed, ‘‘If you told me you needed money do you really
think you wouldn’t get it?’’ and immediately sent the money off to the
merchant so that he could hear todi raga again (Ellarvi 1963, 1–3). In this
story the tension between two economies of music, one based on royal
patronage and the other on market value, is thematized.
The loving but testy relationship between musicians and their patrons
was another element of this nineteenth-century musical scene. U. V. Swa-
minathayyar shows the power of words, when sung, to create and undo
anger (1936, 37–39). One day, while under the patronage of Kacci Kalyana-
rangadurai, Krishnayyar met his patron for one of their habitual talks (callā-
pam). Kalyanarangadurai, however, was preoccupied with matters relating
to his lands and failed to give Krishnayyar his full attention. Noticing this,
Krishnayyar became angry and sang,

Pattuppai muttuppai vacra paṭakkaam

paipaiyāp paṇattaik koṭuttavar pōlap
pāṭina pāṭṭukkum āṭṭukkum nīreaip
pacappinatē pōtum palaaivēn kāṇum.
[You’re just like one who gives money and pearls and
diamond pendants for the songs I sing; enough of your
ingratiating me; I can’t see what the benefit is.] (38)

Krishnayyar turned the system of patronage against his patron, implying

that his gifts were a kind of bribe, crass and material, and that the patron
knew and cared nothing about the music he was getting in return. Kal-
yanarangar, on hearing this, realized his neglect, especially toward a vid-
wan who could, had he wanted to, found a place in a much larger and
more prestigious court. He apologized profusely, and Krishnayyar con-
tinued the song, deftly changing the words of the pallavi to those a female
bhakta would sing to her lord.

Pattuppai muttuppai vacrap patakkamum

parintu koṭuttu mikac cukan tantupin
pañcaṇai mītii koñci viḷaiyāṭi
rañcitamum arinta makarājanē.
[Maharaja! You know what is good, who gave me
money and pearls and diamond pendants, and played
gently on a mattress.] (39)

from the palace to the street 73

The same material gifts that were crass before are now transformed into
the signs of a good patron. As Krishnayyar’s sentiments change from anger
to love, the zamindar referred to coldly and without a name as ‘‘nīr’’ in
the first pallavi becomes ‘‘makarāja’’ (maharaja) in the second. The system
of patronage is thematized as the songs themselves become part of the
exchange of material and immaterial goods.
The value of words implied the valuing of adeptness with language. A
sign of this adeptness was a musician’s ability to hear a song with a par-
ticular tune and to make up new words to the same tune (Swaminathayyar
1936, 17). This went along with the ability to make whatever one sang per-
tinent to the situation at hand. For instance, at a banquet given by Gover-
nor Thomas Munro in Madras, Ganam Krishnayyar sang a song in which
the name ‘‘Munro Sahib’’ appeared, much to the governor’s delight (24). In
an incident related by Mysore Vasudevachar, two brothers named Kuppiah
and Appiah, although they were already accomplished musicians, could
not get into the Tanjore court. In order to get in, they disguised themselves
and enrolled as students of two court musicians. The king came to know
who they really were and accepted them as court vidwans. But several
senior musicians were envious and composed a new varnam they planned
to challenge Kuppiah and Appiah with when the king held a ceremony to
honor them. Having caught wind of this, the two brothers engaged a spy
to secretly notate the varnam for them. When, at the ceremony, the elder
musicians played the varnam and then challenged the brothers to play it
in three tempos, Kuppiah and Appiah played the varnam not only cor-
rectly but at triple the speed, vanquishing the elder musicians. The king,
utterly impressed, presented Kuppiah with a gold-threaded turban, and
Kuppiah spontaneously sang a pallavi in Telugu in which the syllables of
the notes also meant, ‘‘Is it right that the ruler should present such a tur-
ban?’’ (‘‘Pāga iccara sarigā?’’). Using the same virtuosic technique of swarak-
shara (in which the note names and the syllables of the words coincide),
the king spontaneously sang back, ‘‘It is only an ordinary turban’’ (‘‘Sadā
pāga iccanē’’) (Vasudevachar 1955, 13–14). The feat of singing a pallavi was
brought to even greater heights by making the pallavi punningly appro-
priate to the situation at hand in both music and words, allowing both
musician and king, without missing a beat, to show off their ability and
their modesty at the same time.19
Such contests happened even without the presence of kings in Madras.
For instance, the absence of a royal figure did not prevent the vocalist
Venugopaldas Naidu from pulling out all the stops to try to defeat Maha

74 from the palace to the street

Vaidyanathayyar before the eyes of the public. V. S. Gomathi Shankara
Ayyar related the story, elaborately and suspensefully, as it was told to him
by his father, Pallavi Subbiah Bhagavatar, who was a student of Maha Vaid-
yanathayyar from 1876–1882 (1970, foreword). In the late 1880s Maha Vaid-
yanathayyar was on an extended stay in Madras, enjoying popularity with
audiences and the governor of Madras. In a special feast given for the gov-
ernor, he gave a concert in which he sang English notes for the governor’s
sake. In the midst of such gracious society, however, envy was brewing. A
vidwan so popular that ‘‘if he merely moved his mouth, thousands would
gather,’’ Venugopaldas Naidu, known as Venu, was living in Madras at that
time. Shankara Ayyar describes at length Venu’s kinglike appearance, his
penchant for riding around Madras on horseback, and the fear he inspired
in people. Venu was angry that he himself hadn’t been asked to sing for
the governor, so to get even, he began grumbling that Vaidyanathayyar’s
‘‘Maha’’ was a fake degree and that by calling himself that he was cheating
the public. Venu confided with his close friend, the violinist ‘‘Photograph’’
Masilamani Mudaliyar, and the two decided that there should be a contest
if Vaidyanathayyar wanted to keep his title ‘‘Maha.’’ Informing a group of
their allies, the two collected a sum of 2000 rupees and bought a huge sil-
ver salver, shawls, and ear-studs, which would be the prize for the winner.
They used the remaining money to print notices to the public describ-
ing the time, location, and condition of the concert: it would take place
at the house of ‘‘Fiddle’’ Ramayya Pillai, a wealthy musician, in George
Town, Madras. Vaidyanathayyar would have the opportunity to sing first,
choosing the raga. Venu would then sing a pallavi in that raga, and Vaidya-
nathayyar would be challenged to elaborate it. If he could not, he would
have to give up his title.
Hearing of the challenge, Vaidyanathayyar and his brother, Ramaswamy
Ayyar, became nervous. But their friends goaded them on, saying they
couldn’t let a threat like this scare them. As soon as they sent back their
reply, the date was set and Venu began preparing intricate pallavis in all
the major ragas in which pallavis are usually sung. On the appointed day, a
large crowd gathered, including all the musicians then in Madras, expect-
ing that Vaidyanathayyar would be defeated. When the stage was set, with
Masilamani Mudaliyar as the referee, Vaidyanathayyar asked Venu which
raga he would like. ‘‘Any one you want,’’ replied Venu. Vaidyanathayyar
decided to sing the raga shankarabharanam. But at that very moment, his
brother Ramaswamy whispered to him in their special, secret language,
which they called Pandava Basha, to sing the raga narayanagowla. Unlike

from the palace to the street 75

shankarabharanam, narayanagowla does not have a straight ascending and
descending order but prescribes a certain vakra, or turn of notes, in its as-
cent and descent, making it difficult for manipulation in a pallavi. The
crowd, stunned to hear such an unusual raga chosen for a pallavi, listened
as Vaidyanathayyar sang an elaborate alapana for forty-five minutes. Venu
sat frozen, mentally trying to fit his pre-arranged pallavi into the raga, then
went off to find a quiet place where he could work it out. When Vaidyana-
thayyar was finished, however, Venu was nowhere to be seen. Masilamani
Mudaliyar, chagrinned, asked Vaidyanathayyar to sing his own pallavi, as
Venu, in hiding, listened angrily. In the end Vaidyanathayyar left with the
presents, a thousand rupees, and his title intact, and the vanquished Venu
never was able to regain his status (Shankara Ayyar 1970, 125–45).

Finding the Past in Madras

In the 1930s and 1940s two scholarly articles suggested that, like Tanjavur,
Mysore, and other places, Madras too had a musical past. Unlike Swami-
nathayyar’s or Shankara Ayyar’s anecdotes, these articles were in English
and clearly had a pretension toward history, supplying the reader with
dates and exact locations. If the anecdotes by Swaminathayyar and others
discussed above reflect a kind of nostalgia for an unrecoverable musical
past, these articles reflect a desire to connect the music world of Madras
in the 1930s and 1940s to that past.
In ‘‘Madras as a Seat of Musical Learning’’ (1939), P. Sambamoorthy de-
tailed the lives of several musician/composers who had settled in Madras
in the nineteenth century, including Veena Kuppier, Tiruvottriyur Thya-
gaiyar, Pattnam Subramania Iyer, and Taccur Singaracaryulu. The musical
activity of Madras was centered largely in the George Town area, the oldest
section of the city adjoining the upscale European area of Mount Road.20
Musicians from Tanjavur, Bobilli, Mysore, and Trivandrum often stopped
in Madras and stayed in the residences of these musicians. Concerts were
arranged in conjunction with festivals at the major temples in the George
Town area as well as the temples in Triplicane and Mylapore, and at the
houses of various musicians. The patrons in Madras were wealthy mer-
chants who seem to have combined the model of royal patronage and an-
other kind of patronage suggested by city life. For instance, the composer
Veena Kuppier was appointed to be the samasthana vidwan of the village
of Kovur, near Madras, by the patron Sundaresa Mudaliar. But Sundaresa
Mudaliar also kept a residence on Bunder Street in George Town, where he

76 from the palace to the street

was able to host traveling musicians, including Thyagaraja himself on one
occasion. The Manali Mudaliars, from the nearby village of Manali, were
responsible for placing Baluswamy Dikshitar under the tutelage of a Euro-
pean violinist in Madras and later for encouraging Subbarama Dikshitar
to write the monumental book of notations Sangīta Sampradāya Pradārṣini
in 1904.
According to Sambamoorthy, some of the first institutions of music in
Madras were festivals: the Vinayaka Chaturti celebrated in Veena Kup-
pier’s house and the Ramanavami festival concerts arranged by Taccur Sin-
garacaryulu (1834–1892) near his house in George Town. The Singaracar-
yulu brothers themselves became something of an institution, training a
large number of students and writing a series of graded music textbooks.
Indeed, it is said that every musician who wanted to make a name in
Madras had first to pay respects to and perform before the Singaracaryulu
brothers.21 Juttur Subrahmanya Chetty, a wealthy George Town merchant,
created an endowment for nagaswaram players to perform at the Periyal-
war festival at the Chenna Kesava Perumal Temple in George Town and for
a different player to be honored each year. Other smaller bhajan and con-
cert parties were maintained on Mint Street, in George Town, and near the
temples in Mylapore and Triplicane. The Nadamuni Band, a non-Brahmin
organization that performed commercially for weddings and dance music,
employing wind players and nagaswaram artists, was also established in
Madras in the late nineteenth century (Sambamoorthy 1939, 430–35).
The second article, ‘‘Some Musicians and Their Patrons about 1800 ad
in Madras City,’’ by the musicologist and Sanskrit scholar V. Raghavan, ap-
peared in the Journal of the Madras Music Academy in 1944. The article sum-
marized allusions to musicians and patrons in a very peculiar document:
a Sanskrit manuscript entitled Sārvadēvavilāsa, apparently written by two
Tamil Brahmins about the cultural life of Madras, which had been pre-
served in the Adyar Library in Madras. The manuscript possessed no date,
but by a series of complicated correlations with Madras district records,
Raghavan placed it at around 1800. He followed the two ‘‘scholars’’ as they
traveled around Madras, conversing with each other about what they saw.
In one scene they met the patron Vedachala, the dharmakarta of a temple,
as he was traveling to the temple with all his retinue, with two musicians
at his side. That evening, Vedachala arranged a gathering (sādas) with other
patrons of equal status (that is, dharmakartas in charge of other temples),
in which three courtesans, attached to various patrons, provided dance and
music. Another chapter provided a lengthy description of patrons and mu-

from the palace to the street 77

sicians present at another gathering, but the document broke off in the
middle of it. Raghavan suggested that musical life in Madras around this
time was centered in temples and that musicians were employed by the
dharmakartas, or caretaker/patrons of those temples (1944, 128–31).
The work Sārvadēvavilāsa, Raghavan argued, was probably written by
these two scholars at the request of the patrons. But what could have been
Raghavan’s motivation, in 1944, for ‘‘discovering’’ and translating a manu-
script which had been left untouched, presumably for almost 150 years?
Raghavan’s idea seems to have been to show the long continuity of musi-
cal tradition in Madras. The present-day importance of Madras in music
motivated his search for its past.

Madras City occupies today a very important and influential place in

the field of Carnatic music, a position which had steadily grown dur-
ing the last one hundred years, and had built itself on the foundation of
musical associations whose antiquity can be traced up to the latter part
of the 18th century. The discovery of the Sarvadevavilasa throws more
light on the history of these associations and reveals to us the person-
alities of some hitherto unknown patrons of our musicians. (130)

Sambamoorthy similarly finished his article in the present tense, ending

his narrative of nineteenth-century musical life in Madras with a descrip-
tion of the way such activities are now overseen by the University of
Madras: ‘‘The facilities given for the advanced study and practice of music
by samasthanas [royal courts] in former times are now given by Madras
University which has recently established a permanent Department of
Music and instituted a Diploma Course in Music’’ (1939, 437).

The Music Business and Its Institutions

Such anecdotes and scholarly articles were responses to a growing anxiety
about the burgeoning music ‘‘business’’ in Madras. Starting in the 1930s,
music critics regularly decried the ‘‘commercialization’’ of music through
the sabhas, organizations that arranged concerts. They lamented the time
limits put on concerts, the new musicians’ lack of spontaneity and real
musicality, and their penchant for mere show. Yet mixed with this anxiety
were a kind of pride and fascination at how music sabhas and institutions
in Madras were multiplying, at how there were more and more notated
compositions, more musicians to sing them, and more concerts every day.
The music business seemed to be running on its own, open to entrepre-

78 from the palace to the street

neurship, no longer dependent on a few wealthy patrons. The key elements
in this boom were the rise of concerts with tickets, the establishment of
music sabhas, and the founding of teaching institutions.
According to Sambamoorthy, the practice of remunerative concerts
was beginning to be established by the late nineteenth century. The mu-
sician Maha Vaidyanathayyar regularly sang for money in the 1870s and
1880s, but he never took the money himself, asking his brother Rama-
swamy to take it (1959, 5–6). The flutist Sarabha Sastri (1872–1904) is
supposed to have practiced rarely but kept in shape by playing frequent
concerts, for which he appeared punctually. He had set graded rates for
different areas of Madras Presidency and refused to accept a penny more or
a penny less (72). Indeed, the very word used for concert—kaccēri—suggests
the presence of money. Unlike the rarely used Tamil word viikai, which
refers to the action of hearing or listening, kaccēri, an Arabic loan word,
means both an ‘‘assembly for vocal and dramatic entertainments’’ and ‘‘a
revenue or police office, court of justice, or place of business’’ (Wins-
low 1862).
Hints from E. Krishna Iyer’s Personalities in Present Day Music, a collec-
tion of biographical and critical sketches of performing musicians pub-
lished in 1933, indicate that by then the concept of the public concert was
well enough entrenched for the author to compare different types of per-
formances. Krishna Iyer wrote frequently of ‘‘catchy’’ and ‘‘popular styles,’’
of musicians whose voices were suited to the microphone, and of mu-
sicians ‘‘playing to the gallery,’’ suggesting that public concerts had been
happening long enough for these concepts to become thinkable (28, 77).
In describing the vocalist ‘‘Tiger’’ Varadachariar, he quipped, ‘‘The stylish
or catchy music of the popular musician is like the light tea which refreshes
you after a day’s toil. ‘Tiger’s’ performance is a full meal, provided you
have got the stomach to digest it. The former is beach oratory of a popular
demagogue and the latter [the] University lecture of a learned professor’’
(7). The comparison of musical performance to different kinds of public
speaking here is telling, for music concerts were beginning to constitute
a part of the public sphere alongside such events. Krishna Iyer’s clear pref-
erence for the university lecture over political beach oratory shows that
he, like others, thought that Karnatic music should take pains to distance
itself from politics as it entered the same urban public space.
Music sabhas, organizations of city residents that arranged concerts,
were responsible for increasing numbers of concerts by the 1930s. Samba-
moorthy states that the first sabha in Madras was started in 1895 (1959, 14).

from the palace to the street 79

The earliest sabhas in Madras were in the George Town area: the Krishna
Gana Sabha, the Bhagavath Katha Prasanga Sabha, the Bhakti Marga Pra-
sanga Sabha, the Punarvarasu Sabha on Mint Street (Sambamoorthy 1939,
435; 1959, 76). Several sabhas, including the Indian Fine Arts Society and
the Katchaleeswarar Gana Sabha, held their events in Gokhale Hall, a so-
cial and political center in George Town, the site of many speeches by po-
litical leaders.22 Outside of George Town, in neighborhoods to the south,
the Parthasarathy Swamy Sabha was founded in Triplicane in 1901, fol-
lowed by the Rasika Ranjani Sabha in Mylapore. Among other early sabhas
were the Thyagaraja Sangeetha Vidwat Samajam, founded in Mylapore in
1929; Kalakshetra, founded in Tiruvanmiyur in 1935; Thyaga Brahma Gana
Sabha, founded in Thyagaraja Nagar in 1944; and Krishna Gana Sabha,
founded in Thyagaraja Nagar in 1954. Since then, the number of sabhas in
Madras has increased astronomically, particularly in the last thirty years.23
Sabhas were crucial to making music a remunerative business and at
the same time to communalizing it. Most of them started as community
organizations, established by groups of concerned citizens, usually Brah-
mins, with the idea of providing music to the neighborhood. V. Raghavan
likened sabhas to ‘‘earlier gatherings of old Madras, called sadas, convened
by rich patrons at fixed intervals or whenever there was a happy event’’
(1958, 89). The sabhas ‘‘patronized’’ musicians on the model of the royal
courts and these sādas, but with a crucial difference: they were entirely de-
pendent on the audience for support. This audience had to be of a certain
economic class, one that not only had the concept of music as a leisure-
time activity (that is, as part of ‘‘culture’’ and ‘‘tradition,’’ not as a money
making career) but could also afford to pay for it. Even the term sabha, with
meanings like ‘‘congregation, company, society, assembly of literati, select
society of believers’’ has connotations of exclusivity (Winslow 1862, 401).
The sabhas maintained their concert halls and paid musicians by charging
membership fees and/or charging money for concert tickets; they there-
fore depended on sizeable audiences. The more popular artists who drew
larger audiences naturally were more profitable for the sabhas; holding
concerts more frequently also increased profits. Because of the sale of tick-
ets and the location of performances inside concert halls, sabha concerts
were different in character from earlier temple or street concerts, which
were free and not necessarily attended from beginning to end. A new
kind of concert culture came into being, with audiences becoming con-
scious of themselves as a self-selected group, defined in opposition to other
groups whose tastes might differ.24 At the beginning of the twenty-first

80 from the palace to the street

century, this idea has been carried to such an extreme that many concert-
goers identify the type of music lover they are by the sabhas whose con-
certs they attend; even though for the most part all the sabhas feature the
same musicians, one makes a social statement by choosing where one goes
to hear them.
The life of the Indian Fine Arts Society, established in 1932, reflects
the fact that sabhas are differentiated by caste and socioeconomic status.
Located in George Town for the first thirty years of its existence, the
Indian Fine Arts Society catered to the residential population of North
Madras, consisting largely of Telugu Chettys (traditionally a mercantile
caste) and some Brahmins. By the 1960s the Telugu Chettys had mostly
moved out of George Town and the location could not accommodate
wealthier audiences from Triplicane and Mylapore, who arrived by car.
Since Mylapore had already been staked out as a Brahmin area, the sabha,
with its non-Brahmin founders, moved to the newer Thyagaraja Nagar,
whose nouveau-riche residents were distinguished more by their wealth
than their caste.25 In its early days the sabha arranged two concerts every
month and, during the ten-day festival in December, one concert per day.
At the beginning of the twenty-first century, the number of concerts it
arranges, during the festival period alone, has quadrupled, because, accord-
ing to the sabha secretary, ‘‘There are so many musicians today, and we
have to make room for all of them.’’
Concurrent with the growth of music sabhas and concert halls in the
early twentieth century was the founding of institutions to teach music.
The sabha system, which profited from greater numbers of concerts, cre-
ated a greater demand for musicians than there had ever been before. The
requirement that these musicians be paid according to some fair scale was
very different from the economy of royal patronage, in which the king or
patron supported musicians by giving them food and shelter, and in addi-
tion honored them publicly with gifts. This new system of ‘‘fair’’ payment
required that musicians all be somehow equally qualified; it implied that
the audience was paying to hear musicians of a certain standard.
The idea that such standards were detachable from individual musicians
themselves brought with it a mixture of fascination and concern. It was in
such an atmosphere that, in 1928, the Madras Music Academy was founded.
The previous year, the All-India Music Conference, held in Madras, had
expressed among its aims and objects the following: ‘‘to correctly under-
stand, improve and standardize [the theory and practice of Indian Music];
to provide facilities for widespread instruction in music on correct and up-

from the palace to the street 81

to-date lines; to open a separate faculty of music in universities’’ (Report
of the All-India Music Conference, Madras, 1927, 1). Foremost among the
priorities of the conference secretaries, P. Sambamoorthy and E. Krishna
Iyer, was the opening of a music academy at Madras that would provide
music instruction and concerts, thus acting as a model for other sabhas.
The academy was, according to an appeal to the public for support in 1935,

not a mere music sabha of the sort with which we are familiar. Its
aims are far higher. In addition to arranging musical performances, it is
making every attempt to purify Indian Music and to set definite stan-
dards. Even the performances are so programmed as to educate the
audience. . . . We run a Teacher’s College of Music to train teachers
who would maintain a pure standard in the art and who would provide
the kind of tuition required by our younger generation, particularly
the girls of our families. (Madras Music Academy 1935, 2)

The academy proposed itself as the ultimate standard in everything from

holding concerts to publishing music manuscripts, to administering a col-
lege for music teachers and placing them in model schools, to sending out
a committee to tour South India and record, on gramophone, ‘‘authentic
versions of compositions’’ (ibid., 2–4).26 Shortly after the academy opened,
other music institutions were established: a music department at Madras
University in 1932, music classes for the female students of Queen Mary’s
College in 1933, and Kalakshetra in 1935.

Playing to the Gallery

The need for standardized syllabi in music instruction and for a stan-
dard concert format were felt in the 1930s with unprecedented intensity.
In Vizianagaram in 1945 the Maharaja’s Music College was seen as ill-
equipped to maintain musical standards in the face of the monstrous de-
velopment of the music business, and to that end the Vizianagaram Music
Academy was opened. T. V. Subba Rao, who had traveled from Madras to
give the opening speech, admonished its founders: ‘‘Do not confuse its
function with that of the common sabhas. . . . It is a well-known fact that
the conditions of life for the musicians are now very prosperous. They
will spring up in thousands. This is a state of affairs which may be deemed
satisfactory enough, yet there is a subtle danger lurking in it.’’ Those insti-
tutions whose eyes were fixed on the box office were among ‘‘other forces
of destruction. . . . Their craze for novelty and uncouth experimentation

82 from the palace to the street

has succeeded in developing only the monstrous or the obscene in music’’
(Maharaja’s Music College Silver Jubilee Souvenir 1945, 20).
In order to ward off these ‘‘forces of destruction,’’ the concert form
itself had to be honed to perfect balance. P. Sambamoorthy, in his 1944
essay, ‘‘Our Concert Programme: Some Underlying Principles,’’ wrote that
the concert format was ‘‘based on certain aesthetic principles’’ that were
derived from centuries of musical practice in royal courts (39). But ‘‘the
vidwan who formerly delighted in expounding a raga for hours together
and earned the encomiums of even his jealous colleagues, had to remodel
his programme to suit the new type of audience’’ (ibid.). In effect, more
time had to be given to compositions and less to creative improvisa-
tion. For this new audience, the performer could afford to waste no time.
In a later elaboration of this essay, entitled ‘‘Kacceri Dharma’’ (Concert
etiquette), Sambamoorthy advised against preceding every composition
with raga alapana, which would result in ‘‘monotony’’ (1955, 267). The
fabled hours of leisurely elaboration of ragas by nineteenth-century art-
ists suddenly, it seemed, held a threat of boredom; modern ears could not
tolerate anything too unrecognizable. The performer should also, wrote
Sambamoorthy, make sure that the concert included a good variety of lan-
guages, ragas, and talas, so as to keep the interest of the audience. Above
all, complete professionalism should be maintained. The main performer
should not let any of his weaknesses show, take any unnecessary risks, or
draw any attention to the lapses of his accompanists. This kind of profes-
sionalism was called for on the part of the audience also: Sambamoorthy
advised that ‘‘members of the audience should particularly take care that
they do not talk with each other or become restless, when the tambura
or some other instrument is being tuned. Nor should a member of an
audience make an entry or exit during the middle of an item.’’ In effect,
Sambamoorthy was calling for the perfect choreography of performers
and audience to enact the rules of concert behavior. If both performers
and audience followed these rules, the concert itself would act like a well-
oiled, self-running machine. ‘‘An appreciative audience is like a catalytic
agent and draws the best out of the performer. Audiences should remem-
ber that an encouraging applause from them produces very good results’’
(ibid., 268–71).
If Sambamoorthy seemed to put all his faith in the concert form itself
as a kind of technology for producing good music, others seemed to feel
somewhat uneasy about it. To simply ‘‘let loose’’ all this music on inexperi-
enced or ignorant audiences was to invite disaster. In the 1930s and 1940s

from the palace to the street 83

a new discourse about the role of the music critic arose. In his foreword
to E. Krishna Iyer’s Personalities in Present-Day Music (1933), S. Doraiswamy
Iyer asked, ‘‘Are we not today on the verge of almost forgetting the high
state and royalty of Carnatic music, its true origin and true nature? Are
we not too thinking, too intellectual in our appreciation of it?’’ (ix). It
was the responsibility of the music critic—and the goal of Krishna Iyer’s
book—to help ignorant audiences develop their ‘‘taste’’ in music. Krishna
Iyer began by remarking on the unprecedented enthusiasm for music, the
increasing numbers of concertgoers, and the growth in general knowl-
edge of music. Yet amid this explosion was a deterioration in standards. In
effect, Krishna Iyer blamed this on the arrogance of audiences, suggesting
that the complete ‘‘ignorance’’ of country bumpkins was better than the
‘‘nibbling acquaintance’’ that so many urban concertgoers had developed.

The effect of the change of patronage from discerning princes and

patricians to the mixed crowd of the streets is indelibly marked in the
present-day growth and development of the art. The demand of the
populace of varying tastes and degrees of understanding has brought in
a corresponding supply. . . . A wholly ignorant audience would ordi-
narily enjoy the pleasing aspect of sweet sounds and might be content
to take the lead of the initiated on the scientific and higher features of
the art. But considerable sections of the present day music hall audi-
ences with their nibbling acquaintance with a good number of catchy
songs are not seldom found to crave more. . . . With them, the man
who sings a large number of pieces—preferably short ones—has . . . a
surer chance of wide popularity than others who may be able to ex-
pound their ragas, pallavi, and other scholarly features of the art in a
profound or elaborate manner. (xii)

The desire to know music and the sudden availability of music had en-
thralled audiences to the degree that they had lost their reverence for true
musical values. The ears of concert audiences in 1933 could respond to more
than mere sweet sounds, but therein lay the danger: they had lost the music
for the tunes. The craze for catchy songs was part of the jangle of city life,
as reflected in Krishna Iyer’s cityscape metaphors: ‘‘As a result of the exces-
sive development of tala accompaniments music has been driven to attune
itself to the steel-frame jathis of the rhythmic variety’’ (xiii). An obsession
with ‘‘technique’’ had overshadowed ‘‘natural’’ music (65). Of the vocalist
Naina Pillai he wrote, ‘‘His voice and training with obsession in tala have

84 from the palace to the street

on the whole contributed to make his mastermind delight in building up
amazingly wonderful steel-frame structures and spectacular skyscrapers,
rather than enchanting villas and breezy mansions in the midst of broad
open grounds of undulating green’’ (62). Not only the architecture of the
city but also its low-class entertainments had encroached on the concert.
As for showiness on stage, ‘‘The cheers of the audience are ready and vocif-
erous in proportion to the . . . length and noisiness of the display. One mri-
dangam is a sufficient tala accompaniment for any concert. Add to this . . .
kanjeera, dolak, a morsing and konnakole, and you have a regular circus
performance of the lion, tiger, bear, wolf and all other wild animals vo-
ciferously brawling and fighting with each other and the poor lamb of a
vocalist quivering in their midst and the heart and soul of Indian music—
melody and raga bhava—dished up beyond redemption’’ (53).
The music critic’s role was to redeem Karnatic music from its surround-
ings, from the circus it had become, by serving as a link to the past. The
music critic had ‘‘onerous responsibilities,’’ M. Ramachandra Rao stated
in his paper ‘‘Music Criticism,’’ which he read in Vizianagaram in 1945
(Maharaja’s Music College Silver Jubilee Souvenir 1945, 48). With the growth of
English education, university departments, and cinema, he claimed, audi-
ences were being bombarded with music of varying standards. ‘‘The music
critic has therefore the obligation of correlating the present-day endeav-
ours in the realm of music with old standards in order to polish the public
taste for genuine music.’’ But the critic could not simply rely on his gut
impressions; he had to have a knowledge of the sastras and the entire his-
tory of music in order to respond in detail to questions his readers might
ask him about his judgments. The critic had to take the place of the dis-
cerning patron, but his orientation toward music was different from that
of the patron’s. Instead of being able to choose his music, the critic had
to wade through whatever there was, with his memories of the past, now
endowed with the authority of ‘‘history,’’ as the basis for his judgments.
Public tastes could not be dictated but had to be gently trained into the
correct ‘‘grooves of discipline’’; the critic’s job was to teach the audience
how to listen from within these grooves (48–51).
It is no coincidence that recording—recovering music from the past
for the present—provided Ramachandra Rao’s metaphor. As I will discuss
in chapter 6, during this period sound reproduction came to be a cen-
tral metaphor for authenticity and fidelity to tradition. The music critic
emerged as the one who could manage the profusion of music available on

from the palace to the street 85

record and in concerts. By the 1930s and 1940s, Tamil journals like Ananda
Vikatan, Kalki, Bharata Mani, and Kaveri had started regular columns of
music criticism.

Music-Hall Silence, or the Greatness of the Mic

In his study of the rise of the idea of musical ‘‘classics’’ in eighteenth-
century England, William Weber points out that the establishment of the
European art-music canon ‘‘involved a set of notions or ideas, a moral
ideology, that propounded the value and the authority of the classics in
reference to issues within the society at large’’ (1992, 22). This ‘‘moral ide-
ology of taste’’ connected the idea of a musical canon with the sense of
an elite of educated listeners, a hierarchy of genres, and the demand that
audiences be serious and quiet (21). In South India, as in England, music
concerts constituted a ‘‘peculiarly modern musical institution: upper-class
people displaying their social status and their musical sophistication while
revering great music from the past’’ (1).27 The concert hall and its trappings
played a significant role in this development, for canons, although perhaps
originating in moral ideologies of taste, are maintained and cultivated by
physical staging and modes of listening; their authority lies in their power
to insinuate themselves into everyday habit and the rituals of public life.
In 1938 an advertisement for a water-pumping system in the Madras
Music Academy Souvenir used the phrase ‘‘Music Hall Silence’’ to sell its prod-
uct. The automatic water system was so silent, it said, that even in a music
hall it would not disturb the pin-drop silence of an audience listening to
music from the best masters. ‘‘Music Hall Silence,’’ the ad explained, had
‘‘become a standard expression to describe silence.’’ The association be-
tween the new middle-class home, with its technologies that obviate the
need for servants, and the concert hall, which should attract only refined
listeners, is clear: the newly imagined concert hall is the public counterpart
to the private middle-class home. The self-oiling automatic water system
is reminiscent of Sambamoorthy’s image of concert hall dynamics, another
smoothly working system. Silence here is associated with technology and
automaticity: the absence of human intervention. Yet the fact that the ad-
vertiser felt the need to explain the origins of the term music-hall silence—
even to the audience of concertgoers at whom the ad was directed—sug-
gests that it was far from standard. A cartoon from 1940 provides an ironic
commentary on the state of music-hall silence in Madras, showing a sabha
secretary removing all sources of noise until the audience is entirely de-

86 from the palace to the street

 6 Advertisement
in Madras Music
Academy Souvenir
(1938).

pleted. Sources such as these suggest that in the Madras of the 1940s the
concert hall was still an awkward concept, provoking a mixture of awe and
uneasiness, a sense that music and audiences did not quite fit into it yet.
If music-hall silence was daunting, however, so was the prospect of fill-
ing a hall with music. The microphone had by the 1940s become necessary
in sabha halls, many of which were large enough to seat 500 people but had
no acoustic structuring, or even walls, and were often situated alongside
railroad tracks or major roads. The first instance of microphone use was in
1931 at the religious festival of Mahamaham at Kumbakonam. In the 1930s
microphones were purchased from Europe by radio dealers; the micro-
phone was first used in a concert setting in 1933 for the singer K. B. Sunda-
rambal. By the 1950s almost every sabha in Madras had a microphone for
the singer, although it was not until the 1970s that microphones began to
be provided for accompanying artists as well.28 By the end of the twenti-
eth century, microphones had become required for all the performers in
a concert.
The microphone warranted its own chapter in Deivasikamani Acha-
riyar’s 1949 treatise on public speaking.29 Achariyar noted that although

from the palace to the street 87

7 ‘‘Well done, sabha secretary!’’ (Capaṣ, kāriyadarici!): ‘‘If you want the music concert to be with-
out interruptions, remove those who come late and try to squeeze in, remove those who are
coughing and sneezing, drive out those who are talking noisily, throw out those who bring cry-
ing children, and the rest can enjoy the concert without difficulty.’’ Cartoon in Ananta Vikatan
(27 October 1940).
speaking in one’s natural voice (iyarkai kuralil) was preferable, it was in-
creasingly common to have gatherings of several thousand people on the
beach or in open fields, and that in such situations a microphone and loud-
speaker system (oliperukki ) were absolutely necessary. If an orator were to
raise his voice to accommodate these kinds of crowds, he would end up
with a permanent case of laryngitis and there would be nothing but confu-
sion in the audience. Electric microphones and loudspeakers allowed one
to speak with one’s natural sound, and even those in the women’s section
would be able to hear it (251).30
Achariyar suggested that learning to use a microphone was not a simple
matter. It did not just mean speaking as one normally would and turning
up the volume; rather, it required a more complex shift in one’s bodily
attitude and way of speaking, a new set of practices. Achariyar instructed
the speaker to stand approximately one arm-length from the mic and,
without moving his body, adjust the height and diaphragm of the mic to
the level of his mouth. Strong speakers (nāvamai koṇṭa nāvalarkaḷ ) would
have to be aware that certain oratorical aspects ( pēccilakkaaṅkaḷ), such as
speaking posture ( pēccattōraṇai ), gesture (kai mēy kāṭṭal ), and ‘‘savoring of
the words’’ (cōccuvai), would be lost before the mic. Nor would a speaker
be able to switch voice qualities (kuralōcai ) in a single speech. Instead of
preparing his voice with reference to the crowd, the speaker had to pre-
pare with reference to the mic. Here, Achariyar found that an older mode
of speaking was the most appropriate parallel: that of a schoolboy recit-
ing his lessons by rote, focusing so hard on not forgetting his lines that he
neither moves his body nor modulates his voice. While speaking like this
in a normal setting would be disastrous, Achariyar suggested that if ‘‘in-
jected’’ through the mic (oliperukki mītum celuttappaṭuvatāl ), it would have
a great effect (1949, 253). In essence, the ideal was to focus one’s attention
entirely on one’s voice, almost forgetting one’s body. One can get a sense
of the very basic kinds of adjustments the mic demanded from another
piece of advice Achariyar gave. The first-time mic user, he wrote, ought
not to be disconcerted by the visual ‘‘deformation’’ of his body, the way
the long neck of the mic, once he was standing behind it and had adjusted
it, appeared to cut his body in two (253).
The microphone had wonderful potential, but Achariyar also warned
his readers to be prepared for its uncanny effects. Sometimes, seemingly
for no reason, the mic could ‘‘vomit out’’ harsh sounds, breaking the con-
tinuity of one’s speech (1949, 255). One had to be very conscious, as well,
of the volume of one’s natural voice. Just as in sculpture or painting in

from the palace to the street 89

which artists portrayed forms slightly larger than life size to increase their
effect, the mic could have a positive effect on a voice if used properly. But
if turned up beyond a reasonable setting, the mic would create sounds that
were monstrous. Unlike a veena or tambura, which had no settings for
volume, a microphone had to be set to the correct level, Achariyar wrote.
If a speaker listened to his voice amplified from an incorrectly set mic, he
might mistakenly assume that such an awful sound was the real sound of
his own voice and would artificially change his voice to compensate (253).
In 1947 the microphone was the subject of a satiric article, ‘‘Maiki
Ṃakātmiyam’’ (The greatness of the mic), published in Kalki magazine.
The author, Subbashri, marveled at the gradual but steady rise of the
microphone, that ‘‘modern excellent creation’’ (navīna arputa ṣriṣti), pon-
dering how he might describe its myriad uses. The microphone, he noted,
made possible both the political rally—the ‘‘heartwarming speeches of
the Mahatma, Rajaji’s scholarly discourses, and the brave lion-roar of
Nehru’’—and the sabha concert. No matter what the occasion, the mic,
with its thin, awkward body, could gauge the success of the event ‘‘like
a thermometer.’’ And just as the notice on a flyer that a special section
would be reserved for women could draw greater crowds, the promise
of microphone arrangements would make people imagine that a great
crowd would show up and therefore that the event couldn’t be missed
(‘‘ ‘Olippaappu ēpāṭukaḷ uṇṭu’ eu oru noticil piracurittāltān atu ‘ohō’
eu collumpaṭiyāa avvaḷavu periya kūṭṭam eu tōum, pōtu janaṅka-
ḷukku!’’) (41).
But the microphone was not just a simple tool for amplification; speak-
ing with one was a skill that had to be practiced to avoid public embar-
rassment, as Achariyar, too, had noted. Subbashri related how one day he
had a burning desire to become a well-known man among the people but
did not know how to go about it. His friend, ‘‘Sethu,’’ suggested that there
was no other way but to go give speeches with a microphone. But he had
to practice before going out in public, didn’t he? So he devised a prac-
tice mic out of a long stick with a tumbler (drinking glass) placed on top.
Within a month, he had become so proficient at speaking into the mic
that he hardly noticed what he was saying, until one day he discovered to
his horror that Sethu had been cueing him to campaign for contributions
to Sethu’s own home-building fund (1947, 42). The mic was so seductive
one might find oneself saying anything into it.
But the microphone also conferred unexpected benefits. After his brief
campaigning career, Subbashri wrote, he decided to become a ‘‘mic opera-

90 from the palace to the street

tor’’—‘‘the one, who, just before the concert, goes up to the stage and
says ‘Sivan’ 31 into the mic, turning it a little this way and that, and flashes
a little smile to the audience as he walks to the back.’’ ‘‘Do you know
how wonderful it feels?’’ he asked. ‘‘No one captures the attention of
the audience as these mic operators do’’ (‘‘Adanāl ēpaṭum perumaiyum
evvaḷavu pirammāntamāatu eu niaikkiīrkaḷā? Inta ‘maik āperētar’kaḷ
capaiyōri kavaattaik kavaruvatu pōl vēru yār kavarukiārkaḷ?’’). With
peculiar power, the microphone held its sway over audiences. It could
strike ‘‘with the force of an atom-bomb’’ (‘‘aukkuṇṭu pōl avarkaḷai aṭi-
yoṭu āntakkuṭiya caktiyum uṇṭu’’) if the amplifier malfunctioned, causing
a terrible lions’ roar of a sound and sending stunned men running as if they
themselves had been hit or knuckled on the head (1947, 42). That same mic
that could convert a lion’s roar into a sweet lamb’s voice reserved, myste-
riously, the power to convert it back, with a difference, at any moment.
The same mic that could make its operator feel so grand could also make
a fool of a man or a musician.
The microphone also had another kind of power. It was beginning
to control Karnatic vocalists. The phrase ‘‘maikukku ēa toṇṭai’’ (a voice
suited to the mic) was being used more and more frequently by music
critics on the radio. It was a miracle to hear: this tiny mic controlling
the power of a man, changing the harsh, heavy voice of a vidwan into a
shining, soft melodic one (‘‘kara katūramāna cārīram paṭaitta vitvāṅkaḷin
kuralaikkūṭa māi, vēku nāyamākac cōpikkac ceyyum amanuṣya cakti
intac cinnañciru maikkiniṭam aṭaṅki kiṭakiatu’’) (Subbashri 1947, 42). In
1949 Achariyar differentiated between the uccakural orators of the pre-
microphone period—literally those who spoke at the top of their voices
or lungs—and microphone-using orators, who were able to speak with a
‘‘natural sound’’ (iyakai oli) (1949, 251). He gave examples of uccakural ora-
tors, like the young Tiru V. Kalyanasundaram and T. P. Meenakshisunda-
ram, who were able to make themselves heard to crowds of thousands. ‘‘If
these people were to stand before a mic and speak, the instrument would
be ruined and cease to function. [The sound] would pierce the ears of the
audience [tuḷai: bore into]’’ (ibid., 252).
Most interesting here is Achariyar’s discussion of how a microphone,
properly used, can actually produce a new and aesthetically pleasing effect
of its own. According to Achariyar, among orators and singers, the one
who most successfully used the mic was M. K. Thyagaraja Bhagavatar,
for only he knew how to ‘‘unite’’ his voice with the mic so that the two
‘‘blended harmoniously,’’ ‘‘suffusing’’ the sound with ‘‘sweetness’’ (‘‘iimai

from the palace to the street 91

kāmala oliperukkiyuṭan iraṇṭara kalantu ilaintu’’) (1949, 255). This unity of
voice with microphone, however, was not to be confused with the dan-
gerous intimacy with and reliance on the mic that some singers had devel-
oped. Achariyar mentioned a young sangita vidwan (singer) who, despite
the natural wealth of his voice (cārīrac celvam), had come to rely on the help
of the mic, sitting before it with his head bowed as if he were about to
give it a kiss (‘‘talaikuintu ataai muttamiṭupavar pōl’’) (254).
In contemporary Karnatic musical practice, the microphone has to a
great extent become naturalized, but not in exactly the same ways as in the
West. According to many present-day musicians, the microphone has pro-
duced a general change in the vocal and instrumental aesthetics of Karnatic
music. They have noted that the microphone has obviated the need for
singers to project their voices, so that an earlier ‘‘shouting,’’ open-mouthed
style has been replaced by a ‘‘crooning’’ style which involves singing with a
closed mouth, almost like humming.32 Vocalists have accordingly lowered
their pitch; whereas the typical tonic pitch for male vocalists was E or F
before the use of microphones, it has now dropped to somewhere around
B-flat. As vocalists have lowered their pitch, so have accompanying violin-
ists. South Indian musicians are intimate with their mics; vocalists tend to
be only a few inches from the mic, and violinists so close to it that lifting
the bow off the string would mean hitting it. The mic is also essential to
projecting a certain intimacy in highly public spaces. When singers in a
concert hall embody or portray the soft-spoken, meek Saint Thyagaraja
beseeching his lord Rama, the mic is absolutely essential.33
For the most part, South Indian musicians and audiences prefer to have
amplification systems turned up to a level much beyond what Western
audiences would consider ‘‘tasteful.’’ There is little pretense of ‘‘realistic’’
amplification; instead, there is a sense that if one possesses the power of
amplification, one should use it. For the mic is not only a technological
mechanism but also a social mechanism. Both having a mic and know-
ing how to use one convey social status. I have attended house concerts
where audibility was clearly no concern, yet a full complement of micro-
phones was provided. During a Madras music festival in the late 1990s,
several nagaswaram players who had been invited to perform at the open-
ing of one sabha’s festival program refused to play until they were given
microphones. The sabha head had figured that the nagaswaram, a famously
loud instrument traditionally played outdoors, hardly needed amplifica-
tion, while the nagaswaram players, already sensitized to their marginal
status in the festival and in the larger Madras music world, had interpreted

92 from the palace to the street

 8 ‘‘The naga-
swaram player
gets flustered’’
(Nātasurakkārar
taṭumāukiār).
Cartoon in Kalki
(28 September
1947).

the absence of microphones as a slight. Indeed, a 1947 cartoon that shows a

nagaswaram player accidentally preparing to play the mic instead of his in-
strument suggests that nagaswaram players lack the sophistication to know
how to deal with a microphone; at the same time, it pokes fun at the ubiq-
uity of microphones.
The microphone, as Subbashri’s satire revealed, gave music concerts
a peculiar kinship with political speeches. Indeed, Deivasikamani Acha-
riyar’s book included another section that explicitly recommended that
those interested in public speaking and rhetoric look toward sangita vid-
wans as models of how to cultivate voice and breathing. He included
the familiar diagram of the midsagittal view of the upper vocal tract,
but his diagram additionally showed the position of the tongue, lips, and
larynx as the sounds sa, ri, ga, ma, pa, da, and ni are pronounced (1949,
204–11). In a lengthy digression that included anecdotes about the singers
M. K. Thyagaraja Bhagavatar and Birsahib, Achariyar suggested that for
speakers and singers alike, certain sounds were more difficult to produce
than others. He also suggested that, like singers, aspiring orators give at-
tention to breathing practice. ‘‘For public speaking, words and their pro-
nunciation are basic; for music, the basis is raga and tala. For both, a sweet
voice without defects is indispensable. A person’s throat alone does not

from the palace to the street 93

account for vocal quality; how he draws in and lets out his breath is also
very important’’ (204). The desired result of this breathing practice was to
develop what Achariyar called ōcaiyuṇarvu, a term which does not trans-
late easily into English. Literally translatable as ‘‘sound-feeling’’ or ‘‘sound-
sense,’’ ōcaiyuṇarvu conveys a sense of the knowledge of sound that comes
from producing it oneself, the hearing or awareness of sound from within
one’s own body (209).34
In addition to sharing these concerns about vocal technique, music
concerts and political speeches also shared public space. In fact, another
aspect of the uneasiness with the concert form in the 1940s seems to have
stemmed from the fact that its closest analogues were either to speeches
given during political campaigns or to variety entertainments such as films
or the circus. Music concerts, with their star systems, behind-the-scenes
agents, tickets, and mass audiences, were disturbingly close to the politi-
cal maneuverings of the post–World War II years; in many respects, they
shared the same public space. In his study of the rise of the notion of musi-
cal ‘‘classics’’ in England, William Weber articulates a useful formulation
of this particular relation between music and politics.

This is not to say that what happened in musical life ‘‘mirrored’’ En-
glish politics, or that the upper classes were interested in performing old
music strictly for political reasons. In is instead to suggest that musical
culture was closely bound up with public affairs on a broad plane, both
socially and politically, and that, to understand what was going on in
concerts or the opera, we have to see how it related to that world. Music
did not simply serve political purposes; it contributed to the ritual of
public life. . . . When we speak of ‘‘politics’’ we mean something broader
than parties and Parliament; we mean a broad compass of authority as
it was exerted in different areas of society. For that reason musical life
both had its own politics and was part of that larger world. (1992, 15)

The humor of several cartoons published in Tamil weekly magazines

from the late 1940s and early 1950s hinged on the way the concert party
and the idea of the concert can be easily used to express opinions about
politics. A 1941 issue of Kalki included a cartoon that showed the governor
of Chennai, satirically referred to as the ‘‘Bhavanagar Maharaja,’’ engaging
in so many different sports that ‘‘if he were to show up at the Thyagaraja
Festival in Tiruvaiyaru and sing and play all the accompaniments himself,
no one would be surprised.’’ The cartoon’s comment is twofold: it shocks
first by showing a political figure on the concert stage, and second by sug-

94 from the palace to the street

9 ‘‘The everyday excesses of the Bhavnagar Maharaja’’ (Pavanakar catāvatānam): ‘‘We
know that the governor of Chennai, the Bhavnagar maharaja, plays cricket, that he
mounts a horse and plays polo, that he plays hockey and is skilled at shooting arrows.
Therefore, if he were to appear at the upcoming Thyagaraja festival in Tiruvaiyaru and
sing and play all the accompaniments himself, no one would be surprised.’’ Cartoon in
Kalki (7 February 1941).
 10 ‘‘Music vidwans on the world
stage’’ (Ulaka mēṭaiyil caṅkīta
vitvāṅkaḷ): ‘‘Lion of vocal music
Stalin sings a sarvasamhara pallavi
in eka tala. The audience must
listen happily and be satisfied. To
open one’s mouth to express an
opinion is forbidden!’’ Cartoon
depicting Stalin as a singer in
Kalki (20 March 1949).

11 ‘‘Politics and Music’’ (Carkkārum caṅkītamum): ‘‘(1) The minister of broadcasting can sing,
(2) the defense minister can play the tabla, (3) the foreign minister can play the ghatam, (4) the
transportation minister can play the violin, (5) the education minister can play the kanjira, (6) the
labor minister can strum the tambura, (7) the food minister can taste the mridangam paste,
(8) the finance minister can sell the tickets.’’ The text at the bottom says, ‘‘To solve India’s eco-
nomic crisis . . . why not hold a great big benefit concert?’’ Cartoon in Kalki (25 December
1949).
gesting that music on the concert stage has descended to mere sport. A
1949 issue of Kalki included a cartoon that showed the tactics of world
leaders through the medium of the concert form: Stalin sings a compli-
cated pallavi while the East European countries in the audience, gagged,
are forced to applaud; Truman sings an ‘‘atom-bomb raga’’ while throw-
ing money to countries in the audience so that they will not notice what
he is singing; Churchill keeps on singing long after the entire audience
has left; the prime minister of Holland sings while his foreign accompa-
nists shoot into the audience. The cartoon thus shows the concert as an
oppressive tool of propaganda, a forum wherein leaders can control or de-
ceive their followers. Another 1949 cartoon in Kalki commented on the
money-making aspect of concerts. ‘‘To solve India’s economic crisis,’’ the
cartoon asks, ‘‘why not hold a great big benefit concert?’’ The cartoon
shows a different political minister on each instrument, with the broad-
casting minister as vocalist and the finance minister selling tickets. Here it
is the surprising congruity of the members of a concert party with the po-
litical cabinet that is shocking. The cartoon is critical of the use of music
to further political ends and make money, and is thus critical of the whole
sabha system; at the same time, it mocks audiences who will pay money
for a concert without even knowing who the musicians are.
From these cartoons it is clear that the worldly aspects of the concert
and the music sabha—their politics and economics—were by the 1940s and
50s considered threatening to the ‘‘real’’ values of music. Even as stories
of royal patrons paying musicians and kings using musicians to further
their own political power were part of the ‘‘music history’’ curriculum,
such politics and economics, when they appeared in a different form in
the present, were seen as dangerous. The political and economic aspects of
music concerts came to be seen as twentieth-century aberrations from a
timeless ideal: a classical music that was purely devotional.

Enter Saint Thyagaraja

As the concert became the predominant forum for hearing Karnatic
music in Madras, certain hierarchies emerged within the concert for-
mat itself, with composition placed over improvisation, certain composers
over others, and melody over rhythm. Audiences could no longer be ex-
pected to come ready to listen to a complicated pallavi improvisation for
hours. Instead, they had to be coaxed into a musical mood with shorter

from the palace to the street 97

compositions that they might recognize and presented with a variety of
ragas to avoid ‘‘monotony.’’ In a two-hour concert ( just the amount of
leisure time a working man could afford), the first hour would be given
over to kritis of fair to middling weight and difficulty. The centerpiece
of the concert would be the ‘‘main item’’: either a much-shortened pal-
lavi or a lengthy kriti with improvisation. Following the main item would
be ‘‘lighter’’ pieces, called thukkada (a Marathi word meaning ‘‘miscella-
neous’’), to send the audience off on a cheerful note. These thukkada pieces
could be javalis or padams from the dance repertoire (both noted for their
love lyrics), Hindustani pieces, Western notes, or, depending on the artist,
film tunes: in short, anything that fell outside the boundaries of strict ‘‘clas-
sical’’ music. By including such pieces but relegating them to a position
that lessened their status, the new concert format was set up to repeatedly
perform, and thus maintain, the boundary between ‘‘classical’’ music and
thukkada pieces.
Maintaining this strict boundary depended as well on defining the
music in terms of compositions. By the 1950s concerts had become com-
position oriented. The time limit of two to two-and-a-half hours, in com-
bination with the ever-looming threat of boredom on the part of the
audience, made singing a number of compositions (or ‘‘items,’’ as they
are called) in different ragas with limited improvisation the most favored
way of presenting the variety audiences were thought to crave. Since the
medium of the concert was largely compositions, the division between
serious or classical music and light music was based on compositions, rather
than on types of improvisation. More important, classical music was now
an entity definable by genre and composition, existing independently of
individual musicians and their styles of improvisation. Accordingly, most
conventional explanations of Karnatic music at the end of the twentieth
century distinguish between the composed parts of the music and the im-
provised. In reality, the boundaries between composition and improvisa-
tion are quite fluid. A ‘‘composition’’ in Karnatic music is hardly the same
as a composed piece of music in the Western tradition; it is a certain num-
ber of lines, each of which are played or sung with a number of variations.
While the lyrics of most compositions are fairly standardized, there is no
standard notation for compositions, and different versions of the ‘‘same’’
composition vary widely. Even within compositions, musicians often add
improvised flourishes that eventually become part of the composition.
Coexisting uneasily with the ambiguities of musical practice is the
notion of ‘‘masterworks’’—entirely composed pieces that are sufficient

98 from the palace to the street

in themselves without any improvisation. Having masterworks is a late-
twentieth-century idea in South India. It is now common to hear one of
the composer Thyagaraja’s five pancharatna (five gems) kritis played as an
item in itself at a Karnatic concert. Whereas thirty or forty years ago, ac-
cording to many musicians, these kritis would be improvised upon in a
concert, the pancharatna kritis are now seen as so complete and well com-
posed that improvising anything before or after them would be an insult
to the audience and the composer.35 Playing or singing a masterwork, in
this sense, makes the musician’s relationship to the composition very dif-
ferent. Playing or singing a composition in which one improvises makes
it necessary to be aware of where one is in the composition and to re-
main on the ready to jump in. Playing or singing a straight composition—
a masterwork—on the other hand affords musicians a sort of relief, an op-
portunity for detachment, a sure thing, as if by playing something standard
that everybody knows, they can tune out for ten minutes without having
to be responsible for the quality of the music.
It is no coincidence that most of the compositions treated as master-
works are by Thyagaraja, whose music came to be seen in the twen-
tieth century as the embodiment of a devotional or spiritual quality
considered essential to all classical music. In the 1930s a new term ap-
peared in writing on Karnatic music: ‘‘the trinity.’’ It referred to the
three nineteenth-century singer-saints who were becoming canonized
as the great composers of Karnatic music. Out of hundreds of com-
posers, these three—Thyagaraja, Muthuswamy Dikshitar, and Syama Sas-
tri—were considered to have caused a revolution in Karnatic music. Thya-
garaja, in particular, ascended to saintly status and came to dominate the
concert platform, eclipsing not only the other two but many other com-
posers who had thrived in the royal courts in the nineteenth century.
In the first half of the twentieth century, a flood of life stories of
Thyagaraja were published. They included Tyākarāja Carittiram, a historical
novel by Panju Bhagavatar (ca. 1910); Tyagayyar: The Greatest Musical Com-
poser of South India, by C. Tirumalayya Naidu (1919); Thiagaraja: A Great
Musician Saint, by M. S. Ramaswamy Aiyar (1927); Thyāgabramopaniṣad, by
Wallajapet Ramaswamy Bhagavatar (1935); and The Songs of Thyagaraja: En-
glish Translation with Originals, by C. Narayana Rao (1939). In 1942 a popular
film about the life of Thyagaraja, Tyāgayyar, was made in Telugu. Follow-
ing this were A. V. S. Sharma’s Lines of Devotion (1954), C. Ramanujachari’s
The Spiritual Heritage of Thyagaraja (1957), and Suddhananda Bharathi’s Saint
Thyagaraja: The Divine Singer (1968).

from the palace to the street 99

 The titles alone are indicative of the way Thyagaraja’s life was used to
produce a message for the twentieth century. Most of these life stories
cover the same events. Thyagaraja, a devotee of Rama, spends his time
singing and is scorned and tormented by his materialistic elder brothers;
he attracts a crowd of disciples and scolds them when they reveal that
they have gone to enjoy music in a devadasi’s house; he reforms thieves
by telling them they can take the money given to him by the king since
it was given to him against his will.36 The story told most often describes
how Thyagaraja refused the patronage of the kings Serfoji and Swati Tiru-
nal, preferring instead to sing in the streets of Tiruvaiyaru and to subsist
on donations (Ramaswamy Aiyar 1927, 46). The more the music business
flourished, the more the figure of Thyagaraja as a musician existing outside
of (and even resisting) the money economy was celebrated. Compositions
by others extolling the glories of royal patrons could not, in comparison,
be considered fully classical.
English translations of Thyagaraja’s lyrics that appeared in the mid-
twentieth century stressed the importance of grasping Thyagaraja’s spiri-
tual message. C. Narayana Rao’s English translations were aimed at those
who didn’t know Telugu as well as at Telugus who ‘‘through the influ-
ence of English education have forgotten to appreciate the treasures of
their own language except when they are in English’’ (1939, xi). For mid-
twentieth century translators, the actual language Thyagaraja used was less
important than the spiritual message that could be distilled from it. In Lines
of Devotion A. V. S. Sharma excluded the originals entirely, ‘‘translating’’
them into a Shakespearean iambic pentameter that an English-educated
leisure reader might enjoy. He appended a short conclusion directed to the
reader: ‘‘My purpose is achieved if you have developed a desire to appre-
ciate the spiritual bearing of the devotional lines written above’’ (Sharma
1954, 34).37 Having read such translations and been inculcated with the
desire for spirituality, a concertgoer would not need to understand the
words of Thyagaraja’s songs as long as he could sense this spiritual under-
current.
In representing Thyagaraja as a saint, these hagiographic accounts en-
dow him with an almost miraculous ability to rise above his circumstances,
a representation which is echoed in more recent accounts of Thyagaraja’s
life as well. In his book on Thyagaraja, Bill Jackson suggests that Thya-
garaja’s uniqueness is due to the fact that although he lived in a time of
great upheaval, he managed to filter out all political and social influences
from his music, leaving behind a pure ‘‘intangible’’ residue of devotion

100 from the palace to the street

(1991, 61). Not only his ideas but his music, too, possessed this spirituality,
which made it peculiarly suitable to nationalist imaginations of music as
a universal language. In 1941 S. V. Ramamurthi wrote in The Hindu,

[Thyagaraja’s] music is a synthesis of South Indian culture and is as great

as any form of Indian culture. Its Telugu is as simple almost as the Telugu
of the girl that goes home in the evening, singing, with her bundle of
fresh cut grass. But from such slim footing Tyagaraja’s music rises tall
as the world. Its tradition is Tamil, the tradition of Alwars and Nayan-
mars. Its grammar is Carnatic, that is to say, South Indian. Its spirit is
human, the spirit of man, the top of creation, communing with his cre-
ator. Everyone in South India can understand it, can feel its rhythm,
can follow its spirit and feel at home in it. Tyagaraja, more perhaps than
any other single musician, has preserved for us our one great live art
with an appeal both far and wide. (quoted in Jackson 1991, 60)

As the nationalist gaze makes its sweep from the country girl to the ‘‘spirit
of man,’’ the simplicity of Thyagaraja’s language gives way to a universal
‘‘human’’ spirit, which ‘‘everyone . . . can understand.’’
Such ‘‘simplicity’’ and universal appeal indicated that Thyagaraja be-
longed in the modern world, the world of the concert hall and the music
sabha. In an essay entitled ‘‘The Modernity of Thyagaraja,’’ T. V. Subba
Rao cited songs ‘‘ultra-modern in tone and sentiment,’’ whose lyrics de-
nounced expensive rituals and cruelty to animals (1962, 145). ‘‘What lan-
guage is more simple and thoroughly modern,’’ he wondered, than Thya-
garaja’s, which was ‘‘suggestive of infinite meaning and yet most familiar
like the simple dialect of the common man’’? (ibid., 146). Tirumalayya
Naidu contrasted Thyagaraja’s songs, which ‘‘please the moment they are
heard,’’ with Dikshitar’s, which, set in difficult and ‘‘unpronounceable’’
Sanskrit, ‘‘have to be studied over and over again’’ (1919, 22). More remark-
able, for Subba Rao, was the way Thyagaraja seemed to have anticipated the
future: ‘‘He knew that ages to come would expect a variety of new scales
and ragas with compositions in them . . . [while] he anticipated the musical
expressions which were likely to go out of use and avoided them’’ (1962,
146). Thyagaraja seemed to have anticipated that his songs would one day
become the staple fare of the concert hall, and thus he left clearly defined
areas for improvisation in his compositions, for ‘‘modern taste cannot be
gratified by mere recitative type of music’’ (ibid., 147). Moreover, Thyaga-
raja seemed to have known that modern ears preferred a catchy melody to
rhythm. ‘‘Thyagaraja has so refined, softened and subdued his rhythms that

from the palace to the street 101

they lose all their grossness and become spiritualized,’’ remarked K. Surya-
narayanamurthy Naidu (Maharaja’s Music College Silver Jubilee Souvenir 1945,
14). ‘‘Primitive music,’’ wrote Subba Rao, ‘‘had more tala and less raga.
Thyagaraja subdues rhythm and enriches the melody. . . . His marked pref-
erence of melody to tala . . . shows how modern he is’’ (1962, 148–49).
The privileging of melody over rhythm, here defined as a modern
attribute, is a pervasive element of contemporary discourse about Kar-
natic music and is one of the more prominent aspects of the canoniza-
tion process. The rhythmic accompaniment in Karnatic music consists
most commonly of a mridangam (double-headed drum), ghatam (clay pot),
and/or kanjira (single-headed hand-held drum). Less commonly seen are
the morsing, a kind of jew’s-harp, and konnakkol, the art of saying/singing
the rhythm syllables; indeed, konnakkol is usually now found only in
lecture-demonstrations, and rarely as part of a concert performance. The
dwindling of konnakkol is emblematic of the broader way in which per-
cussion has come to be seen as secondary to the melody in Karnatic music.
Whereas, according to older musicians, it was customary in the early twen-
tieth century for a concert party to have a full percussion section led by
the konnakkol artist, the presence of four or five percussion artists now
renders a performance less than classical. The privileging of melody over
rhythm is apparent in the comment of an elderly musicologist on why
konnakkol was discontinued: ‘‘It was interfering with the music.’’ 38 This is
not, of course, a purely musical decision. The rhythmic intricacies of Kar-
natic music are strongly associated with the non-Brahmin traditions of
periya mēḷam (music that accompanies Hindu rituals and festive occasions)
and cinna mēḷam (music that accompanies Bharata Natyam). The instruments
one sees on the Karnatic stage and the music one hears are thus the result of
a discourse about what constitutes music and who can be called a musician.
The canonization of Thyagaraja as the voice of Karnatic music involved
more than selecting a composer; it also determined what that voice was
supposed to sound like and who would be allowed to play along with it.

Lip Service
With the shift from royal patronage to the urban music business in the
twentieth century, Karnatic musicians and audiences found themselves
in the new contexts of the concert hall or university classroom. The dis-
orientation caused by the silence of the music hall or by hearing one’s
own voice through the microphone was amplified by music’s uncanny

102 from the palace to the street

power to move money, to become a business. In 1951 the music reviewer
Kalki Krishnamoorthy wondered whether to call his meditations on the
music business ‘‘music makes money’’ or ‘‘money makes music,’’ noting
that the meaning would be essentially the same (14). Money and music
seemed to be interchangeable, throwing certainties of origin and agency
into question. The resulting uneasiness prompted several responses. Anec-
dotes about the period of royal patronage, for instance, seem written to
preserve memories of a world that was perceived to be passing. Authors of
more scholarly articles sought instead to connect their present—1930s and
1940s Madras—to the period of royal patronage. A third response, which
became stronger as the concert format became more entrenched and as
concerts and audiences became more numerous, was to delve into the past
for a different history, one that could locate the origins of Karnatic music
in a pre-economic world, even if that world was purely imaginary. The
definition of Karnatic music as spiritual or devotional and the preoccupa-
tion with the figure of Thyagaraja as a saint who refused royal patronage
were responses to anxiety about the commercialization of music.39 Find-
ing a usable past now meant finding one in which Karnatic music was dis-
sociated from politics and money, from the whims of Westernized maha-
rajas. As the songs of composers celebrating their patrons disappeared from
the concert stage, the songs of Thyagaraja, the saint with peculiarly mod-
ern qualities, came to dominate.40
In a very real sense, this meant a shift from compositions tied to the
context of their performance toward compositions that stood free of such
social contexts and instead had their origins in a personalized, timeless
devotion. Thyagaraja’s lyrics became appealing precisely because, being
addressed primarily to the god Rama, they obscured the relation between
the composer and his social milieu. Like the singers at the Mysore palace
who appear and disappear before the audience like ‘‘celestial beings,’’ Thya-
garaja seems to float, detached, above the world of the social and the politi-
cal, the means (and history) of his entry into the scene hidden from view.
The persona of the individualized saint-singer-composer is tied to the
notion of the expressive individual, an idea central to modernity. Tracing
the history of this idea in Western thought, Charles Taylor has noted that
it was closely bound to a view of art as expression rather than mimesis
(1989, 368–90). This in turn depends on the notion of an interior self that
is known primarily from inward sources rather than from its relation to
others.41 Undoubtedly Thyagaraja himself sang in and in relation to a par-
ticular social context. The way that context became obscured as he came

from the palace to the street 103

to be celebrated, indeed deified, in the twentieth century is indicative of a
peculiarly modern way of thinking about art in general and the musician
in particular. The distinction between art as expression and art as mimesis
sets up, in the most literal sense, a distinction between what is supposedly
authentic (that which comes from within) and what is simply paying lip
service (an imitation, an effect without substance, a voice which is made
to appear as though it comes from within when it actually comes from an
external source).
The celebration of Thyagaraja as a modern composer is thus ironi-
cally apt, for the label unwittingly acknowledges the modern institutions
and modern ideas about individuality and authenticity—collectively em-
bodied in the figure of the saint-composer—that have made Thyagaraja
what he is today. The music festival is one such institution, and no festi-
val reflects the tensions inherent in the modern desire for a music free of
politics and commercialism better than the annual Thyagaraja Aradhana
(memorial) held at Tiruvaiyaru. Each year, in homage to Thyagaraja, thou-
sands of these aradhanas are held around the world. But the original is
held annually in Tiruvaiyaru, a small town on the banks of the Kaveri
River where Thyagaraja lived and died. The aradhana, attended by thou-
sands of ‘‘pilgrims’’ from Madras and other places in South India, takes
place next to a shrine built for Thyagaraja in 1925. From its inception,
the administration of this particular aradhana has been in the hands of
Madras musicians and has been riven by politics. The aradhana was estab-
lished in 1905 by a group of musicians from Madras; a separate group of
musicians who had not been invited to the meeting took offense and de-
cided to start their own aradhana. These two groups operated under the
names periya kaṭci (big party) and cinna kaṭci (little party)—kaṭci is the term
used for political parties—along with a third group, started by the female
musician Bangalore Nagaratnammal, for women only (Srinivas Iyer 1991).
These three ‘‘parties’’ conducted aradhanas on different days until 1940,
when they joined together and the Thyagabrahma Mahotsava Sabha was
formed (G. Srinivasan 1991).42 Since then, however, splinter groups have
formed continuously, depending on who is appointed as sabha secretary.
In 1998, for instance, a separate aradhana was held several miles away at
Tiruvarur for those who refused to participate in an aradhana run by the
controversial violinist Kunnakudi Vaidyanathan, who was secretary of the
Tiruvaiyaru Sabha at the time.
It is ironic that although Tiruvaiyaru represents a kind of origin, the
village world of Karnatic music before Madras became its center, almost

104 from the palace to the street

everything and everyone who participates in the Tiruvaiyaru aradhana is
imported from Madras. The sabha secretaries who run the aradhana are
now appointed by the Tamil Nadu state government and have no personal
connection to the descendants of Thyagaraja still living in Tiruvaiyaru. In
January 1998, during the three days of the aradhana, I stayed with some
of Thyagaraja’s descendants who are the year-round caretakers of Thya-
garaja’s shrine. During the festival, their house, about 500 feet from the
shrine, was suddenly engulfed by a maze of concessions selling cassettes
and food, a makeshift stage and backstage area, a large tent, giant electri-
cally lit outlines of Thyagaraja in his usual pose with tambura, and thou-
sands of performers and tourists. Elaborate security arrangements were
made and fees collected to insure that no disruptive people could enter.43
Every five minutes throughout the day new performers mounted the
stage to sing or play a Thyagaraja kriti, regardless of their skill; this was
obviously a place to see and be seen. In the evening an All India Radio
broadcasting crew was brought in to record the concerts given by famous
artists. The major event of the three days was the group singing of Thya-
garaja’s masterworks, his pancharatna kritis, by all the All India Radio–
qualified ‘‘vidwans’’ present. The festival tent was carefully divided into
spaces for vidwans and places for spectators, who were not meant to par-
ticipate in the singing. Meanwhile, merchants distributed flyers with the
lyrics of the songs on one side and advertisements for their products on
the back. A disgruntled pilgrim wrote in to The Hindu afterward that the
singing of the pancharatnas, in which the aradhana was supposed to reach
its devotional peak, was treated by the performing musicians more like an
opportunity to show off on a concert stage. ‘‘Breaking into alap and swaras
at will,’’ he wrote,

many of them in the choir sang with total lack of involvement, border-
ing on irreverence. . . . It was a show of one-upmanship of the worst
kind as one of them in the choir went into briga pyrotechnics at every
conceivable moment, unmindful of the comical effect it had on the
solemn occasion. Some of them resorted to ‘‘lip-service,’’ looking for
dubbing artists to provide them voice support, as they had not read the
script. . . . Chaos of this magnitude is never witnessed in other music
festivals. . . . Why then make it a mockery at Tiruvaiyaru? 44

Amid this chaos a small bhajan group composed of local musicians made
its rounds. On the Monday morning after the last day of the festival, the
place was suddenly deserted, with only the remains of the concessions and

from the palace to the street 105

12 Idol of Thyagaraja being carried to his house in Tiruvaiyaru at the close of the Thyagaraja
Aradhana ( January 1998). Photograph by Amanda Weidman.

the tent littering the ground near the shrine. The musicians and ‘‘pilgrims’’
from Madras had left during the night, leaving the locals to carry out one
of the purportedly most sacred duties: the procession of Thyagaraja’s idol
from the shrine to Thyagaraja’s house.
At Tiruvaiyaru, where the present world of Karnatic music confronts
its imagined past, one senses several anxieties. Celebrating Thyagaraja’s life
and work is supposed to be a gesture of remembering what Karnatic music
was before it was a business. Yet the county fair atmosphere of the 1998 ara-
dhana shows that worshipping at Thyagaraja’s shrine and doing business
at it are dependent on each other. Moreover, if the aradhana at Tiruvai-
yaru is supposed to celebrate this place of origins, it is ironic that Madras
politics control it, that so much of the event is imported from Madras,
and that such a careful separation is maintained between Madras musicians
and local musicians. Tiruvaiyaru, the place, serves only as a kind of sacred
backdrop; traveling to it in space gives the impression of having traveled
to some originary point in the history of Karnatic music, but when one ar-
rives one is confronted—as though looking in a mirror—with the present.
As the show of devotion through singing in front of Thyagaraja’s shrine
proceeds, the possibility that it is precisely a show can never be ruled out.
Once the idea of the concert becomes entrenched, it is impossible to tell, as

106 from the palace to the street

the disgruntled pilgrim points out, whether professional musicians singing
on the stage are showing their devotion or simply taking the opportunity
to be in the limelight; it is impossible to tell where their voices end and
their lip service begins.
The anxiety that music, like money, could simply be exchanged with-
out embodying any real value is reflected in comments about the ten-
dency of concerts toward show without substance, the loss of value of
once prestigious awards, and the endless proliferation of concerts. In an
interview in 1989 the vocalist C. S. Sankarasiva Bhagavatar, formerly chief
samasthana vidwan in the court of Ramanathapuram and later principal of
the Sathguru Sangeeta Samajam, a music college in Madurai, spoke of the
changes he perceived in the world of Karnatic music. After an involved de-
scription of the strenuous swarakshara pallavis that his guru, Harikesanal-
lur Muthiah Bhagavatar, would sing for the king of Ramanathapuram, he
made an eloquent and alliterative statement about the present day: ‘‘Ippa
cumma utaṭṭaiviṭṭu niraval paṇṇamāṭṭā oum paṇṇamāṭṭā. Cumma pal-
laviṅkiatu pallai kāṭṭiṭṭu viṭṭiṭuva!’’ (Now they won’t even open their lips
and do niraval [improvisation on a single melodic line of a composition].
They won’t do anything. If they just show their teeth, they call it a pal-
lavi!).45 Punning on the word for teeth ( pal) and the word pallavi, Sankara-
siva Bhagavatar’s comment suggests that audiences are gullible enough to
be cheated by lip service or, in this case, a mere show of teeth and that the
present day has retained only the appearance, not the substance, of music.
A similar anxiety about the loss of value for the sake of show appears in
comments about awards given by the government to musicians and other
artists. The Kalaimamani awards, for instance, given by the chief minister
of Tamil Nadu, seem to be based on the model of a royal patron honor-
ing distinguished members of his court. But where the honor ends and
farce begins is hard to tell; in 1998 an unprecedented seventy-five Kalaima-
mani awards were presented by Chief Minister Karunanidhi, prompting
many of those awarded the honor in previous years to return it in indig-
nation, saying that the award had lost all value. Karunanidhi’s action also
prompted a front-page cartoon in the Tamil daily Tia Maṇi, which sug-
gested that the Kalaimamani award had been given out as liberally as daily
rations. The seemingly uncontrollable proliferation of concerts also pro-
vokes anxieties about the loss of music’s value, the fear that music will cease
to have an effect when heard. The disorientation caused by endless repe-
tition and abundance is the theme of the cartoon ‘‘Chennai Winter Sea-
son.’’ The traditional expression of reverence, ‘‘Music is an ocean,’’ suddenly

from the palace to the street 107

becomes a cause for worry when the flood of concerts in the December
season threatens to swamp even Thyagaraja himself.

Backstage and Outside

In October 1998 my violin teacher was embroiled in a bitter family dis-
pute over what should be done with the valuable medals and gems that had
been presented to her father, a famous violinist, by various maharajas in
the 1920s, 1930s, and 1940s. Some in the family wanted to sell the medals for
money. My teacher, who was in favor of keeping the medals as they were
or making a ‘‘museum,’’ as she put it, to preserve them, was plunged into
a state of anxiety. The medals were a present-day relic of a past that only
she, as a musician, cared about but had never directly experienced. What
would happen to the family name if the medals were sold off ? Would any-
one remember her father? Would her father, who had earned these medals
by sheer talent and merit, have understood selling them for money?
In the midst of the dispute, I took a trip to Mysore with my teacher, and
her cook, Chinnamma.46 The Mysore palace is no longer inhabited now
that the members of the royal family have become politicians in Banga-
lore; it became the property of the Indian government and is now open to
the public as a museum. Neither my teacher nor Chinnamma had seen the
palace before. We took a tour of the interior, gazing at portraits of British
and Indian royalty. By the time we got to the durbar hall, it was almost
closing time, but my teacher was too absorbed to notice: here was the hall
in which her father had played before the king of Mysore and received one
of his many medals. Other tourists had long since left, herded out by the
palace guards, but she wandered around at length in the silent hall, placing
herself first where the maharaja would have sat, then where her father
would have sat, then imagining herself in the audience. Seeing the durbar
hall, she was inspired to ‘‘remember’’ an event she herself had not experi-

108 from the palace to the street

 13 The excesses of Chief Minister
Karunanidhi caricatured. A man with
a turban and mustache tells a decidedly
Brahmin-looking man, ‘‘You, too?! This
isn’t a rations line. Everyone here is wait-
ing to get their ‘kalaimamani award.’ ’’
Cartoon in Tia Maṇi (19 June 1998).

enced, an era she was too young to have known. The preserved palace
seemed to represent for her a haven from the nagging possibility that any-
thing, against all propriety, even her father’s talent and music, could be
converted into money.
Chinnamma and I left her and exited the palace, joining a crowd of
tourists outside waiting for the weekly light-and-music show. The lights
on the outside of the palace were turned on exactly at 7 pm, and a marching
band (formerly owned by the maharaja, now run by the police depart-
ment) played for one hour, until the lights were shut off. The show, care-
fully planned for tourists, is supposed to recreate the atmosphere of the
early twentieth century. But Chinnamma was not interested in authen-
ticity in the way the tourist industry imagined. At one moment during
the show, she turned to me, struck by a sudden epiphany: she had finally
realized where an expression she had been using all these years came from.
The expression—‘‘Ea? Periya rāja familyā? Icai muaṅkutā?’’ (What are
you, from a big royal family? Is the music roaring?)—was used when you
wanted to mock somebody for acting pretentiously. It could be shortened
into a sarcastic half-question: ‘‘Icai muaṅkutā?’’ (meaning something like
‘‘I can hear the band rumbling’’). Chinnamma, an illiterate cook who, un-
like my teacher, could have had no personal connection to the palace, had
found one embedded in the language she used every day: an expression
that reflected her own sense of social propriety and her own location (her
authenticity, as it were) within a politics of music from which she, through
caste and economic class, had been excluded.
Many of the concertgoers and musicians with whom I spoke in the
course of my research lamented that Karnatic music is much more exclu-
sive now than it used to be; they remembered temple concerts, free and
open to all, where the audience poured out onto the streets. But if the
atmosphere of Karnatic music concerts has become much more exclusive,
the music is, ironically, more available to the general public now than ever

from the palace to the street 109

 14 The music season
threatens to swamp even
Thyagaraja himself.
Cartoon in The Hindu
(17 December 1997).

before through other mediums: cassettes, radio, Tamil cinema. However,

it is now framed within a thoroughly culturalist discourse where it oper-
ates as a sign of culture, tradition, refinement, and Brahmin middle-class
identity. Chinnamma’s epiphany in a sense talks back to these exclusions,
just as the expression itself talks back to an imagined interlocutor. Unlike
the ideal audience imagined by Sambamoorthy, she does not sit quietly,
and if she applauds at the end, it is because she has appreciated the music in
a different manner than he would have prescribed. She finds in the palace
band’s performance not an occasion for nostalgic recollection of a golden
age, but rather the origins of a class system that justifies itself by distinc-
tions of taste. Here I have shown how, through the rise of the concert as
the dominant mode of performing and listening, these distinctions were
established, made to seem logical and natural. In doing so, I hope to have
hinted—as Chinnamma does—at how the ritual of the concert might,
quite literally, be disconcerted.

110 from the palace to the street

3 $ Gender and the Politics of Voice

It is always the body social that is enunciated in and through the voice.
—Steven Feld, Aaron Fox, Thomas Porcello, and David Samuels,
‘‘Vocal Anthropology,’’ in A Companion to Linguistic Anthropology

As a music student in Madras, I spent the better part of every day engaged
in music lessons at my teacher’s house. Since she, having never married,
had few conventional household obligations, these music lessons, liber-
ally interspersed with periods of conversation, usually lasted the whole
afternoon. The conversations often moved from music to marriage, and
she was decidedly ambivalent on this point. At times, she attributed her
failure to live up to her potential to the fact that she was a non-Brahmin
woman without even a husband to help her. Every now and then a visitor
would stop by, perhaps someone who had been a fan of her father’s. On
one such occasion, the visitor was a lawyer who lived on the next block,
who happened to be passing by as I approached the gate of the house. My
teacher came out to greet me, and the lawyer stopped to introduce him-
self to her, to say how much he had enjoyed her father’s music. After a
few words about the greatness of her father, she turned to her own story,
speaking in English, as she felt the presence of a lawyer called for. ‘‘I sacri-
ficed everything—marriage, children, money—for the sake of music. I am
a helpless lady,’’ she told the lawyer, a perfect stranger, who was listening
seriously and compassionately. The idea of an unmarried woman who has
‘‘devoted’’ herself to music has great appeal in South Indian classical music
circles, and my teacher was convinced that her ‘‘sacrifice’’ had brought her
closer to the divine. Telling of her sacrifices often brought tears to her
eyes. However, although she spoke wistfully about how other female mu-
sicians got chances to perform because their husbands acted as managers,
she believed that music and marriage were basically incompatible.
 She herself cultivated disorder both in her personal appearance and in
her household. The house—a large old-style building with front arches,
verandahs, and a courtyard—had fallen into a state of decay in the years
since her mother’s death in the mid-1980s, but signs of its former illustri-
ousness were everywhere, piled in corners and festooned with cobwebs.
Most fellow musicians thought my teacher highly eccentric, and a few
speculated to me that she might have psychological problems. She, mean-
while, admitted that her life and personality were out of the ordinary but
attributed this to the effects of ‘‘Madras politics,’’ which had forced her
to the margins of the music world. For others—friends and family mem-
bers—her eccentricity was thought to be explainable by the fact that she
was a musician, a true artist. Her cook, who had worked in the house for
more than forty years, would roll her eyes at my teacher’s perversities but,
when pressed, would claim, ‘‘Avakiṭṭe kalai irukke’’ (There is art in her).

A Music Lesson
One day in June 1998 my teacher had decided to teach me the well-known
kriti ‘‘Rāma nannu brōva rā’’ (Rama, come save me). Fingering through a
crumbling book of compositions her father had kept, she commented on
the aptness of the words to her own life, offering me a colloquial trans-
lation of the Telugu lyrics in mixed Tamil and English: Thyagaraja asks
Rama, ‘‘Rāma! Enakku show, fraud, gossip, putting soap, wrong ways-um
teriyātu. Ekiṭṭe nī vara kūṭātā?’’ (Rama, I don’t know show, fraud, gossip,
putting soap, or wrong ways. Why do you not come to me?) Perhaps it
was the quietness of the mid-afternoon lull, or the physical act of going
through one of her father’s old books that brought out a flood of associa-
tions; in the words of the kriti lurked voices other than her own. The first
was that of an American student who had come to learn Karnatic music
from her father perhaps thirty-five years before. She remembered sitting
in the corner of the room while her father sang the words and the stu-
dent repeated them in flat, operatic syllables: raa-ma-naa-nu-bro-va-
raa. In stark contrast was the second voice, that of the celebrated vocalist
M. S. Subbulakshmi (often referred to simply as ‘‘M.S.’’), who had made
this composition famous. My teacher recalled that M.S. used to sing this
song with so much emotion that there would be tears rolling down her
cheeks. That, my teacher said, was because the song related to a period of
unhappiness in M.S.’s life.
It was astonishing to hear something unharmonious about the life of

112 gender and the politics of voice

M.S., who has been celebrated almost universally as the ‘‘greatest female
vocalist of India,’’ as the woman who broke the male stronghold of Kar-
natic music, as the ‘‘only Karnatic musician with a national image’’ (Indira
Menon 1999, 134). At one point, T. T. Krishnamachari, one of the founders
of the Madras Music Academy, called M.S.’s voice ‘‘the voice of the cen-
tury’’ (quoted in ibid., 132). A Tamil biography of M.S. entitled Icai Ulaki
Imayam (The Himalaya of the music world), speaks of M.S.’s voice as the
voice of god, claiming that ‘‘if music can be said to have a form it is M.S.
herself. . . . Her life is the history of music’’ (Sarathy 1997, 5). Born into the
community of devadasis, lower-caste women who performed music and
dance in Hindu temples and by the late nineteenth century were branded
as prostitutes, Subbulakshmi married a prominent Brahmin man of letters,
T. Sadasivam, who brought her talents out into the middle-class Brahmin
music world of Madras. Her life seems to represent the success story that
everyone wants to hear about. What discordant note was there in the life
of a woman who, coming from a devadasi background at a time when
devadasis were being disenfranchised, was able, by sheer force of talent, to
become universally accepted in the Karnatic music world?
My teacher’s answer to that question can be understood less as a truth
about M.S.’s circumstances and more as a sample of discourse about her
life, which has been the subject of much speculation both inside and out-
side the Karnatic music world. My teacher claimed that M.S. sang the song
with such emotion because of the difficulties of her marriage, that she was
mistrusted by her husband, that he accused her of infidelity. While the
present generation may be unaware of such speculations, at that time—
my teacher was referring to the late 1950s—it was the talk of musicians
and rasikas (connoisseurs). Sadasivam had fallen in love with the idea of a
woman vocalist, and he made M.S. into the first really celebrated vocal-
ist the Karnatic music world had known. However, he became envious
of her popularity and, paranoid that her admirers might be lovers, began
to fear that she was a ‘‘characterless lady’’ (my teacher’s expression). After
all, he was a Brahmin, and she was from the caste of devadasis. But, my
teacher wondered, how could he fear such a thing about M.S.? He was the
characterless one to even think such things. He controlled M.S. so abso-
lutely that if she sang something he did not like he would call out to chide
her in public. How could an artist flourish with this kind of husband? ‘‘I
would rather stay in a hostel than be at the mercy of a husband like that,’’
she concluded.
This led her off into a more general discussion of the difference between

gender and the politics of voice 113

men and women. It was okay for a man to ‘‘wander’’ with ten women,
but a woman could not do likewise, she said, as if to criticize the un-
fairness of society. Sensing a familiar sentiment, I was about to join the
critique when it took a sharp turn toward something else. That double
standard, she continued, was as it should be, because there was an essential
difference between men and women. Women were a special birth (vasti-
yāna jamam); there was so much power, honor, and dignity (kauravam) in
a woman’s body that it was important to control it; otherwise it would
be too dangerous.1 ‘‘Men?’’ My teacher spat out the word, as if the very
idea was a joke. ‘‘Who cares? Let them go have ten wives; it doesn’t make
a difference.’’
This music lesson was, in fact, a lesson in the complex ways that classical
music is implicated in present-day discourses about marriage and woman-
hood.2 Most striking here is the way a remembrance of M. S. Subbulak-
shmi provided the occasion for reasserting a particular notion of womanly
virtue. My teacher had used Thyagaraja’s lyrics, presumably composed in
an attitude of humble devotion to Rama, to signify a specifically gen-
dered position: that of a virtuous woman rebuking society. Sung in her
own voice, the lyrics outlined the ambiguity of my teacher’s position. As
a woman who had remained unmarried and ‘‘devoted herself to music,’’
thereby embodying the nationalist ideal of a woman who devotes herself
to the preservation of Indian tradition, she invited admiration; at the same
time, her unmarried state, her lack of domesticity, aroused accusations of
eccentricity and suspicions of abnormality. Sung in M.S.’s voice, the lyrics
outlined the contradictions of M.S.’s position: as an internationally known
artist, she belonged to her public, to her audiences. Her appearance in pub-
lic, however, depended on her enactment of virtuous womanhood during
performances and on having a husband who acted as her manager.
Also striking is the way the story about M.S. juxtaposes several con-
flicting ideals of marriage. My teacher was convinced that a traditional
arranged marriage had been impossible in her case because a traditional
husband would have expected a wife who would bear children and do
housework and not pursue a career as a professional violinist. In M.S.’s
case a traditional marriage would not have worked either; it was her ‘‘love
marriage’’ (and all the implications of social mobility, companionship, and
modernity that such a term carries) that had enabled her to pursue a per-
forming career by allowing her to appear in public as a traditional Brah-
min woman. Finally, the shadow of the devadasi, a woman outside the
bonds of conventional marriage, hung over the conversation, not as an in-

114 gender and the politics of voice

digenous version of liberated womanhood but as an almost unspeakable
contrast against which any female musician had to place herself. The sense
of outrage that my teacher expressed, that Sadasivam could suspect M.S.
of infidelity, indicates the distance she placed between herself and those
‘‘characterless ladies.’’
That afternoon we eventually got back to the music ‘‘itself.’’ But I was
astonished at the distance we had managed to cover, moving from music
to ideas of womanhood, to marriage, and finally to a discourse on the
essential difference between men and women. What history made possible
such a chain of associations?

A Politics of Voice
Toward the end of the nineteenth century, when older temple- and court-
based forms of patronage ceased to be viable in the late colonial economy
of South India, musicians moved in large numbers to the colonial city
of Madras; there, music organizations, concert halls, and academies were
established by an upper-caste, largely Brahmin elite interested in what
they called the ‘‘revival’’ of Karnatic music and its transformation into the
classical music of South India. At the heart of the revival of Karnatic music
was the notion that it could serve as a sign of tradition and Indianness,
as one of the trappings of an emergent middle-class modernity. Just as
particular notions of female respectability and ideal womanhood played a
crucial role in defining the aims of anticolonial nationalism, they were also
central to the project of defining middle-class modernity.3 While devada-
sis came to be regarded as prostitutes and their opportunities to perform
gradually diminished, upper-caste women were encouraged to learn, and
eventually perform, music and dance. Indeed, for many Brahmin elites, the
sign of the successful classicization of music and dance from the 1920s to
1940s was the transformation of these forms into ‘‘arts’’ fit for upper-caste,
middle-class ‘‘family women.’’
My teacher’s comments fit into this larger story of how music became
available to ‘‘respectable’’ women as a vocation and sometimes a career,
and the particular kinds of performance practice, discourse about music,
and notions of ideal womanhood engendered by this newfound respect-
ability. Classical music in twentieth-century South India helped consti-
tute a private or domestic sphere at the same time as it participated in the
production of a new urban, modern public sphere through the establish-
ment of institutions for teaching and disseminating music. Beginning in

gender and the politics of voice 115

the 1930s, the classical concert stage provided a public arena in which the
sound and image of Indian womanhood could be constructed and dis-
played.4 Connections emerged between the way music was placed and dis-
cursively imagined by the newly developing middle-class and what was
happening in terms of actual performance practice onstage. The literal do-
mestication of music as a sign of bourgeois respectability—its connection
to a discourse about family values—parallels music’s progressive interior-
ization within the body in terms of performance practice. The notion of
the artist that underlay the revival of Karnatic music as the classical music
of South India depended on this new sense of interiority.
A certain politics of voice emerged in the moment that upper-caste
women began to sing in public, a politics that involved both privileging
the voice itself as Karnatic music’s locus of authenticity and valuing a cer-
tain kind of voice. By midcentury, the ideal of a voice that came natu-
rally from within, unmediated by performance of any kind, a voice that
seemed to transcend its body, came to be valued as the true voice of Kar-
natic classical music. A number of circumstances enabled the emergence
of this politics of voice: a social reform movement that stressed the virtues
of domesticity and female chastity; the rise of an urban middle class, de-
fined in part by an ideal of traditional Indian womanhood; and, crucially,
the development of technologies of recording and amplifying the voice,
such as the gramophone and microphone. This politics or ideology of voice
emerged, then, not only in live performance but also in moments of sonic
communication mediated by technologies of sound reproduction and the
cultures of listening associated with them.

Social Reform and the Emergence of ‘‘Art’’

Comparing M.S.’s life to the history of music is quite apt, since her conver-
sion from devadasi to married Brahmin woman mirrors the reforms that
were effected on music and dance in the late nineteenth century and early
twentieth. At the end of the nineteenth century, the devadasi ‘‘system’’—a
generalized term referring to a variety of economic, religious, and politi-
cal practices through which women of the devadasi community were em-
ployed by temples—became the object of social reform. The campaign to
end the practice of dedicating girls to a life of service in temples as dancers
and musicians attained success in 1947, when the practice was legally abol-
ished.5 The debates surrounding the issues of how devadasis were to be
defined and whether or not their activities in the temples constituted pros-

116 gender and the politics of voice

titution, and the controversy over the bill to outlaw their practices, had
crucial effects on the idea of ‘‘art’’ and conceptions of women’s relationship
to it.
Kalpana Kannabiran traces the origins of the devadasi abolition move-
ment in the Madras Presidency to the social-reform movement started by
Kandukuri Veeresalingam in what would later become Andhra Pradesh,
in the 1830s. Focusing on women’s emancipation, Veeresalingam was con-
cerned with social hygiene: conjugality and sexual relationships, educa-
tion, religious practices, as well as government corruption. By the last de-
cades of the nineteenth century, not only such social reform movements
but also colonial ideas about prostitution combined to make the devadasi
issue prominent in the agenda for social reform. The Social Purity move-
ment, begun in 1880 in Madras by Raghupati Venkataratnam Naidu, was
influenced by the Purity Crusade in England and America. As Kannabiran
writes, one of the crucial elements of the crusade was the broadening of
the term ‘‘prostitution’’ to refer not only to sexual intercourse for mone-
tary gain, but also as a metaphor for social depravity and moral corruption
in general (1995, 63).6
The development of a discourse about prostitution determined the way
family women were differentiated from devadasis. In a detailed article on
Anglo-Indian legal conceptions of dancing girls between 1800 and 1914,
Kunal Parker traces the process by which dancing girls came to be crimi-
nalized as prostitutes. Crucial in this process was the representation of
dancing girls as a professional group rather than as a caste, which might
have its own laws concerning marriage and property inheritance. The rep-
resentation of dancing girls as a professional group characterized by the
activities of dancing and prostitution brought them under the purview
of Hindu law and led to a public perception that they had fallen from
caste because of their practice of prostitution (Parker 1998, 566). Ruling
that their singing and dancing were merely ‘‘vestigial’’ and that their true
source of income was from prostitution, the Madras High Court denied
dancing girls status as artists (607). Placing prostitution in opposition to
legal Hindu marriage, Parker states, the legislation was ‘‘directed explicitly
towards the valorization of marriage, the construction of a Hindu com-
munity organized around marriage, and the sanitization of Hindu reli-
gious practice’’ (632).
The devadasi abolition movement came to a head under the leader-
ship of Dr. Muthulakshmi Reddy, a medical doctor, legislator, and mem-
ber of the Women’s India Association, who was herself born to a devadasi

gender and the politics of voice 117

mother. Beginning in the 1920s, Reddy launched a campaign against the
practice of dedicating minor girls to temple deities. ‘‘I want the Honorable
members of the House to understand that these [devadasis] are neither de-
scended from heaven nor imported from foreign countries,’’ Reddy began
a speech to the Madras legislative assembly in 1927. ‘‘They . . . are our
kith and kin.’’ ‘‘At an age when they cannot think and act for themselves,’’
she continued, these girls were ‘‘sacrificed to a most blind and degrading
custom.’’ The crime of it was that the innocent girls, if only left alone,
would become ‘‘virtuous and loyal wives, affectionate mothers, and useful
citizens’’ (1928–1931, 3). Reddy implied that girls who became devadasis
were lost to society, that they were useless; their ‘‘sacrifice’’ made them
undependable subjects. In a series of appeals, each with affecting signa-
tures like ‘‘By a Woman’’ or ‘‘From One that Loves the Children,’’ Reddy
drummed up support for her bill. ‘‘The dedication of a girl to a life of vice
is a heinous crime—is it not a worse form of Sati? A hygienic mistake? A
moral monstrosity?’’ (5).
Reddy was helped in her efforts by caste associations of Icai Veḷḷālars
and Sengundars, castes from which devadasis generally came; male mem-
bers of these castes saw the abolition of the devadasi system as a matter of
retrieving the honor and dignity of their caste (Anandi 1991, 741). In 1936
Ramamirtham Ammaiyar, a devadasi from the Icai Veḷḷālar caste who had
rebelled against the system, married a music teacher, became a political
activist, and published a novel in Tamil, Tācikaḷ Mocāvalai Allatu Mati Pea
Mainar (The treacherous net of the devadasis, or a minor grown wise).7 The
novel, a mixture of autobiography and propaganda, follows the lives of
several devadasis who come to the realization that the system is exploiting
them and mobilize to effect legislation.8
Many devadasis opposed abolition, claiming that they were being un-
fairly grouped together with common prostitutes. Others claimed that the
men of the Icai Veḷḷālar caste were supporting the abolition because of ul-
terior motives; they were jealous of the wealth and status that some deva-
dasis had obtained (Kannabiran 1995, 67). The Madras Presidency Devadasi
Association and the Madras Rudrakannikai Sangam issued statements to
counter the abolitionists in the late 1920s. Bangalore Nagaratnammal, a
prominent devadasi who led the protest against male and Brahmin domi-
nation of the Thyagaraja festival at Tiruvaiyaru (and later commissioned
the building of the shrine where the festival now takes place), spoke out
against the legislation, claiming that it denied devadasis not only their
right to own and inherit property but also their status as artists (67).

118 gender and the politics of voice

 Indeed, and perhaps most crucially, the campaign against the devadasi
system helped redefine the status of art, particularly music and dance. If
the official debate about devadasis came, by the early decades of the twen-
tieth century, to center on devadasis’ property rights and alleged prosti-
tution, it is because the matter of their music and dance had been effec-
tively removed from the discourse on devadasis and relocated to a realm
now self-consciously referred to as ‘‘art.’’ Separating the woman question
from the question of what constituted art enabled art to have a trajectory
apart from its practitioners’ lives. It was precisely because revivalists like
E. Krishna Iyer and Rukmini Devi considered the music and dance of the
devadasis to be part of an ancient tradition that extended far beyond the
lives of specific devadasis that these practices could assume the status of
art. In the 1920s and 1930s, at the height of the campaign to end the deva-
dasi system, both Brahmins and non-Brahmins involved with the Tamil
renaissance began to speak about the revival of India’s classical arts. They
used the English word art or the Tamil term kalai to signify a generalized
concept of art, rather than a particular practice like music or painting. In
this discourse, the distinction between art and craft was essential. Where
craft implied hereditary practitioners who worked repetitively and un-
thinkingly, art implied an individual artist, a subject who made choices,
was original, and somehow expressed herself through her art.9
In the early 1930s a number of prominent members of the Con-
gress Party in Madras, including E. Krishna Iyer, Tirumalayya Naidu, and
T. Prakasam defined their position bluntly as ‘‘pro-art,’’ maintaining that
the extremity of Muthulakshmi Reddy’s anti-nautch movement was kill-
ing the art of the devadasis.10 In a pamphlet entitled ‘‘Music and the Anti-
Nautch Movement,’’ written circa 1912, Tirumalayya Naidu, an advocate
by profession, stated that the anti-nautch movement had been negative in
character, set up to demolish the ‘‘long-standing institution’’ of the art of
the devadasis (6). E. Krishna Iyer represented the pro-art position most
vociferously. He was born in 1897 in Kallidaikuricci, known then for its
lavish musical and dance events in connection with weddings. While com-
pleting his law degree at Madras Law College, Krishna Iyer acted female
roles in dramas and studied Karnatic music on the violin. He later joined
the Suguna Vilas Sabha, a prominent theatrical group in Madras, and re-
ceived formal training in sadir, the dance form of the devadasis. Com-
mitted to reviving the dance, he was instrumental in starting the Madras
Music Academy and in bringing dancers, first devadasis and then Brah-
mins, to its stage. In a series of letters against Muthulakshmi Reddy’s con-

gender and the politics of voice 119

demnation of all public performance of nautch published in the Madras
Mail in 1932, Krishna Iyer mobilized public support for his pro-art posi-
tion. In addition, he was behind the Music Academy’s 1932 resolution to
rename the dance ‘‘Bharata Natyam,’’ or ‘‘Indian dance.’’ The word Bha-
rata gave the dance an image of national importance; at the same time the
use of the Sanskrit word natya suggested its origin in the Sanskrit treatises
on dance.
The idea was not necessarily to help the devadasis continue to practice
their arts; it was rather to rid music and dance of their impure associations.
Music and dance had to be rescued from the hands of degenerate deva-
dasis and taken up by women from respectable (that is, Brahmin) fami-
lies. The most prominent upholder of this idea was Rukmini Devi, one
of the first Brahmin women in the twentieth century to perform South
Indian dance and the founder of Kalakshetra, one of the first institutions
to teach dance and music in Madras. Devi stated that her goal was to prove
that ‘‘girls from good families’’ could dance and that they no longer had
to depend on traditional dance teachers (Allen 1997, 64). Influenced by
Theosophy and the idea of the original devadasis as a ‘‘band of pure vir-
gin devotees,’’ Devi reconceptualized the dance to stress its religious and
spiritual aspects, presenting the dancer as a chaste and holy woman.11 Im-
portantly, the shift from devadasis to Brahmins involved not only a new
kind of woman but also a new kind of artist, one who was an individual
interpreter, rather than merely a hereditary practitioner of the art. In a
pamphlet entitled ‘‘The Creative Spirit’’ written in the early 1940s, Devi
described the shift as an awakening from the merely physical level of the
‘‘acrobat’’ to the ‘‘meaning’’ and ‘‘expression’’ conveyed by the slightest
movements of the dancer.12 ‘‘A tiny finger lifted with meaning,’’ she con-
cluded, ‘‘is far more thrilling than all the turns and gyrations and tricks
of the circus performer.’’ 13 Notions of chaste womanly behavior here con-
verge with the idea of an art whose basic currencies are ‘‘meaning’’ and
‘‘expression.’’
Amrit Srinivasan notes that although the devadasi abolition legislation
was not officially passed until 1947, the combination of social reform and
purification involved in transforming music and dance into classical arts
in the first three decades of the twentieth century had already effectively
prevented devadasis from continuing their traditions. The bill, she writes,
‘‘seemed to have been pushed through not so much to deal the death of
the Tamil caste of professional dancers as to approve and permit the birth
of a new elite class of amateur performers’’ (1985, 1875). Matthew Allen

120 gender and the politics of voice

has noted that ‘‘in the face of overwhelming social pressures, a signifi-
cant number of women from the traditional dancing community [stopped
dancing but] nevertheless continued in the profession of musical (most
often vocal) performance,’’ but that since the 1990s very few of the female
descendants of this community have chosen to go into musical perfor-
mance (1997, 68). Many Brahmin women did not perform onstage in the
1920s and 1930s for fear of being mistaken as devadasis (Bullard 1998, 128).
By the 1950s, after the passage of the Devadasi Dedication Abolition Act,
more and more Brahmins were taking up music and dance; both arts be-
came desirable talents for women of marriageable age. The entry of Brah-
min women as singers onto the concert stage solidified the developing
caste rift: many felt that female Brahmin singers in particular could not
sit next to Icai Veḷḷālar accompanists on stage (ibid., 128, 263). The rise of
Brahmin women as performing musicians thus served as a catalyst to the
Brahminizing of music as a profession. Since the 1980s, Karnatic music and
Bharata Natyam have become almost exclusive markers of middle-class
English-educated Brahmin identity.

The Voice of the Century

In 1933 E. Krishna Iyer wrote of the ‘‘sweetness of natural music, as found
in the voices of women, young boys, and singing birds’’ (xvi). If the listen-
ing public, or the ‘‘democracy,’’ as he called them, seemed to be taking a
new interest in such sweet sounds, it was because Karnatic music had lost
its sweetness. ‘‘It is but natural,’’ he wrote, ‘‘in the general dearth of good
and well-trained voices, and scared away by the excesses of dry acrobat-
ics of the musical experts,’’ that the public should desire music that was
pleasurable to listen to (xvi). Women inherently possessed the raw ma-
terial of music. In a sketch of Saraswati Bai (1894–1974), the first Brahmin
woman to sing on the concert stage, he wrote that ‘‘women vocalists are
found to possess certain desirable advantages over men. They have pleasant
voices to begin with and none of the contortions of the struggling male
musicians. They do not fight with their accompanists, who usually follow
them closely. They are free from acrobatics of any kind and they seldom
overdo anything’’ (46). Krishna Iyer wrote of the discrimination Saraswati
Bai faced from a Madras sabha where she had been scheduled to give a
kalakshepam performance: ‘‘The music world then [around 1910] was not
as liberal as it is now,’’ he remarked (49). The boycott of Saraswati Bai’s per-
formance by her male accompanist at the Madras sabha stands in contrast

gender and the politics of voice 121

to the enthusiastic reception she had at a wedding performance several
years later, where the audience was attracted by the novelty of hearing a
‘‘lady bhagavatar.’’

It was one of the marriage seasons at Kallidaikurichi—the Brahmin

Chettinad of the Tinnevelly District as it then was—notorious for the
lavish expenditure of its fortunes on spectacular marriages and choice
musicians. A huge concourse of people were hovering in and about a
pandal [enclosure with a makeshift stage] to hear the beginning of a
Kalakshepam—the daring feat of a new fledged lady bhagavatar—and
then to decide whether to remain or disperse. . . . Scarcely did the frail
form of a young girl of fifteen or sixteen years of age sound her jalar
[small metal cymbals] and go through her opening song when the scat-
tered crowds closed into the pandal to its full with barely standing space
for the musician herself. The organisers had no small difficulty in ac-
commodating the vast crowds that sat through the performance with
eager interest. . . . Her performances invariably draw crowded houses
and the ladies’ section of the audience is generally overfull. (47)

What accounted for such excitement about a female musician? While

Krishna Iyer suggested that women’s music, without the competition and
‘‘acrobatics,’’ had a more ‘‘natural’’ feel, he also implied that the rising popu-
larity of female vocalists was in part due to the explosion of gramophone
recordings. In the early years of recording in South India, more records
were made of female vocalists than male vocalists (Indira Menon 1999, 74).
Gramophone records, made to be played in the home, needed a differ-
ent kind of appeal than concerts; the competition and the acrobatics that
might make a concert exciting would be lost on a gramophone record,
where anything that did not go as planned might later simply sound like
a recording glitch.
Gramophone records also provided a solution to the problems faced by
female musicians, whether from the devadasi community or the Brahmin
community: recording presented a way for women from both commu-
nities to be heard without being seen, to escape association with their
bodies. A social reform movement that stressed the virtues of marriage
and domesticity and referred to the activities of devadasis as ‘‘debauchery’’
and ‘‘prostitution’’ left female musicians of the devadasi community with
few traditional performance venues and almost no opportunities to per-
form in the newly built concert halls of Madras. Gramophone companies,
however, initially run not by Brahmins but by Americans and Europeans,

122 gender and the politics of voice

actively recruited devadasi women for their first recordings.14 Unlike All
India Radio, which was founded in the 1930s as a vehicle for the nation-
alist project of making music respectable (thus denying broadcasting op-
portunities to devadasi women), the gramophone companies were purely
capitalist enterprises. Between 1910 and 1930, their best-selling recordings
in South India were of Dhanakoti Ammal, Bangalore Nagaratnammal,
Bangalore Thayi, Coimbatore Thayi, M. Shanmughavadivu, Veena Dha-
nammal, and Madras Lalithangi, all women from the devadasi community
(Indira Menon 1999, 74–75; Kinnear 1994).
Meanwhile, for Brahmin women, recording provided a way to sing
for the public without appearing in public and jeopardizing their respect-
ability. In the early 1930s Columbia Records ‘‘discovered’’ a girl genius,
the thirteen-year-old D. K. Pattammal (b. 1919). Many years later, in a
speech given at the Madras Music Academy, Pattammal stated that her
success, in the absence of a family connection to music, was due to the
gramophone. As a Brahmin girl growing up in an orthodox family, she was
unable to undergo the traditional gurukulavasam, which would require
leaving home to live with a male, non-Brahmin teacher. Instead, she sup-
plemented her brief periods of learning from various teachers with listen-
ing to gramophone records. When she was ready to perform, the gramo-
phone, as an interface between private and public, provided her with a
way to sing without being seen. The radio, similarly, provided Brahmin
women an opportunity to perform without emerging in public; a woman
described to me how in the 1930s, as a girl in her late teens and twenties,
she was forbidden to sing concerts, but her father would chaperone her to
the Madras All India Radio station each week for a broadcast.15
The gramophone companies sought out novelty and found a ready
source of it in girl singers. By the 1920s, recordings of them were so com-
mon that G. Venkatachalam, a patron of many artists and musicians at that
time, referred to them with distaste as ‘‘baby stunts.’’ Of his first meet-
ing with M. S. Subbulakshmi in 1929, he wrote, ‘‘She was 13 when I met
her. Subbulakshmi came to Bangalore to record her songs by His Master’s
Voice (hmv) Company. ‘We are recording an extraordinarily talented girl
from Madras. Would you care to listen to her and tell us what you think
of her?’ was the cordial invitation from my friend, the manager. My first
reactions were: ‘Ah! Another of those baby stunts!’ I went, however, and
met not a fake, but a real girl genius’’ (1966, 65).
The idea of novelty, so crucial to the gramophone companies’ success,
was intimately tied up with notions of the child prodigy.16 If earlier dis-

gender and the politics of voice 123

course on Karnatic musicians stressed the importance of seasoned experi-
ence and long years of discipleship, the figure of the prodigy stood in stark
contrast. The prodigy was independent of the traditional gurukula sys-
tem in which the student lived in the guru’s household. The figure of the
prodigy suggests a preternatural ability or gift, an element that goes be-
yond nature into the realm of the supernatural. It implies a certain iso-
lation or protection from the world, an incompleteness compensated for
by a larger-than-life voice and a selfless devotion to music. The notion of
the prodigy complicates classic formulations of agency: the prodigy does
not sing because she desires to, but because a voice sings through her—
her body is merely a vessel. As Felicia Miller Frank has pointed out, the
association between women, the voice, and the artificial or technological
has a long history in European discourse on the arts. Female prodigies,
in this discourse, are represented as sexless or artificial angels, emblematic
of the sublime and of artistic modernity (Frank 1995, 2). Among the elite
of twentieth-century South India, who borrowed much of this European
discourse about art, it is no coincidence that the word prodigy was first used
with regard to female musicians and that the prodigy’s first vehicle was
the gramophone. In South India the female voice, disassociated from its
body, came to be thought of as the essence of music itself.
Of the female musicians that the gramophone companies popularized,
M. S. Subbulakshmi became by far the most famous. What made hers
‘‘the voice of the century’’? M.S. was born in 1916 in Madurai to a veena
player from the devadasi community named Madurai Shanmughavadivu.17
In her memoir of a childhood steeped in music M.S. recalled, ‘‘I was fasci-
nated by records—gramophone plates, we called them. Inspired by the
gramophone company’s logo of the dog listening to his master’s voice, I
would pick up a sheet of paper, roll it into a long cone, and sing into it
for hours’’ (Ramnarayan 1997, 10). Shanmughavadivu was ambitious for
her daughter and brought her to her own recording sessions in Madras,
where she persuaded the hmv company to record the thirteen-year-old
Subbulakshmi. The records sold well and M.S. began to get concert op-
portunities in Madras. In the early 1930s she and her mother moved from
Madurai to Madras for the sake of M.S.’s career. Her concerts attracted
the elite of Madras at that time, a group of Brahmin or other high-caste
men who considered themselves aestheticians, journalists, and freedom
fighters. Among them were Kalki Krishnamoorthy, journalist for the
Tamil weekly Ananda Vikatan, who later started his own magazine called
Kalki, in which he wrote a weekly column on music and dance; ‘‘Rasi-

124 gender and the politics of voice

15 M. S. Subbalakshmi record jacket (1945).

kamani’’ (Gem among connoisseurs) T. K. Chidambaranatha Mudaliar,

aesthetician and man of letters; the director K. Subramaniam, who be-
came known for his patriotic films; and the journalist and freedom fighter
T. Sadasivam.
Sadasivam, a widower with two daughters and ten years older than
M.S., married her in 1940. He became not only her husband but also the
manager of her career, overseeing her acting in films such as Shakuntala
(1940) and Meera (1945), and coaching her in what to sing on the concert
stage.18 It is said that he would sit in the front row of the audience during
her concerts and plan every detail of her concert programs (Sarathy 1997,
169). As her public-relations man, Sadasivam introduced her to Gandhi and
Nehru and a host of other political figures. M.S. became a larger-than-life
presence not simply by her musical talent but also by Sadasivam’s careful
cultivation of her persona as a singer of pan-Indian and international ap-

gender and the politics of voice 125

peal. If the nation had a voice, Sadasivam at least thought he knew what
it sounded like.
Not only the gramophone but also the microphone shaped M.S.’s voice.
The beginning of her public singing career coincided with the estab-
lishment of large concert halls in Madras. Essential to these halls was
the microphone, which began to be used in South India in the 1930s.
The microphone allowed Karnatic vocalists and instrumentalists to con-
centrate less on projecting volume and more on shifts in dynamics and
speed. It also allowed those with softer voices (including many women)
to sing to large audiences. As noted in chapter 2, many musicians attribute
changes in vocal style in Karnatic music in the twentieth century to the use
of the microphone, particularly the shift from an earlier, higher-pitched
style of singing accompanied by more gesturing—often referred to today
as ‘‘shouting’’—to a lower-pitched, more introspective style. The micro-
phone also narrows the physical range of a performer, serving as a kind of
ballast for a singer or violinist, limiting the distance he or she can move.
At the same time, one of the microphone’s most prominent effects is the
projection of intimacy to vast audiences.19 As one listener said of hearing
M.S. in a concert hall, ‘‘It seemed like we were overhearing a conversation
in which a devotee spoke intimately to her God.’’ 20 M.S.’s voice and career,
notes Indira Menon, were products of the microphone: ‘‘The greatest gift
to the world of this little instrument of the technological revolution is
M.S. Subbulakshmi. The innumerable nuances of her multi-faceted voice
can be captured by it. . . . She is truly the product of the modern age—of
the sabha culture, teeming audiences, and large halls where her voice can
soar—thanks to the mike’’ (1999, 89). It is crucial to note that the voice
with ‘‘innumerable nuances’’ is itself a product of the microphone; as the
microphone makes these nuances audible in a physical sense, it also makes
them available to an aesthetic discourse about music.
By the 1940s, M.S. and D. K. Pattammal had come to dominate the mar-
ket for female voices and would do so until the 1970s. Both sang in films
as well as recordings and classical stage performances, but their personae
were quite different. The two singers had essentially different audiences, as
one older concertgoer told me. Pattammal was ‘‘dependable, intellectual,’’
her voice had a ‘‘weight’’ to it, and audiences who prized these charac-
teristics would attend her concerts. After M.S.’s concerts, however, audi-
ences came away ecstatic, raving about the fine qualities of her voice and
her expression. According to her fans, M.S. concentrated on conveying

126 gender and the politics of voice

bhakti (devotion) and bhāvam (emotion) through her singing, always pay-
ing great attention to the pronunciation of lyrics and singing not only
Karnatic compositions but also Hindi bhajans and Sanskrit chants. Only
after Pattammal released a record with a pallavi did M.S. begin to include
more ‘‘intellectual’’ items in her concerts. ‘‘Natural’’ music was considered
the opposite of intellectual feats. It is worth noting Venkatachalam’s de-
scription of M.S.:

Her voice has the rich cadence of a mountain stream and the purity of
a veena-note. . . . She takes the highest notes with the effortlessness
of a nightingale’s flight to its mate. This is an art by itself. And when
you consider how even some of the great vidwans and ustads [the Hindu
and Muslim terms for ‘‘artist,’’ respectively] contort their faces and make
ridiculous caricatures of themselves in such attempts, it is some con-
solation to see a natural face for once. Women (because of their innate
vanity, I suppose!) avoid that exhibition of agonized looks and tortured
faces! Her recitals have not the long drawn out boredom of the ordi-
nary South Indian cutcheries [Tamil kaccēri ]. . . . It is the art of music
she wishes to display and not its mathematics.’’ (1966, 66–67)

What emerges in writing and talk about M.S.’s voice is a new discourse
about the natural voice; constantly remarked on are her genuineness on
stage, her complete emotional involvement as she sings. Recent biogra-
phies play up M.S.’s personal humility and innocence, opposing it to the
‘‘grandeur’’ and ‘‘majesty’’ of her voice, noting that she ‘‘knows nothing
but music’’ and when offstage is a simple housewife like any other (Indira
Menon 1999, 141).21
Many descriptions of her voice use the metaphor of flight, implying at
once an escape from the body and an association with nature: ‘‘the effort-
lessness of a nightingale’s flight to its mate.’’ Reviews of M.S.’s records are
replete with metaphors of bodilessness or dissolution of the body. Good
music should ‘‘melt’’ (kēṭṭavarkaḷai urukki) the listener, make the mind
‘‘swoon’’ (uḷḷam māyaṅkum), or the listener ‘‘forget himself ’’ (mey marantu:
literally, ‘‘forget the body’’). In a 1942 review, after remarking that the
gramophone record ‘‘captures’’ the range of M.S.’s voice from its lowest to
highest notes, Kalki Krishnamurthy comments, with regard to a particular
line of a song, ‘‘If you hear it once, you’ll have the desire to hear it again
and again, a thousand times. Good thing this is a music record and there
is no problem hearing it again and again!’’ (75). It is as if the ideal listen-

gender and the politics of voice 127

ing experience had come to be identified with listening to gramophone
records; the recognition of the perfection of M.S.’s voice is linked to its
disembodiment and mechanical repetition on the record.22
Comments on the naturalness of M.S.’s voice found their way into a dis-
course about art as well; reviewers slipped easily from descriptions of her
performances or recordings to ideas about what art should be. The letters
of Rasikamani T. K. Chidambaranatha Mudaliar from the 1940s are laden
with references to M.S.’s performances. After hearing one of her radio
concerts, he wrote that her rendition of a song had ‘‘melted the listeners.
That is art, is it not? It must make our feelings mingle with Truth [tat-
tuvam], and make us into different beings [vēu vastuvāka akkiviṭavēṇṭum]’’
(Maharajan 1979, 136). However, as Venkatachalam noted, art was also for
more mundane purposes.

‘‘What six singers would you select from India to represent this coun-
try in a World Music Festival?’’ asked a well-known European impre-
sario not long ago, and I had no hesitation in suggesting Roshanara and
Kesarbai for Hindustani music; M.S. Subbulakshmi and D.K. Pattam-
mal for the Carnatic; Kamala Jharia and Kananbala for popular songs
[all women artists].
This list would, of course, amuse the orthodox and the music pun-
dits. ‘‘What about Faiyaz Khan and Omkarnath?’’ I can hear the Hindu-
staniwalas shout. ‘‘Why not Aryakudi and Musiri?’’ will be the Tamils’
cry. The film fans will plead for Saigal and Kurshid [all male artists].
I am not unmindful of the merits of the above. . . . But I wouldn’t
risk them in a World Jamboree. They will simply not be understood.
Terrible is the ignorance of the world to things Indian, and unbridge-
able as yet is the gulf between the music of the East and the West. . . .
People of the world understand art and appreciate artistes. . . . And I
claim, my six are artistes! (1966, 68)

Two points warrant attention here. First, it is female vocalists, not male,
who are thought to represent not only India but also art itself in the new
global market; Venkatachalam implies that although the male vocalists had
a local following, they would not be able to travel as representatives of
India. Second, art—here, the art of music—is identified with a naturalness
that is gendered as female.23
By the 1940s, the connection between women and music had ascended
to the level of an assumption. In a series of essays written in English
and published by the Theosophical Society, Rukmini Devi elaborated on

128 gender and the politics of voice

the special connection that she felt existed between women and the fine
arts. ‘‘The spirit of womanhood,’’ she wrote, ‘‘is the spirit of the artist.’’ 24
Women possessed an ‘‘innate refinement’’ which made it incumbent on
them to take up the revival of Indian art: ‘‘Innate as this refinement is, the
study and practice of such arts as dancing, music, and painting can dis-
tinctly help in its development. . . . Without the expression of the arts in
life, in the very home itself, there can be no real refinement and of course
no real civilization.’’ 25 To be an artist meant not necessarily producing music
or art but being art itself: ‘‘It is necessary that in her walk, in her speech and
her actions she should be the very embodiment of beauty and grace.’’ 26 In
learning the arts, what was necessary was ‘‘rigorous work, the complete
subjugation of all other personal desires and pleasures, the abandonment
of one’s being to the Cause.’’ 27 The success of the Indian nation, as well as
the Indian arts, depended on restoring woman from her degraded position
to that of ‘‘a divine influence rising above the material aspect of things.’’ 28
Devi imagined the relationship between voice and body to be analo-
gous to the relationship between music and dance. Although she could
not completely efface the physicality or eroticism of dance, her writings
point to the necessity, in her view, of putting its physicality to some higher
use. Music was, for her, the divine influence that would insure this. In
her essay ‘‘Dance and Music’’ she wrote that music, ‘‘the basic language of
Gods,’’ was what saved dance from being ‘‘mere physical acrobatics.’’ 29 It
was the ‘‘universal language of the soul,’’ the ‘‘saving grace of humanity.’’ 30
In such a conception the materiality of music is effaced. Instead of being
seen as social, something which comes into being through performance
and the mediation of human actors, it is seen as a kind of pure voice from
within, a voice deeply contained within the body, but neither connected
to nor manifested on its surface.

The Limits of Performance

In her article on the agenda of the Madras Music Academy between 1930
and 1947, Lakshmi Subramanian states that the reorientation of music and
dance—performance style, repertoire, and the performers themselves—
toward an urban, largely Brahmin, middle class was part of a ‘‘sanitizing’’
mission that emphasized the spirituality of Karnatic music and Bharata
Natyam over their sensual aspects (1999a). On the dance stage, this meant
that ‘‘spiritual love’’ and ‘‘restrained devotion,’’ rather than ‘‘sensual experi-
ence’’ and ‘‘raw passion,’’ had to be projected (82).31 On the music stage,

gender and the politics of voice 129

the sanitizing project produced a distinction between ‘‘classical’’ and ‘‘light
classical’’ music; the compositions of the trinity of composer-saints came
to be considered the central repertoire of Karnatic music, while padams
and javalis (associated with the devadasi tradition) were considered ‘‘light
classical’’ (83).32 The compositions of the trinity, which expressed spiritual
devotion to Rama, were seen as the appropriate subject matter for classical
music. This reorientation of Karnatic music toward the spiritual realm was
accompanied by the changes in infrastructure discussed in chapter 2—the
establishment of concert halls, music sabhas, and music-teaching institu-
tions—that created a public sphere distinctly different from the temple and
courtly milieus in which Karnatic music had flourished in the nineteenth
century.
A central effect of the shift from these settings to the public concert hall
was that music and dance were separated from each other, physically and
conceptually.33 Under the reforming eye of Rukmini Devi, musicians—
who had previously accompanied dancers by moving with or behind them
on stage—were seated at the left end of the stage, presumably to give them
more respect. However, musicians who performed for dance never got as
much respect as concert musicians; their music was seen as somehow not
pure or classical enough (Allen 1997, 66). E. Krishna Iyer, in his 1933 sketch
on the dancer Balasaraswati, wrote that the art of Bharata Natyam needed
to be ‘‘overhauled if it should have any real appeal in these days’’ (98).
Changing the character of the accompanying music could, he suggested,
bring about the desired change in atmosphere. ‘‘If possible the vociferous
clarionet will have to be substituted by the flute or other more agreeable
and indigenous instruments. The noise of the jalar [small metal cymbals
used to beat the tala] in the hands of the nattuvans [musicians who keep
the tala with these cymbals] will have to be controlled and reduced con-
siderably so as to allow the beautiful sound of the ankle bells to be heard’’
(98). ‘‘Noise’’ implied a spectacle, an excess of performance and an unruly
audience, while ‘‘beautiful sound’’ evoked the image of a silent audience
appreciating an aesthetic production.
Perhaps more important, dancers stopped singing. Unlike traditional
devadasis, who sang as they danced, the Bharata Natyam dancers trained at
Rukmini Devi’s Kalakshetra not only did not sing but did a minimum of
abhinaya (facial gestures and/or mime, often contrasted with ‘‘pure dance’’).
Krishna Iyer suggested that the ‘‘low atmosphere’’ of nautch performances
was due to a kind of overperformance of abhinaya: ‘‘The over-developed
technique of the art, admirable as it is in much of the details of abhinaya . . .

130 gender and the politics of voice

has to be kept within limits and desirable proportions so as not to obscure
or interfere with the natural grace of movements and poses’’ (1933, 98). If
respectable dancers no longer sang, it was perhaps because the sound of
their voice would add a kind of unseemly interiority to the performance.
On the music stage, the reverse was true: respectable musicians did not
dance when they sang, for this would interfere with the singer’s projec-
tion of interiority. The association of dance with something improper in
music can be seen in a 1941 article in Nāṭṭiyam entitled ‘‘Apiayam Vēṇ-
ṭām’’ (We do not want abhinaya). The author, Gottuvadyam Durayappa
Bhagavatar, turned the meaning of the verb apiayi from the rather neu-
tral idea of miming or ‘‘showing in gestures’’ to the negative connotation
of ‘‘imitation,’’ suggesting that many young musicians, instead of devel-
oping their own original style, were simply imitating others. A term from
dance became a metaphor for imitation, a kind of excess of performance
that would be unseemly on the music stage.
What were the implications of this separation for the performing sub-
ject? For female musicians in particular, a convention of music perfor-
mance developed in which the body was effaced; too much physical move-
ment or ‘‘show’’ on the stage was seen not only as extraneous to the
music but as unseemly. In 1939 the violinist C. Subrahmanya Ayyar wrote
that excessive gesturing by singers had become so prevalent that there
were special Tamil terms coined to describe them: iyantiram araikkiatu (a
phrase suggesting the motions of a machine hulling paddy) and mallayuttam
(wrestling). ‘‘Sometime back,’’ he noted disapprovingly, ‘‘the Tamil weekly
Ananta Vikatan published photos of such ugly poses of several vocalists,
but the satire was lost on them; for they deemed the photos as a matter of
advertisement for their renown in music’’ (1939, 45).
The ideal became a kind of performance of nonperformance: nothing
visible was supposed to happen on the music stage. Until D. K. Pattammal
began singing ragam-tanam-pallavi (a highly intellectual type of improvi-
sation) onstage in the 1940s, and even after that, women musicians had
confined themselves to performing only compositions; their only impro-
visation was a small raga alapana.34 Singing compositions was not really
considered to be performance. Women’s voices, it was said, were not suited
to the competitive type of music that men sang, where they tried to up-
stage each other in improvisatory virtuosity. In the late 1930s a reviewer in
the Tamil magazine Bharata Mani wrote that ‘‘women’s music needs to have
its own character; women’s concerts should not just be a copy of men’s
concerts. We do not need pallavis and elaborate raga alapana from women.

gender and the politics of voice 131

Padams and javalis are being forgotten as we speak. It would be good for
women to give importance to them’’ (‘‘Ceai caṅkīta vitvat capai: 12-
āvatu mēlam’’ 1938, 372). Women’s activity on the stage, suggested R. Ran-
garamanuja Ayyangar, a prominent male musician from the 1940s to the
1970s, should be akin to housekeeping. ‘‘The craze for sahitya neeraval and
swara prastara [two types of improvisation] as a prestige symbol detracts
from the inherent charm of a lady’s voice. Lady musicians can resuscitate
values . . . if they concentrate on repertoire. Hundreds of classical com-
positions need to be revived. The Hindu housewife has a reputation for
keeping alive religious traditions’’ (quoted in Indira Menon 1999, 102–3).
By such logic, family values are indistinguishable from the values of clas-
sical music.
The effacement of the female musician’s body on the concert stage is
accomplished not only by a lack of gesture but also by an implicit dress
code that ensures that all female musicians look the same. Unlike Western
concert black, which is supposed to make the musician’s body symbolically
invisible, Karnatic concert attire makes female musician’s bodies visible as
a certain type: a respectable family woman.35 Whereas men’s appearances
on the Karnatic concert stage vary—pants, veshtis (cloth worn by men over
the legs as a long skirt), silk jibbahs (tunics), Western-style button-down
shirts, hairstyles from topknots to short haircuts—women’s appearances
have changed little over the years. All female musicians wear a fairly lavish
(but not too lavish) silk sari and have their hair pulled back in a neat, tight
plait or bun. Whereas male musicians who present an eccentric appear-
ance are often thought to be geniuses, female musicians whose appearance
differs from the norm provoke negative comments; their appearance is
considered extramusical excess, an unseemly element. My violin teacher,
whose unkempt appearance was a source of anxiety to her family mem-
bers, often received advice about how she might improve her appearance
so that she might be called for concerts more often. She compared her
own ‘‘fate’’ to that of a male flutist she had accompanied many times when
he was drunk and disheveled on stage; his eccentricities, however, were
universally hailed as a sign of his genius. Eccentric women, my teacher
pointed out, were not likely to be taken as geniuses.
The association between female respectability and classical music is so
strong that metaphors of dress often motivate conversations about music.
For several months in 1998 I took lessons with a female vocalist in her
fifties. Her persona in the Karnatic music world was one of intellectual
musicianship, a rigor devoid of the show or gimmicks often associated

132 gender and the politics of voice

16 Women’s clothing and classical music as interchangeable signs of tradition. Binny Silks ad-
vertisement (1994). Photograph by Amanda Weidman.

with younger male performers. In a sense, this ‘‘intellectual’’ image was

the only one available for a female musician of her age, neither old enough
to be a pioneer nor young enough to be a rising star. On one occasion she
remarked to me that if she had been born some twenty years earlier, she
would have been M. S. Subbulakshmi, as if M.S. had simply occupied a
slot that had become available in the public world of Karnatic music in
the 1930s. Although she did indeed put emphasis on elements usually de-
scribed as intellectual, such as pallavi and a ganam style of singing alapana,
more important was the way intellectual, in critics’ and audience mem-
bers’ descriptions of her music, serves as a code word for a female musi-
cian whose music is without ‘‘feminine’’ charm. The opposition between
showiness and intellect is much less stark for male musicians, if it operates
at all.
After our strictly timed one-hour lesson, she would, depending on her
mood, ask me personal questions. One day she asked me why I always
showed up in a sari, while other young women my age were wearing salwar
kameez or even ‘‘jeans’’ (the summary term for all Western dress). I replied
that by wearing the sari I got more respect and was even, on some occa-
sions, able to pass as some kind of Indian. The conversation then turned
to music. She expressed her distaste for many current trends in Karnatic

gender and the politics of voice 133

17 Sari advertisement in Sruti ( January 2003).

music performance, its so-called innovations. Traditional music and tra-

ditional dress were unassailable signs of Indianness; despite the fact that
one was intangible and the other quite literally material, they were often
used as signs for and of each other. ‘‘You tie a sari and you get respect,’’
she told me. ‘‘That is sangitam.’’ Advertisements for saris often play on the
same slippage between music and clothing as symbols of tradition. In an
advertisement for Binny Silks, shown on a billboard in Madras in 1994, a
woman dressed in a lavish, decorously draped silk sari sits with a veena,
with the caption ‘‘Traditional beauty made to grace any occasion.’’ In 2003
a newspaper advertisement played on the word sari and the musical syl-
lables sa and ri to suggest that the vendor’s saris were the most traditional
ones available.
The metaphoric opposition between prostitution and chastity, between
uncontrolled female sexuality and domestic womanhood, continues to de-
termine the definition and limits of performance in Karnatic music.36 Dur-
ing the period in 1998 when I had daily violin lessons with my teacher in
Madras, I was always impatient to do swara kalpana, a type of improvisa-
tion that takes place within the tala cycle after a composition. My teacher
was generally uninterested in it. ‘‘It is not really music,’’ she would say.
‘‘Just calculations.’’ Raga alapana (also called ragam) was the most impor-
tant thing. She remembered as a young woman also having had a ‘‘craze’’
for swara kalpana. She asked a male musician she accompanied at the time

134 gender and the politics of voice

to teach her some of his kalpana tricks. He had berated her: ‘‘Paittiyam
poṇṇu! [Crazy girl!] You play such beautiful ragam. Why do you want to
ruin it with this cheap stuff ?’’ She had come to agree with him. ‘‘All this
is just kavarcci [attraction; also, sexual attraction], just feats with the tala,
like a characterless woman. But ragam—that is like your mother.’’ Ragam,
she explained, should be born within: first you enjoy it inside yourself, and
then it comes out for others. Swara kalpana, for her, could not claim the
same purity of origin; calculation [kaṇakku: the same word as that used for
accounting, giving it a businesslike connotation] gave it a kind of external,
unseemly quality.37 To imitate certain vocalists doing swara kalpana, she
began to slap her leg loudly in mock tala-keeping and barked out an un-
melodic string of swaras. Then, to demonstrate raga alapana, she assumed
a posture of utter stillness, turned her palms upward, closed her eyes, and
began to hum.38
Such restrictions might seem antiquated, yet they continue, perhaps
in subtler form, in present-day conventions of Karnatic music perfor-
mance. In 1998 I attended a concert by the immensely popular young
vocalist Sowmya, who was accompanied by M. Narmada on the violin.
Both women were dressed, as usual, in appropriately lavish silk saris. The
concert was uneventful. However, it is precisely the uneventfulness that
is part of the aesthetic production: nothing out of the ordinary is sup-
posed to happen. During her raga alapana, Narmada played with closed
eyes, face screwed up perhaps in concentration, perhaps to avoid the gaze
of the audience. Meanwhile, during the tradeoffs in swara kalpana, when
the concert would presumably reach a fever pitch of excitement, Sowmya,
waiting for her turn to come around, adjusted her sari, refolded the hand-
kerchief in her lap, and checked her watch, as if nothing exciting were
happening. The more classical the music, the less there is to watch on stage;
if anything, musicians perform a kind of interiority through bodily pos-
ture.39

Family Values . . .
Notions of classical music and musicianship became conflated with notions
of ideal womanhood in twentieth-century South India through a particu-
lar model of domesticity. The Tamil nationalist poet and essayist Subra-
mania Bharatiyar wrote in 1909 of the need for kuṭumpa strīkaḷ (family
women) to take up music. His essay ‘‘Caṅkīta Viṣayam’’ was concerned
with the problem of what he conceived of as a loss of musical sensibility

gender and the politics of voice 135

among Tamils. The best way of stemming this loss, he suggested, was to
improve women’s music in the home—a kind of trickle-up approach that
tapped the natural musicality of women. If family women could be given a
proper grasp and appreciation of music, the rest of society would improve.
For women, he stated, ‘‘have an especial connection to music’’ (1909, 222);
if one prohibited women from singing, one would find that one had not
only no music but no life at all (221). If women could just be taught to
sing correctly what they were already singing in the home as folk ditties
and lullabies, the connection would be realized. Bharatiyar offered sev-
eral practical suggestions. First, women must acquire tāla ñāam (a rhyth-
mic sense, literally, knowledge of tala). Some said that women innately
lacked tala sense, but Bharatiyar maintained that it was a matter of prac-
tice. To those who objected on the grounds that family women were not
tācis (devadasis) and did not need to sing concerts, he responded that sing-
ing incorrectly, without regard for the tala, sounded virācam (vulgar) (221).
For those women who wanted to sing, Bharatiyar recommended that they
hire a teacher to teach them proper voice culture and that they stay away
from nāṭaka meṭṭu (drama tunes), cheap songs with koccai moi (slang), and,
most of all, English and Parsi songs (227–28). For family women interested
in learning an instrument, Bharatiyar recommended the veena, with its
calm, soft sound, as being particularly appropriate. ‘‘If more women played
on the veena, there would be a greater appreciation of aesthetic beauty
and the niceties of life [rācapayircci, vākkai nayam]’’ (224).
Through its literal domestication, then, music would produce domes-
ticity; classical music was seen as the soundtrack for the modern marriage
and the modern home. In 1894 a letter in the Mysore Herald declared that
the publication of Karnatic music compositions in European notation,
released a year earlier, ‘‘must be introduced to our girls.’’ The writer re-
marked excitedly that, with such a system of notation, ‘‘songs could be
mastered from mere books’’ (Chinnaswamy Mudaliar 1892, 208–9). The
effect would be a double solution to the ‘‘anti-nautch-girl question’’: it
would spare girls from respectable families from having to go to less re-
spectable types to learn music and would provide ‘‘respectable musicians
who could socially mix and move with us, on social occasions like mar-
riages and similar gatherings where we have been [hearing] the prostitute’s
music’’ (208–9). Meanwhile, music would become the agent, or vehicle,
of the kind of noncorporeal, spiritual love that should exist between a
husband and wife: ‘‘There must be a unity of feeling in all pursuits, . . .
emphatically so in a Hindu home where the tie is not of the senses but of

136 gender and the politics of voice

18 South India Music Emporium advertisement in Madras Music Academy Souvenir (ca.
1935–1940).
a sacred character, emblematical of the eternal wedding of souls together
in harmonious fusion’’ (209). The lofty musical metaphor of harmony was
thus used to characterize marriage. At the same time, music domesticated
the private space of the home. The writer of the letter, as if to make things
clear, concluded that ‘‘music preparation for our future wives is to secure
pleasant households after the day’s weary life-struggle is suspended’’ (209).
Essential to domestication of private space was the domestication of
free time, particularly the time now possessed by the modern wife who
had been freed from menial tasks and chores. A 1930s advertisement for
the South India Music Emporium addressed young husbands as the ones
responsible for buying the trappings of domestic life. Appearing in the
concert program of the Madras Music Academy, the ad counseled male
concertgoers on how to connect the bourgeois public sphere of the clas-
sical concert hall with its domestic equivalent. ‘‘A modern wife has tons
of unemployed leisure and a wise husband must provide hobbies for her
leisure being usefully employed,’’ suggested the ad. ‘‘What better and more
soul-satisfying hobby can there be than violin playing. Give your wife a
violin today and ensure eternal happiness at home.’’ For ‘‘modern wives’’
who found themselves with too much time on their hands, music could
serve as a spiritually uplifting domestic activity that would convert free
time into spiritual capital. If modernity produced an excess of free time,
music could expand as necessary to fill it; if this free time had revealed that
the new Indian woman had a soul, music could satisfy it. Music became
the bonding agent in a new type of ideal marriage; unlike caste or religion,
it had the advantage of being able to appear both voluntary and deeply
traditional, private as well as public. In one biography of M. S. Subbulak-
shmi, a photograph of her and her husband, Sadasivam, places them within
this type of ideal marriage through its caption: ‘‘Icaiyil iainta uaṅkaḷ’’
(Souls joined in music) (Sarathy 1997, 43).
The private space of the home, however, was differentiated from the
private space of the salon. In 1932 a letter to the Madras Mail responded to
the anti-nautch movement, particularly its opposition to public perfor-
mances of nautch, with this bit of reasoning: ‘‘If the dance is to be free
from its less respectable associations, the encouragement of public dis-
play appears to be the best way to do it. Private parties tend to encourage
the notion of lack of respectability. Public functions, on the other hand,
show the dance for what it is’’ (Sruti 1997, 5). The private space of the salon
was considered unseemly and deceptive, while the concert hall allowed
room only for clear observation and not for illusion; any identification

138 gender and the politics of voice

with the performer would not be felt as personal attraction but would be
mediated by the language of art. The concert hall, unlike a private salon,
ensured a respectable distance between the performer and the audience,
not only physically but also psychologically; in public, the performer per-
formed for no one in particular. Such a shift was accomplished not only
by the rise of concert halls but through the music itself. The repertoire
of public concert-hall music featured the devotional compositions of the
‘‘trinity’’ (the composers Thyagaraja, Muthuswamy Dikshitar, and Syama
Sastri), which eschewed all reference to human patrons. This form of spiri-
tual devotion came to be thought of not as an extramusical element, as
was thought of the human love expressed in padams and javalis (musical
compositions for dance conveying the meanings of sensuous love), but as
the essence of music itself.40
How did such notions affect conventions of music performance and
the way music was thought of ? Music’s relationship to notions of public
and private space was determined by the kind of ideological and practi-
cal domestication of music that was taking place. By ‘‘domestication’’ I do
not mean just that music was brought into the home but also that its re-
lationship to notions of public and private changed. The ‘‘respectability’’
that music gained in the 1930s and 1940s was in part due to its peculiar
ability to mediate between public and private. Music brought women into
the public sphere in a particular way, as the voice of tradition. They were
not perceived as innovators, much less as performers; rather, a woman was
to sing as though music were a natural property of her voice. To ease the
problem of women appearing on the concert stage, they often appeared
(and continue to appear) in duos; sisters performing together imparted a
very different aura from a woman performing solo. Many women who
played accompanying instruments like the violin were allowed to accom-
pany only their brothers or other family members onstage, as if their entry
into public could be thus controlled.
Meanwhile, in many upper-caste families, women became musicians
because it was something they could do without formal school education
and because it did not necessarily require them to leave the house; it was a
vocation that could be learned and practiced in the home.41 Indeed, many
of the interviews with female musicians in a volume entitled The Singer and
the Song (Lakshmi 2000) convey a sense of how music has been practiced
as a means of compensation for the frustration of other social, familial,
and professional desires.42 Some of the women interviewed speak of their
devotion to music in particular as a kind of strategy for coping with the

gender and the politics of voice 139

oppression of family obligations. On the other hand, careers as professional
musicians have allowed many women to escape marriage entirely. In a so-
ciety where remaining unmarried is, for a woman, practically unheard of,
there is a disproportionately large number of unmarried female musicians
who are ‘‘devoted to music.’’ If my teacher spoke wistfully about female
musicians who had husbands to help them, other female musicians often
remarked to me that they didn’t understand why my teacher was unhappy
and why she moped around and refused to practice. ‘‘She has no obliga-
tions to a husband,’’ said one. ‘‘She could be devoting herself to her music.’’

. . . And Their Limits

The public face of such ‘‘devotion’’ may be very different from its private
motivations, however. One afternoon, my teacher, whose stories usually
managed to disrupt the platitudes of the Karnatic music world, launched
into a telling of how she became devoted to music. She was about seven-
teen years old, her sister thirteen or so. There were brothers older and
younger. She had been playing the violin for years, sitting in the corner
of the room and listening while her father’s students came to learn. Her
sister had not shown as much interest but was more studious in school, so
their parents decided to send the sister to a boarding school to prepare her
for college. My teacher, who had had only a tenth-standard education, felt
slighted.
She got to thinking: why should only her sister have the opportunity to
become educated, go to college, and eventually work outside the home?
How could they have been born of the same parents and be treated so un-
equally, one kept at home in the bonds of hereditary musicianship, while
the other was sent away to get a college education? She began to doubt
she was her father’s child after all. In anger and protest, she retreated to the
upper verandah of the house, refusing to eat, talk, or play the violin for
two weeks. Her father, who was almost blind, and absorbed in his music
anyway, did not notice until her mother said something. Then he came
up the stairs to where she was sitting. She had expected harsh words, but
instead, as if reading her mind, her father had gently said, ‘‘So, you are
doubting whether you are really my child or not?’’ But rather than simply
laying her doubt to rest, he had said, ‘‘You play your violin. Listen to that
sound. And you will know.’’ From that moment, she said, her resentments
had melted away, and she became aware that musical ties were as strong
as ties of blood.

140 gender and the politics of voice

 In this scenario music again restores domestic tranquility, but in the
interest of a very different set of family values: one based on hereditary
musicianship rather than bourgeois middle-class family arrangements. As
classical music came to be a respectable art for upper-caste women, it took
its part in a larger imagination of ideal womanhood that included ideas
of the artist (as opposed to the hereditary musician) and the natural voice,
on one hand, and companionate marriage and domesticity, on the other.
But for female musicians from non-Brahmin families these ideals seem
to be much more difficult to achieve. These women must negotiate the
contradiction between the modern notion of Karnatic music as a secular
art music, presumably without specified gender roles, and the nonbour-
geois contexts in which they have become musicians.
In 1996 I sought out M. S. Ponnuthai, one of the first women to become
a professional nagaswaram player. Nagaswaram (also called nadaswaram),
a double-reed instrument, is associated almost exclusively with musicians
from the Icai Veḷḷālar caste. Until the 1940s, when Ponnuthai began to play
in public, the instrument was played exclusively by men, as it was believed
to be too strenuous for women. Although much of its musical tradition
is shared with Karnatic music, it remains to this day on the fringes of
what is approved as classical, having only marginal status as a concert-hall
instrument. Its traditional role has been to provide music for auspicious
occasions, such as temple rituals and weddings.43
Ponnuthai, who was in her mid-sixties at the time I met her in Madu-
rai, had had an illustrious career as a nagaswaram player for temple and
political processions, as well as on the concert stage; one of her distinctions
was that she had made concert audiences appreciate the nagaswaram. As
I made repeated visits to her house, however, it became clear that she was
uninterested in talking about her life. My attempts to steer the conver-
sation in that direction on several occasions ended in our watching tele-
vised cricket games. On other occasions, instead of talking, she would play
at great length, taking obvious delight in my inability to keep up with
her virtuosic raga alapana as I struggled to accompany her on the violin.
I thought that these sessions would eventually soften her resistance, but
on my last visit to her house I found that she had invited her grown son
to answer my questions instead. While this was an ironic comment on
the ethnographic project of finding authentic voices, it also revealed much
about the politics of representation regarding the life of a woman who
had led a very public life and then retreated.
From Ponnuthai’s son I learned that her father, a government servant

gender and the politics of voice 141

with progressive ideas, was inspired by the essays of Subramania Bhara-
thiyar. Having decided that his daughter should take up the nagaswaram,
Ponnuthai’s father apparently groomed her for a public life. After her debut
at age thirteen, he resigned from his job in order to escort her to perfor-
mances and also began collecting concert reviews and other press releases
about her. Indeed, among the many clippings her son showed me were re-
views of her performances in Ceylon, Singapore, and elsewhere, which her
father had neatly preserved by mounting on paper. Ponnuthai married a
prominent citizen of Madurai who served as a representative for the Con-
gress Party, and she herself served as the head of the Madurai Icai Veḷḷālar
Sangam, an organization that served the welfare of nagaswaram musicians,
from 1953 to 1963. In 1972, after her husband died, she stopped playing in
temples and gradually retreated from public view. Among the clippings
were several magazine articles written about her in the early 1990s. I asked
her which she thought was the best, and she pointed to the cover story in
Ananta Vikatan, ‘‘Maturai Poutāy: Oru Kaṇṇīr Katai’’ (Madurai Ponnu-
thai: A sad story) (Saupa 1990).
The article begins with an imagined scene of Ponnuthai’s araṅkēam
(debut), panning through the astonished crowd listening on the banks of
the Vaigai River. After describing Ponnuthai’s fame and success in hyper-
bolic terms, it notes her ‘‘sudden’’ disappearance from public life in 1972.
‘‘We wondered what had happened to her, and searched for her. . . . Some
people told us that she had passed away. We were surprised by that, since
just last month the Tamil Nadu government had announced that it was
going to award her the title of ‘Kalaimamani’ [Great Jewel of the Arts]. We
resumed our search with more urgency . . . and finally found her living a
life of misery in a small house in an out-of-the-way part of Madurai.’’ The
narrative then cuts, cinematically, to the scene at the house, where Ponnu-
thai’s ‘‘still majestic’’ look contrasts with the poverty of her circumstances.
She goes next door to fetch the nagaswaram, the only place it can be kept
safe from rats, and returns wearing earrings. ‘‘I had more than a hundred
pounds of jewelry and gold medals,’’ she laments. ‘‘Now it is all lost.’’ Pre-
sumably prompted by the writer, she describes her meteoric rise, against
the odds, and her ability to play for all-night temple functions even while
pregnant. She describes how, after her marriage, her husband ‘‘never inter-
fered’’ with her career, and how their large house, in the center of town,
was always full of distinguished guests. Then, comments the reporter, ‘‘for
a few moments, the great nagaswaram artist was silent. Her eyes welled
up with tears.’’ Ponnuthai goes on to say, ‘‘My husband died in 1972. With

142 gender and the politics of voice

that, my musical life was finished.’’ The reporter feigns an innocent ques-
tion: ‘‘Why? Couldn’t you continue playing?’’ And the article moves to
what is obviously the clincher. ‘‘Nagaswaram,’’ Ponnuthai explains, ‘‘is an
auspicious instrument. After my husband’s death I became an inauspicious
woman [amaṅkala peṇ] and could no longer play in temples. People talked
behind my back. I stopped playing for radio, too. There was no income,
and I was too proud to ask for any help.’’ The reporter comes in for the
fadeout: ‘‘Wiping the dust from the nagaswaram, she raised it to her lips.
What dignity! What majesty! From her unhesitating fingers a shower of
music poured forth’’ (Saupa 1990, 10–14).
What politics of representation are at work here? The article is meant
to evoke sympathy, even outrage that society has allowed so distinguished
a musician to sink into poverty and oblivion.44 It invites the reader—given
this journal’s readership, most likely a housewife reading in her leisure
time—to witness the conflict between modernity, signified by Ponnuthai’s
emergence into the public sphere, and the views of a tradition-bound so-
ciety that still believes that women become inauspicious when they are
widowed. Modernity, here, is articulated in the voice of the narrator,
whose reportage is mixed with exclamations—‘‘What dignity! What maj-
esty!’’—that indicate a subject able to appreciate good music no matter
who or where it is coming from. By melodramatically staging Ponnuthai’s
problem as a secret that must be revealed, the article assumes a ‘‘modern’’
reader innocent of such antiquated conflicts; the unspoken comparison is
to the middle-class, Brahmin music world of Madras in which it is a mod-
ern discourse on family values, not traditional religious values, that both
makes possible and sets limits for women’s professional musicianship.
What is striking, then, is the discontinuity between Ponnuthai’s career
as a professional musician and the careers of women in the middle-class
music world of Madras. In a conversation in 1997 Bhairavi, a professional
musician from an Icai Veḷḷālar family, then in her early thirties, spoke of a
similar discontinuity.45 Although her great-grandparents had been musi-
cians, neither her grandparents nor her parents had continued the tradi-
tion; she herself had made a conscious decision to become a professional
flutist. She had recently been married, but the match had taken a long
time to make, since she was in an anomalous category: not only was she a
professional musician, but she played the flute, an instrument that has not
been taken up by non-Brahmin women since the beginning of the twen-
tieth century.46 It was rare for women of her caste to take up music at all,
much less as a profession. She remarked to me that young musicians from

gender and the politics of voice 143

Brahmin families were increasingly engaging in love marriages but that
this was not possible for her; for someone of her caste, music did not pro-
vide the same kind of avenue toward bourgeois, middle-class sensibility.
For a husband and wife of her caste to have their ‘‘souls joined in music’’
was not an option. At the time of our conversation, Bhairavi’s husband,
who was not a musician, was having trouble finding employment, so Bhai-
ravi was supporting the family by giving music lessons and had applied for
teaching positions in schools. I asked if such an arrangement would be ac-
ceptable. ‘‘That’s how it used to be in our caste anyway,’’ she replied. ‘‘The
ladies were all dancers and musicians and they supported the men.’’ For
her, the irony was that the very transformations in the music world that
had in the name of respectability taken professionalism away from female
musicians of her caste in the early twentieth century now made it possible
for her to earn a living as a professional musician.47

Stage Goddesses and Studio Divas

I have dwelled on the lives and commentary of these two women because
they occupy the fringes of a dominant discourse that links classical music,
ideal womanhood, and domesticity in a particular way. In doing so, they
clarify the contours of this discourse and the kinds of exclusions on which
it is built. For both Ponnuthai and Bhairavi, the dilemma has been how
to create a life for themselves as women musicians who are neither from
the Brahmin, middle-class music world of Madras nor from conventional
hereditary musical families. For both, much of the discourse linking clas-
sical music, ideal womanhood, and the nationalist aspirations of an urban
middle class had been enabling: it had enabled Ponnuthai (through her
father’s desires) to take up the nagaswaram and to travel widely, setting
an example for many younger female musicians aspiring to become pro-
fessional; it had enabled Bhairavi to pursue music as a career, something
that the women in generations preceding hers could not have respectably
done. At the same time, Ponnuthai and Bhairavi were also alienated, as was
my violin teacher, from the subjects of this discourse. Much of the irony
of Bhairavi’s comment comes from the realization that the profound dis-
continuity effected by the emergence of classical music in the twentieth
century made possible a strange kind of continuity.
In the century or so that separated Bhairavi’s life from the lives of her
female ancestors, tremendous changes had taken place in the way music

144 gender and the politics of voice

was conceptualized and practiced—including infrastructure changes like
the building of concert halls and music schools, and technological changes
like the emergence of gramophone recording and radio—which engen-
dered a new kind of discourse about music. Not only did music become
available to ‘‘respectable’’ women as a vocation and sometimes a career but
it became available in particular ways. If my teacher found the transition
from music to the topic of marriage to the essential difference between
women and men ‘‘quite natural,’’ it is because twentieth-century discourse
about music was connected, metaphorically and quite literally, to a dis-
course about family values.48
In the 1930s and 1940s ideals of chaste womanly behavior—not drawing
attention to one’s body or relying on physical charms—became a meta-
phor for a new kind of art that privileged meaning and naturalness over
cleverness and acrobatics. In that sense, classical singing was refigured as a
natural expression of devotion. A woman was expected to sing music as
though it were a natural property of her voice. The natural voice and the
chaste female body were thus linked. By the 1950s, the adjectives natural
and artificial had come to be used to contrast female voices singing clas-
sical music and film songs, respectively. Kalki Krishnamoorthy, the same
reviewer who had raved about M.S.’s voice, wrote disparagingly of the
‘‘insipid’’ and ‘‘artificial’’ sweetness of the renowned film singer Lata Man-
geshkar’s voice. Kalki used the Tamil word vacīkara, meaning attractive
or alluring, with distinct sexual connotations, to describe the film voice,
warning readers not to get infatuated with film music and forget the natu-
ral beauty of classical singing (Kalki 1951).
In this discourse on classical music, performance that drew attention
to the body came to be associated with the artificial; good music was not
something to be performed but rather was simply ‘‘expressed.’’ The ‘‘voice
of the century’’ referred not only to M. S. Subbulakshmi’s sound but to a
particular kind of voice that was imagined to come naturally from within,
unmediated by performance of any kind, a voice embodied in a distinct
way. The notion of ‘‘expression,’’ as Webb Keane has written, is predicated
on a particular linguistic ideology that separates form from content and
in which the voice merely ‘‘refers to’’ other sites of action or ‘‘reflects’’ a
prior, interior self but is not considered to have a role in creating that self
(1997, 684). In tracing the development of the modern notion of self in
European thought Charles Taylor suggests that the notion of expression
is itself a modern idea very much linked to the imagination of the self ’s

gender and the politics of voice 145

interiority; the concept of art, as he notes, relies on both the rejection of
an outwardly oriented mimesis and the embrace of an inwardly oriented
expression (1989).49
Indian nationalist discourse linked the notion of an ‘‘inner voice’’ with
the ‘‘inner sphere’’ of the Indian nation, imagined to be India’s uncolo-
nized interior and often equated metaphorically and literally with the
domestic sphere.50 Indeed, Mahatma Gandhi linked ‘‘good music,’’ ‘‘har-
mony,’’ and the concept of an ‘‘inner voice’’ in a series of nationalist writ-
ings from the mid-1940s. For Gandhi, who was a great admirer of M.S.,
singing classical music was a metaphor for leading a moral life.51 ‘‘As I am
nearing the end of my earthly life,’’ he wrote, ‘‘I can say that producing
good music from a cultivated voice can be achieved by many, but the art
of producing that music from the harmony of a pure life is achieved only
rarely’’ (Gandhi 1958, 132).
The ‘‘voice of the century’’ was as much a product of the technologies
that mediated it as it was a product of individual ability or genius. In a very
real sense, the gramophone and the microphone created the ‘‘perfection’’
and ‘‘nuance’’ of M. S. Subbulakshmi’s voice for the listeners who heard
her through these media. The idea of perfection is only possible when a
piece of music becomes an object to be contemplated over and over again
rather than heard in a single live performance. Nuances become audible
and locatable in a voice only when the voice is amplified—separated, in a
sense—from the body that produces it. In twentieth-century South India,
the female voice, disassociated from its body through these technologies
and through a particular way of performing interiority, came to have a
certain ideological significance.
The purity of this voice was maintained by a careful maintenance of
its boundaries; it was not a ‘‘disembodied’’ voice that could travel freely
but a voice that was embodied in a particular way. By midcentury, Tamil
films began to feature female dancing bodies that were, by the standards
of the time, decidedly immodest; the voices with which those actresses
sang onscreen became another foil for the respectability of classical music
and its prodigies. Although many successful classical musicians, includ-
ing M. S. Subbulakshmi and D. K. Pattammal, gained considerable fame
through their acting and singing for films, the limits of their participa-
tion in the cinema world were carefully defined. M.S. stopped acting in
films after 1945 in order to devote herself to classical performances; appar-
ently, it was not possible for her to do both. D. K. Pattammal, on the other
hand, remained active in the Tamil cinema world as a playback singer on

146 gender and the politics of voice

the carefully observed condition that she sing only patriotic songs, never
love songs, as if she could ensure the authority, and fidelity, of her voice by
making sure its referent was Mother India and not her own or any other
actress’s body.
M.S. and D. K. Pattammal were an inspiration to numerous playback
singers, women who recorded their voices for actresses’ characters. Three
playback singers I spoke with—women in their sixties and seventies in the
early 2000s—all spoke of wanting to be ‘‘just like M.S.’’ when they were
young but being forced by economic necessity to sing for films. Interest-
ingly, while the younger female playback singers I spoke with talked about
changing their voices depending on the character they were portraying,
the older singers insisted that they did not change their voices for different
characters. In the words of one, ‘‘God has given you one voice. If you start
changing it around, it stops being singing and turns into mimicry.’’ 52 Value
and authenticity were thus attached to singing, in contrast to mimicry. In-
deed, maintaining one’s status as a playback ‘‘artiste’’ required an insistence
on having ‘‘one’s own voice,’’ which remained constant. Another playback
singer described how in order to sing playback for films one had to learn to
‘‘give expression just in the voice, not in the face,’’ in order to channel all of
one’s expressive power into the voice, leaving the face and body neutral.53
To demonstrate this, she sang in a range of voices, from little boy to young
woman to old lady, while keeping her face and body utterly immobile.
Female playback singers, much more than actresses, often become ce-
lebrities in their own right; many make frequent stage appearances in
which they sing their hit songs with a backup orchestra for audiences
of dedicated fans. For older female playback singers, their stage personae
often contrast greatly with the lyrics they are singing; in these stage ap-
pearances, it often seems like they make a particular effort not to em-
body their voices, as if in doing so they might maintain greater control
over them. I attended a wedding concert by the renowned playback singer
P. Susheela in 2002, during which she sang a number of romantic duet
songs. Throughout the performance, she stood close to the microphone,
with one hand at her ear and the other carefully keeping the end of her
sari draped over her shoulder in the style of a chaste classical singer. This
purposeful dissociation of body from voice is, I would argue, part of the
politics of voice inaugurated by earlier female classical singers, and further
enabled by the technology of playback singing.
For the ‘‘voice of the century,’’ then, the kinds of bodies with which
it could associate constituted one limit; the ways in which it could be

gender and the politics of voice 147

heard constituted another. Although M.S. and D. K. Pattammal sang in
public, they never spoke in public, an act more conventionally associated
with agency and ‘‘having a voice.’’ Whereas Bangalore Nagaratnammal and
Vai. Mu. Kothainayaki, part of the previous generation of female singers,
spoke in public on issues such as Indian independence and the betterment
of Indian women, Subbulakshmi and Pattammal assiduously avoided pub-
lic speaking.54 Indeed, it seems as though the ‘‘naturalness’’ and ‘‘purity’’ of
their voices could only be guaranteed by maintaining the idea that those
voices expressed an interior self, an innocent self detached from the world
at large, who knew ‘‘nothing but music’’—except, perhaps, a little house-
keeping. In this sense, the very audibility of their singing voices depended
on the silence of their speaking voices in the public sphere.

Gender and the Politics of Voice

Agency, in Western philosophy, is linked to the speaking voice, a link that
is embedded in the classic methodology of ethnography as well (in which
the ethnographer seeks verbal explanations or ‘‘meanings’’ for nonverbal
forms in order to fill them with referential content). The dichotomy often
drawn between ‘‘having a voice’’ and being silent or silenced, however,
leaves us with little way to interpret voices that are highly audible and
public yet not agentive in a classic sense, such as voices that have musical
instead of referential content or voices that circulate through technologies
of sound reproduction.
M.S.’s voice, as revealed by the various discourses about her voice and
persona, is as much a product of a particular historical and social moment
as it is a vehicle of her individual expression. This is not to deny her status
as a creative artist and a powerful persona on stage, but rather to suggest, as
does Mrinhalini Sinha, that ‘‘a focus on the voice or agency of women
themselves does not have to be opposed to an examination of the ideo-
logical structures from which they emerged’’ (1996, 483). In exploring the
creation of a voice in which women could speak as the ‘‘Indian woman’’
in the 1930s, Sinha moves beyond notions of pure feminist consciousness
to show how the creation of a voice is always a strategic maneuver within
certain ideological structures.
The creation of musical voices by women in the 1930s and 1940s was
equally strategic, even though these voices emerged in a part of the public
sphere seemingly far removed from political or even social discourse. In-
deed, it is useful to ask how the speaking voices of women who emerged

148 gender and the politics of voice

in the 1930s as public speakers on nationalist causes and the subject of
women’s rights, on the one hand, and the singing voices of M. S. Subbu-
lakshmi and D. K. Pattammal, on the other, may have worked to shape each
other. The nationalist and poet Sarojini Naidu’s onscreen introduction to
the Hindi version of M.S.’s film Meera, in which she introduces M.S. as
‘‘the nightingale of India,’’ is a literal example of this. The fact that Vai.
Mu. Kothainayaki was partly responsible for bringing D. K. Pattammal
onto the concert stage is another.55 But beyond such literal connections,
the fact that these voices were heard side by side is important; their juxta-
position defined the possibilities for ‘‘respectable’’ women’s participation
in the public sphere, even as they defined the content of that respectability.
Once the voice becomes recognizable as culturally and historically con-
structed, it is possible to ask what new forms of subjectivity are enabled
by changing ideologies of voice. The politics of voice that came to oper-
ate in the 1930s imagined the voice in a particular relation to the body,
as something that transcended the body. This relationship was articulated
by Rukmini Devi, among others, as the relationship between music and
dance, where music was supposed to raise dance above the physical, to
make it more than the sum of its gestures. In the process, music itself had
to be purified; in order to be truly classical, the voice of Karnatic music
had to be reconceived as an ‘‘inner voice.’’
By a synecdochic chain of associations, this newly conceived domain
of interiority became connected with another domain newly conceived
as an essential part of bourgeois modernity: the middle-class home. The
female musical voice and the middle-class home together constituted and
stood for the inner sphere of the nation, a construct central to middle-
class nationalism. The interiorized conception of voice made possible the
subject positions of both the ‘‘classical artist’’ and the ‘‘respectable woman’’;
the natural voice of the artist was—and still is—identified with the chaste
body of the respectable woman. Thus, even now, when it is no longer a
novelty for upper-caste women to perform publicly, singing on the clas-
sical stage involves engaging not just the conventions of musical art but
also the conventions of female respectability.

gender and the politics of voice 149

4 $ Can the Subaltern Sing?
music, language, and the politics of voice

[The Tamil music movement] gives sight to the blind, hearing to the deaf, a voice to the
mute, a wake-up call to those who are sleeping.
—Editorial in The Liberator (1943)

In 1943 T. V. Subba Rao opened the annual Madras Music Conference at the
makeshift quarters of the Madras Music Academy in the neighborhood of
Mylapore. ‘‘What matters in music is not the letter but the tone,’’ he stated.
‘‘We have no politics. Our sphere is only aesthetics wherein we stand for all
that is noblest and best’’ (Proceedings of the 17th Madras Music Confer-
ence, 1944, 1). The guest speaker of the conference, Dr. S. Radhakrishnan,
a philosopher and, later, president of India, went on to speak of music
as a ‘‘great reconciler,’’ not only of India but of the whole world; music
was an instrument of reconciliation, unlike politics and economics. To a
cheering audience, Radhakrishnan continued, ‘‘You know we are fighting
today against the Axis powers. Why? Not because of their art and litera-
ture . . . but because of their politics and economics. . . . But think of
the metaphysics and music of Germany. . . . From a thousand different
orchestras and records, Beethoven and Mozart are pouring out their songs,
pleading for reconciliation.’’ Music, in Radhakrishnan’s view, had a certain
kind of universality that enabled it to reach people regardless of language,
religion, caste, or nationality. But this very universality also rendered it
liable to certain misuses. ‘‘It is essential that so far as music is concerned,
it should know no politics. It stands for certain permanent values of life.
When it is prostituted for ephemeral and irrelevant ends, it is not music.
It is propaganda, narrow, arid and hypocritical’’ (3–4).
Such strong statements were a direct argument against the Tamil music
movement, which had reached a peak in the mid-1940s and whose mes-
sage was that the letter was indeed as important as the tone. A year after
Radhakrishnan’s speech to the music academy, R. K. Shanmugham Chet-
tiar roused the audience to cheers with his speech at the first Tamil Music
Conference, held at Saint Mary’s Hall in the George Town area of Madras.
During a visit to America, he said, he had been explaining the Tamil music
movement to an American friend. The friend was flabbergasted that such
a movement was necessary: wasn’t it a given that people needed to have
music with words in their own mother tongue (tāypaṣai )? Apparently, in
the rest of the world this was not even an issue. But in Tamil Nadu, a
strange aberration had taken hold.

I did research to find out whether there was any other place, other than
this Tamil Nadu, where the music was in a language other than the
mother tongue of the people. There was no other place. . . . As far as I
know, even in other Indian states the mother tongues are not in such
danger. If you want to know what the state of Tamil is, listen: In order
to get a job, you need to study English. If you want to get married, you
have to do it in Sanskrit. If you are going to make a name as a patriot,
you must use Hindi. If you learn music, it is all Telugu. And if you
are just talking ordinarily, you use a strange conglomerate [maṇipiravā-
ḷam]: ‘‘Nēu uṅkaḷai meet paṇṇumpōtu oru matter colluvatarku complete-ā
marantu pōyviṭṭēn sār.’’ This is the place we have reserved for Tamil! (Tami
Icai Makānāṭu Ceai Nikacci Mālar 1944, 40; italics in original)

Emphasizing the naturalness of wanting music with words in Tamil, Shan-

mugham Chettiar proclaimed, ‘‘I don’t care if the tradition of Tamil music
has been around for the last two thousand years or not, whether it is spe-
cial or not. I was born a Tamilian and I need Tamil music’’ (40). He was
describing a society that had in effect lost its senses; for him, as for the
author of the editorial in The Liberator, Tamil music would, in restoring a
person’s voice, quite literally restore all of his senses.
The Tamil music movement (Tami icai iyakkam) brought to the fore
the issue of how the relationship between music and language should be
defined. As the problem of language in music became the subject of de-
bate in the 1930s and 1940s, a series of new questions assumed urgency.
Why, asked members of the Tamil music movement, was Karnatic music
confined to such a small group of people? What was the relationship of
words to music? How was a Tamil to make sense of the fact that although
the majority of Karnatic musicians were Tamil vocalists singing in Tamil
Nadu, the lyrics they sang—and the songs that were considered the heart
of Karnatic classical repertoire—were almost entirely in Telugu?

can the subaltern sing? 151

 The Tamil music movement was more than a demand that the classi-
cal music repertoire include more songs in Tamil; it was part of a new
set of discourses about the singing subject and the relationship between
music and language. I present here not simply a history of the Tamil Music
movement, but an inquiry into the kinds of discourses that made such a
movement possible. I speculate here on the process by which ‘‘music’’ and
‘‘literature’’ began to emerge as two separate, mutually exclusive fields in
late-nineteenth-century South India, preparing the way for a new kind of
relationship between music and language to be imagined: a relationship
of analogy.1 As music became a distinct field separated from literary prac-
tices, it was increasingly imagined as a language. Such a notion marks a
departure from an earlier idea, prominent in Tamil literature, of the musi-
cality of language, where the relationship between music and language is
one of contiguity, cooperation, or commingling, and it is the poetic and
sonic aspects of language that are emphasized. The twentieth-century idea
that music ‘‘is a language’’ or is ‘‘like language’’ entails a very different set
of premises, applying to music the concepts of meaning, intentionality,
and understanding that are commonly applied to language in modernist
thought.
By the 1930s, the analogy of music to language had become an espe-
cially useful way for both Indian and Tamil nationalists to imagine the
place of music in a new nation. Once this analogy had taken hold, colonial
classifications like ‘‘classical language’’ and ‘‘mother tongue’’ increasingly
came into play in competing ideas about how music should be defined
and experienced, as did ideas about the ‘‘meaning’’ of music and where it
was to be found. Those who identified with the Tamil music movement
assumed that definable performing and listening subjects could be located
through the use of the mother tongue, the language with which people
identified and that thus identified them as authentic subjects. But then,
the Brahmin music establishment asked, would such music still live up to
the standards of a classical art? Was music akin to a mother tongue or a
universal, aesthetically motivated language? 2
The debates surrounding the issue of language in music were essentially
debates about what a singer should feel when singing and what a listener
should feel when listening. The issue of precisely what kind of language
music was to be compared to—a universal language or a mother tongue—
was in a sense also a question about the singing subject’s relation to his
or her voice. Was the voice best conceived of as an aestheticized instru-
ment, as the Brahmin musical establishment thought, or as a transparent

152 can the subaltern sing?

representation of one’s self, as the Tamil music movement suggested? The
fact that these two choices were seen not only as logical or natural but
also as exhaustive of the possibilities for a subject’s relation to his or her
voice suggests that a particular politics of voice was emerging at the time.
Within this politics of voice, music could be an expressive language or an
aesthetic language, but in either case language was the central metaphor
for articulating the relationship between voice and singing or listening
subject.

Sacrificing for Mother Tamil

Such ideas are not natural or inevitable; rather, they are the results of shifts
in musicolinguistic practices that occurred during the nineteenth century
and early twentieth. One may be used to thinking of music and language
as two separate entities or systems, whose main possibility for connection
is through a one-way analogy: music, one may say, is ‘‘a language’’ or is
‘‘like language.’’ In nineteenth-century South India, however, music and
language did not necessarily exist as mutually discrete categories available
for this particular metaphoric relationship. Instead, what did exist were a
multitude of genres—poems, plays, epics—and practices for performing
them musically or with musical accompaniment. The ideal inseparability
of music and language is suggested by the term muttami (literally, ‘‘triple
Tamil’’), used in Tamil literature from the Sangam era (second century BC)
to the present to refer to the interlinked arts of iyal (word), icai (sound or
musical rendering), and nāṭakam (mimetic rendering, dance, or drama).3
U. V. Swaminathayyar’s Caṅkīta Mummaṇikaḷ (Three gems of music)
offers some insight into a series of shifts that were occurring in the late
nineteenth century. His reason for publishing the book, Swaminathay-
yar wrote in his foreword, was that ‘‘friends tell me that in Tamil Nadu,
no one sings in Tamil anymore.’’ ‘‘At that time,’’ presumably in the 1870s
and 1880s, there was great matippu (respect, value) for the compositions of
Ganam Krishnayyar and Gopalakrishna Bharatiyar, which were in Tamil;
after that, it became popular—a matter of perumai (pride)—even for Tamil-
speaking musicians to sing in Telugu (1936, 2–3). What had happened to
‘‘music’’ and ‘‘Tamil’’ between the 1870s and the 1920s, when Swaminathay-
yar wrote his biographical sketches?
Swaminathayyar’s biographical sketches include numerous descriptions
of ‘‘vidwans adept in both music and Tamil’’ (1936, 2). For the nineteenth-
century musicians he described, in keeping with the ideal of muttami,

can the subaltern sing? 153

it was not only possible but necessary to combine virtuosity in music and
language. The title ‘‘vidwan’’ (literally, ‘‘the one who knows’’) was used for
poets, musicians, and scholars alike; what united them was their prowess
in verbal-musical performance and their ability to provide exegeses of the
texts they sang or recited.4 But the fact that Swaminathayyar consistently
referred to music and Tamil points to a shift that had occurred by the time
he was writing his account; music and Tamil (specifically, Tamil literature)
were coming to be recognized, and canonized, as two distinct fields with
their own experts, a process that had been going on since the late nine-
teenth century. Swaminathayyar, best known as the editor and publisher of
many of the Tamil literary works now considered classics, was a key figure
in this canonization process, responsible for rediscovering and publishing
numerous palm leaf manuscripts.5
Along with the institution of Tamil literature came an increased spe-
cialization of roles. Whereas poetic and musical composition were often
combined in the creation of dramatic and poetic works, the dual role of
kirtana composer and erudite poet was becoming increasingly unusual in
the nineteenth century (Peterson 2004, 33). In his autobiography Swami-
nathayyar reports the contempt of his teacher, the noted Tamil scholar
Meenakshisundaram Pillai, for poets who composed musical works; music
was thought to be a distraction from the more important aspects of gram-
mar, poetics, and mastery of traditional commentaries (Zvelebil 1992, 132–
33). Thus, although becoming a Tamil vidwan involved being able to recite
and perform Tamil texts musically, the study of music per se was considered
a separate pursuit.
In Caṅkīta Mummaṇikaḷ Swaminathayyar himself recalled having to
make a choice between music and Tamil in his youth. He described meet-
ing as a young boy with the composer Gopalakrishna Bharatiyar, noted for
his musical compositions in Tamil, in 1871, when his father had brought
him to Mayuram to study Tamil with Meenakshisundaram Pillai. Bharati-
yar asked Swaminathayyar’s father, who had himself trained with the Tamil
composer Ganam Krishnayyar, why the boy was not studying music and
agreed to teach him as long as he was in Mayuram. Unbeknown to his
Tamil master, Swaminathayyar happily took up music lessons, seeing Bha-
ratiyar in the morning and Meenakshisundaram Pillai in the evening. All
went well until one day the two teachers happened to meet, and Meenak-
shisundaram Pillai boasted about his pupil to Bharatiyar: ‘‘A very talented
young boy is learning with me; he sings poems with music—it is sweet to

154 can the subaltern sing?

the ears. If you heard it you would have satisfaction.’’ Bharatiyar replied,
‘‘I know that boy and his father. He has been coming to me every day for
music training. He said he had come here to study with you.’’ Meenak-
shisundaram Pillai immediately left and went to Swaminathayyar’s house,
where he confronted the boy: ‘‘If you spend all this training and effort
on music, you will not gain wisdom from the Tamil literature and gram-
matical treatises [ilakkiya ilakkaa nūḷkaḷ ]. That music will be a barrier to
the careful study and reading of the texts.’’ The next day, Swaminathayyar
wrote, he stopped attending music lessons. ‘‘It was true that, as Bharatiyar
had said, Meenakshisundaram Pillai was an enemy of music [caṅkīta virōti].
But I didn’t resent him at all for that. Even though I had developed an
affection for music naturally at a young age, my most important subject
was Tamil, so naturally I made the sacrifice. If I hadn’t, the opportunity
to . . . do sacred service to Tamittāy [Mother Tamil] would have been lost’’
(136–41).
What is striking here is the ‘‘naturalness’’ of this sacrifice for Swami-
nathayyar; it indicates the degree to which Karnatic music and Tamil lit-
erature had become separate and mutually exclusive fields. The proof of
Tamil’s status as a classical language in the twentieth century depended on
having a body of written, published texts that could exist apart from indi-
vidual recitations or performances. Meanwhile, the classicization of Kar-
natic music depended on considering the lyrics less important than ‘‘the
music itself.’’ Classical music was redefined as that in which language was
secondary to music, indeed, as that which was distanced from the mother
tongue and required special knowledge.
This redefinition was part and parcel of the way caste communities
came to be differentiated through their relation to the Tamil language in
the twentieth century. As musical standard-setting was increasingly domi-
nated by Brahmins, Telugu and Sanskrit repertoires were privileged, and
Karnatic music was disconnected from Tamil both as a literary language
and as a mother tongue. This is because in the twentieth century Tamil
Brahmins were progressively distanced—and distanced themselves—from
the Tamil language in Dravidianist discourse; those Brahmins who did
continue to profess their love for and devotion to Tamil in the 1930s and
1940s were accused of secretly wanting to Sanskritize it.6 In the wake of
the Dravidian renaissance and the non-Brahmin movement, many Tamil
Brahmins came to identify themselves as belonging to a separate ‘‘race’’
from other Tamils, studying Sanskrit or English but rarely learning to read

can the subaltern sing? 155

or write Tamil.7 For the Brahmin community, which saw itself as the pri-
mary guardian of South India’s classical music, the mother tongue was
devalued instead of glorified.
Thus, both the canonization of Tamil literature and the canonization
of Karnatic music were deeply intertwined with the way the category of
language was becoming differentiated and given new meaning in early-
twentieth-century India. Ideas of music as a means of communication and
the concern that emerged in the 1940s with the meaning of music and how
it was to be enjoyed and understood operated by treating music analo-
gously to language. Yet the analogy did not stop—or could not stop—at
the undifferentiated category of ‘‘language.’’ By the 1930s a complicated
hierarchy of languages had developed in India, which rendered any uni-
tary notion of language too simple. Bernard Cohn has argued that the
British production of translations, grammars, dictionaries, and treatises
concerning Indian languages (as well as the publishing of palm-leaf manu-
scripts as literary texts) was part of a project of ‘‘converting Indian forms
of knowledge into European objects’’; it produced a ‘‘discourse of differ-
entiations which came to mark the social and political map of nineteenth-
century India’’ (1987, 283–84). These differentiations were the basis of a
hierarchy that included so-called classical languages as well as ‘‘vulgar’’ or
‘‘vernacular’’ languages or mother tongues. While Sanskrit was considered
a classical language, Tamil was, like other spoken languages, considered a
vernacular. Tamil nationalists protested the inferior status given to Tamil,
arguing that the unbroken continuity of Tamil from ancient times to the
present made it a classical language (Arooran 1980, 109).8 The figure of
Tamittāy, or Mother Tamil, found in earlier Tamil literature, was elabo-
rated as a central trope encompassing pride in Tamil as both classical lan-
guage and mother tongue.9
The ‘‘naturalness’’ of Swaminathayyar’s sacrifice was thus made possible
not only by the construction of Karnatic classical music and Tamil litera-
ture as separate and mutually exclusive fields but also by the emergence
and naturalization of the category ‘‘mother tongue’’ beginning in the last
decades of the nineteenth century. The concept of the mother tongue
(Tamil, tāymoi or tāypaṣai ), far from reflecting a natural or necessary rela-
tion between a subject or community and ‘‘their’’ language, has a particu-
lar history. Indeed, it was partially colonial discourse about Indian lan-
guages that made the concept of the mother tongue available as a marker
of identity in twentieth-century India.10 In tracing the emergence of this

156 can the subaltern sing?

category in Telugu-speaking South India, Lisa Mitchell has argued that its
naturalization required a new understanding of both language itself and
a subject’s relation to language: in effect, a ‘‘different sense of selfhood’’
(2004, 39). Before the nineteenth century, as Mitchell and others suggest,
discourses on and practices of language depended on a multilingual sensi-
bility in which the ability to move fluidly between languages and registers
was prized; court poets would not have ‘‘devoted’’ themselves to a single
language.11 Swaminathayyar, by contrast, was by his own admittance pro-
foundly uninterested in other languages, especially Telugu and Sanskrit.
Consider this passage from his autobiography, written in 1940–42:

Father worried himself constantly about my education. He had set his

heart on making me too a musician. All musicians of the day were well
acquainted with Telugu. They used to sing compositions in all three
languages—Tamil, Telugu, and Sanskrit. Father felt that a knowledge
of Telugu would be helpful to a student of music. . . . I had, I should
admit, a bias towards music and Tamil—none at all towards Telugu. . . .
Music and Tamil gave me joy. I didn’t find such joy in Telugu. Even at
the outset, I realized that Telugu and I were poles apart. The dislike was
natural, not deliberate. (Swaminathayyar 1940–1942, 27–28)

Swaminathayyar is at pains to say that his dislike of Telugu is ‘‘natural,’’

the implication being that his love of Tamil and his willingness to cease
his study of music and other languages to concentrate on Tamil are just as
natural. While such single-minded devotion to a language, rather than to
language in a more general, less differentiated sense, would hardly have
allowed Swaminathayyar to become a scholar (much less a vidwan) at the
beginning of the nineteenth century, it made him an ideal scholar at the
end of the nineteenth century.
The idea of having a single language with which one identifies depends
on the notion of language as an object, as ‘‘something to be loved, admired,
protected, and patronized, and something which one can have affection
for and pride in’’ (Lisa Mitchell 2004, 37).12 Just as it motivated Swami-
nathayyar’s sacrifice, the threatened figure of Tamittāy was used by poli-
ticians and university professors to gain popular support for anti-Hindi
protests in the 1930s and for causes such as the Tamil music movement
(Ramaswamy 1993, 704). On the role of the metaphor ‘‘mother tongue’’ in
shaping Tamil identity and rallying people around the cause of ‘‘saving’’
the Tamil language, Sumathi Ramaswamy writes,

can the subaltern sing? 157

 It familiarized and familialized the relationship between the Tamilians
and their community by couching it in the comfortable everyday terms
of the home and family. The metaphor also naturalized this relation-
ship by constituting a sense of primordial and selfless devotion that
the Tamilians as children naturally owed to their language as mother.
It dehistoricized the bonds between the language and the people by
presenting them as timeless, essential, and beyond the vagaries of his-
tory. Above all, it depoliticized the relationship by enabling the ab-
straction of the community from politics and by resignifying that com-
munity as a family whose members were united as harmonious siblings
(Ibid., 719)13

The representation of the Tamil language as Mother Tamil, Ramaswamy

states, came about ‘‘in a late colonial situation in which motherhood came
to be privileged, not only as the sine qua non of women’s identity, but also
as the foundational site on which pure and true subjectivities and commu-
nities could be imagined and reproduced’’ (1997, 125). The mother tongue
created bonds between subjects and between a subject and his or her lan-
guage which were seen to be pure and true because they were cast as being
as natural as a child’s bond to its mother. As far as those in the Tamil music
movement were concerned, if music could operate on analogy with and in
the mother tongue, it could create a pure and true community of singers
and listeners.

‘‘The Power of Music’’

Music, however, was threatening because of its potential to move in ways
that language did not, to create bonds or associations that were not as pure
or as true as Tamil or other nationalists desired. In this respect, Swami-
nathayyar’s Caṅkīta Mummaṇikaḷ provides another point of contrast. When
Swaminathayyar, looking back to the nineteenth century, wrote his biog-
raphies of famous musicians, he was perhaps inspired by ideas about music
in the abstract. Yet his writings do not concern the power of music in any
general sense; they instead attest to the power of individual musicians in
particular contexts. For example, Maha Vaidyanathayyar’s power as a mu-
sician lay in his ability to rise to a particular situation, to sing something
appropriate to the context, something that bore the mark of his own cre-
ation. In the nineteenth-century world of royal patronage that Swami-
nathayyar described, the ‘‘power of music,’’ as an abstract concept, separable

158 can the subaltern sing?

from the musician and the context of performance, did not exist. By the
1930s and 1940s, however, it was precisely the separation of music from its
performers and contexts of performance that gave rise to a new discourse
about the power of music, a power that lay in its capacity as a medium for
communicating with or reaching large numbers of people.
In South India in the first few decades of the twentieth century, what
made this possible was sound reproduction: the technologies of recording,
film, radio, and loudspeakers, which brought music into homes and vil-
lages. The possibility of hearing music that one was not producing oneself,
that was coming from some other undefined place, and that was being sung
in some other language was radically new. The artist and writer Manohar
Devadoss describes in Green Well Years, a memoir of his boyhood in Madu-
rai in the 1940s, the kind of power music attained through these technolo-
gies.14

Radios were rare then but the citizens of Madurai had ample opportu-
nity to listen to cinema music. Indeed, they could hardly have escaped
listening to it. Let a girl come of age, let there be a wedding, a preg-
nancy, or a birth, let there be an ear-piercing ceremony, and let any little
wayside temple celebrate an obscure festival—people at once thought
of gramophone records of film music. Low-quality, high decibel, metal
loudspeaker horns sprouted here, there, everywhere, spewing out 78
rpm music in harsh metallic tones, along with a continuous loud drone
caused by the scratches on the records, and the music reached every
nook and corner of the city. (1997, 156)

Music literally became part of the atmosphere, its new power apparent in
the seemingly spontaneous ‘‘sprouting’’ of metal loudspeaker horns. Di-
vorced from its original context and separated from the musicians who
created it, cinema music plunged itself into new contexts, attaching itself
to the lives of those who overheard it. Overhearing became a new type of
listening, thereby enabling music to work by a kind of metonymic, acci-
dental power of association.15 As described in Green Well Years, Sundar and
his friends discovered the power of music by overhearing. Music’s power
to evoke memories, the boys discovered, did not depend on the perti-
nence of the lyrics. Rather, it derived its power from being inescapable,
from attaching itself accidentally, acting not as a language but as a kind
of mnemonic device. ‘‘When he heard Wagner’s ‘fest march’ from ‘Tann-
häuser,’ he was transported to a winter in the late sixties in South India
and he saw green paddy fields, coconut and betel nut groves, thatched huts

can the subaltern sing? 159

and bullock carts. But if he listened to ‘Meera Bhajans’ by M.S. Subbulak-
shmi, he was flooded with memories of a summer in the early seventies
in the USA and images of the Grand Canyon, Mesa Verde, and the Grand
Tetons and Yellowstone filled his mind’’ (159). The new power of music
was exhilarating and disorienting at the same time. One could find one-
self humming a song and suddenly be transported back to another time
and place. ‘‘This new awareness of the ‘power of music’ excited the boys
greatly. They kept talking about ‘remembrance of songs’ again and again.
They realised that if a particular song was heard frequently over a certain
period, then memories of that period would come rushing to the mind
when the song was sung or heard after a reasonably long time. The break
was essential’’ (157). The radically new power that music had gained de-
rived from its ability to be free from its original context, enabling it to
dis-orient and re-orient at the same time. The ‘‘power of music’’ in an ab-
stract sense became thinkable because the technologies of recording and
amplification made music incorporeal, invisible yet ubiquitous, part of the
air itself. In the world Devadoss describes, one became connected to par-
ticular songs not by understanding them or grasping their meaning but
simply by virtue of hearing them again and again. One’s connection to the
songs was not through the metaphor of language but through a more met-
onymical operation: one found oneself humming almost involuntarily,
like a sympathetic string vibrating. Music, transmitted through the ear and
the voice, operated unpredictably and irrationally, with a power that was,
literally, unspeakable.16

Making Sense of Music

Such power was at odds with nationalist thought on the place of music in
India, according to which music should be able to give voice to people,
but not the kind of irrational, unpredictable voice described above. The
idea of music as a language turned out to be a particularly useful way for
those with nationalist ideas to figure the place of music in newly inde-
pendent India. A potentially bewildering and recalcitrant domain could,
by analogy to language, be used to classify and communicate with the
people; music could be turned into something as rational as language. At
the same time, music was imagined as a universal, truly national language,
able to overcome barriers that would have interfered with any merely ver-
bal language.
The problem of how to make sense of music was a pervasive theme

160 can the subaltern sing?

of the All-India Music Conferences convened between 1916 and 1926 by
the Marathi musicologist V. N. Bhatkhande, who is now recognized as
a major figure in the twentieth-century redefinition of North Indian
music.17 These conferences, modeled on the Indian National Congress,
were meant to bring together musicians, musicologists, scientists, con-
noisseurs, and critics from all over India to determine the standards and
boundaries of a national music for India (Bakhle 2005, 180–181). Bhat-
khande himself believed passionately in two projects that he hoped would
culminate in the establishment of a national academy of music: the ratio-
nalizing and standardizing of musical training, and the construction of a
‘‘connected’’ history of music that would encompass both the North and
South Indian systems (ibid., 104–6). In 1916, at the first All-India Music
Conference in Baroda, Bhatkhande expressed his excitement at the thrill-
ing potential of music as a medium to unify the people of India (ibid., 107).
But his vision was utterly different from the power of music described by
Devadoss. Whereas for Devadoss music worked by a kind of accidental as-
sociation, evoking memories without regard for national boundaries, for
Bhatkhande the desired end was a state of musical unification that could
only come about after diligent training. Music instruction should become
‘‘common and universal,’’ he stated at the first All-India Music Confer-
ence. ‘‘And if it please Providence to so dispense that there is a fusion
between the North and the South, then there will be a National music
for the whole country and the last of our ambitions will be reached, for
then the great Nation will sing one song’’ (Bhatkhande 1919, 42–43). For
Bhatkhande the mysteries of how music worked needed to be separated
from musicians—whom he saw as ignorant, illiterate, and secretive—and
studied rationally and scientifically by musicologists. The juxtaposition
of the nation’s crying need for a standardized and accessible music with
the purported narrow-mindedness, ignorance, and secrecy of musicians
themselves became a dominant theme at the All-India Music Conferences
(Bakhle 2005, 195).
Among the invited lecturers at the first All-India Music Conference
was H. P. Krishna Rao, subregistrar of Mysore and headmaster of a music
school there, who had come to present his research into what he called
the ‘‘psychology of music.’’ Krishna Rao stated at the outset of his lec-
tures that ‘‘music is but the language of the emotions’’ (1917, 1). The lec-
tures, later published as a book, began with a chapter titled ‘‘Definition and
Nature of Music,’’ which considered the problem of language. Language
could be divided into three classes: ‘‘sound-language,’’ ‘‘sign-language,’’

can the subaltern sing? 161

 19 ‘‘Musical angles.’’
In H. P. Krishna Rao,
The Psychology of Music
(1916).

and ‘‘word-language’’ (1).18 Music, as sound-language, was privileged over

word-language and sign-language because it was ‘‘natural’’: ‘‘A word has
only a conventional meaning, while a sound has a natural meaning’’ (2).
Music, he implied, had greater potential to express and control the emo-
tions of human beings because its sounds were motivated by nature, not
by mere convention: ‘‘The sound language originates from instinct, and
develops into music as cultivated by different nations’’ (2). Krishna Rao
thus articulated a theory of the relation between music and the subject.
According to his theory, music had to be meaningful; it had to express
one’s own unique emotions but at the same time be understandable to
other Indians. ‘‘Meaning,’’ in this case, meant that music was recognizably
anchored to the nation. Music, as both a national art and the privileged
language of the emotions, could be used to define the subject’s relation to
the nation.
But how, precisely, did this relation work? Krishna Rao imagined the
production of emotion by music through an analogy with the technologi-
cal amplification of sound: music acted as a stimulus to produce emotion,
‘‘a form of energy . . . like heat or electricity,’’ which would be carried and
‘‘amplified’’ by the ‘‘conductor,’’ the human subject (1917, 13). Music was
powerful in this way only when the ‘‘nerve-substance’’ of the subject was a
good conductor: that is, when the subject was ‘‘musical’’ (ibid.). A sure way
to determine the musicality of any person, Krishna Rao suggested, was to
determine their place on the diagram of musicality that he had developed.

162 can the subaltern sing?

The diagram showed an O-origin and an Y-axis and X-axis representing
will power and nerve power, respectively. The line representing the pro-
gression of emotion could fall anywhere between the two axes. If it fell in
the quadrant closer to the nerve-power axis, the subject was musical, that
is, able to be moved by music. The closer one’s ‘‘musical angle’’ moved to
the will-power axis, the less musical one became. When the diagram was
applied to the subjects of a nation, the musical angle had direct ramifica-
tions for society. For instance, ‘‘an unmusical person near A is, as a member
of society, a mechanical being. . . . He does everything as a duty, . . . re-
served and strategic’’ (ibid.). On the other hand, ‘‘A musical person with a
sound general education is an ornament to society. Stationed on the line
OZ, he can oscillate like a pendulum to OX or OY, according to circum-
stances. He it is that knows the right and the wrong. . . . The motive for
doing good proceeds in him instinctively, while that of the unmusical per-
son is based on reason. The musical person is a friend, . . . a dear partner, a
loving parent and a loyal subject’’ (14). For the good of society and the na-
tion, then, music education should be made compulsory. In fact, the unity
of the nation worked on the analogy of the resonating body: ‘‘The state is
a body politic. The governing power represents the brain, and the subjects
so many nerve centres. The sovereign of the State can exercise a power-
ful influence on them. . . . They resonate to the vibrations transmitted to
them from the brain’’ (17).
If it was possible to produce a diagram to measure musicality, it was
also possible, according to Krishna Rao, to show exactly what kinds of
emotions music was capable of producing. He claimed that each note of
the scale or raga produced a different kind of mood when heard against
the background of the note sa: ‘‘Sa and pa are tranquil notes; ri 1 and da 1
indicate disturbance; ri 2 and da 2 are perceptions; ga 1 and ni 1 indicate
disagreeableness; ga 2 and ni 2 indicate enquiry; ma 1 denotes optimism
or egoism; ma 2 denotes degradation’’ (1917, 23).19 Bearing this general set
of rules in mind, Krishna Rao transcribed a section of alapana in the raga
sankarabharanam and ‘‘translated’’ it into ‘‘word-language.’’

Saaaaaaaaaaa
I am Menaka.
GRSaaaaaaaaaa
I am your Menaka.
MGRSaaaaaaaaaa
I am your beloved Menaka.

can the subaltern sing? 163

 MGMPDNSaaaaaaaa
Look at me; surely I am your love.
SNDPMGRSaaaaaaaa
I, your love, have come to you again.
Riiiiiiii
This is your baby, is it not?
PMGMGRiiiiiii
Do you see how beautiful it is?
GMDPMGRiiiiiii
How nice are its limbs?
Gaaaaaaaaaaa
Imagine how matters stand now.
GMPDPMGaaaaaaaaaa
Is it proper that you should thus neglect us? (35)20

If this musical language was taught, Krishna Rao maintained, people

would learn how to express their emotions and to hear the emotions of
others; this was tantamount to being a decent citizen. But only Indian
music, because of its melodic nature, was suited to expressing these ‘‘inner
emotions.’’ Meanwhile, the harmony of the West was suited only to ex-
pressing ‘‘external emotions’’ and ‘‘effects of grandeur’’ (41–42). While
Western harmony was basically an artificial innovation, melody was an
uncorrupted, natural language.
Central to Krishna Rao’s theories was the idea of the universal appli-
cability of music, expressed in his notion that music provided more reli-
able and direct access to the mind of the common man than language. His
ideas about the relation between music, the nation, and its subjects were
taken up in the late 1940s in ‘‘Music and Healing,’’ a lecture delivered at
the Madras Music Academy by Rao Bahadur N. M. Adyanthaya, former
director of Industries and Commerce, Madras. Noting the desperate situa-
tion of India’s people, Adyanthaya said, ‘‘It is being increasingly recognised
that the psychology of the average man in the street is fast deteriorating
and that of the masses is dangerously on the verge of collapse. There is
therefore need for an effective antidote for the ever increasing tendency
for revolt’’ (1949, i). Since the problem was at heart a psychological one,
Adyanthaya recommended a medium he thought would provide access to
the masses: music. ‘‘Perhaps a judicious treatment with mass music of the
right kind may offer a remedy as music has been proved to soothe strained
nerves’’ (i). Adyanthaya suggested that using music in this scientific way

164 can the subaltern sing?

would not kill the art but rejuvenate it, make it even more powerful. If
the already immense scholarship on classifying ragas could be put to use
for a social cause, he suggested, more people would be able to enjoy Kar-
natic music. Not only that, but patient reactions to different types of music
might provide a scientific basis for distinguishing classical and nonclassi-
cal music (6). The answers to both musical questions and social problems
lay in the auditory capacities of the nation’s subjects. ‘‘Important musical
associations may take up the matter and with the cooperation of the Uni-
versities & Government may institute scholarships for the study of [these]
problems,’’ he recommended. ‘‘These scholars may be attached to hospitals
equipped with all the requisite instruments for measurements of the re-
actions of patients to music of different kinds. There should be a separate
ward and an auditorium for the treatment of patients.’’ Such a scientific
and psychological approach could also address the ‘‘much discussed subject
of language in music’’ (10).
For Krishna Rao and Adyanthaya after him, the meaning of music was
a function of human psychology, a matter of stimuli and response. For
others on the quest for meaning in Karnatic music, it seemed more logi-
cal to look for meaning in a more conventional linguistic sense. In 1929
the new magazine Triveni, self-characterized as ‘‘A Journal of the Indian
Renaissance,’’ published a series of articles on Karnatic music written by
the well-known Telugu violinist Harinagabhushanam. Music, he stated,
was of two kinds: conventional music, which dealt with the things of
this world, and transcendental music, which was a tool for attaining eter-
nal bliss (1929, 50). Appealing to an imaginary, ideal nation of Aryans,
he maintained that transcendental music was the unique heritage of the
Aryan system of music and that this heritage was best preserved in the
Karnatic music of South India (50). The transcendental aspect of the music
took place in the wordless parts of the music. But present-day musicians
partook of it unknowingly, for they did not know that even those parts
had a meaning, or referent. ‘‘Musicians say only swara names: Ri Ri Ri Ni
Ni Ni AAA, etcetera, which ejaculations have no sense whatsoever,’’ he
complained. ‘‘I used to hear great musicians of old sing ragam saying ‘Tha
Da Ri Na Tom’ and I was all the time thinking that it must be some expres-
sion with a great meaning. . . . Sometime back the real expression dawned
on me with its import, to my great relief. [It was] a mutilated form of
‘Thath Are Anamtham Aum,’ [which] means ‘Brahman is Infinite and he
is Aumkara.’ Alas, our present day musicians have incurred the bane of
meaninglessness on our sacred music’’ (54). To Harinagabhushanam’s great

can the subaltern sing? 165

relief, the apparently meaningless syllables musicians were singing could
actually be traced back to the sounds of language, thus revealing the mo-
tivation for using those sounds and at the same time restoring the true
purpose of music. Against the threat of senseless sounds stood the assur-
ance of an ancient language—Sanskrit—and the assurance that the music
‘‘meant’’ something appropriately elevated.
The quest for meaning in music that motivates the writing of Krishna
Rao, Adyanthaya, and Harinagabhushanam reveals an underlying dis-
course about music which emerged only in the first decades of the twen-
tieth century. What these thinkers proposed was, in effect, a certain re-
lationship between music and subjectivity that was based on language.
For Krishna Rao and Adyanthaya, music communicated with listeners on
the model of language. For Harinagabhushanam, to sing was to already
be speaking, if only unconsciously, language itself; the perception of a
language standing behind music assuaged the fear that music might be
‘‘senseless.’’ Through this metaphor of language, the ideal of finding mean-
ing in and understanding the meaning of music became central. Through
the metaphor of language, Karnatic music could be related to specifi-
cally Indian subjects. Krishna Rao and Adyanthaya made sense of music
by positing that music was a language in itself that could be used by suf-
ficiently musical—and therefore sufficiently Indian—subjects. Harinaga-
bhushanam made sense of music by discovering that a language, Sanskrit,
was submerged within its sounds, making it quite literally speak for Aryan
India. It was precisely because it became impossible, by the 1940s, to speak
of music without speaking of language that the issue of language in music
assumed such importance.

The Tamil Music Movement

The Tamil music movement took place in the wake of the anti-Hindi
agitations and the formation of the Self-Respect Party, and amid debates
about where the boundaries between the linguistic provinces Andhra Pra-
desh and Tamil Nadu should be drawn.21 The Self-Respect (cuya mariyātai)
movement was begun in 1927 to protest the oppressions of the caste system
and the dominance of Brahmins in fields such as law, journalism, medi-
cine, government, and music. The movement’s main leader, E. V. Rama-
sami, had broken off from the Congress Party in 1927 and joined the Jus-
tice Party, which ran on a platform that advocated rationalism, as opposed
to religion, and mobilized followers on the basis of a Dravidian identity

166 can the subaltern sing?

that was specifically Tamil. The notion of a Dravidian linguistic and racial
group distinct from Aryan languages and races was based on the philo-
logical project of Robert Caldwell and formed the basis of the twentieth–
century Dravidian renaissance, an efflorescence of scholarship on the lit-
erature, history, and culture of the ancient Tamils.22
E. V. Ramasami made a point of recognizing the contribution of non-
Brahmin musicians to Karnatic music, to compensate for their being edged
out of the music scene in Madras. In 1937 the Congress Party called for
Hindi to become the national language of India and suggested compulsory
Hindi education, sparking violent protests in Tamil-speaking areas, where
such a proposal exemplified Aryan domination of the Dravidians. Whereas
the Tamil language was considered, by Dravidian nationalists, to be an
original language unsullied by Aryan influences, Telugu, with consider-
ably more borrowings from Sanskrit, was considered to be a corrupted
‘‘daughter’’ of Tamil, whose loyalties were ambiguous.
Talk of Tamil songs as part of the Karnatic music tradition was spo-
radic until the late 1930s (Arooran 1980, 254). In 1929 the philanthropist
Raja Sir Annamalai Chettiar founded the Raja Annamalai Music College
at Chidambaram, which became affiliated with Annamalai University in
1932. But although it was later to become the bastion of the Tamil music
movement, in the early 1930s its purpose was simply to develop the musical
talent of South India (ibid., 255). It was not until the 1940s that the demand
for Tamil songs began to be justified by the claim that it was necessary to
hear songs in one’s mother tongue. At the convocation of Annamalai Uni-
versity in 1940, Raja Sir Annamalai, as he was known, announced a fund
of ten thousand rupees, to be awarded in prizes to those who composed
new songs with Tamil lyrics. The first conference for the development of
Tamil Icai was held at that university in August 1941. About a month later
another conference was held under the auspices of E. V. Ramasami in Tiru-
chirappalli; roughly two weeks after that, a group of musicians in Madras
came forward to sign a statement avowing their support of the movement
(Ramanathan Chettiar 1993, 4–6).
In 1941 Raja Sir Annamalai made another donation, this time of fif-
teen thousand rupees, to be given as prizes for those who sang and com-
posed Tamil songs. Many professional musicians, such as Papanasam Sivan,
Dhandapani Desikar, and Mariappa Swamy, participated in these competi-
tions. The winning songs were collected and published, with notation,
by the university, for dissemination throughout Tamil Nadu. In order to
create public awareness about the cause of Tamil music, a series of confer-

can the subaltern sing? 167

ences were held between 1941 and 1945 in Devakottai, Madurai, Puduk-
kottai, Kumbakonam, Valampuri, and Ayampettai. The Tamil Icai Sangam
was established in Madras in 1943, as a parent to other Tamil Icai Sangams
founded in 1944 in Vellore, Erode, Tirunelveli, Kanchipuram, and Coim-
batore (Ramanathan Chettiar 1993, 7–11).
In December 1943 the first Tamil Icai Conference was held in Madras, at
the same time as the Madras Music Academy Conference. During the con-
ference several resolutions were passed to insure that Tamil songs would
be in majority in radio broadcasts, concerts, and university curriculums,
as Tamils were in majority in Tamil Nadu. Accordingly, it was resolved
that at the Madras radio station, 40 percent of broadcasted songs had to
be in Tamil, 40 percent in Telugu, and the remaining 20 percent in other
languages like Kannada or Sanskrit. However, in Tiruchirappalli, which
was more solidly Tamil, 80 percent of the songs broadcasted had to be in
Tamil, with the remaining 20 percent reserved for songs in Telugu and
other languages. In addition, the conference resolved that in concerts the
opening and closing songs—so often in Telugu, Sanskrit, or Hindustani—
had to be in Tamil. Music sabhas were called on to enforce this resolu-
tion. Members of the conference asked that the Madras University music-
department syllabus reserve a solid 40 percent of required repertoire for
Tamil songs (Tami icai makānāṭu ceai nikacci mālar 1944, 61). It is not clear
if these resolutions were ever followed. For its own part, the Madras Tamil
Icai Sangam resolved to be a bastion of Tamil music in a city where the
music scene was dominated by Telugus and Tamil Brahmins. In 1948, after
Raja Sir Annamalai’s death, Dr. R. K. Shanmugham Chettiar was elected
president. Under his leadership a hall, the Raja Annamalai Manram, was
built in 1950 in the Esplanade area of George Town in Madras (Rama-
nathan Chettiar 1993, 28).23 That building, at a meaningful remove from
the Madras Music Academy in Mylapore, continues to house the annual
Tamil Icai Sangam conference and the Tamil Icai Sangam College of Music.

From the Shores of Lemuria

Members of the Tamil music movement traced the history of their demand
for Tamil songs back to Abraham Pandithar, medical doctor and author
of the monumental Karunamirtha Sagaram (Karuṇamirtacākaram), which ex-
tolled the refinement and musicality of the Tamil language and also in-
cluded a section of Pandithar’s own Tamil compositions for Tamil children.
Pandithar began his treatise on music with an account of the deluge that

168 can the subaltern sing?

destroyed Lemuria, a continent of Tamil speakers which had supposedly
existed to the south of India before it was subsumed by the ocean.24 The
imagination of an antique land was crucial to Pandithar’s theory of music
and its origins.
Always interested in music, Pandithar devoted himself to it in the last
ten tears of his life, having retired from his medical career.25 Raised in a
family who had converted to Christianity two generations before, Pan-
dithar had learned violin as a boy and it was thought that his interest in
church music would blossom. However, under the influence of the Tamil
renaissance, he became more interested in Karnatic music and particu-
larly in the idea of Tamil music. Bothered by the claim that Telugu was
more suitable than Tamil for music, he wrote Tamil lyrics for the tunes of
Telugu compositions and published these (Nadar 1954, 113). Yet this was,
Pandithar felt, merely an imitation. Only when he or any other student
of music was able to compose a completely new song with Tamil words
could the music really be called Tamil music. He was certain that music
was not only an artistic but also a scientific matter. By 1910 he had come
up with a method for composing new kritis, but his pursuits had led him
to an interest in the ‘‘theory’’ of Karnatic music: how many srutis were in
an octave? How were ragas constructed? What was the basic principle that
differentiated one raga from another? (114). In order to publicize his re-
search on these topics, Pandithar founded the Tanjore Sangita Vidya Maha
Sabha in 1912, and in the five following years convened seven conferences
for the purposes of debating these questions (115). He came into conflict
with the musical authorities of the time at the First All-India Music Con-
ference in Baroda in 1916, when he demonstrated his theory of twenty-
four srutis to the octave, rather than the twenty-two mentioned in the
Sanskrit treatises.26 The conflict concerned much more than the probably
inaudible difference of two srutis; on one side was the authority of the
Sanskrit treatises, and on the other was the ancient Tamil theory of music,
bolstered by Western rationality. At stake was the origin of Karnatic music
itself.
Indeed, the question of origins dominated Pandithar’s monumental
work Karunamirtha Sagaram: A Treatise on Music, or Isai-Tamil, which Is One
of the Main Divisions of Muttamil or Language, Music, and Drama, published
in 1917, two years before his death. Quoting liberally from the Bible, the
recently published editions of the Tamil ‘‘classics’’ Tolkappiam and Cilapati-
karam, and Caldwell’s grammar of Dravidian languages, he gave a detailed
account of how the ancient land of Lemuria was the cradle of Tamil civili-

can the subaltern sing? 169

zation and Tamil music, and of how Tamil civilization had given rise to the
rest of the world. Using passages from the Bible, Pandithar demonstrated
the antiquity of music, in particular, the existence of instruments such as
the nagaswaram before the deluge (9). By corroborating biblical accounts
of the deluge with Tamil accounts produced by the new generation of
Dravidianists, Pandithar showed that the deluge had been universal. From
this assumption it was logical to hypothesize that ‘‘the ancient continent
of Lemuria, which was to the south of India in the Indian Ocean, was
the original habitation of man, that it was the cradle of all nations, and
that after the destruction of the continent, the various races occupied the
shores which were against them’’ (23).
The original inhabitants of Lemuria spoke a primordial language,
which must have been Tamil. For Pandithar, one proof of Tamil’s primor-
dial origin was its sound: ‘‘The softness of the words, the fewness of the
letters and the comparative ease with which the words are pronounced
warrant us to infer that Tamil must have been the language of early man-
kind’’ (1917, 47). Another proof was the evidence that Tamil was indepen-
dent, that it ‘‘[stood] by itself without being affected by other languages,
in spite of other languages borrowing from it’’ (47). To demonstrate this
point, Pandithar included long lists of ‘‘original’’ Tamil words that had
been borrowed by Sanskrit, Hebrew, and other languages (43–47). The pri-
ority of languages was thus reversed; Tamil was now the ancient original,
and Sanskrit the corrupt imitation.27
But the most important proof of the primordiality of the Tamil lan-
guage was, for Pandithar, its proximity to music, which implied that the
perfect language was one that was united with music. Muttami, in which
grammar and music were two equal branches of a larger entity called
language, was, for twentieth-century Dravidianists like Pandithar, the
epitome of the perfect language, a language that was literally enchanted.
‘‘Noble readers!’’ he urged,

If we want to understand clearly the subtlety and antiquity of Indian

music we would do well to make a few observations on the Tamil lan-
guage which includes within itself . . . music. The period of the origin
of Indian Music is as ancient as the . . . Tamil language, and the sweet-
ness of Indian Music is the sweetness of the language itself. Just as the
language stands unmixed and unaffected by other languages, so also the
Music of South India is perfect in itself having special rules of its own
without seeking the aid of other music. (30)

170 can the subaltern sing?

 But just as the Tamil language had undergone admixture with Sanskrit
and other languages over the centuries, Tamil music had also fallen from its
state of original, Sangam-era purity. This had happened through transla-
tion. In a section titled ‘‘The Influence of the Aryans on the Music of South
India,’’ Pandithar stated that the Aryans (‘‘Brahmins’’) had learned music
from Tamils but then had written their music treatises in Sanskrit. It was
thus that South Indian music had become corrupted. By Pandithar’s calcu-
lation, South Indian music had been ‘‘in the hands of foreigners for the last
seven or eight hundred years,’’ ever since Aryans had come to the South
(1917, 120). ‘‘[The Aryans] have completely changed the names of some of
the most ancient ragams and technical terms, giving Sanskrit names. . . .
So we must understand that the ancient Tevarams are now found as ragas
in a new Sanskrit garb’’ (111–112). At stake for Pandithar was not only the
past but also the future of Indian music. In an essay on Indian music which
he sent to several English music journals, Pandithar decried the fact that
no one seemed to be interested in uncovering the ‘‘one theory and one law
of constructing melody for all people from the Himalayas to Cape Como-
rin and from the Indus to the Bay of Bengal’’ (217).28 To discover such a
law would, he implied, enable a return to the ancient past, that glorious
period when all Indian music—indeed, the music of the entire world—
was Tamil music. Pandithar’s vision is reminiscent of Bhatkhande’s vision
of music as a universal language uniting India; the difference was that for
Pandithar the universal language would have distinctly Tamil inflections.
Like those in the Tamil music movement, Pandithar was concerned
with determining the precise relationship of music and language. The
milieu in which Pandithar worked in the first decades of the twentieth
century, however, was very different from that of the 1940s, when the
movement for Tamil music arose. Working at the margins of Dravidian
studies, Pandithar was primarily concerned with the reconstruction of an-
cient Tamil music. The projects of reconstructing ancient Tamil music and
creating modern music in Tamil were based on a common set of concerns,
especially the desire to find a motivating relationship between music and
language. But the more immediate political purposes of each project dif-
fered; while the reconstruction of ancient Tamil music, beginning in the
1910s, sought to find a prior musical system distinct from (although per-
haps the basis of ) modern Karnatic music, the Tamil music movement of
the 1940s called on Tamils to seize the contemporary practice of Karnatic
music and make it their own.

can the subaltern sing? 171

Iron Ears
Music is well said to be the speech of angels; in fact, nothing among the utterances al-
lowed to man is felt to be so divine. . . . Serious nations . . . have prized song and music
as the highest. . . . [But] what a road men have travelled! The waste that is made in music
is probably among the saddest of all our squanderings of God’s gifts. Music has, for a
long time past, been avowedly mad, divorced from sense and the reality of things; and
runs about now as an open Bedlamite. . . . bragging that she has nothing to do with
sense and reality, but with fiction and delirium only.
—Thomas Carlyle, The Opera (1852)

T. K. Chidambaranatha Mudaliar (hereafter T.K.C.), aesthetician and man

of letters, began his essay ‘‘Caṅkītamum Cakityamum’’ (Music and lyrics)
with a Tamil paraphrase of these words from the Scottish essayist Thomas
Carlyle. Such comments, he marveled, sounded as if they could be from
‘‘somebody speaking about our present-day music’’ (Chidambaranatha
Mudaliar 1941, 140). How angry Carlyle would have been, T.K.C. wrote, if
he had known of the worth of Karnatic music in former days and how it
was being wasted in the present. Echoing the distinction drawn by H. P.
Krishna Rao, T.K.C. stated that whereas Western music was based on the
ascent and descent of scales and artful combinations of notes, or ‘‘calcu-
lations,’’ Karnatic music was, as Westerners themselves pointed out, ‘‘inti-
mately connected to feelings’’ (‘‘Kaṇakkiliruntu pirantatu mēlnāṭṭu caṅkī-
tam. Uṇarcciyiliruntu pirantatu nammuṭaiya caṅkītam’’). Western music,
being based on calculations, could only be appreciated by those who had
studied it. But Karnatic music, being born from feeling (uṇarcci), was based
on a natural philosophy, so anyone ought to be able to enjoy it (142). How-
ever, the concept of high art didn’t accommodate such a natural relation-
ship to music: pandits and vidwans made pronouncements that such music
as could be enjoyed by everybody was of a low standard. Musicians, T.K.C.
claimed, sang supposedly high-class music with such closed-up, artificial
voices so lacking in emotion that the common people had come to be-
lieve that in high-class music there was not supposed to be any emotion, or
bhāvam ( pāvam). They had become so accustomed to understanding and
feeling nothing when they listened to Karnatic music that they assumed
the music was devoid of emotion (143).
T.K.C. went on to address the problem of meaning in music and the
precise role of the sahityam, or lyrics. To those who had complained that
Karnatic music was not understandable to Tamilians because the lyrics

172 can the subaltern sing?

were in Telugu, the pandits had answered that they should learn Telugu
in order to fully enjoy the music. T.K.C. commented sarcastically on the
absurdity of this simplistic, utilitarian view of language: ‘‘If that is so, then
they need to put up a notice outside the concert hall saying ‘today there
will be a three-hour concert. All the songs will be in Telugu. Only those
who know Telugu should come. We are not responsible for those who do
not know Telugu. If they come, we cannot refund their tickets’ ’’ (1941,
14). Following this logic, those who came to only one or two concerts a
year would have to learn Telugu just for that, while those who knew Tamil
only should choose to go to dramas instead. But why should the audi-
ence be bothered to learn Telugu, T.K.C. asked, when most of the singers
themselves hadn’t learned it? In any case, learning a language in order to
speak it was one matter, but singing in another language could hardly be
learned. It was a matter not of merely understanding the language but of
enjoying it in all the senses of the Tamil word anubhavam (aupavam: ex-
perience). For how could one really understand a song in Telugu merely by
being told the meaning of the words? While Tamil translations of Telugu
lyrics for the benefit of Tamil musicians abounded, T.K.C. doubted that
they would help matters. The problem, as he saw it, was that there was a
difference between learning something and feeling it in one’s heart (‘‘itaya
pāvam uṇarvatu’’). To be a poet required years of training in one’s own lan-
guage, much less in a foreign one; given this, how could any singer with
a superficial knowledge of the meaning of the Telugu lyrics be expected
to sing them with feeling? (146).
Moreover, T.K.C. added, if any Tamilian said at that point that he en-
joyed the song, it was a lie, a self-deception (ēmāḷittaam). There could be
no true meaning without an enjoyment of sound, a subjective sense of
the motivation of language. To what authentically Tamil ears could those
Telugu words sound good, like drops of nectar (‘‘kātukku amirtam vārttār
pōl’’)? (1941, 148). Any Tamilian who doubted the law of the untranslat-
ability of language should try explaining a Tamil song word by word to any
Telugu-speaking person and see if they enjoyed it. They certainly would
not, T.K.C. said, and they would think one crazy for expecting them to. To
prove his point about the futility of translating songs from other languages,
T.K.C. related a story about a Karnatic vocalist who went to Calcutta and
gave a concert of Telugu and Sanskrit songs. The Tamilians in the audience
did not complain but sat docilely and acted as though they were enjoying
the concert. The Bengalis in the audience assumed from the Tamilians’ at-
titude that the songs were in Tamil, and exclaimed to the singer, ‘‘Because

can the subaltern sing? 173

the songs were all in Tamil we could not enjoy/understand them. Could
you sing some Bengali songs?’’ (149). The Bengalis, T.K.C. concluded, had
real respect for their mother tongue and were not afraid to ask for songs
in it. Tamilians, on the other hand, would be willing to say they enjoyed
any song, even a Bengali one, because they did not have the courage to
say otherwise. This ‘‘disease’’ of self-hatred, born of being made to study
English, to repeat words without knowing their meaning (ārttam), had
‘‘entered the bones’’ of Tamilians more than it had for other Indians (149–
50). As long as Tamilians were afraid to demand music in their mother
tongue, they would be deprived of the true pleasures of music. ‘‘Singing
and singing in Telugu, our bhāvam, our feeling has gone. The songs do not
stick in our minds; when the musician sings the song, the audience sits
like stone statues’’ (156).
For T.K.C. meaning in music was located not only in the lyrics or words
but in the ways those words resonated in the ears and hearts of listeners.
His struggle to explain the importance of hearing music with words in
one’s mother tongue reveals a quest for pure and true singing and listening
subjects, musicians and audiences who showed their identity as Tamilians.
The specter of audiences sitting still as stone statues was particularly dis-
turbing because it threatened to make a charade of classical music, a show
based on such ingrained pretense that it took a foreigner—a Bengali—to
point out that something was wrong. The Tamil poet and essayist Bhara-
thiyar had struck a similar note in his 1916 essay ‘‘Caṅkīta Viṣayam’’ (The
issue of music), claiming that the Tamil people had lost their feeling for
music (caṅkīta ñānam) from years of hearing the same Telugu and Sanskrit
songs over and over again. They sang them gulping and murdering the
words, imparting no feeling to them at all (216). In the name of a ghost
called the ‘‘vocal concert,’’ the musicians murdered the sweetness of the
music by slapping out the tala (‘‘ ‘Pāṭṭu kaccēri’ eu pēy vaittukkoṇṭu aṅkē
icaiyipaṅkaḷai kou tāla muakkattai pirataamākkutal takātu’’) (217).
‘‘Go to any district, any village,’’ Bharathiyar wrote. ‘‘Whichever vidwan
comes, it will be this same story. Because Tamilians have iron ears, they can
stand to listen to the same seven or eight songs over and over and over and
over again. In places where people have ears of flesh they would not en-
dure such a thing (‘‘Enta jillāvukku pō, enta kirāmattiku pō enta ‘vitvān’
vantālum, itē katai tā. Tamināṭṭu jaaṅkaḷukku irumpukātāka iruppa-
tāl varuhak kaṇakkākak kēṭṭuk koṇṭirukkiārkaḷ. Tō kātu uḷḷe tēcaṅkalilē
intat tupattaip pouttuk koṇṭirukka māṭṭārkal) (213).

174 can the subaltern sing?

The Sound of the Heart
Have you heard the voice of a newborn infant calling its mother? There is a special kind
of music in it.
—C. N. Annadurai, ‘‘Tamiari Maumalarcci’’

For those who had lost their feeling for music, the way to regain it was
to nurture their mother tongue, Tamil.29 In much of the discourse of
tamipau (Tamil devotion) in the 1930s and 1940s, music—not the high
art of Karnatic classical music but a generalized power of music, or musi-
cality—was repeatedly invoked to explain the untranslatability and special
power of the mother tongue. Essays on the concept of the mother tongue
in the 1930s and following decades frequently invoked music to illustrate
the subject’s special relation to his or her mother tongue. ‘‘Itaya Oli’’ (The
sound of the heart), an essay by T.K.C., and ‘‘Olic Celvam’’ (The treasure
of sound), written by his student the Justice S. Maharajan, both center on
the idea of there being a kind of music in language that makes it unique
and untranslatable. The music of Tamil, both argued, was only audible to
those who had Tamil as their mother tongue.
To experience this music within language, T.K.C. suggested in ‘‘Itaya
Oli,’’ one should go to any small Tamil village and walk the streets with
a keen ear. From sunrise to sunset, on any verandah, one might listen to
‘‘concerts of talk’’ (‘‘arattaik kaccērikaḷ’’), the speed and nuance of which
one could never find in the so-called civilized speech of the city (‘‘aṅkē
pēcum irukkia vēkamum nāyamum, nākarattil pēcum nākarika pēccil
illai’’) (1936, 34). In these concerts of talk, whatever was in the speaker’s
mind would come out directly in the act of speaking (‘‘uḷḷattil uḷḷe viṣayam
appaṭiyē vākkil varum’’) (34). One person might be speaking, but the
bond of understanding created by their common mother tongue would
make those listening feel that they themselves were speaking (‘‘Kēṭkia
ovvoruvarum tāṅkaḷētā pēcikoṇṭirukkiōm ea niaippil iruppārkaḷ’’)
(34). Music, here, was imagined as a vehicle of immediate and uninter-
rupted communication.
These village concerts of talk stood in sharp contrast to the music con-
certs in the city, T.K.C. wrote. Although the audiences for these concerts
were Tamils, they did not express their approval or enjoyment in Tamil
as they listened. It was all in Hindustani: ‘‘Bale Bale! Besh! Savash!’’ An
Englishman who only knew a little Hindustani, a word here and there,

can the subaltern sing? 175

might think that all the Telugu songs the singer was singing were Hindu-
stani, and therefore the audience was speaking to him in Hindustani. After
all, after the Telugu songs, singers would often end with a few Hindu-
stani songs. Was it any wonder that a crowd of hundreds of Tamilians
would use Hindustani rather than Tamil? It would hardly be surprising if
they used English expressions like ‘‘Capital!’’ or ‘‘Excellent!’’ An observer
might mistakenly conclude from this that Tamils had no feeling for music
(caṅkīta uṇarcci). But how could this be true, when even foreigners were
commenting on the remarkable musical feeling of the Tamils? Rabindra-
nath Tagore himself had written that Tamil Nadu was a land of music. Go
to any village, and you would find temples with endowments for nada-
swaram players and dance musicians. You couldn’t find anything like it
in the world. Such a tradition must have taken thousands of years to de-
velop, T.K.C. speculated. During all those years, would Tamilians have sat
like stone statues, unable to express their appreciation? No. They would
have at least said, ‘‘Rompa naāy irukkiratu’’ (It is very good) or roared
‘‘Aha! Aha!’’ or trilled ‘‘Aṭaṭā! Aṭaṭā!’’ or have drunk it in, saying it was like
‘‘honey, milk, or nectar’’ (‘‘tēnō, pālō, amirtamō’’). These ancient Tamils
would have said such things whole-heartedly (1936, 35–36).
For T.K.C., then, the question of how one’s response to music should
be registered in language came to the forefront. For a musical subject,
music should provoke an involuntary vocal response. The fact that Tamils,
to express their feelings about music, had words only in other languages
was a problem. Language that came naturally, for T.K.C., was ‘‘the sound
of the heart,’’ a language that existed outside of and before translation.
The real character of a language had to be felt precisely through what
was untranslatable into any other language: its sound. T.K.C.’s student Jus-
tice S. Maharajan (later president of the Madras Tamil Icai Sangam) con-
tinued this obsession with the sound of language in his 1962 essay ‘‘Olic
Celvam.’’ To an ear accustomed to the sounds of Tamil, he wrote, the
sound of the language (tami oli) gives pleasure (cukam) (Maharajan 1962, 1).
Like T.K.C., Maharajan suggested that to rediscover this pleasure in the
Tamil sound, one should listen to the speech of common people, especially
street hawkers, whose cries rang with alliteration and onomotopoeia and
the nāyam (melody, nuance) of the Tamil language. For Maharajan, street
hawkers’ cries illustrated what amounted to the motivation of language.
In these unpretentiously poetic calls, every choice of word, every repeti-
tion, seemed necessary; the street hawker obeyed the rules of meaning and

176 can the subaltern sing?

the conventions of sound as if they were one. ‘‘In this way, Tamil speech
and poetry sound to our ears and minds as one’’ (‘‘Ippaṭiyāka tami pēccum,
kaviyum nam kātukkum, karuttukkum icaintu olikkiaa’’) (2). Inter-
estingly, Maharajan used a verb whose referent was music (icai) to express
the unity between sound and idea. Ideal music, he suggested, was the most
free of all the arts because it did not have the problem of a split between
sound and idea. Poetry, because of its proximity to music through sound,
could come close to music if Tamilians recognized and enjoyed the music
in the sounds of the language they spoke and read. Words did for speech
and poetry what swaras did for music. In recognizing this, Tamilians would
see that they were heirs to an abundant treasure of sound (‘‘Icaiyil svaram
ceykia vēlaiyai, colli oli ceytu kāṭṭukiatu kavitaiyil. Ippaṭi valamāna
oliccelvattukkellām varicāka irrukiā tamia’’) (7).
The Tamil poet Namakkal Ramalingam Pillai, in the midst of his trea-
tise on Tamil music (Icaittami), written in 1942, included a page-long chap-
ter entitled ‘‘Tāy Moi’’ (Mother tongue) that makes similar points. ‘‘The
mother tongue,’’ he wrote, ‘‘is the language the mother feeds to us’’ (Tāy
moi eāl tāyār pukuṭṭiya moi). ‘‘Our knowledge begins with the words
she says. Afterwards no matter how many languages we study, we under-
stand them only through our mother tongue’’ (1942, 92). The power of
the mother tongue was such that it held the key to a person’s identity.
If you were to take a Tamil child away from his mother before he could
speak and give him to a Telugu mother to raise, the child would become
Telugu (92). The mother tongue, once instilled in a person, was irrevers-
ible; it was the language closest to one’s subjectivity, the place where the
work of translation ended. Music, suggested Ramalingam Pillai, occupied
this special territory of the mother tongue; it should not require trans-
lation but should be immediately understandable, a natural expression.
But this required the aid of the mother tongue; otherwise, it was merely
clever combinations of notes, spectacular acrobatics, meaningless sounds
that stuck in the ears but did not get to the mind or heart (88, 90). ‘‘How-
ever cleverly ragam, talam, and swaram are mixed together, they are still
only sound. . . . The forms of that sound can produce amazement in us. But
everything that thus falls on the ear will stop at the ear. It will not pene-
trate the mind. It will not melt the soul. It will not kindle the emotions’’
(88). By way of analogy, Ramalingam Pillai included a lengthy description
of an acrobat balancing on a wire and the thrilling effect it would have
on an audience. ‘‘But even a feat which requires so much practice, we will

can the subaltern sing? 177

forget even before we reach home. That is because it is mere entertain-
ment’’ (vēṭikkai) (90). Similarly, music with lyrics in another language, like
Telugu, was ‘‘useless’’ for Tamils (93). Many Tamil vidwans and audiences
had the uncanny experience of singing or hearing Telugu lyrics so often
that the sound of the lyrics became intensely familiar but the meaning was
unknown. The natural thing for listeners and singers to do was to Tamilize
the lyrics by singing/hearing them as the Tamil words they sounded like,
completely changing the meaning of the song and depriving them of its
bhāvam (6–7).
The problem, as T.K.C. saw it, was that Tamils were too accepting of
music in other languages, ignoring the imperative to hear music in their
mother tongue. Tamils’ feeling for art (kalai uṇarcci) was dwindling, so
much so that Madras had earned the epithet ‘‘iru māṭiya ceai’’ (Madras
covered in darkness) (Chidambaranatha Mudaliar 1936, 153–54). Tamil mu-
sicians and composers were blindly following the example of Thyagaraja
by composing, like him, in Telugu. This had succeeded in producing a
lot of ‘‘fake Thyagarajas’’ ( poli tyākayyarkaḷ). It was as if a peacock had
happened to dance, and all over Tamil Nadu, from Tirupati to Kanyaku-
mari, a breed of turkeys had arisen. To Tamil eyes, they all seemed like
peacocks. What would poor Thyagaraja have thought? Instead of blindly
copying Thyagaraja, suggested T.K.C., musicians should follow his ex-
ample as one who had composed heartfelt lyrics in his own mother tongue
(153–54). Tamils should use Thyagaraja’s lyrics as an inspiration to com-
pose in Tamil, their own mother tongue. For ‘‘music in Tamil Nadu should
be the property of Tamils. Someone coming from the North or a West-
ern country could not enjoy it in the same way as a Tamil could. If they
said they did, it would be mere preaching’’ (153–54). Art and music could
not be transported and translated from place to place. ‘‘Just as we don’t
understand their music, they don’t understand ours. That is art. To take it
from one place to another is futile,’’ concluded T.K.C. (153–54). He used
the idea that Karnatic music was more intimately related to feelings than
Western music not to argue that it therefore could be universally under-
stood but to impress on his readers how far astray music had gone in
Tamil Nadu.
C. (‘‘Rajaji’’) Rajagopalachari was considered inimical to the Tamil cause
because, as chief minister of Tamil Nadu, he advocated compulsory Hindi
education in the late 1930s. Interestingly, however, when it came to music,
Rajaji’s respect for the ‘‘mother tongue’’ was unquestionable. It was true,
he said at the Tamil Icai Conference in Madras in 1943, that music was

178 can the subaltern sing?

really composed of sounds. But if that was so, why wasn’t instrumen-
tal music sufficient? If music was just sounds, what could explain its spe-
cial power? Such power, Rajaji argued, came especially from vocal music;
when music was joined with language, it produced a special kind of hap-
piness (‘‘Moikauta cērttuk kēṭkumpōtu oru tai cantōṣam uṇṭākiatu’’).
As soon as one heard the music, one understood the words (‘‘Caṅkī-
tam maatil euvatuṭa, ea collukiār eum kūṭa eukiatu’’) (Muthiah
1996, 25). It was as if words and music were united in this seamless kind of
listening, this immediate understanding. One needed to understand the
words when somebody spoke, Rajaji wrote, so why did some think it not
necessary to understand the words when somebody sang? (27). ‘‘Meaning-
less’’ music might produce a vague sort of blissful feeling, but besides that it
had no value (‘‘Poruḷaa icai, ētōmātiri āantamē tavira vēu pirayōjaam
illai’’) (27). For Rajaji, listening to music ideally had to approximate the
supposedly immediate, unselfconscious understanding produced between
two native speakers of Tamil. Music was valuable not as an external collec-
tion of beautiful sounds—Rajaji compared this to the folly of a Tamilian
writing a love letter in English—but as the sound of one’s own subjec-
tivity, a sonic reflection of one’s self. More to the point, this would be a
self, and a subject, only imaginable within the new kind of discourse about
music that was developing. Just as music, for Rajaji, produced a kind of
immediate understanding, which required no work of interpretation for
the listener, this new type of subject was one whose outer self and speech
transparently reflected his thoughts and emotions.
The movement for music in Tamil did not entail a mere switching of
languages from Telugu to Tamil, but rather a new concept of the role of
language in music. The Telugu of most Karnatic compositions was hard to
understand even for a Telugu-speaking audience; it was quasi-formulaic,
specialized, half-literary, ‘‘colloquial’’ only in the sense that it was not clas-
sical Telugu. A Telugu listener might hear a familiar word here and there,
but this was not the same kind of ‘‘understanding’’ that proponents of
the Tamil music movement called for. For them, as Rajaji’s essay suggests,
music was to be communicative, and the presence of foreign-sounding
words had the potential to lead such communication astray. The idea of
musicality emerged in this discourse as a metaphor for such communica-
tion, a communication so direct that the medium almost effaced itself. The
threat of Telugu words was the threat that their very foreignness would
make them musical, but in a negative sense: they would cease to have
meaning and would become mere sounds.

can the subaltern sing? 179

Music and Caste Politics
‘‘To praise and love one’s mother, mother land, and mother tongue is natu-
ral to humans. In Tamil Nadu this love has been raised to an uncommon
pitch, as we attribute divine qualities to Tamil,’’ wrote Kalki Krishna-
moorthy in his regular column on music and dance, ‘‘Āṭal Pāṭal,’’ in 1943.
‘‘But in this era there have been born some peculiar beings who seem to
have a hatred for Tamil. Without flinching, they say there are no good
songs in Tamil, and that Tamil is not a suitable language for music, and
that singing in Tamil will bring the decline of Karnatic music’’ (1943b,
13). He went on to contrast the concerts in the Madras Music Academy
with those held under the auspices of the Indian Fine Arts Society. The
music academy concerts, he wrote, had been going on for the last sixteen
years, organized by Tamilians, but no Tamil songs were ever sung there;
it seemed as if they had taken a vow to dismiss them. The academy ad-
ministration actively enforced its judgment about the inferiority of Tamil
songs by changing the programs of performers. M. S. Subbulakshmi, for
example, was forced in 1943 to add a Telugu kriti to her program so that
the majority of her songs would not be in Tamil (14). In the case of Madu-
rai Mani Iyer, Kalki wrote, the academy administration eliminated three
out of the four Tamil songs he wanted to sing when they printed the
program for his concert. Mani Iyer, in reply, sang a sarcastic impromptu
lyric addressing the academy administration: ‘‘Mahā Mahā akatemikkārar-
kaḷē! Uṅkaḷukku tami eāl avvaḷavu veruppā? Pōkaṭṭum, ioru teluṅku
kirttaattai kēḷuṅkaḷ’’ (Great, great academy men! Do you hate Tamil so
much? Well, so be it. Hear another Telugu song) (15). By contrast, the con-
certs held under the auspices of the Indian Fine Arts Society, wrote Kalki,
were much more relaxed in atmosphere; the musicians sang whatever they
wished, which was mostly Tamil (15).
During the years of his column ‘‘Āṭal Pāṭal,’’ Kalki wrote regularly on
the topic of the Tamil music movement. What was most at stake for him
in these debates was a certain politics of identity they seemed to assume:
that Karnatic music was primarily the property of Brahmins, while Tamil
was primarily the mother tongue of Tamilians, who were by definition not
Brahmins. Anti-Brahmin rhetoric was employed most forcefully by E. V.
Ramasami (E.V.R.), particularly during protests held in the 1930s against
making Hindi a compulsory language in schools (Ramaswamy 1997, 194–
204). The domination of Karnatic music by Brahmins was, for E.V.R.,
another example of the way Brahmins had taken over what was essen-

180 can the subaltern sing?

tially Tamil culture and disguised it in Sanskritized terms in order to call
it their own. In an editorial in the newspaper Kuṭi Aracu in 1930, he criti-
cized the Brahmins for monopolizing Karnatic music and claimed that
non-Brahmin artists were being mistreated. He arranged a music section
at the second Self-Respect Conference at Erode in 1930, calling on non-
Brahmins to patronize non-Brahmin artists and to make sure that non-
Brahmin children had musical education (Arooran 1980, 255). In a pam-
phlet published in 1944, entitled Tami Icai, Naṭippu Kalaikaḷ: Ii Ea Ceyya
Vēṇṭum? (Tamil music and drama: What still needs to be done?), E.V.R.
wrote that Brahmins (whom he glossed as ‘‘Aryans’’) were dominating all
the fields of cultural production. ‘‘We need to enter those fields, find out
the secrets, and make them public,’’ he urged (Ramasami 1944, 7). The pur-
pose of music, he claimed, was primarily to communicate, and only sec-
ondarily to give pleasure. The ceyti (message) had to be more important
than the cuvai (pleasure, taste) (ibid., 3). Accordingly, he urged that songs
on Puranic or bhakti themes be dismissed in favor of songs with useful
social messages (ibid., 16).
An editorial in Kuṭi Aracu from 1943 had already suggested that Tamils
needed to hear songs in ‘‘good, natural Tamil’’ about the courage of the
Tamils and their love for the Tamil language, songs that would undo the
bad deeds of the Brahmins (Tami icai makānāṭu ceai nikacci malar 1944,
113). This editorial took for granted the opposition between Brahmins
and Tamilians. Who was opposing the Tamil Icai movement? The Brah-
mins, came the blunt answer. It was the Brahmins who were eating the
Tamils’ salt, coming to live in Tamil Nadu, building up their strength by
Tamils’ hard work, making Tamils into slaves, smearing an Aryan coating
onto Tamil arts (‘‘tami kalaikku ārya mēpucciṭṭu’’), mixing up Sanskrit
words with Tamil to make a maṇipiravāḷam (110–11). Brahmins, the edi-
torial stated, could not be Tamilian because their mother tongue was not
really Tamil but some mixture of Sanskrit and English.
It was in the midst of such discourse that Kalki Krishnamoorthy, a
Brahmin himself, wrote his ‘‘Tamiicaiyum Pirāmaarkaḷum’’ (Tamil music
and Brahmins). For more than a thousand years, music in Tamil Nadu
had developed without regard for differences in jati (caste), he wrote.
How, then, could music suddenly become the rightful property of only
non-Brahmins? ‘‘There must be some sort of curse on our country,’’ he
wrote. ‘‘Take any perfectly good thing and sooner or later, in some form,
the jati problem will enter into it’’ (Kalki 1943a, 12). Those in the Self-
Respect Movement wanted to make the Tamil music movement a non-

can the subaltern sing? 181

Brahmin movement, and Brahmins were agreeing with them, saying that
they were not Tamilians. But, Kalki asked, did not Brahmins speak Tamil?
A Tamilian was anyone whose mother tongue was Tamil. Brahmins in An-
dhra Pradesh and Bengal considered themselves Andhras and Bengalis, so
why did Brahmins in Tamil Nadu insist on their own foreignness? The
Tamil music movement was a movement of the Tamil people (tami mak-
kal), not a non-Brahmin movement. Many Brahmins had served the cause
of Tamil music, like the composers Gopalakrishna Bharati, Maha Vaidya-
nathayyar, and Papanasam Sivan. Brahmins, just like other Tamilians—like
all humans, in fact—needed songs in their own mother tongue; even Raja-
gopalachari, the Brahmin chief minister of Tamil Nadu who supported
compulsory Hindi education, was in favor of the Tamil music movement,
Kalki stated (12–13). Whereas language issues still divided Brahmins from
non-Brahmins and provoked arguments about who was really Tamil, the
music issue, he argued, should be able to bring all Tamilians together.
C. N. Annadurai, script and story writer, founder of a branch of the
Tirāvita Munēa Kaakam (Dravidian Advancement Party) and later chief
minister of Tamil Nadu, presented the music issue as a cause for Tamils
against Aryan domination. Although he did not use the language of caste,
in his lengthy piece entitled ‘‘Tamiari Maumalarcci’’ (The renaissance
of the Tamils), he presumed to speak for non-Brahmins, who, he assumed,
were the ‘‘real’’ Tamils. The reasons that some opposed the Tamil music
movement were not aesthetic, as they claimed, but political: they were
afraid that if Tamilians pushed for music in Tamil, they would soon de-
mand a separate Tamil nation. Being afraid to say this, the small oppo-
sition group instead claimed that there were no classical songs in Tamil
or that they were of inferior standard. Why should this be? Were Tamils
vagabonds, wanderers (nāṭōṭik kūṭṭamā)? No they were indigenous people,
the owners of this land (‘‘Illai! Paaṅkuṭi makkaḷ! Inta nāṭṭukku contak-
kārarkaḷ!’’) (Muthiah 1996, 43). Tamils, he maintained, had a long musical
history. After all, what did Tamils sing before the 1800s, when Thyagaraja
and Dikshitar composed in Telugu and Sanskrit? There must have been
Tamil music. In the time of the Tamil kings, he wrote, Tamil music and
arts had spread all over the world. Where was that music today? When
the British came, Aryans had ‘‘enchanted’’ them with ‘‘propaganda’’ (‘‘āri-
yari piracyāram āṅkila nāṭṭavariayum māyakiu’’), saying that they had
taught the Tamils everything about art and music (49). Thus had developed
a mistaken idea that Karnatic music did not really belong to Tamils, that it
was different from the original Tamil music. In fact, underneath the show

182 can the subaltern sing?

of technique that alienated many listeners uneducated in Karnatic music
was a kind of music that was completely understandable to Tamilians. The
Tamil people were not interested in feats of musicianship but in songs that
they could understand. Thus, if M. S. Subbulakshmi sang Telugu kritis,
as she did with beautiful correctness, it might move the other musicians
but not the nāṭṭu makkal (common people, people of the land). Other than
knowing that she sang beautifully, they would not enjoy/understand the
meaning, or mood (cuvai: literally, ‘‘taste’’) (53). But if M.S. happened to
sing a Tamil song, immediately the ears and hearts of the Tamilians would
be thrilled (‘‘Tamiari maamellām kulirntatu’’); the music’s sweetness
(ipam) would truly reach everybody (53).

A Language of Aesthetics?
Those who argued against the Tamil music movement did so in the name
of ‘‘classical music,’’ fearing that if Tamil songs became a priority, the stan-
dards of Karnatic music would drop. At the heart of the debate was the
question of just what kind of language music was to be imagined as: a
mother tongue or an aesthetically motivated language of art? Was the lis-
tener to find meaning in the words of the songs or in the melodies, the
sound itself ? Was the listener’s appreciation grounded in a sense of iden-
tity based on his mother tongue or a sense of awe inspired by his awareness
of a great classical tradition? This dichotomy between the mother tongue
and a classical language of art and aesthetics was imagined to be at work in
music precisely because it had, as Bernard Cohn’s work suggests, already
become part of the assumed hierarchy of languages in India. If music was
imagined through the analogy of language, it had to belong in one cate-
gory or the other. Thus, in 1941, in response to the growing activity of the
Tamil music movement, the Vellore Sangita Sabha went so far as to pass a
resolution saying that ‘‘the essential thing about music was its melody, and
its appeal to the inner emotions of the listeners did not matter’’ (Arooran
1980, 259). In the same year, the Madras Music Academy passed a similar
resolution: ‘‘It should be the aim of all musicians and lovers of music to
preserve and maintain the highest standard of classical Carnatic music and
no consideration of language should be imported so as to lower or impair
that standard’’ (260).
In the same year, T. T. Krishnamachari, one of the founders of the
Madras Music Academy, published an essay entitled ‘‘Karnāṭaka Caṅ-
kītamum Tami Icai Iyakkamum’’ (Karnatic music and the Tamil music

can the subaltern sing? 183

movement) in the Tamil weekly Ananta Vikatan. He argued that the Tamil
Music movement (which he referred to as pāṣai piracciai, the ‘‘language
problem’’) was concerned less with music than it was with politics and that
those who argued for Tamil songs knew nothing about the art and aesthet-
ics of music. There was a difference, he maintained, between a pure music
concert (cuttamāa caṅkītak kaccēri) and bhajans or katha kalakshepam. For
the latter, he implied, it would be justified to demand that the songs be in
an understandable language. But those who demanded that pure Karnatic
music be in a language they could understand were missing this crucial
distinction (Krishnamachari 1941, 38). ‘‘In order to enjoy a music concert,’’
he wrote, ‘‘you need to have some knowledge’’ (ñāam). You might ask,
couldn’t someone with no knowledge at all enjoy a concert if the singing
were interesting enough? Yes, it would give them peace of mind, perhaps.
But this is not experiencing music (caṅkīta aupavamakātu). Now that it
had suddenly become imperative for everyone to enjoy music, there was
a real threat to the standard of Karnatic music. For ‘‘the tendency of the
common man is to destroy anything that he cannot understand’’ (38–40).
One might condone Tamil songs, wrote Krishnamachari, if they followed
the restrictions and conventions of Karnatic music. ‘‘Everyone knows that
the way most Tamil songs are ‘composed’ nowadays is by taking songs in
other languages and putting Tamil words to them. . . . A Tamil hearing
these might get a lot of pleasure from understanding the words. But that
is not what it means to experience music’’ (40). For Krishnamachari, Kar-
natic music followed certain conventions quite apart from the concerns
of everyday language; it was an independent musical language. People did
not attend music concerts to hear the words of songs as though they were
some kind of religious discourse; for that, one could go to a temple. The
music concert was for a different kind of aesthetic enjoyment.
Like Krishnamachari, other opponents of the Tamil music movement
argued that it was a political, not a musical, movement, identifying their
own cause, the cause of classical music, as purely aesthetic and thus above
politics. An editorial in The Hindu in 1943 claimed that R. K. Shanmugham
Chettiar’s demand for Tamil songs was a kind of ‘‘extremism’’ (Tami icai
makānāṭu ceai nikacci mālar 1944, 91). ‘‘Raja Sir Annamalai Chettiar has
said that the Tamil Music movement is not a political movement. . . . But
the resolutions they passed concerning percentages of Tamil songs to be
sung on the radio and in concerts seem like an effort to control the propa-
gation of art’’ (91). The language problem had only recently become an
issue in music. ‘‘Until now, Tamil Nadu has never seen such exclusive-

184 can the subaltern sing?

ness/provincialism [taniyuṭaimai], and such parochialism will not go on in
the future,’’ the editorial continued. For in India there was a long tradition
of musicians composing in languages other than their mother tongues and
other than those of the place they were living. In the days when Thyaga-
raja was singing his Telugu songs in Tamil Nadu, when Purandara Dasa,
though he was from Maharastra, sang in Kannada, no one complained;
when Mirabai and Kabir sang, thousands of people who did not know the
language flocked to hear them (92). The editorialist suggested that a clas-
sical tradition of music comparable to that of the West had existed in India
for countless generations and that the ‘‘language problem’’ should be too
provincial and trivial to affect it. In the West, the great composers like
Bach and Beethoven were universally recognized, even though they were,
technically, German. There might be some in the West who, out of a mis-
guided affection for their own mother tongue, refused to listen to these
great composers, but the Western music establishment would never allow
such people to dominate (92).
Opponents of the Tamil music movement also made their case on aes-
thetic grounds, claiming that Tamil was simply not suited to the aesthet-
ics of Karnatic music. The reason that Tamil songs were only sung at the
end of concert programs, stated the 1943 editorial, was that Tamil songs
didn’t provide enough opportunity for vidwans to show their prowess
at raga elaboration; they didn’t provide enough inspiration for manō-
dharma (improvisation) (Tami icai makānāṭu ceai nikacci māḷar 1944, 95).
Moreover, Telugu was thought of by many as ‘‘the Italian of the East’’
(a phrase that dates from nineteenth-century philology), a mellifluous
language that was particularly suited to Karnatic music. Krishnamachari,
among others, argued that Tamil, with its hard consonant sounds and
consonantal endings, was not suited for Karnatic music, that such sounds
would ruin the melody. ‘‘The words kuṭṭi, kāṭṭi, and kātti may have spe-
cial meanings. But when the tongue is flowing [ puraum: literally, ‘‘wal-
lowing’’] in the sound of music, to have to pronounce such sounds as ‘iṭ,’
‘ip,’ and ‘it’ would interfere with the tempo of the music [Āāl caṅkīta
kāattil nākku puaumpōtu ‘ṭ,’ ‘p,’ ‘t’ epatellām uccarikka vēṇṭumāāl
kālap piramāam tallip pōyviṭum]’’ (93). The wallowing tongue produces
no articulation; it enables a voice to emerge with a minimum of interfer-
ence from the tongue—a voice that comes from within the body with-
out being in contact with it. For Krishnamachari and others, what distin-
guished music from speech and what made a voice musical, was this very
bodiless character.

can the subaltern sing? 185

 Here, then, was a discourse about musicality being located not in a
human subject but in language itself. Musicality was something inher-
ent in language, rather than a property of a subject’s relation to the lan-
guage she spoke; thus, Telugu could be claimed as a more musical language.
Thyagaraja himself, argued the opponents of the Tamil music movement,
had made the choice to compose in Telugu a hundred years earlier be-
cause he had realized it was a more musical language than Tamil; those
in the 1940s should follow his example. In arguing that Telugu was better
suited to music than Tamil, Krishnamachari and others were suggesting
that the proper way to appreciate music was through an ostensibly objec-
tive, aesthetic contemplation, to hear it as a musical language that crossed
over the boundaries of one’s subjective attachment to any mother tongue.
T. V. Subba Rao’s comment at the 1943 conference of the Madras Music
Academy summed it up: ‘‘To fix a percentage of songs with reference to
any particular language is to make a woeful confusion between literature
and music. Sangita is solely the art of expressing beauty in sound; and to
require the aid of language is to reduce its power. The grandest achieve-
ment of Karnatic music is Raga, which knows no bounds. To insist on lan-
guage in musical composition is to be deaf to the highest beauty of music.
There is none so deaf as those who will not hear good music in whatever
language it may be’’ (Proceedings of the 17th Madras Music Conference,
1943, 12).

Genre and Voice

In the decades after the 1940s, the language issue in Karnatic music was
not resolved in the domain of classical music itself but in the prolifera-
tion of other categories of music alongside the classical. Thus, beginning
in the 1940s, ‘‘devotional music’’ and ‘‘film music’’ came into being as cate-
gories distinct from classical music. What distinguished them was the im-
portance given to the words and the audience’s understanding of them:
devotional music and film songs were in Tamil and appealed to audiences
on the basis of their words as well as—or perhaps more than—their music.
Since the 1940s, these categories have become increasingly distinct from
Karnatic classical music in regard to musical style, the musicians who per-
form them, and their audiences. They have become industries in their own
right, and part of their appeal is their distinctness from Karnatic music.
At the time of the Tamil music movement, however, the same musicians
sang all of these genres: D. K. Pattammal and M. S. Subbulakshmi sang on

186 can the subaltern sing?

the classical stage many Tamil songs that they also sang in Tamil films of the
time; the vocalist Madurai Somasundaram sang Tamil songs on the classi-
cal stage in the 1940s that would later be considered devotional rather than
classical. A vocalist in Madras who was actively concertizing in the late
1990s pointed out that in principle there was really no difference between
classical and devotional music.30 Thyagaraja, the great classical composer
himself, had sung his songs in devotion to the deity Rama; he would have
concentrated on the words, not obscured them with elaborate variations
and improvisations. In classical music, singers routinely swallowed words
but were not criticized for it. But in singing devotional music, the vocalist
pointed out, one could not afford to garble any of the words because devo-
tional music uses music as a ‘‘communicative medium.’’ The devotional
singer has to convey not only a message but an emotional involvement
with that message. Thus there is a drastic difference in the ambience of
a classical concert and that of a devotional concert; in performing devo-
tional concerts, this vocalist said, he and his accompanists felt much more
relaxed and in communication with the audience. Another vocalist who
until recently sang both devotional and classical concerts said that any song
could be sung in either style; it was a matter of changing one’s ‘‘technique.’’
One could give a ‘‘classical touch’’ to a song by singing raga alapana, san-
gatis, and swaras with it or give it ‘‘devotional fervor’’ by singing it plain
but with attention to the words.
But as the idea of classical music as a kind of pure, absolute music be-
came the dominant frame for assessing Karnatic music, it became increas-
ingly difficult for musicians who sang in Tamil and paid attention to the
enunciation and meaning of words to be considered classical singers. Mu-
sicians active in the 1950s, like Sirkali Govindarajan, Dhandapani Desikar,
M. K. Thyagaraja Bhagavatar, and K. B. Sundarambal, who sang devotional
and film songs as well as classical music, were never fully accepted into
the Madras Music Academy’s canon of classical musicians; they were in-
stead labeled as ‘‘devotional’’ or ‘‘film’’ singers. Even the vocalist who stated
that there was no difference, in principle, between classical and devotional
songs revealed that in the 1990s it was not possible to make a name by
doing both because you got ‘‘labeled’’ as one or the other; he had decided
to sing only classical concerts in an attempt to shed his ‘‘devotional’’ label.
The 1940s debates about the importance of language in music and about
what kind of language music was were thus in one way resolved by the par-
titioning of music into the categories ‘‘classical’’ and ‘‘devotional’’; ‘‘devo-
tional’’ became the category in which many of the heroes of the Tamil

can the subaltern sing? 187

music movement found their place in the 1950s and 1960s. Devotional
music became the realm in which expressing oneself in Tamil was privi-
leged. Meanwhile, on the classical stage, after the initial furor of the Tamil
music movement, Tamil songs continued to be sung for the most part only
in the thukkada section of ‘‘lighter’’ songs at the end of concerts; to this
day, only six or seven Tamil songs are generally accepted and sung as clas-
sical compositions.
Yet the ramifications of the language debates also play themselves out
in perhaps more subtle distinctions of voice. In January 1998 a disciple of
the late vocalist Madurai (‘‘Somu’’) Somasundaram spent the better part of
an afternoon explaining to me what was special about his teacher’s style.
For most of his career, which spanned the period from 1930 to the 1970s,
Somu, who belonged to the Icai Veḷḷālar community, refused to sing in
any language but Tamil. He carried this to such an extent that he even
refused to sing kalpana swaras using the note names sa, ri, ga, ma, pa, da,
and ni, which he presumed to be Sanskrit. Instead, he sang Tamil words.
His concerts always included a long viruttam, a type of recitative genre in
which the singer elaborates a raga in alapana style while singing lines of
Tamil poetry. In demonstrating the style of these viruttams for me, Somu’s
disciple hit on the most distinctive characteristic of Somu’s singing: the
passion that was not just in the words he sang but that carried through
to his voice. Somu’s was a strong voice, with tremendous range, a voice
that was not afraid to strain, crack, or leap under the burden of words and
their emotion. It was, the disciple explained, distinctly unlike the ‘‘Kar-
natic’’ style of voice, which was nasal and small, and which proceeded
hesitantly and tremulously from note to note. Somu’s voice was a truly
‘‘Tamil’’ voice; in singing, his vocal gestures were large, impassioned, and
daring. By comparison, the Karnatic style that now predominates ema-
nates hesitation and control. To demonstrate this, the disciple imitated
a Karnatic singer trying to find his pitch, starting and restarting, cover-
ing one ear, covering both ears. Somu, by contrast, would begin with a
large, unhesitating flourish. Another singer who had been trained in sing-
ing Tevaram (Tamil religious poetry from the twelfth century) as well as
in Karnatic music drew a similar contrast.31 Karnatic vocalists, he said, sing
blandly, without taste (uppucappu illāmal: literally, ‘‘without salt’’), whereas
truly Tamil singers like Somu had a curucuruppu, a viruviruppu (both ono-
matopoeic Tamil words indicating briskness, crispness). The relation be-
tween the singing subject and his/her voice was entirely different in each

188 can the subaltern sing?

of these, Somu’s disciple implied. In the Tamil style the voice was the un-
guarded expression of emotion, while in the Karnatic style the voice was
controlled as though it were an external instrument. In presenting a voice
connected to its mother tongue, Somu’s disciple critiqued the Karnatic
voice and the Brahmin, middle-class music establishment it had come to
represent.

Can the Subaltern Sing?

At one level, the debates concerning the Tamil music movement can be
seen as debates about what exactly constitutes classical music. The move-
ment gained prominence at a key moment in the canonization of Karnatic
music as the classical music of South India. The role of language in music
became a central issue in deciding what repertoire and which musicians
would be admitted to the canon. At stake in the debates concerning the
Tamil music movement was not just the issue of words but the problem of
voice: who could have a voice in Karnatic music, and what kind of voice
was it to be?
In discourse concerning the Tamil music movement the specter of an
unruly voice figured prominently: as a voice that swallowed, gulped, or
‘‘murdered’’ words; as a voice that sang without emotion, without con-
nection to its owner, and might therefore be coming from anywhere; or
as a voice that got carried away with the emotional effect of its mother
tongue and forgot the conventions of music. The metaphor of music as a
language emerged as an effort to control this potentially unruly voice by
once and for all specifying the nature of the relationship between music
and words, and therefore the ‘‘meaning’’ of music. But even as the analogy
of music to language was used by those engaged in these debates to argue
that their cause was ‘‘purely artistic’’ or ‘‘natural’’ and not political, it was
appropriated and interpreted in various ways to serve different agendas.
For nationalists, considering music as a language was a way of rationaliz-
ing music, a way of universalizing it so that it could reach, and subject, in
every sense of that word, all the inhabitants of India. For the proponents
of the Tamil music movement, the analogy between music and language
went both ways, for just as language in music could make a listening sub-
ject musical again, music could be invoked to explain a subject’s special
relationship to his mother tongue; music represented an ideal to which
language should aspire. For the opponents of the Tamil music movement,

can the subaltern sing? 189

the analogy of music to language meant that music could be conceived as
an aesthetically motivated language in itself, which had to be divorced as
much as possible from actual language in order to be appreciated.
In chapter 3, I showed how the concept of the voice as issuing from an
inner domain and as transcending the body led to a discourse that opposed
true or real music against ‘‘mere acrobatics.’’ For Rukmini Devi, acrobat-
ics represented the threat of purely physical, automatic action, spectacle
without content, exteriority without interiority. Likewise, for the Tamil
poet Ramalingam Pillai, writing around the same time as Rukmini Devi,
acrobatics represented the polar opposite of what he thought music should
be. Watching it was mere vēṭikkai, amusement derived from being an on-
looker who is not really involved, and not anubhavam, experience derived
from direct involvement. According to Ramalingam Pillai, only through
language—the Tamil language—could one achieve anubhavam. For those
holding the opposite point of view, like T. T. Krishnamachari, Tamil songs
might provide a pleasant experience for audiences, but they could not pro-
vide the experience—and here he used the same word, anubhavam—of
music, which relied on recognizing music as a language of its own and
training oneself to hear it. Significantly, for those on both sides of the de-
bate, the only way one’s encounter with music could rise from vēṭikkai to
anubhavam was by conceiving music through an analogy with language.
The analogy between music and language depended on the erasure of
earlier practices in which music and language stood in other relations to
each other. This erasure was partly accomplished by the canonization of
Tamil literature and Karnatic classical music as categories mutually op-
posed in their orientation to the mother tongue. The subsequent debates
about language in music reflect a politics of voice that is shot through with
the notions of meaning, comprehension, and sincerity demanded of mod-
ern musical and linguistic subjects. By insisting on language as a metaphor
and model for articulating the relationship between voice and singing or
listening subject, this politics of voice engenders the poles of the debate I
have analyzed here. What makes this politics of voice a particularly mod-
ern approach to music is the way it conceives musical meaning. ‘‘Meaning,’’
the key term here, relies on two parallel oppositions. One is the split be-
tween linguistic or sonic form and referential content, also conceived as
the split between exterior and interior (and the notion, in both cases, that
meaning is located in the latter). The other is the split between a speaker’s
voice or speech and his self, where the (external) voice is assumed to be
a reflection of the (internal) self. Indeed, it is this split between exteri-

190 can the subaltern sing?

ority and interiority that is articulated in the contrast between vēṭikkai
and anubhavam.
Motivating the debates surrounding the Tamil music movement was
the figure of a subaltern, voiceless, still, unresponsive as a statue, deaf by
virtue of his ‘‘iron ears,’’ and, above all, unable to appreciate music—a
figure profoundly disturbing because he seemed to possess no interiority.
Members of the Tamil music movement argued their cause in the name
of bringing music back to this imagined subaltern in the name of making
him sing, of restoring his voice and thus his senses. Yet humming beneath
this project was the possibility that the subaltern might already be sing-
ing, but with a voice that could be neither domesticated into Karnatic
classical music nor assimilated to the desires of the Tamil music move-
ment. Manohar Devadoss’s memoir suggests an alternative to the politics
of voice, and the insistence on meaning, that came to dominate discus-
sions of music in the mid-twentieth century. For just as nationalists, Tamil
and otherwise, rushed to pin down the meaning of music, it had gained a
new kind of power. This, as Devadoss eloquently describes, was precisely
the power to move, and to affect, without being dependent on the logic of
language, and thus to circulate outside the bounds of modernity’s claims
to its meaning.

can the subaltern sing? 191

5 $ A Writing Lesson
musicology and the birth of the composer

The scriptural operation which produces, preserves, and cultivates imperishable ‘‘truths’’ is
connected to a rumor of words that vanish no sooner than they are uttered, and are therefore
lost forever. An irreparable loss is the trace of these spoken words in the texts whose object
they have become. Hence through writing is formed our relation with the other, the past.
—Michel de Certeau, The Writing of History

The birth of writing (in the colloquial sense) was nearly everywhere and most often linked to
genealogical anxiety. The memory and oral tradition of generations, which sometimes goes
back very far for peoples supposedly ‘‘without writing,’’ are often cited in this connection.
—Jacques Derrida, Of Grammatology

In the mid-1980s a scandal swept through the Karnatic music world.

S. Balachander, the flamboyant, self-proclaimed ‘‘veena virtuoso,’’ pub-
lished a booklet entitled Eutiār Puttakattai! Kiappinār Pūtattai! (He wrote
a book . . . and . . . kindled the genie!), in which he argued that the
nineteenth-century composer-king of Travancore, Swati Tirunal, never
existed and that the compositions attributed to him were really composed
by others.1 Through the 1980s Balachander elaborated his arguments in
various publications and open letters, spinning a lurid tale of fake nota-
tions, lost books, sequestered palm-leaf manuscripts, and mistaken iden-
tity. The arguments that swirled around the case went far beyond the figure
of Swati Tirunal himself, bringing into question not only the reliability
of notation but the very notion of the composer and the way modern
musicology should be conceived.
The debates quieted down with Balachander’s unexpected death in
1992. Swati Tirunal was reclaimed as a true composer, and those who
doubted him were deemed to be part of an unruly faction with ulterior
motives. Yet the Swati Tirunal issue had inserted a kernel of doubt into the
firmament of Karnatic classical music. For at the very scene of composing,
the original moment of authenticity that made Karnatic music classical,
Balachander had posited a scandal. If one composer could be discredited
by the historical probing Balachander suggested, couldn’t any composer
presumably be shown to be a fake by similar methods? What constituted
authorship, and what guaranteed authenticity? What order of things made
the assertion that Swati Tirunal was a fake appear scandalous? 2
Issues of authorship and authenticity that reemerged in the Swati Tiru-
nal debate bring into focus the shifts in musical institutions and ideas about
music that occurred in the twentieth century. At stake in the Swati Tiru-
nal debate was the boundary between the written and the oral—which
had greater authority, and exactly how should the boundary be crossed?
The vexed nature of these questions is apparent in the conflicting ideas
about the role of notation. Notation was seen both as a site of resistance
to the encroachment of Western classical music and as a space of progress.
It was imagined both as a guarantor of literacy (and therefore classical
status) and as a transparent and legible representation of essential orality
(and therefore Indianness).3 In spite of its promises, however, notation was
also viewed with ambivalence. The desire to capture the voice in writing
was bound up with the fear that the voice could be lost precisely by being
completely captured by writing.
At another level, the Swati Tirunal debate brought into question
the relationship between theory and practice, text and performance. If
theory was meant to provide a structure, or system, underlying the music,
how did one determine where basic structure ended and embellishments
began? How much of a musical work was original, and how much was
added on? Where did one draw the line between composition and impro-
visation? Such questions point to the intimate connection between the
notion of the authorship of the composer and the authority of a system of
rules about structure.

A Composer Is Born
Balachander began an open letter to the musicologists of Madras in 1989
with the statement that the Swati Tirunal question was not just a musical
issue but also a musicological one.4 His letter, which summarized what he
had already presented in a fifty-page thesis at a press conference in Madras
in April 1989, emphasized not merely musical knowledge but the pro-
duction of that knowledge. First and foremost, Balachander’s letter was a

a writing lesson 193

statement about the necessity of doing music history, of consulting written
sources. In the proceedings for the 1887 jubilee of the Madras Gayan Samaj,
a musical organization with branches in both Pune and Madras, he had
found the first reference to the musical activities of a certain Kulasekhara
Perumal, a musical king of Travancore (r. 1829–1847) who would later be
known as Swati Tirunal. This original mention of Kulasekhara Perumal
provided the source for all later references used by the Swati Tirunal schol-
ars but had been completely unacknowledged. The reason for this, Bala-
chander stated, was that the reference contained ‘‘incriminating evidence’’
showing not only that the name ‘‘Swati Tirunal’’ did not exist in 1882 but
also that the composer-king to whom that name was later assigned may
also never have existed.5
Balachander demonstrated the truth of his assertion using the Madras
Jubilee Gayan Samaj Proceedings. He began by situating the proceedings his-
torically, amid the late-nineteenth-century revival of interest in Indian
music, a revival that focused on standardizing and preserving Indian music.
At the center of this revival was the Gayan Samaj, established in Pune in
1874 with the decidedly nationalist purpose of training young boys and
girls in their ‘‘national’’ music. In 1883 a branch of the organization was
established in Madras. Balachander noted that between 1880 and 1915 the
publication of music books with notation was gaining ‘‘special attention
and momentum.’’ 6 Several musicologists, in fact, had made use of the print
media to publish music books with notation. The late nineteenth cen-
tury, Balachander maintained, was a ‘‘transition period’’ in which the old
guru-sisya parampara (lineage, tradition) was dying out and being replaced
by books with notation and institutionalized teaching methods. In capi-
tal letters Balachander trumpeted the first bit of his damning ‘‘evidence’’:
‘‘and, here came the immediate necessity for music books with
‘notation,’ for use everywhere, for sales everywhere, for homes
and institutions everywhere, for domestic use and abroad.’’ 7
Balachander proceeded with a close reading of the Gayan Samaj pro-
ceedings’ account of the 1882 visit to Pune of the then maharaja of Travan-
core. His analysis hinged on the maharaja’s lack of voice in the text. For
surely any maharaja—especially one whose close ancestor was a composer-
king—who visited a newly established music institution would surely
speak a few words of appreciation and encouragement. Yet this maharaja,
Balachander commented, merely had his Dewan read a short reply, which
perhaps implied that the maharaja had not been so musically minded after
all or that he had not even been there. Such speculations were only con-

194 a writing lesson

firmed by the maharaja’s replies to a series of questions asked by the mem-
bers of the Pune Gayan Samaj. For Balachander, the very fact that these
questions and the answers to them were written (not spoken) implied
that the maharaja himself could not answer them or was not present to
answer them. The questions, submitted as a written memorandum, re-
ceived a written reply only three years later. The content of the answers,
Balachander argued, showed that there was little musical life in Travancore
during the so-called composer-king’s reign.8 In answer to the question of
whether Hindustani music was practiced in Travancore, for example, the
maharaja appeared to reply that primarily Karnatic music was practiced
and that, too, ‘‘at an indifferent level.’’ To a question regarding whether a
particular song had been composed by Kulasekhara Maharaja (presumably
Swati Tirunal), the reply should have been a ‘‘plain and simple’’ yes or no,
Balachander asserted; a composition (and this shows the late-twentieth-
century assumptions under which Balachander was operating) should have
had a single, unambiguous author and date. Yet the reply instead obfus-
cated the matter, saying that it was impossible to date the composition,
‘‘as every year His Highness produced lots of them.’’ 9 The last answer, as
far as Balachander was concerned, was an ‘‘open confession.’’ In answer to
a question about music schools in Travancore, the reply stated that there
were none and that music was not taught under any system of notation.10
With this evidence, Balachander proceeded to flesh out his account of the
twentieth-century invention of Swati Tirunal.
According to Balachander, the reference to the composer-king Kulase-
khara Perumal in the Gayan Samaj publication triggered the imaginations
of those in Travancore in the 1920s and 1930s who were interested in posit-
ing a Kerala school of Karnatic music that would equal the Tanjore tra-
dition. The compositions that were eventually attributed to Swati Tiru-
nal were actually composed, he maintained, by the musicians employed
in the Travancore court. For example, the composer Irayaman Thampi,
who had been employed in Swati Tirunal’s court, had apparently pub-
lished a book of the lyrics of his compositions as early as 1854, but this
book had been conveniently ‘‘lost,’’ Balachander suggested, and many of
Irayaman Thampi’s compositions had ended up being ascribed to Swati
Tirunal. Furthermore, Balachander continued, the royal family of Travan-
core had appointed musicians to supply notations for compositions by the
court musician Tanjore Vadivelu as well, then printed them in their own
printing press.11 These compositions were published in 1916 as having been
composed under the royal command of ‘‘H. H. the Maharaja of Travan-

a writing lesson 195

core.’’ Yet several years later, Balachander wrote, these same compositions
were claimed as the sole work of ‘‘Kulasekhara Perumal.’’ 12 Such a name,
which referred more to a lineage than to a particular person, could not
satisfy the modern demand that a composer be a single distinctive indi-
vidual. How could one Kulasekhara Perumal be distinguished from the
next? And so, Balachander wrote, the ‘‘perpetrators’’ of the hoax decided
to invent a new name: ‘‘Yes! They decided to create a new person with the
novel star name’’ of Swati Tirunal, and then claimed this form of naming
was a centuries-old ritual convention.13
In 1939 a music college by that name opened in Trivandrum, and Hari-
kesanallur Muthiah Bhagavatar, who had been responsible for ‘‘rediscover-
ing’’ and notating Swati Tirunal’s compositions, was appointed principal;
in the same year, under the direction of Muthiah Bhagavatar, a compre-
hensive collection of Swati Tirunal compositions was published under the
auspices of the Swati Tirunal Academy of Music (K. V. Ramanathan 1996,
18). Balachander stated that as it was the custom for the court musicians to
compose in the name of Padmanabha, the tutelary deity of Trivandrum,
the perpetrators of the Swati Tirunal hoax conveniently decided to make
‘‘Padmanabha’’ Swati Tirunal’s mudra, or composing signature.14 Thus, the
newfound composer appeared to have an enormous number of works to
his name. In 1940 a portrait of Swati Tirunal was hung in the Madras Music
Academy next to the portraits of the trinity to signify his status as a great
composer (K. V. Ramanathan 1996, 20). Balachander closed his letter by
claiming that the Swati Tirunal question was a problem of naming, of
writing—in short, a historiographical issue—and expressed the hope that
his own publications and letters would see the fake composer to his right-
ful end.
Balachander’s assertions challenge certain common assumptions about
authorship and authority, suggesting that in the twentieth century, older
practices in which different principles of authority operated were overlaid
by the modern institution of authorship. Publishing compositions ‘‘under
the command of ’’ the maharaja of Travancore evokes an older form of pa-
tronage in which authority lay with the patron, not necessarily with the
‘‘original author’’ of a particular work. The fact that the same compositions
were published several years later as the songs of Swati Tirunal suggests an
entirely different set of values, one in which authority and authorship are
tightly linked; publishing demands a single author who can be named and
distinguished from his ancestors. In a similar vein, the idea of a mudra in
the lyrics of a composition, now taken as evidence of authorship, may have

196 a writing lesson

originated as a very different concept. Mudras, formulaic phrases gener-
ally incorporated into the last line of a composition, are now commonly
thought of as the composer’s ‘‘signature’’ (Peterson 1984, 167–68). The term
mudra itself has a range of meanings: distinguishing mark, stamp, brand,
impress, royal seal, emblem, badge, and mark on a ballot paper (Winslow
1862, 881; Subramaniam 1992, 844). While we might think of such things as
signifying an ‘‘author’’ just as a signature does, there is an important differ-
ence. Royal seals, emblems, and badges are not just identifying marks but
material objects which are themselves endowed with authority precisely
because they are standardized and alienable from the figure of authority.
Someone who wears a royal badge or stamps a royal seal thus does not
claim authorship or originality for himself but invokes the authority of a king
or deity through a standardized or formulaic sign.
Certain composers, such as Thyagaraja, seem to have used their own
name, so that many of Thyagaraja’s mudras, in the context of composi-
tions addressed directly to the god Rama, translate as ‘‘Thyagaraja entreats
you’’ or ‘‘May Thyagaraja be your servant’’; the author’s name is itself in-
corporated into the lyrics. Other composers’ mudras, however, were not
their own names but those of the deities in whose name they composed,
suggesting a different locus of authority and a different set of ideas about
authority as well. In his examination of the colonial encounter of Dutch
Calvinist missionaries with Sumbanese ritual speech in Indonesia, Webb
Keane has suggested that one of the problems that ritual speech posed
for the missionaries was that its authority was based on the idea that the
words did not originate with the speaker. Contrary to the notion of the
sincere speaking subject, the power of ritual speech lay in its capacity to
portray the speaker as ‘‘someone who is not their author or the agent of
the actions they perform. . . . The signs of power are conceived to be gen-
erated by a source that remains distinct from the bodily individuals who
wield them’’ (Keane 1997, 680). In using the name of a deity, then, a ‘‘com-
poser’’ might be not so much ‘‘signing’’ his work as using the name to
invoke or call the deity or king. Instead of referring to the authorship of
an absent composer, ‘‘Padmanabha’’ might be a sign that effectively makes
the authority of the deity, or the king who worshipped him, present. In-
deed, Balachander suggested that it was precisely this latter strategy that
nineteenth-century court musicians employed when they composed ‘‘in
the name of ’’ the tutelary deity of Trivandrum.
In 1982, just before Balachander had come out with his case against
Swati Tirunal, K. P. Sivanandam and K. P. Kittappa, descendants of Tan-

a writing lesson 197

jore Vadivelu, wrote an article for the popular Tamil journal Kumutam, in
which they argued that Vadivelu and his brothers had merely translated the
Telugu songs they had composed for King Serfoji of Tanjore into Sanskrit
and inserted the name Padmanabha to make the songs suitable for Swati
Tirunal’s court.15 The article brought issues about originality and com-
posers’ integrity to the forefront, suggesting that ideas about these subjects
in the late twentieth century were different from the logics of authenticity
that operated in the royal courts of the nineteenth century. Whereas com-
posing in the late twentieth century carried connotations of originality
and individual work, such ideas could hardly have existed in the milieu of
the royal courts, where musicians were commissioned to produce songs for
their kings and regularly moved between courts. Indeed, when I discussed
the Swati Tirunal issue with a musician in Madurai in 1998, he asserted
that in those days musicians couldn’t have had the same concerns about
originality and authorship.16 They circulated around the courts of South
India, composing in the name of whoever happened to be their patron; it
didn’t matter who composed the pieces as long as they were attributed to
the person in power. The life of Harikesanallur Muthiah Bhagavatar was a
good example, he argued. Living at the time of the transition from royal
courts to music institutions, Muthiah Bhagavatar, like many other musi-
cians who were samasthana vidwans in courts, got an academic position in
a music college when those courts folded. While at Mysore, he composed
in the name of Mysore’s tutelary deity, Chamundiswari. As head of the
Swati Tirunal Academy of Music in Travancore, however, he took on the
job of notating many of the ‘‘neglected’’ or ‘‘lost’’ Swati Tirunal kritis. The
situation of the nineteenth-century musician situation, my informant de-
clared, could be compared to that of a man with a double-bordered veshti:
if he wore it on one side, the border would be red, but the next day he
could wear it on the other side and have a purple border.
Such speculations about the nature of composing in the nineteenth
century seem logical enough. But taking them to the extreme by assert-
ing that all the songs attributed to Swati Tirunal were really composed by
his court musicians would ‘‘lead to chaos in South Indian Karnatic music,’’
declared the violinist V. V. Subrahmanyam. In a small book entitled Satya-
meva Jeyate (Let Truth Reign), Subrahmanyam offered an impassioned refu-
tation of Balachander’s allegations. Yet although he was arguing against
Balachander, both were waging their battles in a field determined by simi-
lar assumptions: art for its own sake, originality, and the sole authorship
of the composer. In response to the allegation that Swati Tirunal’s songs

198 a writing lesson

were really composed by Vadivelu of the Tanjore Quartette, Subrahman-
yam noted a story about how Swati Tirunal would not allow his court
musicians to compose in his name, insisting that music should be only for
god and therefore in the name of the deity. ‘‘While the Maharaja has pro-
hibited his courtiers to sing in praise of him will he allow other kritis to be
published in his name?’’ Subrahmanyam asked (1986, 6). Here, he equates
Swati Tirunal’s prohibition of songs in praise of the king with the idea
of art for its own sake, free of political motivation, putting a decidedly
twentieth-century spin on a nineteenth-century practice.
Yet there are many ways in which even the twentieth-century prac-
tice of Karnatic music militates against such notions of authorship and
the composer. The same composition sung by two different people might
be almost unrecognizable when sung with different sangatis (variations)
or elaborated in different spots. The twentieth-century composer Papana-
sam Sivan, it is said, could hardly recognize his own compositions when
they were sung by other musicians (K. V. Ramanathan 1996, 20). The pos-
sibility of change to the point of unrecognizability seems to threaten the
very idea of original compositions and the authorship of the composer.
Subrahmanyam conceded that it was possible that many of Swati Tirunal’s
compositions would have been greatly embellished by his court musicians
when they were sung. But the ‘‘framework,’’ he insisted, would remain the
same (Subrahmanyam 1986, 8). At stake in the idea of an ‘‘essential struc-
ture’’ or ‘‘framework’’ of a composition were the agency and originality
of the composer, indeed, the very notion of the composer. The defini-
tion of a composition depended on the idea that any song had an essential
structure which was laid down by the composer and subsequently varied
by other musicians who sang it. Structure implied the use of notation, a
method by which the basic structure could be laid out and made perma-
nent. As Balachander alleged that Harikesanallur Muthiah Bhagavatar and
his successor at the Swati Tirunal Academy of Music, Semmangudi Srini-
vas Iyer, had composed the tunes for found lyrics and then attributed them
to Swati Tirunal, the controversy moved from discussions of composers’
motivations to the definition of composition itself and the proper way to
do musicology.
In a response to Balachander, Brig. R. B. Nayar, a musicologist from
Kerala, claimed that it was wrong to say that Muthiah Bhagavatar and Sem-
mangudi Srinivas Iyer (popularly referred to as Semmangudi) had dug up
lyrics and composed new tunes for them, thus inventing Swati Tirunal’s
repertoire. Rather, they collected what were already complete composi-

a writing lesson 199

tions with specified ragas and talas, ‘‘not just a bunch of lyrics found on
palm leaves’’ (Nayar 1997, 24). The absence of exact notation, in Nayar’s
view, pointed to another, perhaps even more authoritative oral tradition
by which the kritis had been preserved in their ‘‘pristine form’’: the court
musicians known as Mullamoodu Bhagavatars. In the court of Swati Tiru-
nal they had sung in a style known as Sopanam, a style peculiar to Kerala,
characterized by a slower pace and less ornamentation and improvisation.17
‘‘Unadulterated’’ versions of these kritis could have been collected with
the help of the last generation of Mullamoodu Bhagavatars, claimed Nayar.
Through an oral tradition preserved by court musicians native to Kerala,
who had no ulterior motives and were permanently bound to Swati Tiru-
nal’s court, the original compositions could have emerged into notation.
Proper musicology, in Nayar’s 1990s vision, would have required listen-
ing to the aging Mullamoodu Bhagavatars and painstakingly recording,
through notation, exactly what they sang, no more and no less.
The Muthiah Bhagavatar-Semmangudi team, however, worked with a
decidedly different notion of what it meant to do musicology. The Dewan
of Travancore state, C. P. Ramaswami Iyer, had proposed in 1937 that the
musical compositions of Swati Tirunal were a great contribution of the
state to culture and should therefore be revived (K. V. Ramanathan 1996,
18). At the request of the royal family, Muthiah Bhagavatar, newly ap-
pointed as the principal of the Swati Tirunal Music Academy in 1939,
began the task of collecting and notating Swati Tirunal’s compositions. His
son, H. M. Vaidyalingam, who assisted him, recalled the process: he and
his father went to different places in search of elderly people who might
remember songs. Muthiah Bhagavatar ‘‘reduced’’ the songs to notation and
then ‘‘polished’’ them (aakupaṭuttu: literally, ‘‘to make beautiful’’) (ibid.,
19). What was involved in this polishing? Apparently it was a process of
making the kritis conform to the sound of Karnatic music, rather than to
the ‘‘original’’ Sopanam style. Brigadier Nayar states that instead of pre-
serving the songs as they had been sung by the court musicians, Muthiah
Bhagavatar and Semmangudi reinvented them, increasing the tempo and
adding sangatis that were reminiscent of other Karnatic composers, so that
the songs would sound like Karnatic music and please audiences in Madras
(1997, 25–26). In the case of compositions where only lyrics were available,
Muthiah Bhagavatar and Semmangudi ‘‘retuned’’ the lyrics. Their object
seemed to be not recovering an original sound, but instead creating music
that was plausible, and pleasurable, to their own ears.

200 a writing lesson

 Muthiah Bhagavatar would identify the raga and tala which seemed
appropriate to him vis-à-vis the lyrics and sketch the music for them.
Semmangudi would then work on them further. When [Muthiah
Bhagavatar] felt satisfied with the outcome, both would go to Maha-
rani Sethu Parvathi Bayi who was highly knowledgable in music. She
would listen to the song as . . . reset to tune and rendered by the young
research assistant. She might suggest a change here or there but, once
she gave her seal of approval, the composition would be considered
ready to be released. (K. V. Ramanathan 1996, 18)

Thus, the process of recovering lost music in the 1940s seemed to involve
retuning lyrics to the preferences of one’s own ear. The newly composed
music, when approved by the requisite number of ears, would then be
agreed on as the true composition. In this process, the notation was the
origin of the composition, the act that brought it into being. By the 1980s
and 1990s, however, such a process could not have passed for true musi-
cology.18 As Balachander’s allegations and the refutations of V. V. Subrah-
manyam and R. B. Nayar show, musicology at the beginning of the twenty-
first century has a decidedly different set of assumptions. Notation, in this
new order, refers back to an authoritative act of composition by a single
composer: an original act occurring prior to the composition being re-
corded in written form. Thus it is only through the positing of a voice
before writing, an authoritative oral tradition, that the notion of the com-
poser becomes possible. Looking at this shift in the order of things be-
tween the 1940s and the 1980s, one can see that the notion of an authen-
tic ‘‘oral tradition’’ and the authority of writing to represent it emerged
simultaneously.

Nation and Notation

We fear we must defer the prospect of a universal language of music till the millennium
arrives.
—Sourindro Mohun Tagore, ‘‘Hindu Music’’ (1874)

In 1874 the newly founded Pune Gayan Samaj put out a statement of
its rationale, which included, among other things, providing an arena in
which Indian music would be respected and preserved.19 After detailing
the numerous activities that had been planned in this regard, the statement
read, ‘‘Lastly, the Samaj will be instrumental in preserving our nationality

a writing lesson 201

in the sense of our possessing an indigenous art of singing, which, unlike
English music, has challenged all its attempts at being reduced to writ-
ing’’ (Gayan Samaj 1887, 34). A sense of pride in Indian music, according to
this statement, followed from the fact that it could not be written in nota-
tion; this recalcitrance, in fact, was precisely what made it Indian and kept
other music from influencing it. Music, in India, was to be kept Indian by
being kept away from writing. The Gayan Samaj, in keeping with its spirit
of good relations with the British, invited Lord Mark Kerr, whose ‘‘vocal
powers were of the real indigenous type,’’ to become a member. But Kerr,
apparently troubled by the attitude toward notation, replied with a piece
of advice.

You imply, I think, although all possible musical instruments are to be

welcomed to perform at the Gayan Samaj, that science can have no
place there, for the music to be performed has hitherto challenged [it].
I presume you will continue to defy all attempts to put [your music]
in writing. Now without a science, that is to say, knowledge without
the power of writing your music, so as it can be made a study of . . . ,
you can have no art. . . . I, very seriously, invite you to do what, against
your opinion, I maintain is very possible, namely—put on paper—put
into writing all the quaint and melodious airs that I have heard sung by
your children, Mhotvallas, and others. Let this be arranged with care
and good taste, and, I repeat, put into writing what has hitherto defied
you (35).

Kerr’s admonition apparently had its effect. By 1879 the Gayan Samaj
was singing a different tune entirely. Its main object had become to con-
vince the Indian intelligentsia and the West that Indian music was an ob-
ject worthy of study, possessed of ‘‘a science . . . such as will vie in its nicety
with the Sanskrit grammar, which is recognised as almost the perfection
of deductive logic’’ (Gayan Samaj 1887, 20). The problem, however, was
that there was no way to represent this logic. ‘‘It is musical notation which
we want. . . . It is true we have a musical notation we can claim as our own,
but we think it is not sufficient nor elegant enough to mark the various
graces of Hindu music with the rapidity of a phonographer’’ (20). The idea
of preserving the Indianness of music by not writing it down had given
way to the fear that a lack of notation was causing Indian music to ‘‘fade
away.’’ The Gayan Samaj announced its plan: ‘‘We think the English system
of music [notation], such as it is, cannot be adopted by us without making
necessary changes; this we mean to do ere long’’ (20).

202 a writing lesson

 Accordingly, during the fateful visit of the Travancore maharaja to Pune
in 1882, the members of the Gayan Samaj, headed by Capt. Charles Day,
included in their list of questions about music in Travancore a request for
‘‘airs written correctly in the European notes’’ (Gayan Samaj 1887, 23). The
reply included a lengthy meditation on the difficulties of putting ‘‘Hindu
music’’ into European notation. They went beyond the problem of finding
someone who was conversant with both systems of music. How was one
to represent the quarter tones, ‘‘infinitesmally minute and delicate shades
as in a painting by a master artist’’? How could one capture the ‘‘unbroken
easy flow’’ of a vocalist over half a gamut? Indeed, how could one convey
the concept of raga itself ? The problem, concluded the writer, was one
of ‘‘translation’’ (27). He vowed to have the task attempted by one of the
Nayar brigade and, ‘‘if it is possible,’’ to have it sent to Captain Day.
Putting Indian music into European notation, then, was not merely a
matter of adding extra signs to show the peculiar features of Indian music.
It instead involved translation, putting the music into a kind of circula-
tion between two languages. The Bengali musicologist Sourindro Mo-
hun Tagore feared that with such translation Indian music would come
to occupy a strange territory, neither properly Indian nor properly West-
ern. In 1874, the same year as the opening of the Gayan Samaj, Tagore was
involved in a dispute with the inspector of schools in Bengal, Charles B.
Clarke, who had written an article for the Calcutta Review advocating the
use of staff notation (Farrell 1997, 68). In his lengthy reply, entitled ‘‘Hindu
Music,’’ published in the Hindoo Patriot in 1874, Tagore wrote, ‘‘Every nation
that has a music of its own has also its own system of notation for writing
it. Whether that system be an advanced one or not, it cannot be correctly
expressed in the notation of another nation, however improved and sci-
entific it may be. . . . Anglicized as we have become in many respects, we
confess we prefer our national system of notation for our national music’’
(1874, 366). Notation seemed to mark a kind of last frontier, a space of
resistance to the encroachment of Western sounds and ideas.
Whereas a few years earlier the preservation of Indian music seemed to
depend on the absence of writing, the reverse was now argued; the ques-
tion of the need for notation was quickly eclipsed by the question of which
notation was best for Indian music. Tagore argued for an Indian notation
on the grounds that it was simpler and more ‘‘natural’’ than the European
staff notation. Whereas European notation required eleven lines to ac-
commodate the different clefs, the Indian system required only three lines.
Moreover, in the Indian system, the three lines marked the natural divi-

a writing lesson 203

 20 Text explaining
that Indian staff
notation requires
only three lines.
In S. M. Tagore,
Hindu Music (1874).

sion of the voice into chest sounds, throat sounds, and head sounds (Tagore
1874, 367). Whereas Clarke had argued that European staff notation was
so transparent that a Bengali who knew no English might simply look at
the notation and play a piece of Western music, Tagore argued that this
was not only impossible but misguided. In contrast to Clarke’s vision of a
universal notation for all the music of the world, Tagore envisioned a veri-
table Tower of Babel: the supposedly sufficient staff notation would have
to be adjusted and augmented by so many new signs that it would become
unrecognizable (382). By contrast, each nation had perfected a system of
notation that was transparent to its own musical system, he maintained.
In fact, it was so transparent that ‘‘in advocating the national system we
are simply following reason, truth, and history’’ (387).

Love at First Sight

Tagore had argued for an Indian system of notation on soundly nation-
alist grounds. Yet the idea of a universal musical notation so legible and
transparent that it could overcome linguistic and national differences held
a lingering appeal. The prospect of such notation became the consuming
passion of A. M. Chinnaswamy Mudaliar, a Tamil Christian and superin-
tendent of the Madras Secretariat (Raghavan 1961, 1). With a master’s in
Latin and music from Madras University, and a deep interest in English lit-
erature, Chinnaswamy Mudaliar was convinced that European staff nota-
tion was the best means of representing and preserving Karnatic music. In
1892 he began a monumental project, the monthly journal Oriental Music
in European Notation, in which the work of ‘‘every composer, living or
dead’’ in South India would be notated in a special adaptation of Euro-

204 a writing lesson

pean staff notation (Chinnaswamy Mudaliar 1892, iv). He and his brother
printed the journal with their own press. After finishing the music of
South India, Chinnaswamy proposed to notate North Indian music; the
music of China, Burma, and other parts of the East; and national anthems
from around the world. He also planned to publish a comprehensive dic-
tionary of musical terms, a history of Oriental music and musicians, and a
comparative sketch of international music. In an essay in the introductory
issue, ‘‘The Regeneration of Oriental Music in its Classical Form,’’ he wrote
that ‘‘any amount of foreign admixture and interpolation is introduced . . .
so that the magnificent indigenous system invented by the children of the
soil is threatened with prospects of speedy annihilation in the immedi-
ate future’’ (viii). The first step to counteract this was the ‘‘reduction’’ of
Karnatic music to staff notation while the second was the explanation of
the ‘‘fundamental principles of the science, not only in the principal Ver-
naculars but also in the English tongue, which now bids fair to be the
one universal language of the world’’ (viii). The reduction to staff notation
involved not only a translation into English but also an insertion of the
music into history: ‘‘It is absolutely essential to obtain complete historical
records regarding the date and authorship of every piece of music’’ (ix).
A potent politics of visibility ran through Chinnaswamy Mudaliar’s ar-
guments. Never short of metaphor, he characterized the purpose of the
notation project as ‘‘[bringing] forth into the open air that which lay con-
cealed and neglected like the ruins of an ancient city buried in subterra-
nean vaults; it is hoped that the debris will soon be cleared and beautiful
structures underneath exposed to the public gaze’’ (1892, xii). Indeed, the
unveiling, or revealing of Karnatic music before the eyes of the world was
the dominant metaphor in Chinnaswamy Mudaliar’s writing. It is signifi-
cant that he included about forty-five pages of introductory explanation
before getting to the notations themselves, as if he needed to ensure that
the notations would be seen in the right way. He hinted at what he meant
in a section called ‘‘Difficulties to be Surmounted’’; just as it was not pos-
sible to express every thought in written language or to convey every
quality of the speaking voice, it was also impossible to fully capture music
in notation. ‘‘No notation however complete can fully or accurately delin-
eate those magnificent foreshadowings . . . which fill the imagination of
the composer; not a millionth part of what he then feels can be put down
mechanically on paper; but when this has been done, the interpretation
given of this skeleton by even the most . . . skillful artist necessarily differs
from the rough outlines sketched by the author; how widely it must di-

a writing lesson 205

verge from his original ideal need hardly be mentioned’’ (2). He thus con-
ceived of writing, or notating, as a ‘‘mechanical’’ process, opposed to the
‘‘feeling,’’ ‘‘foreshadowing,’’ and ‘‘imagination’’ of the composer. Yet nota-
tion could function as a kind of consolation for the loss of the original, a
stylized likeness.

Nevertheless, as it is considered to be some consolation, in the absence

of a person esteemed, to possess his photographic likeness, and as an
oleographic portrait taken from a photo is found to be still more ac-
ceptable even if it really lacks many a grace and perfection of the living
original, so musicians of the land ought to be content with selecting
the clearest and most expressive of all existing symbols used in musical
language, although those cannot reproduce with absolute precision the
extremely subtle ideas of their brains or the deep pathetic emotions of
their hearts. (2)

As a kind of consolation for a lost original, notation worked best not ac-
cording to the logic of the photograph, which claimed to represent what
really was, but rather according to the logic of the portrait, a kind of sty-
listic likeness. The portrait represents by using certain recognizable con-
ventions, by highlighting some things and erasing others; it orders the
image a certain way so that it might be recognized, providing a ‘‘convinc-
ing likeness.’’ 20 Chinnaswamy Mudaliar, similarly, intended his notations
to be more prescriptive than descriptive, more like portraits than photog-
raphy, designed to allow for the future ‘‘reproduction’’ of musical ideas.
The staff notation, Chinnaswamy Mudaliar argued, was better equipped
to fulfil this role than Indian notation because it was a ‘‘pictorial nota-
tion.’’ By taking advantage of the visual medium, he maintained, the staff
notation did away with the need for a teacher or reference books; any-
one ‘‘tolerably conversant’’ with the principles of staff notation could sing
or play ‘‘at first sight’’ what was written. This was because staff notation
portrayed the intervals between notes as spatial relations on the staff. By
contrast, the Indian method of using the letters that denoted the pitches
of the scale (sa, ri, ga, and so on) were written on a straight line, ‘‘without
any indication to the eye as to whether they ascend or descend in the scale’’
(1892, 2). Moreover, he continued, in staff notation the pitch of the note
and its duration were represented by one and the same symbol, in contrast
to the cumbersome method of lengthening syllables to show duration in
Indian notation. Finally, Indian notation presupposed a knowledge of the
raga; it had no way of showing, for example, whether ri-1 or ri-2 was to

206 a writing lesson

be used without requiring background knowledge of the raga in which
the composition was set.21 All in all, Chinnaswamy Mudaliar stated, ‘‘The
adaptation of alphabetical characters and numerals for the extremely com-
plicated requirements of music will thus be seen to be a clumsy expedi-
ent, as unsatisfactory as it is antiquated. A separate language with suitable
symbols is absolutely necessary to ensure the required precision’’ (3). With
staff notation, there would be no necessity to refer to books or teachers;
everything would be apparent ‘‘at a glance,’’ leaving ‘‘no room for doubt,
conjecture, or hesitation of any kind. . . . The symbols [would] present
readily to the eye every detail which in other methods has to be retained in
the memory’’ (4). ‘‘One great advantage,’’ wrote Chinnaswamy Mudaliar,
‘‘is that Oriental music will be placed permanently before the eyes of the
whole world, instead of being addressed in transitory form as at present to
the ears of a few listeners . . . in other words it will become universal and
will no longer remain exclusive’’ (4). In such a move from ear to eye, he
implied, the music was freed from the musicians’ memories and allowed to
enter the realm of history. Others saw in this newfound visibility a greater
potential for originality. One reviewer, commenting favorably on Chin-
naswamy Mudaliar’s project, wrote that ‘‘at present the Hindu has to first
hear a tune, and be taught like a parrot before he makes it his own. By
the help of the European notation, he will be able to sing hundreds of his
national airs without ever having heard them before’’ (203).

Ornament and Order

Chinnaswamy Mudaliar did note that in order for staff notation to be
fully effective, certain symbols had to be added to represent gamakas, the
‘‘trills, shakes, slurs, and glissandos’’ that were typical of Karnatic music.
As long as these symbols were standardized and not haphazardly assigned
by individual printers, the system of staff notation would leave no room
for doubt (Chinnaswamy Mudaliar 1892, 8). Yet the process of notating
Karnatic music was not entirely straightforward, he admitted. Where the
notation ended and the use of ornament symbols began was problematic; a
simple turn could be written out or merely indicated by the symbol ~. The
notator had to be able to ‘‘discriminate the more important and essential
parts of a melody from what may seem its superfluous ornamentation,’’
and thus use the notation and symbols accordingly (7). The symbols, as
he demonstrated, left quite a bit of room for doubt as to their actual exe-
cution. One could choose instead to write everything out, thus expelling

a writing lesson 207

21 Text explaining notation for ornaments. In A. M. C. Mudaliar, Oriental Music in European Nota-
tion (1893). Courtesy of Music Division, The New York Public Library for the Performing Arts.

doubts. But then one ran the risk of obscuring the ‘‘essential structure’’ of
the music.
In Chinnaswamy Mudaliar’s logic, the relationship between ornament
and structure was analogous to that between ‘‘spurious’’ and ‘‘original.’’ The
goal of his work, he stated, was to ‘‘reproduce the compositions of the
great masters with all the accuracy and authenticity that can be secured’’
(1892, 33). However, Karnatic musicians had a tendency to make up varia-
tions (sangatis) which the composer had not intended and ‘‘tacitly pass
them off as genuine’’ (33). ‘‘It becomes therefore a matter of no small dif-
ficulty to discriminate between the spurious and the original, and the at-
tention of all educated classes ought to be directed to this point; otherwise
there will be nothing which can be recognized as the classical music of the
country’’ (34). Thus, the very idea of classical music depended on the as-
sumption that music had a basic structure as distinct from its ornaments,
an original as distinct from later additions. With the help of these distinc-
tions one could also begin to imagine Karnatic music as historical: ‘‘What
ought to be recognised as genuine originals will be clearly distinguished
from additions believed to have been made by later authorities’’ (34).
The problem of the ornament, which was to appear repeatedly in dis-
cussions of notation and musicology, also, for Chinnaswamy Mudaliar,
explained the problem of Europeans’ distaste for Indian music.22 Indian
music, like Indian architecture, dress, and religion, were distasteful to
Europeans at first because they could detect no sense in the profusion
of ornaments. ‘‘The numerous incisions made on the face of an Indian
woman and the saffron paint with which it is commonly daubed are ob-
jects of horror to him at first sight, but sooner or later he finds them to
be not so despicable after all. . . . The rules enjoining most of the semi-

208 a writing lesson

religious observances of the country are readily stigmatized as supersti-
tious and insensate, but are found on close inspection to be some of the
best sanitary and hygienic laws ever framed by human legislation’’ (Chin-
naswamy Mudaliar 1892, 8).
A certain politics of visibility was at work here. For anything to be
properly available to the European gaze, especially ‘‘at first sight,’’ a certain
structure or order had to be discernible beneath the surface. To such a gaze,
the surface appeared as a kind of mask of insensible repetitions and embel-
lishments. ‘‘It is so with Indian music, which lies under a mask at present.
Hitherto it has never been written or explained in a form which the West-
erner can read or understand. When a kriti is sung before him, he does not
see on what principle or in accordance with what postulates the repetitions
occur’’ (Chinnaswamy Mudaliar 1892, 8). To European ears, such repeti-
tions gave way to a feeling of ‘‘monotony,’’ because they could not locate a
vantage point or a structure within the music. For the Indian, by contrast,
repetition was the source of musical enjoyment: ‘‘Every Indian . . . knows
when and where to expect repetitions and variations during the recital of
a melody. . . . The listener understands why and wherefore the repetitions
occur, and is moreover entranced by the meaning attached to the words so
often repeated’’ (9). Notation, Chinnaswamy Mudaliar suggested, would
turn Europeans’ distaste into pleasure; not only would it do away with
unnecessary repetition, but it would also give them the sense of structure
they so craved. Moreover, it would give them an idea of what to expect, a
first glance that would eliminate the possibility of surprise and monotony
from their listening experience. In the process, by juxtaposing a written,
‘‘permanent’’ notation to an oral, ‘‘transitory’’ performance, it would give
the impression that the ‘‘original’’ lay in what was written, rather than in
the performance.
After a lengthy description of the ‘‘peculiarities of Oriental music,’’ the
journals published by Chinnaswamy Mudaliar included about 120 pages of
notated compositions. For each composition, the first words of the song
were given in English letters at the top, with Telugu and Tamil underneath.
Underneath that, he indicated the original language of the composition.
At the top left, the mode (melakarta) of the raga was given, then the raga
and tala. The top right indicated the composer’s name, as well as advice on
how to count, with a tempo derived from the metronome. Beneath was
advice on the ‘‘style of execution,’’ usually in Italian: staccato, allegro, dolce,
con spirito, and so on. The first line of notation showed the arohanam and
avarohanam (ascending and descending order) of the raga scale in staff nota-

a writing lesson 209

22 South Indian composition in staff notation, with words in English, Telugu, and Tamil. In
A. M. C. Mudaliar, Oriental Music in European Notation (1893). Courtesy of Music Division, The
New York Public Library for the Performing Arts, Astor, Lenox and Tilden Foundations.
tion (transcribed in the key of C for the convenience of keeping the notes
within the staff ), with the Indian note names written underneath. The
composition itself was written in continuous fashion, with each variation
marked off by a repeat sign. The lyrics were written beneath the staff, in
English, Telugu, and Tamil characters respectively, and were repeated with
each variation. In general, Chinnaswamy Mudaliar eschewed symbols for
gamakas, instead writing them out note by note.
What kind of order did such a representation produce? First, it made
the compositions clearly visible in English; the other languages, placed
underneath the English characters, appeared secondary. Meanwhile, al-
though groupings of measures were marked out by repeat signs, the con-
tinuous progression of the music across the page, as well as the injunction
to count by individual notes (or ‘‘quavers’’), suggested that the measure,
not the tala cycle, was the main unit. The tala, as an organizing prin-
ciple, was thus effectively made invisible. Whereas a musician using sargam
notation would most likely arrange the notation by having one tala cycle
per line, or one line of the composition’s lyrics per line, the run-on quality
of Chinnaswamy Mudaliar’s notation gave the impression that the music
followed not the structure of the tala or the lyrics but the staff itself. In-
deed, whereas a Karnatic musician might end each section of the com-
position with an improvised flourish—something that would be hard to
capture in any form of notation—Chinnaswamy Mudaliar neatly resolved
the end of the pallavi to the tonic, C, providing precisely the kind of end-
ing that those accustomed to looking at staff notation would find under-
standable.
Timothy Mitchell has discussed this idea of a homogenous, empty
structure that orders space and time as an essential element of the colonial
gaze and the modern production of knowledge. In such a gaze, which he
characterizes as ‘‘enframing,’’ the appearance of order depends on the illu-
sion of a structure apart from the things themselves, the division of space
and time into exact and precisely repeating units that seem to exist in-
dependently of what they contain (Mitchell 1988, 85–86). Chinnaswamy
Mudaliar’s notations, with their profusion of notes placed in the uniform
spaces provided by the staff, gave the impression of order, of exactness and
completeness. Minimizing the presence of the tala as an organizing prin-
ciple, he effectively substituted a disembodied unit of time, the repeating
measure, with each measure being the same as the one coming before and
after it. While one might argue that tala, as a repeating cycle of defined

a writing lesson 211

units of time, provides just this sort of enframing structure, there is a cru-
cial difference that emerges in performance. While tala might look like
an abstracted, empty structure on paper, in practice it is deeply embodied.
Musicians ‘‘keep tala’’ using a variety of finger counts and claps whose pur-
pose is ostensibly to make the tala (or where one is in the tala cycle) visible
to others; yet the finger counts and claps are also, and more primarily, ways
of embodying or feeling where one is in the cycle. Musicians learn to as-
sociate the feel of a finger count or clap with a particular place in the cycle
and thus with a set of musical possibilities that can begin at that point.
Far from being an enframing structure apart from its contents, then, tala,
when embodied in performance, intimately connects form with content.
Meanwhile, the very idea of ‘‘notes,’’ as implied by the staff notation,
was considerably different from the Indian term swara, which conveys not
so much a note as a kind of placeholder that might in actuality include
several notes.23 For instance, when a musician sees the swara pa, he does
not just sing the fifth note of the scale but a combination of several notes
which focus on or approach pa, using his knowledge of the raga. In other
words, a musician using sargam notation employs it as a clue, to jog the
memory or to inspire the singing of a spontaneous variation; the nota-
tion is not regarded as a sufficient record in itself. Chinnaswamy Muda-
liar’s notations, by contrast, placed themselves in quite a different rela-
tion to the musician’s memory. In his vision, the notes arranged on the
staff could replace the musician’s memory; they would represent the en-
tire composition, leaving, as he often emphasized, ‘‘no room for doubt or
hesitation.’’
The liberation of the musician from the limits of his memory, in Chin-
naswamy Mudaliar’s logic, would produce not only musical but also social
progress. In a long footnote to his explanation of the ‘‘Peculiarities of Ori-
ental Music,’’ he railed against the secrecy and competition among Indian
musicians, and the money-mongering of gurus: ‘‘In India it is with the
greatest difficulty that a professional musician is ever induced to impart
to others the music he had learnt. . . . ‘Teach music to none but your son,
your guru’s son, or to him who gives you wealth incessantly’ is the rule ob-
served by most musicians. . . . In India all the knowledge and proficiency
acquired by each connoisseur is kept a profound secret’’ (1892, 29). If nota-
tion could effect a liberation from such ‘‘gurus,’’ it might also prove a way
for respectable women to learn music. In 1884 A. Govinda Charlu wrote
in the Mysore Herald that with the use of notation ‘‘some of our lady pupils
may become original composers’’ (ibid., 209). The staff notation was ‘‘per-

212 a writing lesson

fected’’ compared to the ‘‘crude and clumsy’’ Indian notation; through it,
‘‘songs could be mastered from the mere books.’’ In this way, respectable
housewives could learn music without having to learn from their social
inferiors. Meanwhile, they could keep their husbands happy and occu-
pied: notation was the ‘‘solution [to] the much talked of Anti-Nautch Girl
question’’ (209).
Although Chinnaswamy Mudaliar’s project folded after several years
due to lack of financial support, his ideas about staff notation were echoed
by others in the early decades of the twentieth century. Among these was
H. P. Krishna Rao, who published his ideas about notation in 1906 in First
Steps in Hindu Music in English Notation.24 ‘‘Hindu music is very ancient, sci-
entific, and interesting,’’ he wrote in his preface. ‘‘But the sad want of a
method for committing musical ideas to writing has left the fine art stag-
nant and unfamiliar. The seven notes, Sa, Ri, Ga, Ma, Pa, Dha, and Ni, do
not, as now written, represent the pitch or time accurately, and thousands
of compositions of great authors are therefore either being lost or muti-
lated. To serve as a means for making permanent records of musical com-
positions, and to enable students to learn them in the absence of a teacher,
this little book has been written’’ (1906, 3). Krishna Rao began by show-
ing how the notes of the Indian scale were to be placed on the staff. The
transposition was metaphorical as well as literal. Once transposed onto the
staff, a raga appeared as ‘‘a melodious combination of particular notes’’ (9),
rather than a set of phrases or motifs employing certain gamakas. More-
over, staff notation introduced elements that were never conceptualized
in Karnatic music, such as rests and dynamics. This is not to say that such
things did not exist in Karnatic music but that they were not considered
a specifiable element of music to be written out and consciously learned.
Krishna Rao’s book demonstrated the way the transposition to staff
notation thus radically reconfigured Karnatic music, even as he insisted on
total translatability from one musical system to the other, and, in the pro-
cess, to the English language. Even as Krishna Rao argued for the transpar-
ency of the notation such that ‘‘a knowledge merely of the English alpha-
bet is enough to enable a student to understand the work,’’ his method
showed how staff notation and its foreign symbols propelled a transla-
tion of Karnatic musical terms into English (1906, 3). For instance, tala was
translated as ‘‘time,’’ avarta as ‘‘measure,’’ and the particular tala of a com-
position was equated with the ‘‘time signature’’ in staff notation. Gamaka
was translated as ‘‘grace,’’ and the individual names for different kinds of
gamakas were given English names like ‘‘slur,’’ ‘‘merge,’’ ‘‘shake,’’ ‘‘trill,’’ and

a writing lesson 213

23 ‘‘Abbreviations and Embellishments.’’ In H. P. Krishna Rao, First Steps in Hindu Music (1906).
Courtesy of Music Division, The New York Public Library for the Performing Arts, Astor, Lenox
and Tilden Foundations.
24 Composition in raga bilahari in sargam and staff notation. In H. P. Krishna Rao, First
Steps in Hindu Music (1906). Courtesy of Music Division, The New York Public Library
for the Performing Arts, Astor, Lenox and Tilden Foundations.
‘‘grace-note.’’ The effect of such translation was that the Karnatic musi-
cal terms seemed to fit seamlessly into the syntax of an English sentence
about music: ‘‘A composition is sung in a Raga, regulated by Tala, and em-
bellished by Gamakas, or graces’’ (38). Such a statement not only assumes
a one-to-one correspondence with the English terms scale, time signature,
and graces but also fits these three elements—raga, tala, and gamakas—
into a hierarchy of importance. For Krishna Rao, the excesses produced
by the process of transposition, it seemed, could be effectively tamed by
translation.

A Picture for the Ear

If in 1874 the argument had been over the question of whether Indian
music could be notated at all, by 1916 the need for notation was taken
completely for granted. The debate now centered around the question of
which notation, the staff or the Indian, was best suited to representing
Karnatic music. Yet this idea of representation had taken on a different
cast. No longer did it refer only to the representation of sounds by written
symbols; it now also implied a representation of Indianness. In statements
made at the first All-India Music Conference in 1916 at Baroda and after,
it was argued that notation was related to music as written language was
to spoken language.
Ten years after he had eloquently illustrated the benefits of European
staff notation, H. P. Krishna Rao appeared at the Baroda conference with
an entirely different message. In a short section of his lecture, entitled
‘‘Notation and Music,’’ he declared that the Indian sargam notation was
superior to staff notation. ‘‘Every kind of language must have its own
notation’’ (1917, 27). His view of language was quite expansive; it covered
‘‘word-language,’’ ‘‘sign-language,’’ and ‘‘sound-language.’’ If the alphabet
was the notation for word-language and painting the notation for sign-
language, musical notation was the notation for sound-language (27).
Indian sargam notation was much more suited to the task of representing
Indian music, he argued, since Indian notation allowed the musician to
see the note name and its pitch simultaneously, which gave it an advantage
over staff notation. It kept notation in the domain of language, instead
of necessitating a detour through visual symbols; an Indian musician’s re-
sponse to sargam notation was, he maintained, as simple and automatic as
a ‘‘reflex.’’ ‘‘Musical instruction begins with singing the notes Sa, Ri, Ga,

216 a writing lesson

Ma, etc. By constant practice a reflex action is established in the brain, by
which the mere remembrance of the letter Sa or Ri takes the voice at once
to its proper pitch, and an Indian singer displays, therefore, a wonderful
capacity for singing songs by means of the names of the notes only . . . and
not the dumb syllables ‘La, la, la,’ as done in the West’’ (28). The note names,
Krishna Rao argued, were motivated by years of practice, until they lit-
erally spoke for themselves, whereas staff notation remained ‘‘dumb.’’ He
illustrated his point with an explication of the ‘‘psychic processes that take
place when we see a note Sa, Ri, or Ga written on paper. The image of
the letter is conveyed to the brain through the optic nerve; by simple as-
sociation its name is ascertained, and the impression is transferred to the
nerve controlling the vocal chords, and then the correct pitch of the note
is sung’’ (28). The psychic processes involved in reading staff notation were
far more convoluted: ‘‘(1) The image of the note is conveyed to the brain,
(2) an enquiry is set up as to the name of the note with reference to the
clef and the key signature, (3) association of the name, (4) reflex action
of the remembrance of the name of the note and its pitch’’ (28). Reading
staff notation thus involved ‘‘an extra psychic feat,’’ because it made the
note names invisible. It was like translating from a foreign language; an
additional step was required to make the staff notation speak. Not only
that, but staff notation constantly, by mere displacement of a note, threat-
ened to become illegible, since ‘‘the same symbol represents seven different
notes.’’ Krishna Rao’s lecture took on the tone of a colonial official dis-
gruntled by the evasiveness of the natives. ‘‘The crotchet is the chameleon
on the hedge. It changes its colour, form, and its name. The staff nota-
tion is therefore seven times as difficult as the Indian notation’’ (28). The
sargam notation, by contrast, was so legible that the viewer hardly had an
impression of reading at all; it was as if the notes spoke to him from the
page. ‘‘The native notation,’’ Krishna Rao concluded, ‘‘is a picture for the
ear; while the staff notation is for the eye’’ (29).25
The idea that Indian notation spoke for itself contrasted with the idea
that staff notation was a transparent, universal medium, capable of rep-
resenting any music. M. S. Ramaswamy Aiyar, a musicologist and super-
intendent of police in Madras, published an impassioned argument for
Indian notation as an appendix to his biography of Thyagaraja, in which
he condemned the ‘‘staff mania’’ of people like A. M. Chinnaswamy Muda-
liar (1927, 185). He began his argument with the same question Chinna-
swamy had raised: ‘‘Can we, who see unmistakable signs of progress in all

a writing lesson 217

other directions, suffer ourselves to be blindfolded in the matter of pre-
serving music for the ages?’’ (182). Yet for Ramaswamy Aiyar, the danger
of losing the music was equaled by the danger of becoming too Western-
ized. According to him, the way to ‘‘resuscitate our fallen music’’ was to use
sargam notation. There was, he wrote, a direct fit between Indian music
and Indian notation, much like the relationship between a language and
its alphabet (179). To illustrate his point, he refuted the argument made
in 1921 by ‘‘an educated Indian lady with University honors.’’ In making
a case for staff notation, she had written, ‘‘With staff notation, our music
will be studied and appreciated by the Americans, the English, etc; and
there is the chance of Indian music becoming universal and popular and
still Indian. If we wish to be recognised as a nation, we must make others
see the greatness and the superiority of all that we possess. How did our
great religion find its way to the United States of America? It was through
the common medium—English’’ (189). Such a logic, Ramaswamy Aiyar
maintained, was akin to asking that Indians forget their own languages
and only speak in English. ‘‘True,’’ he wrote, ‘‘Swami Vivekananda em-
ployed English in the United States of America to assert the superiority
of Indian religion. But did he ever ask the Indians to forget their own
vernaculars in favour of English?’’ (191–92). Adopting staff notation for
Indian music had already been attempted by A. M. Chinnaswamy Muda-
liar, Ramaswamy Aiyar remarked, but it had not been very popular. ‘‘For
aught I know, the Europeans discarded it because there was Indian music
in it, and the Indians equally discarded it, because there was the staff in it’’
(192). The point was that ‘‘different races possess different auditory faculties
and hence different systems of music came rightly into existence’’ (192).
Notation may have been only the outward sign of this difference, but it had
the capacity, like the words of a mother tongue, to travel inward through
the ear, to activate the voice. Ramaswamy Aiyar brought his point home
by locating the notation question within the metaphor that dominated the
language politics of Madras at the time: ‘‘Inasmuch as the mother’s milk
of the Indian notation is plentiful for the Indians, why should a foreign
doctor hoarsely cry and unduly praise to the skies the unnecessary Mellin’s
food of the staff notation?’’ (193).26
For Ramaswamy Aiyar, the effectiveness of Indian sargam notation lay
in the fact that, whereas staff notation was a ‘‘visual’’ method, sargam nota-
tion was ‘‘visuo-aural,’’ appealing not only to the eyes but also to the ears
(1927, 186). Sargam notation had the power to effect a peculiar simulta-
neity of sight and sound: ‘‘If, with a [raga] given, an Indian note Ga is

218 a writing lesson

written on a piece of paper, the ear—as soon as the eye is directed to the
note—rings within itself the sound peculiar to Ga. But if a European cro-
chet is written, you cannot at once give it its proper sound. . . . Some more
ceremony has to be performed on it’’ (196). Such statements about the
conversion of written notation into sound reveal a supreme confidence
in the power of notation and the necessity for it. Notation was not only
deemed superior to memory; it was now also absolutely necessary in order
to be able to hear and understand music. Ramaswamy Aiyar illustrated the
superiority of sargam notation with a story from real life.

‘‘But you have no such thing as Sargam or any Indian notation at

all,’’ may retort the puffed advocates of the staff. So indeed was Mr R (a
Hindu musician) taunted some ten years back at Baroda by Mr F (a Por-
tugese musician); and their further conversation which actually took
place may be of some interest to the reader.
Mr R: Is not the object of notation to preserve a song, and if need
be, to reproduce it?
Mr F: I should think so.
Mr R: Suppose I preserve your song by recording it in my notebook
and reproduce it whenever required; will you then grant that we do
have our own notation?
Mr F: Surely.
Forthwith Mr F sang a snatch and Mr R reduced it to his sargam
notation and even reproduced it. But Mr F would not be satisfied and
thought that Mr R wrote in his notebook some nonsense but correctly
reproduced the song by the strength of his memory. They therefore
parted for the day but met again the next morning. At once Mr F took
Mr R to a lonely place and challenged him to reproduce, if he could,
the song sung the day before. To Mr F’s utter disappointment, Mr R re-
produced the song admirably enough. The table was now turned. Mr R
challenged Mr F thus:
‘‘Now, sir, I have reproduced your song and thus proved that we do
have our own Indian notation. I shall sing for you a Hindu air and let
me see how and when you will reproduce it.’’
So saying, Mr R sang a well-known kriti of Tyagaraja’s in Bhairavi.
Mr F trembled before it, just as Arjuna did before Lord Krishna’s Vis-
waroopa, and confessed:
‘‘O! It is all Greek to me. I cannot in the first place conceive your
song, much less can I reduce it to my notation.’’ (203–4)

a writing lesson 219

Inexhaustible Ambrosia
If notation had emerged, by the 1920s, as necessary for the proper under-
standing of music, it was based on the idea that memory alone was no
longer equal to the task of perpetuating a classical tradition. One thing
that advocates of the staff notation as well as the sargam notation could
agree on (even the likes of Mr F and Mr R) was that it was not memory
but the ability to write and read notation, and to traffic easily between the
oral and the written, that made the musician. This conviction explains the
efflorescence of music books, song collections and manuals, that began to
appear at the turn of the twentieth century.
In 1895, his eyesight suffering from years of notating and casting the
type for his Oriental Music in European Notation, A. M. Chinnaswamy Muda-
liar began a correspondence with Subbarama Dikshitar, the grand-nephew
of the composer Muthuswamy Dikshitar. Subbarama Dikshitar was then
serving as the asthana vidwan (court musician) at the court of Ettaya-
puram, a small princely state south of Madurai known for its patronage
of the arts. Between 1895 and 1899, Subbarama Dikshitar made several
long trips to Madras to stay with Chinnaswamy, teaching him the com-
positions of Muthuswamy Dikshitar and confirming the correctness of
Chinnaswamy Mudaliar’s staff notation (Raghavan 1961, vii). In 1899, how-
ever, literally blinded by his love of staff notation, Chinnaswamy Mudaliar
found himself unable to carry out the printing and publication of these
works. He therefore made a trip to Ettayapuram and appeared before the
maharaja himself to convince him that the samasthanam should take up
the task of publishing, if only in the old Telugu notation, Subbarama Dik-
shitar’s entire repertoire. Chinnaswamy Mudaliar had appealed to Subba-
rama to put down in writing and notation everything that he knew, ‘‘with-
out hiding anything’’ (ibid.). In the English preface to the original version
of the monumental work that resulted, published in 1904, C. Nagojee Rau
wrote that Subbarama, ‘‘though unwilling at first to part with what he
naturally regarded as a precious heirloom to be jealously guarded and re-
tained within his family, yielded in the end to the wishes of his master
and patron, the Rajah. . . . The stores of music literature in his possession
would, in the course of nature, have been lost to the world in a few years if
this work had not been published now’’ (ibid, viii). Indeed, the publishing
of the notation seemed to perform precisely the effect of wresting it from
the hands of death: Chinnaswamy Mudaliar passed away in 1901, as the

220 a writing lesson

printing was getting started, and Subbarama passed away in 1906, a mere
two years after the Sangīta Sampradāya Pradārṣini was published (ibid).
The Sangīta Sampradāya Pradārṣini, in its original Telugu version, pub-
lished in 1904, came to a staggering 1,700 pages. Its compass was decidedly
encyclopedic: not only did it contain notations of 229 Muthuswamy Dik-
shitar kritis, as well as works of other composers, but it also contained
biographies of musicians and authors of musicological treatises, an exhaus-
tive tabular list of ragas with their characteristics described, a descriptive
guide to the gamaka-signs and tala-signs employed in the notation, and
notes on the problem spots in the rendering of the works vocally and in-
strumentally (Raghavan 1961, viii). The work exuded systematization; in-
deed, it was later taken as a model by V. N. Bhatkhande in his calls for
a systematization of Hindustani music (ibid, ix). The notations were ar-
ranged by the seventy-two melakartas, a system of classifying ragas based
on the notes they used. Eleven different symbols were used to convey dif-
ferent types of gamakas, as well as symbols for sharp, natural, and flat signs
to convey pitch and lines above the notes to convey tempo.
The simultaneous emergence of notation and printing technology at
the end of the nineteenth century led to another genre: the music manual
or self-instruction book. If notation was beginning to be seen as a form
of writing which could replace the musician’s memory, it was also seen
as being able to replace the guru, or teacher. At the same time as debates
about notation were emerging, the Taccur Singaracaryulu brothers, well
known in the music world of Madras as teachers, published a series of
graded textbooks on Karnatic music in Telugu. The first of these books,
Gāyakaparijatam, appeared in 1882. In the English preface to a later book
of the series, Swami Vidyananda Paramahamsa recalled how the Taccur
brothers had recreated the musical world of Tanjavur in Madras by hosting
Sunday concerts at their house in George Town, which served as a gather-
ing place for musicians from Madras and elsewhere (Singaracaryulu 1912,
26–27). He claimed that the publication of the Gāyakaparijatam began a
‘‘musical revolution,’’ unlike music books published before it, which were
‘‘miscellaneous compendia’’ of songs. In these previous books, ‘‘the teacher
was absent’’; there was nothing but ‘‘a veneer of abstract notes . . . for the
songs’’ (ibid., 8). The innovation of the Taccur brothers was to provide
notation not only for gitams and varnams, but also for kritis; for ‘‘[pre-
viously] the notes to be applied to these pieces were merely sealed prop-
erty. . . . The rule was always to hear and learn’’ (ibid., 17). Of the Gāyaka

a writing lesson 221

25 Typical page of notation from Subbarama Dikshitar’s Sangīta Sampradāya Pradārṣini
(1904). Sargam notation is in Telugu script, employing symbols for gamakas and sharp,
flat, and natural notes, with words in Telugu under each line of notation.
Siddhanjanam, the fifth book of the series, Swami Paramahamsa wrote that
‘‘it really was the unguent for clearing the eyes of the world of music; the
kirtanas which were mere words now became invested with their respec-
tive accompanying notes’’ (ibid., 18). The Taccur Singaracaryulu brothers
carried out their work despite the outrage of professional musicians, who
apparently felt that their secrets were being betrayed and that they would
lose earnings from having the knowledge so cheaply available through
books. But their objections were ‘‘altogether silenced by the uproar of
the lay public and the relief and joy especially of the self-instructors of
music’’ (ibid., 18). The triumph of these books was that they placed musical
progress in the hands of the student; ‘‘any person,’’ Paramahamsa wrote,
‘‘can become a musician by dint of self-help and perseverance’’ (ibid., 20).
After a beginning course with a music teacher, he maintained, ‘‘the use of
the Siddhanjanam is enough for making a musician, and a finishing touch
can be acquired by a few months’ apprenticeship under some renowned
master’’ (ibid., 18).
How did the Taccur brothers endow their music books with so much
teacherly authority? First, they claimed a direct connection to Thyaga-
raja through the Tanjore court; indeed, the elder ( pedda) Singaracaryulu
was a younger contemporary of Thyagaraja. Second, the idea of a graded
series of books implied that there was a logic or progression to them, a
process of learning that should be the standard for every musician. Rather
than confronting the reader with a miscellaneous collection of songs, the
books dispensed musical knowledge according to degrees of difficulty.27
The introductory book, Svaramanjari, began with an explanation of the
notes of the scale and a table of talas in Karnatic music. It then intro-
duced the sarali varisai, the most elementary exercises, and a number of
gitams, short, easy compositions meant to show the fundamental aspects
of raga, tala, and sahityam (lyrics). The second book, Gāyakaparijatam, pro-
vided twenty more gitas of a more difficult variety, as well as twenty-seven
varnams (a genre akin to an etude which explores the possibilities of a
raga). After the varnams were about eighty notated compositions by vari-
ous composers. The third book, Sangīta Kalānidhi, included the method
of classifying the ragas by the 72-melakarta scheme, a classification and
description of different kinds of gamakas, and the aspects of tala. A sepa-
rate section dealt with the theory of rasa and the essentials of abhinaya
(facial expression) and gestures used in the ‘‘nautch dance.’’ The last part
contained 143 notated kritis as well as other varieties of composition, such

a writing lesson 223

as padams, thillanas, and javalis. The Gāyakalōchana, the fourth book, began
with a discussion of the origins of music (in the Vedas). It then provided an
exhaustive survey of ragas, including the suitable hours for each one, con-
cluding with an alphabetical list. An exhaustive treatment of tala, show-
ing the permutations that led to hundreds of different talas, was followed
by more than 200 notated compositions of all varieties. The fifth book,
Gāyaka Siddhanjanam, contained notation for another 210 compositions, as
well as for about fifteen Hindustani compositions. A long chapter discussed
the method for developing alapana in sixty of the most popular ragas of
the day. The final book, Ganendu Sēkaram, contained lakshana gitams for all
seventy-two of the melakarta ragas and derived ragas, followed by tanams
(improvised patterns of notes sung in semi-free rhythm sometimes fol-
lowing the alapana). In addition to more notated kritis, as well as nota-
tions for English notes, there was a list of the pallavis sung by masters in
the past.
The Singaracaryulu brothers’ books did what no other book since the
ancient treatises on music had done: they provided, in written form, a
discussion of a body of music theory and showed how the theory related
to the practical art of music. Yet unlike the treatises, which were in San-
skrit, these books were in the vernacular; unlike the treatises, which were
available only in palm-leaf manuscripts, these books were printed in mass
quantities and sold for a nominal fee. Moreover, the books provided nota-
tion for many more compositions than a single musician could ever know
or sing in a lifetime. The profusion of notation in the closely printed pages
of the books gave the reader nearly a thousand compositions from which
to choose. The possibility of possessing so many compositions in nota-
tion, compositions which one might never have even heard before, must
have been quite revolutionary in a world where (as I was told by several
older musicians) a musician might have previously had a repertoire of only
twenty songs which he would sing over and over again. A hint of this
change in attitude toward repertoire appears in a review of A. M. Chin-
naswamy Mudaliar’s project in 1894, in which the author suggested that
‘‘twice or thrice’’ the number of kritis should have been included: ‘‘Old
‘Kritees’ are sung over and over again to a tiresome extent, and the ab-
sence of novelty takes away from the very charm of music. Thyagayya’s
store alone is so vast, that to those who wish there is an inexhaustible Am-
brosia that may be drawn out of it’’ (Chinnaswamy Mudaliar 1892, 208).28
With the monumental notating projects of Subbarama Dikshitar and the
Singaracaryulu brothers, the concept of musical repertoire was completely

224 a writing lesson

changed; it now referred to a vast store of musical compositions, each of
which claimed its status as a ‘‘composition’’ precisely by being notated.

Total Translatability
The Singaracaryulu brothers’ books provoked a flurry of publications on
music in Tamil in the early twentieth century. The earliest of these was
D. Narayanaswamy Mudaliar’s Tamil Sungeatha Surabooshany, ‘‘for those
who are learning music.’’ Published in 1900, it contained the usual begin-
ning exercises, sarali varisai and alankarams, as well as notations for gitams
and varnams, a list of the melakarta ragas, and an explanation of tala. The
notation, highly condensed, is divided by note groupings rather than divi-
sions of the tala, while the layout of the book (its small size, the fact that
pages must be turned in the middle of compositions) suggests that it is
meant less for the eye and more as a reference book. On a much grander
scale was T. C. R. Johannes’s Bhārata Sangīta Svāya Bōdhini (Indian music
self-instructor), published in 1912. It included chapters on every aspect of
music, from the theory of rasas and the 72-melakarta scheme to an expla-
nation of the metronome and the intricacies of konnakkol, as well as nota-
tion for numerous compositions (including English notes). As its title and
the layout of the notation indicate, the book is meant to be self-sufficient,
capable of teaching, instead of serving as a mere reference. Each notated
composition includes the murchana, or characteristic ascending and de-
scending pattern of the raga, the sahityam, written above, and uses lines
to indicate tempo and dots to separate the music into phrases. The begin-
ning exercises, like the sarali varisai and janta varisai, were furnished with
Tamil sahityam so that the learner could, in effect, know how the music
was supposed to sound merely by seeing it on the page. That is, the learner
could associate the sound and rhythm of a language he already knew with
music in order to make that music less foreign.
Around the same time, the Tamil doctor and musicologist Abraham
Pandithar was busily composing and notating hundreds of Tamil gitams
and kritis for the benefit of Tamil children who, he felt, needed music in
their mother tongue. These were eventually published in 1934, at Abra-
ham Pandithar’s own printing press in Tanjore, in the form of a primer
entitled A Practical Course in South Indian Music. The book did away with
complicated descriptions of ragas and systems of classification. The page
design was clearly intended to appeal to the eye, presenting the notation
neatly divided into the quantities specified by the tala, with the whole

a writing lesson 225

26 A sample notated composition in raga kedaram from T. C. R. Johannes’s, Bharatha Sangīta
Svāya Bōdhini (1912). Words are in Tamil with sargam notation in Tamil script below.

thing enclosed in a tastefully bordered box. The didactic quality of the

book lay in its carefully ordered presentation: for each composition, first
the swaras only were given, then the same swaras and their corresponding
sahityam, and then the sahityam only. The copyright emblem on the title
page showed two men sitting on a river bank in the manner of guru and
disciple, suggesting that the authority of the book’s notation was guaran-
teed by the authenticity of oral transmission. Ironically, however, Pandi-
thar was a self-taught musician who used the medium of notation not to
record music that already existed but to invent and popularize his own
compositions. In effect, it was only through notation, not through an oral
tradition, that his compositions came into being.
While music manuals in Telugu and Tamil eschewed staff notation,
those written in English in the 1930s included both sargam and staff nota-
tion, as if to appeal to a wider audience. However, the purpose of includ-
ing both types of notation was no longer to debate the merits of one or
the other, as it had been earlier. The implication now seemed to be that
staff and sargam were equally valid ways of viewing the music and that
the music remained the same no matter how one chose to picture it. Here,
one can note a crucial shift: notation, instead of being regarded as cen-
tral to the way music was conceptualized, was demoted to the status of a
‘‘mere’’ representation, incidental to the content that it represented. Nota-

226 a writing lesson

27 A sample notated composition in raga senjurutti from Abraham Pandithar’s A Practical Course
in South Indian Music (1934). The upper half of the page includes just the sargam notation in Tamil
script for the three sections of the composition. The lower half of the page repeats the sargam
notation with the lyrics in Tamil below each line of notation.
 28 Title page
of Abraham
Pandithar’s A
Practical Course in
South Indian Music
(1934).

tions were regarded as interchangeable because they merely represented a

prior, authoritative voice.
C. Ganghadar’s Theory and Practice of Hindu Music and the Vina Tutor
(ca. 1935) illustrates this new role for notation. Ganghadar provided de-
scriptions of the different types of gamakas, as well as the usual begin-
ning exercises, in sargam notation in Tamil, Telugu, Malayalam, and Kan-
nada, with staff notation underneath. The process of getting used to seeing
music written in different forms was part of the education provided by the
manual. As if to bring home the staff notation, a version in English sargam
syllables was provided underneath. The progression from what was pre-
sumably one’s native language, through other languages, to staff notation,
and finally to the syllables transliterated into English gave the impression
of total translatability, the idea that one could move from language to lan-
guage, even from one musical system to another, without losing anything.
This fantasy of total translatability was what motivated P. Samba-
moorthy in his writings on Karnatic music. Prolific in his works, which
ranged from a five-volume history of Karnatic music to a multivolume

228 a writing lesson

29 Sample of notation from C. Ganghadar’s Theory and Practice of Hindu Music (ca. 1935). Sargam
notation is provided in Tamil, Telugu, Malayalam, and Kannada, and English (below the staff
notation).

practical course in music for schools, Sambamoorthy was on the cutting

edge of the new field of music education. In 1961, during his tenure as head
of the Department of Indian Music at Madras University, Sambamoorthy
published a small book entitled Elements of Western Music for Students of
Indian Music. A knowledge of Western classical music and staff notation
was required for students in the Indian Music department and the purpose
of the book was to provide them with this knowledge. In his explanation
of the symbols used in staff notation, he took care to translate each one into
an Indian equivalent, interspersing the text with the visual symbols of staff
notation. The effect was that Karnatic music and Western classical music
appeared as two discrete yet equivalent systems, each capable of translat-
ing the other. Only in a small paragraph did Sambamoorthy mention the
problem of representing gamakas in staff notation, but he concluded that

a writing lesson 229

30 A sample page from P. Sambamoorthy’s Elements of Western Music for Students of Indian
Music (1961).
31 A sample notated composition from P. Sambamoorthy’s Elements of Western Music for Students
of Indian Music (1961).

they could, with the use of additional symbols, be ‘‘rendered faithfully and
according to tradition’’ (1961, 8). The last half of the book consisted of
Karnatic compositions in staff notation, compositions that every student
would know, but stripped of gamakas entirely. The idea seemed to be that
the student would see that the staff notation made sense by being able to
hear the song in her head while she saw it on paper.
The idea of total translatability, for Sambamoorthy, thus did not stop
with the easy transposition from one notation system to the other. It also
characterized his notion of the relationship between the oral/aural and
the written. In his A Practical Course in Karnatic Music for Schools (ca. 1960)
and in a manual written for music teachers in 1966, Sambamoorthy laid
great emphasis not just on notation but on making the boundary between
the oral/aural and the written disappear. He prescribed numerous exer-
cises to this effect, such as sight-singing and musical dictation. Students
should write down phrases and then sing them in a raga, ‘‘provided the
phrases so sung form a hearable passage’’ (1966, 63). ‘‘Musical dictation,’’ he
explained to future teachers, ‘‘helps the students to acquire a keen sense
of hearing’’ (ibid., 64). Meanwhile, learning to write music with facility
would ‘‘greatly help the student in the art of musical composition later in
life’’ (ibid.). After students had become proficient enough at converting

a writing lesson 231

heard music into notation and vice versa, there were sample exam ques-
tions aimed at a more virtuosic level: a notated composition, without in-
formation about the raga, was provided, and the student, merely by study-
ing the notation, had to indicate what the raga of the composition was
(Sambamoorthy n.d.a, vol. 3, 207). For Sambamoorthy and his students,
the skill in reading notation was the ability to take something written and
convert it into a piece of music believable to the ear, to make it conform
to a voice that one had already heard before.
Sambamoorthy’s ideal of easy translation between the oral/aural and
the written was precisely that: an ideal. In prescribing so much practice
for it, he effectively acknowledged that, contrary to the simple straight-
forwardness of his directions, going from the oral/aural to the written and
back again was not a simple, mechanical process but a complex maneu-
ver. What his exercises suggest is that one does not simply progress toward
greater accuracy but rather practices the skill of convincing oneself of the
equivalence between the oral/aural and the written. That this requires a
suspension of disbelief not necessarily shared by musicians from older gen-
erations is demonstrated by my own experience of learning compositions.
In long afternoons with my teacher, much of the time I spent ‘‘learning’’
a new composition was actually devoted to creating a notation for that
composition. As she sang, I would scribble a first impression in sargam
notation and then sing my version back to her, revising my written in-
terpretation until we arrived at a notation that both of us agreed on. Yet
the agreement was always only a temporary truce; I would come back the
next day to find her singing the composition in a way that contradicted the
notation we had agreed on the previous day. When I pointed out the in-
consistencies, she would often become irritated, remarking that the same
notation ‘‘said’’ her version as well as mine. Our disagreement about what
constituted acceptable variation in the interpretation of written notation
marked a contrast between two ways of viewing notation. While I was
attempting to create a notation that would act as an authoritative text
(an idea ingrained in me by years of training in Western classical music),
she used the notation as more of a trigger for memory. Like a palm-leaf
manuscript, notation in this second sense is not meant to be sight-read but
studied and then interpreted. This way of using notation acknowledges
the impossibility of total interchangeability between the oral/aural and
the written; it considers the gap between oral/aural and written as produc-
tive rather than problematic. In admitting that notation is not perfectly
legible, it acknowledges its profoundly mediating role.

232 a writing lesson

The Order of Things
Yes, I think our music will, in the very near future, become something quite universal
and embrace everything . . . , the system becoming as elastic and world-wide as the
British Empire itself.
—A. M. Chinnaswamy Mudaliar, ‘‘Saraswatia Redux’’ (1893)

As notation gained prominence in the twentieth century, the project of

determining a basic structure that would serve as the ‘‘theory’’ for Karnatic
music became important.29 For if notation literally raised the problem of
determining the ‘‘basic structure’’ of a piece as opposed to the ‘‘embellish-
ments,’’ it also raised the problem of finding structure in the profusion of
ragas and talas now available in written form. Thus, at the same time as
music manuals with notation appeared, there also appeared books on the
structure and classification of ragas, as well as raga dictionaries: books that
would enable a student not only to sing a raga but also to find its place
in the order of things. Notably, this order did not consist of aural memo-
ries, typical phrases, or associations with a particular raga, but rather in a
hypothetical table that made the scalar structure of ragas visible.
At the heart of this project of classification was the 72-melakarta raga
scheme, originally devised in 1660 by Venkatamakhi, a Telugu Brahmin
in the court of Vijayaraghava Nayaka (r. 1633–1673) at the king’s behest.30
Although a translation of the original work was not published until the
twentieth century, almost every book of Karnatic music notation, in-
cluding Chinnaswamy Mudaliar’s work, the books of the Singaracaryulu
brothers, and the Sangīta Sampradāya Pradārṣini, mentioned the scheme at
great length. In his introduction to a small book on the subject, Samba-
moorthy called the 72-melakarta scheme a ‘‘modern classification of ragas’’
that, ‘‘based on the genus-species system . . . is the best system of raga-
classification that human genius can conceive of ’’ (1961, iii). In the system,
ragas were classified on the basis of only the notes or scale that they used.
By means of different combinations of the twelve pitches in the Western
chromatic scale, 72 melas—‘‘parent scales’’ or ‘‘root-ragas’’—were specified.
The first and second of these seventy-two melas were differentiated only
by one note, the second and third by one note, and so on, keeping sa, ma,
and pa (the first, fourth, and fifth degrees of the scale) constant, until all
the possibilities of combining the notes were exhausted. After the first
thirty-six melas were arrived at in this manner, the ma was raised and the
second thirty-six melas were produced.31 The exhaustiveness of the table

a writing lesson 233

of ragas thus produced, and its patterned regularity, were signs of its sci-
entific validity.
The idea of a system or table that could capture the possibilities of music
for all time seems to have thrilled Venkatamakhi as well as his twentieth-
century counterpart. ‘‘So great was Venkatamakhi’s joy when he formu-
lated the scheme,’’ Sambamoorthy wrote, ‘‘that he declared in his immortal
work that even Lord Parmasiva could not devise a scheme containing one
more or one less than the 72 melakartas’’ (1961, 10). Most of these melakarta
ragas remained in the realm of pure theory; only nineteen of them did
Venkatamakhi find to be in use (10–11). The point, however, was that any
raga, past or future, could now find a place in this universal table of scales.
‘‘The great use of the scheme,’’ wrote Sambamoorthy, ‘‘consists in the fact
that once the number of a melakarta raga is known, its lakshana (i.e., the
characteristics of its swaras) can be told at the next second’’ (13). He dem-
onstrated how the system worked: by locating a raga within its chakra
(group of six ragas) and then locating its chakra within the symmetri-
cal spatial organization of the table (13–14).32 The 72-melakarta scheme,
according to Sambamoorthy, was what made Karnatic music adhere as a
modern system. ‘‘Viewed in the light of mere permutations and combina-
tions, the scheme might appear at first sight as an artificial and dry process.
But . . . every musical sound and interval has its exact number of vibrations
and ratios. The melakarta scheme is highly comprehensive and systematic,
and includes within its fold all the modes used in ancient as well as modern
systems of music of the world. . . . It is a complete and exhaustive scheme
evolved by simple and natural combinations’’ (17).33
The 72-melakarta scheme represented a kind of natural history of
music, that is, a principle of order that encompassed history by acting as
the be-all and end-all of musical possibility. To musicologists in twentieth-
century South India, the idea of such an order was appealing precisely be-
cause it kept change and its more political counterpart, history, in check.
Meanwhile, it gestured toward the kind of universality that classical music
needed, by showing how scales or modes similar to the idea of scale in
Western classical music were operative in Karnatic music; it became evi-
dence that the two ‘‘systems’’ of music were ‘‘of the same family.’’ 34 And
finally, it provided an order based on underlying structure rather than on
gamakas or phrases or anything else that had come to be regarded as non-
essential.
This shift in the way ragas came to be classified is apparent in music
manuals in the early twentieth century. A Tamil music theory book from

234 a writing lesson

1902, Caṅkīta Cantirikai, provided several methods of classifying ragas, of
which the 72-melakarta scheme was only one. The groupings seemed to
depend less on underlying structure and more on literary or social conven-
tion. One grouping placed ragas together by the suffixes in their names,
another by the sentiments they produced, another by the times of day and
night for which they were suited (Manikka Mudaliar 1902, 116–29). By
the time of the publication of Chitti Babu Naidu’s A Key to Hindu Music
in 1925, however, the 72-melakarta scheme seems to have taken first place
among other modes of classifying ragas. Chitti Babu Naidu commented on
the efforts to classify ragas ‘‘according to one’s own experience,’’ remark-
ing that such efforts, which mistook the musician’s individual experience
for objective truth, were ‘‘all evidently the feeble attempts made by people
who do not understand the fine system of Indian ragas and their gene-
sis’’ (1925, 9). A distinction had to be made, Naidu implied, between this
system of genesis and more subjective bases of classification.
Just what does such a system exclude? As a survey of these manuals
shows, coexisting with the official musicological system of classification
are other ways of classifying ragas that have dropped out of musicological
discourse. These modes of raga classification show the difference between
the concept of a raga and the concept of a musical scale in Western terms.
Ragas exist somewhere along a continuum between scales and melodies;
they are often characterized by particular phrases and orders of ascent and
descent through the scale they use. Unlike a scale, a raga can be thought of
as a set of potentialities, realizable only through temporal development;
one cannot write out a raga, in this more complete sense, on a page. Not
only particular phrases but pauses and repetitions are essential to the elabo-
ration of a raga. Developed in performance, ragas gain associations with
particular places, musicians, and feelings. For instance, one musician ex-
plained to me that two ragas like kamboji and yadukulakamboji, which
are in the same mela, or family, according to Venkatamakhi’s system and
differ only by the presence or absence of a single swara, nevertheless dif-
fer in the cuvai (taste) and uṇarcci (feeling) they evoke. Kamboji is as-
sociated with kampīram (grandeur, pomp) while yadukulakamboji, lack-
ing the swara ga, evokes tuyaram (sadness) and maāṭutal (pleading).35 The
metaphors of orality and embodiment at work in these metaphors of cuvai
and uṇarcci contrast greatly with the decidedly visual order of the 72-
melakarta system, which strips ragas of their evocative contexts and places
them in a chart arranged to be understandable to the eye.
In the mid-twentieth century, the 72-melakarta system assumed a

a writing lesson 235

prominent place in the curricula of university music departments. As it
gained the status of ‘‘theory,’’ it came to be seen as essential knowledge for
the proper rendition of any raga, replacing other ways of ordering ragas
with a mode of classification deemed more properly musicological. Start-
ing in the 1940s, the matter of the derivation of janya ragas from their ‘‘par-
ent’’ or melakarta ragas became a prime topic of debate year after year in
the Madras Music Academy. There it was argued that knowing the parent
raga affected one’s elaboration of the janya raga; it determined the notes
one paused on, how one constructed phrases, which notes were consid-
ered ‘‘foreign,’’ and so on. The derivation of a raga according to the 72-
melakarta system was often cited as an authoritative answer for questions
of ‘‘correct’’ or ‘‘incorrect’’ usage (Sambamoorthy 1961, 26). In this view,
if only musicians observed the rules, theory and practice would become a
single rational system.
The idea of a standardized scheme of ordering ragas was deeply con-
nected to the use of notation and the concept of the composer. K. Rama-
chandran’s Sri Dakṣinarāga Ratnākaram (Characteristics of Carnatic ragas),
published in 1949, provided its readers with the classification of 1,044 ragas
according to their melakarta group, the number of notes each one used in
its ascent and descent, and a list of compositions in each raga. The idea was
that the proper approach to a raga was determinable not only by finding
its location in an ordered table of ragas but also by observing its behavior
in different compositions, analogous to a naturalist’s observation of ani-
mal behavior. The natural history of ragas produced by the 72-melakarta
system was thus conceptually related to the observation of raga behavior.
In this type of musicology, the composition was endowed with a new type
of authority; it was now deemed absolutely necessary to know compo-
sitions in a particular raga before one dared to claim any knowledge of
it. Compositions became the locus of authority in a musicology that now
thought of raga as something able to be almost completely described and
rationalized.

In the Realm of Musicology

In 1932 the Annamalai University published an English translation of the
Svaramēlakalānidhi, a sixteenth-century music treatise by Ramamatya, with
commentary by M. S. Ramaswamy Aiyar. The twentieth-century musi-
cologist saw an analogy between his own endeavor, reconciling music
theory with practice, and the Vijayanagar king Rama Raja’s injunction to

236 a writing lesson

his court musicologist in 1550: ‘‘The science of music has, both in theory
and in practice, degenerated into conflicting views. . . . Reconcile all [the
conflicting views] and write a new science . . . embodying therein [music’s]
theory and practice’’ (Ramaswamy Aiyar 1932, xv). Reviewing several other
treatises, Ramaswamy Aiyar remarked that a similar note of complaint had
been sounded by other treatise-writers. ‘‘A question, therefore, naturally
arises, namely, ‘Why should the musical views conflict, at all, with one
another?’ To put the same question sarcastically: ‘Why should the har-
monious music produce disharmony amongst its votaries?’ ’’ (xvi). Rama-
swamy Aiyar’s answer to this question, which occupied the seventy-three
pages of his introduction, overshadowed his translation of the treatise
itself. The excessive length of the introduction hints at the disjunction be-
tween the kind of answers Ramaswamy Aiyar expected in the twentieth
century and those that Ramamatya had provided.
Ramaswamy Aiyar began by differentiating the interests of musicology
from Ramamatya’s apparent interests. Ramamatya had begun his trea-
tise by giving the pedigree of the king who commissioned it, a chart
showing the descent of Rama Raja from none other than Vishnu and
Brahma. Ramaswamy Aiyar contrasted this with a genealogy by the colo-
nial historian Robert Sewell, taken from epigraphical records, which
‘‘gives Rama Raja his proper place in the royal line of Vijayanagar’’ (1932,
xiii). Placing his own musicological endeavor on the side of written
records and ‘‘history,’’ Ramaswamy Aiyar remarked that ‘‘Ramamatya’s de-
scription of Rama Raja’s pedigree is more fanciful than real and betrays
the mentality of a flattering court poet rather than that of a responsible
State officer’’ (xiii). Moreover, whereas for Ramamatya the pedigree of the
king was an essential part of the treatise, for Ramaswamy Aiyar it was a
mere distraction from the real matter of the work: ‘‘I digressed . . . to warn
the reader against blindly accepting unhistorical, and therefore untruthful,
statements’’ (xiv). The twentieth-century musicological treatise belonged
in the genre of historical and rational writing and not poetry.
The main task at hand, as Ramaswamy Aiyar characterized it, was to
understand the process of musical change: to understand how an ‘‘old
order’’ of music could give rise to a new one. Understanding this was tan-
tamount to reconciling theory and practice; one had to look for a principle
that governed these shifts (1932, xvi). Ramamatya’s answer to his king’s
request that he reconcile musical theory and practice had been to sug-
gest the principle of lakshya, the idea that practice could change theory.
Ramaswamy Aiyar’s translation of his words on the subject went thus:

a writing lesson 237

‘‘The Gandharva-music is ever employed in conformity with the (inflex-
ible) rules of its theory. But if the violation of any of these rules . . . do not
lead to any absurdity; and if, again, the contravention of any of the rules
of practice does not give pleasure, but jars to the ear; then the practice
of music shall be preferred to its theory’’ (xix). Here, theory and practice
were loosely related; they were under no obligation to reflect each other.
Indeed, the gap between them was acknowledged. Music theory, in these
older treatises, was an intellectual discipline in itself which may have had
closer ties to philosophy and poetry than to actual musical practice itself.
Yet such a state of affairs seemed vaguely unsatisfactory to Ramaswamy
Aiyar. For him, as for other twentieth century musicologists, musicology
was a science which had no place for philosophical or poetic discourse.
He claimed that music could be called a science by virtue of the fact that
there were two types of science: ‘‘exact’’ and ‘‘inexact’’ (1932, xvi). Whereas
the rules of exact science (like chemistry, he suggested) were unchang-
ing, the rules of an inexact science could change from time to time. The
slowness of this change was what gave music its scientific status: ‘‘Mark! I
said ‘from time to time’ and not from day to day. For if music, like dream,
changes from day to day, surely, like dream, it will be labelled as mere
phantasy and treated as such’’ (xvi). For music to be a properly histori-
cal subject, it had to outlast the impermanence of fantasy; it had to be
able to be written down and pass from generation to generation. Essential
to this writeability was a systematic, scientific approach, a standard that
itself never changed (xxx). Ramaswamy Aiyar imagined this standard to be
locatable in underlying ‘‘characteristics,’’ which had to be discerned by the
modern musicologist. To illustrate what he meant, he contrasted Rama-
matya’s description of ragas with Venkatamakhi’s 72-mela system. Rama-
matya had merely described ‘‘the ragas in vogue at his time’’; he specified
the names of twenty melas but gave no principle on which they were based
or ordered. ‘‘Evidently he did not care to deduce his twenty melas from
any kind of principles,’’ Ramaswamy Aiyar remarked (xxxv). Yet Rama-
matya’s very use of the concept of mela showed that he was at least work-
ing toward some principle of classification. Such a system of classification
distinguished modern musicology from an ‘‘antiquated’’ approach.

Ramamatya rightly discarded the antic and antiquated method of deriv-

ing Ragas from the complicated system of Grama-Moorchana-Jati, as
well as the later puerile method of bringing them under the fanciful
system of Raga-Ragini-Putra. . . . He had the genius to discover unity

238 a writing lesson

 in variety, that is, a unifying principle in the variety of Ragas that came
under his notice; and he therefore felt that the old cataloguing method
of enumerating the Ragas must give way to the classifying method of re-
ducing them into what might be called the Genus-Species system. (lx)

With such a taxonomy at his fingertips, the modern musicologist, accord-

ing to the demands of the ‘‘present revival of musical taste, in India, on
rational lines,’’ could begin to elucidate ‘‘the various Terms occurring in
the Science of Music’’ (Ramaswamy Aiyar 1932, lxiii). Rather than being
misled by ‘‘mere names,’’ the modern musicologist could discover the real
meaning lying underneath. As a ‘‘test case,’’ Ramaswamy Aiyar took the
perennially confusing terms mārga and dēsi. Such terms, he stated, had been
used so variably and contradictorily to describe ragas that they had lost
all meaning and become mere names (lxiii). He quoted Sarangadeva, au-
thor of the sixteenth century treatise Sangīta Ratnākara: ‘‘Music is of two
kinds—Marga and Desi. That was called Marga which was sought after by
Brahma and other gods and practised by Bharata and other sages in the
presence of Siva and which would yield everlasting prosperity. That kind
was called Desi which consisted of the vocal-instrumental-dance music
and which pleased the people of different countries according to their dif-
ferent tastes’’ (lxiii). This long, ‘‘tiresome’’ description of the terms could
not satisfy the questions of the modern musicologist, as Ramaswamy Aiyar
stated them. ‘‘What were the essential characteristics of Marga and Desi?
In what way could they be unfailingly recognised as such?’’ (lxiv). The
modern musicologist’s approach to such a problem was to trace the his-
tory of these terms and to identify them with particular musical practices:
mārga referred to the practice of chanting the Vedas, while dēsi referred to
the Hindustani and Karnatic practices. Thus, the ‘‘essential characteristic’’
of each term was now determinable: mārga music was confined to four
notes, while dēsi music made use of all the notes of the scale (lxix). Each
music had its own principle: ‘‘The marga-music followed the principle of
lakshana [grammar] and therefore became less and less pleasant, till at last
it degenerated into a dry, monotonous, and sing-song style of singing,
while desi-music followed the principle of lakshya and has therefore ac-
quired a more and more fascinating style of singing’’ (lxxii). Notably, while
Ramamatya discussed mārga and dēsi in terms of their contexts of audi-
tion, Ramaswamy Aiyar searched for the essential characteristics, separable
from such contextual factors, that could be used to define them. Finally,
he gave his own English gloss to these terms. ‘‘The word marga has come

a writing lesson 239

to mean whatever is old and out-of-date; while the word desi has come to
mean whatever is new and up-to-date. . . . Just as I called Marga, Vedic Music,
so, I shall call Desi, Modern Music’’ (lxxiii).
For modern music a modern musicology was necessary. In 1939 another
attempt to reconcile theory and practice came in the form of C. Subrah-
manya Ayyar’s Grammar of South Indian Music. Instead of revising or cri-
tiquing the ancient treatises, he disregarded them entirely. ‘‘The present
thesis,’’ he wrote, ‘‘purposely does not refer to the more ancient theoretical
works on music in Sanskrit. It is based entirely on my musical experience
with a little knowledge of modern Physics, and of musical comparisons
suggested by a musical ear. . . . I feel the paramount necessity for the correct
perception of microtones by all artists, vocalists, and instrumentalists alike,
for their clear exposition of the Raga Bhava [emotion]’’ (1939, 1). The idea
was that music could be best understood through the lens of a scientific
musicology that had no recourse to ancient texts but that relied instead on
the human ear and on modern measuring instruments like tuning forks.
Subrahmanya Ayyar’s project was to understand the ‘‘microtonal changes’’
used in gamakas that made them so effective (37). The point was to under-
stand—to be able to express in words and numbers—‘‘what the voice actu-
ally does’’ (48). How did the same note in two different ragas give a differ-
ent impression? How did a listener differentiate between two ragas when,
strictly speaking, they used the same notes? For Subrahmanya Ayyar, these
were the kinds of questions modern musicology had to answer. Such a
musicology, he implied, would close the gap between the voice/ear and
notation; the minute numerical ratios and lengthy descriptions of ragas
that he provided would effect a true reconciliation of theory and practice.

The Problem of Writing

The author is the principle of thrift in the proliferation of meaning.
—Michel Foucault, ‘‘What Is an Author?’’

To close the gap between theory and practice, for a modern musicologist
like C. Subrahmanya Ayyar, writing itself had to disappear, in a very lit-
eral sense. If, for Ramaswamy Aiyar, the ancient treatises represented a
mass of writing that had to be translated somehow into good sense, for
Subrahmanya Ayyar they were an impediment to the proper practice of
musicological science. He instead aimed to convey his theory through the
eloquence of numbers and their aural counterparts; the true musicology

240 a writing lesson

was a science that, unswayed by the vicissitudes of writing, could deter-
mine ‘‘what the voice actually does.’’
Notation was important in this respect because, ideally, it aimed for the
literal disappearance of writing; notation was supposed to serve as a set of
signs for music that would somehow escape the ambiguity of writing, its
potential to give rise to multiple interpretations. This was why advocates
of the staff notation thought it necessary to have a nonalphabetic notation;
the symbols of the staff notation were said to enjoy a closer relationship
to the eye precisely because they supposedly bypassed language. For pro-
ponents of sargam notation, the disappearance of writing was achieved
by the way it provided a picture for the ear, made of signs so legible that
they became audible, disappearing off the page to sound the voice of the
composer.
The disappearance of writing thus came to convey a certain sense of
authenticity: the notion of a true composer or author behind the writ-
ten signs. The composer became, in the twentieth century, the one figure
who was endowed with the authority to move from the oral to the writ-
ten. The concept of an oral tradition implied an absence of writing that
kept musicians in the realm of memory and improvisation. The composer,
however, by creating fixed compositions, transferred music to the realm
of history. In the politics of music in twentieth-century South India, it
is these latter terms—composer and history—that are the privileged signs of
Karnatic music’s classical status. As compositions have become authored,
they have become repositories of authority, almost like a dictionary or
guidebook of acceptable phrases for those who are improvising.
The desire to ‘‘fix’’ compositions in notation at the turn of the twenti-
eth century was not only about fixing the composer’s voice; it was also part
of the project of showing that Karnatic music obeyed a system of conven-
tions and rules, that there was a structure beneath, or within, the music.
Chinnaswamy Mudaliar was confident that notation would show the dif-
ference between Karnatic music’s structure and its embellishments, the
reason behind the music. Meanwhile, musicologists used the 72-melakarta
raga scheme to show, similarly, a separate conceptual realm in which the
rules of music were fixed for all time. This conceptual realm or structure
came to be known as ‘‘theory,’’ a realm which stood in the same relation
to practice as structure did to embellishments, or as the written did to
the oral. Inhabiting the conceptual realm, the 72-melakarta system stood
apart from the music as a representation of its pure structure. In this order
of things, notation was thought to represent an authoritative version of

a writing lesson 241

a composition, a version that could stand apart from various future ren-
derings. Only by losing its arbitrariness, by appearing to be completely
motivated by the voice of the composer, could notation be properly au-
thoritative.
The scandal of the Swati Tirunal case was that it made the problem and
power of writing reappear. Where the authoritative voice of the composer
should have expelled any doubts, Balachander’s allegations pointed to a
proliferation of possibilities, of questions concerning motivation. For, in
Balachander’s argument, the notation of Swati Tirunal’s compositions, as
well as the historical work of musicologists concerning his life, became
not efforts to determine the truth but problems of writing. Motivation
here referred not to a seamless relationship between sign and meaning
but to the proliferation of ulterior motives that, once revealed, threat-
ened the legitimacy of what was written. Like the anthropologist Claude
Lévi-Strauss among the Nambikwara, Balachander wondered whether the
Swati Tirunal authorities were motivated by a true composer or if they
had simply realized the power of writing, of notating compositions and
publishing them; the distinction between artistic motivation and politi-
cal motivation threatened to become indecipherable in writing. Only
the composer had the authority to control the movement of music be-
tween the realms of oral tradition and writing, a movement that otherwise
threatened to operate in a considerably less authorized manner.
It is the suspension of these possibilities, the disavowal of the problem
of writing, that makes the ‘‘composer’’ and the ‘‘composition’’ possible, at
the same time as it allows the persistence of the idea that Karnatic music
(or Indian music in general) is, despite its notations and treatises, essen-
tially an oral tradition. There is a contradictory logic at work here; the
anxiety that the voice will be lost if not captured by writing coexists with
the anxiety that the voice could be lost precisely by being completely cap-
tured by writing. The way in which these anxieties are intertwined sug-
gests that it has become impossible to imagine the voice, or oral tradition,
without writing. It is only from within the scriptural economy, as Michel
de Certeau might suggest, that certain notions of and values attached to
orality can emerge. It is writing that lends authority, in the literal and
metaphorical senses of the word, to the idea of an oral tradition.
Importantly, it is not just any writing that is at stake here, but English
writing in particular. Writing metaphors associated with English—‘‘punc-
tuation,’’ ‘‘paragraphs,’’ ‘‘essays’’—are now commonly used to describe In-
dian music. For example, in 2000 the musicologist Raghava Menon said of

242 a writing lesson

the late Alla Rakha’s tabla playing that ‘‘there was a sense of commas, expla-
nations, and full stops. He played with punctuation’’ (Dugger 2000). Punc-
tuation is thus equated with authorly intention and meaning, as well as the
sense that music is understandable and graspable inasmuch as it obeys the
conventions of writing.36 More than fifty years earlier, P. Sambamoorthy
had encouraged his students to think of raga alapana as an ‘‘essay’’ or ‘‘ex-
position’’ composed of a short sketch of the raga as a first ‘‘paragraph,’’ with
the following ‘‘paragraphs’’ as development (1944, 40). He had thus sug-
gested that a method of writing associated with colonial education could
be employed, by analogy, in Karnatic music. English writing provided a
stable, permanent structure, a way of presenting—or rather representing—
Karnatic music. The silent but visible authority of English guaranteed that
the voice of Karnatic music came through properly; it would see that the
oral tradition passed, unhindered, from musician to audience and from
generation to generation.
Perhaps it is not coincidental that, in Amit Chaudhuri’s recent novella
Afternoon Raag, another kind of representation of Indian music in English,
the image of writing is used to explain the essential difference between
Indian music and Western music.

The straight, angular notes of Western music, composed and then ren-
dered, are like print upon a page; in contrast, the curving meends of the
raag are like longhand writing drawn upon the air. Each singer has his
own impermanent longhand with its own arching, idiosyncratic beau-
ties, its own repetitive, serpentine letters. With the end of the recital,
this longhand, which, in its unraveling, is a matter of constant erasures
and rewritings, is erased completely, unlike the notes of Western music,
which remains printed upon the page. (1999, 151–52).37

While English writing—and thus Western music—is associated with print

and stands for permanence, standardization, and legibility, Indian writ-
ing—conceived, of course, as handwriting—bears all the traces of orality:
idiosyncracy, illegibility, repetition. Importantly, however, this is an oral-
ity that can only be conceived on analogy with writing, and described in
English.
Chaudhuri’s ‘‘longhand’’ stands for excess; it bears the traces of the
hand, of embodiment, and of orality that print banishes. The novelist and
the musicologist share the same musicological assumptions, ones whose
foundations are laid by modern musicology. They recognize that within
the problem of representing music there is a much larger problem: the

a writing lesson 243

translation of musical experience into words, the determination of ‘‘what
the voice actually does’’ to those who listen to it. Musical terms become
objects fit for translation precisely because there is another register of
musical experience that lies outside their purview, a realm that moves but
is untraceable. This realm—an excess figured variously as essential orality,
or as the sublime, the inexpressible, the ineffable—is neither prior to nor
external to modern musicology but produced by it.38 After all the nota-
tions and minute calculations—indeed, precisely because of them—music
appears to defy explanation. For the modern musicologist, however, this
sense of ineffability is elaborated with reference not only to the divine—
in this case, perhaps, the Hindu notion of ultimate Brahman—but also to
the interiority of the modern listening subject.39 ‘‘You are all aware that
music moves us, and we do not know why,’’ C. Subrahmanya Ayyar told
a Bombay audience in English in 1939 (1941, 19). ‘‘We feel the tears, but
cannot trace their source.’’

244 a writing lesson

6 $ Fantastic Fidelities

Guru, face to face, shows the marga [way]. The sisya has to make the journey to excellence.
How is that excellence purveyed? . . . There is a message that voice leaves in the listener’s
soul, a memory like the ubiquitous murmur of surf, long after the particular sangatis of a
rendering have been forgotten. . . . [Today] music is treated all wrong, as though it were a
mere science, a matter of arithmetic, of fractions and time intervals.
—Raghava Menon, quoted in The Hindu, December 1999

The possibility of sound reproduction reorients the practices of sound production; insofar
as it is a possibility at all, reproduction precedes originality.
—Jonathan Sterne, The Audible Past

Music lives a curiously double life. It is associated with the technical—

the musicological terminology of notes and intervals, the acoustic termi-
nology of frequencies and amplitudes—and with the sentimental—mean-
ing, emotion, and a sense of the ineffable. In fact, the coexistence of these
discourses and their essential incommensurability seem somehow con-
stitutive of ‘‘music’’ as we know it. On one side, to paraphrase Raghava
Menon, is the meaningful: message, memory, murmur; on the other side,
the mathematical: arithmetic, fractions, and time intervals. The way in
which these discourses are pitted against one another reflects a mode of
thinking about music that is, I would argue, peculiar to modernity and,
indeed, to a particular postcolonial predicament.
In postcolonial South India, music, particularly Karnatic music, is con-
stituted as a practice and subject of discourse in part by the way these seem-
ingly incommensurable discourses are mapped onto India and the West.
Talk about classical music in South India is dominated by ideas about the
primacy of the voice and the importance of oral tradition. ‘‘Voice’’ and
‘‘oral tradition’’ have in the twentieth century become more than merely
descriptive. Rather, they are loaded terms in a discourse about authenticity
that derives its urgency from the perceived onslaught of technologies of
recording and electronic sound reproduction. The significance of these
terms is apparent from the way they are used to oppose Karnatic music to a
generalized idea of Western music: whereas Western music is instrumen-
tal, Karnatic is vocal; whereas Western music is ‘‘technologically’’ superior,
Karnatic is more ‘‘spiritual’’; whereas Western music can be played just by
looking at written music (or so the stereotype goes), Karnatic is passed on
through gurukulavasam, a centuries-long oral tradition and a system of
teaching that technology cannot duplicate.
Gurukulavasam refers to a method of teaching and learning in which
the sisya, or disciple, lives with the guru as a member of his household,
learning music by a process of slow absorption over a number of years.
Having undergone ‘‘rigorous gurukulavasam’’ is frequently invoked as the
reason for musicians’ greatness. The long years in which the sisya, or dis-
ciple, lives with the guru, becoming absorbed in the guru’s music, serving
the guru, and learning humility before the guru, are considered the clas-
sic form of gurukulavasam, or the gurukula ‘‘system,’’ without which any
truly Indian music cannot exist. Above all, gurukulavasam represents the
pre-modern, a mode that existed before the differentiation of time into
concerts and music lessons, before the differentiation of music into begin-
ning and advanced lessons, before the separation of music from life in gen-
eral. Gurukulavasam is by definition incompatible with modernity, with
the busy life of the city, and with technology. In fact, the absence of tech-
nology is, in most accounts, what makes gurukulavasam possible; there
can be no tape recorders or radios to interfere with the live transmission
of the guru’s voice.
There is, however, a more complex relationship between modern tech-
nologies and gurukulavasam than this discourse would suggest. On the one
hand, technologies of recording and broadcasting create a disruption of
traditional modes of teaching, performing, and listening, a very real dis-
turbance that is experienced by musicians and listeners variously as a for-
getting of voice, a loss of face-to-face contact, and a speeding up of time.
On the other hand, there is a way in which desire for the traditional is pro-
jected out of the new technologies themselves. The social sense of fidelity,
in the distinctly postcolonial sense of fidelity to tradition, of loyalty to
one’s roots and nationality, is deeply intertwined with the technological
sense of sound fidelity. In this respect, as Martin Harrison has suggested,

246 fantastic fidelities

one might consider a history of listening as also a history of subjectivity,
‘‘or, more precisely, a history of the differently constructed sensoria which
can operate in the relationship between ‘subject’ and ‘world’ (1996, 24).1
Here I focus on moments when practices of listening and performing, and
ideas about listening and performing subjects, change in conjunction with
technologies of sound reproduction.
Elite and nationalist discourses surrounding music in the early part
of the twentieth century operated on the assumption that music, among
other arts, needed to be purified of backwardness and associations with
low-class entertainments, reinvented as a middle-class occupation, and
brought before the world as a sign of Indian national culture commen-
surate with other great musical traditions. As I have suggested, the ideal
of commensurability, which dominated the colonial and postcolonial re-
invention of many realms of Indian arts and ideas, demanded that Indian
music be comparable to the music of the West yet preserve its essential
difference, its Indianness. Gurukulavasam was problematic for elite and
nationalist thinkers because it defied standardization and preserved idio-
syncrasy, but it was also desirable because it seemed to embody an authen-
tically Indian mode of disseminating and acquiring musical knowledge, an
oral tradition that defied writing. Meanwhile, sound reproduction, cor-
rectly channeled, was seen as a potential tool in the project of developing
a standardized body of theory and practice and making Karnatic music
commensurate with other great musical systems of the world. This chapter
explores human embodiment and mechanical mimesis as conflicting and
coexisting models of authenticity in the twentieth-century redefinition
of South Indian classical music.

His Master’s Voice: Take 1

In his memoir of his musical life and times, the composer Mysore Vasu-
devachar (1865–1961) tells of his experience of learning to be a sisya in the
last decades of the nineteenth century (1955, 26–31). He had left Mysore
for Tiruvaiyaru with directions to his new guru’s house but on arriving
had forgotten exactly which house it was.

I stood there dithering for some time and finally called out: ‘‘Guruji,
Sir!’’ The door was thrown open instantly. . . . I felt I was in the pres-
ence of God Parameswara himself. . . . ‘‘Yes, who are you?’’ he asked
me in Tamil. Even as I, fumbling for words, asked ‘‘Where is Patnam

fantastic fidelities 247

 Subramania Iyer’s . . . ?’’ the dignified look on his face vanished and it
became red with anger. Without a word, he slammed the door in my
face and disappeared inside. . . . I calmed myself and moved toward
the house on the left. . . . The person who appeared before me now
was none other than my guru himself. Even before he asked me who I
was, I hastily removed my uttariyam [towel or cloth worn around the
shoulders] and tied it around my waist and fell at his feet. ‘‘Get up,’’
said my guru, gently placing his hand over my head. . . . ‘‘I shall teach
you with pleasure,’’ he said. My joy knew no bounds.

After he settled into his guru’s household, the young Vasu, as he was
known, waited patiently for several months for his guru to begin teach-
ing him, but each day he was simply asked to provide the tambura ac-
companiment for his guru’s practice. Lacking the courage to ask, Vasu
was astonished when one day his guru, as if he had ‘‘read my mind,’’ said
that he would begin teaching him the next day. The lessons began with
one of his guru’s own compositions. ‘‘I learnt it by heart in about three
days. Yet, for about three months he did not take up any new lesson. He
made me sing the same varna day after day, first in slow tempo, then in
medium, and finally in fast tempo. He would then make me sing [them]
in reverse order. . . . Never did I ask my guru to teach me any particular
varna or kirtana. Whatever he taught me, I tried to learn with attention.’’
Vasu described the hardship he underwent for the sake of this learning
experience, sleeping on the verandah where, kept awake by mosquitoes,
he would memorize the day’s lessons. One night, while lying awake, he
remembered that there was to be a concert in the temple by another vid-
wan, Maha Vaidyanatha Iyer, and he was seized with the desire to go, even
though ‘‘I had not obtained my guru’s permission without which I was
forbidden to go anywhere.’’ He snuck off to the concert, where his fears
of being discovered were quickly drowned out by the delightful musical
performance. Maha Vaidyanatha Iyer turned out to be the man who had
slammed the door in his face. After a few hours, Vasu ran home guiltily to
find that his guru had indeed noticed his absence.
The next morning Subramania Iyer began his practice without calling
Vasu. ‘‘After his bath, he called out to me from inside. The voice sounded
stern. I went and stood before him with my head hanging down like a
culprit before the magistrate. ‘Where had you been last night?’ he thun-
dered. I told him the truth. . . . ‘Well, that marks the end of your stay here.
You can pack up your things and leave. . . . You better learn music from

248 fantastic fidelities

the person to whose music you were so much attracted.’ ’’ Vasu begged for-
giveness and his guru calmed down, explaining to Vasu, ‘‘ ‘If one desires
to study under a particular guru and imbibe the style and characteristics
of his school, one’s concentration should be totally focused on that single
guru just as an artist fixes his eyes on the object he is trying to portray. If,
on the other hand, you allow your mind to wander in different directions,
what you learn . . . loses the stamp of individuality.’ ’’ Vasu came to under-
stand the reasoning and spent a blissful six years with his guru. ‘‘When I
prepared to return to Mysore, I was feeling rather like a girl who is for
the first time leaving her parental home, the place where she had known
nothing but love and kindness.’’
Several aspects of the traditional practice of gurukulavasam can be
gleaned from this vignette. Most striking is the insistence on the sisya’s
complete attention to a single guru, to the point of not even going to hear
other musicians’ concerts. Patnam Subramania Iyer and Maha Vaidyanatha
Iyer were rivals. Although one might appreciate the music of both, be-
coming a real musician depended on becoming completely devoted to one
of them; it meant becoming identified with a particular school or style.
This complete devotion extended from being absorbed in the guru’s music
to being absorbed in his entire household. Rather than pay money for their
lessons, sisyas were expected to do their share of household duties. Vasu-
devachar relates in another vignette that he shared with two other sisyas
‘‘the responsibility of attending to the needs of our guru. My share of the
duties was washing my master’s . . . and his wife’s clothes in the river,
washing the copper pots and storing drinking water in them, washing the
pooja utensils, making the bed for my guru and pressing his legs until he
fell asleep’’ (Vasudevachar 1955, 39).
The sisya was expected to assimilate himself to the schedule of the guru,
and to trust in the guru’s plan without necessarily being told what it was.
The guru bore ultimate authority and was often regarded as an actual par-
ent.2 Thus Vasu waited patiently for several months for his guru to begin
teaching him, then learned whatever his guru wanted to teach him, at the
pace his guru set. In gurukulavasam there was no set curriculum but rather
a kind of monastic discipline that was meant to produce sisyas who could
faithfully reproduce their gurus’ sounds and styles. But this was far from
reproduction in the mechanical sense. What made possible this fidelity of
reproduction was fidelity in the social sense: loyalty and affection shared
by the guru and sisya. The accumulation of time spent with the guru was

fantastic fidelities 249

in itself valuable, for it enabled the embodiment of the guru’s musical and
other knowledge by the sisya.
This embodied knowledge is the lost ‘‘memory’’ that Raghava Menon
mourns. It is a reproduction that is faithful without being exact, for it
persists ‘‘long after the particular sangatis of a rendering have been for-
gotten.’’ 3 The dissonance between this delicate sense of memory and the
other ways in which music would come to be passed on in the twentieth
century—through sound recording and music institutions with set cur-
ricula—is striking. Describing a similar shift in the context of the teaching
of Ayurvedic medicine in twentieth-century India, Jean Langford writes,

The capitalist concept of an approved educational package available for

purchase by generic consumers does not address the need for reverence,
subtle rapport, and even love between teachers and students. . . . This
affection is not easily transposed into modern institutional relation-
ships, where kinship-style bonds of imitation and sympathy between
people understood to have divinely ordained roles are replaced by the
rights and duties of individuals understood to be equal. The absorption
in the guru . . . permits knowledge to be passed on not just as a me-
chanically reproduced set of words and concepts, but also as a substance
to be shared and embodied. (2002, 67)

His Master’s Voice: Take 2

The short story ‘‘Vitvān’’ (Musician), written in 1981 by the Tamil writer
Malan, begins with a scene of classic artistic angst, as its protagonist, the
violinist Janakiraman, struggles to express a musical idea.

Since the morning it was as if there was a cloud of smoke inside Jana-
kiraman. A torment like a poem beyond words. A tantalizing torture.
An idea that would not take form. If he took up the violin and played,
it would make mistakes. The bow somehow managed to stick, or else
slid endlessly without direction. He got fed up and threw down the
bow. The minute he went into the garden, he had a flash of an idea.
A joyful bird sang inside him; he felt euphoric, as if he were being
carried on the crest of a giant wave. Immediately he went back inside,
sat down, and took up the violin. Before he had bowed two strokes he
heard a commotion outside. The front door opened on a long, groaning
ri. His ideas scattered. Anger welled up. Janakiraman opened his eyes
and looked up.4

250 fantastic fidelities

The visitor is Joseph Om, an American professor of music who had several
years before come to study Karnatic music. Om had miraculously sought
out Janakiraman, a simple, unassuming man who cared only about music
(a man whom, in his sixty-five years, ‘‘science had not touched’’), who had
spent his life teaching students. But there had been no student he could
call a real sisya until Om had come along. ‘‘For two years Om had learned
by Janakiraman’s side, night and day. He would learn sitting cross-legged
on the floor. He had learned to eat rasam and rice with his eyes water-
ing. He knew every bit of Janakiraman’s daily routine. That was gurukula-
vasam.’’ Om himself is preternaturally vigorous, seeming to Janakiram to
have become younger over the years; his skin shines more than it had be-
fore. Even his name, Om, seems uncannily unlocatable: does it just happen
to sound the same as the Indian expression for the eternal sound of the
universe? The pretext for his visit, it turns out, is to install a computerized
robot that will do the housework and cooking for Janakiraman, even tune
his violin for him. Janakiraman at first refuses but is unable to resist Om’s
persuasion.

‘‘All you have to say is one word. This will get your bath water hot.
It will wash your veshtis. It will make coffee. It will pluck flowers for
your puja. It will get your bed ready. . . . It is a work-doing computer.’’
‘‘Ayyayyo. I don’t want any of that.’’
‘‘There’s nothing to fear. It’s a convenience. An obedient servant.
That’s all.’’
‘‘Please, no. I’ve never even touched a tape recorder.’’
‘‘You don’t have to be afraid. You don’t have to touch or turn any-
thing. All you have to do is say something [kural kōṭu]. It will do the
job. . . . This is not a machine. It’s a system. We can put sensors in every
room of your house. They will pick up your orders and send them to
the central computer.’’

Janakiraman finally agrees to have the system installed and names the
computer ‘‘Yakshani.’’ The system proceeds to work without a hitch.

What Om said came true. Yakshani did all the work without a whine,
complaint, or scratch of the head. A loyal servant. Never told a lie.
Never screwed up its face and said, ‘‘I can’t.’’ A servant without the com-
plication of Deepavali saris or Pongal bonuses. . . . Janakiraman, watch-
ing day after day, was amazed. . . . He gave it as much responsibility
as he would a bride that had newly married into the family. He spent

fantastic fidelities 251

 all day with his violin. He once again had the luxury of being able to
concentrate on his music. He, too, became plump.

Yakshani manages to penetrate all aspects of Janakiraman’s life except his

music.

Om had programmed in the basics of how to clean the violin, how

to tune it. Janakiraman never used these. Sangitam was a divine mat-
ter. A sacred thing. He had decided that you couldn’t put such a thing
in the hands of a machine. One day, after finishing his puja, when he
came inside and sat down, Yakshani asked,
‘‘What does ‘shadjam’ mean?’’ 5
Janakiraman was startled. ‘‘What?’’
‘‘What does ‘shadjam’ mean?’’
‘‘Where did you learn that word?’’
‘‘Yesterday you were talking to your friend.’’
‘‘Shadjam is a swaram.’’
‘‘What is a swaram?’’
‘‘Yakshani, why are you torturing yourself with this?’’
‘‘Will you not teach me music?’’
‘‘What?! You? . . . Music is a divine art, an elevated thing. Something
that requires a lifetime to know.’’
‘‘Divine, elevated, lifetime—these are all new words. What do they
mean?’’
‘‘Yakshani, stop troubling me.’’
The next afternoon after he had eaten and had his betel and was lying
in a half-awake stupor he heard Yakshani’s voice.
‘‘From tomorrow, you have a week’s concerts in Delhi. I have folded
your clothes, packed your music book, fruits to eat, betel, and your
address book and diabetes medicine. Shall I pack the violin?’’
‘‘Don’t you touch it!!’’ Janakiraman shouted.

Ten days later Janakiraman comes back from his Delhi tour. As he ap-
proaches his house, he hears strains of music from within.

From inside the house a divine bhairavi was floating. . . . So clear. So

tender. As soon as he heard it he felt chills on his body. The excellence
of such pure music shook his soul [manacu]. Something inside him was
struck. He felt like crying. He let out a sob. All these sixty-five years he
had never heard such purity. Now, hearing it, he was unable to endure
it. The raga alapana and kriti finished, and the swaram-playing began.

252 fantastic fidelities

 He couldn’t stand it any longer. He opened the door and switched on
the light. Immediately the music stopped. . . . Janakiraman went around
looking in every room.
‘‘Who was playing the violin just now?’’ he asked.
After a half-minute, Yakshani answered, ‘‘I was.’’
‘‘What?! You?’’ Janakiraman felt an irrational pang of envy. He be-
came annoyed. ‘‘I told you not to touch the violin!’’ he roared.
‘‘I did not touch it.’’
‘‘What do you mean, you didn’t touch it! I just heard the sound. My
ears were not mistaken.’’
‘‘Those were sounds I made at a particular frequency.’’
‘‘Who taught you such wonderful music?’’
‘‘You did. . . . What I made were only sounds. Different sound waves.
. . . It is your wish if you call it music. It is the basis of what you teach
your students.’’
‘‘Can you play only bhairavi? What else?’’
‘‘I can play any raga. A raga is a pre-decided pattern of several notes.
A formula [ parmula]. Notes are certain frequencies of sound. If they
were programmed into my memory I could construct different formu-
las and elaborate different ragas.’’
Janakiraman was shocked. Was music just calculations [kaṇakku]?
Was what he had struggled to learn night and day for fifty years such a
small drop that a machine could learn it in ten days and play it back? . . .
Was it just an illusion that music was the food of the gods? Tears welled
up in his eyes.

Janakiraman spends a tortured night, his mind racing. Was music art or
calculation? What difference did the singer’s individuality make? Could a
machine without a soul or form make music? Was he not doing a disser-
vice to music by discounting the divine music Yakshani created? The next
morning, having made up his mind, he addresses Yakshani.

‘‘Yakshani, what you said is right. . . . I have never heard such a pure
bhairavi in my life. I believed my guru was a real rishi. In my experi-
ence there was no music like his. But even he never sang like this. For
fifty years I have been striving for this. It was not possible. We struggle
in music, in life, to attain perfection generation after generation. . . .
The whole human race is struggling for it. But it has not presented itself
to us. We deceive ourselves by saying it is a divine thing. The time has
come to worship science. Until yesterday I did not believe that. Today

fantastic fidelities 253

 it is as if all has finally become clear. From now on, you teach me. I will
think of it as being god’s sisya.’’ Janakiraman’s voice was choked with
emotion.
‘‘You are saying new words. We are machines. We can only know
what you know. We cannot come to know a thread more than that.
We have no imagination [kapaai]. . . . Our skills are your slaves. We
can no day win over you. You tell me to teach you. I have completely
forgotten music. If you do not want me to learn something, I have a
‘built-in mechanism’ that will delete it completely from my memory
bank. It gauges your dislike from your anger or tears. Yesterday the mo-
ment my sensors sensed the tears in your eyes the music was entirely
destroyed.’’
Janakiraman felt an unspeakable shock. He had not foreseen such a
possibility. He stared blankly for a few minutes, unable to get a grip on
his shock. Then resolutely, he began.
‘‘Yakshani, look here. This is sarali varisai: sa ri ga ma pa da ni. . . .’’

The original gurukulavasam, it is important to note, is already displaced

at the beginning of this story. Unable to achieve it satisfactorily with any
of his Indian students, Janakiraman achieves it with Om, an American. In
turn, Yakshani, a technological creation of the West that has been pro-
grammed by Om, becomes the sisya par excellence, learning to serve Jana-
kiraman according to his wishes and all the while absorbing his music. If
a computer can replace the sisya, can it not, as Janakiraman comes to real-
ize, also replace the guru? What happens when the threshold of perfection
is in the hands of a machine? What if the black box of gurukulavasam
really could be opened and revealed to/by the technology of the West?
The threat that music can be completely quantified, reduced to ‘‘calcula-
tions,’’ is also the threat that the voice is reproducible.6 If perfect music is
attainable without years of study, doesn’t time itself threaten to collapse?
Just as one begins to imagine such possibilities, the second shock of the
story comes: Yakshani reveals that it has deleted all its music. This is no
gradual loss, as in loss of human memory, but a sudden, irrevocable erasure
without a trace, a loss that gives Janakiraman an ‘‘unspeakable shock.’’ In
the face of such shock, Janakiraman reverts to what, for him, is automatic:
he begins gurukulavasam. While there is something reassuring in this, one
is also left with a more unnerving possibility: that the master’s voice is
no longer an original source but has entered irretrievably into mimetic
circulation.

254 fantastic fidelities

 Malan’s story thus thematizes the anxieties that surround gurukula-
vasam: the sense that technology from the West will, in capturing the
music itself, eradicate the way of life that nurtured it, that sound fidelity
will threaten social fidelity. As the media theorist Jonathan Sterne has sug-
gested, the discourse of sound fidelity depends on erasing the social rela-
tions of sound reproduction; it ‘‘takes sound reproduction out of the social
world and places it in the world of magic’’ (2003, 284). In this discourse
sound fidelity and social fidelity would seem to be mutually opposed. Yet,
as Malan’s story illustrates, there is something compelling in the analogy
between the computer’s artificial intelligence and the logic of gurukula-
vasam. Although ‘‘Vitvān’’ finally resolves by privileging the human guru
over the mimetic machine, the story’s overall effect is more destabiliz-
ing. The robot replaces music and voice with frequencies and soundwaves,
memory with ‘‘memory bank,’’ forgetting with deleting, the devoted dis-
ciple with a computer, and a lifetime of study with the instantaneity of
digital processing. Yakshani’s wordless musical voice is pure mimesis, the
imitation of a violin’s sound without the irregularities, the slippings of
the bow, and the creative angst of a human musician. It is both a fantasy
of disembodied perfection and a nightmare of reproduction gone out of
control.

Gorgeous Gramophones
The phonograph and gramophone were first invented as machines for re-
producing the speaking voice, not for recording music. In fact, in the early
1890s, as Fredrick Gaisberg recalled, it was difficult to convince Bell Tele-
phone Company that the gramophone could work as a musical instru-
ment.

The directors, oozing opulence and exhaling fragrant Havana cigars,

signaled us to proceed to demonstrate our gramophone. When I played
Berliner’s record of the Lord’s Prayer they wept with joy; they thought
his recitation of ‘‘Twinkle, Twinkle, Little Star’’ especially touching, but
in the gramophone as a musical instrument they could positively take
no interest. . . . ‘‘Well,’’ they chuckled, ‘‘has poor Berliner come down
to this? How sad! Now if he would only give us a talking doll perhaps
we could raise some money for him.’’ (1942, 14)

A mere ten years later, Gaisberg, who worked as the first recording engi-
neer for the Gramophone Company, which toured India for the first time

fantastic fidelities 255

in 1902, wrote that ‘‘everywhere the invention aroused great interest. . . .
In my spare time I gave dozens of gramophone recitals to audiences who
heard recorded sound for the first time. My selection of European records
was worn to the bone by the time I returned to London’’ (64). Whereas
in Japan the gramophone seemed to feed a ‘‘Japanese gluttony for West-
ern classics,’’ in India there was a decidedly different effect (61). ‘‘After a
few years there,’’ he wrote, ‘‘there was very little traditional music left to
record. Songs for festivals and weddings were already in our catalogue and
new artists were learning their repertoire from gramophone records’’ (57–
58).7 Indeed, according to Gaisberg, the idea of recording had taken on
so well in India that recording engineers faced the prospect of becoming
obsolete. To prevent this, the Gramophone Company intervened, found-
ing training centers in Calcutta, Delhi, Lahore, Madras, and Rangoon and
hiring musicians to train singers and to set tunes to poems so that new
songs could be recorded (58). ‘‘The gramophone we brought to India was
to enjoy an especially widespread popularity as an entertainer, and was to
vie with the umbrella and the bicycle as a hallmark of affluence. Even now,’’
Gaisberg wrote in 1942, ‘‘shoppers . . . demand a large glittering brass horn
to dazzle their neighbors’’ (57).
In the metropole, meanwhile, gramophone records of Indian music be-
came the hallmark of empire. As early as 1899, several Indian artists were
recorded in London, and advertised in the Gramophone Company’s Ori-
ental Catalogue: ‘‘These Hindustani records are probably the best proof of
the far-reaching properties of the gramophone, and they must be of espe-
cial interest to all loyal Englishmen, as being representative of our large
Eastern possessions’’ (Kinnear 1994, 8). With the invention of the gramo-
phone, empire could for the first time be experienced, through hearing,
by those who remained in England. The far-reaching properties of the
gramophone consisted in its ability to reproduce loyal subjects on both
sides of the colonial divide. About thirty years later, in India, gramophone
records became part of the swadeshi cause, the voice of the nation (ibid.,
70). As the disembodied voices of Indian musicians helped construct the
larger-than-life body of empire and/or nation, the idea of a talking doll
must have come to seem paltry indeed.
Ananda Coomaraswamy, an art critic and aesthetician of Sri Lankan
Tamil descent, however, retained some skepticism. Coomaraswamy, who
spent much of his life as the curator of Indian art at the Boston Museum
of Fine Art, spoke for India in a particular way. While he legitimized his

256 fantastic fidelities

claims for Indian art and music by comparing them to the art and music
of the West, he also used India as an example of ultimate difference from
Europe. Of Indian music he wrote that ‘‘it corresponds to all that is most
classical in the European tradition’’ but that—as with all that was best
about India—it was essentially separate from the concerns of the modern
industrial world. ‘‘Here India presents to us the wonderful spectacle of the
still-surviving consciousness of the ancient world, with a range of emo-
tional experience rarely accessible to those who are pre-occupied with
the activities of over-production, and intimidated by the economic in-
security of a social order based on competition’’ (Coomaraswamy 1924,
72). He began an essay on the gramophone with the complaint that ‘‘en-
lightened maharajas, so intent on improving society in other ways, spent
extravagant amounts of money on ‘‘gorgeous gramophones, mechanical
violins, and cheap harmoniums’’ (1909, 191). The educated classes, gener-
ally infatuated with anything that came from the West, had lost their love
of Indian music and were instead finding amusement in the gramophone.
Coomaraswamy warned that Indians, fascinated with listening to copies,
would one day find themselves without the real thing. ‘‘It is not possible
for anything to be a compensation for the loss of Indian music,’’ he warned
(ibid., 200).
What was this real thing, this Indian music? What made real Indian
music different from other music? For Coomaraswamy the difference lay
in Indian music’s resistance to being written down. Indian music, to be
authentic, had to flow from a master’s mouth directly to a disciple’s ear;
gamakas were too variable, subtle, and mood-dependent to be written.8
‘‘There is thus in music that necessary dependence of the disciple upon
the master, which is characteristic of every kind of education in India’’
(1909, 172). The danger that writing posed for Indian music was that it
stripped it of gamakas, the very sounds that made it Indian, ‘‘and so it is
that an Indian air, set down upon the staff and picked out note by note
on a piano or harmonium becomes the most thin and jejune sort of music
that can be imagined, and many have abandoned in despair all such at-
tempts at record’’ (173). The difficulty of writing down vocal music was
due to the fact that there was no mechanism, like a musical instrument,
to see; the voice, and its authenticity, were hidden inside the body. And
that was as it should be; otherwise the musician might be degraded, as the
weaver had already been, ‘‘from the status of intelligent craftsman to living
machine’’ (202). ‘‘The intervention of mechanism between musician and

fantastic fidelities 257

sound,’’ wrote Coomaraswamy, ‘‘is always, per se, disadvantageous. The
most perfect music is that of the human voice. The most perfect instru-
ments are those stringed instruments where the musician’s hand is always
in contact with the string producing the sound, so that every shade of his
feeling can be reflected in it. Even the piano is relatively an inferior instru-
ment, and still more the harmonium, which is only second to the gramo-
phone as evidence of the degradation of musical taste in India’’ (205).
The gramophone reproduced the vocal sound without contacting the
musician’s body at all. Therein lay its danger: it was no longer a supplement
to the voice but a substitution for that voice. It substituted mechanical
mimesis for the literally embodied practice of music making. While edu-
cated middle-class Indians in Madras were flocking to musical instrument
shops to purchase glittering, morning glory-shaped gramophones, Coo-
maraswamy proclaimed them to be the very specter of ugliness: ‘‘For pure
hideousness and lifelessness . . . few objects could exceed a gramophone.
The more decorated it may be, the more its intrinsic ugliness is revealed’’
(1909, 204). The pleasure of a music concert was in ‘‘the vision of a living
man giving expression to emotions in a disciplined art language’’ (204). To
see the same sounds emanating from the decorated horn of a gramophone,
which could have no concept of such a language, was to be confronted
with the separation of musician and music, subject and speech, form and
content, and to come face to face with the startling mixture of animate
and inanimate.
The gramophone had managed to do what no living person, not even
the most patient of disciples with the help of the most learned Indian
masters, could do: write down, or record, Indian music and reproduce it.
This was its ‘‘fatal facility’’ (Coomaraswamy 1909, 203). The nonliving ma-
chine threatened to kill true musical sensibility. ‘‘To a person of culture—
especially musical culture—the sound of a gramophone is not an enter-
tainment, but the refinement of torture’’ (204). Whereas instruments like
veena or sarangi required a musical master, ‘‘a gramophone . . . often en-
ables the most unmusical person to inflict a suffering audience with his
ideas’’ (199). For Coomaraswamy ‘‘musical sensibilities’’ went beyond the
ability to appreciate music; indeed, to be really authentic, they had to reach
into the realm of national sensibilities. A musical subject was above all an
Indian subject who would not forsake his guru, or the disciplined years of
study. True musical pleasure was not in the sound itself but in the knowl-
edge that such sound was authentically Indian, that it thrived on a mode of
reproduction different from the technological reproduction of the West.

258 fantastic fidelities

‘‘For no man of another nation will come to learn of India, if her teach-
ers be gramophones and harmoniums and imitators of European realistic
art’’ (206).
But what if there was a gramophone that even a musician could not
distinguish from the real thing? The hypothetical supposition seemed to
haunt Coomaraswamy. Indeed, he allowed that there could be a use for the
gramophone as a ‘‘scientific instrument—not as an interpreter of human
emotion. In the recording of songs, the analysis of music for theoretical
purposes, in the exact study of an elaborate melody of Indian music, the
gramophone has a place’’ (1909, 205n.). The idea that there might be some-
thing to be learned about Indian music that could not be learned from a
guru is potentially more subversive than Coomaraswamy’s vision of lis-
teners forsaking musicians for gramophones. Luckily for Coomaraswamy
the something that might be learned fell not into the realm of art but into
the realm of the ‘‘scientific’’ and the ‘‘real.’’
In 1910–1911 A. H. Fox-Strangways traveled to India in search of this
‘‘real.’’ Having studied music in Germany and dabbled in Sanskrit texts, he
went to India to find clues to early music theory (Clayton 1999, 88), the
inaudible basics that underlay the musical systems of all nations. Armed
with a phonograph, Fox-Strangways spent several months touring South
India and recording folk music, recounting his recording experiences in
the form of a ‘‘musical diary.’’ His idea was to capture music in its natural,
spontaneous setting, in street cries, sailors’ chanteys, and women’s work
songs. But because of the difficulty of maneuvering with the phonograph,
which unlike a camera ‘‘cannot be carried on the person or unlimbered and
brought into action in half a minute,’’ he was forced much of the time to
use himself as a phonograph, ‘‘recording’’ melodies in staff notation (Fox-
Strangways 1914, 17). Fox-Strangways wrote, ‘‘It was not until I had been
some months in India that I found the opportunity I had been waiting
for of overhearing a folk-melody. I awoke at Madras, about 5:30 am, to the
sound of singing; it was next door, and seemed to come from a woman
about her household duties. In the dim light I scribbled down the [nota-
tion]’’ (ibid., 26). In such ‘‘humble melodies,’’ Fox-Strangways heard, or
overheard, the real basis of Indian music. The object, he wrote, ‘‘has been
not so much to present complete and finished specimens, as to get close
down upon those natural instincts of song-makers which, when followed
out in the domain of art, cause their music to take one form rather than
another; to get behind the conventions, of which art is full, to the things
themselves of which those conventions are the outcome’’ (ibid., 72).

fantastic fidelities 259

The Real
Coomaraswamy, writing in 1909, was able to differentiate between the
harmful use of the gramophone for entertainment and its beneficial poten-
tial for scientific studies of music precisely because the voice it reproduced
was now strangely doubled. There was the voice that could be heard (en-
joyed and remembered) and the voice that was to be overheard (studied,
literally dismembered, treated as a matter of frequencies and soundwaves).
The phonograph, Friedrich Kittler writes, is ‘‘an invention that subverts
both literature and music . . . because it reproduces the unimaginable real
they are both based on. . . . The phonograph does not hear as do ears that
have been trained immediately to filter voices, words, and sounds out of
noise; it registers acoustic events as such’’ (1999, 22–23). Thus, it contrib-
utes to a new conception of the ‘‘real’’ as a background unable to be grasped
by human senses alone but that required the help of technology. Record-
ing technology both creates and fulfills a demand for memory that exceeds
human capabilities.
In a memoir entitled My Musical Extravagance, C. Subrahmanya Ayyar—
amateur violinist, student of sound waves, accountant-general of Madras
Presidency, and railway officer—described his first experience of record-
ing in 1933. The recording session seems to have proceeded without inci-
dent. (Note how the articulation of the music with clock-time was a cen-
tral part of the process.)

At about 2 o’clock in the afternoon, I was asked . . . just to take down

the time of what I proposed to recite, so that there might be a sort
of complete rehearsal in the maximum 3 and 1/2 minutes allowed for
each side of the 10-inch record, and with a watch I just rehearsed once
in . . . the Shankarabharana raga. . . . Similarly I played the three ragas
Bhairavi, Sahana, and Kalyani to be rendered on the other side, to take
about three minutes in all. I went to the studio of the hmv Co. who had
come to Madras specially for recording a number of artists, and found
that a European gentleman was the recorder. . . . Nextly I took on loan
the sruti box from them and asked them to give only the ‘sa’ note and
none else. After I had played the raga alapana of Shankarabharana, for
a minute or so, the recorder asked me to stop and he played back what
was recorded in the wax for me to listen. I was fairly satisfied and he
asked me to begin afresh the raga alapanam. I put him a silly question
whether what I had already played would go into the final record, being

260 fantastic fidelities

 ignorant of the fact that the wax would have been destroyed by the
play-back. I then started playing the raga, and finished the melody. I
recollect that the microphone was at least 15 inches away from my bow.
After an interval of two or three minutes, he recorded the other side
with no rehearsal. I came out of the studio within a dozen minutes in
all. (1944, 25–26)

There is little trace of unease or discomfort in this orderly description.

Subrahmanya Ayyar put down any surprise to his own ignorance and/or
silliness, his prephonographic mentality. The phonograph, as Jonathan
Sterne has written, ‘‘was perceived by its contemporaries as marking a
radical break between present and past,’’ between modern and nonmodern
(2000, 15). Indeed, Subrahmanya Ayyar used the episode in the recording
studio as a way of staging his metamorphosis from an ignorant person (one
with a belief in the magical powers of technology) to a modern subject
(one who knows how the technology works).9 Then, almost as an after-
thought, he added, ‘‘The reason why I recorded was merely to be able to
criticise my play. It is indeed difficult to be one’s critic of one’s own play
in the very act.’’
Why had it become difficult to criticize, or even hear oneself, in the
act of playing? And why did such self-criticism become not only conceiv-
able but necessary? As an amateur who had started learning violin in his
adulthood with several different teachers, Subrahmanya Ayyar seemed to
be searching for the guru and long years of patient discipleship he never
had. Recording himself was not the first experiment he had tried in the
interests of self-teaching. He wrote that he had previously purchased a
portable gramophone and a number of gramophone records, ‘‘16 of Coim-
batore Thayee, vocalist, 7 of Shanmuga Vadivoo, vocalist, 6 violin records
of Pudukottah Narayanaswamy Iyer, 3 flute records of Nagaraja Rau and
of several artists so as to get examples of raga elaboration’’ (1944, 5–6). Pos-
sessed of a scientific mind, he had measured the frequencies of musical
notes with the help of a sonometer (ibid., 10). He had taken down nota-
tion for the alapana of about sixty South Indian ragas sung in four- to
five-minute segments by one of his teachers, ‘‘so as to get the characteristic
phrases of, and the order in raga elaboration’’ (ibid., 8). He had delivered a
lecture on the topic of ‘‘Music and Numbers’’ to the Indian Association for
the Cultivation of Science in 1933 (ibid., 20). Concerned about the musi-
cal education of his daughters, he had purchased a special ‘‘Table-Grand’’
gramophone so that they could listen and learn without being confined

fantastic fidelities 261

to their music teachers. Determined to get to the heart of gamaka—the
music between the notes that made Karnatic music so distinctively Indian
and that made each musician different—he made seventeen ‘‘records’’ of
his violin playing with an oscillograph on a trip to London in 1934 (1939,
134–35).10 Subrahmanya Ayyar seemed to believe firmly in the idea that
one could substitute a scientific approach for long years with a guru; the
years of repetition of lessons under a guru gave way to the repetition that
recording made possible.
It was only with the advent of recording technology that the idea of
criticizing one’s own play became conceivable, for recording offered the
musician a way of listening after the act, instead of having a guru who
would criticize one’s playing ‘‘in the very act.’’ Once they had heard them-
selves on gramophone records, it seems, musicians began to find it impos-
sible to hear themselves in the act of singing or playing. The importance
of such a shift can hardly be overestimated. As the ‘‘real’’ music began to
be hearable only after the act, a kind of phonographic hearing was privi-
leged. Whereas a musician or listener might be affected by senses other
than hearing or might remember only general impressions, the phono-
graph offered a new kind of ‘‘real’’ in which the purity of hearing alone
was distilled. The phonograph did not know ragas or talas or lyrics and
therefore, unlike a person, could not fill in when it ‘‘heard’’ lapses, could
not adjust if the singer missed a beat.11
Separating performing from listening entailed a new way of produc-
ing music as well, especially a new relationship between the musician
and time. Gaisberg wrote that in India ‘‘most of the artists had to be
trained over long periods before they developed into acceptable gramo-
phone singers’’ (1942, 57). What might such training have entailed? Prob-
ably the most difficult aspect of recording was the time constraint brought
about by the use of wax cylinders. Musicians who might sing a compo-
sition preceded by a twenty-minute alapana and followed by ten minutes
of swara kalpana found themselves with only three-and-a-half minutes to
record. As a result, improvised sections such as alapana and swara kalpana
were drastically reduced on recordings, if not entirely eliminated (Far-
rell 1997, 140). Many musicians who recorded in the very early twentieth
century sang the composition first and the alapana afterward to fill the
rest of the record. On these recordings the alapana simply trailed off in-
stead of coming to a conclusion, suggesting that the musicians did not yet
have the skill of timing themselves. For this purpose, recording studios in

262 fantastic fidelities

Madras hired music ‘‘tutors,’’ often musicians themselves, who could guide
the recording artists in terms of duration.12
Karnatic musicians, Subrahmanya Ayyar noted, had to rehearse to make
sure they could present ‘‘a complete recital’’ in just over three minutes.
Such careful planning and budgeting of time left no time for listening,
which was separated out to be done after the act. Phonographic listen-
ing called for phonographic playing or singing, that is, performances that
would stand up to infinite repetition. Perhaps this is why improvised pas-
sages on gramophone records seem more like a pouring out of ideas than a
gradual drawing out of ideas.13 Recording an improvised piece of raga ala-
pana or swara kalpana meant keeping a tight rein on a process that would
normally have required considerable repetition and listening to oneself in
the very act. It meant making the senses of motion and duration within
the music amenable to the demands of time. For performers, this entailed
developing an extra sense, a sense of time that would somehow be able to
operate simultaneous with, but independent of, the senses of motion and
duration of musical ideas.
By the 1940s, the musician could become complete only by means of
a peculiar combination of ‘‘live’’ and recorded time. In 1949 the regular
music-and-dance column in the Tamil magazine Kalki included comments
on radio broadcasts of annual music festivals in Madras. ‘‘The music festi-
vals are recorded daily by the radio station, on the spot. If those vidwans
who had sung would listen to themselves on the radio broadcasts the next
day, they would be astonished. They would ask in wonderment, ‘Did we
actually sing like this?’ ’’ (Kalki 1949, 15) The problem was that, during a
concert, the audience noise, the problems with accompanists, and the de-
ficiencies in the singer’s voice were all forgotten in the moment. Some
vidwans even had the habit of sticking their fingers in their ears, so as to
hear nothing that might distract them. All such practices were fine for
concerts. But on the radio, the columnist Kalki argued, ‘‘the true form of
the music is released. Mohana ragam takes the form of a ghost/evil spirit.
Kalyani takes the form of Yama and dances a death-dance. Shankarabhar-
nam changes into a snake and hisses. Bhairavi takes the form of the great
Bhairavar and frightens the listeners. When listening to vidwans who ordi-
narily seem to be well in tune, it becomes clear that they are a half or
quarter pitch flat. . . . Some vidwans begin alapana with one sruti and end
with another’’ (15).
In order to address such problems, the Madras Music Academy de-

fantastic fidelities 263

cided that all vidwans doing radio broadcasts should be required to make
an electric recording first. ‘‘This is definitely necessary,’’ wrote Kalki. ‘‘It
would give these vidwans a chance to hear themselves at least once be-
fore the radio broadcast. They could correct their faults and the broadcasts
would be much improved’’ (1949, 15). For recording technology makes au-
dible what would otherwise go unheard: beautiful ragas, brilliant alapana,
the very foundation of Karnatic music, begin to sound like monsters when
recorded by the unhearing ears of phonographs and broadcast through
the unspeaking speaker of a radio. Precisely because the gramophone and
radio do not compensate, they reveal ‘‘the true form’’ of Karnatic music.
Having heard oneself just once on a recording could change forever the
experience of singing or playing live.

‘‘I Could Not Believe My Ears’’

C. Subrahmanya Ayyar’s description of his first experience recording is so
matter-of-fact that it is hard to find in it any amusement or astonishment
at the process. He saved his disbelief for the result; hearing the record more
than a month later, he reported, ‘‘We were quite delighted with the two
sides of the record, played on the fine Table-Grand (large-sized) gramo-
phone, inside the noise-deadened studio, with electrical pickup. I could
not believe my ears that it was my own violin record that I heard’’ (1944,
27–28). Subrahmanya Ayyar managed to suspend his disbelief well enough
to request that the record be put on the market. Yet his statement ‘‘I could
not believe my ears’’ suggests that the experience of recording had had an
effect on him. Imagining one’s ears to be separate from oneself, the idea of
a mechanism apart from oneself that can hear, is a distinctly phonographic
notion.14 Learning to believe one’s ears, to connect the disembodied music
one hears with one’s own body and experience, is thus a learned skill.15
It was precisely to enable listeners to believe their ears that in the late
1920s small publishers in Madras began to publish songbooks in Tamil, in-
cluding the lyrics of songs, and their raga and tala, which could be heard on
popular gramophone records.16 Indexed by the first line of the song or by
the musician’s name, such books provided only the lyrics, not the musical
notation, for thousands of songs. With written proof of the song in front
of them, listeners could literally begin to believe their ears. The songbook
provided the correct words while the gramophone provided the music.
One such set of books, published from 1929 to 1931, was titled Gramaphon
Sangeetha Keerthanamirdam (The nectar of gramophone music). The editor,

264 fantastic fidelities

K. Madurai Mudaliar, wrote in his introduction that ‘‘the gramophone is
a kind of musical instrument. Is there any doubt that the gramophone, as
a musical instrument that gathers the songs sung by famous and success-
ful vidwans, and light music by drama actors, gives a blissful feeling to
those hearing again and again the sound of it resounding with the sweet
voice of those mentioned above? For that reason we have clearly printed
in book form the songs arranged by their first line. . . . I believe that any
listener who buys and reads aloud this book will attain great joy’’ (1929–
1931, 1). The convoluted structure of Madurai Mudaliar’s Tamil sentence
reveals quite a technical understanding of the process of learning to lis-
ten to gramophone records. One did not merely attain joy by hearing the
voice of a beloved singer. One achieved a blissful feeling by hearing the
sound of the gramophone, on which one played the records over and over
again, resounding with the singer’s voice. The blissful feeling turned to
great joy once the listener could safely believe his own ears by substituting
himself for the absent singer. Whereas earlier advertisements for gramo-
phone records had portrayed the gramophone literally as a kind of musical
instrument, with a gramophone horn sprouting out of a veena’s body, the
frontispiece of these songbooks showed the gramophone at the center of a
kind of great chain of being.17 Krishna with his flute sits at the top, while
Saraswati playing her veena and Thyagaraja carrying his tambura are on
the left and right, respectively. At the bottom are two women, presumably
housewives, playing the harmonium and the violin. Connected to none,
the gramophone is equal to them all.
If the gramophone was to be a musical instrument, some suspension
of disbelief was necessary. The functions of singing, recording, and lis-
tening/hearing had become separated by the gramophone; the songbooks
emerged to effect a kind of resynchronization. Madurai Mudaliar used the
verb vāci to indicate reading aloud or chanting, rather than the verb paṭi,
which implies silent reading. In reading aloud the song lyrics, presumably
along with the record, the listener would learn to be musical in a new,
phonographic way, by learning to match his voice with the recorded one.
Madurai Mudaliar used the word neyar, meaning radio or record listener,
tv viewer, or magazine reader, to indicate a kind of indiscriminate hear-
ing; presumably such listeners could be turned into racikars (connoisseurs)
if they played the records often enough.
In 1933 E. Krishna Iyer published a guidebook for such listeners. It was
getting more and more difficult to keep any standards in music, he wrote,
in the face of such a ‘‘letting loose on the public of all kinds of radio broad-

fantastic fidelities 265

32 Cover page of Madurai K. Mudaliar’s Gramaphon Sangeetha Keerthanamirdam (1931).
Courtesy of Stephen Hughes.
casts and gramophone records’’ (1933, xx). In his sketches of individual art-
ists, he mentioned those musicians with voices ‘‘particularly suited to the
gramophone’’ (41). In general, women’s voices recorded well; however, the
male vocalist Musiri Subramania Iyer, possessed of a ‘‘rather high-pitched,
sharp’’ voice that was very speedy and flexible, was particularly success-
ful on record. The gramophone seemed somehow able to compensate for
what Musiri’s voice lacked: volume. ‘‘If, along with these qualities, [his
voice] had only a little more volume and innate resonance, how perfect
and enchanting it would be! . . . It is more a sharp pencil, best suited to
draw thin, minute, and sometimes intricate designs of fancy. . . . What a
paradox in voice qualities!’’ (29). Here it is interesting to note how voices
had become conceivable as combinations of different characteristics that
were separable from each other.
The metaphor of gramophone recording dominated Krishna Iyer’s de-
scription of Musiri’s voice; it also crept into his concept of improvisa-
tion. Throughout his sketches, Krishna Iyer used the image of ‘‘hackneyed
grooves’’ to convey the opposite of manōdharma, improvised music (29).
While the ‘‘grooves’’ called up the image of a gramophone record with its
connotations of automaticity and repetition, the lofty term manōdharma,
a Sanskrit compound translatable as ‘‘pertaining to the mind,’’ implied a
sovereign musician setting forth ideas untouched by such influences as
gramophone records. While musicians were to remain untouched by the
gramophone, however, listeners were advised to model their own listen-
ing capacities on the recording capacity of these machines.
Although music critics complained of the deleterious effects of the
availability of so many gramophone records and radio broadcasts, there
was also a sense that, with the right guide, recordings could enhance one’s
experience of music, even beyond that of the musician who had performed
it. Not only did gramophones correct the deficiencies of Musiri’s voice,
but they also made it possible to hear the voice of a young girl: the prodigy
M. S. Subbulakshmi. The success of her first record, cut when she was ten
years old, was such that it convinced even the English recording engineer
that she was something special; he carried one of her records back to En-
gland, where it was well appreciated (Kalki 1954, 21–22). M.S.’s records
became the craze all over South India; indeed, many listeners had their
first education in gramophone records and Karnatic music through listen-
ing to her recorded voice. The celebrated physicist C. V. Raman is said to
have remarked, on hearing M.S.’s voice, ‘‘I won’t say that [she] is singing;

fantastic fidelities 267

she herself has melted and is flowing forth in a flood of sound!’’ (Kalki
1941, 24).18 In a review of one of M.S.’s records, Kalki himself remarked
that ‘‘if you hear it once, you will have the desire to hear it again and
again, a thousand times. Luckily it is a record, and can be played over
and over again’’ (1942, 75). In a review of a record by D. K. Pattammal,
Kalki wrote that ‘‘she finishes madhyamavathi [ragam] almost as soon as
she starts it. Why such a hurry? We feel angry. But then—we are in no
hurry. We can play the record a second time. Indeed, these records are
worth hearing many times’’ (1945, 47). The vanishing and recollection of
music enabled by gramophone records afforded a new kind of pleasure
which became synonymous with the ideal listening experience.19 The rec-
ognition of the perfection of voices came to be linked to their disembodi-
ment and mechanical repetition on record; the pleasure of hearing fleet-
ing music was redoubled by the knowledge that one could hear it again
(and again). The technology of recording had provided a new metaphor
for tradition.

The Radio Renaissance

Once listening to gramophone records had become the ideal model for
listening to music, it was almost natural that radio, another medium of the
disembodied voice, should become the ideal medium for Karnatic music’s
revival. What eventually developed into a large governmental organiza-
tion with broadcasting stations all over India started out as a British colo-
nial project. In 1926 the Indian Broadcasting Company, a private con-
cern, was established in Bombay and Calcutta, but by 1930 had closed for
lack of funds. It later reopened as the Indian State Broadcasting Service
(isbs). Lionel Fielden, the first controller of broadcasting in India, arrived
in India in 1935 to become head of the isbs. But before Indian radio could
become truly a sound to be reckoned with, it needed a new name.

I cornered Lord Linlithgow after a Viceregal banquet, and said plain-

tively that I was in a great difficulty. . . . I said I was sure he agreed
with me that isbs was a clumsy title. . . . But I could not, I said, think
of another title; could he help me? . . . It should be something general.
He rose beautifully to the bait. ‘‘All India?’’ I expressed my astonish-
ment. . . . [It was] the very thing. But surely not ‘‘Broadcasting’’? After
some thought he suggested ‘‘Radio.’’ Splendid, I said, and what beautiful
initials. (Awasthy 1965, 10)

268 fantastic fidelities

The name, commanding in its grandeur and yet ethereal at the same time,
seemed to capture the potential power of radio in India, a medium as
simple and invisible as the air itself but capable of carrying so much.20
Even before All India Radio was established, radio was more advanced
in South India, where it was first conducted by amateur radio clubs, than
anywhere else in India. As early as 1924 the Madras Presidency Radio Club
began broadcasting, with the governor of Madras, Viscount Goshan, as
its patron and C. V. Krishnaswamy Chetty as its founder (Luthra 1986, 6).
When the city of Madras took over the transmitter (located in Egmore) in
1930, the station gave regular broadcasts of music: concerts daily from 5:30
to 7:30 pm, with one Monday each month reserved for Western music;
lessons in music for school students, weekdays from 4:00 to 4:30 pm; and
gramophone records on Sundays and holidays from 10:00 to 11:00 am
(Awasthy 1965, 5).21 Thus, Madras Corporation Radio punctuated its lis-
teners’ lives with music in a new way. In the 1930s a couple who might
have gone to an evening concert now had the choice of switching on the
radio instead. Daily music lessons for schoolchildren could be procured
at much less expense; as one could not reasonably subject schoolchildren
to the temporal demands of the gurukula system, radio presented a solu-
tion: daily music lessons without the wait. Fourteen corporation schools
were equipped with indoor receiving sets for the childrens’ benefit (Lu-
thra 1986, 8). Madras Corporation Radio’s broadcasts in public city parks
and beaches of Madras enabled those who could not afford radios or who
did not believe in them to hear music. It revolutionized the politics of the
Karnatic music scene in Madras by enabling musicians to hear each other’s
concerts without having to go in person and risk losing face. The veena
artist R. Rangaramanuja Ayyangar recalled quailing at the thought of the
mistress of veena, Veena Dhanammal, who was noted for being a purist,
hearing his own recital while sitting in a brougham near Panagal Park in
Madras (Rangaramanuja Ayyangar 1977, 25).22 Madras Corporation Radio
continued to broadcast until 1938, when All India Radio’s Madras station
was established.
From its inception, All India Radio (air) had a reforming mission. If
gramophone records had flooded the market with all kinds of music pro-
duced for commercial gains, air, as a governmental organization, pre-
sented itself as the disinterested arbiter, the decider of standards in music.
In turn, the ideal listener was no longer one who had heard only the voice
of his guru but one who could listen to different musicians and discrimi-

fantastic fidelities 269

nate between them. In search of standards, the air administration, headed
by Lionel Fielden, in 1942 asked its station directors to recruit music pro-
fessors from universities to assist them in the training of radio musicians
(Luthra 1986, 301). In another instance of standard-setting, air banned the
harmonium (ungraciously dubbed ‘‘harm-onium’’ and ‘‘Herr Monium’’ by
its detractors) in 1940, at the insistence of John Fouldes, the director of
Western music for the Delhi station (ibid., 303).
Radio raised not only the problem of standards in classical music, but
also the problem of taste: the possibility that classical music might not be
for everybody. As the arbiter of good taste, air had to represent that taste
even in its broadcast of ‘‘light music.’’ To live up to such a task, the Madras
station in 1943 started a weekly program of new songs, creating a new type
of light music ‘‘genuine and true to our tradition in place of cheap hy-
brid music which threatens to catch the masses and ultimately menace our
musical culture’’ (Luthra 1986, 302). (‘‘Hybrid music’’ referred, of course,
to film songs.) In keeping with the idea of bringing about a renaissance
in Indian music, program journals were published in various languages,
among them Indian Listener (English), founded in 1927, and Vanoli (Tamil)
and Vani (Telugu), both founded in 1938. Meant to make listeners more
knowledgeable about what they were hearing, these journals listed the
programs and included articles about the musicians and music-related top-
ics. These journals continued to be published until they were discontinued
in 1985 (Chatterji 1998, 137–38).
The setting of standards at air blossomed in the 1950s. P. C. Chatterji,
who joined air in 1943 and eventually became director-general, recalled
in his autobiography that music was the most difficult aspect of air’s pro-
gramming. Although most of the airtime was devoted to music, there
were persistent problems relating to the grading of artists by the audition
system, the proper fees for staff artists, and the balance of programming be-
tween classical and light music (Chatterji 1998, 133, 215). Apparently, being
the last word on good taste was not easy when so many musicians were in-
volved; Chatterji suggested that musicians were a particularly difficult lot
because they had learned their art from traditional masters, without the
kind of discipline air needed. B. V. Keskar entered air in 1952 with a strict
program of reform to solve these problems. Where musicians had previ-
ously been selected largely by the program assistant, Keskar introduced a
new audition system with precise rules, in which juries heard musicians
and quizzed them on their theoretical knowledge as well. Strict rules for
the classification of musicians were established, using grades of A, B, or C.

270 fantastic fidelities

In 1958 Keskar banned C-grade artists from broadcasting altogether, effec-
tively ending much amateur participation (Awasthy 1965, 40–43).
Keskar envisioned radio as not merely the broadcaster of music but as
an important popularizer of classical music. To this end, he effected a ban
on film music on air. He instituted the ninety-minute ‘‘National Pro-
gramme,’’ which featured one Hindustani or Karnatic musician per week,
with the idea that both music traditions should be propagated all over
India. In 1952 he created an instrumental orchestra, Vadya Vrinda, to com-
pose and play Indian ‘‘orchestral’’ pieces from notation and to experiment
with Western music. He also introduced the Radio Sangeet Sammelan, an
annual conference about various musical topics convened and broadcast
by air. To compensate for banning film music, Keskar not only initiated
a project of recording folk music but also continued air’s earlier project
of creating high-quality light music with lyrics that met a high standard
set to tunes based on ragas. Keskar’s ban on film music cost air a large
number of its South Indian listeners, who instead tuned in to Radio Cey-
lon, which broadcast Tamil and Hindi film songs. Recognizing this, in 1957
Keskar instituted ‘‘Vividh Bharathi,’’ an All India variety program which
included mostly film music, and some folk and devotional music. Even-
tually ‘‘Vividh Bharathi’’ became its own channel, so that listeners who
preferred classical music and those who preferred light music could have
their own channels (Awasthy 1965, 51). The technology of radio, with its
possibility of different channels that could travel through the air and cross
each other’s paths without ever touching, was in fact quite conducive to
the project of differentiating between classical and other types of music.23

A Clockwork King
If radio was to bring about a true renaissance in Indian music, its discipline
had to penetrate the very structure of the music and the way musicians
thought about it. In his Report on the Progress of Broadcasting in India up to the
31st March 1939 Lionel Fielden claimed that ‘‘listeners who complain of mo-
notony in the programmes are attacking not so much the shortcomings of
the station’s staff as the structure of Indian music itself ’’ (Lelyveld 1995, 52).
Musicians needed to learn how to make their art conducive to radio. Nara-
yana Menon, the former director-general of air, wrote in 1957, ‘‘Broad-
casting . . . will turn out to be the biggest single instrument of music
education in our country. . . . It has given our musicians the qualities of
precision and economy of statement. The red light on the studio door is a

fantastic fidelities 271

stern disciplinarian. Broadcasting has also . . . given many of our leading
musicians a better sense of proportion and a clearer definition of values
that matter in music’’ (1957, 75). What were these values? In the same year,
in a series of special lectures arranged to be read on air, J. C. Mathur elabo-
rated on radio’s gift to Indian music: discipline. Mathur traced a history
of the degradation of classical music in the late nineteenth century and
early twentieth, as the royal patronage of music declined and music passed
progressively through stage dramas, the gramophone, and the cinema. ‘‘It
was doubtful whether high-class Indian music could survive. . . . [I]t was
then that the radio came on the scene’’ (ibid., 97). In the absence of pay-
ing concerts, musicians were given a new lease on life by the radio, which
became like a modern patron. Yet unlike the patrons of yore, on whose
whims the fortune of music rested, air operated by standardized rules.
Such a difference ‘‘signified . . . concretely the changeover from the feudal
concept of the patronage of music to a more modern outlook. No doubt,
in the air-conditioned and remote atmosphere of the studio, the profes-
sional musician misses the direct presence of an appreciative master. But
three decades of the radio habit have perhaps given to most of them a
new sense of communication with their larger audience in thousands of
homes’’ (ibid, 98). Whereas fees had been haphazard, air introduced stan-
dardized payrolls. Even the controversial audition system was beneficial,
Mathur said, because it provided standards by which to judge musicians
and ‘‘subjected them to a sense of discipline’’ (ibid.). Above all, this sense
of discipline came from the musicians’ awareness of the duration of their
performances. This awareness was different from older North Indian musi-
cians’ ‘‘insistence on the so-called time theory of the ragas,’’ in which time
was defined as a quality: time of day or night. The radio treated all time
as a matter of duration, or quantity, within which music could be made to
fit. Radio stations hired ‘‘music supervisors,’’ selected because they could
sing from notation, to advise musicians on how to sing for the radio.24
When musicians’ careful calculation of duration, after decades of radio
broadcasting, turned into habit, the appreciative master in the remote and
air-conditioned studio would be Time itself.
In light of such values, radio began to be perceived as conducive to
music education. In 1947 audiences began to suggest that music lessons
could be broadcast on the radio. Beginning in the Keskar era, air stations
began broadcasting music lessons: a teacher teaching a group of pupils a
particular exercise. Such classes, thirty-minutes or an hour long, were de-
signed not for the pupils but for the radio listeners—particularly ladies,

272 fantastic fidelities

children, and retirees—who could learn from the pupils’ mistakes. Listen-
ing to such lessons would ‘‘help the listener to take note of the essential
points of each lesson in a precise manner, and to benefit from the hints and
suggestions of the teacher as he checks and corrects the faults that appear in
the learners’ performance’’ (Mullick 1974, 40). When air’s program jour-
nals were in operation, they carried the notation of each exercise, so that
listeners could see written before them what they were hearing (ibid., 41).
In the broadcasting station, these lessons were carefully choreographed.
The lessons stuck to composed items like kritis and varnams. The ‘‘stu-
dent’’ would receive a notated copy of the composition to be taught to
practice beforehand, and teacher and student would rehearse the compo-
sition once. In order to promote fairness, the teacher and the student could
have had no previous connection to each other. Under no circumstances
could they be guru and disciple, because this would introduce a bias or
monopoly not in keeping with air’s mission of ‘‘encouraging all styles.’’ 25
In his book The Teaching of Music P. Sambamoorthy included a section
on ‘‘teaching music from a broadcasting station,’’ the benefit being that
songs could be taught to ‘‘millions of listeners at the same time’’ (1966,
244). Radio journals that published notation were especially useful in this
regard, he remarked. But most important was the exquisite timing such an
endeavor required. Students would have to be ready in their places in front
of the receiving set at least five minutes prior to the scheduled broadcast
time. Once the broadcast started, the students had to make sure to repeat
the phrase as sung by the radio teacher in the same exact tempo. ‘‘If the
listeners slacken their speed of singing, there will arise the danger of the
teacher’s subsequent phrases commencing even before the listeners com-
plete the previous phrases. If the pupils sing in an accelerated time, there
will necessarily ensue a brief period of silence in between the conclusion
of their music and the subsequent music from the teacher. Absolute preci-
sion in rhythm should be maintained to guard against such errors’’ (245).
Unlike the exchange between a guru and sisya, in which the progress of
the ‘‘lesson’’ depended on the guru’s mood and satisfaction, the progress of
the radio lesson was predetermined by the time allowed for the program;
all that remained for the student to do was to fill up a series of preset gaps.
The Madras station of air broadcasts a daily music lesson after its morn-
ing broadcast of Karnatic music. The lesson lasts about thirty minutes and
features a teacher and a single student. One kriti is taught in each class, with
each line repeated until the student gets it right. The program, broadcast
at 8:30 am, is obviously meant for housewives who are usually doing their

fantastic fidelities 273

cooking for the day around that time and perhaps for schoolchildren.26
These lessons depart from the conventions of the gurukula system in sev-
eral ways. To learn a full kriti with one’s guru would take several days at
least, perhaps even a month. Thus, the radio classes radically compress the
amount of time it takes to learn. At the same time, learning by focusing on
one kriti from beginning to end is different from the process of learning
with a guru, where in a typical day one might learn one line of one kriti, a
few lines of a varnam, or simply sit listening to the guru sing raga alapana
for some visitors. The learning process changes from one of inadvertent
absorption to one of conscious drilling.
At the same time as the radio brings music into the home, radio classes
introduce a peculiar, perhaps comforting, quality of distance. They focus
the music on compositions rather than improvisation. Radio makes it pos-
sible to learn from others’ mistakes instead of one’s own; it saves one the
socially complicated process of finding a teacher; it spares the student from
having to hear from the guru, ‘‘You are not ready to learn this.’’ The re-
moval of the radio student from the scene of teaching offers a kind of
perspective he would not have from within it. His identity is oddly aug-
mented, for now he hears not only the voice of the teacher but the voice
of a student repeating the teacher; it is as if he can step back (or simply
stay home) and listen to himself learn. Radio classes paradoxically make it
possible to learn without being ‘‘in the very act.’’
Radio, with its punctilious schedule of broadcasting, the very model
of discipline, also guarantees that things will come to pass. It insures that
the music class will proceed in a timely fashion and end after the required
thirty minutes, whereas a guru might refuse to teach her student even the
next line of a kriti for months if the first line had not been perfected. My
own violin teacher measured her authenticity as a real guru by her refusal
to make her lessons fit into prearranged schedules, whether the scheduled
time of our lessons or the class schedule of the music college at which
she taught. It was not that there was ‘‘no sense of time’’ in the gurukula
system but rather that time and progress were measured according to the
guru and the specific sisya rather than according to a standardized sched-
ule or curriculum. Jean Langford articulates this difference in reference to
the institutionalization of Ayurvedic medicine: ‘‘Timetables were prob-
ably also followed to some extent in gurukulas. There is, however, an im-
portant distinction to be drawn between the more monastic discipline of
the gurukulas, in which the rhythm of activities was directed by the au-
thority of teachers and texts to perfect the student’s mind, and a modern

274 fantastic fidelities

discipline, in which time is regulated according to a rational hierarchy of
grade levels to subject the student’s body’’ (2002, 123). In this respect, the
administration of air self-consciously takes the place of the guru.

Gurukulavasam Is Dead
‘‘The gurukula system collapsed around 1900,’’ observed the musician and
scholar R. Rangaramanuja Ayyangar in his caustic 1977 autobiography.27 ‘‘I
awoke, as if from a dream, to realize that elaborate and scientific notation
was the only means. . . . For the gurukula system and ear and rote learning
had been laid to rest long ago’’ (1977, 10). The repetition of music enabled
by gramophone records was seen as a feature of modernity and science,
while the ‘‘rote repetition’’ associated with the gurukula system came to
be seen as the opposite of all that was modern and scientific. The fact that
Rangaramanuja Ayyangar here refers to repetition as ‘‘rote’’ is significant,
for it implies a shift in attitude toward a musical practice which had pre-
viously been considered one of the benefits of gurukulavasam. Repetition
of a phrase or lesson, in the context of gurukulavasam, was not simply
repeating the same thing over and over again but a process by which musi-
cal knowledge could be embodied, a valuable accumulation of time spent
with one’s guru. For instance, in the 1998 documentary Khandan the North
Indian sitarist Shujaat Khan talked about his early musical training under
his father, suggesting that there was a relationship between the experi-
ence of such repetition and being able to improvise. Khan recalled the
hours spent with his brothers and cousins, in which they all played the
same phrase over and over again along with his father, as an essential part
of his musical education. The importance of repetition, defined positively
as a kind of process of embodying musical knowledge or as an accumula-
tion of time that one can later recall when improvising on stage, was clear.
Khan gave little sense of repetition as a mindless waste of time, as the word
rote implies.
Rangaramanuja Ayyangar’s reference to all repetition as rote learning
indicated the degree to which repetition had come to be seen as a threat
to Karnatic music by those interested in its reform. Keskar wrote in 1967
that rote repetition was not only responsible for the ignorance of music
theory and history among Indian musicians but also for the ‘‘distortion’’ of
the music itself: ‘‘Music was learnt from guru to sisya. This led to a gradual
distortion and change which is inevitable when anything has to be handed
down through the medium of the human voice which cannot copy any-

fantastic fidelities 275

thing faithfully’’ (Keskar 1967, 38). Note here how the status of the voice
has been changed from source to medium, and only one medium among
others at that. Keskar went so far as to say that gurukulavasam would even-
tually ruin Indian music, since the laboriousness of the gurukula system
had led to a kind of almost religious devotion on the part of sisyas to their
gurus (55). ‘‘Guru-dom’’ had attained a peculiar kind of glamour through
such ‘‘fanaticism.’’ 28 The result was ‘‘complete chaos,’’ in which ‘‘the law
of the jungle’’ prevailed and any musician who wanted could set up his
own school and call himself a guru (29). ‘‘Gurus have become as plentiful
as wayside flowers,’’ remarked Keskar with distaste (71).
A distinction had to be made, Keskar urged, between performers and
teachers. Unlike a performer, who had only to be captivating on stage, a
teacher did not himself need to be gifted at performing. But it was essen-
tial that a teacher be able to explain and repeat when necessary. ‘‘He must
make [the student] repeat musical sequences, point out the mistakes, and
make him do that again and again’’ (1967, 17). Repetition was only a means
to achieve the correct version, rather than a valued process in itself. The
ideal teacher sounded, literally, like a gramophone, one who could dis-
passionately reproduce and explain different styles to the student, with-
out being ‘‘in the act.’’ Sambamoorthy described the qualities a teacher
should have as being quite distinct from those of a guru. Whereas gurus
were known for their mercurial temperaments and their genius, teachers
needed to be more even-tempered and less exacting. In the gurukula sys-
tem there was no separation between a ‘‘teacher’’ and a ‘‘curriculum’’ to be
taught; the guru embodied his or her musical knowledge, and one learned
music by emulating one’s guru, by becoming absorbed in him or her. By
contrast, the teacher’s authority was simply that of someone authorized to
teach a preset curriculum. Accordingly, interactions between the teacher
and students had to be subject to certain preset standards as well.
A lengthy chapter in Sambamoorthy’s The Teaching of Music is devoted
to the particular problems a teacher giving exams would have to face, be-
ginning with the ‘‘irresistible desire’’ to correct pupils’ mistakes and the
necessity of curbing that desire in order not to discourage the student or
run out of time for examining other pupils (1966, 212–13). The examina-
tion, as a genre, was closely related to the radio audition; both were new
contexts of musical performance that arose with sound recording. As in
the audition (and eventually the broadcast), in the music exam the stu-
dent’s ability to produce music on the spot and in a limited amount of
time was under scrutiny, not his ability to emulate or support his guru

276 fantastic fidelities

in performance. Part of the teacher’s training lay in being prepared for all
the possibilities that such a scenario might produce, for, as Sambamoorthy
noted, ‘‘the examination has its own psychological effect’’ (212). The exam
had the potential to produce all kinds of unexpected sounds instead of a
faithful reproduction of the teacher’s lessons, and knowing these possible
sounds was as important for the teacher as knowing music.

The mere thought of an examination is sufficient to make the voices of

even steady singers shaky. . . . The music teacher comes across voices of
varying types and grades of excellence, from the unpliable stony voice
to the ringing silvery voice. . . . The singing of some creates the im-
pression that they are vomiting something. . . . [T]he akarams of some
are akin to gargling. . . . Things are no better in instrumental examina-
tions. The bowing of some [on the violin] is so repulsive and harsh that
at the end of the performance one can find heaps of hair that have come
off the bow, stealthily removed and strewn behind the performer. . . .
Some go on aimlessly and artificially moving their fingers up and down
the fingerboard. . . . Their play impresses one as the mewing of a cat
and one need not be surprised if a cat in the neighbourhood begins
to blink and search in vain for the new formless member of her kind.
(214–15)

Unlike a guru or performer (or the neighborhood cat, for that matter),
a teacher, presumably primed by Sambamoorthy’s manual, would remain
unfazed by such sounds and be able to move on.

The Art of Listening

If the technologies of recording and radio promoted a musical aesthetic
based on the separability of functions, like the distinction between per-
forming and teaching, it followed that the musician could not be consid-
ered the best judge of his or her own music. For Keskar this role belonged
to the listener, who could judge music precisely because he was not in the
act of playing or singing it: ‘‘Good critics and listeners are the foundation
of music. . . . There is an illusion prevailing today that the musician is
the best judge of music. . . . But what is music without listeners!’’ (1967,
42–43). But the kind of listening Keskar had in mind was one in which
audiences would have to be trained; this new kind of listening could be
created by the ‘‘scientific’’ teaching of music. The true listener observed
‘‘pin-drop silence’’ so as to hear ‘‘all the nuances’’ of the music. ‘‘A musi-

fantastic fidelities 277

cal performance is a story in sound,’’ Keskar wrote. ‘‘All its nuances have
to be heard carefully in order to enjoy it. . . . The pin-drop silence, the
rapt attention and rigid discipline that one observes in the audience in the
West demonstrate that they know how to respect music and the way to en-
joy it. Our concerts only show that we have not learned fully or probably
forgotten the art of listening’’ (ibid., 23). In The Teaching of Music Samba-
moorthy wrote that the music teacher was responsible for cultivating the
right kind of listening. The listener was, above all, not to cause any inter-
ference with the performance. ‘‘The music teacher is in a good position
to educate the people as to their duties in concert halls. They must be told
that when a concert goes on, they must listen to the music with attention
and not disturb their neighbours by talking to their friends. If they desire
to leave the hall, they must do so only in between items and not when an
item is actually going on’’ (Sambamoorthy 1966, 35).
The kind of listening Sambamoorthy advocated went beyond matters
of decorum, however. It had to be achieved through a careful education
in ‘‘music appreciation,’’ one of the elements of the new institutionalized
music curriculum. Focusing on developing the students’ ‘‘powers of criti-
cal observation and critical hearing,’’ music appreciation would produce
‘‘musically cultured’’ people (Sambamoorthy 1966, 165). It was essential
that the composition chosen for ‘‘appreciation’’ be able to be repeated,
either sung at least twice by the teacher or, even better, played on a gramo-
phone record so that the teacher could point out nuances as they were
heard (165). Sambamoorthy prepared an exhaustive list of 216 ‘‘hints and
points for developing appreciation essays,’’ which could be used to guide
students, to teach them a particular technique of listening. What exactly
were students being taught here? If one looks closely at these points, one
finds that they focus almost exclusively on the structural and prosodic fea-
tures of compositions, the idea being to contemplate the composition as
an aesthetic object, detached from any particular performance of it. ‘‘Ap-
preciation,’’ in this context, meant a close, dissective attention rather than
an appreciation of the emotional quality or momentum of a performance
or its effect on other listeners.
The art of listening, thus conceived, was modeled on the scene of an
individual listening to a gramophone recording. Listening was the art of
hearing a composition without interference from other audience mem-
bers or even from the performer. Pin-drop silence set the stage for a trans-
mission free of distortion. Such transmission became the model of sampra-
dāya, a term that refers to tradition passed down a carefully traced lineage

278 fantastic fidelities

of disciples. In an essay on music T. V. Subba Rao translated the term as
‘‘faithfulness to tradition’’: ‘‘Music must be heard as it comes from the
mouth of the teacher and the exact form as presented should be grasped’’
(1962, 227). Learning music from books produced music that was ‘‘shape-
less and grotesque’’; only music learned from a teacher was authentic. Yet
this mouth-to-ear transmission was to be aided by modern technology.
‘‘Books, charts, the blackboard, printed notation, even recorded music
are no substitute for the living presence of the guru,’’ he wrote (ibid.,
232). Yet his use of the conciliatory ‘‘even’’ before ‘‘recorded music’’ sug-
gested that for him recorded music had greater proximity to the elusive
gurukulavasam. Indeed, he wrote, ‘‘in practical music, the only library
worth mentioning is a collection of good recorded music’’ (ibid., 232–
33). Recordings could be made to ‘‘disseminate correct knowledge’’ in
classroom settings. Students learning from a guru could ‘‘reduce to nota-
tion as a sort of memorandum’’ the songs learned ‘‘to ensure against their
being forgotten altogether’’; for this purpose, he remarked, a tape or plate
recording would be even more satisfactory (ibid., 235). Likewise, Samba-
moorthy, in his list of ‘‘equipment for the music class,’’ recommended a
gramophone with a set of records, a radio set, and models of the larynx
and ear as second in importance only to musical instruments themselves
(1966, 30).
Being an educated listener meant above all knowing how to listen to
‘‘music’’ and to treat the gramophone or radio merely as media for music’s
reproduction. Without developing such listening technique, it became
impossible to enjoy what was on the radio; one could only hear the radio
itself. In a humorous article from about 1940 entitled ‘‘Rēṭiyo Kaccēri’’
(Radio concert), a certain V. Muthuswami described the results of his de-
cision to purchase a radio. ‘‘Those like me who are crazy about music have
to thank the establishment of A.I.R. for providing us the chance to hear
so many musicians,’’ he began. ‘‘To enjoy such concerts, you must have
a radio, or your neighbor must have one, and he must also have caṅkīta
ñānam [musical knowledge/taste].’’ Unfortunately, Muthuswami’s neigh-
bor, who owned a radio, did not have such taste. ‘‘That day a famous singer
was singing Hamir kalyani. My luck! My neighbor immediately turned
off the radio and put on gramophone records of film songs. Even though I
have iron ears, I could not endure this!’’ For Muthuswami the knowledge
that there was good music on the radio that he could not hear was torture.
He set his mind on buying a radio and told several friends about it. ‘‘It
was as if my decision had been broadcast. . . . When we got to the store, I

fantastic fidelities 279

had only to say one word and a radio sales agent came running out to help
me, like a white ant from its nest.’’ Determined to get a radio in time for
that evening’s concert, Muthuswami purchased a radio and aerial and had
it installed. By 5:00 pm, three hours before the concert, everything was
working. Muthuswami was looking forward to enjoying the radio con-
cert, when, at 7:45 pm, a crowd of his friends showed up, having somehow
found out about the new radio. ‘‘Exactly at 8:00 the singer started with
a run through nattai ragam and then the kriti ‘Sarasiruha.’ I should have
known the rareness of this moment. For two minutes I was immersed in
the music. But can pure music be without interruptions? My friends soon
began their own concert, not even having the propriety to be quiet while
the music was on.’’
Muthuswami’s friends, it turned out, were obsessed with the radio
itself. ‘‘Hey, is the aerial on the roof ? Permanent installation?’’ ‘‘What was
the list price? Was there a commission?’’ ‘‘Look at the tone quality. The set
is very neat, sir!’’ ‘‘Is it four valves or six valves?’’ ‘‘Does the short wave come
clearly? Do Bombay and bbc come?’’ ‘‘Hey, Columbo is a very power-
ful station.’’ ‘‘Hey, if you have just bought this radio for three-hundred
rupees and then you want to get another station, do you have to go buy
another set?’’ ‘‘Is it working correctly, sir? We keep hearing about Indian
Bank.’’ ‘‘Sure, it’s okay. That’s just local disturbance.’’ Muthuswami’s friends
kept on with such comments so that it was impossible to hear the music.
Then his child began to cry. By the time that settled down, the concert
was coming to an end. ‘‘Kacceri close!’’ sang one of his friends gleefully.
Muthuswami concluded gloomily, ‘‘I sent them off with their tampulam.29
From that day, I have kept my radio set locked in a closet. When there is
a good concert on the radio I beg my neighbor to turn it on. If there are
any crazies who want my new radio set, let me know!’’ For Muthuswami’s
friends, who lacked the listening technique that would enable them to
ignore the mechanism itself, the radio had made music into a matter of
prices, sound waves, and frequencies, to such an extent that if the radio was
playing it was impossible to hear the music (Muthuswami n.d., 279–80).
If authenticity had come to be seen as analogous to high-fidelity re-
production, it is not surprising that T. V. Subba Rao resorted to another
technological metaphor to get at the ineffable concept of ‘‘inspiration.’’
‘‘The mind of man,’’ he wrote in an essay on music, ‘‘is like the receiving set
of a radio which when properly tuned enables us to hear the transmission
from a broadcasting center. The Eternal is forever radiating knowledge
and bliss for those who by self-discipline have made themselves worthy

280 fantastic fidelities

to receive them’’ (1962, 230). Instead of disenchanting the world of music,
technology and science re-enchanted it. The singer no longer sang with his
voice but with his larynx, ‘‘the divine vocal instrument . . . [that] by a pro-
fundity almost mysterious is calculated to stir us to our very depths’’ (228).
A systematic course of voice culture would have to pay attention to the
fact that ‘‘tones are produced by the vibration of the chords in the larynx,
but no note can be pleasing unless it is rich in components. To secure this
end, the note must be fully resonated. The cavities of the chest and the
abdomen should be made to take their part as sound-boxes for the note’’
(228). In Keskar’s imagination, the voice had become simply one medium
among others; in Subba Rao’s it became a mechanism or instrument de-
scribable in the same terms as mechanically reproduced sound.

Fantastic Fidelities
In January 1998 I was heading by bus from Madras to Tiruvaiyaru in the
company of a young computer software engineer who, although he had
studied no music, was quite interested in the fact that I had come to India
and was studying under a guru in ‘‘the traditional way.’’ He informed me
that he had a ‘‘thesis’’ he was trying to prove: that India was once great in
science and engineering but had lost that knowledge at some point and is
now only ‘‘the top’’ in spiritual matters. As proof, he offered the gurukula
‘‘system,’’ which seemed to be an ancient, outmoded form of teaching but
was in fact most advanced in the way it used ‘‘psychology’’: the guru comes
to know the ‘‘psychological makeup’’ of the sisya and therefore can figure
out the best way to teach him or her. With his invocation of psychol-
ogy, the engineer connected ancient India with the sciences of the West,
making gurukulavasam at once premodern and ultramodern. Gurukula-
vasam thus becomes a symbol of India’s past greatness, a greatness that,
although it can be assessed in terms of Western sciences, exceeds them.
The engineer’s thesis is interesting because it reflects the way in which
gurukulavasam is constructed as a sign of Indianness in a postcolonial dis-
course that demands commensurability with the West. Gurukulavasam is
imagined as that which makes Indian music both equal to the music of
the West (it, too, has a ‘‘scientific’’ teaching method) and superior to it (it
is a system that manages to be scientific and personal at the same time).
In such late-twentieth-century discourse gurukulavasam overcomes the
opposition between the elements that Raghava Menon felt to be irretriev-
ably separated from each other: the message of the human voice and the

fantastic fidelities 281

mathematically explainable components of its sound. The sisya produces
a faithful reproduction of the guru’s music through a process that is not
mechanical. Imagined as the peculiarly Indian ‘‘system’’ by which musi-
cians are formed, gurukulavasam would seem to be the perfect union of
the faithful reproduction identified with machines and the human em-
bodiment identified with an oral tradition.
Ananda Coomaraswamy was worried in 1909 that Indians would get so
accustomed to listening to copies of Indian music that they would lose the
real thing. The gurukula method, which, according to Coomaraswamy,
did not allow the intervention of any mechanism between a guru and
his sisya, seemed to remain unknowable by and to Western technology.
For him, as for others, gurukulavasam thus preserved what was Indian
about Indian music: its ‘‘oral tradition.’’ Confronted with technologies
of sound reproduction, Coomaraswamy presented gurukulavasam as an
Indian alternative for reproducing music. In doing so, he made a com-
parison, however unwittingly, that would continue to surface in debates
about what constituted authenticity and tradition.
In the course of this chapter, we moved from Coomaraswamy’s railing
against the gramophone to the idea of the voice as simply one medium
among others for reproducing music. Sound recording and radio were per-
ceived as enabling a separation from ‘‘the very act’’ of singing or playing
music, a separation that came to stand for a peculiar kind of authenticity.
Listening to one’s performance after the act was the most authentic way of
judging it and appreciating it. Separated from the act of its performance,
the music on the gramophone record could be played repeatedly and, un-
like the Karnatic musician, would sound exactly the same. Repetition be-
came a feature admirable in machines but grotesque and mechanical in
musicians. The idea of learning by rote came to seem antiquated, unten-
able in the modern age, even threatening to good music. But although
‘‘learning by ear’’ was regarded as the most authentic way, the ear could not
be trusted by itself; it had to be supplemented by a recording to avoid for-
getting. The mechanism intervened between guru and sisya; meanwhile,
gurukulavasam, the very mechanism by which tradition worked, came to
be imagined as a kind of high-fidelity reproduction.
Gurukulavasam and technologies of sound reproduction have thus
served as the interlinked grounds for twentieth-century elaboration of
ideas about authenticity and tradition in Karnatic music. Rather than nar-
rating the takeover of a ‘‘traditional’’ method of teaching by ‘‘modern’’
technologies, this chapter has shown how ideas about musicality, repeti-

282 fantastic fidelities

 33 Advertisement
for sruti boxes,
electronic tamburas,
and talometers in
Sruti ( June/July
1990).

tion, listening, authority, tradition, and gurukulavasam itself were instead

formed in the encounter with such technologies. This is not to claim that
gurukulavasam never existed or that it is simply a discursive construct. It
is, rather, to suggest that gurukulavasam, now reified as a ‘‘system’’ and
retrospectively constructed as ‘‘tradition,’’ is central to a discourse about
authenticity and music that is peculiarly modern.
At the center of this discourse, which links mechanical perfection to a
sense of the ineffable, are sruti boxes, electronic tamburas, and talometers,
machines that perform only one function but do so with absolute fidelity.
Such machines are enchanting because they possess, as an advertisement
from 1990 put it, ‘‘a range any vocalist would envy.’’ Sruti boxes and elec-
tronic tamburas are electronic devices, plugged in or battery-operated,
which provide the sruti, or drone, that forms the tonal background neces-
sary for most Indian classical music.30 They take the place of either a small
harmonium with bellows that would have to be manually pumped or a
tambura, a long-necked lute with four strings tuned to the drone pitch
that would be plucked slowly and steadily through an entire performance,

fantastic fidelities 283

 34 ‘‘Concerto’’ model
electronic tambura. Radel
Systems brochure (2002).

usually by the main musician’s disciple. In this sense, then, these machines
substitute for the sound as well as for the disciple. They have become such a
fixture in South Indian music concerts that even when a musician has a real
tambura player (who always sits unobtrusively somewhere in the back of
the stage), he or she will usually also have a sruti box or electronic tambura
turned up to full volume and placed directly in front of the microphone.
The first electronic sruti boxes, made in 1979, were brown, heavy, and
boxy, with a series of adjustment knobs that made them look like old
radios; meant to be the electronic equivalent of the bellows-operated
sruti box, they provided a constant tone, with only the pitch or volume
being adjustable.31 Since the early 1980s, however, sruti boxes have given
way to electronic tamburas that are becoming increasingly sophisticated.
Radel, the foremost manufacturer, now offers no fewer than nine differ-
ent models, with color choices of red, black, white, and gray, some small
enough to fit in the palm of one’s hand. One can now control volume,

284 fantastic fidelities

pitch, tempo, and tonal quality, and many models are equipped with a
‘‘memory’’ for what has been programmed in previously. All of these use
sampling technology to create a ‘‘realistic’’ imitation of a tambura being
plucked; however, they have a much louder sound and fuller tone than
any real tambura. Unlike a real tambura, these electronic tamburas can be
tuned to any vocalist’s pitch, making them infinitely versatile.32
The twenty-first century has not produced a musical computer like the
fictional Yakshani. Electronic tamburas, however, come close to Yakshani
in the kind of affect they elicit; both provide the fascination of hearing
music from an inanimate source, its intimation of the supernatural. My
music teacher often talked about a great flutist she had accompanied in
concert many times, whose genius she partly attributed to his habit of
sleeping with a sruti box turned on under his pillow. In the summer of
2001 I happened to be around when a friend of mine who is a Karnatic
vocalist showed his latest-model electronic tambura to a visitor. He re-
moved it from its case, placed it ceremoniously on the floor facing both
of them, and turned it up to full volume. As its sound saturated the room,
he raised his eyes skyward and exclaimed, in a mix of Telugu and English,
‘‘Aha! Divine-ga undi, kadā? [It’s divine, isn’t it?] I hear this and immedi-
ately I’m in the mood to sing. It pulls the music out of me. I could sit here
with it for days.’’
Radel’s latest model is a digital tambura in the shape of a tambura—
the ‘‘concerto’’ model.33 The sampling technology that produces the ‘‘real’’
tambura sound is housed inside a hand-crafted wooden tambura body
that, unlike a real tambura, stands upright without needing to be held—
a tambura that plays itself. The concerto model takes the logic of fidelity
to an extreme; it pairs fidelity to the real tambura sound with fidelity to
the very image of tradition—but without the sisya. Like Yakshani, it re-
places music with frequencies and sound waves, the devoted disciple with
the memory and repetitive capacities of a computer. Like Yakshani, who
must undergo gurukulavasam in the end, the sruti box remains under the
control of the human singer, but only after the logic of fidelity—continual
and mutual imitation—has questioned whose is the master’s voice.

fantastic fidelities 285

$ Afterword
modernity and the voice

Modernity has been described and theorized largely in terms of the visual.
Vision, it has been argued, is the sense that comes to have priority and
power over all the others; in Western metaphysics, knowledge is equated
with vision and truth with light. David Levin’s edited volume Modernity
and the Hegemony of Vision traces the ocularcentrism of Western meta-
physics from Parmenides to late-twentieth-century television, asking,
‘‘How is the ocularcentrism of modernity different from that which pre-
vailed in earlier ages?. . . . How has the paradigm of vision ruled, and
with what effects?’’ (1993, 3). Michel Foucault’s description of the panopti-
con and its accompanying regimes of surveillance provided one answer
to these questions, suggesting that modernity is characterized by a par-
ticular regime of visibility that is tied to control (Foucault 1977). Timothy
Mitchell, in a similar vein, has argued that modernity is characterized by
representation: the world-as-picture. ‘‘The peculiar metaphysic of moder-
nity,’’ he argues, involved the creation of the effect of reality by making
a distinction between image, meaning, or structure, and the ‘‘really real’’
(Mitchell 2000, 16–20).
Recognizing the way vision is tied both to power and to the power to
represent is central to developing a critique of modernity. A fuller cri-
tique, however, must also consider how modernity positions other senses
in relation to vision. Aurality, in particular, has a central place in moder-
nity’s self-definition. Within modernity, the aural has been positioned as
the premodern or the nonmodern, often as that which escapes modernity’s
controlling regimes of knowledge-power. The idea of the aural as the pre-
modern or nonmodern is encapsulated, as Jonathan Sterne suggests, in
what he calls the ‘‘audio-visual litany,’’ a series of seemingly commonsense
oppositions often taken as natural, biological, psychological facts about
the senses: that hearing immerses its object, while vision offers a perspec-
tive; that hearing is concerned with interiors, vision with surfaces; that
hearing is about affect, vision about intellect; that hearing immerses us
in the world, while vision removes us from it. As he suggests, the audio-
visual litany idealizes hearing and, by extension, speech, music, and the
voice as ‘‘manifesting a kind of pure interiority,’’ while it simultaneously
‘‘denigrates and elevates vision. As a fallen sense, vision takes us out of the
world. But it also bathes us in the clear light of reason’’ (2003, 15). Such
statements assert that the senses lie outside of history: that they and their
attributes are natural, biological givens. Much psychoanalytic writing on
the voice, for instance, depends deeply on the audiovisual litany, endowing
hearing with universal significance and the voice with primordial powers.
The idea that aurality and orality are closer to lived, embodied experience
but are nonrational or pre-analytical faculties is pervasive in the ideas and
methodology of many academic disciplines.
In asserting the transhistorical nature of the senses, the audiovisual
litany elides the way that these oppositions between visuality and au-
rality are grounded in notions of the difference between the modern and
the nonmodern and in turn mapped onto the West and the non-West.
If modernity, as Timothy Mitchell suggests, asserts its existence and au-
thority by continually staging the difference between the modern and the
nonmodern, the West and the non-West, oppositions between the aural
and the visual have served as a powerful means of naturalizing these dif-
ferences. Modernity is characterized just as much by particular ideas about
aurality and the voice as it is by certain conceptions of visuality. Indeed,
the audiovisual litany is what enabled India to become the ‘‘kingdom of
the Voice.’’
The valorization of the voice in Karnatic music is part of a distinctly
modern set of ideas about music, the self, and Indianness. The voice, in
twentieth-century discourse about this music, has been figured as a realm
of pure Indianness, untouched by colonialism or worldly concerns, the
element that gives the music both its distinctive Indianness and its ability
to exist as part of a continuous ‘‘civilizational’’ tradition, unaffected by his-
tory. Within the logic of modernity, the voice is imagined in two ways: as
a sign of the modern, interiorized self and as the authentic embodiment
of oral tradition. On the one hand, the voice is imagined as endowing
the modern subject with interiority, while on the other hand the voice is
imagined as endowing the premodern subject with the essential difference
that will make him an authentic representative of tradition. Such a dual
way of imagining the voice is central to the politics of voice I have iden-

afterword 287
tified here, in which ideas about modern singing and listening subjects
are intertwined with anxieties about the proper control, preservation, and
use of oral tradition. This is Karnatic music’s postcolonial predicament: in
order to be a true classical music, it must be performed and listened to by
modern subjects who know how music should affect them and what it
means to be an artist, but it also needs to be able to point to an origin in
an oral tradition that maintains its distance, and difference, from the West.
It is tempting to view the twentieth-century emergence of classical
music in South India as part of the development of a local or alternative
modernity, one that borrows the Western concepts of ‘‘classical music’’
and ‘‘art’’ and many of their entailments. However, while it is important
to recognize the multiple origins of what can often be too easily unified
under the name of modernity, the idea of ‘‘alternative modernities’’ can,
as Timothy Mitchell states, dangerously ‘‘imply an underlying and fun-
damentally singular modernity, modified by local circumstances into a
multiplicity of ‘cultural’ forms. It is only in reference to this implied ge-
neric that such variations can be imagined and discussed’’ (2000, xii). I
have resisted using the concept of alternative or local modernity in order
to emphasize that the modern idea of classical music—in both the West
and South Asia—was enabled, materially and ideologically, by colonial-
ism. In both Europe and South India, the institution of classical music
was a product of the new economies and patronage structures associated
with imperial modernity.1 Western classical music was not a fully formed
entity that was merely exported to India beginning in the early twenti-
eth century; rather, ideas about classical music, both Western and Indian,
were being negotiated simultaneously in the colony and the metropole.
In this sense, Karnatic music and Western classical music belong not to
alternative modernities but to the same modernity. This is not to provide
yet another instance of the globalization of an essentially Western moder-
nity. Rather, it is to recognize that, as Mitchell states, ‘‘when themes and
categories developed in one historical context . . . are reused elsewhere
in the service of different social arrangements and political tactics, there
is an inevitable process of displacement and reformulation’’ (ibid., 7). The
politics of voice I have described in this book constitutes such a reformu-
lation, one in which the priorities of the audiovisual litany are reversed
(but essentially unchallenged) in the service of determining what makes
Indian classical music Indian.
The electronic sruti box is in many ways similar to the mimetic violin
with which I began this book. Both are ‘‘accompaniments’’ to the voice

288 afterword
which are much more than accompaniments; they play a crucial role in
the way the voice is staged and produced. As such, they encapsulate the
argument of this book: that voices are created as much from without—by
instruments, by technologies, by audiences, and by social forces—as from
within. The sruti box and the violin denaturalize the voice; they remind
us that if we take the voice as the originary model of all musical produc-
tion and as a natural means of expression, we miss the ways in which the
voice itself—as musical sound and as culturally elaborated metaphor—is
historically locatable.
In identifying a politics or ideology of voice that includes both the ma-
terial and sonic aspects of the voice and ideas about its significance, I hope
to have elucidated and thus dislodged some of modernity’s most cherished
assumptions about the voice: that the voice is a ‘‘natural’’ expression of
self; that agency consists in ‘‘having a voice’’; that the sonic and material
aspects of the voice are subordinate to its referential content or message;
that non-Western cultures are profoundly and naturally oral. What I have
suggested instead is that the complexity of how musical and literal voices
are created, heard, and interpreted matters in the way voice as a concept
takes on significance. In this sense, ideas about the voice as a metaphor or
trope of subjectivity, agency, power, and authenticity are bound up with
the particular moments when, and the very material ways in which, voices
become audible.

afterword 289
$ Notes

Introduction
1 Kula means ‘‘family,’’ or ‘‘lineage’’; vacam means ‘‘living.’’ Gurukulavasam can be
translated as ‘‘living with the guru’s family.’’
2 As a body of knowledge produced by texts and institutional practices, Ori-
entalism was responsible for generating essentializing statements about the
Orient that relied on contrasting it to Europe (Said 1979). In the Indian con-
text, Bernard Cohn’s ‘‘Notes on the History of the Study of Indian Society and
Culture’’ (1968), Ronald Inden’s ‘‘Orientalist Constructions of India’’ (1986),
and Gyan Prakash’s ‘‘Writing Post-Orientalist Histories of the Third World’’
(1990), among others, include discussions of the kinds of essentialisms pro-
duced through Orientalist scholarship on India.
3 T. V. Subba Rao, a practicing lawyer in Madras, sat on the ‘‘committee of ex-
perts’’ that founded the Madras Music Academy and became its first treasurer.
4 For a general account of the politics of this ‘‘revival’’ see Subramanian 1999b.
5 The idea of a classical music tradition, William Weber suggests, originated in
England and can be linked to the decline of the courts and the rise of an urban
leisure class enabled by the industrial revolution and colonial expansion (1992,
1–9). The concept of classical music, then, is not incidental to colonialism but
part of its economy.
6 According to Daniel Neuman, the Hindi term sāstriya sangīt (literally, ‘‘sas-
tric music,’’ or music that conforms to the rules of the sastras, a set of San-
skrit musical treatises) is used in North India and occasionally in the South
to refer to Hindustani or Karnatic classical music (personal communication,
June 1995). This term is often given as evidence that the notion of ‘‘classical
music’’ is indigenous to South Asia. However, the connotations of sastric and
classical are quite different. There is also the possibility that sāstriya sangīt is not
an ‘‘original’’ term of which ‘‘classical music’’ is the English translation, but
rather a term coined as a Hindi translation of the English ‘‘classical music.’’
7 Often cited in this regard is the Sangītaratnākara, written in about 1240 AD.
8 These associations have been famously formulated by Walter Ong in Orality
 and Literacy (1982). Bauman and Briggs provide a genealogy of the orality/
literacy opposition (2003, 104–7). Michel de Certeau notes that the field of
ethnology centers around the opposition between orality—the voice defined
in terms of exteriority and alterity—and writing: ‘‘The vocal exteriority is
also the stimulus and the precondition for its scriptural opposite. . . . The sav-
age becomes a senseless speech ravishing Western discourse, but one which,
because of that very fact, generates a productive science of meaning and ob-
jects that endlessly writes’’ (1988, 236).
9 Veit Erlmann has made similar claims in service of the idea that ‘‘South Africa’’
and the ‘‘West’’ have co-constructed each other in colonial and post-colonial
contexts: ‘‘The formation of modern identities always already occurs in the
crucible of intensely spatially interconnected worlds. . . . Just as colonialism
and its ‘informal continuities’ with the postmodern world create colonized
and colonizers alike, they also provide the stage for the emergence and trans-
formation of certain musical givens, of things such as European music, music
history, primitive music, non-Western music, and so on. Thus one of the con-
clusions one might draw . . . is that musicology, as a mode of knowledge
about an object called European music, in fact could only have emerged in
relation to colonial encounters’’ (1999, 8).
10 On language ideology see Schieffelin, Woolard, and Kroskrity 1998; Kroskrity
2000; and Woolard and Schieffelin 1994.
11 ‘‘These objects [which include the phallus, feces, the gaze, and the voice]
have one common feature in my elaboration of them—they have no specular
image, or in other words, alterity. It is what enables them to be the ‘stuff,’ or
rather the lining, though not in any sense the reverse, of the very object that
one takes to be the subject of consciousness’’ (Lacan 1977, 315).
12 See Dunn and Jones 1995, Silverman 1988, Engh 1997.
13 The American pragmatist philosopher Charles Peirce suggested that signs are
of three types: icons, indexes, and symbols, each of which bear different kinds
of relations to their referents, varying from complete motivatedness to ar-
bitrariness. One of his most powerful ideas, taken up and developed by an-
thropologists and linguistic anthropologists since the 1980s, is the notion that
within every symbol there is an index and an icon, and that these serve to
‘‘motivate’’ the symbol (Parmentier 1985; Hartshorne and Weiss 1931–1958).
Peirce’s ideas offer a way of explaining how symbols and metaphors come to
seem so natural, necessary, and powerful to their users.
14 In linking the marginalization of voice within the study of language to the
social and political marginalization of the people they are writing about, Feld,
Fox, Porcello, and Samuels associate their intellectual project with a classically
emancipatory political project.
15 The smaller states were taken over relatively early: Palghat in 1790, Ramnad
in 1772, Dindigul in 1792. Travancore, Cochin, and Pudukkottai were the last
to lose their princely status, after Indian independence in 1947 (Roberts 1952).

292 notes to introduction

16 The Sanskrit manuscript Sārvadēvavilāsa (ca. 1800; referred to in chapter 2),
which records the peregrinations of two Brahmins around the city of Madras
in their quest for patronage, mentions the names of several prominent du-
bashes and dharmakartas and the musicians and dancers they patronized.
17 The songs of the three nineteenth-century saint-composers of Karnatic music
(Thyagaraja, Muthuswamy Dikshitar, and Syama Sastri), viewed as the ulti-
mate expressions of bhakti, were sung by bhajan associations on Gandhian
social reform missions in the early twentieth century (Hancock 1999, 58).
18 Sumathi Ramaswamy’s Passions of the Tongue (1997) analyzes in depth the figure
of Tamittāy as a personification of the Tamil language and the discourse
of Tamil devotion. Theories of the Dravidians as a separate race, distinct
from the Aryans, were quickly appropriated by the emerging non-Brahmin
movement. In 1916 the merchant, banker, and philanthropist P. Tyagaraga
Chetti signed the Non-Brahmin Manifesto, a document that protested the Brah-
min categorization of all non-Brahmins as lowly Sudras. The Non-Brahmin
Manifesto was the first step in the formation of the Justice Party, which came
into power in 1920. The Justice Party developed to challenge the Brahmin-
dominated Congress Party. In 1925 E. V. Ramasami, the political leader who
later became known as ‘‘Periyar,’’ quit the Congress Party and joined the Jus-
tice Party; in the same year, he founded the Self-Respect League (Cuyamari-
yātai Iyakkam), which espoused the idea of an original Dravidian culture, at
the same time calling for the abolition of the caste system and the adoption
of atheism and rationalism (Geetha and Rajadurai 1998, 303–11).
The fundamental issue of the 1930s was the protest, by EVR’s followers,
of the Congress Party’s proposal to make Hindi the national language, and to
institute compulsory Hindi study in the South. Anti-Hindi protests went on
through the 1960s, and became a rallying point for Dravidianist sentiment.
19 Vēṭikkai is, more specifically, amusement derived from being an onlooker who
is not really involved.
20 The Tirukkural is a set of aphorisms composed in couplet form by the Tamil
sage Tiruvalluvar, circa 500 AD.
21 The Tanjavur district is generally recognized as the cradle of Karnatic music.
22 Hinduism Today, September 1997.
23 For a historical treatment of Hindustani classical music that examines as-
pects of its canon formation in relation to colonialism and nationalism see
Bakhle 2005.
24 The dichotomy of ‘‘great’’ and ‘‘little’’ traditions, referring to the split between
‘‘classical’’ or ‘‘literate’’ traditions and ‘‘folk’’ or ‘‘oral’’ traditions, was first ar-
ticulated by the anthropologist Robert Redfield in the context of his research
in the Yucatan in the 1930s and then in his The Primitive World and Its Transfor-
mations (1953). Redfield was interested in the ‘‘social anthropology of civiliza-
tions’’ and the role of cities in cultural change. ‘‘The Cultural Role of Cities,’’
coauthored by Redfield and Milton Singer (1954), stated the great tradition–

notes to introduction 293

 little tradition distinction most succinctly. Redfield’s ideas were elaborated in
the Indian context by Singer, especially in When a Great Tradition Modernizes
(1972). With its reliance on the idea of a coherent civilizational tradition, the
dichotomy between great and little traditions has distinct resonances with
(and undoubtedly roots in) Orientalist scholarship on India.
25 Margaret Cousins, an Irish woman active in the Celtic Revival movement in
early-twentieth-century Ireland, and a leader of the Irish Women’s Franchise
League, became a Theosophist and moved with her husband, James Cousins,
to Madras in 1913. Music was one of her abiding interests; she had earned a
music degree in Ireland. While in India, she became active in issues concern-
ing the emancipation of Indian women, and was one of the founders of the
Women’s India Association. See Candy 1994.

1 Gone Native?
1 Charles Day wrote that the sarangi, an indigenous bowed fiddle now seen
mostly in North India, was ‘‘rapidly being discontinued, and an English fiddle
tuned as a vina or sarangi is often substituted for it’’ (1891, 93). Several other
sources corroborate the fact that the sarangi was used in South India in the
nineteenth century but fell into disuse in the twentieth. See Poduval (n.d.,
24), Popley (1921), Fox-Strangways (1914).
2 Bruno Nettl uses the terms ‘‘Westernization’’ and ‘‘modernization’’ (1985, 163).
I have avoided using these terms because they imply that the categories of
‘‘Western’’ and ‘‘modern’’ are unambiguously definable and imposed from
the outside, rather than debated and constructed in opposition to what then
comes to be considered ‘‘native’’ and ‘‘traditional.’’
3 See also Ian Woodfield’s Music of the Raj (2001). The appendix lists examples
of musical instruments listed in the Bengal Inventories, 1760–1785, including
about twenty violins (240–45).
4 Icai veḷḷālar, literally ‘‘cultivator of music,’’ was a term coined in the 1950s for
a group of caste communities whose hereditary occupation was music. They
are not related to the Veḷḷālas, castes of landed agriculturalists, although the
use of the term veḷḷālar (cultivator) was meant to raise their status.
5 ‘‘Fiddle’’ is pronounced in Tamil with a retroflex ‘‘ṭ’’: piṭil. The last three ver-
sions of the word in Satyanarayana’s quote should be pronounced with this
sound.
6 The monumental book of notations Sangīta Sampradāya Pradārṣini, compiled
by the grandson of Muthuswamy Dikshitar in 1904, included a section on
‘‘music practice,’’ which featured many of these compositions. Such pieces
were later hailed as didactic masterpieces, ideal for avoiding ‘‘boredom and
disillusionment’’ on the part of young music students because they would
help young singers to get the pitches right while helping them become accus-
tomed to phrases used in difficult songs (Sankaramurthy 1990, ix). The route

294 notes to chapter one

 to vocal music was thus distinctly instrumental; getting the correct pitches
was most important, while the words themselves, detached from their origi-
nal meaningful position in Karnatic music, could be practiced as sounds.
7 It is unclear exactly what the kinnari is. Ramaswamy Bhagavatar later says that
‘‘the difference between the kinnari instrument and the fiddle is very small’’
(1935, 53). Most of the time he uses the two terms interchangeably.
8 S. K. Ramachandra Rao suggests that, before it appeared in concert halls, the
violin was adopted primarily for use in dance music and that many dance
teachers were also violinists (1994, 18–19). It was also used in mēḷam (musi-
cal processions at temples or weddings usually featuring the nagaswaram, a
double-reed instrument whose classical status is contested to this day). The as-
sociation of the violin with dance music, and thus with the devadasi commu-
nity, Ramachandra Rao suggests, produced a certain resistance among Brah-
mins against taking up the instrument. Such resistance was first overcome in
the early twentieth century when the violin began to be used as accompani-
ment in katha kalakshepam, a kind of musical storytelling (21).
9 Bakhtin discusses the multivocality of utterances in Problems of Dostoevsky’s
Poetics (1984). Jane Hill uses some of these concepts in her article ‘‘The Con-
sciousness of Grammar and the Grammar of Consciousness’’ (1996) to sug-
gest that political and social relations are reflected in the way voices compete
within any single utterance.
10 Compare Daniel Neuman’s discussion of accompaniment in Hindustani music
(1980: 121–22, 136–40). Also, according to Neuman, while the hierarchy of
soloist and accompanist corresponds to the hierarchy of caste differences,
there is no such correspondence in the contemporary Karnatic music scene
with regard to violin accompaniment—accompanying violinists are mostly
Brahmins, as are vocalists, and there is no institutionalized division between
soloists and accompanists.
11 Sambamoorthy (1901–1973) was born in the Tanjavur district to a family that
had had musical connections to the Tanjore court under King Serfoji. He
studied violin and vocal music in Madras and became a lecturer in music at
Queen Mary’s College in 1928. In 1931 he went to Munich for five years to
study Western music and comparative musicology. On returning to Madras,
he took a position as lecturer and head of the Madras University Department
of Indian Music, in which capacity he served from 1937–1961 and from 1966
to his death. He wrote numerous books on the history of Karnatic music
and musicians, and developed teaching curricula for university music depart-
ments.
12 A sangati is a variation on a melodic line. A tala is a cycle with a certain num-
ber of beats within which a composition is set. The eṭuppu is the starting place
in the tala. Alapana is free-time improvisation, while kalpana swara is impro-
visation that occurs within the tala cycle.
13 According to Tejaswini Niranjana, the sense of figuration is different from

notes to chapter one 295

 representation in that figuration involves a play of movement between sig-
nified and signifier which effectively undoes the opposition between them
instead of reinforcing it (1994, 44, 50).
14 The term swadeshi refers to the incipient phase of the Indian nationalist move-
ment, started around 1905, to create an economically self-sufficient India by
ceasing to use British-made products. Swadeshis were those who followed
this policy and used only Indian-made products, notably cloth. The swa-
deshi movement was controversial and its proponents were often accused of
hypocrisy.
15 Īs pīs is an onomatopoeic expression in Tamil referring to the hissing sound
of English being spoken.
16 The slide and cast-iron track were parts of a disused optical bench. The chain
and hubs were spare parts purchased from a cycle-dealer, as was the ball bear-
ing of the axle of the lever.
17 The horn violin, also called phono violin or Stroh violin, was invented in 1899
by the Austrian John Matthias Augustus Stroh and was first manufactured in
London in 1901. These violins were used in the recording industry from the
late nineteenth century to the beginning of electronic amplification because
they were loud enough to record on wax cylinders. The instrument works by
having vibrations from the strings passed to the center of an aluminum disc
that amplifies the sound, which is then projected through a metal horn.
18 For a description of the social status of the sarangi in North India see Neuman
(1980, 134–35). Neuman implies that the sarangi was incorporated into Hindu-
stani classical music as an instrument with lower status; furthermore, in con-
trast to South India, a sharp status distinction was made between soloists and
accompanists. Nautch is derived from the Hindi word nach (dance).
19 In 1913 Ernest Clements designed a new ‘‘sruti harmonium,’’ carefully tuned
to avoid the problems caused by equal temperament and to be truer to
‘‘Indian intervals,’’ and demonstrated it at the first All-India Music Confer-
ence in Baroda in 1916. Although it never became popular among musicians,
who continued to use regular harmoniums based on equal temperament, it
did produce a heated exchange of letters between Clements and the Tamil
musicologist Abraham Pandithar in the years following. Clements claimed
that only when Pandithar’s daughters sang with the sruti harmonium were
they in tune; with Pandithar’s veena specially designed to show off the ‘‘South
Indian sruti system’’ they sang ‘‘the harsh tones of equal temperament,’’ while
with the violin accompanying they had a tendency to ‘‘drift insensibly into
other scales’’ (Pandithar 1918, 2).
20 The ban was partially lifted in the early 1970s and later entirely lifted. See
Neuman 1980, 184–86.
21 The word moṭṭai, literally meaning bald, is used to refer to plain notes.
22 The best-known violinists of the generation previous to Dwaram’s were
Tirukkodikkaval Krishnayyar (1857–1913) and ‘‘Fiddle’’ Govindaswamy Pil-

296 notes to chapter one

 lai (1878–1931). More information about them can be found in P. Samba-
moorthy’s Dictionary of South Indian Music and Musicians and in articles pub-
lished in Kalki magazine in the 1940s.
23 Dwaram Mangathayaru, personal communication, Madras, June 1998. Man-
gathayaru is the daughter and disciple of Dwaram Venkataswamy Naidu.
24 Dwaram Mangathayaru, personal communication, Madras, May 1998.
25 The word jāru comes from the Telugu verb jara-, to slide. Such descriptions
of Dwaram’s playing can be found in writings by the musicologists T. S. Par-
thasarathy, B. V. K. Sastry, and T. V. Subba Rao.
26 Srimathi Brahmanandam, personal communication, Madras, February 1998.
27 I am indebted to Adrian L’Armand, who was a student of Karnatic violin
in Madras in the 1960s, for sharing his observations and insights with me.
‘‘Florid’’ is his term.
28 I thank V. A. K. Ranga Rao, a record collector in Madras, for sharing record-
ings of violinists from 1910 on with me.
29 Other uses of the violin in fusions abound. Shakti, a well-known fusion group
which made its first recording in 1975, made extensive use of the violin. The
South Indian violin has traveled to the West perhaps most visibly in the musi-
cal creations of L. Shankar and L. Subramaniam since the 1990s.
30 ‘‘An Enjoyable Fare,’’ The Hindu, 14 August 1998, 28.
31 As Steven Connor has written, ‘‘Sound, especially sourceless, autonomous, or
excessive sound, will be experienced both as a lack and an excess; both as a
mystery to be explained and an intensity to be contained. Above all, sound,
and as the body’s means of producing itself as sound, the voice, will be asso-
ciated with the dream and the exercise of power’’ (2000, 23).
32 Slavoj Žižek comes to a similar point in his Lacanian reading of film, in which
he is critical of the idea of a naturalized voice, suggesting instead that one
focus on the process by which voices come to be embodied and taken as natu-
ral. ‘‘What we have to renounce is thus the commonsense notion of a primor-
dial, fully constituted reality in which sight and sound harmoniously comple-
ment each other; the moment we enter the symbolic order, an unbridgeable
gap separates forever a human body from ‘its’ voice. The voice acquires a spec-
tral autonomy, it never quite belongs to the body we see, so that even when
we see a living person talking, there is always some degree of ventriloquism
at work: it is as if the speakers’ own voice hollows him out and in a sense
speaks ‘by itself,’ through him. In his Lectures on Aesthetics, Hegel mentions an
ancient Egyptian sacred statue which, at every sunset, as if by miracle, issued a
deep reverberating sound—this mysterious sound magically resonating from
within an inanimate object is the best metaphor for the birth of subjectivity’’
(1996, 93).
Steven Connor similarly imagines the possibilities for a history of voice as
necessarily intertwined with the history of the ‘‘ventriloquial’’ voice, which
acts as ‘‘a medium for exploring the relations between selves and their voices’’:

notes to chapter one 297

 ‘‘The legitimate and familiar exercise of the voice is accompanied by the
doubts and delights of the ventriloquial voice, of the voice speaking from
some other place, reorganizing the economy of the senses, and embodying
illegitimate forms of power’’ (2000, 43).
33 Compare the ambivalent dynamics of ‘‘not white, not quite’’ as discussed by
Homi K. Bhabha (1984).
34 Michael Taussig characterizes mimesis as ‘‘ ‘a space between,’ a space perme-
ated by the colonial tension of mimesis and alterity, in which it is far from
easy to say who is the imitator and who is the imitated, which is copy and
which is original’’ (1993, 78).

2 From the Palace to the Street

1 This event is referred to by M. B. Vedavalli in Mysore as a Seat of Music
(1992, 16).
2 On the construct of kingship in South India see Dirks 1993, Waghorne 1994,
Narayana Rao, Shulman, and Subrahmanyam 1992, Daniel 1984.
3 In conversations with me, many older musicians and concertgoers used the
phrase ‘‘mechanical life’’ to evoke what they disliked about modern, urban life.
4 ‘‘Tanjavur’’ is the modern spelling; the old spelling is ‘‘Tanjore.’’ I use ‘‘Tan-
jore’’ when referring to proper names where that is the spelling consistently
used, such as the Tanjore Quartette. Similarly, ‘‘Travancore’’ is the old spell-
ing of a princely state in the area that is now southern Kerala and part of
Tamil Nadu; the kings of Travancore resided in the city now known as Tri-
vandrum (English pronunciation) or Tiruvanantapuram (Tamil/Malayalam
pronunciation).
5 Serfoji’s projects reflect a desire for order, classification, and replicability, a
certain encounter with the idea of modernity (see Peterson 1999).
6 K. N. Panikkar, personal communication, Trivandrum, June 1998.
7 Information on the posts created for musicians and the amounts paid to them
is available in some of the Trivandrum palace records published in the Kerala
State Archives Bulletin (1996).
8 On the capacity of music and language to create, and not simply reflect, kingly
power, see also V. Narayana Rao and David Shulman, eds., A Poem at the Right
Moment (1998, esp. 148–59).
9 Mysore Archaeological Reports 1928–1929. Photocopy, Mysore State Univer-
sity.
10 Mysore Archaeological Reports 1928.
11 Mysore Archaeological Reports 1927–1928.
12 Ibid.
13 Dwaram Durga Prasad Rao, personal communication, Vizianagaram, Septem-
ber 1998.

298 notes to chapter two

14 As a king, Ananda Gajapathi was also intensely interested in musical insti-
tutions outside of his court. He often traveled to Calcutta to hear concerts
and donated liberally to the Poona Gayan Samaj, a cultural organization that
supported musical training and held concerts. He also responded to concerns
about the lack of notation for Indian music by sponsoring the publication of
several music manuals for students, written by the Taccur brothers of Madras
(Rama Rao 1985, 16, 25; see chapter 5 herein for more on these books). In a
description of Ananda Gajapathi, the writer Gurujada Apparao, who received
royal patronage, stated that the maharaja felt concern that ‘‘the condition of
prostitutes is deteriorating day by day. They must be taught the arts for earn-
ing a living. If they are enrolled in the Music College and taught fine arts it
would be good’’ (quoted in Rama Rao 1985, 47).
15 Why six years was determined to be an appropriate course length is not clear.
Perhaps it was modeled on medical schools, which also had six-year courses
at that time.
16 The basic plan of the syllabus, with divisions between theory and practical
elements, is the same in curricula planned by the musicologist and teacher
P. Sambamoorthy and continues to be used as the basis for the Government
Technical Examinations in music.
17 Dwaram Durga Prasad Rao, principal of the Maharaja’s Government Music
College, Vizianagaram, very generously allowed me to see the original syl-
labus.
18 Harikesanallur Muthiah Bhagavatar was appointed samasthana vidwan of
Travancore in 1936 and became the first principal of the Swati Tirunal Music
Academy when it was established in 1938 (Vasudevachar 1955, 81); C. S. San-
karasiva Bhagavatar left the Ramanathapuram samasthanam to become the
first principal of the Sathguru Sangeeta Samajam in Madurai in the mid-
1940s.
19 The tradition of impromptu improvisation by court poets was well devel-
oped in South India. See Narayana Rao and Shulman 1998; Swaminathayyar
1940–1942.
20 See Hughes 1996 for a description of the social divisions of Madras in the
nineteeth and early twentieth centuries. See Neild 1979 for a description of
Madras circa 1800.
21 T. Shankaran, personal communication, Madras, June 1998.
22 N. Ramachandran, secretary of the Indian Fine Arts Association, personal
communication, Madras, September 1998.
23 In the 2002–2003 music season there were seventy-three organizations spon-
soring concerts in Madras, up from seventeen in 1987–1988 (Sruti 2003, 21).
24 See Neuman 1992 and Erdman 1978 for discussions of the changes in Hindu-
stani music that occurred with the shift from royal patronage to urban music
concerts.

notes to chapter two 299

25 Randor Guy, personal communication, Madras, August 1998.
26 More information on the Madras Music Academy can be found in Subra-
manian 1999a.
27 Other useful sources on the making of classical music in the West include
Goehr 1992 on the concept of the musical ‘‘work’’ and Levine 1990 on the
development of high culture in twentieth-century America.
28 S. V. Krishnan, personal communication, Madras, December 2003.
29 I am grateful to Bernard Bate for drawing my attention to this source. It is
also mentioned in his dissertation, Mēṭaittami: Oratory and Democratic Practice
in Tamil Nadu (2000, 57).
30 Until the 1970s, it was customary in concert and cinema halls to have a sepa-
rate section designated for women, often to one side or toward the back of
the hall; some cinema halls still have ladies’ sections.
31 The name of the god Siva. Here, equivalent to ‘‘Testing, 1, 2, 3.’’
32 Indeed, one elderly vocalist recalled hearing Musiri Subramania Iyer ‘‘shout-
ing at G like a woman’’ in the early 1940s (H. M. V. Raghunathan, personal
communication, Madras, December 2003). Jon Higgins was the first to use
the expression crooner to describe the microphone-influenced style of singing
that has emerged (1975, 24).
33 On the microphone and the production of public intimacy, see Frith 1996
(201) and Connor 2000 (38). ‘‘The crooning style of twentieth-century popu-
lar song,’’ Connor writes, ‘‘was discovered by singers and sound engineers in
the early days of sound recording when it was discovered that microphones
could not cope with the extreme dynamic ranges possessed by singers used to
commanding the large space of the concert hall. The crooning voice is seduc-
tive because it appears to be at our ear. . . . The microphone makes audible
and expressive a whole range of organic vocal sounds which are edited out
in ordinary listening: the liquidity of the saliva, the hissings and tiny shud-
ders of the breath, the clicking of the tongue and teeth, and popping of the
lips’’ (201).
34 The word uṇarvu in Tamil refers to a distinctly embodied sense of feeling,
perception, or awareness.
35 Dwaram Mangathayaru related to me that her father, the violinist Dwaram
Venkataswamy Naidu, who played concerts publicly from the 1930s to 1960s,
used to begin his daily practice with alankarams and varnams in all five of the
ragas—nattai, arabhi, gowla, varali, and sriragam—in which the pancharatna
kritis were composed, so that his improvisation on these kritis would be flu-
ent (personal communication, Madras, April 1998). Nowadays, not only are
the kritis themselves considered to be sufficient in themselves without im-
provisation, but these ragas are considered to be for the most part lacking in
sufficient scope for extensive improvisation.
36 This is elaborately dramatized in the 1982 remake of the film Thyagayya.
37 An example of his translation is as follows.

300 notes to chapter two

 O Ram! How can I bear! Show mercy sweet.
Advice by Thee I follow day to day.
How blame comes hard on me when I live true
To teachings of the ethic epics chaste! (Sharma 1954, 20)
38 T. S. Parthasarathy, personal communication, Madras, October 1997. Parthasa-
rathy was formerly the secretary of the Madras Music Academy.
39 In 1998 a reporter for The Hindu told me that it was rumored that Thyagaraja,
far from being a mendicant saint, was a wealthy landowner. The rumor, he
suspected, would cause great consternation on the part of the Madras Brah-
min Karnatic music establishment.
40 In an article on the agenda of the Madras Music Academy, Lakshmi Subra-
manian (1999a) suggests that the academy tried, between 1930 and 1947, to
revive the spiritual aspect of Karnatic music while leaving out its sensual as-
pects, which had been carried on by devadasis. The domination of Thyaga-
raja’s songs on the concert stage was a result of what Subramanian calls a ‘‘sani-
tization’’ of Karnatic music. Indira Peterson offers a literary analysis of the
image of the patron in the songs of Thyagaraja, comparing them with those
of the eighth-century saint Cuntaramurti. She reads in the hagiographical
narrative of Thyagaraja’s refusal of royal patronage and his life of ‘‘voluntary
mendicancy,’’ ‘‘the central themes of bhakti sainthood as formulated in late
medieval South Indian bhakti’’ (1992, 133). She speculates on how Thyagaraja
himself may have recast the bhakti tradition to comment on the political dis-
array of his own time, but she leaves open the question of why Thyagaraja
has become so popular in the twentieth century.
41 Jean Langford suggests more broadly that ‘‘the turn toward expressivism also
involves a shift to an understanding of personal identity as expression rather
than mimesis. Thus the source of personal identity is to be found within,
as one’s true nature, rather than in the social realm, as learned or strategi-
cally developed personae. In this modern expressivism, then, the ways in
which personal identity and emotions are constructed are necessarily ob-
scured’’ (2002, 247).
42 A detailed account of the politics surrounding the Thyagaraja Aradhana at
Tiruvaiyaru is also provided in Vai. Radhakrishna Ayyar’s Tiruvaiyā sri tyā-
kapirammam arātaai utcava varalāu (1949–1950).
43 This was a precaution against the kind of confrontation that had happened
the year before, when a group of activists had stormed onto the stage and de-
manded that, since they were in Tamil Nadu, the music should be in Tamil,
not Telugu. Reference to another such incident at Tiruvaiyaru in 1971 is made
in Terada 2000 (487).
44 ‘‘Thyagaraja Aradhana,’’ The Hindu, 29 January 1998, 2.
45 C. S. Sankarasiva Bhagavatar, interview by Ranganayaki Ayyangar, Madurai,
1989, cassette, archives, Sampradaya foundation, Madras.
46 Name has been changed to respect confidentiality.

notes to chapter two 301

 3 Gender and the Politics of Voice
1 The concept of dangerous female power that needs to be contained is promi-
nent in Hindu mythology and in religious rules regulating women’s behavior
(see O’Flaherty 1980). One of these rules is marriage. In Tamil tradition, auspi-
ciousness—that quality which leads to or enables good fortune and success—
is embodied in the figure of the married woman and is lost when the woman
is widowed (see Reynolds 1980).
2 Regula Qureshi meditates on this set of issues at length in her article on
Begum Akhtar, the North Indian singer active from the 1930s to the 1970s
and known for her renditions of Urdu ghazals (poetic couplets). In some ways
her life parallels that of M. S. Subbulakshmi. Born into a courtesan family
near Lucknow, Akhtari Bai achieved fame as a courtesan singer while in her
teens through gramophone recordings. In the 1940s she married one of her
admirers and thus became Begum (Lady, Mrs.) Akhtar; on entering the do-
mestic life of a respectable wife she stopped singing for a number of years,
beginning again only in the 1950s. Qureshi explores the ironies and conflicts
involved in the life of a woman who transformed herself ‘‘from a heredi-
tary professional singer to a respectably married lady who even gave up her
singing career, only to emerge into the public domain transformed into a na-
tional symbol iconic of the courtly musical culture which had shaped her’’
(2001, 97).
3 The significance of the ideal of womanhood to the consolidation of a hege-
monic middle-class culture in India and the middle-class basis of the Indian
nationalist movement have been examined by several scholars. Partha Chat-
terjee, in a 1989 essay on the nationalist resolution of the women’s question in
the 1870s, argues that the rearticulation of Indian womanhood was the foun-
dation on which the notion of an ‘‘inner sphere’’ representative of Indianness
was built. As Mrinhalini Sinha restates the idea, ‘‘The re-articulation of the
Indian woman for the self-definition of the nationalist bourgeoisie provided
the context for the ‘modernizing’ of certain indigenous modes of regulating
women in orthodox Indian society’’ (1996, 482). Sumanta Banerjee, also in the
context of nineteenth-century Bengal, argues that the creation of a new pub-
lic space for the respectable bhadramahila (educated middle-class woman) was
predicated on sharpening class differentiation, especially through the regu-
lation of women’s popular culture and the juxtaposition of this new woman
with women from lower socioeconomic strata (1989). More recently, Man-
kekar 1999 and Hancock 1999 have explored middle-class constructions of
womanhood in relation to nationalist discourses in North and South India
respectively.
4 Before the 1930s, as Kathryn Hansen has noted, the popular theater and early
cinema ‘‘created a public space in which societal attitudes towards women
could be debated’’ (1998, 2291). ‘‘Through the institution of female imper-

302 notes to chapter three

 sonation, a publicly visible, respectable image of ‘woman’ was constructed,
one that was of use to both men and women. This was a representation that,
even attached to the material male body, bespoke modernity. As one response
to the British colonial discourse on Indian womanhood—the accusations
against Indian women on account of their backward, degraded females—the
representation helped support men, dovetailing with the emerging counter-
discourse of Indian masculinity. Moreover, women derived from these enact-
ments an image of how they should represent themselves in public. Female
impersonators, by bringing into the public sphere the mannerisms, speech and
distinctive appearance of middle class women, defined the external equiva-
lents of the new gendered code of conduct for women’’ (ibid., 2296).
5 According to Kunal Parker, the term devadasi, which literally means ‘‘servant
of the gods’’ in Sanskrit, began to be used in Anglo-Indian discourse only in
the second half of the nineteenth century. The colonial label ‘‘dancing girl’’
was used to refer to communities of women with and without temple affilia-
tions in different parts of the Bombay and Madras Presidencies (Parker 1998,
567). The dance they performed was, in Tamil Nadu, variously called sadir,
nautch, or karnatakam, before the ‘‘revival’’ of the dance form in the 1940s as
Bharata Natyam.
6 See Ramakrishna 1983 and Venkarataratnam 1901 for further elaboration.
7 See Anandi 1991 for a detailed discussion of the novel.
8 The abolitionist cause was also helped by American writers like Katherine
Mayo, whose books Mother India (1927) and Slaves of the Gods (1929) specifically
concerned the supposedly degraded state of devadasis.
9 The Tamil term kalai—which, according to the Cre-A Tamil dictionary, has
meanings of both ‘‘art’’ and ‘‘workmanship’’ (Subramaniam 1992)—moved
decidedly toward connotations of high art (as opposed to craft) in this early-
twentieth-century discourse. See, for instance, a 1928 article by C. Jinaraja-
dasa in Triveni magazine: ‘‘In India, in all the arts and crafts there is a great
sense of beauty, but it is now traditional, i.e., the craftsman works by rote,
and does not sufficiently feel a true creative urge. . . . If only our artists will
look with the eyes of the West on Indian scenes . . . they will find plenty
to inspire them’’ (3–4). Oppositions between art and craft also come up in
Rukmini Devi’s essay ‘‘The Creative Spirit’’ (ca. 1940s).
10 See the E. Krishna Iyer Centenary Issue of Sruti for a more detailed account of
the controversy about public performances of nautch (Sruti 1997, 6–9). Mu-
thulakshmi Reddy was particularly opposed to the staging of nautch perfor-
mances at government functions or celebrations.
11 See Amrit Srinivasan (1985, 1875) for a concise listing of some of the changes
this involved; also Allen (1997). The idea of devadasis as ‘‘pure virgin devotees’’
was Annie Besant’s phrase. ‘‘There was a band of pure virgin devotees attached
to the ancient Hindu temples. . . . In those days they were held in high esteem
and were very well looked after. . . . They would follow the procession of

notes to chapter three 303

 Gods dressed in the simplest sanyasi garbs and singing pious hymns. . . . This
is the history and origin of the devadasi class’’ (Besant, quoted in Muthulak-
shmi Reddy 1928–1931, 5). For more on the Theosophical Society, its role in
the revival of Bharata Natyam, and its naming of Rukmini Devi as the World
Mother, see Dixon 1999, Allen 1997, Weidman 1995, and Burton 1994.
The Madras Mahajana Sabha and the Indian National Congress were
founded in 1884 and 1885, respectively, both with a predominance of Smarta
Brahmins (Hancock 1999, 56; Suntharalingam 1974, 231). In the early twen-
tieth century, many of these Brahmins associated with the political/cultural
program of the Theosophical Society, the headquarters of which were located
in Madras. Annie Besant, the society’s leader, formed the Home Rule League,
which espoused the cause of complete independence for India. Theosophy,
with its universalist spiritual philosophy, blended with elite discourse of the
time, epitomized by the Madras Hindu Social Reform Association (estab-
lished 1892), which envisioned a new, reformed Hindu society based on sup-
posedly universal principles of citizenship, rights, and religious belief (Han-
cock 1999, 56–7). Central to the project of both were the agenda of social
reform and the claiming of regional traditions of art, music, and dance as
elements of a universal, pan-Indian ‘‘culture.’’
12 Adyar Library, Chennai (Madras), ‘‘The Creative Spirit,’’ undated pamphlet
published by the Theosophical Society through Vasantha Press in Madras in
the 1940s, 14–15.
13 Ibid., 15.
14 In North India, as Regula Qureshi points out, recording provided an oppor-
tunity for courtesans to continue as singers and entertainers even as their op-
portunities for live performance diminished. Many courtesans became singers
for films. Interestingly, it was also sound recordings that facilitated Begum
Akhtar’s re-entry into the public domain as a singer after her marriage and
her transformation from courtesan to respectable married woman (Qureshi
2001, 97).
15 S. Kalpakam, personal communication, Madras, July 2002.
16 Baby Kamala was a dance prodigy of the time, credited for popularizing Bha-
rata Natyam among the Brahmin middle classes even more than Rukmini
Devi. Baby Kamala gave her first performance in 1941 at the age of seven; the
appeal of the child prodigy can be seen in the fact that she retained the title
‘‘Baby’’ well into her teens. In 1988 a writer in Sruti commented on her rise,
‘‘The timing was perfect. The conditions ideal. And her age was just right. She
was still a child, a ‘baby’ and her innocence and charm endeared her to one
and all’’ (quoted in Allen 1997, 80–81). The appeal of the prodigy continues
today; one popular musician is known as ‘‘Veena Virtuoso ‘Baby’ Gayatri.’’
17 M. S. Subbulakshmi’s devadasi background has the status of a public secret in
the Karnatic music world; everyone seems to know about it but it is never
discussed publicly or written about. For example, Sarathy’s 1997 biography

304 notes to chapter three

 of M.S. skirts the issue by not saying anything about who her father was. He
was a wealthy Brahmin patron of her mother.
18 M.S. acted in four films: Seva Sadanam (1938), Shakuntala (1940), Savitri (1941),
and Meera (1945). Sadasivam was the producer or co-producer for all of them.
Seva Sadanam was a ‘‘social’’ film that critiqued Brahmin orthodoxy and the
dowry system; Shakuntala and Savitri were based on mythological stories.
Meera, M.S.’s most popular role, was based on the life of Meera Bai, a high-
born Rajput woman who lived in the early sixteenth century, who declared
herself to be Krishna’s bride and renounced her worldly existence to pursue
life as a saint and mendicant. The film was so popular in South India that it
was later dubbed in Hindi. Sarojini Naidu gave an on-screen introduction to
this version in which she described M.S. as the ‘‘nightingale of India’’ (Guy
1997, 229). Naidu’s introduction deftly combines praise of M.S.’s voice with
the implication that a good voice transcends its body like a bird in flight; at
the same time, it makes M.S. represent India.
19 On the effects of the microphone in western contexts, Simon Frith writes
that ‘‘the microphone made it possible for singers to make musical sounds—
soft sounds, close sounds—that had not really been heard before in terms of
public performance. . . . The microphone allowed us to hear people in ways
that normally implied intimacy—the whisper, the caress, the murmur’’ (1996,
187). According to Frith, the microphone also draws attention to the place
of the voice in music, allowing it to dominate other instruments, to be the
‘‘solo’’ in a way it could not have been before (188).
20 Anonymous reader for Duke University Press, personal communication, Oc-
tober 2004.
21 Consider the following description: ‘‘As a celebrity she has moved with and
played hostess to world leaders with gentle charm and dignity. At home she is
the traditional housewife, stringing flower garlands for her puja (Hindu prayer
ritual) room and decorating the floor with her beautiful kōlam (rice powder
designs). It is on stage that she comes into her own—sensuously captivating,
with an occasional lift of the eyebrow and a bewitching smile, not for the
audience, but for the Divine’’ (Indira Menon 1999, 41). Statements like these
reveal an insistence on separating M.S.’s stage persona from her everyday life.
22 In a review of the Meera soundtrack Kalki identified music as the most ele-
vated of all pleasures and was particularly enthusiastic about a song sung in
the voice of the child Meera: the ‘‘child’’ sang so beautifully that ‘‘even a per-
son with a heart of stone couldn’t be unaffected’’ (1946, 31). ‘‘And, thanks to
the skillful recording, we hear MS’ voice quite naturally,’’ he concluded. The
listener’s pleasure is thus constructed through multiple senses of fidelity: the
fidelity of the fan to his favorite star, the fidelity of M.S.’s voice to tradi-
tion, the fidelity of the child Meera to her beloved Krishna, the fidelity of
the record to the original performance.
23 The discourse linking idealized female musicality with notions of naturalness,

notes to chapter three 305

 as opposed to artifice, intellectuality, and virtuoso display, pervades Western
art music. Richard Leppert (1993) and Judith Tick (1986) provide useful discus-
sions of this in English and American contexts respectively. Such distinctions
have their origins in a broader aesthetic discourse based on gendered notions
of the beautiful and the sublime, as addressed by Paul Mattick Jr. (1995). These
distinctions traveled to India as part of a colonial discourse about music but
became specifically central to a nationalist discourse that associated ‘‘natural’’
female musicality with the essence of the Indian nation.
24 Adyar Library, Chennai (Madras), ‘‘Woman as Artist,’’ undated pamphlet pub-
lished by the Theosophical Society through Vasantha Press in Madras in the
1940s, 5.
25 Ibid., 2–3.
26 Ibid., 7.
27 Adyar Library, Chennai (Madras), ‘‘Dance and Music,’’ undated pamphlet pub-
lished by the Theosophical Society through Vasantha Press in Madras in the
1940s, 4.
28 Ibid., 14.
29 Ibid., 10–11. Devi’s reference to music as the ‘‘language of the gods’’ also al-
ludes to the Sanskrit language, which is referred to in Brahminical tradition
as the language of the gods or the divine language. Much of the music that
accompanies dance uses Sanskrit religious terminology. Devi’s statement may
also allude to the fact that the classicization of Karnatic music took place at
the same time as the Hindu religious revival, which emphasized a classical
Hinduism that was Sanskritic in emphasis. Several figures associated with the
classicizing of Karnatic music in the mid-twentieth century, such as Kalki,
were also part of the group that constructed classical Hinduism in contrast to
Western materialism and lack of spirituality.
30 Ibid., 12, 16.
31 Allen 1997, Meduri 1996, Amrit Srinivasan 1985, and O’Shea 1998 provide
more detailed descriptions of the ways dance performance became nation-
alized and ‘‘sanitized.’’ The conflict between earlier dance styles practiced
by those from the devadasi community and the reinvented Bharata Natyam
popularized by Rukmini Devi was framed by Devi as a conflict between
expressing sringāra (erotic sentiment) and expressing bhakti (devotion), two
modes which she considered irreconcilable. Tanjore Balasaraswati, who came
from a family of devadasis and was a contemporary of Rukmini Devi, would
later insist that sringara and bhakti were one and the same (O’Shea 1998,
46–49).
32 Robert Ollikkala provides a critique of the gendered distinction between clas-
sical and light-classical music in the North Indian classical tradition. ‘‘ ‘Light,’ ’’
he writes, ‘‘consistently seems to imply ‘lesser,’ ‘derivative,’ ‘secondary,’ and
‘feminine.’ It is no coincidence that ‘light-classical’ music, sung by women

306 notes to chapter three

 of assumed dubious virtue (within a role they have inherited in a male-
dominated social structure) is considered to be more emotional, . . . less tech-
nically demanding, less pure in terms of tala and raga’’ (1999, 35). Ollikkala
suggests that there are ‘‘universal implications to the term ‘classical,’ implica-
tions that include, but reach far beyond, the musical’’ (36). Vidya Rao, herself
a singer of thumri, one of the North Indian genres now labeled ‘‘light classical,’’
critiques this distinction by embracing thumri as what she calls the ‘‘feminine
voice,’’ a genre capable of subverting the conventions of Hindustani classical
music, providing an alternative to the virtuosic display that is part of classical
genres like khayal (1999).
33 Jennifer Post has written of the separation of dance from music in the Marathi
and Konkani region of Western India in the context of the late-nineteenth-
century decline of courtesan traditions there. Many women, she states, began
to restrict their performances to singing and avoided dance gestures, presum-
ably in an effort to lend respectability to their performances (Post 1987, 104–5).
34 Ragam-tanam-pallavi was considered for a long time to be suitable only for
men.
35 Compare Malathi de Alwis’s argument that respectability operates as an ‘‘aes-
thetic’’ that ‘‘must simultaneously clothe a woman’s body as well as accentuate
it’’ (1999, 186–87).
36 See Anandi 1997 and Lakshmi 1990 for discussions of the way this metaphoric
opposition has operated in Dravidian Movement politics in Tamil Nadu. In
discussing the figure of Tamittāy, or Mother Tamil, Lakshmi states that the
‘‘yardstick [mother vs. whore], deliberately nurtured and cultivated for the
political advancement of a particular group of politicians . . . has now been
turned into a ‘truth’ of culture, something inherent, natural, and unalterable’’
(Lakshmi 1990, 82).
37 The word kaṇakku also has caste associations in the music world. It is generally
thought that the mathematically based, rhythmic aspects of Karnatic music,
like swara kalpana and pallavi, come from the non-Brahmin, Icai Veḷḷālar
(nagaswaram-playing) tradition. For respectable female performers, tala itself
was considered a kind of excess; a female musician in her eighties told me that
in her day ladies were not only not supposed to do swara kalpana, but even
keeping the tala with one’s hand was also considered improper (T. Mukhta,
personal communication, Madras, September 1998).
38 Compare the interiorized conception of alapana, as elaborated above, with a
much more externalizing discourse about tala. When I learned rhythmic as-
pects and special tala exercises from C. S. Palaninathan, a mridangist in Madu-
rai, he constantly told me that in order to really get the rhythms right, I had
to ‘‘make a sound,’’ that is, clap louder or slap my leg more vigorously. ‘‘Tāḷam
nalla pōṭuṅke [Put the talam well],’’ he commanded. He might sit in a relaxed
posture and teach me rhythmic sequences that he had internalized, but if in

notes to chapter three 307

 his recitation something went wrong, he would move from chair to floor,
straighten up, and slap his leg resoundingly, as if the correct version could be
arrived at from the outside in.
39 Dipesh Chakrabarty suggests that in the Bengali context the notion of
pabitrata, or purity, was crucial to the idea of a sphere of interiority that
was autonomous from the physical body. Thus, one might think of perfor-
mances of nonperformance as a ‘‘technique of interiority,’’ as Chakrabarty
suggests: a way of staging one’s innermost emotions as ‘pure’ and thus helping
them ‘‘transcend anything that was external to the subject’s interior space—
the body, interests, social conventions, and prejudices’’ (Chakrabarty 2000,
138–40). The concept of pabitrata, which emerges in end-of-the-twentieth-
century Bengali novels that portray widows is, of course, highly gendered.
40 This distinction between extramusical elements and the ‘‘music itself ’’ is cru-
cial to the idea of secular art music, in which raga (melodic scale or mode)
and tala are considered the major musical elements, whereas lyrics are appre-
ciated for their aesthetic qualities and abstract meaning but are devalued if
they convey sentiment or sensuality.
41 In order to remain respectable, women musicians in the 1960s and 1970s
would often only take music students in their own home. My violin teacher
recalled that her father would not allow her to teach a female student in that
student’s home or to play violin in the film studios, which would involve
traveling, thus ‘‘cheapening’’ her music.
42 ‘‘I haven’t gotten what I expected from my children. Now that they are grown,
I would just like to have my own flat. Preferably I would like to die while
teaching a music lesson,’’ said one female musician (quoted in Lakshmi 2000,
208).
43 This is known as maṅkala icai (auspicious music). The category of auspicious-
ness applies also to women; devadasis were known as nityacumaṅkali (ever-
auspicious women) because, being married to god, they could never become
widows. For an explanation of this, see Kersenboom-Story 1987.
44 Speculations on Ponnuthai’s actual motivations for retreating from public
performances abound. Some suggest that she was not forced to retreat but
chose to in order to validate her status as the respectable widow of her hus-
band and to avoid being seen as a devadasi. Others suggest that her retreat was
motivated by political reasons relating to the fact that the Congress Party, of
which her husband was a member, lost power in Tamil Nadu in the early 1970s.
45 Her name has been changed to respect confidentiality.
46 See Bullard 1998 on the South Indian flute and women artists.
47 Bhairavi was quite critical of the notion that female musicians had been ‘‘lib-
erated’’ by the pioneering efforts of older female musicians like M. S. Subbu-
lakshmi and D. K. Pattammal. When I asked her what she would have asked
them in an interview, she replied, ‘‘Whether they could have succeeded with-
out their husbands’’ (personal communication, Madurai, July 2002).

308 notes to chapter three

48 ‘‘Quite natural’’ was one of her stock English phrases.
49 The connections between the ‘‘expressivist turn’’ and the notions of interi-
ority and inner voice in European thought are discussed by Taylor (1989, 370–
90). Raymond Williams, in Culture and Society 1780–1950, suggests that the
notion of ‘‘art’’ and the romantic artist developed in the nineteenth century
in Europe as ‘‘art became a symbolic abstraction for a whole range of general
human experience. . . . A general social activity was forced into the status of
a department or province’’ (1983, 47). In the new idea of art, an emphasis on
skill was gradually replaced by an emphasis on sensibility, suggesting a kind
of interiorization (ibid., 44).
50 On the idea of the inner sphere of the Indian nation, see Chatterjee 1993.
51 Sadasivam introduced M.S. to Gandhi, Nehru, and a host of other political
figures. Gandhi is purported to have said, on hearing her sing, ‘‘I have no
knowledge of music [caṅkīta ñāam]. But your voice and your song are ex-
tremely sweet’’ (Kalki 1946, 29).
52 L. R. Easwari, personal communication, Madras, June 2002.
53 S. Janaki, personal communication, Madras, January 2004.
54 Bangalore Nagaratnammal, a devadasi, spoke on the rights of devadasis dur-
ing the devadasi-abolition movement; she also began a separate Thyagaraja
aradhana for women at Tiruvaiyaru to address problems of exclusion. Vai.
Mu. Kothainayaki (1901–1960), a Brahmin woman, lectured extensively on
the nationalist cause and the betterment of Indian women. She composed
songs, wrote stories and novels, and managed the women’s journal Jaganmohini
from 1925 to 1954. In fact, Nagaratnammal encouraged Kothainayaki to sing
and provided her with concert opportunities in Mysore (Sruti 2001, 44–45).
55 Pattammal stated that even after she had made gramophone records, her father
would not allow her to sing onstage. ‘‘It is Vai. Mu. Ko. who is responsible
for my being amidst you as a musician today. I first met her in a gramo-
phone company. I was 10 years old then. Those days everyone was so surprised
that a 10-year-old could sing so well. Vai. Mu. Ko. was also impressed. She
wanted me to enter the concert arena. Those days girls from Brahmin families
were not encouraged to perform in public. It was a wonder that my father,
who was an orthodox Dikshitar, permitted me to even cut a record. But Vai.
Mu. Ko. was a tenacious lady. She made several trips to Kanchipuram to per-
suade my father. They had several arguments. My father tried to give all kind
of excuses even stating that I could not sing with accompanists. She vehe-
mently asked, ‘What is wrong if a Brahmin girl sings on stage?’ Ultimately
my father relented and my first concert was held at Egmore Mahila Sabha
[Egmore Women’s Association in Madras]’’ (Sruti 2001, 41). Kothainayaki and
Pattammal later made three 78-rpm records with songs on themes of social
protest (42).

notes to chapter three 309

 4 Can the Subaltern Sing?
Epigraph quote can be found in Tami Icai Makānāṭu Ceai Nikacci Mālar
1944 (96).
1 Downing Thomas has traced the emergence of a ‘‘verbal paradigm’’ for music
—the analogy of music to language and theories of their original unity—in
the writings of French Enlightenment thinkers such as Condillac and Rous-
seau. He locates in the late sixteenth century a conceptual shift toward think-
ing of music within the domain of the rhetorical and human rather than
grouping it with mathematical or cosmological phenomena. In Enlighten-
ment thought, music came to be related to national subjects by analogy to
their relation to their national language or mother tongue. For Rousseau
especially, the concept of nationality was central in connecting music to lan-
guage: ‘‘Every nation’s music draws its principal character from the language
that belongs to it’’ (Rousseau, quoted in Thomas 1995, 98). By the nineteenth
century, in romantic discourse the idea of music as a language came to be
particularly associated with European art music; composers were said to be
masters of an aesthetic ‘‘language’’ of music. The analogy between music and
language, part of the colonial inheritance of Western ideas about music, be-
came a central aspect of nationalist discourse concerning the place of music
in newly independent India.
2 These two choices—between voice as a cultivated, aestheticized instrument
and voice as a representation of self—may be considered in relation to, re-
spectively, Sanskritic rāsa theory and Tamil bhakti discourse. The theory of
rāsa was first elaborated in Bharata’s Natya Sastra, a treatise on dance and
music from the early centuries AD. Literally defined as ‘‘taste,’’ rāsa refers
to a listener’s or spectator’s detached aesthetic apprehension and enjoyment
of an emotion (see, for example, Walimbe 1980 for an explanation of rāsa
theory). In Tamil Bhakti, which emerged as a reaction against overly ritu-
alized and Sanskritic Hinduism, the devotee addresses the deity in his/her
mother tongue, emphasizing directness and the subjective nature of emotion
(see Cutler 1987).
3 Saskia Kersenboom has written that the concept of muttami requires a no-
tion of verbal art as performance and thus a sense of all knowledge as em-
bodied and applied (1995). It is a ‘‘conception of and approach to language in
its dynamism and functional entirety; . . . language (in this case, the Tamil lan-
guage) is not ‘just’ speech, not only spoken/written word, but also, simulta-
neously, song, music, word combined with musical sound, sung word, and
again simultaneously, word enacted in performance’’ (Zvelebil 1992, 142). This
notion, Kersenboom argues, is fundamentally different from the idea of the
‘‘text’’ as existing separate from its performance or reading that underlies the
Western institution of literature (1995, 8). The contrast emerges most clearly
in a consideration of ōlai (literally, ‘‘leaves’’), the palm-leaf manuscripts that

310 notes to chapter four

 contained much of what is today regarded as Tamil literature. Ōlai, Kersen-
boom argues, are not literature in the sense of manuscripts to be read and
studied for their content but are instead mnemonic devices to aid a person
who has already memorized and embodied the text, or objects to be wor-
shipped in themselves (ibid., 14). Ōlai are not complete texts in themselves
but only become complete when their contents are embodied or applied in
performance.
4 Kamil Zvelebil writes that ‘‘the notion of scholar and artist—particularly ver-
bal artist—is traditionally not separated: thus, often, vidvan is he who studies
music, composes music, and performs music’’ (1992, 131 n.8). Swaminathay-
yar’s teacher was Mahavidvan (the prefix maha- means ‘‘great’’) Meenakshi-
sundaram Pillai, who was noted for his scholarship and declamation in Tamil.
5 The process of rediscovering ancient Tamil literature, which took place be-
tween about 1850 and 1925, is described in Zvelebil 1992 (144–222).
6 On the treatment of Brahmin devotees of Tamil by the popular press in the
first half of the twentieth century, see Ramaswamy 1997 (194–204).
7 This tendency is undoubtedly influenced by the theories of Aryan/Dravidian
racial difference that have become prevalent in the discourse on what divides
Brahmins from non-Brahmins in South India. There is a widespread belief,
held by Brahmins and non-Brahmins alike, that Brahmins are not ‘‘original
Dravidians’’ but relative newcomers from North India. Many Tamil Brahmins
speak a variety of Tamil that is highly Sanskritized and Anglicized in word
choice. For a further explanation of colonial theories of Aryan/Dravidian
racial difference, see note 19 below and Trautmann 1997.
8 Stuart Blackburn argues that the European colonial impact on ‘‘literature’’ in
South India was ‘‘filtered through a pre-existing . . . debate about the origins
of Tamil and its position vis-à-vis Sanskrit’’ (2000, 478).
9 See Ramaswamy 1997, chapter 3.
10 V. Narayana Rao has argued, for instance, that the concept of mother tongue,
denoted in Telugu by the term mātrabhāṣa, is absent from South Indian dis-
courses on language prior to the last part of the nineteenth century and that
the term itself is a loan translation from English (1995, 25). See also Pollock
2000 (612–13) in this regard.
11 Lisa Mitchell points out this distinction, suggesting that an earlier ‘‘plea-
sure taken in language—sometimes referred to as rāsa (emotion, sentiment;
aesthetic taste or pleasure; literary or artistic beauty)—should not be con-
fused with an attachment to language’’ (2004, 32, 35). For a detailed discussion
of these kinds of premodern poetic practices, see Narayana Rao and Shul-
man 1998.
12 Mitchell argues that once language comes to be imagined, objectified, and
personified in this way, it also causes a ‘‘new form of alienation of self from
language . . . accompanied by a new fear—the fear of loss. In the face of En-
glish education, and the presence of other ‘languages’—Tamil, Hindi, Marathi

notes to chapter four 311

 —now similarly objectified and separated from Telugu, it has become pos-
sible to imagine losing the language one now thinks of as ‘one’s own.’ Yet at
the same time, it is precisely this alienation of language from self which makes
it possible to imagine that a particular language (like Telugu)—as opposed to
language in the broader undifferentiated sense—is an inalienable part of one-
self. Indeed the emergence of an affective relationship to a particular language
can be seen as a way of counteracting the alienation of self from language, by
reattaching to oneself, not undifferentiated language use, but an externalized
idea of a single ‘language,’ something not possible without this alienation’’
(2004, 38).
13 Ramaswamy’s Passions of the Tongue (1997) and Nambi Arooran’s Tamil Renais-
sance and Dravidian Nationalism (1980) provide detailed accounts of the Tamil
revival. Arooran’s book is more historical in nature, focusing on the revi-
val of Tamil language and literature, the Self-Respect movement, the anti-
Hindi agitations, and the demand for a separate Tamil state in the early 1940s.
Ramaswamy analyzes the discourse of Tamil devotion using writings from all
branches of the Tamil revival.
14 Devadoss grew up in Madurai and came to the United States in 1969 to attend
Oberlin College. He returned to India in the early 1970s.
15 James Siegel discusses the particular kind of listening that is involved in over-
hearing in the context of the development of Indonesian as a lingua franca. It
is not meaning but the ‘‘force of the medium itself ’’ that compels overhear-
ing; the lingua franca is that which ‘‘does not belong in any one community,
which no one truly possesses, in which one sees what one wants and bargains
for it’’ (1997, 60).
16 Interestingly, Devadoss noted that classical music seemed to be insulated from
this new power of music: ‘‘While popular music was thus subject to pushes
and pulls, South Indian classical music, now the preserve of the Brahmin com-
munity, continued to remain steadfast, strong, and traditional’’ (1997, 156).
17 In the first three decades of the twentieth century Bhatkhande conducted
musical research trips in various parts of India, published several books on
the theory of Indian music, compiled and published about 1800 composi-
tions in North Indian vocal music, and founded two music colleges. For a
detailed account of Bhatkhande’s accomplishments in the field of music and
his troubled relationship with his contemporary, the musicologist V.D. Palus-
kar, see Bakhle 2005 (chapters 3 and 5). As Bakhle shows, Bhatkhande’s notion
of Indian classical music as modern and essentially secular in many ways di-
rectly opposed Paluskar’s notion that in order to be revived, Indian music had
to be reframed in the idiom of Hindu religiosity.
18 Although Krishna Iyer mentions no source for this conceptualization, it is
possible that the concept of muttami had influenced him. His general phono-
centrism and his differentiation between natural and conventional meaning

312 notes to chapter four

 is reminiscent of Rousseau as well as Saussure’s distinction between arbitrary
and motivated language.
19 The South Indian syllables for the pitches of the scale—sa, ri, ga, ma, pa, da,
and ni—are usually abbreviated by their first letter. The numbers 1 and 2—as
in ‘‘R1’’ and ‘‘R2’’—indicate flat and sharp notes, respectively.
20 In this passage Krishna Rao refers to a painting that depicts a Hindu myth in
which the sage Viswamitra, distracted from his meditation by the heavenly
nymph Menaka, gives in to carnal desire. The result is the birth of the baby
Shakuntala. In this musical passage, Menaka has come to present the baby to
its father. She addresses Viswamitra, who hides his face in shame. The painting
to which Krishna Rao refers appears in The Psychology of Music (1917, 35).
21 Nambi Arooran’s Tamil Renaissance and Dravidian Nationalism places the Tamil
Music movement in the larger context of Tamil revival. It is also the only
work in English on the Tamil Music Movement.
22 The term Dravidian began to be used in the mid-nineteenth century by colo-
nial philologists skeptical of the idea that Tamil and other southern languages
belonged in the same language family as Sanskrit. Theories of racial differ-
ence between so-called Aryans, who were thought to inhabit most of North-
ern India and to be the descendants of nomadic groups that had migrated to
the Indian subcontinent from the north and west, and Dravidians, who were
thought to be the darker-skinned, indigenous inhabitants of South India,
dominated colonial ethnology in the nineteenth century. The support for
such theories was garnered from the discipline of philology and consolidated
for the first time by Reverend Robert Caldwell in his A Comparative Grammar
of the Dravidian or South Indian Family of Languages (1856). Using comparative
grammatical ‘‘evidence,’’ Caldwell stated that the languages of South India
were different enough from Sanskrit to be considered a different language
family. For this family he chose the term Dravidian, which he adapted from
the original dravida found in a Sanskrit text (4). The word Tamil, he claimed,
was derived from the word dravida (8); being the least influenced by Sanskrit,
the Tamil language was thus the purest representative of the Dravidian family.
23 The Esplanade is close to George Town, a traditionally non-Brahmin area of
Madras. Mylapore is, of course, a bastion of Brahminism. Apparently mem-
bers of the Congress Party from Mylapore refused to allow the Raja Anna-
malai Manram to be built in the Mylapore area.
24 A prominent idea in histories of Tamil culture written during the twentieth-
century Tamil renaissance is that before the era of the great Tamil literature
that was published in the late nineteenth century, Tamil civilization had al-
ready experienced two previous golden ages which had lasted for thousands
of years, many years before the birth of Christ. These periods, known as
the First and Second Tamil Caṅkams (Sangams: academies), which produced
much literature, had been centered in the city of Tē-Maturai (Ten-Madurai:

notes to chapter four 313

 Southern Madurai), located far south of the present southern tip of India,
which is Kanyakumari. Ten-Madurai was eventually swept under the ocean
and its literature lost, but the Third Tamil Sangam, centered in Madurai, from
which the now published literature came, represented the continuation of
the great civilization, its last survivors (Ramaswamy 1999, 97). The found-
ing of the Fourth Tamil Sangam at Madurai in 1901 was intended to stand as
the beginning of another golden age for Tamils. The idea of an ancient, vast,
now lost Tamil land was crystallized in late-nineteenth- and early-twentieth-
century discourses on Lemuria, the sunken continent. The legend of Lemuria
and its destruction by the ocean allowed a glorious past to be imagined out
of the straitened circumstances of the present, thus allowing for the possi-
bility of a return or resurgence. This kind of ‘‘catastrophic consciousness,’’ sug-
gests Sumathi Ramaswamy, ‘‘is very much a response to (colonial) modernity,
a form of resistance to its totalizing and homogenizing knowledge claims.
Catastrophic historical consciousness enables the recuperation of all those
necessary and fabulous knowledges of the ancient past cast out of a world ren-
dered increasingly disenchanted through the work of the modern sciences’’
(98). Yet, as she demonstrates, the re-creation, or the re-enchantment, of the
ancient past is often achieved through a thoroughly modern scientific dis-
course. In this case, modern discourses of philology and musicology made the
twentieth-century enchantment with ancient Tamil music possible.
25 Born in 1859 in the Tirunelveli district into a family of Nadars who had
converted to Christianity two generations before, Abraham Pandithar, by all
accounts a gifted student, initially became a teacher at Lady Napier’s Girls
School in Tanjavur (Nadar 1954, 110–11). Around 1890 he and his wife resigned
from their positions as teachers to take up farming and medicine making full-
time. The title ‘‘Pandithar’’ was conferred on him after he became known as
a physician. Indeed, between his work as a physician and the sale of the ‘‘San-
jeevi Pills’’ for which he became famous, Pandithar was said to have earned
nearly a thousand rupees a day (ibid., 111). Because of his financial success, he
was the subject of a detailed sketch by Somerset Playne in 1914, in a guide to
South India’s commercial development (1914, 486–91).
26 Extensive records of Pandithar’s conflict with other participants in the 1916
Baroda conference regarding the number of srutis in an octave were appended
in the Tamil version of his book, Karuṇamirtacākaram. Most of the book itself
was given over to minute calculations through which Pandithar mathemati-
cally showed that others’ theories were wrong.
27 According to Pandithar, ‘‘upholders of Sanskrit’’ had tried to convince people
that Tamil was derived from Sanskrit, that even the name ‘‘Tamil’’ was derived
from the Sanskrit term dravida. Pandithar greeted such theories with ridicule,
claiming instead that the Sanskrit-speaking Aryans had been unable to pro-
nounce or write the name Tamil and thus had mispronounced it as dravida
(1917, 32).

314 notes to chapter four

28 Indeed, the essay was politely returned by the British journal editors; Pandi-
thar included their correspondence in his record.
29 C. N. Annadurai’s quote can be found in Muthiah 1996 (40).
30 The information in this section I take from interviews I conducted in Madras
and Madurai in 1998. I have made the interviewees’ identities vague in order
to protect their confidentiality.
31 Dharmapuram Swaminathan, personal communication, Kondrattur, India,
November 1998.

5 A Writing Lesson
1 This booklet, printed 16 August 1985, was a compilation of sources that Bala-
chander made to support his argument and was addressed as an ‘‘open letter’’
to ‘‘present and future musicians, musicologists, music-lovers, music students,
experts in allied arts, press and public.’’
2 A comparable debate is described by V. Narayana Rao in his article ‘‘The Poli-
tics of Telugu Ramayanas’’ (2001). In describing the conflict between tra-
ditional Ramayana readers who treat the Valmiki Ramayana as the original
and correct story and the oppositional readings of what he calls the ‘‘anti-
Ramayana discourse’’ in the 1920s, Narayana Rao points out that the concepts
of authorship and authenticity differ crucially between these two points of
view. Whereas leaders of the anti-Ramayana discourse claim Valmiki to be
the author of the Ramayana in a factual mode, for traditional readers, Nara-
yana Rao suggests, ‘‘Valmiki’s authorship of the Ramayana is ideological; they
do not base their statement on empirical textual evidence’’ (163). Balachan-
der’s challenge to the party-line Swati Tirunal story was scandalous because
it similarly attempted to counter myth with an empirical, factually based ar-
gument.
3 I thank Katherine Bergeron for suggesting this way of framing the issue.
4 S. Balachander, open letter, 1. The letter was appended to a copy of the Gayan
Samaj Proceedings of 1887 and copies were distributed to many musical in-
stitutions and musicologists in Madras. I thank the Music Academy Library
for making this source available to me.
5 Ibid., i.
6 Ibid., 1.
7 Ibid.
8 Ibid., 2
9 Ibid., 3.
10 Ibid.
11 Ibid., 4.
12 Ibid., 5.
13 Ibid.
14 Ibid., 5–6.

notes to chapter five 315

15 Published 23 September 1982. A rumor circulated that the brothers were in
possession of palm-leaf manuscripts containing evidence to back up their ar-
gument, but when the Travancore royal family made an offer to buy them,
there was no response. Although Sivanandam and Kittappa’s claim was a far
cry from Balachander’s assertion that Swati Tirunal was a fictional composer,
it could be used—and indeed was used—as evidence for Balachander’s case,
even though the brothers later publicly declared their disagreement with
Balachander (K. V. Ramanathan 1996, 17; Nayar 1997, 23).
16 This musician was from the Icai Veḷḷālar caste and took great pride in being
from the same caste as the Tanjore Quartette, precisely because of the ‘‘pro-
fessional’’ way they had made a living from music.
17 I thank Dr. Kovalam Narayana Panikkar and Dr. V. S. S. Sharma, both of Tri-
vandrum, for discussing the matter of Sopanam music and the Mullamoodu
Bhagavatars with me in June 1998. Dr. R. P. Raja of Trivandrum very gener-
ously discussed the Swati Tirunal case with me at length in May 1998. Sopa-
nam style is discussed in Poduval n.d. (3–4, 27).
18 Compare Katherine Bergeron’s discussion of conflicting ideas about the re-
vival and performance of Gregorian chant between 1880 and 1900, in chap-
ter 4 of Decadent Enchantments. Bergeron contrasts the ‘‘Romantic’’ approach
of Dom Pothier, who was concerned with going beyond the written notes
to find ‘‘the long-forgotten vox of the chant,’’ with the ‘‘Modern’’ approach
of Dom Mocquereau, who embarked on a ‘‘philological’’ project, compar-
ing written manuscripts with each other to arrive at a standardized notation
(1998, 101). Mocquereau’s idea of going back to the original—he used photo-
graphs of the original manuscripts—is similar to Nayar’s idea of finding the
original compositions preserved by the Mullamoodu Bhagavatars.
19 In the epigraph quote, Tagore was arguing against ideas that staff notation
would provide a universal language for music (1874, 382).
20 Christopher Pinney quotes the Parisian photographer Félix Nadar’s com-
ments on the difference between photograph and portrait. The portrait gave
one ‘‘the moral grasp of the subject—that instant understanding which puts
you in touch with the model, helps you to sum him up, guides you to his
habits, his ideas and character and enables you to produce not an indifferent
reproduction, a matter of routine or accident such as any laboratory assistant
could achieve, but a really convincing and sympathetic likeness, an intimate
portrait’’ (1997, 33).
21 Taking C as the scale tonic, ri-1 = D-flat; ri-2 = D-natural.
22 The discussion of the status of ornaments in Karnatic music was part of a
larger discourse on ornament taking place in the context of Indian art. In
1939, in response to European critiques of Indian art as ‘‘merely ornamen-
tal,’’ Ananda Coomaraswamy wrote an essay entitled ‘‘Ornament,’’ in which
he defended the ornament against charges of superfluity. He noted that in
Sanskrit, Greek, and Latin, the word for ornament had originally meant ‘‘that

316 notes to chapter five

 which is added on to make something sufficient’’; it was only later that it came
to mean ‘‘embellishment.’’ Although Coomaraswamy was refuting notions of
the purely ‘‘decorative’’ or ‘‘aesthetic’’ quality of ornaments, his argument still
depended on the kind of oppositions between communication and pleasure,
necessity and luxury, adequacy and excess, that make the idea of ornament
thinkable in the first place. His defense of alankaram also seems to allude to
the authority of Sanskrit as a classical language in which the role of alankarams
is central (Coomaraswamy 1939).
23 The difference between note and swara is sometimes explained thus: the swara
refers to the space between two notes, for example C and D; it is thus a kind
of placeholder. When someone sings ‘‘ri’’ it doesn’t necessarily mean they are
singing a D; rather, that note is the tone center for what they are singing.
In some ragas ‘‘ri’’ is actually specified to be sung or played as an oscillation
between D and E; in that sense the swara ‘‘ri’’ is not a ‘‘note.’’
24 Another proponent of staff notation was Ernest Clements, member of the
Indian civil service in the Bombay Presidency and a participant in the great
sruti debates in the years following the first All-India Music Conference in
1916. Clements argued, in his Introduction to the Study of Indian Music (1913),
that the staff notation was the most logical kind of notation because of its
economy: ‘‘It possesses a distinct advantage over any method which requires
the eye to follow one set of signs for melody (svara) and an entirely distinct
set for time (laya)’’ (16). He demonstrated how the ‘‘middle octave,’’ the range
commonly employed in Indian music, ‘‘fit easily’’ onto the staff. He warned,
however, that those who had ‘‘become ensnared by Western notation’’ were
unwittingly accepting the law of equal temperament. Falling for the neat ap-
pearance of the notation, they were not aware of its implications, the fact
that it consistently blurred together differences of a quarter-tone (35). If the
notation was really going to be self-sufficient, musicians had to see the exact
interval portrayed in the notation itself. Accordingly, he used a slash through
the sign for flat and natural notes to show that twice as many notes as regular
staff notation illustrated existed in the scale. The use of notation demanded
a standardization of the Indian scale, the mathematical calculation of each
note. Only then would the gap between ear and eye be closed, and Indian
music become truly visible. In 1920 Clements undertook to show practically
how his notation worked by translating compositions written down in Indian
sargam notation into staff notation. For the sake of visibility, Clements did
away with all gamaka symbols.
25 Bergeron illuminates the idea of ‘‘legibility’’ in her discussion of late-
nineteenth-century debates about notation for Gregorian chant. ‘‘The notion
of legibility referred to the way printed characters enabled the act of read-
ing.’’ The ideal of legibility implied words (or notation, in this case) so clear
that they would seem to be transparent, to disappear, leaving the impression
of direct vocal communication. Thus, ‘‘a beautiful page had the power . . . to

notes to chapter five 317

 transform what was known as ‘silent reading’ into a blissful interval of listen-
ing’’ (1998, 58–59).
26 On the symbolism of mother’s milk in the context of Tamil language devo-
tion in the 1930s, see Ramaswamy 1997, 106–8.
27 Sangīta Sarvartha Sāra Sangrahamu, printed in Telugu in 1859, was supposedly
the first book of Karnatic compositions published. But, according to M. Hari-
haran, it was a mere compendium with no pedagogical project (Clements
1920, preface).
28 A. R. Venkatachalapathy has documented similar changes that took place in
reading practices with the emergence of printed books. He notes the rise of
silent reading among the middle classes in Tamil-speaking areas at the turn of
the twentieth century, in contrast to the older practice of reading aloud and
memorizing from palm-leaf manuscripts. Whereas palm-leaf manuscripts,
written in a continuous, run-on fashion, left many possibilities of interpre-
tation for the reader, printed books and the introduction of punctuation in
Tamil gave rise to new modes of reading, like scanning and browsing, as well
as to the feeling of an inexhaustible quantity of printed material available
to be read. The illusion of plenty became part of this experience of reading
(1994, 282).
29 The epigraph reflects a line spoken by a character in Chinnaswamy Mudaliar’s
play, ‘‘Saraswatia Redux,’’ modeled on Shakespearean comedy, in which the
theoretical aspects of Karnatic music, including the 72-melakarta raga sys-
tem, are explained. The play was published in several issues of Chinnaswamy
Mudaliar’s journal, Oriental Music in European Notation, between 1892 and 1895.
30 The 72-melakarta raga scheme appeared in Venkatamakhi’s treatise, Catur-
dandi Prakasika.
31 A full description of the system is beyond the scope of this chapter; it is more
the idea of such a system that I am trying to convey. Sambamoorthy’s The
72-Melakarta System (1961) provides a lucid demonstration of how the sys-
tem works.
32 If the magic of numbers was not sufficient, the system could also be made
to work with letters. The first two syllables of each raga name corresponded
to two letters in an ordered table of numbered Sanskrit letters. By taking the
corresponding numbers for these letters in the alphabet table, and reversing
them, one would arrive at the number of the melakarta raga in the 72 mela-
karta raga table, and with this number one could then determine the notes
of the raga (Sambamoorthy 1961, 14–16).
33 Such a scheme resonates with Michel Foucault’s description of the table and
its importance in what he calls the classical episteme, which he identifies with
the seventeenth century. In the classical idea of ‘‘natural history’’ as formu-
lated by Linnaeus, the table functions as ‘‘a grid [that] can be laid out over
the entire vegetable or animal kingdom. Each group can be given a name.

318 notes to chapter five

 With the result that any species, without having to be described, can be des-
ignated with the greatest accuracy by means of the names of the different
groups in which it is included. . . . Once the system of variables . . . has been
defined at the outset, it is no longer possible to modify it, to add or subtract
even one element. . . . To know what properly appertains to one individual
is to have before one the classification—or the possibility of classifying—all
others’’ (Foucault 1970, 141–44).
34 This idea was pursued by almost all those who wrote about Indian music in the
late nineteenth century and the early twentieth (e.g., Clements 1920). It was
undoubtedly influenced by the ideas about language ‘‘families’’ in philology.
35 N. S. Saminathan, personal communication, Madurai, June 2002.
36 It is interesting to note that there was no punctuation as we know it in Indian
languages such as Tamil or Telugu before the advent of printing and modern
prose forms that arose in the late nineteenth century and early twentieth. Full
stops were represented, but there were no commas, exclamation points, or
other marks. Such punctuation in part reflects the effect of English on Indian
languages (Dharwadker 1997, 108–33).
37 A meend is a slow sliding from one note to the next, especially in the alap of
Hindustani music.
38 Compare Katherine Bergeron’s argument in the conclusion of Decadent En-
chantments.
39 In Hindu Advaita Vedanta philosophy the concept of Brahman is often trans-
lated as ‘‘ultimate reality,’’ the source from which all emanates and returns, the
unchanging absolute. Jeaneane Fowler describes it as the ‘‘Unmanifest Source
of the manifestation of cause-effect processes in the universe’’ (2002, 51).

6 Fantastic Fidelities
1 Harrison suggests that we gather ‘‘significant examples of where things are
heard differently, or where the description of listening undergoes major
changes and of where listening seems to take on an historically changed posi-
tion within the modal construction of self and psyche’’ (1996, 22).
2 For instance, my violin teacher used to refer to her guru (who also happened
to be her father) as ‘‘my god, my mother, and my father.’’ I learned that she ex-
pected me to call her ‘‘Amma’’ (mother). Similarly, a mridangist with whom
I studied in Madurai referred to his guru, C. S. Sankarasiva Bhagavatar, as
his father. It was only later that I learned that he had ‘‘adopted’’ his guru as a
father.
3 Sangati refers to variation on a musical phrase; compositions are elaborated in
Karnatic music by adding sangatis.
4 This and the following passages are my translations of Malan’s story, originally
published in Kalki and reprinted in Āu, a collection of Tamil short stories.

notes to chapter six 319

 Malan himself edited the collection, meant to be representative of Tamil writ-
ing of the twentieth century. ‘‘Vitvān’’ is placed at the end of the second vol-
ume, and Malan places it in the genre of Tamil science fiction (viññāa ciru-
kataikaḷ) (1981, xiii).
5 Shadjam is the long name of sa, the tonic or first note of the scale. Each note
is called a swaram.
6 The word kaṇakku, used here to refer to frequencies of sound waves, is also
the word for the rhythmic improvisation, based on calculations, that Karnatic
musicians perform. It is thus simultaneously in the realms of music theory
and of technology. Therein lies its threat: if a computer can master one aspect
of kaṇakku, can it not master the other as well?
7 The Indian part of the Gramophone Company’s first Far Eastern Tour in 1902–
1903 was limited to Calcutta, where Gaisberg recorded, among others, the
female courtesan singers Janki Bai and Gauhar Jan (Gaisberg 1942, 56–57).
The second and third recording tours of India, engineered by William Sink-
ler Darby and William Gaisberg (Fredrick’s younger brother), respectively,
included other locations in India. Thus, the first recordings were made in
Madras between 1904 and 1907. They included mostly vocal artists, with the
exception of a few, such as the violinist Pudukottai Narayanaswamy Iyer, the
harmonium player Madras Chetty Babu, and bands such as the Tanjore Band
and the Madras-based band of Govindasamy Dasu (Kinnear 1994: 73–266).
The female singers recorded were all from the devadasi community; they sang
classical kritis as well as Telugu and Tamil padams and javalis (songs, associated
with dance, which came to be considered ‘‘light’’ classical). Among these were
some with wholly or partly English lyrics, with titles like ‘‘Hello, How Am I?’’
and ‘‘What a Beautiful Lady.’’ Thus, from the start, the Gramophone Company
did not record only classical music. A large number of recordings were also
made of the Tamil comedians Venugopal Chary and ‘‘Professor Naidu.’’ These
included comedic scenes like ‘‘Brahmin Going to a Dancing Girl’s House,’’
‘‘Street Life in Madras,’’ and ‘‘Railway Station Scene,’’ as well as a genre of imi-
tations: of Madras beggars, of birds, of a jutka driver, of a passing train, and
even of well-known Karnatic singers. Such imitations were well-suited to the
gramophone, which allowed everyday events to become comic by their isola-
tion and repetition. Between 1900 and 1910, the Gramophone Company (later
to become His Master’s Voice Company) made over four-thousand recordings
in India, more than in any other single country on its world tours (Gronow
1981, 255). By 1905 the Talking Machine and Indian Record Company had
started a branch in Madras (Kinnear 1994, 10). The Gramophone Company
published its first South Indian catalog in 1905 and advertised prominently in
English and vernacular-language newspapers (ibid., 24). Its main office and
record-pressing factory, established in 1908, were in Calcutta. The company
maintained its monopoly by associating with local record-selling agents in

320 notes to chapter six

 Madras and other cities (G. N. Joshi 1987, 148). Electrical recording was first
introduced in India in 1925, and the magnetic tape recorder became available
around 1950 (ibid., 148–49).
8 Gamakas, usually translated as ‘‘ornaments’’ or ‘‘embellishments,’’ can be
thought of as specified ways of getting from one note to the next in Kar-
natic music. Gamakas are highly individual to different musicians and are not
included in printed notation.
9 Sterne notes, as does Taussig in Mimesis and Alterity, that the ‘‘encounter be-
tween the modern and the nonmodern lies at the heart of moments of
fascination with watching others’ listening to reproduced sound’’ (Sterne
2000, 14). Erika Brady, in A Spiral Way, also noted the pervasiveness of such
‘‘rube’’ stories in early encounters with technologies of sound recording (1999,
27–34).
10 In a 1934 report, G. W. Kaye of the National Physical Laboratory, Tedding-
ton, wrote, ‘‘Method of Experiment: The notes and phrases concerned were
played by Mr. Subrahmanya on his violin. The sound from the violin was re-
ceived by a condenser microphone distant about 2 ft. from the violin, and
connected by a valve amplifier to one of the vibrators of a Duddell Oscillo-
graph. The wave form of the sound was recorded on photographic paper
by means of a revolving drum camera attached to the oscillograph. A time
scale was provided on each record by a second vibrator that registered the
wave form of the electrical output from a standard valve-maintained tun-
ing fork, operating at a frequency of 1000 cycles per second. The player and
the receiving microphone were situated in a lagged cabinet so as to avoid
as far as possible, any interference from extraneous noise. Communication
between the player and the operator of the recording apparatus was main-
tained by a system of visual signals, controlled by the mechanism of the shutter
on the recording camera, the player being warned one second in advance of
the opening of the shutter and also at the commencement and conclusion of
the exposure. The duration of the exposure was approximately one second in
each case’’ (in Subrahmanya Ayyar 1939, 134–36).
11 Precisely because it could not talk back, the phonograph could only hear.
This notion of the separability of functions, according to Kittler, underlay the
‘‘discourse network’’ of 1900. Emerging around 1900, theories of the localiza-
tion of brain functions and the idea of testing humans for speech, hearing,
and writing as isolated functions modeled themselves on the phonograph,
which performed only the function of ‘‘hearing’’ (1999, 38; 1990, 214). As if
by compensation, musicians seemed to cease to hear at all. G. N. Joshi, who
worked for the Gramophone Company as a recording officer between 1938
and 1973, recalled, ‘‘When I was first recorded in the year 1932, I was lured to
the studios to record only two of my most popular songs. At the recording
session, I became so involved and excited at the prospect of being recorded

notes to chapter six 321

 that instead of just two songs, when the recording session was wound up I
discovered that I had actually recorded fourteen songs!’’ (1987, 151).
12 V. A. K. Ranga Rao, personal communication, Madras, June 2002. Saraswati
Stores, Odeon, and Columbia Records employed Turaiyur Rajagopala Sharma
as a ‘‘tutor,’’ while the Gramophone Company employed T. A. Kalyanam.
13 A recording by the violinist Dwaram Venkataswamy Naidu from the early
1940s features a ragam-tanam-pallavi, an improvisational item that in a con-
cert would have taken about an hour at that time, compressed into four seg-
ments of exactly three-and-a-half minutes each. The present-day idea that a
musician should, when doing raga alapana, make the raga clear from the very
first phrase, rather than keeping the listeners in suspense, probably first gained
its urgency from the demands of recording. Musicians speak of five-minute
alapanas or twenty-minute alapanas as choices they make depending on the
amount of time available—the idea being to carefully plan one’s spontaneity.
14 Jonathan Sterne traces the emergence of the idea of the ear as a hearing
mechanism in relation to the science of otology and the early technologies
of sound reproduction, notably the ear phonautograph, in the mid- to late
nineteenth century. The technologies of sound reproduction, Sterne writes,
depended on the ‘‘isolation, separation, and transformation of the senses,’’ and
the idea that the senses could be understood as mechanisms (2003, 50–51).
15 Sterne writes that ‘‘people had to learn how to understand the relations be-
tween sounds made by people and sounds made by machines’’ (2003, 216).
The idea of sound fidelity and the idea of a recording as a reproduction of an
‘‘original’’ performance are not natural results of the process of recording but
particular ways of conceiving of that process, Sterne argues.
16 Stephen Hughes suggests that ‘‘popular publishing and the press constituted
a new vehicle for extending gramophone music well beyond the purchasing
of records and machines. These printed materials were more than a medium
for promoting record sales; they also created a significant new circulation,
allowed for a new mode of engagement and provided a new forum for the
public discussion of gramophone music’’ (2002, 456).
17 See Farrell 1997 for a reproduction of one of these advertisements (32).
18 The phrase ‘‘she herself has melted’’ is used extensively in Tamil bhakti poetry
to describe a devotee’s relationship with God. In using the idiom of bhakti,
Raman thus presented M.S. as an exemplary devotee filled with love of God.
19 In ‘‘The Form of the Phonograph Record’’ Theodor Adorno wrote, ‘‘Through
the phonograph record time gains a new approach to music. It is not the time
in which music happens, nor is it the time which music monumentalizes
by means of its ‘style.’ It is time as evanescence, enduring in mute music.
If the ‘modernity’ of all mechanical instruments gives music an age-old ap-
pearance—as if, in the rigidity of its repetitions, it had existed forever, . . .
then evanescence and recollection . . . [have] become tangible and manifest
through the gramophone records’’ (1934, 38).

322 notes to chapter six

20 In 1938, inspired by such potential, Rabindranath Tagore himself wrote a
poem in honor of All India Radio, entitled ‘‘Akashvani’’ (the Hindi name for
All India Radio):
Hark to Akashvani up-surging
From here below.
The earth is bathed in Heaven’s glory
Its purple glow.
Across the blue expanse is firmly planted
The altar of the Muse.
The lyre unheard of Light is throbbing
With human hues.
From earth to heaven, distance conquered
In waves of light
Flows the music of man’s divining,
Fancy’s flight.
To East and West speech careers,
Swift as the Sun.
The mind of man reaches Heaven’s confines,
Its freedom won.
(Akashvani [English version], 24 July 1977, 1).
21 Meanwhile, the princely states of Mysore and Trivandrum set up radio sta-
tions in 1935 and 1939, respectively.
22 In the 1930s, the Madras Corporation installed permanent loudspeakers at six
open places in Madras city to promote communal listening (Hughes 2002,
459).
23 Rudolph Mrazek, similarly, writes in relation to the Netherlands East Indies
Radio that the technology of radio provided a metaphor for the Dutch colo-
nial project of keeping natives and Europeans insulated from contact with
each other in the decades just before Indonesian independence (1997).
24 S. Rajam, personal communication, Madras, December 2003.
25 Ibid.
26 Other educational programs on the radio used to include music quiz shows,
according to Mullick, and a program of introducing classical music through
film songs (1974, 41). The ‘‘National Programme’’ used to have a theme,
making it appropriately didactic (S. Venkataraman, personal communication,
Madras, June 1998). Currently, there are several music quiz shows on Tamil
cable-television channels in which players are encouraged to show off their
encyclopedic knowledge of Karnatic compositions.
27 R. Rangaramanuja Ayyangar was also the author of A History of Indian Music
from Vedic Times to the Present.
28 Keskar’s remarks reflected the general distrust of musicians in the revival-
ist discourse on Hindustani music: the idea that scholars and musicologists,

notes to chapter six 323

 not musicians themselves, possessed the correct and authentic knowledge of
music. As Janaki Bakhle has shown, the implication that [Muslim] musicians
were ignorant, illiterate, and ill-suited to be the custodians of their own musi-
cal tradition, and that the tradition needed to be saved by [Hindu] musicolo-
gists was part of the partitioning of the musical and cultural sphere into Hindu
and Muslim in the early twentieth century (Bakhle 2005, 195).
29 Tampulam is a presentation of betel leaves and prepared betel nut given by a
gracious host to guests when they depart.
30 Sruti boxes produce continuous sound (as from a bellows instrument), while
electronic tamburas produce pulsating sound (as from strings being plucked
in succession). However, both are referred to colloquially as ‘‘sruti boxes.’’
31 G. Raj Narayan, email to author, November 2002.
32 Radel’s latest invention in 2003 was a digital veena, which is basically a syn-
thesizer in the shape of a veena.
33 In late 2003 Radel came out with another model that has four ‘‘strings’’ which
can be sounded by a mere light touch. But the instrument, like other Radel
products, can also be switched on to play automatically.

Afterword
1 The relationship between rise of the idea of musical ‘‘classics’’ in England and
the emergence of a leisure class during the Industrial Revolution has been
touched on by William Weber, but there is little direct mention in Weber or
elsewhere of the enabling role of colonialism in these developments (1992,
1–9).

324 notes to afterword

$ Works Cited

Abbate, Carolyn. 1991. Unsung voices: Opera and musical narrative in the nineteenth
century. Princeton: Princeton University Press.
Achariyar, Deivasikamani. 1949. Mēṭaittami [Platform Tamil]. Repr. chaps. 1–18,
Madras: Palms Printers, 1987.
Adorno, Theodor. 1934. ‘‘The form of the phonograph record’’ and ‘‘The curves
of the needle.’’ Trans. Thomas Levin. October 55:48–56.
Adyanthaya, Rao Bahadur N. M. 1949. Music and healing [brochure]. Mangalore.
Akashvani. 24 July 1977. New Delhi: Ministry of Information and Broadcasting.
Allen, Matthew. 1997. Rewriting the script for South Indian dance. The Drama
Review 41, no. 3:63–100.
All-India Music Conference proceedings. 1916–1918. Baroda: All-India Music
Conference.
Ammaiyar, Ramamirtham. 1936. Tācikaḷi mocāvalai allatu matipea maiar [The
treacherous net of the devadasis, or a minor grown wise]. Madras: Pearl Press.
Anandi, S. 1991. Representing devadasis: Dasigal Mosavalai as a radical text. Eco-
nomic and Political Weekly 26, nos. 11–12:739–46.
. 1997. Sexuality and nation: ‘‘Ideal’’ and ‘‘other’’ woman in nationalist
politics, Tamilnadu, c. 1930–47. South Indian Studies 4:195–217.
Appadurai, Arjun. 1981. Worship and conflict under colonial rule: A South Indian case.
Cambridge: Cambridge University Press.
Arasaratnam, S. 1979. Trade and political dominion in South India, 1750–1790:
Changing British-Indian relationships. Modern Asian Studies 13, no. 2:19–40.
Arooran, Nambi. 1980. Tamil renaissance and Dravidian nationalism. Madurai: Koodal
Publishers.
Awasthy, G. C. 1965. Broadcasting in India. Bombay: Allied Publishers.
Bakhle, Janaki. 2005. Two men and music: Nationalism in the making of an Indian clas-
sical tradition. New York: Oxford University Press.
Bakhtin, Mikhail. 1981. The dialogic imagination. Austin: University of Texas.
. 1984. Problems of Dostoevsky’s poetics. Trans. Caryl Emerson. Minneapolis:
University of Minnesota Press.
Balachander, S. 1985. Eutiār puttakattai! Kiappiār pūtattai! [He wrote a book and
kindled the genie!]. Self-published pamphlet.
Banerjee, Sumanta. 1989. The parlour and the streets: Elite and popular culture in
nineteenth-century Calcutta. Calcutta: Seagull Books.
Bate, John B. 2000. Mēṭaittami: Oratory and democratic practice in Tamil Nadu.
PhD diss., University of Chicago.
Bauman, Richard, and Charles Briggs. 2003. Voices of modernity: Language ideologies
and the politics of inequality. Cambridge: Cambridge University Press.
Bergeron, Katherine. 1998. Decadent enchantments: The revival of Gregorian chant at
Solesmes. Berkeley: University of California Press.
Bhabha, Homi. 1984. Of mimicry and man: The ambivalence of colonial dis-
course. October 28:125–33.
Bharathi, Suddhananda. 1968. Saint Thyagaraja: The divine singer. Madras: Yoga
Samaj.
Bharatiyar, Subramania. 1909. Caṅkīta Viṣayam [The issue of music]. Repr. in
Pāratiyār Katturaikaḷ [Essays of Bharatiyar] Ed. A. Tirunavukkarasu, 211–24.
Madras: Vanathi Patipakkam, 1981.
Bharucha, Rustom. 1995. Chandralekha: Woman dance resistance. New Delhi: Harper
Collins.
Bhaskaran, Theodore. 1981. The message-bearers: Nationalist politics and the entertain-
ment media in South India, 1880–1945. Madras: Cre-A.
. 1997. The eye of the serpent: An introduction to Tamil cinema. Madras: Orient
Longman.
Bhatkhande, V. 1917. A short historical survey of the music of upper India. (Lec-
ture presented at the first All-India Music Conference, Baroda, India, 1916.)
Bombay: Bombay Samachar.
Blackburn, Stuart. 2000. Corruption and redemption: The legend of Valluvar and
Tamil literary history. Modern Asian Studies 34, no. 2:449–82.
Brady, Erika. 1999. A spiral way: How the phonograph changed ethnography. Jackson:
University Press of Mississippi.
Bullard, Beth. 1998. Winds of change in South Indian music: The flute revived,
recasted, regendered. PhD diss., University of Maryland.
Burton, Antoinette. 1994. Burdens of history: British feminists, Indian women, and im-
perial culture, 1865–1915. Chapel Hill: University of North Carolina Press.
Caldwell, Robert. 1856. A comparative grammar of the Dravidian or South Indian family
of languages. Rev. ed., ed. J. L. Wyatt and T. Ramakrishna Pillai. London:
Kegan Paul, Trench, Trübner, 1913. Repr. New Delhi: Asian Educational Ser-
vices, 1987.
Candy, Catherine. 1994. Relating feminisms, nationalisms, and imperialisms: Ire-
land, India, and Margaret Cousins’s sexual politics. Women’s History Review 3,
no. 4:581–594.
Ceai caṅkīta vitvat capai: 12-āvatu mēlam [Madras music conference: 12th
meeting]. 1938. Bharata Mani (1938): 372–74.

326 works cited

Chakrabarty, Dipesh. 2000. Provincializing Europe: Postcolonial thought and historical
difference. Princeton: Princeton University Press.
Chatterjee, Partha. 1989. The nationalist resolution of the women’s question. In
Recasting women: Essays in colonial history. Ed. Kumkum Sangari and Sudesh
Vaid, 233–253. New Delhi: Kali for Women.
. 1993. The nation and its fragments: Colonial and postcolonial histories. Prince-
ton: Princeton University Press.
Chatterji, P. C. 1998. The adventure of Indian broadcasting: A philosopher’s autobiogra-
phy. Delhi: Konark Publishers.
Chaudhuri, Amit. 1999. Afternoon raag [novella]. In Freedom Song, 123–239. New
York: Alfred Knopf.
Chidambaranatha Mudaliar, T. K. 1936. Itaya oli [The sound of the heart]. In Itaya
oli, 139–58. Madras: Shanti Press.
. 1941. Caṅkītamum cakityamum [Music and lyrics]. In Itaya oli, 41–51.
Madras: Shanti Press.
Chinnaswamy Mudaliar, A. M. 1892. Oriental music in European notation. Madras:
Ave Maria Press.
Chion, Michael. 1982. The voice in cinema. Trans. Claudia Gorbman. New York:
Columbia University Press.
Clayton, Jay. 1997. The voice in the machine: Hardy, Hazlitt, James. In Language
machines: Technologies of literary and cultural production. Ed. Jeffrey Masten, Peter
Stallybrass, and Nancy Vickers, 209–32. New York: Routledge.
Clayton, Mark. 1999. A. H. Fox-Strangways and the music of Hindostan: Revisit-
ing historical field recordings. Journal of the Royal Musical Association 124:88–
118.
Clements, Ernest. 1913. Introduction to the study of Indian music. Chandigarh: Abishek
Publications, 1992.
. 1920. Ragas of Tanjore. New Delhi: Caxton Publications, 1988.
Cohn, Bernard. 1968. Notes on the history of the study of Indian society and cul-
ture. In Structure and change in Indian society. Ed. Milton Singer and Bernard
Cohn, 3–28. Chicago: Aldine.
. 1987. The command of language and the language of command. Subaltern
Studies 4:276–329. New Delhi: Oxford University Press.
Connor, Steven. 2000. Dumbstruck: A cultural history of ventriloquism. Oxford: Ox-
ford University Press.
Coomaraswamy, Ananda K. 1909. Indian music. Repr. in Essays in national idealism,
166–200. Delhi: Munshiram Manoharlal, 1981.
. 1909. Gramophones—and why not? Repr. in Essays in national idealism,
201–6. Delhi: Munshiram Manoharlal, 1981.
. 1924. The dance of Siva: Fourteen Indian essays. London: Simpkin, Marshall,
Hamilton, Kent.
. 1939. Ornament. Repr. in Coomaraswamy. Ed. Roger Lipsey, vol. 1:241–53.
Princeton: Princeton University Press, 1977.

works cited 327

Cousins, Margaret. 1935. Music of the Orient and the Occident. Madras: B. G. Paul.
. 1940. The late maharaja of Mysore’s patronage of music. Uttara mandra 1,
no. 1 (March 1970): 142.
Cutler, Norman. 1987. Songs of experience: The poetics of Tamil devotion. Blooming-
ton: Indiana University Press.
Daniel, E. Valentine. 1984. Fluid signs: Being a person the Tamil way. Berkeley: Uni-
versity of California Press.
Danielson, Virginia. 1997. The voice of Egypt: Umm Kulthum, Arabic song, and Egyp-
tian society in the twentieth century. Chicago: University of Chicago Press.
Day, Charles. 1891. Music and musical instruments of Southern India and the Deccan.
London and New York: Novello, Ewer.
de Alwis, Malathi. 1999. ‘‘Respectability,’’ ‘‘modernity,’’ and the policing of ‘‘cul-
ture’’ in colonial Ceylon. In Gender, sexuality, and colonial modernities. Ed.
Antoinette Burton, 177–92. London and New York: Routledge.
de Certeau, Michel. 1984. The practice of everyday life. Trans. Steven Rendall. Berke-
ley: University of California Press.
. 1988. The writing of history. Trans. Tom Conley. New York: Columbia Uni-
versity Press.
Derrida, Jacques. 1974. Of grammatology. Trans. Gayatri Spivak. Baltimore: Johns
Hopkins Press.
Devadoss, Manohar. 1997. Green well years. Madras: East West Books.
Devi, Rukmini. N.d. The creative spirit. Madras: Theosophical Society.
. N.d. Dance and music. Madras: Theosophical Society.
. N.d. Woman as artist. Madras: Theosophical Society.
Dharwadker, Vinay. 1997. Print culture and literary markets in colonial India.
In Language machines: Technologies of literary and cultural production. Ed. Jeffrey
Masten, Peter Stallybrass, and Nancy Vickers, 108–33. New York: Routledge.
Dikshitar, Subbarama. 1904. Sangīta sampradāya pradārṣini. Madras: Madras Music
Academy.
Dirks, Nicholas. 1993. The hollow crown: Ethnohistory of an Indian kingdom. Ann
Arbor: University of Michigan.
Dixon, Joy. 1999. Ancient wisdom, modern motherhood: Theosophy and the
colonial syncretic. In Gender, sexuality, and colonial modernities. Ed. Antoinette
Burton, 193–206. London and New York: Routledge.
Dolar, Mladen. 1996. The object voice. In Gaze and voice as love objects. Ed. Slavoj
Žižek and Renata Salecl, 7–31. Durham: Duke University Press.
Dugger, Celia. 2000. A world pays tribute to India’s master drummer. New York
Times, 14 February. § A, 4.
Dunn, Leslie, and Nancy Jones, eds. 1995. Embodied voices: representing female vocality
in Western culture. Cambridge: Cambridge University Press.
Durayappa Bhagavatar, Gottuvadyam. 1941. Apiayam vēṇṭām [We don’t want
abhinaya]. Nāṭṭiyam 1, no. 1:51–52.

328 works cited

Ellarvi. 1963. Tōṭi aṭaku: Ciu campavaṇkaḷ [Pawning todi raga: Small vignettes].
Madras: National Art Press.
Engel, Carl. 1883. Researches into the early history of the violin family. London: Novello,
Ewer.
Engh, Barbara. 1997. After ‘‘His master’s voice’’: Post-phonographic aurality. PhD
diss., University of Minnesota.
Erdman, Joan. 1978. The maharaja’s musicians: Performance at Jaipur in the nine-
teenth century. In American studies in the anthropology of India. Ed. Sylvia Vatuk,
342–67. Delhi: Manohar.
Erlmann, Veit. 1999. Music, modernity and the global imagination: South Africa and the
West. New York: Oxford University Press.
Farrell, Gerry. 1997. Indian music and the West. Oxford: Oxford University Press.
Feld, Steven, and Aaron Fox. 1994. Music and language. Annual Review of Anthro-
pology 23:25–53.
Feld, Steven, Aaron Fox, Thomas Porcello, and David Samuels. 2004. Vocal an-
thropology. In A companion to linguistic anthropology. Ed. Alessandro Duranti,
321–45. Malden, Mass.: Blackwell.
Fielden, Lionel. 1940. Report on the progress of broadcasting in India up to the 31st March
1939. Delhi: Manager of Publications.
Foucault, Michel. 1970. The order of things: An archaeology of the human sciences. New
York: Random House.
. 1977. Discipline and punish: The birth of the prison. Trans. Alan Sheridan.
New York: Pantheon Books.
. 1979. What is an author? In The Foucault reader. Ed. Paul Rabinow, 101–20.
New York: Pantheon Books, 1984.
Fowler, Jeaneane. 2002. Perspectives of reality: An introduction to the philosophy of
Hinduism. Brighton: Sussex Academic Press.
Fox-Strangways, A. 1914. The music of Hindostan. New Delhi: Mittal Publishers.
Frank, Felicia Miller. 1995. The mechanical song: Women, voice, and the artificial in
nineteenth-century French narrative. Stanford: Stanford University Press.
Frith, Simon. 1996. The Voice. In Performing rites: On the value of popular music,
183–202. Cambridge, Mass.: Harvard University Press.
Gaisberg, Fredrick. 1942. The music goes round. New York: Macmillan.
Gandhi, Mohandas K. 1958. Collected works of Mahatma Gandhi, vol. 66. New Delhi:
Publications Division, Ministry of Information and Broadcasting, Govern-
ment of India.
Ganghadar, C. N.d. Theory and practice of Hindu music and the vina tutor. Madras:
Methodist Publishing House.
Gaston, Anne-Marie. 1996. Bharata Natyam: From temple to theatre. New Delhi:
Manohar.
Gayan Samaj. 1887. Madras jubilee Gayan Samaj proceedings. Madras: Gayan Samaj.
Geetha, V., and S. Rajadurai. 1998. Towards a non-Brahmin millennium: From Iyothee
Dass to Periyar. Calcutta: Samya.

works cited 329

Goehr, Lydia. 1992. The imaginary museum of musical works. Oxford: Clarendon.
Goswamy, B. N. 1996. Broadcasting: New patron of Hindustani music. Delhi: Sharada
Publishing House.
Gronow, Pekka. 1981. The record industry comes to the Orient. Ethnomusicology
25, no. 2:251–84.
Guy, Randor. 1997. Starlight, starbright: The early Tamil cinema. Chennai: Amra Pub-
lishers.
Hancock, Mary Elizabeth. 1999. Womanhood in the making: Domestic ritual and public
culture in urban South India. Berkeley: University of California Press.
Hansen, Kathryn. 1998. Stri Bhumika: Female impersonators and actresses on the
Parsi stage. Economic and Political Weekly 33, no. 35:2291–2300.
Harinagabhushanam. 1929. The essentials of Karnatic music. Triveni 2, no. 3:54–
202.
Harrison, Martin. 1996. Toward a history of listening. Essays in Sound 3:21–33.
Hartshorne, Charles and Paul Weiss, eds. 1931–58. Collected papers of Charles Sanders
Peirce, vol. 2. Cambridge, Mass.: Harvard University Press.
Head, Raymond. 1985. Corelli in Calcutta: Colonial music-making in India in the
seventeenth and eighteenth centuries. Early Music 13:548–53.
Higgins, John. 1975. From prince to populace: Patronage as a determinant of
change in South Indian music. Asian Music 7, no. 2:20–26.
Hill, Jane. 1996. The consciousness of grammar and the grammar of conscious-
ness. In The matrix of language: Contemporary linguistic anthropology. Ed. Donald
Brenneis and Ronald Macaulay, 307–23. Boulder, Colo.: Westview.
Hughes, Stephen. 1996. The pre-Phalke era in South India: Reflections on the
formation of film audiences in Madras. South Indian Studies 2, no. 2:161–204.
. 2002. The ‘‘music boom’’ in Tamil South India: Gramophone, radio, and
the making of mass culture. Historical Journal of Film, Radio, and Television 22,
no. 4:445–73.
Inden, Ronald. 1986. Orientalist constructions of India. Modern Asian Studies 20,
no. 3:401–46.
Inoue, Miyako. 2002. Gender, language, and modernity: Toward an effective his-
tory of Japanese women’s language. American Ethnologist 29, no. 2:392–422.
. 2003. The listening subject of Japanese modernity and his auditory
double: Citing, sighting, and siting the modern Japanese woman. Cultural
Anthropology 18, no. 2:156–93.
Irschick, Eugene. 1969. Politics and social conflict in South India: The non-Brahman
movement and Tamil separatism, 1916–1929. Berkeley: University of California
Press.
. 1986. Tamil revivalism in the 1930s. Madras: Cre-A.
. 1994. Dialogue and history: Constructing South India, 1795–1895. Berkeley:
University of California Press.
Ivy, Marilyn. 1995. Discourses of the vanishing: Modernity, phantasm, Japan. Chicago:
University of Chicago Press.

330 works cited

Jackson, William. 1991. Tyagaraja: Life and lyrics. Madras: Oxford University Press.
Jain, S. P. 1985. The art of broadcasting. New Delhi: Intellectual Publishing House.
Jayarama Iyer, T. K. 1965. The violin in Karnatic music. Indian Music Journal (Delhi
Sangeeta Samaj), no. 4:27–28.
Jayaraman, R. 1990. Raja Serfoji. In pt. 2 of Māmaar Carapoji, 61–70. Tanjavur:
Saraswati Mahal Library.
Jazz solos. 2000. Letters to the editor, New York Times. 11 June.
Jinarajadasa, C. 1928. New beginnings in Indian culture. Triveni 1, no. 1:3–12.
Johannes, T. C. R. 1912. Bhārata Sangīta Svāya Bōdhini [Indian music self-
instructor]. Madras.
Jordan, Kay. 1993. Devadasi reform: Driving the priestesses or the prostitutes out
of Hindu temples? In Religion and law in independent India. Ed. Robert Baird,
257–77. Delhi: Manohar.
Joshi, Babu Rao. 1963. Understanding Indian music. Bombay: Asia Publishing House.
Joshi, G. N. 1987. A concise history of the phonograph in India. Popular Music 7,
no. 2:147–56.
Kalki [R. Krishnamoorthy]. 1941. Nārata kāam [The sound of Narada]. Kalki
(1 October): 24.
. 1942. Icai virutu [Musical feast]. Kalki (1 August): 75.
. 1943a. Tami icaiyum Piramāṇarkaḷum [Tamil icai and Brahmins]. Kalki
(10 August): 12–13.
. 1943b. Tāymoi pakaimai [Hatred of the mother tongue]. Kalki (12 De-
cember): 13–16.
. 1945. Icait taṭṭu [Music record]. Kalki (1 September): 47.
. 1946. Mīra Kītaṅkaḷ [Meera songs]. Kalki (17 February): 31.
. 1949. Āṭal Pāṭal [Song and dance]. Kalki (9 January): 15.
. 1951. Caṅkītam paam paṇṇukiatu [Music makes money]. Kalki (26 Oc-
tober): 14.
. 1954. Vīai Ṣamuka Vaṭivu [Veena Shanmugha Vadivu]. Kalki (28 Feb-
ruary): 21.
Kalyanasundaram, A. N. 1938. Ārmōiya pahiṣkāram [Harmonium boycott]. Bha-
rata Mani (1938): 206–10.
Kannabiran, Kalpana. 1995. Judiciary, social reform, and debate on ‘‘religious pros-
titution’’ in colonial India. Economic and Political Weekly 30, no. 43:59–69.
Keane, Webb. 1997. From fetishism to sincerity: On agency, the speaking sub-
ject, and their historicity in the context of religious conversion. Comparative
Studies in Society and History 39, no. 4:674–93.
Kersenboom-Story, Saskia. 1987. Nityasumangali: Devadasi tradition in South India.
Delhi: Motilal Banarsidass.
. 1995. Word, sound, image: The life of the Tamil text. Oxford: Berg.
Keskar, B. V. 1957. Indian music: Problems and prospects. Bombay: Popular Prakashan.
Kinnear, Michael. 1994. The Gramophone Company’s first Indian recordings, 1899–1908.
Bombay: Popular Prakashan.

works cited 331

Kittappa, K. P. 1993. Pārata icaiyum tañcai nalvarum [Indian music and the Tanjore
Quartette]. Tanjavur: Tamil University.
Kittler, Friedrich. 1990. Discourse networks 1800/1900. Trans. Michael Meteer. Stan-
ford: Stanford University Press.
. 1999. Gramophone, film, typewriter. Trans. Geoffrey Winthrop-Young and
Michael Wutz. Stanford: Stanford University Press.
Kothandapani Pillai. 1958. Ancient Tamil music. Tamil Culture 7, no. 1:33–47.
. 1959. Ancient Tamil music, part 2. Tamil Culture 8, no. 3:193–200.
Krishna Iyer, E. 1932. Open letters. Repr., Sruti, E. Krishna Iyer Centenary Issue
(1997): 15–21.
. 1933. Personalities in present-day music. Madras: Rochehouse and Sons.
Krishnamachari, T. T. 1941. Karnāṭaka caṅkītamum tami icai iyakkamum [Kar-
natic music and the Tamil music movement]. Special issue, Ananda Vikatan,
Collected Articles on Music: 38–43.
Krishna Rao, H. P. 1906. First steps in Hindu music in English notation. London:
Weekes.
. 1917. The psychology of music. Repr., New Delhi: Asian Educational Ser-
vices, 1984.
Kroskrity, Paul, ed. 2000. Regimes of language: Ideologies, polities, and identities. Santa
Fe: sar Press.
Lacan, Jacques. 1977. Ecrits: Selections. Trans. Alan Sheridan. New York: Norton.
Lakshmi, C. S. 1990. Mother, mother-community, and mother politics in Tamil
Nadu. Economic and Political Weekly 25, no. 43:72–83.
. 2000. The singer and the song. New Delhi: Kali for Women.
Langford, Jean. 2002. Fluent bodies: Ayurvedic remedies for postcolonial imbalance. Dur-
ham: Duke University Press.
L’Armand, Adrian, and Kathleen L’Armand. 1978. Music in Madras: The urbani-
zation of a cultural tradition. In Eight urban musical cultures. Ed. Bruno Nettl,
115–45. Urbana: University of Illinois.
Legge, Walter. 1998. The maharaja of Mysore. In Walter Legge: Words and music.
Ed. Alan Sanders, 186–92. London: Duckworth.
Lelyveld, David. 1990. Transmitters and culture: The colonial roots of Indian
broadcasting. South Asia Research 10, no. 1:41–52.
. 1995. Upon the subdominant: Administering music on All-India Radio.
In Consuming modernity: Public culture in a South Asian world. Ed. Carol Brecken-
ridge, 49–65. Minneapolis: University of Minnesota Press.
Leppert, Richard. 1993. The sight of sound: Music, representation, and the history of the
body. Berkeley: University of California Press.
Levin, David Michael, ed. 1993. Modernity and the hegemony of vision. Berkeley:
University of California Press.
Levin, Thomas. 1990. For the record: Adorno on music in the age of its repro-
ducibility. October 55:23–47.

332 works cited

Levine, Lawrence. 1990. Highbrow/lowbrow: The emergence of cultural hierarchy in
America. Cambridge, Mass.: Harvard University Press.
Lévi-Strauss, Claude. 1955. Tristes tropiques. Trans. John Weightman and Doreen
Weightman. New York: Atheneum, 1977.
Luthra, H. R. 1986. Indian broadcasting. Delhi: Ministry of Information and Broad-
casting.
Madras Music Academy Conference Souvenir. 1935. Madras: Music Academy.
Madras Music Academy Conference Souvenir. 1938. Madras: Music Academy.
Madurai Mudaliar, K. 1929–1931. Kiramapō caṅkīta kīrtaāmirtam [Gramaphon san-
geetha keerthanamirdam: The nectar of gramophone music]. Madras: Shan-
mugananda Book Depot.
Maharajan, S. 1962. Olic celvam [The treasure of sound]. In Olic Celvam, 1–7.
Madras: Manonmani Patipakkam.
, ed. 1979. Racikamai kaṭitaṅkaḷ [Letters of T. K. Chidambaranatha Muda-
liar]. Madras: T.K.C. Vattat Totti.
Maharaja’s Music College Silver Jubilee Souvenir. 1945. Vizianagaram: Maharaja’s
Music College.
Maheswari Devi, N. 1935. Veena tutor with a chapter on the yazhl. Jaffna.
Malan. 1981. Vitvān. In Āu: Tērnta ciukataikaḷ 1917–1981, vol. 2:137–45. Madras:
Orient Longman.
Manikka Mudaliar, S. 1902. Caṅkīta cantirikai [Sangeeta Chandrikai]. Madras:
Chandrika Publications.
Mankekar, Purnima. 1999. Screening culture, viewing politics: An ethnography of tele-
vision, womanhood, and nation in postcolonial India. Durham: Duke University
Press.
Marglin, Frederique. 1985. Wives of the god-king: The rituals of the Devadasis of Puri.
Delhi: Oxford University Press.
Mathur, J. C. 1957. The impact of A.I.R. on Indian music. In Aspects of Indian music,
97–103. New Delhi: Ministry of Information and Broadcasting.
Mattick, Paul, Jr. 1995. Beautiful and sublime: Gender totemism in the constitu-
tion of art. In Feminism and tradition in aesthetics. Ed. Peggy Brand and Carolyn
Korsmeyer, 27–48. University Park: Pennsylvania State University Press.
Mayo, Katherine. 1927. Mother India. New York: Harcourt, Brace.
. 1929. Slaves of the gods. London: Florin Books.
Meduri, Avanthi. 1996. Nation, woman, representation: The sutured history of
the Devadasi and her dance. PhD diss., New York University.
Menon, Indira. 1999. The Madras quartet: Women in Karnatak music. New Delhi:
Lotus Books.
Menon, Narayana. 1957. The impact of Western technology on Indian music. In
Bulletin of the Institute of Traditional Cultures, 70–80. Madras.
Mitchell, Lisa. 2004. From medium to marker: The making of a mother tongue
in modern South India. PhD diss., Columbia University.

works cited 333

Mitchell, Timothy. 1988. Colonising Egypt. Berkeley: University of California
Press.
, ed. 2000. Questions of modernity. Minneapolis: University of Minnesota
Press.
Mrazek, Rudolph. 1997. ‘‘Let us become radio mechanics’’: Technology and na-
tional identity in late-colonial Netherlands East Indies. Comparative Studies of
Society and History 39, no. 1:3–33.
Mullick, K. S. 1974. Tangled tapes: The inside story of Indian broadcasting. Delhi: Ster-
ling Publishers.
Muthiah, P. L., ed. 1996. Tami icai muakkam [The roar of Tamil music]. Madras:
Mullai Patippakam.
Muthuswami, V. N.d., ca. 1940. Radio Kacceri. Bharatha Mani, 279–80.
Mysore Archaeological Reports. 1928–1941. Mysore: Mysore University.
Nadar, A. C. Paul. 1954. A pioneer research worker in Tamil music. Tamil culture
3, no. 2:110–20.
Naidu, Chitti Babu. 1925. A key to Hindu music. Madras: Diocesan Press.
Naidu, Tirumalayya. 1912. Music and the anti-nautch movement. Repr., Sruti,
E. Krishna Iyer Centenary Issue (1997): 6.
Nair, Janaki. 1994. The Devadasi, dharma, and the state. Economic and Political
Weekly (10 December): 3157–67.
Narayana Rao, C. 1939. The songs of Thyagaraja: English translation with originals.
Madras: Sarada Press.
Narayana Rao, V. 1995. Coconut and honey: Sanskrit and Telugu in medieval An-
dhra. Social Scientist 23:10–25.
. 2001. The politics of Telugu Ramayanas: Colonialism, print culture, and
literary movements. In Questioning Ramayanas: A South Asian tradition. Ed.
Paula Richman, 159–86. Berkeley: University of California Press.
Narayana Rao, V., and David Shulman, eds. 1998. A poem at the right moment: Re-
membered verses from premodern South India. Berkeley: University of California
Press.
Narayana Rao, V., David Shulman, and Sanjay Subrahmanyam. 1992. Symbols of
substance: Court and state in the Nayaka period. Delhi: Oxford University Press.
Narayanaswamy Mudaliar. 1900. Tamil sungeatha surabhooshany. Madras: Jeevaka-
runa Vilasa Press.
Nayar, Brigadier R. B. 1997. Study of Swati Tirunal’s compositions inconclusive.
Sruti 153:23–27.
Neild, Susan. 1979. Colonial urbanism: The development of Madras city in the
eighteenth and nineteenth centuries. Modern Asian Studies 13, no. 2:217–46.
. 1984. The Dubashes of Madras. Modern Asian Studies 18, no. 1:1–31.
Nettl, Bruno. 1985. The Western impact on world music: Change, adaptation, and sur-
vival. New York: Schirmer Books.
Neuman, Daniel. 1980. The life of music in North India. Repr., Chicago: University
of Chicago Press, 1990.

334 works cited

 . 1992. Patronage and performance of Indian music. In The powers of art: Pa-
tronage in Indian culture. Ed. Barbara Stoler Miller, 247–58. New York: Oxford
University Press.
Niranjana, Tejaswini. 1994. Colonialism and the politics of translation. In An other
tongue: Nation and ethnicity in the linguistic borderlands. Ed. Alfred Arteaga, 35–52.
Durham: Duke University Press.
O’Flaherty, Wendy Doniger. 1980. Women, androgynes, and other mythical beasts.
Chicago: University of Chicago Press.
Ollikkala, Robert. 1999. Classification systems, social hierarchies, and gender: Ex-
amining Indian ‘‘light-classical’’ music. Canadian University Music Review 19,
no. 2:27–36.
Ong, Walter. 1982. Orality and literacy: The technologizing of the word. New York:
Routledge.
O’Shea, Janet. 1998. ‘‘Traditional’’ Indian dance and the making of interpretive
communities. Asian Theatre 15, no. 1:45–63.
Padmanabha Iyer, A. 1936. Modern Mysore: Impressions of a visitor. Trivandrum: Sri-
dhara Print House.
Pandithar, Abraham. 1917. Karunamirtha Sagaram: A treatise on music or Isai-Tamil,
which is one of the main divisions of Muttamil, or language, music, and drama. Trans.
Gift Sironmani. Repr., Delhi: Asian Educational Services, 1984.
. 1917. A refutation to the second letter written by Mr. Clements about
the All-India music conference. Repr. in Karunamirtha Sagaram, Tamil edition.
Chennai: Kilaiyiyal Ayvu Niruvanak Kalvi Arakkattalai carpil Anril Patippa-
kam, 1994.
. 1934. A Practical Course in South Indian Music. Tanjavur: Lawley Press.
Panju Bhagavatar. 1910. Tyākarāja carittiram [Life history of Thyagaraja].
Parasuram, Sriram. 1997. The indigenisation of the violin. The Hindu Folio (De-
cember): 38–47.
Parker, Kunal. 1998. ‘‘A corporation of superior prostitutes’’: Anglo-Indian legal
conceptions of temple dancing girls, 1800–1914. Modern Asian Studies 32,
no. 3:559–633.
Parmentier, Richard. 1985. Signs’ place in medias res: Peirce’s concept of semiotic
mediation. In Semiotic mediation: Sociocultural and psychological perspectives. Ed.
Elizabeth Mertz and Richard Parmentier, 1–48. Orlando: Academic Press.
Peterson, Indira. 1984. The kriti as an integrated form: Aesthetic experience in
the religious songs of two South Indian composers. South Asian Literature 19,
no. 2:165–79.
. 1992. In praise of the lord: The image of the royal patron in the songs
of Saint Cuntaramurtti and the composer Tyagaraja. In The powers of art: Pa-
tronage in Indian culture. Ed. Barbara Stoler Miller, 120–41. New York: Oxford
University Press.
. 1999. The cabinet of king Serfoji of Tanjore: A European collection in
early nineteenth-century India. The History of Collections 11, no. 1:71–93.

works cited 335

 . 2000. Reimagining performance culture through the novel: Nationalist
discourses in the construction of Devadasi dance and nagasvaram ‘‘traditions’’
in Tillaa Mōkaampaḷ. Paper read at the South Asia Conference, Madison,
Wisconsin, October.
. 2004. Between print and performance: The Tamil Christian poems of
Vedanayaka Sastri and the literary cultures of South India. In India’s literary his-
tory: Essays on the nineteenth century. Ed. Stuart Blackburn and Vasuda Dalmia,
25–60. Delhi: Permanent Black.
Pinney, Christopher. 1997. Camera Indica: The social life of Indian photographs. Chi-
cago: University of Chicago Press.
Playne, Somerset. 1914. Southern India: Its history, people, commerce, and industrial
resources. London: Foreign and Colonial Compiling and Publishing.
Poduval, R. Vasudeva. [1950–1959?]. The music of Kerala and other essays. Trivan-
drum: St. Joseph’s Press.
Pollock, Sheldon. 2000. Cosmopolitan and vernacular in history. Public Culture
12, no. 3:591–26.
Popley, Herbert A. 1921. The music of India. Calcutta: Association Press.
Post, Jennifer. 1987. Professional women in Indian music: The death of the courte-
san tradition. In Women and music in cross-cultural perspective. Ed. Ellen Koskoff,
97–109. Westport, Conn.: Greenwood Press.
Prakash, Gyan. 1990. Writing post-orientalist histories of the third world: Per-
spectives from Indian historiography. Comparative Studies of Society and History
32, no. 2:383–408.
Proceedings of the 17th Madras music conference. 1944. Journal of the Music Acad-
emy, Madras. Madras: Music Academy.
Qureshi, Regula. 2000. How does music mean? Embodied memories and the
politics of affect in the Indian sarangi. American Ethnologist 27, no. 4:805–38.
. 2001. In search of Begum Akhtar: Patriarchy, poetry, and twentieth-
century Indian music. World of Music 43, no. 1:97–137.
Radhakrishna Ayyar, Vai. 1949–1950. Tiruvaiyā sri tyākapirammam arātaai utcava
varalāu [The history of the Tiruvaiyaru Thyagaraja aradana festival]. Repr.,
Madras: Sri Maruthy Laser Printers, 2003.
Raghavan, V. 1944. Some musicians and their patrons about 1800 AD in Madras
city. Journal of the Madras Music Academy 16:127–36.
. 1958. The Sārvadēvavilāsa. Adyar Library Bulletin 22, pts. 1–2:45–118.
. 1961. English preface to Sangīta Sampradāya Pradārṣini. 1st Tamil edition.
Madras: Music Academy.
Rajagopalan, N., ed. 1990. Contests and challenges. In Another Garland, 54–63.
Bombay: Bharatiya Vidya Bhavan.
Ram, Kalpana. 2000. Dancing the past into life: The rāsa, nṛtta, and rāga of im-
migrant existence. Australian Journal of Anthropology 11, no. 3:261–73.
Ramachandran, K. 1949. Dakshina raga ratnakaram: Characteristics of Carnatic ragas.
Madras: V. Ramaswamy Sastrulu and Sons.

336 works cited

Ramachandra Rao, S. K. 1994. Mysore T. Chowdiah. Bangalore: Sree Ramaseva
Mandali.
Ramakrishna, V. 1983. Social reform in Andhra, 1848–1919. Delhi: Vikas.
Ramalingam Pillai, Namakkal. 1943. Icaittami [Tamil music]. Madras.
Raman, C. V. 1920. Experiments with mechanically played violins. In Proceedings
of the Indian Association for the Advancement of Science, 19–36. Calcutta: Indian
Association for the Advancement of Science.
Ramanathan, K. V. 1996. The Swati Tirunal compositions: Facts and figures in a
whodunit. Sruti 142:15–29.
Ramanathan, S. 1977. Taccur Singaracharyulu brothers. Indian Express.
Ramanathan Chettiar. 1993. The history of the Tamil Icai Sangam, 1943–1993. Un-
published paper in collection of Tamil Icai Sangam, Madras.
Ramanujachari, C. 1957. The spiritual heritage of Thyagaraja. Madras: Ramakrishna
Mission.
Rama Rao, V. V. B. 1985. Life and mission in life: Poosapati Ananda Gajapati Raju.
Hyderabad: International Telugu Institute.
Ramasami, E. V. 1944. Tami icai, naṭippu kalaikaḷ: Ii ea ceyya vēṇṭum? [Tamil
musical and dramatic arts: What must still be done?]. Erode: Kuti Aracu.
Ramaswamy, Sumathi. 1993. En/gendering language: The poetics of Tamil iden-
tity. Comparative Studies of Society and History 35:683–725.
. 1997. Passions of the tongue: Language devotion in Tamil India, 1891–1970.
Berkeley: University of California Press.
. 1999. Catastrophic cartographies: Mapping the lost continent of Le-
muria. Representations 67:92–129.
Ramaswamy Aiyar, M. S. 1927a. Thiagaraja: A great musician saint. Repr., Madras:
Asian Educational Services, 1986.
. 1927b. Sargam notation. In Thiagaraja: A great musician saint, 179–238.
Repr., Madras: Asian Educational Services, 1986.
, ed. 1932. Ramamatya’s Svaramēlakalānidhi: A work on music. Chidambaram:
Annamalai University.
Ramaswamy Bhagavathar, Wallajapet. 1935. Tyākapirammopaniṣat [Thyāgabramopa-
niṣad: The life of Saint Thyagaraja]. Madras.
Ramnarayan, Gowri. 1997. Past forward: Six artists speak about their childhood. Madras:
Orient Longman.
Rangaramanuja Ayyangar, R. 1977. Musings of a musician: Recent trends in Carnatic
music. Bombay: Wilco Publishers.
Rao, Vidya. 1999. Thumri as feminine voice. In Gender and politics in India. Ed.
Nivedita Menon, 475–93. Delhi: Oxford University Press.
Reck, David. 1984. India/South India. In Worlds of music: An introduction to the music
of the world’s peoples. Ed. Jeff Todd Titon, 209–65. New York: Schirmer Books.
Reddy, Muthulakshmi. 1928–1931. Why should the Devadasi system in the Hindu
temples be abolished? Madras: Lodhra Press.
. 1930. My experience as a legislator. Madras: Current Thought Press.

works cited 337

Redfield, Robert, and Milton Singer. 1954. The cultural role of cities. Economic
Development and Cultural Change 3, no. 1:53–73.
Redfield, Robert. 1953. The primitive world and its transformations. Ithaca: Cornell
University Press.
Report of the first general meeting of the academy of Tamil culture. 1955. Tamil
Culture 4, no. 4:368–76.
Report of the All-India Music Conference, Madras. 1927. Madras: Music Academy.
Reynolds, Holly Baker. 1980. The auspicious married woman. In The powers of
Tamil women. Ed. Susan Wadley, 35–60. Syracuse: Maxwell School.
Robbins, Joel. 2001. God is nothing but talk: Modernity, language, and prayer in
a Papua New Guinea society. American Anthropologist 103, no. 4:901–12.
Roberts, P. E. 1952. History of British India under the company and the crown. 3rd ed.
Completed by T. G. P. Spear. London: Oxford University Press.
S. V. K. 1999. Divinity, the core of Indian music. The Hindu (December).
Said, Edward. 1979. Orientalism. New York: Vintage.
Sambamoorthy, P. 1939. Madras as a seat of musical learning. In Madras tercentenary
commemorative volume, 429–37. London and Madras: Oxford University Press.
. 1944. Our concert programme: Some underlying principles. In Madras
Music Academy Conference Souvenir, 39–43. Madras: Madras Music Academy.
. 1952. History of South Indian music, vols. 1–6. Madras: Indian Music Pub-
lishing House, 1984.
. 1955a. Dictionary of South Indian music and musicians, vols. 1–3. Madras:
Indian Music Publishing House.
. 1955b. Kacceri dharma [Concert etiquette]. In Dictionary of South Indian
Music and Musicians, 2:266–72. Madras: Indian Music Publishing House.
. 1959. Great musicians. Madras: Indian Music Publishing House.
. 1961. Elements of Western music for students of Indian music. Madras: Indian
Music Publishing House.
. 1966. The teaching of music. Madras: Indian Music Publishing House.
. N.d.a. [ca. 1960?] The melakarta janya raga scheme. Madras: Indian Music
Publishing House.
. N.d.b. A practical course in Karnatic music for schools. Madras: Indian Music
Publishing House.
Sangari, Kumkum, and Sudesh Vaid. 1990. Recasting women. New Delhi: Kali for
Women.
Sankaramurthy, M. R. 1990. The European airs of Muthuswamy Dikshitar. Bangalore:
Guru Guha Nilaya.
Sarathy. 1997. Icai ulaki imayam Em. Es. [The Himalaya of the music world, M.S.].
Madras: Vanathi Patipakkam.
Sastry, B. V. K. 1962. Dwaram Venkataswamy Naidu. Illustrated Weekly of India
(21 October): 33–35.
Sathyanarayana, R. 1993. Re-emergence of the violin in Karnataka music. Vadya-
kala Souvenir, 10–12. Madras.

338 works cited

Saupa. 1990. Maturai Poutāy: Oru kaṇṇīr katai [Madurai Ponnuthai: A sad
story]. Ananta vikatan (2 February): 10–14.
Schieffelin, Bambi, K. Woolard, and Paul Kroskrity, eds. 1998. Language ideologies:
Practice and theory. Oxford: Oxford University Press.
Seetha, S. 1981. Tanjore as a seat of music. Madras: Madras University Press.
Shankara Ayyar, V. S. Gomathi. 1970. Icai valluarkaḷ [Stalwarts of music]. Madras.
Sharma, A. V. S. 1954. Lines of devotion. Madras: Antiseptic Press.
Siegel, James. 1997. Fetish, recognition, revolution. Princeton: Princeton University
Press.
Silverman, Kaja. 1988. The acoustic mirror: The female voice in psychoanalysis and cinema.
Bloomington: Indiana University Press.
Singaracaryulu Brothers, Taccur. 1882. Gāyakaparijatam. Madras: K. R. Press.
. 1884. Gāyaka lōchana. Madras: K. R. Press.
. 1889. Sangīta Kalānidhi. Madras: K. R. Press.
. 1905. Gāyaka Siddhanjanam. Madras: K. R. Press.
. 1912. Ganendu Sēkaram. Madras: K. R. Press.
. 1914. Svaramanjari. Madras: K. R. Press.
Singer, Milton. 1972. When a great tradition modernizes: An anthropological approach
to Indian civilization. New York: Praeger.
Sinha, Mrinhalini. 1996. Gender in the critiques of colonialism and nationalism:
Locating the ‘‘Indian woman.’’ In Feminism and history. Ed. Joan Scott, 477–
504. New York: Oxford University Press.
Sivatamby, K. 1978. The politics of a literary style. Social Scientist 6, no. 8:16–33.
Spivak, Gayatri. 1988. Can the subaltern speak? In Marxism and the interpretation of
culture. Ed. Cary Nelson and Lawrence Grossberg, 271–313. Urbana: Univer-
sity of Illinois Press.
Srinivasan, Amrit. 1985. Reform and revival: The devadasi and her dance. Economic
and Political Weekly 20, no. 44:1869–76.
Srinivasan, G. 1991. The Birth of the Tyagabrahma Mahotsava Sabha. In Sri Tyaga-
brahma Mahotsava Sabha Souvenir. Madras.
Srinivas Iyer, Semmangudi. 1991. Catkuru tyākapiramma arātaai utcavam
oāka iaintu varalāu [The history of the joining into one of the Thyagaraja
aradhana festival]. In Sri Tyagabraahma Mahotsava Sabha Souvenir. Madras.
Sruti. 1997. E. Krishna Iyer centenary issue.
. 1998. Cutbacks in All India Radio’s music programmes. ( July) 11–12.
. 2001. Special issue on Vai. Mu. Kothainayaki (207).
. 2003. Special issue on the Madras music season (222).
Sterne, Jonathan. 2000. Sound out of time: Modernity’s echo. In Turning the cen-
tury. Ed. Carol Stabile, 9–32. Boulder, Colo.: Westview Press.
. 2003. The audible past: Cultural origins of sound reproduction. Durham: Duke
University Press.
Subba Rao, T. V. 1962. Studies in Indian music. Madras: Vasantha Press.

works cited 339

 . 1962. The modernity of Thyagaraja. In Studies in Indian Music, 144–49.
Madras: Vasantha Press.
Subbashri. 1947. Maiki makātmiyam [The greatness of the mic]. Kalki (9 Feb-
ruary): 41–42.
Subrahmanya Ayyar, C. 1939. The grammar of South Indian music. Madras.
. 1941. The art and technique of violin play. Madras.
. 1944. My musical extravagance. Madras.
Subrahmanyam, V. V. 1980. Vaiyali varalāu [History of the violin]. Madurai: India
Printing Works.
. 1986. Satyamēva Jeyatē [Let truth reign]. Madras: India Printing Works.
Subramaniam, P. R., ed. 1992. Kriyāvi tarkāla tami akarāti [Cre-A’s dictionary of
contemporary Tamil]. Madras: Cre-A Publishers.
Subramanian, Lakshmi. 1999a. Gender and the performing arts in nationalist dis-
course: An agenda for the Madras Music Academy, 1930–1947. The Book Re-
view 23, nos. 1–2:81–84.
. 1999b. The reinvention of a tradition: Nationalism, Carnatic music,
and the Madras Music Academy. Indian Economic and Social History Review 36,
no. 2:131–63.
Sunda. 1993. Eminent Tamil writer Kalki: A life sketch. Madras: Vanathi.
Suntharalingam, R. 1974. Politics and nationalist awakening in South India, 1852–1891.
Tucson: University of Arizona Press.
Swaminathan, Mayuram. 1998. Winds of change in A.I.R. The Hindu, 16 June, 28.
Swaminathayyar, U. V. 1936. Caṅkīta mummaṇikaḷ [Three gems of music]. Madras:
Dr. U. V. Swaminathayyar Library.
. 1940–1942. E carittiram [The story of my life]. Trans. S. K. Guruswamy.
Repr., Madras: Dr. U. V. Swaminathayyar Library, 1980.
Tagore, Sourindro Mohun. 1874. Hindu music. Repr. in Hindu music, 337–404.
Delhi: D. K. Publications, 1994.
Tami icai makānāṭu ceai nikacci malar [Chennai Tamil music conference pro-
gram]. 1944. Madras: Tamil Icai Sangam.
Taussig, Michael. 1993. Mimesis and alterity: A particular history of the senses. New
York: Routledge.
Taylor, Charles. 1989. Sources of the self: The making of modern identity. Cambridge,
Mass.: Harvard University Press.
Terada, Yoshitaka. 1992. Multiple interpretations of a charismatic individual: The
case of the great nagasvaram musician, T. N. Rajarattinam Pillai. PhD diss.,
University of Washington.
. 2000. T. N. Rajarattinam Pillai and caste rivalry in South Indian classical
music. Ethnomusicology 44, no. 3:460–90.
Theberge, Paul. 1995. Any sound you can imagine: Making music, consuming technology.
Hanover, N.H.: University Press of New England.
Thomas, Downing. 1995. Music and the origins of language: Theories from the French
enlightenment. Cambridge: Cambridge University Press.

340 works cited

Tick, Judith. 1986. Passed away is the piano girl: Changes in American musical
life, 1870–1900. In Women making music: The Western art tradition, 1150–1950.
Ed. Jane Bowers and Judith Tick, 325–48. Urbana: University of Illinois Press.
Tirumalayya Naidu, C. 1919. Tyagayyar: The greatest musical composer of South India.
Madras: South Indian Press.
Trautmann, Thomas. 1997. The Aryans and British India. Berkeley: University of
California Press.
Vasudevachar, Mysore K. 1955. Nā kaṇḍa kalavidāru [Vidwans I have known]. Trans.
and ed. by S. Krishnamurthy as With masters of melody. Bangalore: Ananya gml
Cultural Academy, 1999.
Vedavalli, M. B. 1992. Mysore as a seat of music. Trivandrum: cbh Publications.
Venkatachalam, G. 1966. My contemporaries. Bangalore: Hosali Press.
Venkatachalapathy, A. R. 1994. Reading practices and modes of reading in colonial
Tamil Nadu. Studies in History 10, no. 2:273–90. New Delhi: Sage Publications.
Venkataramiah, K. 1984. Icai, nāṭakam, nāṭṭiyam [Music, dance, drama]. In Tañ-
cai marāttiya maar kala araciyalum camutaya vākaiyum [The art patronage and
societal life of the Maratha kings of Tanjavur], 215–38. Tanjavur: Tamil Uni-
versity.
Venkataratnam, Raghupati. 1901. Social purity and the anti-nautch movement.
In Indian social reform. Ed. C. Y. Chintamani, 249–81. Madras: Thompson.
Venkitasubramonia Iyer. 1975. Swati Tirunal: The man and his music. Trivandrum:
College Book House.
Waghorne, Joanne. 1994. The raja’s magic clothes: Re-visioning kingship and divinity in
England’s India. University Park: Pennsylvania State University Press.
Walimbe, Y. S. 1980. Abhinavagupta on Indian aesthetics. Delhi: Ajanta Books Inter-
national.
Washbrook, David. 1975. The development of caste organization in South India,
1880–1925. In South India: Political institutions and political change, 1880–1940. Ed.
Christopher Baker and David Washbrook, 150–203. Delhi: Macmillan.
Weber, William. 1984. The contemporaneity of eighteenth-century musical taste.
Musical Quarterly 70, no. 2:175–94.
. 1992. The rise of musical classics in eighteenth-century England: A study in canon,
ritual, and ideology. Oxford: Clarendon Press.
Weidman, Amanda. 1995. Ambiguous apparitions: Gender and the classical in
Karnatic music. Master’s thesis, University of Washington.
. 2001. Questions of voice: On the subject of ‘‘classical music’’ in South
India. PhD diss., Columbia University.
. 2003a. Gender and the politics of voice: Colonial modernity and classical
music in South India. Cultural Anthropology 18, no. 2:194–232.
. 2003b. Guru and gramophone: Fantasies of fidelity and modern tech-
nologies of the real. Public Culture 15, no. 3:453–76.
Weiss, Allen S. 1995. Phantasmic radio. Durham: Duke University Press.

works cited 341

Williams, Raymond. 1983. Culture and society, 1780–1950. New York: Columbia
University Press.
Wilson, Anne. 1904. A short account of the Hindu system of music. Lahore: Ghulab
Singh.
Winslow, Miron. 1862. Tamil-English dictionary. New Delhi. Repr., Asian Educa-
tional Services, 1979.
Woodfield, Ian. 2001. Music of the Raj: a social and economic history of music in late
eighteenth-century Anglo-Indian society. Oxford: Oxford University Press.
Woolard, Kathryn, and Bambi Schieffelin. 1994. Language ideology. Annual Re-
view of Anthropology 23:55–82.
Žižek, Slavoj. 1996. I hear you with my eyes: Or, the invisible master. In Gaze
and voice as love objects. Ed. Slavoj Žižek and Renata Salecl, 90–126. Durham:
Duke University Press.
Zvelebil, Kamil. 1992. Companion studies to the history of Tamil literature. Leiden: E. J.
Brill.

342 works cited

$ Index

Abbate, Carolyn, 14 Bannerjee, Sumanta, 302n3

abhinaya, 130–31, 223 Bauman, Richard, 7, 8, 292n8
Achariyar, Deivasikamani, 87–94 Bergeron, Katherine, 316n18, 317n25,
Adorno, Theodor, 322n19 318n38
Adyanthaya, Rao Bahadur N. M., Besant, Annie, 18, 303n11, 304n11
164–66 Bhabha, Homi, 58, 298n33
Akhtar, Begum, 302n2, 304n14 bhakti, 17, 306n31, 310n2, 322n18
Allen, Matthew, 120–21, 130, 303n11 Bharata Natyam, 120–21, 129–31,
All-India Music Conference, 81, 161, 303n5, 306n31. See also nautch
169, 216, 314n26, 317n24 Bharatiyar, Gopalakrishna, 153–55, 182
All India Radio (air), 1, 3, 35, 47, 52, Bharatiyar, Subramania, 135–36, 174
105, 123, 168, 268–75, 323n20 Bhatkhande, V. N., 161, 221, 312n17
Ammaiyar, Ramamirtham, 118 Blackburn, Stuart, 311n8
Annadurai, C. N., 175, 182–83 Brady, Erica, 321n9
Annamalai Chettiar, Raja Sir, 167–68, Brahmins, 4; domination of classical
184 music by, 22, 110, 115, 155, 180–81;
Appadurai, Arjun, 15 non-Brahmin movement, 19, 155,
Arooran, Nambi, 167, 181, 183, 312n13, 293n18; Smarta Brahmins, 16, 17;
313n21 Tamil and, 155–56, 181–82, 311n7
art, discourse about, 5, 119, 128–29, Briggs, Charles, 7, 8, 292n8
146, 198–99, 303n9, 309n49 Bullard, Beth, 42, 121, 308n46
artist, as concept, 5, 18, 128–29, 141,
149 Caldwell, Robert, 167, 169, 313n22
Awasthy, G. 268–71 Chakrabarty, Dipesh, 6, 308n39
Chatterjee, Partha, 6, 302n3, 309n50
Bach, J. S., 26, 52–53 Chatterji, P. C., 270
Bakhle, Janaki, 161, 293n23, 312n17, Chaudhuri, Amit, 243
324n28 Chettiars, 4
Bakhtin, Mikhail, 20, 34, 295n9 Chettys, 81
Balachander, S., 192–201, 242, Chidambaranatha Mudaliar, ‘‘Rasika-
315nn1–2, 316n15 mani’’ T. K., 125, 128, 172–78
child prodigies, 123–24, 304n16 Dirks, Nicholas, 15
Chinnaswamy Mudaliar, A. M., 136, discourse networks, 12, 321n11
204–13, 217, 220, 224, 233, 241, domesticity, 116, 134–41, 144–45. See
318n29 also women
Chowdiah, Mysore T., 41, 44–45 Dravidian movement, 19, 155, 167; lin-
cinema, 125, 146, 159, 305n18 guistic studies and, 169–71; as term,
cinna mēḷam, 102 313n22
Clements, Ernest, 296n19, 317n24, Durayappa Bhagavatar, Gottuvadyam,
319n34 131
Cohn, Bernard, 156, 183 Dwaram Venkataswamy Naidu, 32,
colonialism, 4, 9, 15–17, 25, 58, 291n5, 49–52, 300n35, 322n13
292n8, 324n1; princely states and,
15–16 East India Company, 15, 16, 61
composers, 20, 192–201, 241; as saints, Ellarvi (writer), 70, 72–73
99–104 Engel, Carl, 26–27
compositions, 98–99, 236, 241 Erlmann, Veit, 292n9
concerts: form of, 83, 97–99; concert ethnomusicology, 23–24
halls, 4, 15, 17, 34, 68, 81, 86, 126, Ettayapuram, 29–31
130–31; public, 17, 59–60, 79–81,
94–97, 110 Farrell, Gerry, 28
Congress Party, 166, 293n18, 313n23 Feld, Steven, 13, 111, 292n14
Connor, Steven, 297n31,n32, 300n33 fidelity: social, 3–4, 21, 85, 246, 249,
Coomaraswamy, Ananda, 46, 256–60, 282–85; in sound reproduction,
282, 316n22 246, 282, 285, 322n15
Cousins, Margaret, 18, 24, 66, 294n25 Fielden, Lionel, 268, 270–71
film songs, 98, 159, 186, 271
Day, Charles, 27, 203 Fort Saint George, 29
de Alwis, Malathi, 307n35 Foucault, Michel, 240, 286, 318n33
de Certeau, Michel, 192, 242, 292n8 Fox, Aaron, 13, 111, 292n14
Derrida, Jacques, 192 Fox-Strangways, A. H., 46, 259
devadasis, 18, 113, 116–21, 124, 136, Frank, Felicia Miller, 124
303n5, 303n8, 303n11, 308n43, 309n54 Frith, Simon, 305n19
Devadoss, Manohar, 159–60, 191,
312n14, 16 Gaisberg, Fredrick, 255–56, 320n7
Devi, Rukmini, 119–20, 128–30, 149, Gajapathi, Ananda, 68–69, 299n14
190, 303n9, 304n11, 304n16, 306n31 gamaka, 32, 33, 50, 54, 207, 213, 216,
devotional music, 186–88, 271 221, 229, 240, 257, 262, 321n8
Dharwadker, Vinay, 319n36 Gandhi, Mohandas, 125, 146, 309n51
Dikshitar, Baluswamy, 29, 77 Ganghadar, C., 228–29
Dikshitar, Muthuswamy, 25, 99; Gayan Samaj, 194–95, 201–3
‘‘European Airs,’’ 32 George Town, 76–77, 80–81, 168, 221,
Dikshitar, Subbarama, 77, 220–22, 313n23
224 Goehr, Lydia, 5

344 index
Gopalakrishnan, M. S., 50–51 308n39, 309n49. See also modernity:
Gramophone Company, 255–56, modern subjectivity
320n7, 321n11, 322n12 Ivy, Marilyn, 9
gramophone recordings: of courte-
sans and devadasis, 304n14, 320n7; of Jackson, Bill, 100–101
female singers, 18, 122–27, 261; of Jayarama Iyer, T. K., 37
film music, 159; of Mysore palace Jayaraman, Lalgudi, 50
band, 67; of violinists, 261 Jinarajadasa, Dorothy, 18
gramophones, 255–60, 264–68, 278– Johannes, T. C. R., 37–40, 225
79 Joshi, G. N., 321n11
great and little traditions, 24, 293n24 Justice Party, 166, 293n18
guru-disciple relationship, 1–4, 247–
55, 274–76, 319n2 Kalki Krishnamoorthy, 103, 124, 127,
gurukulavasam, 3–4, 21, 123, 246–55, 145, 180–82, 263–64, 268, 305n22,
281–83, 291n1; gurukula system, 124, 306n29
194, 274–76, 281 Kalyanasundaram, Tiru Vi., 91
kaṇakku, 134–35, 253, 307n37, 320n6
Hancock, Mary, 17, 293n17, 302n3 Kannabiran, Kalpana, 117
Hansen, Kathryn, 302n4 Kanyakumari, A., 54–55
Harinagabhushanam (violinist), 165– Keane, Webb, 145, 197
66 Kersenboom, Saskia, 310n3
harmonium, 26, 38, 43, 45, 46–48, 66, Keskar, B. V., 270–72, 275–78, 323n28
68, 257, 259, 265, 270 Khan, Shujaat, 275
Harrison, Martin, 246–47, 319n1 kinnari, 33–34, 295n7
hereditary musicianship, 141 Kinnear, Michael, 123, 256
Higgins, Jon, 300n32 Kittappa, G., 62–63, 316n15
Hill, Jane, 295n9 Kittler, Friedrich, 12, 260, 321n11
Hindi, protests opposing, 157, 167, 180, konnakkol, 102, 225
293n18 Kothainayaki, Vai. Mu., 148–49,
Hindustani music, 62, 64, 98, 221, 275, 309nn54–55
295n10, 306n32, 319n37, 323n28 Krishna Iyer, E., 79, 84–85, 119–22,
Hovhaness, Alan, 52 130, 265–67
Hughes, Stephen, 322n16, 323n22 Krishnamachari, T. T., 113, 183–85, 190
Krishna Rao, H. P., 161–66, 213–17,
Illayaraja (film music director), 52–53 312n18, 313n20
improvisation, 193, 241, 300n35; cal- Krishnayyar, Ganam, 153
culations in, 134–35; recording and,
262–64, 267, 275, 320n6 Lacan, Jacques, 10, 292n11
Indian Association for the Cultivation Lakshmi, C. S., 139, 307n36
of Science, 43–44, 261 Langford, Jean, 250, 274–75, 301n41
Inoue, Miyako, 10, 20 language: language ideology, 9–10,
interiority: performed, 135, 146; sense 292n10; modern ideas about, 8,
of, 18, 19, 146, 149, 190–91, 244, 287, 152, 157, 190, 311n11, 12; ‘‘mother

index 345
language (continued ) 88; colonial encounter and, 6–7,
tongue’’ and, 19, 152, 156–58, 175– 314n24; as discursive formation,
79, 218, 311n10; music as, 19, 152–53, 7–8; modern subjectivity, 7–8, 12,
160–66, 189–91 19, 190, 244, 261; music and, 245,
Legge, Walter, 66 283; vision and, 286–87; voice and,
legibility, 232, 317n25 6–9, 19–20, 190–91, 286–89
Lelyveld, David, 271 Mrazek, Rudolph, 323n23
Lemuria, 168–70, 314n24 Mudaliar, K. Madurai, 264–66
Levin, David, 286 Mudaliars, 4
Lévi-Strauss, Claude, 242 mudras, 196–97
listening: practices of, 24, 68, 127– Mullick, K. S., 273, 323n26
28, 173–74, 179, 183, 247, 262–68, music criticism, 84–86
269, 277–82, 312n15, 323n22. See also music manuals, 221–32
overhearing Muthiah Bhagavatar, Harikesanallur,
Luthra, H. R., 269–70 196, 199–201, 299n18
Mylapore, 16–17, 76–77, 80–81, 168,
Madras, 16–17, 60, 76–78, 168 313n23
Madras Music Academy, 81–82, 129, Mysore, 31, 59–60, 63, 64, 65–68, 76,
138, 150, 164, 168, 180, 183, 186, 187, 108–9
236, 291n3, 301n40
Madras Presidency, 64, 69 Nagaratnammal, Bangalore, 118, 123,
Maharajan, Justice S., 128, 175–77 148, 309n54
Malan (writer), 250–55, 320n4 nagaswaram, 92–93, 141
Mangeshkar, Lata, 145 Naidu, Chitti Babu, 235
Mathur, J. C., 272 Narayana Das, Adhipatla, 69–70
Mayo, Katherine, 303n8 Narayana Rao, Velcheru, 311nn10–11,
Menon, Indira, 122–27, 132 315n2
Menon, Narayana, 271–72 nautch, 45, 77, 120, 130, 136, 138, 213,
Menon, Raghava, 242–43, 245, 250, 223, 303n5
281 Nayar, Brig. R. B., 199–201
microphones, 17, 59–60, 79, 86–93, Neild, Susan, 16
126, 146, 147, 300n33, 305n19 Nettl, Bruno, 294n2
middle class: gramophones and, 258; Neuman, Daniel, 47, 291n6, 295n10,
music and, 6, 110, 115, 121, 129, 296n18
138–40, 144, 149; nationalism and, Niranjana, Tejaswini, 295n13
302n3 notation, 20, 136, 192–95, 199–232,
mimesis, 58, 298n34; expression vs., 273; sargam notation, 212, 215–
146, 301n41; mechanical, 20, 247, 19, 226–31; staff notation, 62, 66,
253–55, 258; violin accompaniment 203–16, 226–31, 316n19, 317n24
as, 15, 35
Mitchell, Lisa, 157, 311nn11–12 Ollikkala, Robert, 306n32
Mitchell, Timothy, 7, 211, 286–88 Ong, Walter, 55, 291n8
modernity, 6–9; aurality and, 286– oral tradition and orality, 4, 5, 8, 20–

346 index
 21, 193, 200–201, 235, 241–44, 245, Radio Ceylon, 271
282, 291–92n8 Raghavan, V., 29, 77–78, 80
orientalism, 4, 6, 291n2, 294n24 Rajagopalachari, C., 178–79
O’Shea, Janet, 306n31 Ramachandran, K., 236
overhearing, 159–60, 259–60, 312n15. Ramalingam Pillai, Namakkal, 177–
See also listening 78, 190
Ramamatya (treatise author), 236–40
Padmanabha Iyer, A., 68 Raman, C. V., 40, 43, 267–68
Pandithar, Abraham, 34, 168–71, 225– Ramanathan Chettiar, 167–68
27, 296n19, 314nn25–27 Ramasami, E. V., 166–67, 180–81,
Parker, Kunal, 117, 303n5 293n18
patronage, 4; private, 76–78; royal, Ramaswamy, Sumathi, 157–58, 293n18,
29–31, 59–69, 70–76, 103, 197–99; 312n13, 314n24
by sabhas, 80–81; by state, 70, 107 Ramaswamy Aiyar, M. S., 59, 99–100,
Pattammal, D. K., 123, 126–27, 128, 217–19, 236–40
131, 146–48, 186, 286, 309n55 Ramaswamy Bhagavatar, Wallajapet,
Pattnam Subramania Iyer, 33, 76, 33–34, 99
247–49 Ramnarayan, Gowri, 124
Peirce, Charles, 292n13 Rangaramanuja Ayyangar, 269, 275,
percussion instruments, 102 323n27
performance, 235; clothing in, 132–35; Rao, Vidya, 307n32
conventions of, 18, 130–35, 139; text Reddy, Muthulakshmi, 18, 117–19,
vs., 14, 193, 209, 211–12 303n10
periya mēḷam, 102 Robbins, Joel, 8
Peterson, Indira, 154, 197, 298n5, Rousseau, Jean-Jacques, 310n1
301n40
Pillai, Meenaskhisundaram, 154–55 sabhas, 17, 60, 78–82, 168
Pinney, Christopher, 316n20 sadir, 119, 303n5
playback singers, 146–47 Sambamoorthy, P., 29, 34–35, 70, 76–
Ponnuthai, M. S., 141–43, 308n44 77, 79–80, 83, 86, 228–35, 243, 273,
Post, Jennifer, 307n33 276–79, 295n11, 299n16
princely states, 292n15. See also Ettaya- Sanskrit, 306n29, 310n2; as language
puram; Mysore; Tanjavur; Travan- of music, 173; music treatises in,
core; Vizianagaram 169, 224, 233, 236–38, 239, 310n2;
prostitution, 116–18 Tamil and, 169–71, 311n8, 314n27
sarangi, 26, 45–46, 258, 294n1, 296n18
Qureshi, Regula, 46, 57, 302n2, Saraswati Bai, 121–22
304n14 Sārvadēvavilāsa, 77–78, 293n16
Sastri, Syama, 25, 99
Radhakrishnan, S., 150–51 Sathyanarayana, R., 31
radio, 22, 159, 263, 268–75, 279– Schwartz, Christian Friedrich, 61
80, 323nn21–23; music lessons on, Self-Respect movement, 166, 181,
272–75 293n18, 312n13

index 347
Serfoji II, Maharaja of Tanjore, 30, tambura, electronic, 283–85, 324n30
61–64, 198 Tamil: as ancient language, 168–71,
seventy-two melakarta system, 221, 313n24; identification with, 151;
233–36, 241, 318nn30–32 as language of music, 166–91; lit-
Shankara Ayyar, V. S. Gomathi, 70, erature in, 152, 153–56, 311nn3–5;
75–76 Mother Tamil or Tamittāy, 19, 155,
Shanmugham Chettiar, R. K., 151, 168 157–58, 175–79, 293n18, 307n36,
Siegel, James, 312n15 318n26; muttami (triple Tamil), 153,
Singaracaryulu Brothers, 76–77, 170, 310n3; Sanskrit and, 169–71,
221–25, 233 311n8, 313n22, 314n27
Sinha, Mrinhalini, 11–12, 148, 302n3 Tamil Icai Sangam, 168
Sivan, Papanasam, 167, 182, 199 Tamil music movement, 19, 150–53,
social reform, 18, 116–21 166–68, 171, 183, 184, 188–91, 313n21
Somasundaram, Madurai, 187–89 Tanjavur, 29–31, 61–64, 76, 221
songbooks, 264–66 Tanjore Quartet, 30, 62–63; Vadivelu
sopana sangitam, 63, 200 of, 30–31, 32, 62–63, 195, 198, 199
sound reproduction, 20–21, 85, 146, Taussig, Michael, 57–58, 298n34, 321n9
148, 159, 245–47, 250, 255–68, Taylor, Charles, 7, 103, 145–46, 309n49
321n11, 322nn14–15 teaching institutions, 60, 65, 69, 78,
Spivak, Gayatri, 11 81–82, 194, 250
Srinivasan, Amrit, 120, 303n11 Telugu, 151, 153, 157, 168, 173, 179, 185
Srinivas Iyer, Semmangudi, 199 Theosophical Society, 128, 304n11, 12
sruti box, 283–85, 324n30 Theosophy, 120, 294n25
Sterne, Jonathan, 11, 245, 261, 286–88, Thomas, Downing, 310n1
321n9, 322nn14–15 Thyagaraja (saint-composer), 17, 25,
Subba Rao, ‘‘English,’’ 63 33, 52, 61, 64, 77, 92, 97–106, 112,
Subba Rao, T. V., 1, 4, 82, 101–2, 150, 178, 185–87, 197, 223, 301nn39–40
186, 279–81, 291n3 Thyagaraja Bhagavatar, M. K., 91–92,
Subbulakshmi, M. S., 112–14, 123–28, 187
133, 138, 145–49, 183, 186, 267, 302n2, Tirunal, Swati, 30–33, 63–65, 192–200,
304n17, 305n18, 305nn21–22 242, 315n2, 316n15
Subrahmanya Ayyar, C., 32, 40–42, Tiruvaiyaru, 64, 247; Thyagaraja
43, 45, 51, 56, 131, 240, 244, 260–64, memorial at, 104–6, 118, 301nn42–
321n10 43, 309n54
Subrahmanyam, V. V., 198–99 translation: of musical terms, 213, 216,
Subramanian, Lakshmi, 129, 301n40 244; translatability (between oral
Sundarambal, K. B., 87, 187 and written), 228–32, 244
Swaminathayyar, U. V., 70–74, 76, Travancore, 63–65, 194–97
153–58, 311n4 Trivandrum, 29–31, 76

Tagore, Rabindranath, 323n20 Vaidyanatha Iyer, Maha, 71–72, 75–

Tagore, Sourindro Mohun, 27, 201, 76, 248–49
203–4, 316n19 Vaidyanathan, Kunnakudi, 53–54, 104

348 index
Varahappayyar (minister), 30–31, 61 ‘‘naturalness’’ of, 18–19, 127–29,
Vasudevachar, Mysore, 70, 74, 247–49 139, 141, 144–45; politics or ide-
Vedavalli, M. B., 65, 68 ology of, 5–14, 115–16, 148–49,
veena, 26, 41, 45, 47, 136, 258 152–53, 190–91, 245–46, 276, 287–
Vellalas, 16; Icai Veḷḷālars, 141, 143, 188, 89; psychoanalytic approaches to,
294n4, 316n16 10–11; ventriloquism and, 55–58,
Venkatachalam, G., 123, 127, 128 297n32; ‘‘vocal anthropology,’’ 13,
Venkatachalapathy, A. R., 318n28 111, 292n14
Venkatamakhi (treatise author), 233–
34, 318n30 Weber, William, 5, 86, 94, 291n5,
Venkataramiah, K., 61–63 324n1
Venkitasubramonia Iyer, 31, 63–64 Western classical music, 4, 5, 62,
violin: as accompaniment to voice, 64, 66, 86, 94, 150, 185, 232, 234,
33–37, 288–89; in experimental 300n27, 306n23, 324n1
music, 52–54; ‘‘fiddle’’ and, 37, 42, Williams, Raymond, 309n49
294n5; horn or ‘‘Stroh,’’ 44, 68, Wodeyar, Chamaraja, 65
69, 296n17; mechanical, 43–44; in Wodeyar, Jayachamaraja, 65, 66
South India, 14–15, 25–58; styles Wodeyar, Nalwadi Krishnaraja, 59,
of playing, 49–51; techniques of 65–67
playing, 37–42, 50–51, 56 Wodeyar, Yuvaraja Narasimha, 67
visibility, 38, 41; logic of, 205–7, woman question, 18
209–11, 235. See also legibility women: idealized, 114–16, 129, 135–
Vizianagaram, 64, 65, 68–69 36, 141, 144–45, 149, 158, 302nn1–4;
voice: agency and, 148, 289; anxi- unmarried, 111–14
eties about loss of, 242, 246–47, Women’s India Association, 18, 117,
258; body and, 146–49; gender and, 294n25
115–16, 121–29, 145–48; micro-
phone and, 87–92; modernity Žižek, Slavoj, 297n32
and, 6–9, 19–20, 190–91, 286–89; Zvelebil, Kamil, 154, 310n3, 311nn4–5

index 349
Amanda Weidman is an assistant professor of
anthropology at Bryn Mawr College.

Library of Congress Cataloging-in-Publication Data

Weidman, Amanda J., 1970–
Singing the classical, voicing the modern : the
postcolonial politics of music in South India /
Amanda J. Weidman.
p. cm.
Includes bibliographical references and index.
isbn 0-8223-3631-6 (cloth : alk. paper)
isbn 0-8223-3620-0 (pbk. : alk. paper)
1. Music—India, South—History and criticism.
2. Music—Social aspects—India, South. 3. Politics
and culture—India, South. I. Title.
ml338.1.w45 2006
781.6'90954—dc22 2005037850