The Hegemony of Heritage by Mapin Publishing

The Hegemony of Heritage ritual and the record in stone

DEBORAH L. STEIN

The Hegemony of Heritage The Hegemony of Heritage makes an original and significant contribution to our understanding of how the relationship of architectural objects and societies to the built environment changes over time. Studying two surviving medieval monuments in southern Rajasthan—the Ambika Temple in Jagat and the Sri Eklingji Temple complex in Kailashpuri—the author looks beyond their divergent sectarian affiliations and patronage structures to underscore many aspects of common practice. This book offers new and extremely valuable insights into these important monuments, illuminating the entangled politics of antiquity and revealing whether a monument’s ritual record is affirmed as continuous and hence hoary or dismissed as discontinuous or reinvented through various strategies. The Hegemony of Heritage enriches theoretical constructs with ethnographic description and asks us to reexamine notions such as archive and text through the filter of sculpture and mantra. “Makes visible the multiple methodologies that can be mobilized to write nuanced histories of Hindu temple architecture. The author’s approach is both refreshing and new. Skillfully weaving in postcolonial theory, object ontologies, and affect theory, among other approaches, the book opens up an exciting paradigm in the study of South Asian art and architecture.” SUGATA RAY, Assistant Professor of South Asian Art and Architecture, University of California, Berkeley

The Hegemony of Heritage

The Hegemony of Heritage Ritual and the Record in Stone

Deborah L. Stein

First published in India in 2019 by Mapin Publishing Pvt. Ltd 706 Kaivanna, Panchvati, Ellisbridge Ahmedabad 380006 INDIA T: +91 79 40 228 228 • F: +91 79 40 228 201 E: mapin@mapinpub.com • www.mapinpub.com Publication of this book has been made possible with the support of Maharana Mewar Historical Publications Trust, a unit of Maharana of Mewar Charitable Foundation, The City Palace, Udaipur 313001, Rajasthan, India. First published in the United States of America in 2018 by University of California Press Available in South Asia: Mapin Publishing Pvt. Ltd © 2018 by Deborah L. Stein Text © Deborah L. Stein Illustrations © Deborah L. Stein, unless otherwise mentioned. All rights reserved under international copyright conventions. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording or any other information storage and retrieval system, without prior permission in writing from the publisher. The moral rights of Deborah L. Stein as author of this work are asserted. ISBN: 978-93-85360-61-9 Copyediting: Joe Abbott Editorial support (Indian edition): Neha Manke / Mapin Editorial Production: Mapin Design Studio Printed in India

F OREWORD We have had a very long and meaningful relationship with Dr. Deborah L. Stein, who teaches Art History at California College of Arts (CCA) and San Francisco State University, USA. She is richly deserving of our warmest congratulations on the publication of The Hegemony of Heritage: Ritual and the Record in Stone. This volume is based on her dissertation from U.C. Berkeley and a year of fieldwork in southern Rajasthan, India. Universities like U.C. Berkeley are equally worthy of congratulations for providing the academic platforms and rigorous intellectual environment for young scholars such as Dr. Stein to generate research of lasting value. Undoubtedly, with this publication, the University has also played a significant role in building bridges between our nations, communities and cultures. It is indeed laudable and praiseworthy. The Hegemony of Heritage has covered new ground. It has, as Dr. Stein writes, looked at Indian temples in a new way, as ‘catalyst’ agents. She has demonstrated her belief in studying South Asian religious monuments as catalysts, ‘eliciting new modes of art historical inquiry far beyond the buildings or sculptural iconography alone.’ At the Maharana of Mewar Charitable Foundation in Udaipur, we have supported Dr. Stein’s research and this publication for a very long time, and in every which way possible. It showcases our continuing commitment to encourage credible academic research, debate and discussion on the history of Mewar, Rajasthan and India. The ancient region of Mewar, its temple towns and the fort cities of Chittorgarh and Kumbhalgarh provide a seemingly infinite array of research themes even in the twenty-first century. We welcome institutions, scholars and researchers from every discipline to explore and discover its new facets. Our Maharana of Mewar Charitable Foundation is mission-inspired to work towards the preservation of the ‘living heritage’ of Mewar. We are serving society and shouldering the moral responsibilities bequeathed to us as the legacy of the Custodians of Mewar. I pray to the Supreme Lord of Mewar, Parameshwaraji Maharaj Shree Eklingnath ji, to shower His blessings on all those who have worked on this publication and especially on the readers, who will gain from it in the years ahead. Arvind Singh Mewar Chairman and Managing Trustee Maharana of Mewar Charitable Foundation

This book would not have been possible without my grandparents, my parents, and the influence of the late Alan Dundes. This book is dedicated to my husband, Laurent Goldsztejn, and my two sons, Ariel Goldsztejn and Aiden Goldsztejn.

Contents

List of Illustrations Acknowledgements Introduction: The “Hindu” Temple in Diachronic Context

ix xv 1

1. Temple as Geographic Marker: Mapping the Tenth-Century Sectarian Landscape

2. Temple as Catalyst: Renovation and Religious Merit in the Field

3. Temple as Royal Abode: The Regal, the Real, and the Ideal in Fifteenth-Century Mewār

4. Temple as Palimpsest: Icons and Temples in the “Sultanate” Era

120

5. Temple as Ritual Center: Tenth-Century Traces of Ritual and the Record in Stone

149

6. Temple as Praxis: Agency in the Field in Southern Rājāsthan

186

7. Temple as Legal Body: Aesthetics and the Legislation of Antiquity

220

Conclusion: Heritage and Conflict: Medieval Indian Temple as Commodified Imaginary

243

Notes Bibliography Index

251 287 305

Illustrations

0.1. 0.2. 0.3. 0.4. 0.5. 0.6. 0.7. 0.8. 0.9. 0.10. 0.11. 1.1. 1.2. 1.3. 1.4. 1.5. 1.6. 1.7. 1.8. 1.9.

Installation ceremony at Ambikā temple, Jagat, May 2002 2 Śri Ekliṅgjī temple 4 Śri Ekliṅgjī icon 5 Pippalāda Mātā temple (c. 960), Unwās, Mēdapāṭa region 10 Kṣēmaṅkarī, Pippalāda Mātā temple, Unwās 10 Main icon at Pippalāda Mātā, Unwās 11 Old and new architecture and repairs comingle at the Śri Ekliṅgjī temple complex, Kailāśpurī, Rajasthan 12 Map of sites in modern Mewār and Dūṅgarpur 13 Ambikā temple, Jagat 14 Chapati dough ornaments for Daśamātāpūjā 18 Painting the sanctum gold, Ambikā temple, Jagat, May 2002 19 Lakulīśa temple (c. 971), Ekliṅgjī 25 Saraswatī, Lakulīśa temple, Ekliṅgjī 26 Śivēśvara temple (c. 950s–70s), Ekliṅgjī 27 Sūrya on his seven-horse chariot (c. 950–75), Ṭūṣa 28 Surasundarī, Ṭūṣa 29 Door to sanctum on each side for bidirectional circumambulation, Ambikā temple (c. 960), Jagat 35 Durgā zoomorphic, side 1 (south), Ambikā Temple, Jagat 36 Durgā zoomorphic/anthropomorphic, side 2, (east/back), Ambikā temple, Jagat 36 Durgā anthropomorphic, side 3 (north), Ambikā temple, Jagat 37

Illustrations

1.10. An icon cast aside inside the sanctum displays an emaciated form of Durgā killing the buffalo demon. Ambikā temple, quartzite (c. 960), Jagat 38 1.11. Ferocious Śiva (Andhakāntaka?) (c. 955–75), stone, Hita 40 1.12 Nateśa (c. 955–75), stone, Hita 40 1.13. Cāmuṇḍā, Nateśa temple (c. 955–75), stone, Hita 40 1.14. Exquisite śekharī architecture, Nateśa temple (c. 955–75), stone, Hita 40 1.15. Durgā-Mahiṣāsuramardinī in the identical artistic style of Jagat (c. 960), Hita 42 1.16. Guru and disciple on the temple wall (c. twelfth to thirteen century), stone, Gamari 43 1.17. Bestiality on the temple wall (c. twelfth to thirteenth century), stone, Gamari 44 1.18. Mandala in stone, with devotees ladling ghee over a fire (c. tenth or eleventh century), Nāgadā 46 1.19. Mandala made of legumes, Jagat, May 2002 47 1.20. Durgā-Mahiṣāsuramardinī (c. tenth or eleventh century), stone, Nāgadā 48 1.21. Jagat praṇālā, demon’s leg protruding out of the neck of the decapitated form 49 1.22. Toraṇa gate, Nāgadā 50 2.1. Kumbhalgarh fortress 54 2.2. Rampart, fifteenth century, Ekliṅgjī 55 2.3. Marble icon (stolen in 1998), Jagat 60 2.4. New marble image from Jaipur, May 2002 61 2.5. Śubhamaṇḍapa, archival photo showing building foundation (c. eleventh century) and brick layer (c. 1800s–1900s), Jagat 62 2.6. Śubhamaṇḍapa, Jagat 63 2.7. New icon, under worship in the Ambikā temple, 2009 65 2.8. Perishable maṇḍapa built for the goddess installation, May 2002, Jagat 66 2.9. Diagram of the social space of the goddess installation ceremony, May 2002, Jagat 68 2.10. Mudbrick shrine, Āmjhara (near Dūṅgarpur) 70 3.1. Bappa Rāwal, by a French sculptor (c. second half of the twentieth century), Ekliṅgjī 75 3.2. Modern sculpture of Harit Rashi in a Sanskritic haṁsa vehicle (swan boat), Sri Ekliṅgjī temple 75 3.3. Harit Rashi on an exterior wall in Udaipur during Rath Yātrā 76 3.4. Painting of Harit Rashi’s apparition (c. 1850–1950) 76 3.5. Pratap, Rath Yātrā parade, Udaipur, 2002 79 3.6. Kīrtistambha tower (c. 1440–60), Chittauḍgaḍh 81 3.7. Labeled sculpture of servants, interior of Kīrtistambha tower (c. 1440–60), Chittauḍgaḍh 83

Illustrations

3.8. 3.9. 3.10. 3.11. 3.12. 3.13. 3.14. 3.15. 3.16. 3.17. 3.18. 3.19. 3.20. 3.21. 3.22. 3.23. 3.24. 3.25. 3.26. 3.27. 3.28. 3.29. 4.1. 4.2. 4.3. 4.4. 4.5. 4.6. 4.7. 4.8. 4.9. 4.10. 4.11. 4.12. 4.13.

Sculpture of Viṣṇu, interior of Kīrtistambha tower (c.1440–60), Chittauḍgaḍh 85 Deogarh, Madhya Pradesh 87 Narrative scene, Nara Narayana, Gupta Era, Deogarh 88 Harihara (Half-Śiva/Half-Viṣṇu), interior Kīrtistambha tower (c. 1440–60), Chittauḍgaḍh 89 Interior stairwell between floors 2 and 3, Kīrtistambha tower (c. 1440–60), Chittauḍgaḍh 90 Ardenareśvara (Half-Śiva/Half-Pārvatī), interior Kīrtistambha tower (c. 1440–60), Chittauḍgaḍh 91 Allah, Interior Kīrtistambha tower (c. 1440–60), Chittauḍgaḍh 92 Interior stairwell precedent, Jain Kīrtistambha tower, Chittauḍgaḍh 94 Ekliṅgjī temple roof (fifteenth century) 99 Roof of the Samiddhēśvara temple (fifteenth century), Chittauḍgaḍh 100 Mahārāṇā Mokal sponsors the Bhāghelā Tālāv, Ekliṅgjī 101 Jāwar Mātā temple 102 Keśeriyajī/Rishabdeo temple, Delwara 105 The Mīrabai temple, Ekliṅgjī 108 Mīrabai temple, rear view, Ekliṅgjī 109 Viṣṇu icon, Mīrabai temple, Ekliṅgjī 110 Viṣṇu icon in fifteenth-century temple wall, Mīrabai temple, Ekliṅgjī 111 Ramabai temple tank, Jāwar 112 Ramabai icon, Jāwar 113 Palace where the historical Rani Padmini resided within the fortress of Chittauḍgaḍh 115 “The Water Palace of HH Rani Padmawati, Chittauḍgaḍh”, Photographed by J. W. Caplain 116 Queen Elizabeth receiving the tower of “Victory,” Udaipur Palace 118 Trimūrti liṅgaṃ, Kumbhesvara temple, Chittauḍgaḍh 124 Inscription on the Chaturbhuj temple, Khamnor 125 Four-Faced Śiva icon (c. 975), Khamnor 126 Chaturbhuj icon in mirrored hall, twentieth-century mirrorwork, Khamnor 127 Śekharī-style temple (c. eleventh to thirteenth century), Talwara 128 Brahma temple (c. twelfth century), Chinch 129 Raisamand Lake (c. fifteenth/sixteenth century) 129 Ithyphallic Lakulīśa, Menāl 131 Mahanaleśvara temple (c. eleventh century), Menāl 132 Cāmuṇḍā, Menāl 132 Nateśa, Menāl 133 Andhakāntaka (?), stone, Menāl 134 Maṭha (c. tenth century), Menāl 134

xii

Illustrations

4.14. 4.15. 4.16. 4.17. 4.18. 4.19. 4.20. 4.21. 4.22. 4.23. 4.24. 5.1.

Column detail (c. eighth century), monastery in Menāl 135 Śiva temple (c. twelfth century), Bijoliā, Uparamāla 136 Cāmuṇḍā on south side of Śiva temple (like at Jagat), Bijoliā 136 Nateśa, Bijoliā 137 Sahasraliṅga (c. twelfth to thirteenth century), Bijoliā 138 Traces of ritual made permanent in stone, Bijoliā 138 Vindhyāvāsinī temple (c. twelfth century), Kailāśpurī 139 Ganesh, Ambikā temple, Jagat 143 Śri Ekliṅgjī painting (c. 1700), Juna Mahal, Dūṅgarpur 145 Harit Rashi painting (c. 1700), Juna Mahal, Dūṅgarpur 146 Deo Somnāth temple (c. twelfth century), near Dūṅgarpur 147 Woman pours wine into a cup held by a man (detail), Ambikā temple, Jagat 150 Sahasraliṅga (thousand-faced liṅga) may well date to the tenth century, Śri Ekliṅgjī temple compound, lower level 153 Sahasraliṅga, (c. tenth century), Āhaṛ 153 Sadāśiva head, stone, (c. 500–600 CE), thakur’s compound, Rawala, Jagat 154 Four-faced (and four-bodied) liṅgaṃ from Kalyanpur, black schist, (c. eighth century) 155 Stone liṅgaṃ at Ahar (c. tenth century) 156 Detail flowers atop Stone liṅgaṃ at Ahar 157 Toraṇa column detail (c. tenth century), Āaṭ 159 Monastery (c. tenth century), Āaṭ 160 Gaṇēśa (c. 950–60), quartzite, Āaṭ 161 Toraṇa gate into the site of Āaṭ 162 Śiva and Pārvatī (c. tenth century), quartzite, Āaṭ 163 Cāmuṇḍā in situ (c. tenth century), quartzite, Āaṭ 164 A large Durgā-Mahiṣāsuramardinī (c. tenth century), Āaṭ 165 A large lion base and serpent-hood awning (c. third quarter of tenth century), Āaṭ 166 Kṣēmaṅkarī (c. tenth century), Lodravā 167 Vyāla leonine figure next to surasundarī, celestial maiden (c. 960), quartzite, temple, Jagat 170 Circumambulation with priests and women carrying waterpots during installation rites 173 Śiva temple (c. tenth century), Śobhagpura 174 Devotees staring out of veranda (c. tenth century), Ambikā temple, Jagat 175 Kṣēmaṅkarī, lintel above front entrance (c. 960), Ambikā temple, Jagat 180 Ghaṭeśvara Mahādeva temple, Bāḍoli in Uparamāla 180

5.2. 5.3. 5.4. 5.5. 5.6. 5.7. 5.8. 5.9. 5.10. 5.11. 5.12. 5.13. 5.14. 5.15. 5.16. 5.17. 5.18. 5.19. 5.20. 5.21. 5.22.

Illustrations

6.1. 6.2. 6.3. 6.4. 6.5. 6.6. 6.7. 6.8. 6.9. 6.10. 6.11. 6.12. 6.13. 6.14. 6.15. 6.16. 6.17. 6.18. 6.19. 6.20. 6.21. 6.22. 6.23. 6.24. 7.1. 7.2. 7.3. 7.4. 7.5. 7.6. 7.7. 7.8. 7.9.

xiii

Yogis on the Śri Ekliṅgjī temple 187 Worship of a liṅgaṃ from Jagat 188 Śri Ekliṅgjī as featured in the palace calendar, 2002 190 Śri Ekliṅgjī popular prints for sale during Śivratri, 2002, Ekliṅgjī 191 Portrait of the deity Śri Ekliṅgjī with Maharana Bhagwat Singh and the royal family at Śri Ekliṅgjī 192 The wife of a Śri Ekliṅgjī priest pours water into an aqueduct leading into the inner sanctum, July 2002, Ekliṅgjī 197 Clay liṅga, Ekliṅgjī 198 Worshipping cloth, Ekliṅgjī tailor’s fair 199 Ties that bind, fabric strung up the hill at Ekliṅgjī 200 Śītalā-saptamī worship, Mewār 203 Śītalā worship in side niche of Ambikā temple, Jagat 204 New goddess icon, installed May 2002, Ambikā temple, Jagat 206 Mallar Mātā, Mallar Mātā hilltop shrine, Jagat 207 Cāmuṇḍā (c. 960), Ambikā temple compound, Jagat 207 Fragments from Ambikā temple complex (c. 960), Jagat 207 Tenth-century stone Cāmuṇḍā on the exterior wall of the Ambikā temple 208 Sacrificed goat, winter 2002, Mallar Mātā shrine 210 Paddhi channel planted around the temporary pratiṣṭhā pavilion, midfestival and after one-week festival, May 2002, Jagat 211 Pūrṇaghata motif next to woman with overflowing pot on her head, May 2002, Jagat 212 Children enjoying garbha dancing, 2002, Jagat 213 “Portrait” of the goddess Ambā Mātā, Ambikā temple, spring 2002 214 Plastic Meena goddess, later drowned in the well, May 2002, Jagat 215 Tenth-century iconography, Durgā Mahiṣāsuramardinī (c. 961 CE), quartzite, Ambikā temple, Jagat 216 Original Śītalā icon, broken from her frame, covered in dots of sindūr 218 Samadhi Spot of the last nonhouseholding monk of Ekliṅgjī 225 Maṭha (Pāśupata monastery), painting of Puja at the Śri Ekliṅgjī temple (c. eighteenth century), Ekliṅgjī 226 Jantar Mantar (sixteenth century), Jaipur 232 Stepwell (c. ninth century), Ābhānerī 233 Paramāra temples (c. tenth century), Arthuna 234 Lakulīśa (c. tenth century), Arthuna 235 Cāmuṇḍā (c. tenth century), Arthuna 235 A śekharī-style temple, Arthuna 235 New Ambā Mātā sculpture and Śītalā Mātā sculpture, blindfolded, before their eye-opening ceremonies 238

xiv

Illustrations

7.10. Painted fragments from the Mallar Mātā shrine 240 7.11. Worship of Absence, 2000–2002, Ambikā temple, Jagat 241 8.1 Vermilion footprint, twenty-first-century sindūr on tenth-century quartzite stone, Ambikā temple, Jagat 247

Acknowled gements

Vidya Dehejia introduced me to Indian art when I was a freshman at Barnard and, subsequently, as a mentor at the Smithsonian Institute. Since then, so many have guided my projects, and I am grateful to each of them. My senior thesis advisers, Rachel Fell McDermott and Benjamin Buchloh, as well as Keith Moxey, Nathalie Kampen, Barbara Novak, Mme Parlier-Renault (Paris IV), Joan Cummins, Peter Conor, M. Beguin and M. Maucuer (Musee Cernuschi), Paul (librarian at the Bibliothèque du Musée Guimet), Barbara Ford (Metropolitan Museum of Art), Vishakha Desai (Asia Society), Robert Thurman, Anne Boyman, and Gary Tubb played pivotal roles in my introduction to art history, French thought, linguistics, Sanskrit, Hinduism, Buddhism, and Asian Art in museums around the world. Angela Zito served as faculty adviser for the Religious Diversity Suite housing and outreach project that I started at Barnard and as a mentor in Religious Studies and Anthropology. I would like to thank all those who guided the early stages of this project as a dissertation at Berkeley, including Joanna Williams, Greg Levine, Pat Berger, Darcy Grigsby, Margaretta Lovell, and all of my fellow founders and members of the Asian Art Working Group, founded in 1999 and supported to this day by the Townsend Center. Usha Jain, Ved Vatuk, Steven Goodman, Cynthia Col, and Sally and Bob Goldman taught me Hindi, Tibetan, and Sanskrit, without which none of this project could have been completed. In the later stages of the dissertation, many supported the anthropological and ethnohistorical methodologies of my work including my inspiration, Michael Meister, my posthumous mentor, Alan Dundes, and his faithful successor, Whitney Davis, as well as Lucinda Ramberg, Adheesh Sathaye, and all of my fellow graduate students in Alan Dundes’s Methods xv

xvi

Acknowledgements

and Theories Seminar. Priya Joshi, with her courses on Indian cinema, also had a significant influence on the postcolonial perspective of this work. I would also like to thank my fellow students and Tapati Guha-Thakurta for her insights during my last course at Berkeley. Kajri Jain and Tapati both inspired me with their talks at Stanford as well. Friends and peers, including but not limited to, Alka Patel, Jinah Kim, Catherine Becker, Alka Hingorani, Kirtana Thangavelu, Soo Kim, Sharon Yamamoto, Orna Tsultem, Wenshing Chou, Sunglim Kim, Doryun Chong, Debra Diamond, Molly Aitkens, Joan Cummins, Rebecca Brown, Julie Romain, Sugata Ray, Aditi Chandra, Santhi Kavuri-Bauer, Atreyee Gupta and the younger generation of the Asian Art Working Group, including William Ma, Ryosuke Ueda, and Patricia Yu have stood by me over the past twenty years unwaveringly, regardless of my academic status as visiting assistant professor at Mills, adjunct lecturer in the UC system, visitor at SF State, or independent scholar. The generous group of South Asianists at the American Council of Southern Asian Art (ACSAA), including but not limited to, Rebecca Brown, Padma Kaimal, Darielle Mason, the Ashers, the Huntingtons, Walter Spink, Bob Brown, Janice Leoshko, Lisa Owen, Anna Seastrand, Subhashini Kaligotla, Ajay Sinha, Laura Weinstein, Mary-Beth Heston, Deepali Diwan, Melody Rodari, Gary Tartakov, Richard Davis, John Cort, and especially Pika Ghosh (who read and greatly influenced my work) have also provided support and encouragement over the past two decades at our biannual symposia. Deven Patel read drafts of the historical chapters, and Arjun Gupta intervened at critical moments with legal advice and moral support. Robin Tovey and Roger Snell also read and proofed earlier versions of this work. At Berkeley, Alex von Rospatt provided important discussions of pratiṣṭhā at key intervals; and from abroad, Sanjukta Gupta shared her Sanskrit knowledge of goddess texts and traditions. In the field, Tamara Sears and I accompanied each other to Terahi, Chittor, Jagat, Menal, Badoli, Bijolia, and many other sites in both Madhya Pradesh and Rajasthan, measuring monasteries, photographing architecture and iconography, and spending hours together in the back of the car traversing India, discussing ideas, sharing humor and even an occasional song. In India, people across academic, domestic, and intellectual spheres adopted me and supported this project with love, generosity, intellect, and goodwill. Narendra Dashora spent hours poring over the Puja Paddhati with me, and his family who taught me family recipes, included me in every holiday and fed, housed, and loved me like their own daughter. In Udaipur, Annique Prunet graciously offered a francophone home base and a lively interlocutor with whom I could discuss all aspects of fieldwork and who made the most wonderful whiskey sours. In Jagat, I would like to thank the Chauhan family, who housed me, kept safe with a brother, entertained with a sister, well-fed with a mother, and historically informed with a father. I would also like to thank the priest Shankar Giri, who welcomed me at every ritual and was so patient until my Hindi improved enough to joke around with him properly.

Acknowledgements

xvii

At the American Institute of Indian Studies (AIIS) in Delhi, Dr. Mehendiratta and Purnima helped with all administrative aspects of visas and kept me safe and comfortable during my fieldwork in 2002. M. A. Dhaky, Vandana Sinha, and Mr. Yadav were helpful at AIIS in Gurgaon, sharing ideas over lunch, navigating the photo archives, and discovering that library. At the Archaeological Survey of India (ASI), the director and Mr. Bakshi ensured that my photography and research permissions were granted quickly. The Jaipur director of the Rajasthan State Archeological Department granted all permissions and gave full archival and archaeological access. R. C. Agrawala invited me to his home on a moment’s notice and shared the wealth of his personal library and archives. In Jodhpur, Bhavana and Chandrashekar provided a wonderful place to stay, and the staff of the Rajasthan State Archives in Bikaner provided full access to manuscripts. In Udaipur, the palace architect, Batul Raaj, provided endless comradery. She introduced me to Shriji, who gave permissions for photography at Ekliṅgjī and invited me for a memorable evening at the palace with Batul. Since then HRH Śriji Arvind Singh Mewār, as well as his daughter and son in law in the United States, have lent unwavering encouragement and administrative support for this project over the past decade. In Dūṅgarpur, the mahārawal and his family introduced me to their state historian, Mr. Purohit, and gave me full permission to photograph in the old palace and to access their archives. This project could not have been completed without the help of Vanita Ojha, Atulya Vyas, R. K. Purohit, Manish Bharadwaj, and my Mohan Lal Sukhadia University sponsor, Professor Meena Gaur. In the Udaipur Devasthan Department, Director/Ms. Poonam Sagar kept feminism, caste, and class on the radar in important ways. On a return trip in 2009, Santosh Sadanandan, and Shivaji Panikkar and many others joined in fruitful collaborations on the idea of an Archive. I could not have safely reached the many sites photographed in this book without my driver Jagadish and would also like to thank Dadaji for taking me to Eklingji for Monday bhajans whenever I wanted to tag along. Financially, this project could not have been completed without US government support in Hindi through a Foreign Language and Area Studies (FLAS) grant from the state department, multiple years of graduate fellowship and teaching assistantships at the University of California, Berkeley, as well as editorial support from the South Asia Across the Disciplines (SAAD) series collaboration of Columbia, University of Chicago, and University of California Presses for first-time authors of South Asia monographs. After writing and research was completed entirely thanks to these public grants and funds, this project could not have come to press without the generous book subvention from the Maharana Mewār Charitable Foundation, which allowed me to give back as an author by making this research open access and available as a free e-book. Reed Malcolm, my editor at UC Press, made this open access collaboration with Luminos possible, and Zuha Khan, with the help of her interns, shepherded through an art history monograph with twice the number of images normally included in a single book. I would like to thank Sheldon

xviii

Acknowledgements

Pollock, who first encouraged me to submit my manuscript to the SAAD series, and everyone at Columbia and the University of California Press who helped me bring this work to press in print and digitally. My two anonymous readers gave helpful feedback and helped me to advocate for this book’s topic and perspective with their own astute observations about the work in relation to our field. Joe Abbott has served as a dedicated copyeditor, attuned with Indian diacritical spelling and culture, and Francisco Reinking lent his expert hand during production. I would also like to thank Paige MacKay at Ubiquity Press, as well as Bipin Shah, Neha Manke and Gopal Limbad at Mapin Publishing, as well as Mayank Gupta and Śriji Arvind Singh Mewār in India. No project authored by a mother could be completed without a transparent nod to those who sustained me in every way with childcare, love, and support as I first wrote drafts while breastfeeding and completing graduate school. My husband took our first son to the market every Saturday as I wrote my dissertation, accompanied me and our babies on lectures and job talks all over the country, and stood ready to move at a moment’s notice for years on end. My father, Dr. Eugene J. Stein, raised me with lovingkindness to be his feminist daughter and took care of my young son while I taught at Berkeley and Mills. Carole Rathfon, Jo Boyer, and Karen Seidman also watched my children when I wrote and when I taught at Mills and when a caregiver called in sick. In particular, I would like to thank Kavita Singh, Saloni Mathur, Saira Wasim, Suzanne Preston Blier, Steven Nelson, Judith Rodenbeck, Jaleh Mansoor, T’ai Smith, Carole Birkan, Stephane Kerber, Jackie Jung, Stephen Campbell, Danijel Benjamin Ćubelić, Leyla Jagiella, Shivaji Panikkar, Shuddabrata Sengupta, and Muhammad Salman Khan, along with too many more to name here who provided me with an imaginary art history department on Facebook during my difficult years as an independent scholar, and Mary Ann Milford, who has always been a delightful mentor and colleague over many years at Mills College. Finally, I would like to thank my mother, Dr. Diane C. Stein, for reading to me, taking me to the library, and buying me books, and both my parents for lively dinner conversation, dragging me to too many museums and archaeological sites as a reluctant child and enthusiastic teen, and giving me the gift of a Barnard education. My husband, Laurent Goldsztejn, contributed a levelheaded stability and much of the financial support over the many years it took to sustain this work. Through couvade, one might imagine, he even authored his own work of French fiction becoming a fellow published author during this long writing process. This book is dedicated to our two sons, Ariel and Aiden Goldsztejn, who are also authors, and whose lively minds and loving souls inspire me to be my best in every aspect of my life. Lastly, I would like to thank Wanda Chan, Megan Rathfon, Meriem Curtil, Mariel Vazquez, and many more close women friends who have seen me through thick and thin over the years with love and intellect.

Introduction The “Hindu” Temple in Diachronic Context

What do we see when we look at a monument, and how do we come to see what we do? Far from the innocent ravages of time, the calculated aesthetics of the Indian temple today result from the confluence of religious performance, the politics of identity formation, the tension between neoliberal and socialist preservation models, and the display, erasure, and fragmentation of the visual and material record. Architecture gives an illusion of eternal permanence only to reveal a state of perpetual flux both in meaning and in form. Through a thorough examination of two sites in southern Rājāsthan, we gain insight into a process of curating from the field whereby the erstwhile colonial institutions and socialist state compete with a variety of private initiatives for the right to construct the past and future alike. Across India, ancient sites are put back into worship, left untouched, or visited by throngs of tourists and pilgrims. A diachronic history of temples can lead us to examine how various actors claimed power and authority and shaped notions of sacred space and ritual praxis over time. A T E M P L E C OM PA R I S O N : N EW M AT E R IA L I SM I N A R ĀJĀ ST HA N I C A SE ST U DY

Chosen among the Mēdapāṭa regional cohort of temples as the two in most active worship today, the Ambikā temple in Jagat and the Śri Ekliṅgjī temple in Kailāśpurī serve as a case study of the larger pan-Indian phenomenon of putting temples “back” into use and reflect modern people’s praxis in great depth. In light of their particular histories, this book proposes that we look at Indian temples in a new way, as “catalyst” agents—generative architecture that sparks a wide variety 1

Introduction

Figure 0.1. Thakur of Jagat and his wife pour ghee onto a sacrificial fire with priests from Īdar, Gujarat, on the occasion of the installation (pratiṣṭhā) ceremonies for the installation of a new twenty-first-century goddess into the tenth-century inner sanctum of the Ambikā temple, May 2002. © Deborah Stein.

of ritual and other activity, often far from the temple itself. As an active catalyst of a wide variety of human interaction, the temple burns brightly and is never used up (fig. 0.1). Beyond a single region or sectarian origin, the idea of South Asian religious monuments as catalysts elicits new modes of art historical inquiry far beyond the buildings or sculptural iconography alone. Whether we consider the aesthetics of the Taj Mahal today, a Chola bronze of Śivanataraja in a medieval procession, or the famous Jataka scenes on the gateways of Sanci, the idea that these architectural locations are catalysts for whole bodies of diachronic and ephemeral material practices and performative praxis expands our field of inquiry as art historians. Whereas one could misconstrue the core comparison of this book as a reinforcement of a false binary between a tantric/female/rural/populist periphery and an elite/male/dynastic center, the Ambikā temple in Jagat and the Śri Ekliṅgjī temple in Kailāśpurī offer much more than a simplistic dialectic. Both sites offer a radical array of materials beyond style and stone to propose an alternative to high modernist notions of Hindu temple architecture through new materialist approaches to butter, flour, vermilion, and the primary importance of the materiality of the stone itself over its mere figuration. Rather than create a dichotomy between center and periphery in a single time frame, these sites offer multiple perspectives that vary greatly from one era to the next—at times serving as key

The “Hindu” Temple in Diachronic Context

centers of religious and political activities and other moments fading from the historical record entirely. Without the modern performance of the puja-paddhati at Śri Ekliṅgjī, the use of mantra in the installation rites at Jagat would not have had the same clarity, nor would the relationship between the books of ritual liturgy and the ritual performance unfolding. A postcolonial approach to the method of material culture reveals some difficult and at times incongruous ideas. For example, if objects, buildings, and materials have agency, some may argue that this power, inherent in the material world, takes away from the agency of the human agents who engage with those materials.1 When Alfred Gell argues that agency lies in the work of art, does that mean that those who made and use it are erased? Alternately, what are the risks as an art historian of a suspension of disbelief when discussing a religious icon believed by devotees to be alive? This is where the idea of the temple as a catalyst becomes even more important. As an active agent that is never used up, material remains can spark human agents to a wide variety of actions at different points in time. Within each time period, diverse agents interact with those materials quite differently.2 Can objects or buildings speak for themselves? No, of course, these materials cannot. People—consciously or unconsciously, both individually and collectively—leave in these materials traces of their ideas, behavior, and uses of these sites over time. Their ritual residue is the stuff of this study. Material residue, ritual residue, stone residue, aesthetic residue, physical residue, temporal residue, and architectural residue each reflect the material traces that are left behind at religious sites intentionally and unintentionally through ritual practice. The Ambikā temple in Jagat and the Śri Ekliṅgjī temple in Kailāśpurī, as two of the most active sites of worship in southern Rājāsthan today, serve as prime examples of how religious monuments serve as catalysts for a wide variety of praxis in South Asia, often in a radius as large as two kilometers or more from the building itself, and at times in giant networks, such as the goddess network between Jāwar, Jagat, and Īdar. This triangle of sisterly geomantic relationships is documented folklore, such as the sung Jagat Mata ki Katha and the Jāwar Mata ki Katha narratives recorded on cassette tapes and sold at the village bus stand. The material remains of ritual, also referred to as ritual residue, provide physical traces of agents’ actions in relation to these temples as catalysts for praxis. Many temples and sites across India correspond to this phenomenon today, but few articulate as clear a set of diachronic histories as those found in the kingdom of Mewār. This book compares two key tenth-century sites in southern Rājāsthan to reveal very different sectarian foci and histories of religious use. The first is a temple called the Śri Ekliṅgjī temple (fig. 0.2), dedicated to the god Śiva. Today it is said that a god named Śri Ekliṅgjī has ruled the kingdom of Mewār for more than one thousand years (fig. 0.3). An inscription dated to 971 CE corroborates this idea with an early link between the ruling dynasty and the patron saint of Śri Ekliṅgjī’s Śaiva sect named Lakulīśa. Dedicated to this saint, the monastic Lakulīśa temple

Figure 0.2. Śri Ekliṅgjī temple. © Deborah Stein.

Figure 0.3. Maharana Bhupal Singh at Manorath, Śri Ekliṅgji Temple. Maharana Bhupal Singh at worship with the head priest in front of the deity at the Śri Ekliṅgjī Temple, Kailashpuri. Devare & Co., Gelatine silver print. Printed from glass plate negative, 1935–1945 CE. Accession No. 2008.06.0435_R. Image courtesy: The City Palace Museum, Udaipur, © MMCF.

Introduction

displays fine masonry and smooth walls to suggest the focused practice of Śaivaite gurus and ascetics and the architectural location of their intellectual exchange with Jains and Buddhists. In contrast with this imperial center of political and religious authority, the Ambikā temple in the town of Jagat is dedicated to a goddess who quells the buffalo demon to restore cosmic order. The complex figural program on the exterior walls of this goddess’s temple suggests that syncretic modes of practice attempted to incorporate local religion into increasingly systematized modes of brāhmanical Hinduism for a popular audience. Today, the mahārāṇā and the state of Rājāsthan legally contest the Ekliṅgjī temple, whereas the Ambikā temple’s ownership is contested through the display of modern icons in the ancient sanctum after the icon’s theft for the international art market. The Śri Ekliṅgjī temple is currently under a longue-durée trial between the Devasthan Department of Living Temples of the government of Rājāsthan and the Śri Ekliṅgjī Charitable Trust, set up in the 1970s after changes in tax law by the family of erstwhile mahārāṇās of Mewār to protect their Sisodia dynastic royal temple. The Ambikā temple in Jagat recently was the site of a public deity installation ceremony, which could also be used legally to establish the site as a commodified public trust. Both deities question the politics of aesthetic taste in an increasingly global era of world heritage. The aesthetics of temple administration suggest the legal arbitration of taste as a commodity and the role of praxis and agency in the field. Renovation serves as a form of religious merit—a phenomenon witnessed during ethnographic and performance-based fieldwork but also found as early as the mid-tenth century in temple inscriptions from the Mēdapāṭa regional cohort of temples. Each of these individual legal situations leaves these two temples as the only two out of the Mēdapāṭa regional temple cohort with specific historic time periods that seem to alternately illuminate or negate the historical and legal claims being made in the twenty-first century. On the one hand, the Ekliṅgjī temple debates seem to lead the historian to the fifteenth century as a snapshot of the Sisodia clan at the apogee of its power. The royal house, on the other hand, recently claimed an unbroken bloodline back to the eighth century, and in 2012 it began to date its lineage prior to the sixth century. The Ambikā temple inscriptional record leaves a three-hundred-year silence. This sultanate period in the Chhapa and Vagada regions where the Ambikā temple is located reveals a great efflorescence of nondynastic activity from mining, to multisectarian temple patronage, to icon theft and warfare, to fleeting attempts to maintain (or even to establish) some form of dynastic or political hegemony. On the other historical side of the sultanate period is a time before these vast ruptures with the present. Origins have long been privileged as the most “authentic” moments in history, so origins become pregnant with meaning. In this study the temples’ origins are not a unique moment of truth but rather one of four major eras considered in relation to the histories we choose to construct in the present. The second half

The “Hindu” Temple in Diachronic Context

of the tenth century was a time in what is now southern Rājāsthan as well as all across northern India where the fragmentation of the Paramāra and Pratīhāra Empires gave rise to an efflorescence of small kingdoms and new dynasties yearning to legitimize their newfound status through signature architectural styles and lunar or solar divine lineages. This same millennial moment witnessed a great rise in multisectarian, populist movements toward the practice of tantric religion—esoteric to initiated practitioners and unconsciously shaping society for the uninitiated. This tantric shift had a tremendous impact on temple architecture, iconography, and the kinesthetic and philosophical implications of temple sculptural programs that have only recently begun to be studied in detail. Recent breakthroughs in the textual scholarship open new avenues of research for the study of the architectural remains. It is at the confluence of these two major millennial shifts—dynastic-political and tantric-populist—that the Mēdapāṭa regional cohort of temples was built primarily in the AD 960s and 970s. In response to new research and to the intersection of these two millennial changes, I move away from long-established dynastic categories of architecture and style to begin to experimentally map the sectarian landscape, to map east–west fluvial geographies of style (as opposed to the current northwest dynastic axis that is more commonly used), and to map traces of millennial ritual and ephemera. G E O G R A P H Y A N D D E M O G R A P H IC S I N T H E F I E L D

In the twenty-first century the fierce competition between religious use and historical preservation creates a parallel dialectic between these two sites. Increased commodification of culture makes temples, ritual, and even ideas about temples available to be bought and sold.3 The Ambikā temple—halfway between Udaipur and Dūṅgarpur—is situated in a fairly isolated area. Politicians and erstwhile nobles banded together to draw on the numinous and martial powers of the goddess during an installation ceremony in 2002, held far from any political capital in the small village of Jagat during a ceremony that was nonetheless attended by members of the BJP (Bharatiya Janata Party), government officials from the State of Rājāsthan, and members of erstwhile royalty who now live in Udaipur. True to the regal origins of the royal Ekliṅgjī temple complex, current legal debates surrounding the site suggest the continued power of archaeology for the legitimation of kingship—even after the end of monarchy as a result of Indian independence in 1947. At both sites the archaeological remains themselves become hegemonic, at the very same time in the 1990s when Guha coined the phrase “dominance without hegemony” to refer to how power was exercised by the colonial state.4 In fact, it is no longer the bourgeois colonial elite who pretend to a hegemony that would never be theirs; the buildings themselves, as commodified objects and catalysts for praxis, allow new segments of society to stage powerful counterhegemonic

Introduction

performances in and around archaeological sites. The hegemony of heritage in its unique capacity to serve as catalysts of counterhegemonic praxis simultaneously with hegemonic reification of existing power structures is not unique to these two Rājāsthani Hindu temples, nor to India or to Hinduism, nor to South Asia at large from the secular Sikh scholar’s personal experience of art in the National Museum or the multisectarian ecumenical enjoyment of ritual at religious sites.5 Indeed, the dueling hegemony of heritage to produce both hegemonic and counterhegemonic visual discourse seems to be the number one defining scopic reality of the twentyfirst century—as evidenced in the politicized administration of UNESCO and the role of monuments and icons old and new in the visual rhetoric of war. The similarities between the modern commodified lives of these two temples call into question “post”-capitalist accumulation in an era when the old clichéd dichotomy of iconoclasm and iconophilia no longer serve to define what is important about these Hindu temples. Imported largely from a colonial Protestant perspective, and employed above all at the hands of Empire, the idea of destruction or figuration as the central defining feature of an icon has all but evaporated in the South Asian context across more than one religion.6 In fact, recent scholarship suggests that an almost Catholic interest in ritual may have provided an interesting counterpoint to that perspective historiographically when we reexamine the archives of the ASI (Archaeological Survey of India).7 Furthermore, Italian Marxist Antonio Gramsci’s ideas of hegemony have served as a source of inspiration for numerous postcolonial scholars, including Edward Said and, more recently, Hamid Dabashi, among others. The idea of praxis across class and caste lines as a powerful counterpoint to colonial hegemony is borne out quite fruitfully in the tremendous scope of ritual and traditional practices (parampara in Hindi) that take place at these sites today. More and more frequently in the age of “late” capitalism, ancient sites are being put back into ritual use after lying dormant for centuries.8 The two temples that form the subject of this study have each experienced several deaths and rebirths over the past one thousand years.9 Even so, a teleological and chronological approach to their biographies would not suffice to reveal the nuanced complexities of how their histories compete in the twenty-first century and in specific points in time (the fifteenth century, the thirteenth century, and the tenth century of their origin). One history did not blindly and developmentally lead to the next; rather, in each period various actors and agents chose to ignore and to highlight the past in different ways to make political arguments about the present. Both temples have been renovated and used for ritual during periods of time and then left dormant for various reasons before being given new “lives” again. These two sites form prime examples of how the nature of the archaeological enterprise is rapidly changing both in Rājāsthan and in the greater global context today. Current uses of these two archaeological sites produce accumulations of ritual residue that visually record the interests of their respective patrons, makers, and

The “Hindu” Temple in Diachronic Context

ritual participants.10 Different groups currently lay claim to each site. A postcolonial mahārāṇā (CEO) uses his family’s Śri Ekliṅgjī temple complex to solidify the continuity of the House of Mewār in the age of the nation-state in India.11 Simultaneously, householder priests continue to lead services in the complex, while gardener caste Mali women continue to sell flower garlands at the entrance today, as is pictured in eighteenth-century frescos inside the monastery. Meanwhile, in Jagat, multiple castes actively react to the Ambikā temple today. Lower caste, habitually disenfranchised Ādivāsi groups, such as Meenas and Bhils, have slowly expanded the powerful sphere of their local goddess, Mallar Mātā, as they exercise their counterhegemonic praxis in and around the Ambikā temple today, while in daily life the temple remains largely ignored by the economic and urban elite. In the village of Jagat, it is the Ādivāsis who have reconsecrated her sister Cāmuṇḍā within the archaeological compound of the ancient Ambikā temple. This goddess—so popular in the twenty-first century that she often eclipses her sister Ambā Mātā, who is the main icon in the temple sanctum—was also incredibly popular in this region in the tenth century, when the Ambikā temple was built. At the end of this book I will focus more closely on Cāmuṇḍā and her textual and iconographic position in medieval millennial North India as an evidential response to our twenty-first-century frame. The overarching comparison between the Ambikā temple in Jagat and the Ekliṅgjī temple in Kailāśpurī illustrates how kings and nobles are not the only ones involved in praxis. Tailors and gardeners, Ādivāsi and Rajput women and men, city dwellers and countryside locals all use praxis to vie for hegemony and counterhegemonies at these ancient architectural sites. Local groups at both sites enter into dialogue with tourists, the state, and their own imagined pasts and futures through their indexical relationships to these ancient monuments. Ekliṅgjī and Jagat share an important modern tension between history and ritual. Both sites lie at the heart of competitive contests for authenticity. When R. C. Agrawala “discovered” Jagat in the mid-twentieth century, he was interested in iconography, historical analysis of inscriptions, and the preservation of fragments for a museum.12 In a footnote, Agrawala mentions that the entry pavilion (śubhamaṇḍapa) at another tenth-century Mēdapāṭa temple, the Pippalāda Mātā temple in the village of Unwās, “is completely mutilated,” which suggests that people in the village had already undertaken drastic renovations before his article was written in 1964 (fig. 0.4).13 From an architectural historian’s perspective, one cannot glean much more information than the basics of the temple program.14 Ongoing construction, whitewashing, and painting at the site attest, instead, to the continuing power of this goddess (figs. 0.5 and 0.6). One could imagine this “post”-capitalist accumulation practice as a form of theft from history, from the Archaeological Survey, from the state, and, hence, from the people. Or one could argue that these “drastic renovations” result from the use of modern materials to implement ancient forms of renovation as a form of religious duty to instill merit in the patron.15

Figure 0.4. Pippalāda Mātā temple in Unwās, c. 960, Mēdapāṭa region. © Deborah Stein.

Figure 0.5. Kṣēmaṅkarī, Pippalāda Mātā temple, back wall, Unwās. © Deborah Stein.

The “Hindu” Temple in Diachronic Context

Figure 0.6. Main icon, “Pippalāda Mātā (a folk version of Kṣēmaṅkarī?), Unwās. © Deborah Stein.

Can this praxis serve as a counterhegemonic answer to colonialism through use rather than preservation, or does this praxis indicate hegemonic Hindu nationalism with power inherited blindly from the colonial past? Furthermore, is this nationalist discourse hegemonic or counterhegemonic in relation to an increasingly privatized state resulting in increasingly privatized archaeological sites? The “complete mutilation” of the temple in Unwās lamented by R. C. Agrawala, the foremost indigenous scholar of southern Rājāsthani archaeological sites in 1964, begs new questions in the twenty-first century—an era imagined by some, in exile or in limbo, as a time when some global scholars may define their identity as “amphibious,” neither Western nor non-Western but entirely more complicated than that outdated binary. “A post-nativist amphibian intellectual,” according to Dabashi, “has his or her roots in the material reality that embraces both ‘home’ and ‘exile,’ a division that has in effect caused the initial intellectual labor migration.”16 From these new global perspectives, how does the hegemony of heritage reveal the specifics of “post”-capitalist accumulation of icons, buildings, and practices that radiate out from these catalysts? Miles Glendinning, in his chapter “Heritage in the Age of Globalization: Post1989,” addresses the “instability” of the concept of “authenticity” in terms of the

Introduction

Figure 0.7. Old and new architecture and repairs comingle at the Śri Ekliṅgjī temple complex, Kailāśpurī, Rajasthan. Photo by author, 2002. © Deborah Stein.

apogee of a conflict between the global and the local resulting in the 1994 Nara Convention in Japan. According to Glendinning, “Definitions of authenticity, after all, had underpinned all doctrinal definitions from the 1964 Charter of Venice through to the outstanding universal values and operational guidelines of the World Heritage Convention.”17 The clash of local heritage management in Japan and increasingly global attempts to standardize heritage conservation in the early 1990s highlighted the problems with policing authenticity and taste resulting in “the radically new field of intangible heritage.” In a backlash against what came before, “the value attributed to any heritage object began to depend entirely on the present-day host culture.” Agrawala’s bemoaning of a loss of authentic architecture in 1964 may have been celebrated by others in the living buildings of the 1990s, whereas in the twenty-first century it would be interesting to find a way these two visions are not mutually exclusive so that buildings’ histories are not erased and the buildings are able to serve multiple uses, including both local archaeological and religious ritual uses in the field. Recent studies have suggested the antiquity of tracts of land largely beyond state control in the region of Chhapa, where the Ambikā temple in Jagat is found.18 Historically, the Śri Ekliṅgjī complex (fig. 0.7) lay in the heart of ancient Mēdapāṭa, now known as Mewār (fig. 0.8). The Ambikā temple, however, has alternately been ruled from Mewār or Vagada, when it was not in a vacuum of power. Located in the village of Jagat, the Ambikā temple (fig. 0.9) is two hours’ drive south of

Figure 0.8. Map of sites in modern Mewār and Dūṅgarpur—ancient territories of Mēdapāṭa (north, near modern capital of Udaipur), Chhapa (middle Zawar [Jāwar]/Āaṭ/Jagat), Vagada (south, near modern capital of Dūṅgarpur), and Uparamāla (northeast, from Chittauḍgaḍh to Menāl and Bijoliā).

Introduction

Figure 0.9. Ambikā temple, c. 960, Jagat. © Deborah Stein.

the modern capital of Mewār, Udaipur. Jagat lies in a hilly region once known as Chhapa. In the thirteenth century the Guhila ruler Sāmanta Singh made Chhapa a part of the Vagada Empire when he left the Mewār throne to his brother to become the mahārawal of Vagada.19 These two Guhila royal houses still exist today—as do intact communities and villages of Bhils, Meenas, and Rājput descendants near each of the temples in this study. The continuity and rupture found in the material, iconographic, stylistic, inscriptional, and kinesthetic architectural remains at these sites today help to clarify the hegemony of heritage in the twenty-first century. In discussing the nineteenth century in Britain, Hobsbawm reminds us that “ ‘traditions’ which appear or claim to be old are often quite recent in origin and sometimes invented.”20 Much of what we see in the field in Rājāsthan today harks back to the Victorian era as well, when the Royal Titles Act (passed on April 27, 1876) made Victoria the empress inheritor of Mughal power.21 Rather than attempt to specifically date each tradition we encounter at archaeological sites in southern Rājāsthan in the twenty-first century, it seems more fruitful to keep in mind a broader definition of the “invention of tradition,” defined by Hobsbawm as “a set of practices, normally governed by overtly or tacitly accepted rules and of a ritual or symbolic nature, which seek to inculcate certain values and norms of behavior by repetition, which automatically implies continuity with the past. In fact, where possible, they normally attempt to establish continuity with a suitable historic past.”22

The “Hindu” Temple in Diachronic Context

Understanding the role of visual material sheds light on some of the violence surrounding monuments in the postcolonial era. The existing art historical literature on the temples of Mēdapāṭa falls largely on the side of architectural preservation, whereas religious studies scholars privilege ritual practice at ancient sites. Historically, an interest in visual material has led art historians to consider modern renovations as a form of destruction of the archaeological record rather than as an addition to a series of changes. Some are tempted to view practice at archaeological sites as a direct, unchanged continuation of past practices. One of the most cogent reasons to resist the rhetoric of continuity is that religious nationalism often gives rise to communal violence. Religious buildings and icons incite crowds to riot, to burn people alive.23 Moving beyond artistic intention does not negate the importance of the moment of making. A temple is not built by a single mind for a sole purpose; it is a collage of patron’s interests, guilds’ aesthetic habits and choices, and the diverse body of people who consecrate and use the site. Antiquity makes a site better adapted to modern religious activity. Continuity may exist in the same cult using a site over a thousand years, yet site use changes—a reflection of the concerns of the moment. Whether we legitimize the collection of a cult’s expression as a continual chain or as a discontinuous record of rupture is largely political. The physicality of permanent stone and ephemeral offerings tells a story that would be lost in inscriptions and texts alone. Jules Prown has defined material culture as “the study through artifacts of the beliefs—values, ideas, attitudes, and assumptions—of a particular community or society at a given time.”24 One of the most powerful possibilities of material culture lies in the attention to what is not intentional. Prown suggests we must extend our inquiry to include objects beyond art and icon. If we use the term “artifact”—not in the nineteenth-century sense of depreciation of the art of the “other” but rather as permission to include “butter” and “vermilion” alongside “stone” itself—conventional art historical cornerstones such as style can illuminate more when combined with a deeper understanding of context. Man-made, and in this study I would argue woman-modified, objects “reflect, consciously or unconsciously, directly or indirectly, the beliefs of individuals who made, commissioned, purchased, or used them, and by extension the beliefs of the larger society to which they belonged.”25 Material culture offers art historians a chance to expand the realm of objects within reach. Ethnographic practice no longer serves as a colonial key to oppressing the other but rather as a potentially liberating experiment in highlighting counterhegemonic praxis on the part of people whose agency was almost always ignored intersectionally owing to gender and class as well as the counterhegemonic praxis of others who experience themselves as economically empowered local stewards of culture.26 This study examines the sacred tree alongside temple architecture, the photograph at the village bus stand alongside the original black schist goddess up the hill, and the nine kilos of flour

Introduction

Figure 0.10. Chapati dough ornaments, 2002, Jagat. © Deborah Stein.

participation allowed me to experience the ritual through the senses in a way that discussion after the fact would preclude. This embodied approach to fieldwork created a unique source for dialogue with those who habitually performed the rites. Unlike male textual scholars, whose participation often becomes enmeshed with altering the liturgy (or even sponsoring the ritual in the case of Frits Staal),34 my dual status as a young female and as a scholar allowed access to the worlds of children’s rooftops and women’s kitchens, as well as men’s more public, political, and religious spheres. Art history has conventionally dealt with the past. But many have begun to engage the study of ancient Indian art with the present in innovative ways, although few have questioned the impact of history and the present on the future of Indian patrimony. Art historians’ interest in visual material makes them particularly concerned with the future of monuments and artifacts outside the protection of archives, museums, and libraries. UNESCO increasingly interfaces with local heritage groups in an effort to coordinate relevant branches of governments.35 The urgency of these projects stems from unregulated development, theft, looting, and exponential increases in tourism. Even though the discipline of art history and international organizations such as UNESCO understand their mission as preservation, living temples complicate matters. To preserve history is, in many ways, to kill it. This book examines the many ways in which different Indian people continue to use ancient temples to construct

The “Hindu” Temple in Diachronic Context

Figure 0.11. Painting the sanctum gold, Ambikā temple, May 2002, Jagat. © Deborah Stein.

their own histories. The result is not always the preservation of a building. In fact, at times, buildings are defaced when locals decide to refresh a deity’s home. The book explores this tension between preservation and use. The theoretical premises that inform any choice regarding preservation, conservation, renovation, or use directly affect the visual qualities of an ancient site in the present and in the future. If use is privileged to respect living icons in living monuments, then a tenth-century sanctum and sculpted doorframe may be painted metallic gold as part of an installation ceremony (fig. 0.11).

chapter 1

Figure 1.2. Saraswatī, Lakulīśa temple, Ekliṅgjī. © Deborah Stein.

An image of Saraswatī, the goddess of learning and the arts, flanks the entrance of the Lakulīśa temple to indicate a center of learning and may well provide the only extant example of a sculptural version of Saraswatī in situ on a tenthcentury building known to have been associated with the acquisition of knowledge (fig. 1.2). Saraswatī rarely graced sculptural programs of tenth-century temples. Even unusual evidence, such as the inscription at the base of a female figure, does not provide a similar visual example of Saraswatī. Previously known as King Bhoja’s Saraswatī from Dhar, a sculpture now housed in the British Museum bears an inscription linking it to a royally sponsored university.9 Saraswatī’s association with learning and the arts is well known, but the inscribed sculpture provides evidence only for her association with sites of learning and not for her visual representation within an architectural context.10 She holds in her hand an elephant goad, a typical attribute of this goddess, who prods her disciples toward evergreater acquisition of knowledge, an item that is even mentioned in guru-disciple initiation rites in Abhinavagupta’s writings on tantra.11 The paucity of sculptural form aside from this single goddess may reflect the audiences who used the Lakulīśa temple. A monastic audience would not have needed visual support to guide the body through space during circumambulation. Alternatively, this temple may have been used for guru darśan, which did not necessitate circumambulation or any form of practice that would require the animation of the central icon performed with a prayer manual.12 The building

Temple as Geographic Marker

Figure 1.3. Śivēśvara temple, c. 950s–70s, Ekliṅgjī. © Deborah Stein.

may have served as a monastic meeting space or lecture hall. Given the uncertain date of the main icon, it is even possible a living guru presided over some type of assembly. The inscription offers further evidence of the philosophical debates that may have taken place inside.13 The visual pairing of the inscription with a statue of Saraswatī may indicate a learned ascetic audience for this shrine. Those who used this building may not have needed elaborate emanations of deities to guide their already sophisticated practice. This temple may have served a set of Pāśupata ascetics in residence at Ekliṅgjī.14 In contrast to the reserved simplicity of the Lakulīśa temple, the Śivēśvara temple in the Ekliṅgjī temple complex and the Takṣakēśvara temple found in the gorge nearby are much more typical of Mēdapāṭa temples in the second half of the tenth century CE (fig. 1.3). Guardians protect the corners of both buildings and surasundarī figures twist to either side of wall niches. Whereas the Viṣṇu temple has vyālas, the Śiva temple does not. The Śivēśvara temple and the Takṣakēśvara temple are undated but roughly contemporaneous with the Lakulīśa temple. The ornament and the texture of the temple wall are neither sectarian nor chronological. Temples in a spare style and in a more complex style coexist in tenth-century Mēdapāṭa. Although the carving of surasundarīs on the Takṣakēśvara temple is better executed, both temples evidence certain stylistic features common to the Mēdapāṭa School, such as the triangular forms above the window (simhakarna). Although the Lakulīśa temple is relatively unornamented, some decorative features such as the window screen (candravalokana) are typical of the Mēdapāṭa School. The quality of masonry of the Lakulīśa temple, together with the complex, multifigured formula of the Takṣakēśvara temple program, were preceded by an exquisite example of Mēdapāṭa architecture.

chapter 1

Figure 1.4. Sūrya on his seven-horse chariot, c. 950–75, Ṭūṣa. © Deborah Stein.

One of the earliest works of tenth-century Mahā-Gurjara architecture in Mēdapāṭa is the Śaivaite Sūrya shrine at Ṭūṣa built, in all likelihood, under the reign of Bhartṛpaṭṭa in the second quarter of the tenth century.15 Like the Ambikā temple at Jagat, the Sūrya temple at Ṭūṣa repeats three manifestations of the same deity adorning the exterior walls of the garbhagṛha in each of the niches found on the three bhadra wall protrusions (fig. 1.4). In similar fashion Sūrya is flanked by two surasundarīs who are flanked by vyālas. The dikpālas guard the four corners of the shrine. In contrast with the Takṣakēśvara shrine, which shares this basic iconographic structure, the sculpture complements the protrusions and recesses in the temple wall. The presence of couples above the heads of the figures is a shared feature with the Ambikā temple, although here they sit rather awkwardly above the full figures without any architectural articulation, whereas at Jagat the couples sit on shelves within their own defined space. The sculpture of Mēdapāṭa is known for a balance between rounded flesh and crisp, finely linear carving, as is exemplified by the surasundarī in figure 1.5. The tribhanga (three-bend) pose makes it seem as if her girdle is swaying with her movement, and the articulation of her stomach is unsurpassed in this region. The sculptors have captured the way her flesh gently protrudes over the top of her waistline by indicating the beginning of fabric, with only a delicately carved line, and the flattening of flesh above her belt. This way of representing a standard set of female forms atop lotuses exemplifies the regional style of Mēdapāṭa

Figure 1.5 Surasundarī, Ṭūṣa. © Deborah Stein.

chapter 1

which may codify much of what was being innovated through praxis, art, and architecture along the Som River, north of the Narmada River in millennial India.28 At Jagat we do not have a direct linkage of text and image, in either specific time or place, but rather kinesthetic architectural traces of ritual, use, and praxis that suggest a general “tantric” mode of worship using the female form as an alternate model for arousal and disgust in the search for detachment from desire. Writings by Phyllis Granoff and others have emphasized the plurality of subjective positions, even quoting early medieval writings that elucidate and imagine how no fewer than six different demographics would have understood the sexuality of a temple wall at a site like the infamous Kaṇḍāriyā Mahādeva temple in Khajurāho.29 At Jagat the bidirectional circumambulation of the temple program suggests two narratives that would have been available in similar form to men and women alike and of all ages and social stations in varied stages of enjoyment, standard worship, and more esoteric and philosophical though equally populist tantric worship (fig. 1.6). Art historically, the Ambikā temple at Jagat has long captivated the imagination because of the beautiful, largely intact carvings dedicated to the goddess Durgā-Mahiṣāsuramardinī, who vanquished the buffalo demon, Mahiṣā, in cosmic battle when the male gods could no longer restore order.30 The Purāṇic account of the goddess lends itself most aptly to narrative iconography such as the famous frieze at Mamallapuram in Tamil Nadu. The goddess Durgā-Mahiṣāsuramardinī is named for her sacrificial act, when Durgā kills the buffalo demon named Mahiṣā. Durgā-Mahiṣāsuramardinī wielded the borrowed weapons of the male gods to confront demons on the battlefield, issuing new warriors from a light in her forehead. The most famous of the goddesses to emanate from Durgā’s forehead is none other than Kālī, whose long tongue is known to stretch the length of a miniature painting lapping up the blood of demons quelled in battle before the drops could sprout into new warriors. This story, the Devī Māhātmya, the stuff of the oral legends told across Rājāsthan—that is, until the very recent replacement of pan-Indian Gujarati “garbha” dancing—was told in seven hundred verses every day during the autumnal festival of Navratri, the nine nights of the goddess, culminating in the tenth night of Dusserah celebrated across India. But the oral legend, the Purāṇic texts, and the narrative festival traditions seem to have little to do with the highly iconic, monoscenic, or even nonnarrative way in which the artists, masons, and patrons chose to represent the goddess Ambikā at Jagat. The tenth-century temple in Mēdapāṭa is often characterized by the way the icon in the inner sanctum emanates powerfully outward on architectural axes in the four directions. These axial relationships usually take the form of three sculptural icons on the three outer walls of the sanctum, with the fourth direction being the direct gaze of the deity on the worshipper inside the temple. A bhadra is an Indian architectural term for the protrusion of the temple wall that aligns axially with the icon in the inner sanctum of the temple. There are always three bhadras that face three of the four directions and allow the essence of the icon to emanate

Figure 1.6. Door to sanctum on each side allows the bidirectional circumambulation, AmbikÄ temple, c. 960, Jagat. ÂŠ Deborah Stein.

chapter 1

Figure 1.7. Durgā zoomorphic, side 1 (south), Ambikā temple, Jagat. © Deborah Stein.

Figure 1.8. Durgā zoomorphic/anthropomorphic, side 2 (east/back), Ambikā temple, Jagat. © Deborah Stein.

outward in every direction—the fourth direction, of course, is the interior opening to the garbhagṛha sanctum, where the icon gazes out directly onto the viewer during worship. Each bhadra, then, has a niche on the exterior of the three outer walls of the temple sanctum. Exactly on axis with the inner icon, three of the same or three different sculptural deities represent the three directions in which the inner god of the sanctum is exuding power. Each manifestation geomantically reflects the directionality. For example, particularly fierce deities associated with death often face north. These axial protrusions and emanations form a sequence as the viewer circumambulates in the normal clockwise direction, or counterclockwise in an esoteric twist on the sequence that serves to relativize the passage of time or the inherent order of things. The Ambikā temple in Jagat, in particular, has two doorways on either side of the inner sanctum that serve as thresholds between the exterior bhadra niches with their emanations of the inner sanctum. The viewer can then take darśan from the main deity before choosing to circumambulate either direction around the temple sanctum exterior before returning up a few steps back inside on axis with

Temple as Geographic Marker

Figure 1.9. Durgā anthropomorphic, side 3 (north), Ambikā temple, Jagat. © Deborah Stein.

the main icon. On the three walls of the temple sanctum the goddess plunges her trident into the flank of the buffalo demon, and three different times he seems to die in pain on each wall as the devotee circumambulates the structure.31 If one faces the temple in normal circumambulatory order, with one’s right side to the wall in clockwise fashion, three forms unfold on the southern, western, and northern walls. In the first bhadra niche, the zoomorphic form is quelled (fig. 1.7). In the second bhadra niche, the spirit in human forms begins to rise from the buffalo’s sacrificed neck (fig. 1.8). And in the last bhadra niche, the goddess holds the fully human form of the spirit tightly by the hair (fig. 1.9). One could move from animal sacrifice to the vanquished spirit or, alternatively, in a left-handed tantric manner, from the vanquished spirit toward the sacrificed animal form. Either way, the double doorways suggest that the main icon was the fourth and main culminating component in either circumambulatory order. Left-handed and right-handed coexisting circumambulatory paths do suggest tantra, but who exactly was in the main sanctum? Recent worship and thefts have clouded this picture, as have competing archaeological photographs, but it seems that only two viable options remain.32 Dhaky argues the main icon was Kṣēmaṅkarī, which would make sense, given the location of this mantric goddess above the entrance to the temple, where she stands atop two lions in classical iconography.33 A second possibility lies in the emaciated version of the goddess Durgā slaying the buffalo demon that (as of 2009) lay cast aside to the right of the main icon in the

chapter 1

Figure 1.10. An icon cast aside inside the sanctum displays an emaciated form of Durgā in the act of killing the buffalo demon. Ambikā temple, quartzite, c. 960, Jagat. © Deborah Stein.

inner sanctum (fig. 1.10). Either possibility suggests a tantric interpretation of the fourth and most important element in circumambulation. Kṣēmaṅkarī is a beatific version of the goddess, who would encourage a devotee to focus on the seed syllable, or bīja mantra, “Kṣa,” associated with the entire “garland of letters,” as Arthur Avalon would have put it.34 The goddess ferociously and independently sacrifices the evil demon and then is revealed as this pleasant version of Śiva’s wife, Pārvatī, a calm evocation of the power of mantra, or speech incarnate, as Śiva’s śakti, or his cosmic energy. The second icon, still in situ but cast aside, suggests instead that the voluptuous, sensual body of the sacrificing independent goddess, repeated three times on the exterior, becomes emaciated in the inner sanctum. The mounting arousal of the sacrificial moment by the libidinal actor then culminates in the skeletal confrontation of time itself—either in the withering of one’s own female body, the withering of the object of one’s desire for a female body, or the withering of the object of desire—or of desire itself. Whether delightfully mantric, through the personification of speech as Kṣēmaṅkarī, or whether gruesomely corporeal through a reiteration of destiny and death incarnate, either goddess would have cemented the bidirectional tantric circumambulation found at Jagat. Either way, the cycle of samsara is reiterated forward or backward in the kinesthetic architecture of the temple sanctum flanked by two doorways.

Temple as Geographic Marker

Could a temple as unusual as the Ambikā temple in Jagat really exist in isolation as so incredibly unique—a long-lost material key to illustrate the textual scholars’ interpretations of tantric texts? The tenth-century temples in Hita, Āaṭ, and Bāḍolī suggest otherwise. The artists of the Ambikā temple in Jagat worked in their lifetimes on a temple dedicated to Nateśa directly east of Jagat at the previously unpublished site of Hita. Not only do we have a physical trace of a time and place on the map, but we also have an intact iconographic program at Hita that integrates female goddess sculptures into a Śaivaite program—sculptures identical to those found at Jagat, less than sixty kilometers west of Hita. Sculptural fragments at Hita include a Gaṇēśa (very similar in style to the Gaṇēśa at Jagat) and exterior wall sculptures of the female form where the jewelry, flesh, position, finish, carving, and stone seem identical to what is found at Jagat—more than any other two tenth-century sculptures I have found across modern-day southern Rājāsthan. Dancing with his skull staff, Śiva dominates the back wall of the sanctum, which suggests that the tenth-century temple in Hita was devoted to a tantric form of the god in mid-dissolution as Nateśa. Like the Śaiva-Siddhanta lord of dance, Nataraja, found so commonly among Chola bronzes, his northern counterpart, Nateśa, was quite common in Madhya Pradesh along the Narmada River and its tributaries at the turn of the millennium.35 Nateśa also had a long history of association with mātṛkās (mother goddesses) in Rājāsthan.36 This Nateśa temple thus links to the Ambikā temple in Jagat with mātṛkā worship and links to more eastern sites with established centers of tantric worship during the second half of the tenth century. Remains from the Gupta period, such as the mātṛkā sculpture of Aindrī in the Udaipur archaeological museum and a tiny five-faced Pāśupata liṅgaṃ found by the thakur’s (local government official) family and kept in their compound of Rawala in Jagat, suggest that Śiva/Śakti mātṛkā worship and visual icons of Sadāśiva liṅga at the site of the Ambikā temple existed at least four hundred or more years before the tantric temple was built (before similar icons were made in the eighth century at Kalyanpur, and before the infamous one in Ahar— which Lyons argues may have served as a model for Śri Ekliṅgjī himself). The walls of the smaller and more sparsely decorated temple at Hita do not include as many auxiliary semidivine figures as are found at Jagat, but the basic program includes a series of figures on the sanctum walls. The maṇḍapa is connected to the sanctum, and all is severely whitewashed and restored, so no evidence of a bidirectional circumambulatory path remains, except perhaps in the proportions that would leave enough room for a small doorway on either side of the main sanctum. At Hita, as at the better-known goddess temple of Unwās, we are left to work primarily from sculpture left in situ rather than from original intact architectural space. The circumambulatory order at Hita begins with a ferocious form of Śiva (fig. 1.11). Could it be Andhakāntaka, found with Nateśa and Cāmuṇḍā at Menāl and known more famously from Ellora and Elephanta?

Figure 1.11. Ferocious Śiva

Figure 1.12. Nateśa, c. 955–75, stone, Hita.

(Andhakāntaka?), c. 955–75, stone, Hita. © Deborah Stein.

Figure 1.13. Cāmuṇḍā, Nateśa temple, stone, c. 955–75, Hita. © Deborah Stein.

Figure 1.14. Exquisite śekharī architecture, Nateśa temple, stone, c. 955–75, Hita. © Deborah Stein.

Temple as Geographic Marker

Andhakāntaka steps on a demon and stabs another with a trident. Perhaps because of his fangs, he is often misunderstood as Bhairava. This figure on the south side of the temple in Hita seems to swing a drum above his head. The main position on the back wall of the temple faces east. There, in a move similar again to Menāl, Bijoliā, and Bāḍolī in Uparamāla, Nateśa dances his cosmic dance, bending in an almost impossibly limber way (fig. 1.12). His dance is met on the north wall by none other than Cāmuṇḍā (fig. 1.13). Before laying eyes on the ferocious emaciated female form, the circumambulator is confronted with one of the sharpest, most geometrically clean, small śekharī temples in India (fig. 1.14). Located southeast of Chittauḍgaḍh on the way to Jagat via Bambora, the tripartite iconography of this temple recalls temples of Menāl, Bijoliā, and Bāḍolī in Uparamāla. The last image in circumambulation so resembles the sculptural carving style of Jagat that one could imagine only three scenarios. The first is that one artist carved the same sculptures for both temples. A second scenario is that one guild was responsible for each—though they seem even too similar for that. The least likely is that somehow a piece of sculpture got carried from one site to another and reinstalled. Given that there is no record of this temple in the Archaeological Survey of India records, this is unlikely. The concluding sculpture at Hita (fig. 1.15) is the one that bears such a remarkable affinity to sculpture at Jagat, such as the Kṣēmaṅkarī in Dhaky’s photograph.37 Delicate rows of beaded necklaces and girdles decorate the finely chiseled features of voluptuous, fleshy bodies, filled with life-breath yet not overinflated on prana nor overly “medieval” in their columnar elongation. The sensual texture of these bodies makes stone seem like a warm, living place where one could rest one’s head, the jewelry delicately jingling as the carving synesthetically invades one’s ears. The foundations of other temples suggest that Hita may well have been a larger center than imagined today. The location of Hita pushes the spread of the Mēdapāṭa cohort south and east, away from Ghāṅerāo and the Guhila strongholds around Nāgadā/Ekliṅgjī. Jagat’s closest stylistic companion now lies closer to Uparamāla territory and yet significantly south of Bijoliā, Bāḍolī, and other tenth-century sites found on the east–west axis along the Banas River. Pilgrims and travelers may well have stopped in Hita before overnighting in Bambora, where a palace was built a few hundred years later north of what became known as Jaisamand Lake. Bambora, a subregal noble retreat, is currently in the female jagīr (dowry) of the nobles/family of Jodhpur, who married into the family from Dūṅgarpur. Regardless of modern marital property rights and erstwhile kingships, this bustling little town is the closest city east of Jagat and Āaṭ. Just west of Jagat and Āaṭ is the historical zinc mining center of Jāwar. Together, these places form a route from Hita, to Bambora, to Jagat, to Āaṭ, to Jāwar—a route through a region that saw fluctuating and alternating pockets of time with no polity, forming by the sultanate period the “gray areas” of the map in northwestern India.

Figure 1.15. Durgā-Mahiṣāsuramardinī in the identical artistic style of Jagat, c. 960, Hita. © Deborah Stein.

Temple as Geographic Marker

Figure 1.16. Guru and disciple on the temple wall, stone, c. twelfth to thirteen century, Gamari. © Deborah Stein.

In the tenth century, when Hita and Jagat and Āaṭ were small temple towns, they lay on a route along the Som, north of the Narmada, where multiple estuaries connected them back to the great maṭhas of central India and down the river, through the birthplace of Lakulīśa, founder of Pāśupata ascetic Śaivism. Could Pāśupatas, goddess worshippers, and others have traveled from monastery to monastery through this back region of the Narmada between central India and the port of Khambhat—stopping along the way for tantric mediation and worship at three or more interconnected temples in Hita, Jagat, and Āaṭ? Ojha recorded an inscription from Āaṭ in his Dungarpur Rajya ka Itihas, and the remains found there suggest a thriving Pāśupata center with possible Jain remains and radically explicit sexual imagery, better-sculpted and yet linked stylistically through similar representation of pectoral muscles in stone to the bestiality and other motifs found farther south at sites such as Gamari. With uncertain dating, many of the architectural remains of Vagada have remained outside of art historians’ purview. Sites such as Gamari are fairly remote to modern visitors and are not nearly as impressive stylistically as the core of the Mēdapāṭa cohort or even the Paramāra remains at Arthuna. Hastily and simply carved sculptures on the exterior of the temple in Gamari, for example, suggest artistic trends in a more tenuous period of political upheaval in the twelfth and thirteenth centuries (fig. 1.16). Despite a later date, the content and meaning of the carvings suggest an interesting waypoint between sultanate-era Āaṭ, as a functioning tantric center off the beaten path, and a site much closer to the Mahi River as it gushes along the borders of the Malwa plateau and into the Narmada River

chapter 1

Figure 1.17. Bestiality on the temple wall, stone, c. twelfth to thirteenth century, Gamari. © Deborah Stein.

toward Gujarat.38 Whereas the quality of the art element of Gamari would not normally catch the attention of art and architectural historians, the scenes of bestiality (fig. 1.17) and other explicit forms of sexual embrace could be used to trace the religious record in stone—a tenth-century tantric inheritance from Āaṭ and Jagat still important enough in twelfth- or thirteenth-century Vagada to merit the expense and trouble of carving the stone and publicly erecting the stone architecture even though it was unlikely to be sponsored by any dynastic power. Mēdapāṭa used to include Jagat, Ṭūṣa, Nāgadā, Ekliṅgjī, Īswāl, Unwās, and Ghāṅerāo. Uparamāla used to include Bijoliā, Bāḍolī, Menāl, and Chandrabhaga. These regional appellations are found in contemporaneous inscriptions, which suggests that they served as place names for people in those regions at the time. But if we use fluvial patterns to imagine networks of exchange rather than autonomous dynastic regions with fixed borders, we can begin to remap the sectarian landscape over a large region during an increase in production of tantric architecture in the tenth century. Along the Banas River from c. 950 to 1200, an east–west axis reveals active sites of Pāśupata Śaivism where temples are not the only stone markers of ritual. Chandrabhaga, Bāḍolī, Bijoliā, Menāl, Chittauḍgaḍh, Īswāl, Unwās, Ekliṅgjī/ Nāgadā, and Ṭūṣa lie along this route. These sites leave traces of Śaiva-Śakti worship specifically of Nateśa and Cāmuṇḍā. Cāmuṇḍā was once a part of sets of sixtyfour or eighty-one yoginīs or was found as one of the seven mother goddesses. Michael Meister’s work in this very region suggests that the mother goddesses often included Nateśa, a male form of dancing Śiva, as the eighth figure.39 We see

Temple as Geographic Marker

this in the mother goddess shrine at Bāḍolī, for example. From these mother goddess sets, so popular in the Gupta period throughout northern India, Cāmuṇḍā and Nateśa seem to arise as a pair or on their own around 950 CE in Uparamāla, Mēdapāṭa, Chhapa, and Vagada—a large contiguous region north of the Malwa plateau, situated more broadly between the Gangetic Plains and Gujarat’s ports, which open onto the Persian Gulf. Along the Banas and its smaller branch rivers at sites often protected in natural enclaves formed by mountain ranges such as the Vindhyas, or by natural plateaus at Chittauḍgaḍh and seasonal waterfalls at Menāl, ithyphallic club-bearing statues of Lakulīśa held forth (Ekliṅgjī, Menāl, Bāḍolī, and Bijoliā), Nateśa danced his cosmic metaphors of dissolution (Menāl, Bāḍolī, Bijoliā, and Hita), and Cāmuṇḍā held forth in ubiquitous groups of seven (or eight, including Nateśa) mothers and on her own (Chandrabhaga, Bāḍolī, Bijoliā, Menāl, Chittauḍgaḍh [Kālikā temple], Unwās, and Nāgadā). The power and immediacy of tantric worship through tantra, mantra, and yantra was spreading in millennial India, and the sculpture, architecture, and geography leave interesting traces of specific cults along a geographical grid largely independent of dynastic affiliation. At Bāḍolī several fragments reinstalled on-site and not in their original locations, preclude the type of programmatic analysis available for Unwās and Jagat; however, tantalizing iconographic parallels emerge at this tenth-century site in Uparamāla. For example, the Sadāśiva liṅgaṃ found at Jagat in the thakur’s house dated to the Gupta period, the Sadāśiva liṅgaṃ found at Kalyanpur, the Sadāśiva liṅgaṃ installed at Ahar, the Sadāśiva head reinstalled at Ekliṅgjī, and the Sadāśiva head found in the kuṇḍa (tank) at Bāḍolī all manifest four faces, with the fifth, Iśvara, pointing upward. In Uparamāla, too, whether under the Pratīhāras, or in Mēdapāṭa under the Guhilas, or in Chhapa with no reference to a dynasty at all, this form was prevalent from the Gupta period into the fifteenth century and beyond in this region.40 Continuing east to west along the Banas River, near Ekliṅgjī at Nāgadā, we find a stone mandala similar to the ones we find made of legumes and grain laid out at Ekliṅgjī and Jagat to this day (fig. 1.18). Beneath this image, on either side of a sacrificial fire, devotees in stone ladle ghee onto the fire and utter “swaha.” Ritual does leave a record in stone, but this does not make Indian history, art history, or religion timeless. To the contrary, specific continuities and ruptures can be historicized visually in particularly accurate ways owing to the stone record of temples. For example, the mandala made of legumes in 2002 for the goddess installation celebration in Jagat (fig. 1.19) is not an exact replica of the mandala of multiplying figural emanations found at Nāgadā, even though both may speak to a similar mental process of visualization. Temples are geographic and temporal markers of what was or was not happening ritually, historically, religiously, and inscriptionally in a specific time and place. To moderns, the temple serves as a buffer against ahistorical orientalisms. Each period reveals specific ideas different from the next,

chapter 1

Figure 1.18. Mandala in stone, with devotees ladling ghee over a fire, c. tenth or eleventh century, Nāgadā. © Deborah Stein.

yet visually, ritual links, such as the stone and grain mandalas, sometimes remain organic echoes of each other over time. Built around a century later than the Pippalāda Mātā temple, fewer than one hundred kilometers away, the fifteen shrines surrounding the two larger Sās-Bahu temples at Nāgadā are marked by similarly sparse walls but on a more prolific scale. The Sās temple has relatively few sculptures in its program. Two sculptural couples frame the entrance, Śiva/Pārvatī and a broken couple. On the exterior sides of the pavilion Viṣṇu holds the never-ending knot; another representation of Viṣṇu, Śiva, and Brahma appears on either side of the exterior of the inner sanctum; and Viṣṇu

Temple as Geographic Marker

Figure 1.19. Mandala made of legumes in Jagat, May 2002, temporary maṇḍapa during goddess installation. © Deborah Stein.

predictably graces the back niche. This temple has no auxiliary figures whatsoever. Although it was built after the Takṣakēśvara shrine in the gorge at Ekliṅgjī, after the Datoreśvara Mahadev temple at Śobhagpura, and after the Sūrya shrine at Ṭūṣa, this temple has a program similar to the Pippalāda Mātā temple in Unwās. Rather than displaying a rhythmic entourage of deities and attendants, the shrine invokes only the most basic deities. The exterior figures serve to show a correspondence with the interior deities and with the main deity’s emanation on the back wall of the inner sanctum. In keeping with the sparse program of the Pippalāda Mātā temple at Unwās, the Sās-Bahu temple at Nāgadā displays an even more reduced program, with only three exterior figures on the walls of the garbhagṛha: Śiva with a fruit, Viṣṇu, and Brahma. Most of the shrines surrounding these temples share this abbreviated form of iconography. Of note is a small Devī shrine at the back of the complex at Nāgadā. Durgā-Mahiṣāsuramardinī is sculpted on the back wall of the shrine. Mahesvari, with her skull staff, and Cāmuṇḍā accompany her on the two side walls. This Śaiva-Śakti shrine has no main icon, but the Durgā-Mahiṣāsuramardinī is represented with the demon’s human leg protruding out of the neck of his decapitated buffalo form (fig. 1.20). A second goddess shrine close to the entrance gate by the lake is dedicated to Saraswatī. This shrine still has a statue of Pārvatī inside. A Jain complex also suggests that Nāgadā—located adjacent to Ekliṅgjī where the

chapter 1

Figure 1.20. Durgā-Mahiṣāsuramardinī, c. tenth or eleventh century, stone, Nāgadā. © Deborah Stein.

famous 971 CE debate among Jains, Buddhists, and Pāśupata Śaivaites was held— was already a multisectarian center for trade in the eleventh century.41 Several features of the temple complex plan at Nāgadā resonate with other tenth-century sites farther south. At Jagat, shrines also tightly overlap to the point of precluding circumambulation. There the praṇālā shrine would have originally been freestanding, as is evident from the later addition of the stone spout awkwardly joined to the shrine through the middle of a piece of sculpture (fig. 1.21). The plan of the site on a raised platform with several closely spaced shrines is similar to the Śaiva complex at Āaṭ, which also has an elaborate entrance gate similar in

Figure 1.21. Jagat praṇālā, demon’s leg protruding out of the neck of the decapitated form. © Deborah Stein.

chapter 1

Figure 1.22. Toraṇa gate, Nāgadā. © Deborah Stein.

style if not in iconographic content to the gate at Nāgadā (fig. 1.22). Two inscriptions from the eleventh century attest to the growing power of the Guhilas near Nāgadā and Ahar in the Mēdapāṭa region.42 An inscription from Nāgadā dated to 1026 CE describes the town as “a renowned seat of scholars well versed in the Vedas.”43 Unfortunately, the fragmentary inscription leaves the name of the ruler a subject of speculation.44 Viṣṇu is invoked by the name of Puruśottama in this inscription. The inscription at this grand site of Viṣṇu temples terminates with the symbol of the diamond lotus. Whether this motif is considered tantric or merely decorative, it was certainly multisectarian in early eleventh-century Mewār. These traces of ritual and the record in stone suggest that a diachronic approach cannot reveal timeless religious beliefs but rather that elements of current worship have landed vestigially in the twenty-first century and that—together with the inscriptional, architectural, iconographic, and stylistic evidence—we can begin to map specific responses to periodic shifts in religious practice along a grid of specific forms of worship and belief. To this end, if we follow the less-traveled waters of the Som and Mahi Rivers, we find a new set of sites, including two previously unknown to scholars. Along the Som River we find a southern group of temple sites in a region that was historically and to this day largely outside of dynastic reaches. Bhils, Meenas, Lohars, and other Ādivāsi people still form the majority populations there. Careful inscriptional work can even trace exact points when Mērwara tribes joined the

Temple as Geographic Marker

Sisodia elite from Uparamāla in tenuous political relationships reaching as far west as Jhadol. But for the most part, Chhapa, throughout the one-thousand-year history we have explored, remained beyond the reaches of kings. In this fertile region Śakta tantra and Śaiva tantra abound to this day—with the worship of Bērujī and Mātājī predominating at hilltop shrines where bhopas preside with peacock brooms and corn-grain divination around the corner from many of the ancient art historical sites covered in this study. Along the Som tributary to the Mahi River, it is no surprise that from southern Uparamāla into southern Mēdapāṭa, on the northern fringes of the Malwa plateau, we find that Nateśa (the classical version of Bērujī) and Cāmuṇḍā (the classical version of Mātājī) held sway alone more than one thousand years ago at Hita, Jagat, and Āaṭ. During the brief flash of dynastic power held by the Guhilas of Kiṣkindā prior to the tenth century, we find one of the earliest and most impressive four-faced and four-bodied Pāśupata-Śiva icons at Kalyanpur, which dates approximately to the eighth century and provides a precedent for Ahar, Ekliṅgjī, Bāḍolī, and other Pāśupata-Śaiva centers. Contemporaneous sculptures in Āmjhara suggest excellently carved mother goddess sculptures from the eighth to ninth centuries hidden and dispersed over a wide area, perhaps in the wake of destruction of their stone home. Postmillennial sites in the region suggest expansive states attempting often quite precariously to take hold of these largely tribal lands with the construction of fortresses, mines, and public works from Jaisamand Lake to Jāwar mines and multisectarian temples, to the long sculptural and architectural history of Jhadol. This southern stretch was a place where Bērujī and Mātājī were canonized in stone; where Jains bankrolled Sisodia imperial power; and where Rājputs, Afghans, Turks, and many others traveled, looted, negotiated, fought, laid siege, marched, pillaged, but never lingered. It was in this vacuum of statehood that art, architecture, and local religion flourished in creative and unique ways, leaving a record in stone and a rich heritage of practices found today that will leave no ephemeral traces for tomorrow.

Temple as Catalyst Renovation and Religious Merit in the Field

The history of temple renovation sheds light on tensions between preservation and use at archaeological sites in southern Rājāsthan. The word “renovate” in Hindi, nayā karanā, has Sanskrit origins (navī karoti) and contains the root “nayā” (new), just as the English word “renovate” means to make new again. Ancient inscriptions rarely distinguish between renovation and new construction since once an icon or site is jīrṇa (“old” or “tainted”), it should automatically be replaced with something new according to local belief. Renovations have historically ranged from slight modifications to significant additions, to completely rebuilding. This range of renovation activity continues today. Generous ancient definitions of renovation clash with ideas about archaeological preservation inherited from the British. Temple trusts, archaeological departments, and local patrons alike undertake creation in the name of preservation. The aesthetic interpretation of archaeological sites hinges on the subjective notion of taste. In the discipline of art history, beauty has long been a subject of debate.1 When we travel, both temporally and geographically, the issue of taste, of aesthetic judgment, is fraught with difficulty. Taste, according to John Elsner and Roger Cardinal, “is merely another item in the cabinet of social display.”2 They describe the “truly tasteful collector” as someone who creates taste rather than ascribing to it. This creation of taste is grounded in a unique approach valued for its difference. So when we turn to a temple, most often understood by the discipline of art history as a specimen from an archaeological collection, we exercise taste that originated in the colonial British project of collecting patrimony.3 Disdain for current modes of renovation, such as metallic gold paint, reaches far beyond the Ambikā temple. Any Indian urbanite, especially among the rising middle classes,

Temple as Catalyst

may well find metallic paint on the tenth-century stone sanctum of the Ambikā temple in Jagat just as distasteful as most art historians do. Ancient definitions of renovation gleaned from the historical record suggest that inexpensive, modern materials may actually perform a rather traditional function. In Alois Riegl’s terms, modern materials conflict with monuments’ transcendental “age value,” a value that he argues actually interferes with the preservation of monuments.4 The Ekliṅgjī temple complex and the Ambikā temple complex also lose “historical value” and their “original status as an artifact” to white plaster roofs, metallic gold paint, and a twenty-first-century white marble icon.5 Given that “disfiguration and decay detract from [historical value],” one could argue, as well, that the local people see their efforts as the preservation of “historical value” through the erasure of decay.6 With the birth of archaeology in India, the romantic ideal of the ruin implicit in “age value” was replaced by a quintessentially modern concern for “historical value.” Historicity had the power to “single out one moment and place it in the developmental continuum of the past and place it before our eyes as if it belonged to the present.”7 Current uses of archaeological sites in southern Rājāsthan attempt to steal buildings from history to create “intentional commemorative value.” According to Riegl’s definition, “intentional commemorative value aims to preserve a moment in the consciousness of later generations, and therefore to remain alive and present in perpetuity.”8 Sowing the seeds of memory keeps monuments alive and greatly empowers the specific commemorative vision and aspirational zeitgeist of the individual person constructing memory. Those who farm memory attempt to trump death through control of future generations’ harvests. Controversial enough to spark legal battles, the renovation of temple sites is an institution as old as temple building itself. Temples derive much of their meaning from the numinous power of the sites on which they stand. The ability to create links with the past often secures the value of a temple’s future.9 At both the Ambikā temple and the Śri Ekliṅgjī temple the future is woven into the past. This intersection of past and present is often a site of legal contention, moral quandary, and empowering affirmation, where preservation gives way to creation and consecration borders desecration. E K L I Ṅ G J Ī’ S G AT E S

Historical definitions of repair found in inscriptions reveal the amount of physical change and new building considered to be a renovation and not something entirely new. In 1489 CE, Mahārāṇā Raimal repaired the Śri Ekliṅgjī temple and made land grants.10 Buildings in the Nāgadā-Kailāśpurī region had been largely destroyed when the Guhila dynasty was taking refuge at Kumbhalgarh in the preceding century (fig. 2.1). Mahārāṇā Raimal’s inscription suggests that “repairs” often meant completely rebuilding on a sacred site. The Śri Ekliṅgjī temple dates

chapter 2

Figure 2.1. Kumbhalgarh fortress. © Deborah Stein.

to his era (see fig. 0.2). The icon of the god Śri Ekliṅgjī might even date to his time, even though the four-faced black stone is considered to be a self-revelatory “svayambu” icon that had been taken to Ekliṅgjī by Bappa Rāwal. Verse 90 of Mahārāṇā Raimal’s inscription reads: That which is eternal can never be an object of creation, that which is boundless can never have limit, and that which is Çtmanipada [confined to one’s self] can never be Parasmaipada [transferred to another]; but king Śri Rajamalla does make extensive gifts of gold, does encompass all religion, and allows all to stand free and happy.11

This verse is rather vague: it does not make explicit the exact object and architecture being donated. It could be a subtle way of referring to the installation of a new icon without negating the eternal existence of Śri Ekliṅgjī and his abode. At the Ambikā temple in Jagat there are no inscriptions referring to the donation of the goddess Ambikā, even though more than one icon has graced the main niche over the past forty years. The lack of precise written records when a new icon was installed comes as no surprise, since to mark a beginning for an icon is to take away its eternity and, hence, its divinity. Although ancient renovations enjoy a certain romantic authenticity, modern renovations are often dismissed as garish intrusions. The Ekliṅgjī temple complex has witnessed a long history of preservation under various mahārāṇās, whether under Kumbhā and Raimal in the fifteenth century or under Śri Arvind Singh

Temple as Catalyst

Figure 2.2. Rampart, fifteenth century, Ekliṅgjī. © Deborah Stein.

and his father in the twentieth century. But the preservation efforts of these kings differ from those of the Archaeological Survey of India (ASI). For the mahārāṇās, preservation has often not been far from creation. Mokal added a rampart to the complex, whereas Raimal may have replaced the icon itself (fig. 2.2). These innovations are now ancient history. There is little difference between the authenticity of the tenth-century Lakulīśa shrine as a historical site (see fig. 1.1) and the authenticity of the fifteenth-century Śri Ekliṅgjī temple (see fig. 0.2). But if we turn to some of the renovations in plaster and concrete on temple roofs throughout the complex (see fig. 0.7), or to the row of shrines to the left of the main entrance, some may argue these newer repairs detract from the authenticity of the archaeological site. The newer renovations lack the period integrity of the tenth-century Lakulīśa shrine and the fifteenth-century Śri Ekliṅgjī temple. Regal renovations at a site like Ekliṅgjī may produce intense aesthetic shifts; however, the spirit of housing a living being—understood to be the ruler of Mewār—suggests an alternative form of continuity. Like the mahārāṇās of Mewār, the ASI also repairs and occasionally restores archaeological sites. In some ways the ASI is more or less forthright about its projects. It often attempts to perfectly maintain the color and texture of the ancient stone, making it quite difficult to distinguish from the original structure. This creates a visual harmony that is historically discordant. In contrast, mahārāṇās visually delineate and make repair records of the changes they make to the site. Although the regal renovations may

chapter 2

Figure 2.3. Marble icon stolen in 1998, Jagat. © Deborah Stein.

documentation on the subject, especially after theft and damage. Writing in 1964, he reported some very important information about the Ambikā temple’s inner sanctum: “The interior, measuring 7 feet × 7 feet, contained a medieval schist image of goddess Mahiṣāsuramardinī, under regular worship on an altar. Here we notice the demon coming out of the chopped off head of the buffalo (Mahiṣā) under the mighty influence of contemporary art traditions.”22 This description corresponds to the image found on the back exterior wall of the shrine (see fig. 1.8), temporally in-between the zoomorphic form of Mahiṣā found on the south wall and the anthropomorphic form of the demon found on the north wall during the course of circumambulation. Hence, as of 1957, when Agrawala saw it, an ancient statue of Mahiṣāsuramardinī was under worship in the sanctum. This sculpture’s iconography correlated to the iconographical program found on the exterior temple walls.

Temple as Catalyst

Figure 2.4. New marble image from Jaipur, May 2002. © Deborah Stein.

Unfortunately, Agrawala did not include an image of this sculpture. Surprisingly, a photograph from 1963 reveals neither any deity nor any sign of worship whatsoever.23 Was the white marble Mahiṣāsuramardinī sculpture under worship in the sanctum in 1998 (fig. 2.3) actually installed in 1957? Stolen in 2000, the statue had left the sanctum empty when I returned in January of 2002.24 The image believed by scholars to be the original icon of Kṣēmaṅkarī remained cast aside, leaned up against a side wall.25 By May of 2002 the villagers of Jagat and the surrounding area had raised enough money to commission a new marble image made in Jaipur (fig. 2.4). The ensuing installation raised critical questions concerning the value of the site as patrimony, as well as problems with rejecting archaeological death in favor of modern religious and political use.

chapter 2

Figure 2.5. Śubhamaṇḍapa, archival photo (1950s), building foundation, c. eleventh century; brick layer, c. 1800s–1900s, Jagat. © Archaeological Survey of India (ASI). My sincere thanks to all of the ASI officers who worked together to efficiently provide me with copies of relevant photographs.

Preservation often left good records but rarely succeeded in maintaining archaeological material in situ. At Jagat, preservation involved the removal of material considered historically irrelevant. A photograph of the śubhamaṇḍapa (approximately fifty feet in front of the entrance to the Ambikā temple) taken in the 1950s by the ASI (fig. 2.5) reveals that the building originally had a second story of brick.26 A stone staircase from the lower level of this structure leading up to what is now a roof indicates a second story was always an integral part of this structure; holes remaining in the stone suggest masonry to support it (fig. 2.6). The fate of this brick structure is unknown. Traces of white to the right of the portal in the photograph may indicate this structure was originally plastered or stuccoed by the ASI. The śubhamaṇḍapa suggests how preservation colors our understanding of a site’s history. Although preservation implies permanence, sites do change during restoration. While archaeological sites may die as living monuments, they do not remain unchanged. Periods of dormancy and renovation at Jagat are not limited to recent history. A three-hundred-year hiatus was followed by a flurry of inscriptions in the eighteenth century. These inscriptions refer to yatras (pilgrimages) made to Jagat and

Temple as Catalyst

Figure 2.6. Śubhamaṇḍapa, Jagat. Holes in the masonry suggest the second story may have originally been built of wood. © Deborah Stein.

reveal a diverse temple audience. Tours of 1744 indicate that royalty chose the site for pilgrimage and that Sunday was already an auspicious day for the goddess by the eighteenth century. Inscriptions of 1744 and 1745 commemorate the pilgrimage of architects belonging to the Sompurā and Nāgadā guilds. These guilds link this temple to an important set of Mewāri temples to the north. Pilgrimages were jointly recorded by several different castes, including Bhils and Meenas, and not just royalty. An inscription of 1792 lists the names of nine commoners, one of whom may have been a woman, Roopajaa. Renunciants, nobles, Bhils, women, and masons all wanted to leave their trace on the stone temple. Like inscriptions, legal documents attest to the uses of temples. The Ambikā temple at Jagat falls under the jurisdiction of the Udaipur Archaeological Department. When the sanctum was painted gold, my field of inquiry turned to the archaeological department, where the only legal document in the museum dates to the colonial period. The Jaipur Ancient Monuments Act of 1941 makes two claims: (1) a place of worship must not be used for “any purpose inconsistent with its character”; and (2)

chapter 2

when a protected monument is used for religious worship, it should be protected from pollution or desecration.27 The tension between use and preservation remains unresolved. The legal definition of desecration, which loosely implies harming an icon, hateful graffiti, or the destruction of a mosque to build a temple, lacks clarity. In contrast, the code makes no mention of changing the visual culture of an ancient site as a part of consecration. Section 5 provides for maintenance, including “fencing, covering in, repairing, restoring and cleansing of a protected monument.”28 The act seems to refer to restoration associated with preservation in the archaeological sense of the term. The Jaipur Preservation Act attempts simultaneously to protect the sacral quality of monuments and to maintain them as archaeological treasures, seemingly unaware of the tensions between these two models. In the absence of a clear legal mandate, the issue of preservation and use becomes a matter of taste—albeit with significant political ramifications. Although the metallic paint and modern marble sculpture may even be considered kitsch or vulgar according to Western art-historical notions of taste, from a Marxist standpoint these modern renovations may well be the opposite of vulgar. Theodor Adorno writes: Only in mutilated fashion does the vulgar represent the plebian that is held at a distance by the so-called high arts. When art has allowed itself, without condescension, to be inspired by a plebeian element, art has gained in an authentic weightiness that is the opposite of vulgar. Art becomes vulgar through condescension: when, especially by means of humor, it appeals to deformed consciousness and confirms it. It suits domination if what it has made out of the masses and what it drills into them can be chalked up to their own guilty desires.29

If we take Adorno’s definition of vulgarity as a form of condescension, the use of metallic gold paint on the tenth-century sanctum at Jagat could just as well be understood as possessing “an authentic weightiness that is the opposite of vulgar.” The act of painting the shrine metallic gold is also a commemoration of the installation of a new icon, a white marble goddess statue chiseled in Jaipur (fig. 2.7). This piece of sculpture has no place in a museum. The white stone fits neither the rhetoric of modern transnational artists such as Anish Kapoor nor the premodern Hindu art in museum collections. The new icon has no place on the art market, no reason to be stolen. This primarily religious object is not valuable aesthetically yet extremely valuable from a ritual standpoint. The installation of the goddess is a political act of reclaiming ritual space. The authors of Khaki Shorts and Saffron Flags describe how the 1990s’ Hindu right corresponds to a rise in popular goddess worship.30 By painting the ancient Ambikā temple, the participants removed its historical and aesthetic value and replaced it with political and ritual value, thus putting the sanctum in the same category as the new icon it housed. Theft in the eyes of the preservationist, this act was a reclaiming of space in the eyes of the Rājputs who sponsored the goddess

Temple as Catalyst

Figure 2.7. New icon, under worship in the Ambikā temple, 2009. © Deborah Stein.

installation. For the average village local, however, people who would have little impact on their lives or practice were simply making a claim to power. In the case of the śubhamaṇḍapa at Jagat, the performers of ritual usurped the historical site. The grassy jagged lip of the upper wall meets no roof in the 1950s photograph (fig. 2.5). As of 2002 the same structure looked well maintained and

chapter 2

Figure 2.8. Perishable maṇḍapa built for the goddess installation in May 2002, Jagat. © Deborah Stein.

paralleled the form of a perishable maṇḍapa constructed for the new goddess installation at the Ambikā temple (fig. 2.8). Locals asserted their independence from any local, state, or national archaeological administration by staging this ceremony to direct the future of their patrimony. The Hindu goddess Ambikā was installed just after violence broke out in Gujarat. Months of riots followed an attack at the Godhra railway station on February 27, 2002, that burned Hindu activists who were returning from a pilgrimage to Ayodhyā.31 As the sacrificial fire burned in Jagat, the anguish had not been extinguished in neighboring Gujarat. Hundreds of Muslims were living in refugee camps, and the state government was doing little, even participating in the wave of unmitigated killing. In Rājāsthan the threat of violence forced Udaipur, the capital city of Mewār, to close for a seventy-two-hour curfew. Rājput boys in Jagat called on their warrior ancestry, hoping when they grew up either to run for political office or to fight the terrorists.32 This anxious atmosphere may have contributed to the Bharatiya Janata Party (BJP) victory in Rājāsthan, while the left-leaning Congress Party won the national election. This branch of the Hindu right exercises a democratic rhetoric that shallowly masks complacency toward the violent pull of communalism in northern India. After the destruction of the Babri Masjid mosque on December 6, 1992, the BJP’s alliance with extreme groups such as the Rashtriya Swayamsevak Sangh and the

Temple as Catalyst

Vishwa Hindu Parishad became more explicit.33 Ten years later, in 2002, the same communal violence that began with the Hindu right’s destruction of a mosque in Ayodhyā, the legendary birthplace of the Hindu hero Rama, continued with an attack of a train full of pilgrims returning from Ayodhyā to Gujarat. Approximately eight hundred Muslim deaths, out of one thousand total deaths, suggest that the rhetoric of revenge and parity was political rather than factual. Ironically, the Ambikā temple did not see more use with a new icon. The first Navratri nine-day goddess holiday after the installation did not even include the usual buffalo sacrifice. The eight-day installation ceremony of May 2002 culminated in a final fire sacrifice on the last day and the actual placing of the image in the sanctum. Important Rājāsthani luminaries and people from villages around Jagat attended this ceremony. The prince of Jagat, who now runs a heritage hotel in Udaipur, and the Rājāsthan home minister were among the speakers at what appeared to be a right-wing BJP rally delivered to the locals attending the fire ceremony. The focus of the goddess installation was anything but ritual for the majority of participants. A diagram of the social space of the ceremony reveals (1) ritual taking place between the Ambikā temple and the śubhamaṇḍapa, (2) a political rally for village men to the side of the śubhamaṇḍapa, (3) distribution of prasād to women and girls to the side of the temple, (4) a cluster of boys behind the women and men, and (5) the researcher on a (polluting) pile of shoes just outside the temporary maṇḍapa, where she had been given permission to film (fig. 2.9). Although men saw the ritual under the contemporary maṇḍapa, they clearly were listening to political speeches being broadcast in their midst. The women were chatting while nibbling prasād far from the ritual. Only those conducting the ritual paid attention. The installation of a new icon restored the honor of a stolen goddess. Men and women who paid little attention to the installation ritual now use the temple, while Rājputs, priests, and politicians who were staging their power rarely or never return to the site. The quotidian and seasonal celebrations at the site remained completely unchanged by the presence of the new icon. In fact, if anything, the Ambikā temple seemed even less a focus of attention once the new icon was in place. The anticlimax satisfied the hope of reestablishing honor to a site marred by theft. Complacency replaced desire. The collection of money and power along with the ensuing enactment of the goddess installation answered a call to restore the honor of a stolen goddess. Once her honor was restored, her maintenance was turned back to the cluster of Bhil and Meena women who pray to her and to her hilltop sister, Mallar Mātā. As an alternative to preservation, temple use protects the temple from death even though it cannot offer unbroken continuity. A romantic interpretation assumes that if local people control the thousand-year-old Ambikā temple, continuity is maintained—somehow they form an unbroken chain with the past. This

Temple as Royal Abode The Regal, the Real, and the Ideal in Fifteenth-Century Mewār

Many have written about fifteenth-century Mewār over the past century.1 The material residue of imaginaries contrasts with the material and architectural residue still found in situ today. This chapter on Guhila dynastic history marks a moment when the Sisodia dynasty, which claimed descent from the Guhilas, could look back on a past that included more than oral history and the seeds of dynastic legitimization that their Guhila forebears had used back in the tenth century. Tenth-century inscriptions and architecture sought to legitimize the rise of the Guhilas in the vacuum of power that characterized a two-hundred-year period prior. In the fifteenth century the physical record of tenth-century production and monumentalization served to recall Guhila greatness on behalf of its successors—the Sisodias of Mewār. This early modern period in the kingdom of Mewār was characterized by revivalism. At the fortress of Chittauḍgaḍh, carefully labeled iconography in the Kīrtistambha tower indicates an early instance of self-definition and the fear of losing heritage that gives birth to nostalgia. Long bardic inscriptions and early historical texts use the tenth-century past to deal with fifteenth-century insecurities in much the same way that twentieth- and twenty-first-century people make claims of authenticity based on fifteenth-century history to cope with rapidly changing governance and the many insecurities of the modern period. Further art historical and sociocultural comparison with Malwa, Gujarat, Gwalior, Delhi, and Persia would illuminate more about the visual, erudite, and archival impulses that characterized polity over a large multisectarian region in the fifteenth century. Such a geographic scale of comparison, however, lies beyond the scope of this chapter. In an era of encroaching Mughal power from Delhi and sultanate power from Gujarat and Malwa, the kingdom of Mewār used culture to 72

Temple as Royal Abode

produce, not just mark, its borders. In fact, there were no clear borders—just texts, buildings, images, arguments, dreams, anthologies, and the like. Recently, textual historians have mapped inscriptional data to illuminate the extent and limits of Guhila hegemony in the early stages of state formation in Mewār, Chhapa, and Vagada. Unlike the tenth century, when Guhila dynastic identity was under negotiation, the late medieval period was characterized by a tenuous Rājput political hold in the form of intense “cultural” production.2 This fifteenth-century propensity to use archival impulses and quotation in illuminated books, architectural projects, music, food, and other forms of encyclopedia was a multisectarian form of polity that stretched across northwestern India as far as Persia in this period. For example, the illustrated Nīmāt Nāmā cookbook, produced at court in neighboring Malwa during the same era, references specifically Persian modes of kingship and painting styles alongside Indian and Persian foodstuffs. The recording of recipes, like the musical encyclopedia said to have been authored by Mahārāṇā Kumbhā Mewār during the same era, suggests an archival impulse—and an artistic production or expression of that archival desire—as a cornerstone of polity in this time and place. Here I focus on the ebb and flow of architectural and inscriptional production in two different geographic locations. In Mewār the fifteenth and sixteenth centuries were tumultuous times that left architectural traces of Guhila claims to power— buildings that do not directly correlate to written accounts of Rājput and Mughal histories. The mahārāṇās of Mewār sponsored temples and towers at Ekliṅgjī and at Chittauḍgaḍh during the fifteenth and sixteenth centuries. Meanwhile, between Mewār and Vagada to the south, the region of Chhapa witnessed similar vacillations between extensive patronage and cultural silence in the stone record. Are we to understand architectural absence as a corroboration of written records of threat, danger, or even defeat? Local rumor would have us believe that the Ambikā temple in Jagat was buried in sand at one point in history to protect it from destruction, but no proof or even suggestion of dates for this theory remain. The military history of Ekliṅgjī, in contrast, remains quite legible from the inscriptional and artistic record already analyzed in detail by Tryna Lyons.3 This chapter fills the architectural and inscriptional silences in Jagat with the exuberant, active patronage of a multisectarian sacred center in neighboring Jāwar. In an interesting parallel to tenth-century Mēdapāṭa, where the Lakulīśa temple defined a Guhila center at Ekliṅgjī and the Ambikā temple articulated regional style on the border of Guhila territories, the fifteenth-century Mewāri architect Maṇḍana left his traces on the border as well. The successors to the Guhila lineage self-consciously defined their kingdom from the geographic location of an oscillating center between Ekliṅgjī/Nāgadā and Chittauḍgaḍh and left residue of Mewāri history on a border defined by the development of industry rather than the self-conscious construction of history.

chapter 3

The art and architectural history of fifteenth-century Mewār, Chhapa, and Vagada reveals relative silence at the twenty-first-century centers of Ekliṅgjī (Nāgadā/Ahar), and Jagat (near Jālōr, Ranthambhor, Jāwar, and Bambora). Instead, initially Jain centers in the fortress of Chittauḍgaḍh, the mines of Jāwar, and the monastery of Delwara defined the polity of the sultanate period in this region. The Śri Ekliṅgjī temple complex and the Ambikā temple in Jagat thus form a diachronic relationship between the tenth and the twenty-first centuries that nevertheless reveals intense periods of rupture during the very moment of tenuous state formation in the region. The pendulum of architectural production between Ekliṅgjī and Chittauḍgaḍh, on the one hand, and Jagat and Jāwar, on the other, offer a material record of very early instances of self-conscious history making, long after the tenth-century manufacture of political power and foreshadowing the modern period in which the display of art in museums and in situ continues to mark territories of both geography and imagination. BA P PA R ĀWA L : L E G E N DA RY F OU N D E R O F T H E GUHILAS OF EKLIṄGJĪ

One of the most powerful icons of the Guhila/Sisodia rupture between the tenth and the fifteenth centuries is the figure of Bappa Rāwal—identified as the founder of Mewār lineage in the present but not listed as the originator of the Guhila line in tenth-century inscriptions. The aesthetic power of this somewhat obscure lineage debate is evident in a twentieth-century French sculptor’s rendition of Bappa (fig. 3.1), housed within a structure generally attributed to the patronage of fifteenth-century Mahārāṇā Kumbhā. This modern statue of Bappa articulates the claims of the Mahārāṇā Mewār website, where the eighth-century Bappa is linked to the eighth descendant of the Guhila line, Prince Kalbhoj. The most recent Mewār encyclopedia produced by the House of Mewār identifies Bappa with Kalbhoj and more accurately navigates the uncertainties through a description of the relationships between legend and history. There, in the same vein as Col. James Tod’s versions, Bappa is described as the founder of Mewār who received spiritual instruction from the Śaiva acetic Harit Rashi.4 Bappa became a devotee of Śri Ekliṅgjī and was named by his spiritual teacher as the first regent of Mewār in the service of the divine ruler of Mewār, Śiva in the manifestation of Śri Ekliṅgjī. A painting displayed inside the mahārāṇā’s private residence reveals Bappa Rāwal, hands folded in respect. His greeting is aimed at Harit Rashi, who floats above in a very literal iconographic rendition of the haṁsa (incorrectly translated as “swan”) air vessels that are so famous in Sanskrit literature. These protoairplanes date back to the Rāmāyaṇa and the Ṁahābhārata, and it is in a similar vessel that Harit Rashi, in a white plaster sculpture, hovers over the entrance to the Śri Ekliṅgjī temple today (fig. 3.2). One wonders, in fact, if the sculpture was

Figure 3.1. Bappa Rāwal, by a French sculptor, c. second half of the twentieth century, Ekliṅgjī. © Deborah Stein.

Figure 3.2. Modern sculpture of Harit Rashi in a Sanskritic haṁsa vehicle (swan boat), Śri Ekliṅgjī temple. © Deborah Stein.

Figure 3.3. Harit Rashi on an exterior wall in Udaipur during Rath Yātrā. © Deborah Stein.

Figure 3.4. Maharishi Guru Harit Rashi entrusting the State of Mewar to Bappa Rawal, 734 CE. Pannalal Parasram Gaud, Gouache on vasli, c. 1934. Accession No. 2012.18.0147_R. Image courtesy: The City Palace Museum, Udaipur, © MMCF.

Temple as Royal Abode

copied from the painting or vice versa. During the Rath Yātrā parade in Udaipur in 2002, Harit Rashi, in his signature swan boat, greeted people in the streets from a second-floor mural (fig. 3.3) that depicts the sage’s boat hovering above the temple of Ekliṅgjī (in blue) and the large brown mountain of Vindhyāvāsinī. There even appears to be the Rastrasena temple peeking out from behind the mountain atop a green peak in the distance. This street mural contrasts in style, but not in basic iconography, with one of the mahārāṇā’s favorite paintings on view in his home (fig. 3.4). In this less cluttered composition, Bappa Rāwal (in yellow) clasps his hands as he looks up at the sage in the red swan boat. Dating back to 971, the Lakulīśa temple and inscription record a debate that took place among Buddhists, Jains, and the Pāśupata-Śivas. This inscription also links for the first time the Guhila dynasty to the Pāśupata-Śivas. Line 5 of the Lakulīśa temple inscription mentions Bappa, and line 15 references Ekliṅgjī.5 Bhandarkar assumes the 971 inscription as proof of why Bappa remains so important to the mahārāṇās of Mewār. The Atpur inscription of 977 clearly lists the early lineage of the Guhila line as (1) Guhadatta, (2) Bhoja, (3) Mahendra, (4) Naga, (5) Syeela, (6) Aparajīta, (7) Mahindra, (8) Kalbhoj (associated by some with Bappa), (9) Khoman, (10) Bhartṛpaṭṭa, (11) Singse, (12) Śri Ullut, (13) Nirvāhana, (14) Salvāhana, and (15) Śaktikumār.6 Inscriptions within less than a decade and fewer than one hundred or two hundred kilometers of each other reveal slightly different lineages. The legend of Bappa Rāwal found on the Mahārāṇā Mewār website does not correlate historically with tenth-century inscriptions such as the c. 971 Lakulīśa temple inscription, and offers instead a very thorough compilation of oral history around this important figure over the longue durée. Kalbhoj’s name of endearment as “Bappa” meaning father, and “Rāwal” as the traditional Rājput title for a king is translated and explained on the Mahārāṇā Mewār website. A lineage is set forth—one that does not list Bappa Rāwal as the progenitor of the royal line, in contrast with post-fourteenth-century records. D. C. Sircar situates the elevation of Bappa from “petty Rāwal” to “one of the greatest heroes India ever produced” in folklore as a response to status Bappa earned from “the struggle with the invaders in the sixteenth century AD.”7 Nandini Kapur cites the seventeenth-century Hindi poetry of Girdhar Asia and the seventeenth-century history of court official Muhanot Nainsi to conclude, “What Bappa did for the thirteenth century Guhilas, Hammīra did for the fifteenth century Guhilas.”8 Whereas Bappa was the progenitor of the Sisodia line by the fifteenth century, as of the tenth century, Guhadatta was listed as the first Guhila of Mēdapāṭa (later known as Mewār). Tryna Lyons also mentions the Bappa debate, citing the inscriptional lineages as problematic with the Bappa myth.9 In contrast, Kapur seems to take the KalbhojBappa equivalency argument at face value—perhaps owing to her quotations of origin myths from famous seventeenth-century history and poetry. Although she seems to rely largely on Tod and Sharma and their post-1300 dynastic viewpoints

chapter 3

as her sources for nineteenth-century Bappa adoration, her careful reliance on inscriptional data leads her to even more specific dates of rupture between the Guhila and the Sisodia lines.10 Corroborated by Topsfield’s visual history of manuscripts in this region and the work of Lyons with local bards, Kapur’s argument of a break between 1303, when Alāuddīn Khilji sacked Ekliṅgjī, and 1326, when the generically named Sisodia progenitor Hammīra takes “back” Chittauḍgaḍh, seems quite plausible and agreeably specific in relation to many older accounts.11 This break is likely the very reason why the fifteenth-century art history of Mewār reflects an archival impulse to quote the past and an encyclopedic impulse to create the actual monuments listed in Sanskrit and vernacular architectural manuals called śilpaśāstras; whereas, before the break, the fledgling Guhilas built in new ways to legitimize their rule with stone architecture for the first time from the 950s to the 970s CE but did not seek to quote the past or build an archive—there was no strong recent past on which to build. W HO WA S HA M M Ī R A ? SI S O D IA- G U H I L A C L A I M S T O CHITTAUḌGAḌH

If the Sisodia clan seems to appear suddenly when Mahārāṇā Kumbhā’s grandfather ruled the kingdom of Mewār from Chittauḍgaḍh (and not the Guhila stronghold of Ekliṅgjī/Nāgadā), then where did they come from? A largely silent inscriptional record from the time after the Vindhyāvāsinī goddess temple in Ekliṅgjī to Kumbhā’s grandfather in Chittauḍgaḍh suggests that perhaps a new ruling dynasty filled a political vacuum in this region from before the time of Alāuddīn Khilji’s Afghan raid in 1305 to the time of Kumbhā’s grandfather. In his book Objects of Translation Finbarr Flood alludes to these precarious origins when he points out that the name “Hammīra” simply means “a ruler.” Nandini Kapur argues from what inscriptional record remains that the expansion under Rāṇā Hammīra and Rāṇā Lakha “seems to have begun the process of the annexation of Merwara.”12 There was a critical shift in state formation in the fifteenth century, where tribal areas are increasingly incorporated into the Rājput state. Kingship officially ended with Indian independence. Today, Mewār is technically ruled by the state of Rājāsthan under the nation of India. For many in Mewār, their ruler remains Śri Ekliṅgjī, a god—and a god in need of a dīwān at that. It is in this capacity that Śriji Arvind Singh Mewār serves as mahārāṇā in the twenty-first century and the age of the modern nation-state. He and many in modern Mewār find the hegemony of their heritage in the exploits of Rāṇā Hammīra’s grandson. The grandson of this Sisodia “Hammīra” was the famous king Mahārāṇā Kumbhā, who ruled from Chittauḍgaḍh and then constructed Kumbhalgarh at the northern edges of what was Guhila dominion, or at least the area known as Mēdapāṭa, where the old tenth-century Jain temple lies at Ghāṅerāo. He moved his capital from the southeastern edges of the Chhapa/Vagada border with Mewār to

Temple as Royal Abode

Figure 3.5. Pratap, Rath Yātrā parade, Udaipur, 2002. © Deborah Stein.

the furthest point north—never once selecting the Nāgadā/Ekliṅgjī region for his capital. His daughter Ramabai in turn held the mining town of Jāwar as part of her jagīr (dowry) in the heart of Chhapa to the south of Mewār, halfway to the Guhila offshoot kingdom of Dūṅgarpur. In contrast with his lesser-known grandfather, Kumbhā was a very active ruler, patron of the arts, author, and architectural patron. His distinction was heralded by colonial historian James Tod and by contemporary nationalist parties in India to this day. The Sisodia line eventually was known for Kumbhā’s descendant Mahārāṇā Pratap, who is said to have defeated the Mughals at the Battle of Haldīghāṭī in the second half of the sixteenth century—a hundred years after Kumbhā was actively sponsoring architectural projects at Chittauḍgaḍh. Pratap’s legacy has resulted in multiple visual renditions ranging from a large-scale bronze statue of him on horseback that greets the visitors at the airport, to the image carried during a Rath Yātrā parade in Udaipur in 2002 (fig. 3.5). He is recognizable by his red-trimmed blue coat, his portly and confident stance, and his profile—all of which have made his portrait infamous in several renditions from calendar art to palace paintings. This colonial and nationalist lens of Rājput greatness has influenced our understandings and misunderstandings of some of Kumbhā’s most famous monuments to this day.13 More recent studies, such as Upendra Nath’s book about Mahārāṇā Kumbhā and, in 2002, Nandini Kapur’s excellent work on state formation, have begun to correct that picture.14 Recent studies of artistic agency by Tryna Lyons provide an impressive depth of detailed data carefully culled in the field directly

chapter 3

from bards, as well as translated directly from inscriptions and mason marks on monuments and manuscripts alike.15 A closer look at the architectural landscape of sultanate-era Mewār, Chhapa, and Vagada exposes just how precarious the hegemony of Guhila-Sisodia heritage was prior to the fifteenth century. The Kīrtistambha of Chittauḍgaḍh, for example, illustrates the quintessentially fifteenth-century phenomenon of reification of kingship through specifically revivalist and archival architectural projects. One of the most impressive monuments built under Kumbhā’s reign is the Kīrtistambha at Chittauḍgaḍh (fig. 3.6). Previously understood as a Jayastambha, or “Tower of Victory,” the tall stone spire of Rāṇā Kumbhā’s fortress remains to this day the symbol of Chittauḍgaḍh, Rājāsthan, and, consequently, a symbol of Rājput glory in India. The extensive plateau has been alternately ruled and captured for hundreds of years owing to its location in the center of a geographic triangle formed by Gujarat, Malwa, and Agra. In a long line of rulers who used architecture—and, more specifically, pillars, stambhas, towers, and minarets of all types—to define their dominion and to augment their power in South Asia, Kumbhā’s cultural patronage far outweighs his military claims to victory. Moreover, it is exactly this tactic—the privileging of cultural hegemony over political territorial boundaries—that made it possible for this fifteenth-century ruler’s legacy to acquire such an enduring form. In contrast to the typical victory narrative, both the rich sculptural contents and the inscriptions of the nine-story interior tell a different story—a tale that could be romantically coined the making of one of India’s first museums, because it is the story of a collection—and a very permanent collection at that. The combination of two features distinguishes this tower from any previous Indian monument: first, the incised labeling of each image in stone underneath; and, second, an interior turn-square staircase that permits the viewer to travel across nine different interior landings within a span of ten minutes or so.16 The labels fix each sculpture’s meaning in stone—as if it were possible to curate for posterity. The internal square helical staircase collapses the proximity of nine different gallery spaces into a single monument. Rather than travel a distance of a day or more via horse or on foot to see a variety of temple exteriors, the fifteenth-century viewer could experience these levels in intimate proximity to one another. Each level of the tower quotes prior modes of architecture and iconography to create a permanent expression of the artistic canon of the day. Even though we are inside the tower, this collection of iconographic programs quotes temple exteriors and never seems to reveal an inner icon—a crucial distinction from medieval temple iconography and architecture. All prior towers, stambhas, and even kīrtistambhas in South Asia and even western Asia relied on surface decoration of the exterior; only the Kīrtistambha had such rich interior sculptural decoration. Affixed to the inside ceiling of the uppermost gallery, an inscription corroborates the visual evidence of collecting and the creation of a permanent canon

Temple as Royal Abode

Figure 3.6. Kīrtistambha tower, c. 1440–60, Chittauḍgaḍh. © Deborah Stein.

in stone.17 This inscription reveals a reliance on a specific scientific manual, or śilpaśāstra, called the Aparājitapṛcchā, an early medieval architectural treatise that lays out a prescription for the construction and iconographic program of kīrtistambha towers. With the inclusion of a portion of this text on the building itself, we learn that the goal of a kīrtistambha is to mark a royal capital with a

chapter 3

tower meant to include all worldly and celestial things. Although the śilpaśāstra text—revised and reauthored by the king himself and affixed to the monument interior—does not give a specific plan for a square helical internal staircase, nor does it give an iconographic prescription for specific programmatic placement of sculpture in protrusions and recesses of a drawn elevation, it does suggest the purpose of a kīrtistambha and the role of this specific choice of monument in King Kumbhā’s worldview. The story told by Kumbhā’s Kīrtistambha suggests a desire to encompass the entire cosmos within a single structure—an enticing window onto the king’s own individual subjective fifteenth-century “period eye” as an architectural patron in the kingdom of Mewār at a crossroads with an increasingly complex political, ethnic, sectarian, and cultural landscape. Constructed between 1440 and 1460, the Kīrtistambha designed and built by architect Jaita and his sons Napa and Puna visually articulates a claim to kirtti (glory) rather than jaya (victory). The glorious claim of Kumbhā’s regal tower—to encompass the heavens and earth—yields some surprising results. We are left with neither a clear-cut tale of Hindu/Muslim conflict—as many historians in colonial and nationalist modes have previously assumed—nor a tale of multicultural premodern global harmony, as other postmodern historians may hope to find. Of course, no one can make a claim to a truly “authentic” history in any scholarly way, but the visual record does leave behind some important clues about the hopes and dreams of Rāṇā Kumbhā, his architects, and their revivalist claim for the cultural and dynastic place of the Sisodia branch of Rājputs in relation to their Guhila dynasty predecessors in Mewār. How, then, can we begin to understand a royally sponsored monument that gives sculptural form and a wall label to ordinary people, such as servants (fig. 3.7), but does not depict the king? How can we think about a collection that includes the calligraphic presence of the Muslim God, Allah, and multisectarian iconographic sculpture of Hindu deities Śiva, Śakti, and Viṣṇu all under the same roof but leaves out any clear references to Jainism? Could this suggest that the multisectarian iconography reflects a specifically kshetrias, or ruler’s caste, point of view? Does the Jain tower precedent preclude any need for Jain iconography, or does their lack of a godhead similar to Allah, Śiva, or Viṣṇu suggest that a saint is not part of Kumbhā’s cosmos whereas his human servants, dancers, and the like remain an integral part of his world? The iconographic collection of the tower interior delineates complex webs of relationships among a variety of belief systems, and it does so in an almost encyclopedic manner. Far from a random assemblage of imagery, the organization of these images architecturally in relation to one another suggests an archival impulse on the part of the makers. The creation of an archive suggests a desire to classify information and objects—to forge and fix relationships for future generations. As an active patron of the arts, Rāṇā Kumbhā regularly sponsored architectural projects and scholarly works. Kumbhā is credited even with important musical treatises

Temple as Royal Abode

Figure 3.7. Labeled sculpture of servants in a domestic interior. Interior of Kīrtistambha tower, c. 1440–60, Chittauḍgaḍh. © Deborah Stein.

called the Sangita-ratnakara and Sangita-krama-dipaka, which also engages in an encyclopedic enterprise with a revivalist tone and an eye to fixing a contemporary view of the past in a permanent way for future generations. The fifteenth-century Kīrtistambha of Chittauḍgaḍh can be read as a permanent record of one king’s curatorial eye toward the past and his political claims for the future. I N SI D E T H E K Ī RT I STA M B HA : “ T H E P E R M A N E N T C O L L E C T IO N ”

The Kīrtistambha was built by a Vaiṣṇavaite king in a Śaivaite kingdom. Mewāris consider Śri Ekliṅgjī, a Pāśupata manifestation of the god Śiva, as their divine ruler—whereas the human mahārāṇā serves only as “dīwān,” or his divine guardian. Technically speaking, one could argue that Śiva is the ruler of Mewār. As for the antiquity of this claim, an inscription on the Lakulīśa temple dated to 971 records a great debate that took place among the Buddhists, the Jains, and the Pāśupata Śaivas. This inscription establishes a link between the Guhila dynasty and the Pāśupata Śaivas, said to have been the winners of the debate. Although the inscription does not clearly delineate whether Ekliṅgjī was understood as the divine ruler of Mewār in the tenth century, it does establish Mewār as a Pāśupata kingdom. Rāṇā Kumbhā, however, clearly was a devotee of Viṣṇu. He sponsored the famous Mīrabai temple at Ekliṅgjī and even gave his own daughter the Vaiṣṇavaite name “Ramabai.”18 The Kīrtistambha tower he sponsored was also dedicated to Viṣṇu, and it is an image of Viṣṇu that first greets viewers as they enter the tower. The program does not focus uniquely on Viṣṇu by any means. In fact, the complex iconographic program establishes complex relationships among Viṣṇu, Śiva,

chapter 3

and the people of Mewār. In addition, the ensemble of the sculptural program suggests more of an encyclopedic, curatorial eye toward a canon of iconographic traditions rather than a pointed sectarian journey toward any singular religious experience. The interior program includes the following general categories of sculpture: 1. 2. 3. 4. 5. 6. 7. 8. 9.

Male Deities Nondual Deities “Muslim” and “Hindu” Deities Male and Female Deities Humans, Mostly Ordinary People (Nonmythic, Nonnoble, Nonclergy) Goddesses Open Empty Gallery Sealed Empty Gallery Observation Deck

In tandem with medieval North Indian architectural practices, the Kīrtistambha program engages in architectural punning and other metaphors that grow from placement and the visual interaction between sculpture, wall, and building.19 For example, the fifth floor, halfway up the tower, is filled with carefully labeled images of ordinary people like servants, ascetics, drummers, dancers, architects, and scribes. On this level, where the people of Mewār are meticulously represented, the exterior program explicitly alludes to Śri Ekliṅgjī, the four-faced Pāśupata god of Śiva understood today to rule Mewār. Can we speculate that the planners were suggesting that this rich diversity of the human world was literally encased within a particular Śaivaite paradigm? In a second example, the nondual deities on floor 2 precede the viewer’s ascension to floor 3. The first two floors of the tower are larger architecturally and serve as a base for the upper stories. Both inside and outside there is more room, so the architects were able to include thick temple facades with protrusions and recesses on the building’s interior. The figures of Harihara (half Viṣṇu, half Śiva) and Ardhanārīśwara (half Śiva, half Pārvatī) directly precede the juncture between the two larger temple-within-a-temple galleries and the more narrow programmatic displays on floors above. Could the joined figures reflect an intentional pun with the joining of two parts of the building? These two examples of iconographic and architectural metaphor reflect the types of choices architects and patrons must make. The tower interior remains as a remnant of those intentions. On entering the tower, one views a sculpture of Viṣṇu (fig. 3.8). He is easily recognized by his crater crown and holds a discus in one hand and a club in the other. Two of his arms are missing, along with a portion of his legs, yet the pedestal on which he sits seems to have all of its original form intact. A closer look reveals pitted accretions in Viṣṇu’s eyes, nose, and mouth—to an extent where the flattened traces of features become barely legible. In contrast the incised crisscross

Temple as Royal Abode

pattern of the crown remains incredibly intact, as if it had been completed yesterday. Tourists may assume “Muslim invaders”20 attacked the sculpture. In an a priori narrative of iconoclasm a German tourist asked me during my fieldwork inside the tower, “But didn’t you notice that all [emphasis mine] the faces have been destroyed?” During a public lecture, similar questions arose about this very first image in the tower. Could he be proof of Tod’s misreading of the Ain-I-Akbari and his conviction that the tower engages in an intentional victory narrative? If so, one could project a narrative of victory and revenge onto the material cultural remains—but how much of this story derives visually from the remains at hand?

chapter 3

The weight of arms that extend from a bas-relief outward far from the picture plane could easily cause a large chunk of stone to fall over time. The stone had no metal reinforcements to sustain its weight over a long period. As for the face, one would imagine a marauder scenario would involve a club, a sword, or some other tool that would be used to bludgeon a “heathen” god’s delicate stone face. I find it highly unlikely that someone with iconoclastic aims would delicately chip into a face leaving faded features rather than slash an image in half or chip off a chunk of nose. So the visual evidence that looks more like erosion does not support this assumed narrative. Furthermore, recent studies of looting, power, and display of sacred spoils suggests that display in the capitals of the victors or ritualized travel of icons across specific routes followed standardized practices, meant to establish overlordship and hegemony more than to express a universal iconoclastic disgust for the figural image.21 According to the construction date of 1440 to 1460, the sculpture would have to have been destroyed subsequently—but the next capture of Chittauḍgaḍh unfolds in the late sixteenth century, when Akbar wins the fortress in 1567. This Mughal emperor was known for a multisectarian thirst for knowledge. He built the famous kitab-khanna for his capital of Fatepur Sikri. A dyslexic who never learned to read, the powerful Indian-born emperor was an intellectual who sponsored a number of famous illustrated manuscripts. From the “Hindu” Bhagavata Purāṇa to the Persian-style homage to his own reign, the Akbarnama, the sponsored works of the Mughal emperor included rich figural imagery and some of the most impressive Indian painting ever produced.22 Again, I find it hard to imagine his conquest of Chittauḍgaḍh as a blind attempt at in situ iconoclasm. The sculpture could not have become spolia if it still remains in situ. Moreover, Akbar’s sponsorship of so many figural works as an intellectual, an art lover, and a powerful patron makes him an unlikely candidate for iconoclastic-style conquest. Rather than confuse the tests of time with an a priori and ahistorical narrative, we can begin instead with Kumbhā and his architects as they began their impressive architectural project. Not only did their architectural manual specify the kīrtistambha tower type as dedicated to Viṣṇu, but we can imagine as a Vaiṣṇavaite, this choice would have appealed to Rāṇā Kumbhā as a patron. But was this personal deity (fig. 3.8) the icon of the royal tower? If we return to the question of placement, it does not make sense. A main icon normally resides in a sanctum. A main icon unfolds only at the end of a complex iconographic circumambulation. A main icon is housed in a sanctum and approached through a series of pavilions. This Viṣṇu, with fleshy pectorals and a sensual medieval stomach, greets the viewer but does not occupy a position where ritual respects could be paid properly. The sculpture reads more like exterior deities than like a central icon. From the famous eighth-century Śiva liṅgaṃ of Kalyanpur to the tenth-century icon of Śri Ekliṅgjī, most important icons are made from shiny black stone and not the same material as exterior walls. The lack of a sanctum, direct approach, scale, lack of elaborate

Temple as Royal Abode

framing, and materials suggest that this Viṣṇu sculpture served to welcome the viewer rather than as the central icon of a singular cohesive religious narrative. The turn-square staircase leads to a narrow circumambulatory path on the first floor, where a predictable iconographic triad of Viṣṇu, Śiva, and Brahma unfolds on three exterior walls. To fully view the large figures, one must climb back into the stone window seats and enjoy the “achi hawai” (pleasant breeze). From this position the walls look like a typical Gupta temple such as Deogarh,23 where one narrative scene graces each wall without any auxiliary figures or other adornment (fig. 3.9). In monoscenic narrative this Vaiṣṇavaite story is told (fig. 3.10).24 The walls of this first-floor temple-within-a temple gallery remain unadorned and free of typical medieval auxiliary figures. At the time when the Kīrtistambha was built,

chapter 3

one would expect to find leonine vyāla figures, beautiful maidens, and guardians of the corners—possibly even duplicated in two registers. The singular sculptures seem to quote a past idiom, though not verbatim. In contrast with the narrative mode of display for Viṣṇu’s mythology found at Deogarh, the first floor of the Kīrtistambha displays three carefully labeled deities who remain in nonnarrative, iconic poses and have typical attributes. Not only does the sculpture-to-wall relationship suggest a quotation of the Gupta period architectural style with a single deity per wall, but the deities are labeled. Why, in fifteenth century Mewār, would anyone—architect, patron, priest, sage, servant, visitor, or drummer—need a one-word label to explain that Śiva holds a

Temple as Royal Abode

skull and trident? Even hundreds of years later, one need only take a course in basic iconography or read an introductory text on Indian art to know, without a label, that Śiva holds a trident and a skull, whereas Viṣṇu holds a discus and a club. Of course, for more complex iconographic representations the modern viewer might turn to Rao’s Elements of Hindu Iconography, but the labels do not indicate an obscure manifestation of three familiar gods but rather “Viṣṇu Narayana, Mahā-Śiva, and Brahma” tout court. The turn-square staircase leads to the second of two temple-within-a-temple galleries and carries the viewer five hundred years into the future. From the Gupta architectural quotation on floor 1, floor 2 reflects typical Guhila dynasty architectural style. Thickly textured with deep recesses and protrusions, and punctuated with auxiliary lion, maiden, and guardian figures, the temple wall on display could almost be dated to the second half of the tenth century, if it were not for the fifteenth-century sculptural style (fig. 3.11). The extra framing sculptures associated with the early medieval period served as alaṅkāra (ornamentation) but also correlated to specific placements in relation to the central icon. These were not emanations from a vastushastra grid, as Stella Kramrisch may have imagined. To the contrary, the auxiliary sculptures served to encode architectural technology in aesthetic terms. The figures were not part of a narrative program or a mythological relationship to the central icon. They may have served as a celestial court,25 or

chapter 3

palace entourage of sorts, yet they remain a part of a strict system of architectural placement. The guardians stand at each corner of the building, apotropaically protecting the precarious seams of the building. Darielle Mason has demonstrated that the maidens and the lionine figures also correspond to specific sections of the wall.26 The subsidiary sculptures thus correspond to subsidiary projections— the very projections and recesses that make possible the height of soaring North Indian temple spires of the medieval period. Although the second-story temple wall does not support a spire, it does provide a thick base for upper stories. The turn-square staircase continues inside these thick walls where an inner main icon would normally be found in a temple sanctum (fig. 3.12). From this inner staircase one emerges through a doorway onto a set of three smaller galleries stacked one upon the other. Nondual deities celebrate the juncture of the larger initial floors—where entire temple walls seem to be reproduced—with the upper galleries. The sūtradhāra (scribe) has carefully etched Harihara in stone at the base of a figure that sports a crown on one half of his head and an ascetic’s dreadlocks on the other. Half Viṣṇu and half Śiva, Harihara carries attributes typical of Viṣṇu—such as the discus—on one side and attributes typical of Śiva—such as the trident—on the other. Harihara is not that unusual a deity, but he was rarely if ever represented on temple exteriors in the position of a main wall projection, or bhadra. On liminal floor 2 the reference to nonduality cannot be mistaken with the presence of Ardhanārīśwara on the

Temple as Royal Abode

Figure 3.13. Ardenareśvara (Half-Śiva/Half-Pārvatī), interior Kīrtistambha tower, c. 1440–60, Chittauḍgaḍh. © Deborah Stein.

main projection of another wall (fig. 3.13). A bilateral representation half Śiva and half Pārvatī, the universally recognizable female breast of Pārvatī occupies half of the body and the male pectoral of Śiva the other. Elegantly coiffed hair meets an ascetic’s locks, and heavy jewelry gives way to a lightweight renunciant’s brahmin cord. Nondual deities visually depict the liminal state of betwixt and between, neither fully one thing nor the other or, alternatively, both.27 Just as any bilingual person might read newspapers in two languages to understand the real news lies between the lines—often in the truthful story left partially or wholly untold in both languages— the nondual deities reflect a philosophy about what is beyond binary categories.

chapter 3

As one emerges from the doorway of the inner staircase onto the third floor, one clasps the columns of the doorframe, ducks one’s head, and stands facing westward. Right above the place on the column where one’s hand would naturally clasp, the calligraphic form of Allah is carefully sculpted in relief (fig. 3.14). Above these calligraphic sculptures of the Muslim god’s name, small symbols of the architectural plan of a mosque with mihrab are carved deeply into the stone. In contrast with the incised labeling of the other iconography, these verbal and symbolic elements are treated as iconographic sculpture. Looking west, toward Mecca, forced to bow one’s head because of the height of the doorframe, and hands naturally clasping the sculptural calligraphy of the Muslim God’s name, any person who kinesthetically navigates the stairwell inherently includes Allah in Kumbhā’s world. The fourth floor departs from the deities of the earlier levels to include a portrait of the architect and his sons, servants, and other human actors on the interior, whereas the exterior of this level is covered in Pupate Śaiva iconography encasing this representation of all of fifteenth-century human Mewār. The sixth and seventh floors build on the deities and humans of Mewār with a free departure into metaphysics. The sixth floor is populated by goddesses, sa-guna (with form), whereas the seventh floor is left empty with jati lattice windows into a gallery of nothing, nir-guna (without form). The ascent culminates, remarkably, with this abstract philosophical reference to the nonduality of form and formlessness, before transporting the viewer up the steps to the light-laden gallery of the observation deck, surrounded on all sides by elaborate lattice windows. There are many architectural precedents for famous towers and minarets in South Asia: Ashokan columns, the Iron Pillar in Delhi, the Minaret of Jam in Afghanistan, the Qutab Minar in Delhi, and even the Jain kīrtistambha built in the same fortress as

Temple as Royal Abode

the “Jayastambha” only a century or so prior to the one filled with sculpture built by Kumbhā. In the Maurya era these columns were inscribed with multilingual edicts that praised nonviolence from a Buddhist perspective and sought to unite the empire through a common religious perspective and state code of ethics across a very broad territory uniting most of North India. The Iron Pillar, in the same compound as the Qutab Minar built subsequently, also served a political purpose as a marker of Gupta power in tandem with a dedication to the Hindu god Viṣṇu. Both the Minaret of Jam and the Qutab Minar were Ghurid monuments in summer and winter capitals, serving as loudspeakers for the Muslim call to prayer, as well as marking these capitals with tall, regal monuments to be seen far and wide. Up close, both revel in geometric and calligraphic relief, as well as—in the case of the Afghan tower, so close to the lapis lazuli mines—incredible ceramic tile work with blue glaze. The Jain tower in Chittauḍgaḍh, Rājāsthan, that precedes Kumbhā’s Kīrtistambha follows a multisectarian patronage pattern found in other nearby cities such as Jāwar, where Jain patronage follows a mercantile success, often rooted in the exploitation of natural resources by tribal people, harnessed, financed, and traded across Jain networks, and then finally recognized, claimed, and established as royal centers by Hindu Rājput rulers and their direct noble relatives. So if we look back on this very brief history of the tower in South Asia prior to Kumbhā’s Kīrtistambha, we find a multisectarian history of towers in state capitals, often inscribed with religious and/or political texts, increasingly large and impressive over time, quoted often from one dynasty to the next—even across sectarian lines. Given these commonalities, what is so special about a royal tower in the fortress of a Mewāri capital from 1440 to 1460, given that stately towers had been around North India since the Maurya Empire centuries prior and had been used across capitals on a grand scale by the Ghurids long before the Sisodias took back the fortress of Chittauḍgaḍh to turn it into a royal capital of Mewār? Although the Qutab Minar has an internal staircase that winds circularly from a wide base to a narrow top, the Kīrtistambha is the first of its kind to have a turnsquare plan without narrowing at the top. None of these precedents had an internal turn-square staircase, with the exception of the Jain Kīrtistambha in Chittauḍgaḍh (fig. 3.15). Furthermore, dark, narrow, and tiny, this internal staircase is more of a precarious stepladder that climbs steeply and blindly to the expansive 360 degree views from the top gallery across all the plains that surround the plateau on which the fortress of Chittauḍgaḍh sits relatively protected from invasion through its natural geographic features. Kumbhā’s Kīrtistambha takes this technology so much further, where the turn-square staircase becomes a relatively roomy gallery winding around a central column, at times tucking itself under a story to rise to the next level. It is the first of all these famous towers to provide an inner passage lined with thematic collections of sculptural iconography in a set program. This is a radical pictorial, technical, ideological, and political invention that brought the entire cosmos into visual dialogue in an entirely new way.

chapter 3

Kumbhā’s Kīrtistambha collection functions as an archive, a database, and a matrix, although it may not exactly have been planned as one. The Kīrtistambha engages in a particular form of display thanks to the turn-square internal staircase. This creates a stacking of iconographic programs that can lead upward or downward. Each level can be circumambulated in two directions. This creates an architectural matrix of sorts that displays iconographic programs in relation to each other in a fixed set of nonlinear relationships, like a database or an archive. A database is a collection of information arranged for ease and speed of search and retrieval—in this case sculptural, representational, iconographic, and

Temple as Royal Abode

philosophical information—that is normally structured and indexed for user access and review. Databases may exist in the form of physical files (folders, documents, etc.) or as digital files (which combine to form data-processing systems). In this architectural database we have both “physical files” in the sculptural forms and a “processing system” in their architectural programmatic relationships. Meanwhile, a matrix has an even more mathematical definition as a rectangular array of numeric or algebraic quantities subject to mathematical operations, or a rectangular array of elements set out in rows and columns, used to facilitate the solution of problems, such as the transformation of coordinates. One could stretch to imagine circumambulation forward and backward in horizontal space on each floor as the rows, whereas the technical feat of the turn-square staircase exists as the columns—where viewers can travel up or down. In computer science a matrix involves computing rectangular arrays of circuit elements usually used to generate one set of signals from another. How do the signals cross in a stack of interrelated temple walls, deities, and human agents—from architect to servant? Last, a correlation matrix—a matrix giving the correlations between all pairs of data sets—suggests the tower may serve as a matrix in that it creates and encourages relationships through the juxtaposition of different sculptural data sets, sets that almost never inhabit the same structure but that usually lie a significant distance from one another. An archive is a place or collection containing records, documents, or other materials of historical interest. It is a repository for stored memories or information. For example, the Photographic Archive of the American Institute of Indian Studies (AIIS) is available online. Much has been written on both the archive and the photograph in relation to colonial projects in India, but Vidya Dehejia gets at the essence of the photograph itself. She argues that, “The photograph, like the footprint, is treated as an actual ‘trace’—an artifact of the scene it reveals.”28 Building on Roland Barthes’s idea, she explains how “we look straight through the photograph, ignoring its status as a signifier, and seeing only the signified—the image itself.”29 What if we try to imagine a sculptural iconographic archive instead of a photographic one? Would a sculpture of a servant, a named architect, or the goddess Saraswatī act more like the signifier or the signified? To what extent does the label cast the image as a representation, as opposed to an actual deity? Can curatorial display and puja coexist today, and did these two different ways of seeing and being seen coexist in the fifteenth century, when this labeled sculpture was planned? Are the sculptural bas-reliefs that we find on the first floor of the Kīrtistambha depictions of Viṣṇu or icons of Viṣṇu? Is it a portrait of a deity we are seeing or the god himself who stares back at us? Perhaps that intersubjective question of reception lies in each individual viewer regardless of time or place. Could the multiplicity that unfolds from this unique display of iconography suggest that the text that prescribed this particular regal form of a tower to encompass the entire universe foreshadowed the

100

chapter 3

simple lineages put forth in the tenth century.42 Kumbhā brought Vagada under Mewāri control in V.S. 1498 (1441 CE) and forced the mahārawal of Dūṅgarpur to surrender Jāwar to Mewār and to submit Dūṅgarpur to the overlordship of Mewār.43 This mahārāṇā was powerful enough to repair damages done to Ekliṅgjī while the Guhila clan had taken refuge from Nāgadā at Kumbhalgarh.44 Kumbhā is best known for two records composed during his reign: the inscription of the Kīrtistambha and the inscription at Kumbhalgarh. Akshaya Keerty Vyas attributes the contents of the two inscriptions to the Ekliṅgamāhātmyam.45 After praising Gaṇēśa, Saraswatī, and Ekliṅgjī, the inscription continues to invoke many deities “such as [Lamboda]ra, Gajamukha, Vindhyāvāsinī, Ekliṅgjī, Pinakin, Ina, and others.”46 The text also describes many important geographical sites in Mewār, including “the range of hills naturally formed into a triangle within which is situated the temple and town of Ekliṅgjī,” as well as “the goddess Vindhyāvāsinī, whose shrine is situated on the slope of the hill to the north outside the rampart around Ekliṅgjī’s temple.”47 Verses 23 and 24 give a description of the history of the religious compound of Ekliṅgjī, attributing its founding to Bappa Rāwal. It was destroyed by “Taruṣkas” and later repaired by Mahārāṇā Mokal, who furnished the

Temple as Royal Abode

101

rampart. Kumbhakarna “repaired” the ancient shrine, and Raimal is credited for the modern structure, laying the foundation and erecting a new structure. Lines 25–28 credit Bhojabhupa with the formation of Indira Sagar, the pond behind the temple.48 Mokal is credited with the creation of Vāghelava Lake (Bhāghelā Tālāv) in memory of his brother, Bhaghasimha (fig. 3.18).49 Built under Kumbhā’s successor, Mahārāṇā Raimal, the construction of the Śri Ekliṅgjī temple dates to this era as well. The roof of the structure (fig. 3.16) shares much with the roof of the Samiddhēśvara temple at Chittauḍgaḍh (fig. 3.17), yet the structure of the maṇḍapa (see fig. 0.2) visually recalls the twelfth-century Deo Somnāth temple in Vagada. Whereas the Kīrtistambha makes meaning explicit by carving semantic labels under every sculpture from the servants to the architects to the gods, the architectural quotations such as those cited above are no less intentional. Sompurā masons constructed these three buildings; their desire to archive, to make permanent, and to create history in the fifteenth century is corroborated by a collection of fifteenth-century manuscripts held by an architect who claims he is the twenty-second Sompurā descendant from Maṇḍana himself. During the 2002 Navratri celebrations at Jagat, Manish Bharadwaj considered the possibility that the Ambikā temple may well have been the kūldevī shrine of his line of Sompurā masons. Whether or not the Ambikā temple had originally been conceived with this in mind, the fact that a Sompurā descendant was considering

108

chapter 3

Rāṇā Kumbhā’s daughter, Ramabai, was the patron of the beautiful stepwell and Vaiṣṇavaite temple in her name in the mining town of Jāwar. Last but not least, Padmini, who self-immolated with all of her ladies at Chittauḍgaḍh centuries before, may well be the most famous woman in Indian history. Although traces of her story remain in the multiple overlapping mythical accounts so carefully studied by Ramya Sreenivasan, the architectural remains of her act have made Chittauḍgaḍh a site of nationalist pilgrimage in her name. Padmini, Mīrabai, and Ramabai each reveal the hegemony of heritage in Rājāsthan. Here I explore the record in stone to find that whereas women held agency as patrons and poets in the fifteenth century, prior to that time it is the tale

Temple as Royal Abode

109

of a climax of rupture—the end of a lineage and its precarious escape—that holds the most hegemonic grip on the modern imaginings of medieval Rājāsthan. With a rooftop similar to the Śri Ekliṅgjī temple, the Mīrabai temple within the same complex postdates the Śri Ekliṅgjī temple by less than a century (fig. 3.21). The architectural plan consists of a slightly taller version of what one might expect of a tenth-century Guhila temple from Mēdapāṭa, such as the Śivēśvara temple found just a few steps away. A one-story maṇḍapa is joined to a single-register iconographical plan around the three outer walls of the sanctum. The quintessentially fifteenth-century Sompurā roof above the maṇḍapa contrasts with the oversimplified śikhara (spire) when viewed from the front. When viewed from behind, the single register appears anything but tenthcentury as it bursts forth in the exuberant aediculation known to experts as śekharī style—a form that consists of multiple projections that can more easily be discerned in the roof than on the walls (fig. 3.22). The style of this temple and its proximity to the Śri Ekliṅgjī temple suggest something about the desires of its

110

chapter 3

patron. This temple, named after the saint and poet who refused to marry out of her monogamous love for Krishna, is dedicated to Viṣṇu. Unlike the Guhila temple dedicated to Viṣṇu at Īswāl, the Ramabai temple dedicated to Viṣṇu at Jāwar, and even the most famous Gupta-era Viṣṇu temple of all—the Daśāvatāra temple at Deogarh—this temple does not follow a pañcaratha plan. The extreme marriage of form and function between this specifically Vaiṣṇavaite architectural program of a central shrine with four corner detached subshrines suggests that a break from that convention may indicate a different form of ritual or philosophy. Why, then, was the Mīrabai temple placed so carefully next to the Śri Ekliṅgjī temple, in this style, at this time? Mīrabai was not just a Vaiṣṇavaite—she was known locally as a Mewāri Vaiṣṇavaite. Could it be that she was considered a patron saint of Mewār? If so, perhaps her temple was more about the cementing of Mewāri dynastic power than about providing an active space of Vaiṣṇavaite worship (either pañcaratha, as was common in the fifteenth century and earlier, or in a two-story congregational building with the potential space inside and out to sponsor dance rooted in increasing expression of bhakti from the fifteenth to seventeenth centuries). Many textual and religious studies scholars have translated Mīrabai’s poetry and examined her life’s story in great detail. Building on their work, how does one fit this specific temple into a specifically Mewāri construction of post-Kumbhā pride? In this building we find more of the origins of contemporary Sisodia dynastic aspirations than the desire to create an active theological building. Those two

Temple as Royal Abode

111

Figure 3.24. Viṣṇu icon placement, deeply recessed, in fifteenth-century temple wall, Mīrabai temple, Ekliṅgjī. © Deborah Stein.

functions are not mutually exclusive, of course. If we look at the exterior niches, which correspond axially to the inner icon, we find three forms of Viṣṇu. In pradakṣinā circumambulatory order (clockwise), the second back wall iconographic representation displays twelve hands with typical Vaiṣṇavaite iconography including the discus and the conch shell (fig. 3.23). Viṣṇu’s three heads are crowned by his traditional crater-shaped crown and are backed by an elaborately carved halo. He is seated in a posture of royal ease atop, one would assume, his vehicle, a garuda, despite the physique of a runner rather than the typical winged depiction of this magical bird. If we step back from the temple, we see the Viṣṇu icon of the third side within a niche (fig. 3.24). If you look at the temple frontally, without moving your body, you see the icon flanked by two surasundarī figures (celestial maidens), who are in turn flanked by dikpālas (guardians of the corners)—just as one might expect in a tenth-century Mēdapāṭa region temple. In the fifteenth century the projections of the bhadras come much, much farther from the wall, and the sides of the niche also have sculpture at a perpendicular angle. Each surface of each protrusion has a sculptural outcropping. This changes circumambulation. Deeply enshrined icons remain in the shadows of their niches as a richly ornamental temple wall unwinds

112

chapter 3

around them. Kinesthetically, this high medieval temple seems to push the viewer in a serial circular movement with three points of punctuation. As we will see in following chapters, the tenth-century temples use the guardian figures as much more than framing devices. They use those figures to manipulate the viewer’s gaze to preview, view, and review the main icon. The original syncopated circumambulation, which I argue is akin to the sonic resonance of a pūjā-paddhati, gives way to serial circumambulation. This is another sign that these later temples may in some ways intentionally copy earlier architecture even while they function in a kinetic manner similar to contemporary temples and different from the tenthcentury antecedents they mirror. The temple copies aspects of tenth-century Mēdapāṭa temple programs but does not retain the kinesthetic functionalities typical of that era. As I have mentioned, the temple sits right next to the Śri Ekliṅgjī temple, as if these high medieval stone tributes to the ruler of Mewār and Mewār’s female patron saint could sit side by side as revivalist tributes to Sisodia glory, envisioned as historical continuity with Guhila dynastic glory at the location where their sect, the Pāśupata-Śivas, won a theological debate in their territory five hundred years earlier. A second temple, together with a more vernacular piece of architecture— a communal water tank—was also dedicated to Viṣṇu. This time in the form of Rameśvara, the temple’s main icon pays tribute to King Kumbhā’s daughter, Princess Ramabai. It was she who inherited Jāwar as part of her dowry and she who was the patron of this temple and tank (fig. 3.25). This Hindu Viṣṇu temple plan suggests links with other Viṣṇu temples to the north. The Ramanatha temple follows a pañcaratha plan similar to the tenth-century Viṣṇu temple at Īswāl and

Temple as Royal Abode

113

the seventeenth-century Jagannātha temple in Udaipur. Literally translated as a five-chariot temple, a pañcaratha plan consists of a main temple in the center of four smaller subshrines. This architectural pattern is generally associated with Viṣṇu, as seen in the famous Gupta-period Daśāvatāra shrine in Deogarh, Madhya Pradesh (fig. 3.9). In contrast to the majority of architectural stylistic features— where regional style trumps any sectarian orientation—the pañcaratha plan seems to span a long period and wide geographic area, possibly owing to a specific mode of Vaishnavism. Unlike the sectarian necessities of the basic Vaiṣṇavaite pañcaratha plan, the central icon of Viṣṇu in the form of Ramanatha definitely reflects regional style and choice of materials. Black schist was a common material for medieval icons in this region (fig. 3.26). From the eighth-century four-faced Śiva liṇga of Kalyanpur to the south to the fifteenth-century gigantic Lakulīśa icon at Ekliṅgjī to the north, the highly polished, shiny black stone signals a material reserved for special icons placed within inner sanctums. In neighboring Jagat the tenth-century black icons of the goddesses Cāmuṇḍā and Mallar Mātā provide the most geographically close examples of this medieval phenomenon—a trend found in Śaiva, Śakti, and Vaishnava icons alike. The style of carving is also quite similar among these icons and differs from the fastidiously chiseled precision displayed in the ornamentation

114

chapter 3

of exterior walls. The meticulously carved sandstone, quartzite, or marble sculptures of the exterior display more elongated features than their Gupta counterparts and are characterized by a move away from volume toward a celebration of line. In contrast, the black icons of the inner sanctums in medieval Mēdapāṭa—including the Ramanatha icon of Jāwar—suggest a rudimentary folk style with an interest in basic forms rather than ornate ornamentation. And yet this “folk” style, for lack of a better term, was shared by rural and urban alike and was patronized both by those whose history remains unwritten and by nobility. The Ramanatha icon falls in the category of icons sponsored by nobility. A female patron illuminates the political position of the Ramanatha temple within medieval Mewār since she was the daughter of the infamous Rāṇā Kumbhā. Ramabai’s father was a great builder, as well, and a patron of the arts. Though the geographic area of his rule was often tiny, shifting stretches of his cultural prowess were expansive and impressive. He was obsessed with a revivalist desire to canonize the artistic feats of his lineage and to become a steward for future generations. Kumbhā wrote an erudite treatise on Indian music and the aesthetic theory of rasa, and he sponsored the Kumbhalgarh fort as well as the Vaishnava Kīrtistambha (mistakenly known as the Jayastambha, or Tower of Victory) at Chittauḍgaḍh.66 The Ramanatha temple provides an important and noble female patron, a precise date, a clear geographic location, and a sectarian temple with a meaningful pañcaratha plan and an icon in situ in a single architectural example. Perhaps the most official and important of architectural projects found at Jāwar, the Ramanatha temple, can be precisely dated thanks to a 1489 CE inscription. According to this inscription, Kumbhā’s daughter sponsored the Ramanatha temple and tank since Jāwar was part of her jagīr.67 Can we then envision the architectural relationship between the Ramanatha temple and Kumbhā’s projects as a mirror of political relationships between kings and daughters, fathers and sonsin-law, rulers and the noble elite, women and their power as property owners—or alternatively, women as property tied to lands, holdings, and wealth? Certainly we can glean that a marital alliance established a noble Rājput presence at the site of Jāwar at the close of the fifteenth century. Whether that fact is understood as a signal of Jāwar’s prominence and wealth or relative unimportance merits further investigation as we learn more about the history of gender and property in fifteenth-century Mewār.68 Ramabai’s inscription appears at the entrance to a large tank, which adjoins the Ramanatha temple and its four subshrines. This large pool of water would have provided a state-sponsored civic space for the cool purification of water under the powerful gaze of the Ramanatha (Viṣṇu) icon.69 Mahārāṇā Kumbhā’s daughter’s inscription—located at the entrance to a space designed for the congregation of the public—suggests the desire to control, celebrate, and take credit for the economic success of the zinc mining and the rich social fabric that had grown around this natural resource.70 From the initial construction of industrial projects to the

Temple as Royal Abode

115

Figure 3.27. Palace where the historical Rani Padmini resided within the fortress of Chittauḍgaḍh, picturesque view from within the domestic interior of the medieval palace onto a structure rebuilt in the 20th century in the middle of the lake. © Deborah Stein.

architectural phase of thanks for the rewards of that industrial endeavor, the construction of the nobly sponsored Viṣṇu temple signals the creation of communal centers of social exchange. A tank forms the heart of a village and indicates the growth and importance of the sacred center with the expansion of the zinc industry. The third and last piece of architecture dedicated to a woman is a site of pilgrimage in person to this day (fig. 3.27).71 The palace of the infamous Rani Padmini at Chittauḍgaḍh may be the domestic architectural remnant of one of the most potent historical events in India. Chronicled in multiple accounts over the fifteenth through the nineteenth centuries, Rani Padmini is known as the Rājput queen who committed jauhar (ritual suicide) along with all the palace ladies rather than fall into the hands of Alāuddīn Khilji, the Sultan of Delhi. Ramya Sreenivasan has traced the trajectory of Padmini’s story as it was told in 1540 and rewritten over time in different parts of India.72 In contrast to these centuries of old bardic tales, the physical location of Alāuddīn Khilji’s actual 1303 siege was the fortress of Chittauḍgaḍh. Could this “palace” next to the tank be the physical site of a jauhar led by Ratan Singh’s wives? The palace was the residence of Rani Padmini, and not the location of the jauhar, which took place at the jauhar kund. Rani Padmini

Figure 3.28. J. W. Caplain, The Water Palace of HH Rani Padmawati, Chittauḍgaḍh. Albumen print. Contact printed from glass plate negative, 1865–1885 CE. Accession No. 2009.10.0096-00013_R. Image courtesy: The City Palace Museum, Udaipur, © MMCF.

is mentioned in Veer Vinod Part-1, which is the official historical chronicle of Mewar.73 This structure, which looks significantly more modern (it was renovated in 20th century) than c. 1303, serves nevertheless as a site of pilgrimage. The historicity of the Mewāri Queen’s life in contrasts greatly with the artistic depictions subsequent to her time and produced outside of Mewār, each of which reflect above all the time and place where they were made. The bardic tale and “memory” dates historically to 1540 when Malik Mohammad Jayasi wrote the famous poem “Padmawat”. In circulation during Akbar’s reign, one can assume this avid patron of illuminated books and those in his karkhana workshops would have been aware of this story. Moreover, Jauhar is depicted in the Victoria and Albert Museum’s Akbar Nama of c.1590–5 as one of the illustrations of Abu-Fazl’s account of a Mughal siege of Chittauḍgaḍh in 1568, but this Mughal illuminated manuscript does not reference a time when Rani Padmini was in residence at her palace at all and, in fact, postdates the historical dates of Alāuddīn Khilji, the Sultan of Delhi by well over two centuries.74 By the nineteenth century, when Ravi Varma was producing his famous oleographs, and writers such as Tagore and others were writing about Padmini in literature, the focus in this colonial era had

Temple as Royal Abode

117

pivoted from jauhar to nationalism as evidenced in an image entitled “Padmini or Lotus Nymph”—where a female figure is transformed into an allegory of the Indian nation incarnate—her pink sari depicting a map of India in a representation that seems to resemble France’s personification in Delacroix’s 1830 painting of “Liberty Leading the People” more than any direct reference to the historical Rani Padmini of Mewār, her life in early medieval Chittauḍgaḍh, or her palace as pictured here in an archival albumen print from Mewār taken in c. 1865-1885 (Figure 3.28). SE L F- FA SH IO N I N G M O N UM E N TA L I T Y I N T H E WA K E O F DY NA ST IC RU P T U R E

The mid and high medieval periods in Mewār witnessed an efflorescence of selffashioning through the construction of new monuments at old sites of numinous and political power. Heralded earlier in the thirteenth century by a shift in the Guhila origin myth, the fifteenth and sixteenth centuries locate the first instances of self-fashioning through the self-conscious use of tenth-century historical monuments and inscriptions. Whereas the tenth century was a time of solidification of power and the construction of dynastic hegemony through monuments and inscriptions, the fifteenth century could build on more than the seeds of dynastic power or bits of lineage in a void of architectural evidence in stone. True monumentality came when monuments could be constructed next to buildings dating to centuries earlier. Fifteenth-century Sisodia monuments referenced tenth-century Guhila projects through an intentional spread of a new Sompurā-based dynastic style, the labeling of iconography established centuries earlier, and their location on the site of earlier buildings and inscriptions. Like thinking about thinking, these were monuments about monuments. The impact of statehood on Ekliṅgjī and Jagat echoes this process of reification through architectural campaigns yet with references dependent more on the fifteenth-century monumentality than tenth-century remains. The ritual and politics of Ekliṅgjī today continue the tradition of inventing Mewāri identity by defining the center with monuments. It is still the royal family of Mewār who ensures that the archaeology of their family is not read merely as dead history. In the present, Ekliṅgjī serves to define postcolonial kinship most of all. Regardless of Udaipur’s relationship to Delhi, either in the seventeenth or in the twenty-first century, Ekliṅgjī’s role remains constant as the divine ruler of Mewār. Like many Rājput families, the descendants of Mewāri royals have turned their attentions to the hotel business, turning their royal residences into commodities for tourist consumption. As we have seen in this chapter, however, the manufacture of heritage may no longer pass for such a modern pursuit. In the fifteenth century already, with the labeling of the Kīrtistambha iconography Kumbhā and Jaita took

118

chapter 3

Figure 3.29. Queen Elizabeth II of Great Britain and Prince Philip on their visit to Udaipur, January 1961. Press Information Bureau, Govt. of India. Gelatine silver print. Accession No. 2009.09.0296-00001_R. Image courtesy: The City Palace Museum, Udaipur, © MMCF.

an encyclopedic approach to capturing history and freezing meaning in stone for future generations. It worked, as we can see in figure 3.29, a photograph of the presentation of a silver miniature of the tower presented to Queen Elizabeth on a royal visit to Udaipur. The ritual and politics of Jagat today reflect an age-old struggle for power by politically disenfranchised populations on the periphery of dynasty, empire, and state. The increase of sūtradhāra interest in the seventeenth and eighteenth centuries suggests Jagat had already taken on an art historical quality as an exemplary piece of architecture. The last inscriptions in Jagat date to the early eighteenth century. James Tod makes no mention of Jagat, nor are there any inscriptions dating from the nineteenth century onward. The next reference to Jagat is in the form of R. C. Agrawala’s “discovery” in 1957 and a few 1950s photographs in the Archaeological Survey of India photography archive. In the twentieth century most of Jagat’s nobles moved to Udaipur to live an urban existence, gradually transforming their court lifestyles into the hotel businesses after independence, as many powerful Rājput families have done in postcolonial India. This vacuum of power back in the village may echo the period between the fifteenth and sixteenth centuries, when the local tribal peoples lived fairly independently and free of any clear dynastic power—Rājput, Mughal, British, or national. The difference between

GENERAL

The Hegemony of Heritage Ritual and the Record in Stone Deborah L. Stein 318 pages, 82 colour photographs 71 black & white photographs and 1 map 6.75 x 10.60” (171 x 260 mm), plc ISBN: 978-93-85360-61-9 ₹1499 2019

The Hegemony of Heritage makes an original and significant contribution to our understanding of how the relationship of architectural objects and societies to the built environment changes over time. Studying two surviving medieval monuments ´ Eklingj ´ ˙ ı̄ Temple comin southern Rajasthan—the Ambik a¯ Temple in Jagat and the Sri ¯ ¯ı—the author looks beyond their divergent sectarian affiliations and plex in Kailashpur patronage structures to underscore many aspects of common practice. This book offers new and extremely valuable insights into these important monuments, illuminating the entangled politics of antiquity and revealing whether a monument’s ritual record is affirmed as continuous and hence hoary or dismissed as discontinuous or reinvented through various strategies. The Hegemony of Heritage enriches theoretical constructs with ethnographic description and asks us to reexamine notions such as archive and text through the filter of sculpture and mantra. “Makes visible the multiple methodologies that can be mobilized to write nuanced histories of Hindu temple architecture. The author’s approach is both refreshing and new. Skillfully weaving in postcolonial theory, object ontologies, and affect theory, among other approaches, the book opens up an exciting paradigm in the study of South Asian art and architecture.” SUGATA RAY, Assistant Professor of South Asian Art and Architecture, University of California, Berkeley

DEBORAH L. STEIN holds a PhD from UC Berkeley and has

taught at Mills College, UC Berkeley, UC Irvine, UC Santa Cruz, and San Francisco State University. Currently Senior Lecturer at California College of Arts, she is the author of several peer-reviewed articles.

ISBN 978-93-85360-61-9

Cover illustration: Pūrṇaghata motif next to woman with overflowing pot on her head (see p. 212)