After Timur Left: Culture and Circulation in Fifteenth-Century North India 0199450668, 9780199450664

Timur invaded northern India in 1398 but returned to Samarkand a year later. In 1555 the Timurid emperor Humayun came ba

940 51 18MB

English Pages 512 [470] Year 2014

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

After Timur Left: Culture and Circulation in Fifteenth-Century North India
 0199450668, 9780199450664

Table of contents :
Title Pages
Francesca Orsini and Samira Sheikh
Title Pages
(p.i) After Timur Left (p.ii) (p.iii) After Timur Left
Title Pages
(p.vii) Acknowledgements
Francesca Orsini and Samira Sheikh
(p.vii) Acknowledgements
(p.vii) Acknowledgements
(p.ix) Note on Transliteration
Francesca Orsini and Samira Sheikh
(p.ix) Note on Transliteration
(p.ix) Note on Transliteration
(p.ix) Note on Transliteration
(p.xi) Plates and Figures
Francesca Orsini and Samira Sheikh
(p.xi) Plates and Figures
Plates
(p.xi) Plates and Figures
Figures
Introduction
Francesca Orsini and Samira Sheikh
Introduction
Abstract and Keywords
Literature and Writing
Introduction
Literature as History
Introduction
Introduction
Multilingualism and Language Domains
Introduction
Introduction
Introduction
Introduction
Vernacularization
Introduction
Introduction
(p.17) Writing, Scripts and Scribes
Introduction
Introduction
Introduction
Regional Polities
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Geographies in Literature
Introduction
Introduction
(p.36) The Religious Public Sphere
Introduction
Introduction
Introduction
Introduction
Introduction
Notes:
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
Introduction
After Timur Left
Francesca Orsini and Samira Sheikh
After Timur Left
North India in the Fifteenth Century*
Simon Digby
Abstract and Keywords
After Timur Left
After Timur Left
After Timur Left
After Timur Left
After Timur Left
After Timur Left
After Timur Left
After Timur Left
Notes:
After Timur Left
After Timur Left
After Timur Left
Bandagī and Naukarī
Francesca Orsini and Samira Sheikh
Bandagī and Naukarī
Studying Transitions in Political Culture and Service under the North Indian Sultanates, Thirteenth–Sixteenth Centuries*
Sunil Kumar
Abstract and Keywords
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
The Value of Déracinés: Bandagi in the Pre-Mughal Period
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Service with Status and Honour: Negotiating Naukari
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Establishing Homesteads, Alliances, and Natality: Transiting from Bandagi
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
(p.90) Contextualizing the Fifteenth Century: Bandagi, Naukari, and Service
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Afghans between Bandagi and Naukari
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Conclusion
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Notes:
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
Bandagī and Naukarī
The Rise of Written Vernaculars
Francesca Orsini and Samira Sheikh
The Rise of Written Vernaculars
The Deccan, 1450–1650
Richard Eaton
Abstract and Keywords
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
(p.128) Conclusion
The Rise of Written Vernaculars
Notes:
The Rise of Written Vernaculars
The Rise of Written Vernaculars
The Rise of Written Vernaculars
Turki and Hindavi in the World of Persian
Francesca Orsini and Samira Sheikh
Turki and Hindavi in the World of Persian
Fourteenth- and Fifteenth-Century Dictionaries
Dilorom Karomat
Abstract and Keywords
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki in Indo-Persian Dictionaries
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Hindavi in Indo-Persian Dictionaries
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Multilingual Musical Knowledge
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Conclusions
Turki and Hindavi in the World of Persian
Notes:
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Turki and Hindavi in the World of Persian
Local Lexis?
Francesca Orsini and Samira Sheikh
Local Lexis?
Provincializing Persian in Fifteenth-Century North India
Stefano Pellò
Abstract and Keywords
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Notes:
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Local Lexis?
Languages of Public Piety
Francesca Orsini and Samira Sheikh
Languages of Public Piety
Bilingual Inscriptions from Sultanate Gujarat, c. 1390–1538*
Samira Sheikh
Abstract and Keywords
Languages of Public Piety
Languages of Public Piety
Epigraphic Language and Form
Languages of Public Piety
Pre-sultanate Bilingual Epigraphs
Languages of Public Piety
Bilingual Epigraphs on Civic Structures
Languages of Public Piety
Bilingual Edicts and the Ass Curse
Languages of Public Piety
Stepwell Inscriptions in the Sultanate
Languages of Public Piety
Languages of Public Piety
Languages of Public Piety
Bilingual Inscriptions on Water Architecture
Languages of Public Piety
Languages of Public Piety
Languages of Public Piety
‘Muslim’ Epigraphs in Sanskrit and Gujarati
Languages of Public Piety
Languages of Public Piety
Notes:
Languages of Public Piety
Languages of Public Piety
Languages of Public Piety
Languages of Public Piety
Languages of Public Piety
Languages of Public Piety
Universal Poet, Local Kings
Francesca Orsini and Samira Sheikh
Universal Poet, Local Kings
Sanskrit, the Rhetoric of Kingship, and Local Kingdoms in Gujarat
Aparna Kapadia
Abstract and Keywords
Universal Poet, Local Kings
Universal Poet, Local Kings
Hill Forts and Local Kingdoms: Champaner and Junagadh
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
The Narratives
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
The Rhetoric of Kingship
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
(p.234) Political Geography and the Importance of Place
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Conclusion
Notes:
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Universal Poet, Local Kings
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Francesca Orsini and Samira Sheikh
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Ramya Sreenivasan
Abstract and Keywords
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Hinterland Politics and Marginal ‘Courts’ in Northern India C. 1350–1550
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Classical Genres Adapted at Hinterland Courts, 1350–1550
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales in Hinterland Courts, 1350–1550
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Conclusion
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Notes:
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Warrior-tales at Hinterland Courts in North India, c. 1370–1550*
Emotion and Meaning in Mirigāvatī
Francesca Orsini and Samira Sheikh
Emotion and Meaning in Mirigāvatī
Strategies of Spiritual Signification in Hindavi Sufi Romances*
Aditya Behl
Abstract and Keywords
Introduction: Sufism in Hindavi Texts
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Critical Approaches to the Text
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Meaning and Narrative in the Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
(p.290) Rupa in Sufi Romances
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in the Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Notes:
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
Emotion and Meaning in Mirigāvatī
The Art of the Book in India under the Sultanates
Francesca Orsini and Samira Sheikh
The Art of the Book in India under the Sultanates
Éloïse Brac de la Perrière
Abstract and Keywords
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
Classifications
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The First Group
The Art of the Book in India under the Sultanates
The Second Group
The Third Group
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Fourth Group
The Art of the Book in India under the Sultanates
(p.317) The Fifth Group
The Art of the Book in India under the Sultanates
Illustrated Manuscripts and Literary Tastes
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
Illustrations in Sultanate Books
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
Illumination
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
Some Mapping and Conclusions
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
Notes:
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
The Art of the Book in India under the Sultanates
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Francesca Orsini and Samira Sheikh
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Eva De Clercq
Abstract and Keywords
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha and Jain Literary Activity in the Fifteenth Century
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
(p.347) Raidhu
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Raidhu’s Prashastis and Patronage
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Conclusions
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Notes:
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Apabhramsha as a Literary Medium in Fifteenth-Century North India
Early Hindi Epic Poetry in Gwalior
Francesca Orsini and Samira Sheikh
Early Hindi Epic Poetry in Gwalior
Beginnings and Continuities in the Rāmāyan of Vishnudas
Imre Bangha
Abstract and Keywords
Early Hindi Epic Poetry in Gwalior
Fifteenth-century Gwalior
Early Hindi Epic Poetry in Gwalior
Vishnudas’s Epics
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
The Circulation of Vishnudas’s Works
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
(p.377) A Retelling of Valmiki
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
A Translation of Valmiki?
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Oral Embeddedness
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Histories of Prosody and Language: The Chaupai and Madhyadeshi Bhasha
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Emerging Hindi Literary Culture and Vishnudas
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Conclusion
Early Hindi Epic Poetry in Gwalior
Notes:
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Early Hindi Epic Poetry in Gwalior
Traces of a Multilingual World
Francesca Orsini and Samira Sheikh
Traces of a Multilingual World
Hindavi in Persian Texts
Francesca Orsini
Abstract and Keywords
Traces of a Multilingual World
Diglossia, Bilingualism
Traces of a Multilingual World
Traces of a Multilingual World
Literary History and Contact Linguistics
Traces of a Multilingual World
Hindavi Words and Utterances
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
A Fifteenth-century Provincial Sufi and His Poetic Tastes
Traces of a Multilingual World
Traces of a Multilingual World
Poetic Bilingualism
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
(p.427) Formal Literal Equivalence
Traces of a Multilingual World
Refraction and Mirroring
Traces of a Multilingual World
Traces of a Multilingual World
Dynamic Equivalence
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Conclusions
Traces of a Multilingual World
Notes:
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
Traces of a Multilingual World
(p.471) Editors and Contributors
Francesca Orsini and Samira Sheikh
(p.471) Editors and Contributors
Editors
Contributors
(p.471) Editors and Contributors
(p.471) Editors and Contributors
(p.471) Editors and Contributors
(p.437) Bibliography
Francesca Orsini and Samira Sheikh
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.437) Bibliography
(p.477) Index
Francesca Orsini and Samira Sheikh
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index
(p.477) Index

Citation preview

Title Pages

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Title Pages (p.i) After Timur Left (p.ii) (p.iii) After Timur Left

(p.iv) Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trademark of Oxford University Press in the UK and in certain other countries Published in India by Oxford University Press YMCA Library Building, 1 Jai Singh Road, New Delhi 110 001, India © Oxford University Press 2014 The moral rights of the authors have been asserted First Edition published in 2014 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the Page 1 of 2

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Title Pages prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this work in any other form and you must impose this same condition on any acquirer ISBN-13: 978-0-19-945066-4 ISBN-10: 0-19-945066-8 Typeset in Adobe Garamond Pro 10.5/13 by SPEX Infotech, Puducherry, India 605 005 Printed in India at Sapra Brothers, New Delhi 110 092

Access brought to you by:

Page 2 of 2

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Acknowledgements

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

(p.vii) Acknowledgements THE VOLUME IS BASED ON papers presented at the conference ‘After Timur Came: Multiple Spaces of Cultural Production and Circulation in FifteenthCentury North India’ that was held at the School of Oriental and African Studies, London, on 29–31 May 2007. This was the first of a series of workshops, reading sessions, and reading groups and intense interactions that formed the project on ‘North Indian Literary Culture (1450–1650)’ that Francesca Orsini led at SOAS, funded by the Arts and Humanities Research Council (AHRC); the conference also benefited from a British Academy conference grant. We would like to first of all acknowledge the AHRC for its generous funding, and the British Academy for its additional sponsorship—the conference and the project provided wonderful opportunities for experienced and younger scholars and students to come together and develop a truly comparative and multilingual approach together. We also thank Preeti Khosla for the visual materials she brought to the conference, and we are glad that we could include Eloïse Brac de la Perrière’s essay in this volume. We wanted the conference to do for the fifteenth century what Simon Digby had done in his magisterial synthesis, ‘Before Timur Came: Provincialization of the Delhi Sultanate through the Fourteenth Century’ (2004); our title pays direct homage to his essay. We were indeed fortunate in having the late Simon Digby as our keynote speaker, and we only wish that we could have continued to draw upon his (p.viii) unmatched knowledge of all things fifteenth century, but tantalizing references to perfumes, Naths, and precious stones in his talk could not be followed up after his sad demise. It is to him and to his and our friend and colleague Aditya Behl that this volume is dedicated. That their influence goes much beyond their contributions will be clear to any reader of this volume. In the sadness for Aditya’s untimely passing away, it is a source of great satisfaction that his translation of the Mirigāvatī and his long-awaited Page 1 of 2

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Acknowledgements monograph, Love’s Subtle Magic, have now appeared in print, thanks to the editing efforts of Professor Wendy Doniger, whom we wish to thank here. The many activities of the research project could not have taken place without the active and unfailing support of Jane Savory and Rahima Begum in the SOAS Centre’s Office, and we warmly thank them here. For permission to reproduce images from their collections, we are grateful to the British Library in London, the Bibliothèque nationale de France in Paris, the Aga Khan Museum in Toronto, the Museum Rietberg in Zurich, the Bharat Kala Bhavan in Bhopal, the Chhatrapati Shivaji Maharaj Vastu Sangrahalaya in Mumbai, and the John Rylands Library in Manchester. Finally, we want to thank the contributors for their patience, and Oxford University Press for their careful editing. Francesca Orsini London Samira Sheikh Nashville

Access brought to you by:

Page 2 of 2

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Note on Transliteration

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

(p.ix) Note on Transliteration A VOLUME OF THIS KIND inevitably has a large number of transliterated words. To make the text readable without sacrificing its scholarly appeal, we have chosen to use diacritical marks only for book titles and direct quotes. For Devanagari, the transliteration used follows R.S. McGregor, The Oxford HindiEnglish Dictionary (New Delhi: Oxford University Press, 1993), with the only exception that nasalized vowels are transliterated with a tilde (~) over the vowel instead of ṁ. For Persian words, we have slightly adapted existing systems as below. In spite of our efforts, we have not achieved complete consistency.

Page 1 of 3

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Note on Transliteration

‫ﺍ‬A

‫ﺏ‬B

‫ﭖ‬P

‫ﺕ‬T

‫ ﺙ‬S̱

‫ﺝ‬J

‫ ﭺ‬CH

‫ﺡ‬Ḥ

‫ ﺥ‬KH

‫ﺩ‬D

‫ﺫ‬Ẕ

‫ﺭ‬R

‫ﺯ‬Z

‫ﺱ‬S

‫ ﺵ‬SH

‫ﺹ‬Ṣ

‫ﺽ‬Ż

‫ﻁ‬Ṭ

‫ﻅ‬Ẓ

‫‘ﻉ‬

‫ ﻍ‬GH

‫ﻑ‬F

‫ﮎ‬K

‫ﮒ‬G

‫ﻝ‬L

‫ﻡ‬M

‫ﻥ‬N

‫ ﮊ‬ZH

‫ﻕ‬Q

‫ ﻭ‬W,V, Ū, (O only if specified as majhul) ‫ ﻩ‬H ‫ ﯼ‬Y, Ī (E only if specified as majhul) short vowels: a, i, u

Page 2 of 3

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Note on Transliteration (p.x)

Access brought to you by:

Page 3 of 3

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Plates and Figures

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

(p.xi) Plates and Figures Plates (Between pp. 308 and 309) 5.1 ‘Korasht’, children playing, Miftāḥ al-fużalā, by Muhammad b. Da’ud b. Muhammad b. Mahmud Shadiabadi. (British Library, London, Or. 3299, f. 220 verso) 11.1 Qur’an manuscript, completed 801H/1399 in Gwalior Fort. (Aga Khan collection, Acc. No. AKM00281/ex Ms.32) 11.2 ‘The tomb of Alexander the Great’, page from a Khamsa of Amir Khusraw Dihlawi, first quarter of the fifteenth century (Aga Khan collection, Acc. No AKM00015/ex INDM001B) 11.3 One page from a Shāhnāma, first half of the fifteenth century (Bharat Kala Bhavan, Varanasi, Acc. No. 9940) 11.4 ‘Rustam kills the White Dīv’, page from a Shāhnāma, first half of the fifteenth century (Museum Rietberg, Zurich, Acc. No. RIV 962, Gift Rietberg Society © Photograph: Rainer Wolfsberger) (p.xii) 11.5 ‘Ghiyas Shah seated in a garden pavilion, being offered dishes by female attendants’, page from Ni‘matnāma, between 1495 and 1505 (British Library, London, IO Islamic 149/Ethé 2775, f. 40 verso) 11.6 Candāyan, sixteenth century (Chhatrapati Shivaji Maharaj Vastu Sangrahalaya, Mumbai, Acc. No. 57.I/I-68, f. 22). (Courtesy of the Trustees, Chhatrapati Shivaji Maharaj Sangrahalaya [formerly Prince of Wales Museum of Western India], Mumbai, India) 11.7 Candāyan, Malwa, early sixteenth century, (John Rylands University Library, Manchester, Acc. No. Hindustani Ms. 1, f. 146 recto. Copyright of the University of Manchester)

Page 1 of 2

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Plates and Figures Figures 1.1 Biscriptual (Persian ta‘liq and Kaithi) document from the Lodi period, early sixteenth century. Reproduced from M. Mohammad Shafi, ‘Three Old Documents’, Proceedings of the Idara-i Maarif-i Islamia, Lahore, 1936, 281–5 23 11.1 Jaunpur Anthology, Dastūr al-shu‘ara, beginning of the fifteenth century (British Library, London, Or. 4110, f.153 recto) 305 11.2 One page from a missing Sikandarnāma, fifteenth century 312 11.3 One page from a Candāyan, fifteenth century (Bharat Kala Bhavan, Varanasi, Acc. No. 230) 313 11.4 Yūsuf-u Zulaykha, text copied in 950H/1539 by Shāh Mahmūd Nayshābūrī, paintings probably second half of the sixteenth century (Bibliothèque nationale de France, Paris, Supplément persan 1919, f. 17 recto) 316 11.5 Qur’an, sixteenth century (Bibliothèque nationale de France, Paris, Acc. No. Arabe 7260, f. 46 recto) 332

Access brought to you by:

Page 2 of 2

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Introduction Francesca Orsini, Samira Sheikh DOI:10.1093/acprof:oso/9780199450664.003.0001

Abstract and Keywords This introduction sets the agenda for the volume by arguing for a closer look at fifteenth-century north India, a period of multilingualism, vernacularization, the evolution of script conventions, and the emergence of specialist literary and cultural producers who served new patrons. It charts the changing political landscape of the period, with its new towns and regional polities, showing how the new contenders for power from upwardly mobile martial groups were gradually brought under more settled, bureaucratizing regimes. Against prevailing wisdom, it argues for a continuum between oral and literary cultures, and for recovering histories embedded in texts and images generally dismissed as mythic or religious. It delinks religion from script while drawing attention to the proliferation of religious groups and vocabularies in the period. Keywords:   Sultanates, vernacularization, Hindi, Hindavi, bhakha, bhasha, Persian, fifteenth century, bhakti, Sufis, urbanization

Literature and Writing THE ACHIEVEMENTS OF THE MUGHALS throw into deep shadow the kingdoms and cultures that preceded them in north India. The Mughals are assumed to have changed everything: politics, literature, religion, even language. But how can we know what the Mughals changed when we know so little of what happened before?1 What was South Asia like before the Mughals? Most of us associate the preceding period with the ‘high’ Delhi sultanate with its idiosyncratic rulers, bureaucrats, and historians. But this had come to an end

Page 1 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction soon after Timur’s bloody invasion of north India in 1398. Was the invasion the end of an era? What happened after Timur left? The ‘long fifteenth century’ between Timur’s invasion and Humayun’s return to India in 1555, a time of remarkable change and invention in literature, culture, and politics, forms the chronological core of most (p.2) of the essays in this volume. This is a period too often masked by the centralizing categories of the Mughal, colonial, and postcolonial bureaucracies that were to follow. A quarter century ago, historians began to reveal the eighteenth century as a time of cosmopolitan ferment and experimentation; it is now time to unveil new scholarship on another century too long regarded as backward and interstitial. The conventional periodization that blanks out the long fifteenth century from histories is not because nothing happened then. Quite the opposite; it is the diversity and intensity of politics and culture in this period that have rebuffed scholars searching for singular ideologies or narratives. This was a time of cultural production in languages and idioms that ran into each other, in vernaculars ‘literized’ for the first time, in the old classical languages modulated for new patrons. There was a certain ‘democratization’ of written culture: arriviste patrons could have genealogies and tales composed; performers could reinvent the epics for the new world; upwardly mobile chieftains could lay claim to languages and forms from which they were previously excluded. Spiritual yearnings were expressed in new vocabularies: some expressed the injustices of the present, others the worldly and other-worldly aspirations of the new patrons. Political turmoil meant that people travelled: in search of employment or business opportunities, for pilgrimage, war, or pleasure. Thanks to travel, a north Indian vernacular—bhakha—was disseminated and literized across north India while the transregional High Languages of Sanskrit, Arabic, and Persian continued to wield cultural cachet. The decentralization of power also meant that there was a demand for literary specialists as chieftains and merchants all sought poets, composers, and scribes. Many of these specialists were multilingual and travelled widely in search of patrons and opportunities. It is this hybrid, restless, multilingual ferment that nation-centred, monolingual scholarship has been unable to comprehend. How does this plethora of voices and texts relate to fifteenth-century polities? The contribution of this volume is that, for the first time, it foregrounds and embraces the diversity of what we have called the long fifteenth century and investigates the links between politics and cultural production.

Literature as History Textbooks generally start from ‘hard’ evidence and documents such as coins, inscriptions, and historical chronicles typically compiled in (p.3) centres of political power, to which literature and the arts are added as supplementary ornaments, usually under the rubrics of ‘patronage’ and, in the case of vernacular devotional literature, of an undefined ‘popular culture’. But for a richly fluid time like the fifteenth century in north India, literary texts are often Page 2 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction the only way we have to write social history, to write individuals and groups, their self-representation and worldview into the picture, which is otherwise a largely empty and dichotomous one of court and people, rulers and dynasties, Muslims and Hindus, men and, of course, hardly any women at all. To study these voices and texts, and to study them in relation to each other and within a wider comparative framework, means attempting to write a thicker and more comprehensive history than that usually available in textbooks. For this reason we have tried to include in this book the widest possible range, not just of cultural production but also of social contexts and types of self-expression, and to connect and intersect them in all possible ways. The long fifteenth century was not a canon-making period. As Delhi after Timur became just one of many regional power centres (although always one with great symbolic importance), it becomes necessary to adjust our lens and look not for the great bureaucratic projects, imperial histories, or central linguistic and literary experiments of the Mughals, but for other genres of recording and remembering. Simon Digby remarks in this volume that after the relative stability of the greater Delhi Sultanate, the ‘lesser’ was a time during which few histories were written and from which even fewer survive. This remark is true only if we consider political histories from Delhi—Persian court histories continued to be written in Malwa, Jaunpur (although these do not survive), Gujarat, the Deccan, and in smaller sultanates such as Kalpi.2 Nevertheless, once we begin to look at other literary survivals and acknowledge how profuse and diverse they are, we find that each genre marks history in its own way. One basic assumption underlying this book is that producing literature in the form (p.4) of heroic narratives, genealogical accounts, local or caste puranas, or biographies and hagiographies was a way of producing one’s own history or inscribing oneself in larger histories. A large and varied body of texts in this period do exactly that—Aparna Kapadia’s local Sanskrit narratives, Ramya Sreenivasan’s vernacular ones, the Jain genealogies in the Apabhramsha texts of Eva De Clercq’s essay, or the Persian texts by provincial Sufis in Francesca Orsini’s. These are deliberate attempts by kings, merchants, and spiritual figures, through the medium of professional poets or members of their circles, to create narratives that become the history of their family or lineage and insert their protagonists or patrons into the history of a place or of a wider group. Some such histories are in book or manuscript form, while others are carved on stone inscriptions that record individual achievements and family trees on civic and religious buildings. History can also be found in other genres of texts, including in the many glossaries and dictionaries produced at this time, some that offer etymologies and lexical histories in their explanations, others that signal the linguistic needs of their commissioners and compilers. In the visual realm too, painters represented their surroundings even when they deliberately relocated ‘classical’ tales to local landscapes and climates, turning them into

Page 3 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction records of their own times. Many such conceptions of history are explored in this volume. In addition, the long fifteenth century saw the emergence of the powerful voices and personalities of Ramanand (whatever his historicity), Kabir, Nanak, Raidas, Mira Bai, and Surdas, to name but a few. Although their words (bani) did not attempt to produce history in the same way as the ones mentioned above, they produced history all the same. Through their distinctly new forms of articulation, the rapid circulation of their words, and in a few cases, their active proselytizing, these figures prefigured the creation of widely shared discourses and new groups (called panths).3 If a new form of literature and its textualization is always linked to the emergence of a new power (with consequent (p.5) realignments in the field of power), the emergence of these voices denotes a changed polity and the confident assertion of new historical subjects. Even though the words and songs of these charismatic figures were textualized only a century later, they marked as well as produced important socio-historical change.4 In the century following this textualization, their biographies and hagiographies followed in great number as their followers systematized the memorialization of both panth and bani.5 In the absence of massive bureaucratic history-making projects, what we get from this period are texts representing the political and cultural aspirations of many of the stakeholders in the patchwork of authority. While the three warrior tales of Sreenivasan’s article or the Sanskrit compositions described by Kapadia were written in different sectarian milieux, they come out of remarkably similar social circumstances, having been composed for small-time rulers or chieftains of garrison towns, each in tense subjection to an overlord, dazzled by the promise of the city, and beset by anxiety about controlling women for marriage and men for battle. These are romances, but are histories too. Their novelty lies in the adaptation of genres, the choice of literary languages, and the range of media. With the imperial ideology of the Mughals, the entrepreneurial world of such warlords was no longer viable and their attempts to inscribe themselves retreated into the oral realm where they still survive. This is why some of the narratives bear close resemblance to the oral epics that still circulate in South Asia. The eighteenth century was another time (p.6) when such narratives were literized, and literary production from the period was similarly dismissed by scholars as lacking refinement.

Multilingualism and Language Domains While it has long been recognized that this was the period when modern languages came into being, language histories have been overshadowed by the rancour of subsequent community histories. As several scholars have documented, the history of Hindi has been wrenched from the history of Urdu in the twentieth century.6 Thus, Malik Muhammad Jayasi’s Padmāvat (1540) is not seen as part of the history of Urdu because of its Indic vocabulary, while Sufi Page 4 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction tales are seen as too ‘Islamic’ to form part of the history of Hindi. However, many of the surviving narrative texts from fifteenth-century north India arose from a common aspirational landscape and evolving multilingual genre-bending, regardless of sectarian orientation. Another contention of this book is that north India, like any other part of India and most parts of the world in this period, was a multilingual society, and that cultural historians have not yet taken this basic ‘conditioning condition’ (Bourdieu) seriously enough.7 Multilingualism took different forms, particularly because literacy was often limited to particular scribal groups and educated elites, and because diglossia, or the distinction between a High Language and low language, meant that many forms of vernacular textual production, though documented, have come to us only as traces in the archive, and not as texts. This is the case, for example, of the sustained interest that Persianized elites, and provincial Sufis in particular, had in Hindavi songs and narratives (see Chapter 14 in this volume). We find many traces of such multilingualism—‘so-and-so composed songs also in Hindavi’—but very few actual texts, since the protocols of Persian anthologies and biographical dictionaries (tazkiras) meant that only Persian verses (p.7) would be quoted.8 In other cases, multilingualism is outside the text; the sophisticated Sufi authors of Avadhi romances probably read Persian, as their awareness of masnavi conventions shows, but chose to use almost no Arabo-Persian words in their poems. Or, again, the protocols of Persian texts meant that, with very few exceptions, dialogues that explicitly took place in Hindavi would be written down in Persian, leaving the oral vernacular world outside the text (Chapter 14). In other cases, the diglossic relation is flaunted, as in Vishnudas’s supposed ‘translation’ of Valmiki’s Rāmāyaṇa into the vernacular (Bangha), though little of the style of the superimposed language or text remains. While north India was not a homogenous region in political terms, it was a wellconnected cultural and linguistic region. Its linguistic economy can be described as one of ‘multiple diglossias’,9 with several high languages—Arabic, Persian, Sanskrit—and a general spoken vernacular (what we call here Hindavi) written in the Persian, kaithi, or devanagari scripts. We use the term Hindavi, which was the term (together with Hindi and Hindui) used for the north Indian vernacular in Persian sources, intentionally in order to avoid the split history of Hindi and Urdu that has dominated modern scholarship and language consciousness.10 Prakrit had a limited but symbolically important status (p.8) for Jains, who also continued to write in Apabhramsha until the sixteenth century (Chapter 12). A simplified form of Persian seems to have been a spoken lingua franca, while individuals and groups maintained spoken languages such as ‘Turki’ or Pashto for generations. Material traces of this multilingualism are scant yet unmistakeable: Persian dictionaries compiled in India in this period are particularly multilingual and include Turki and Hindavi synonyms (Chapter 5); the poor command over Persian of some Pashto-speaking Afghan amirs is Page 5 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction occasionally commented upon; and a few compositions of the Sikh gurus show that simplified Persian was current as a spoken language in Punjab in the latefifteenth and sixteenth centuries.11 Persian words had, of course, entered everyday language use, particularly in the domains of administration, power, and warfare, and can be found inflected by local phonology in a number of texts. Even a text famous for its low percentage of Perso-Arabic lexical items, such as Jayasi’s romance Padmāvat, employs a number of such words when describing the sultan’s army, suggesting at the same time a highly heteroglot body of soldiers:

Dhani sultān jehika saṃsārū,

Uhai kaṭaka asa jorai pārū.

Sabai turuk sirtāj bakhāne,

Tabala bāja au bāndhe bāne.

Lākhanha mīr bahādur jangī,

Jantra kamānaine tīr khadangī.

Jebā, kholi, rāga sõ maṛhe,

Lejim ghāli irākinha caṛhe. […]

Barana barana au pāntihi

Calī so senā bhāntihi bhāntī.

pāntī, Behara behara sab kai bolī,

Bidhi yah khāni kahān saun kholī.12

Happy is the Sultan who owns the world, and who can assemble such an army! All sing praises of the Turkish chiefs, with drums and war-attire. Thousands of Mirs and brave warriors, with mechanical bows and khadangi arrows. (p.9) Ready with armour, cannons, and leg-covers, with ironstringed bows they mounted Iraqi [horses]. Kind after kind, row after row, the varied army went. All differed in speech—where did God open such a treasure?13

While Persian remained the language of scholarship and political theory of the Delhi sultanates and the regional sultanates, and Persian poetry, both ‘classical’ and Sufi, was taught and written in madrasas, Sufi centres (khanqahs), and courts all over north India, there are signs that parts of administration and cultural production began to encompass both Persian and what, for the reasons explained above, we have called Hindavi. Meanwhile, the knowledge of Persian began to spread among Hindu scribal groups.14 Some of the very few Sultanate documents extant are bilingual, with Persian on top and Hindavi in the scribal kaithi script at the bottom.15 Provincial Sufis clearly wrote and listened to and thought about both Persian and Hindavi poems and songs (see Chapter 14), and Page 6 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction the tradition of writing romances in Hindavi (p.10) begun with Da’ud’s Candāyan that addressed both worldly/courtly and spiritual audiences continued in this period (see Chapter 10). Sanskrit musical treatises continued to be written, but the same musical knowledge became available in Persian and, so we are told, in bhakha at Gwalior thanks to Raja Man Singh Tomar himself.16 Music in north India developed in widely different environments, patronized by Turk, Afghan, and Indian rulers alike.17 In contrast with the later art of the book under the Mughals, among the books illustrated under the regional sultans was a vernacular text, the Candāyan again (see Chapter 11), indicating that the taste for fine books included what was probably the most popular Hindavi romance of the time. Persian itself was ‘provincialized’ in the fifteenth century. After the great transregional moments of Mahmud’s Ghazni, Amir Khusraw’s Delhi, and Timur’s Herat, Persian scholarship in the fifteenth-century regional sultanates, whether concerned with lexicography, medicine, or cooking, took into account local material (Chapters 5 and 6). Large-scale patronage of Sanskrit literature and scholarship in north India was limited and was similarly ‘provincialized’ or adapted to the political and cultural demands of the time. A striking example is the Sulaimaccaritra, a Sanskrit life of King Solomon written by the poet Kalyana and commissioned by a Lodi Afghan amir, Lad Khan, in early sixteenth-century Lucknow,18 or the itinerant poet Gangadhara who tuned the tools of (p.11) Sanskrit kavya to the realities of his patron Gangadasa in the small local court of Champaner (Kapadia). The Sanskrit model letters and documents of the late fifteenth-century Lekhapaddhati include local words and even ‘Sanskritize’ some colloquialisms.19 Sanskrit poets thus turned to new patrons—local rulers or the sultans20—or else turned to the vernacular, or both, as the careers of Keshavdas’s ancestors from Gwalior to the new Bundela principality of Orchha attest.21 Later, as Sheldon Pollock, Muzaffar Alam, and others show, both Persian and Sanskrit would make a spectacular comeback under the new imperial Mughal dispensation, finding new patrons and new social roles. Our strategies for grasping the multilingual quality of the social and literary/ cultural world in this period have been several. One has been to focus specifically on bilingual or multilingual texts (Sheikh, Orsini, Karomat), trying to gauge the conditions under which the different languages were used in the same place, the assumptions behind the use of each language and the relationship between them. For example, if one striking characteristic of Persian dictionaries in this period is that they included among other languages not just Arabic, but also Turki and Hindavi, what kinds of words were included, and were they among the lemmas or synonyms in the definition? Were they the words that the dictionary sought to explain or those the dictionary used to explain? (See Karomat, Pellò). Another strategy has been to focus on a particular place and explore the language and literature traditions that were current there, as in the literature of the Jain merchants and the Tomar court in the Gwalior of De Clercq Page 7 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction and Bangha’s essays.22 A third strategy has (p.12) been to look for traces or mentions of other languages and language tastes in monolingual texts; even if the protocols of transmission of a particular archive exclude verses and narratives in other languages, traces and mentions are enough to show that they were known and circulated at the time. Thus, while seeking to understand the logic internal to the formation and transmission of each archive (courtly and madrasa Persian, Jain, courtly bhasha, Bhakti, Sufi), it has been crucial for us to also question their limits and exclusions, and to place them within a larger framework. It has also been important to consider the world of orality and performance which some of these texts refer to (for example, music treatises, which famously do not include song texts) and the oral-performative aspects of the High Languages, Sanskrit, Persian, Arabic, for words and phrases and verses from the High Languages circulated orally beyond the realm of the educated, and can be found in the songs of ‘unlettered’ bhagats.

Vernacularization The most influential and compelling argument about the relationship between language, literature, and politics, and between a High Language (Sanskrit) and Indian vernaculars has been put forward by Sheldon Pollock.23 Marshalling an impressive range of genres and examples, he argues that: (a) vernacular languages are first literized or written down, usually for documentary purposes, and then, often after a considerable gap, are literalized, or used for literary, imaginative, performative, and expressive purposes (what he calls ‘workly’); (b) any vernacular innovation (p.13) is linked to a reconfiguration of the culturepower order, when in place of cosmopolitan imperial polities more regional, vernacular polities emerge; (c) vernaculars then themselves become cosmopolitan and are used simultaneously, as had been already the case with Sanskrit, both for political and literary discourse, ‘with the court functioning as engine for the stimulation of literary production of a textualized sort’;24 (d) in order to become literarized, the vernacular has to emulate the superimposed models of literature of the cosmopolitan language, for ‘there is no parthenogenesis in culture’.25 As he puts it, ‘vernacular poets achieved literary expressivity by appropriating and domesticating models of literary-language use from superposed cultural formations’.26 Pollock is very clear that what he means by literature is what his authors meant by literature: kavya and sahitya,27 a set of genres and discourses highly regulated by the ‘science of literature’, sahityashastra. ‘By contrast, the world of the “uncultured”, that is, of the uncourtly and noncosmopolitan languages of Place (deshabhasha), was subliterary: a domain of the sung, the unwritten, the oral’.28 And since his Sanskrit theorists disdained orality and literature in non-cosmopolitan languages and consigned songs to a different order of discourse (gita), for Pollock, too, literature that does not follow the courtly practice of kavya is simply not literature at all.29

Page 8 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (p.14) This model—of the emergence of vernacular literary culture with the making of vernacular polities and the literarization of the vernacular through emulation of the superimposed model of Sanskrit by (mostly Brahmin) poets at court who knew Sanskrit well and shared the conceptual framework of Sanskrit poetics but wanted to innovate—works perfectly in the case of Kannada literature at the Hoysala court, but it works less well for north India for a number of reasons. First and more obvious is the substantial presence of a new High Language of literary and political discourse, Persian, which spread over north India through the sultanate administration, madrasa education, and the culture of the Sufis. That Persian was not only a high cultural model but also, in a simplified form, a kind of vernacular in north India is demonstrated by the fact that when Guru Nanak, or later Guru Arjan, used it for hymns, they used a spoken, broken, and already phonologically assimilated Persian and not the Persian of literary models.30 Second, while Sanskrit-educated Brahmin poets later did produce the kind of superimposed vernacular literary culture described by Pollock—notably by Keshavdas at the small Bundela principality of Orchha— they were by no means the only or principal agents of literature at this time.31 As we shall see, many of the authors of poems and narratives in fifteenth-century north India were Muslims, Kayasthas, Jain panditas, or of low-caste, unknown or mixed background, even when later tradition strove to ascribe hidden Brahmin pedigree to them. It is then unsurprising that the literary genres they preferred and the literary models they followed were less Sanskritic than in Kannada and other similar regional literary cultures. As a matter of fact, fifteenth-century vernacular literature consists mainly of songs, dohas or couplets, and kathas or narratives; some were indeed produced at regional or even smaller courts, but others were aired in the open (p.15) ‘Bhakti public sphere’ of towns and villages.32 Songs and singer-composers, vaggeyakaras, were highly prized and at the centre of courtly performances, as well as of devotional practices and temples. Kathas, songs, and dohas were genres practised by a range of different poets—Naths, Sants, Sufis, Jains, bhakha, and also sometimes Persian court poets—and the high degree of intertextuality in their titles, tropes, and images shows that they circulated in all these domains. Literary genres were thus a palimpsest on which every poet wrote from his particular perspective. Instead of Pollock’s model of vernacularization sketched above, then, we propose to understand literary culture in fifteenth-century north India as multilingual and multi-locational, mirroring the social forces that were active and vocal in the polities of the regional sultans and Rajput kingdoms and in the religious marketplace of the time. There were distinct trends to literary production: Persian–Hindavi bilinguality in the political and literary domains of the sultans and of Sufi religious and literary practice (Chapters 14 and 10); Apabhramsha– Hindavi/bhakha bilinguality in Jain circles (Chapter 12); vernacular literary production with significant gestures towards Sanskrit in certain Rajput polities

Page 9 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (Chapters 13 and 8); and the emergence of strong vernacular voices in the ‘Bhakti public sphere’. In terms of language names, the modern regional linguistic categories of Braj Bhasha, Avadhi, Bhojpuri, and Khari Boli are not reflected in fifteenth-century sources, which instead speak of a generic bhakha (bhasha) or Hindavi/Hindui/ Hindi in Persian texts as the vernacular of north India.33 These general terms denote a certain lack of grammatical and taxonomic interest in the vernacular; for example, Da’ud’s romance Candāyan, classed by modern writers as ‘Avadhi’, was recited in Delhi without any comment on the eastern flavour of the language.34 It was (p.16) only late in the sixteenth century, thanks to its status as the language of court poetry with an alamkarashastra pedigree, that Braj Bhasha was recognized as a named language, a ‘cosmopolitan vernacular’ with a standard poetic language that needed to be learnt properly from teachers and through riti-granthas or poetic manuals.35 As is well known, the growing status and popularity of Braj Bhasha poetry induced poets like Tulsidas to use it for his songs and verses.36 The generic terminology for the vernacular also suggests a continuum with locally produced songs and tales that could travel and be understood over the whole of north India. Oral performers and performance contexts were probably crucial in this respect, with performers able to modify inflections and replace words that were too local while keeping to the metrical scheme. Such changes would then be written down by scribes and resurface as linguistic variations in texts, thus the great differences remarked between the ‘eastern’ and ‘western’ recensions of Kabir’s songs.37 Writing was in some cases intrinsic to the production of literature—as in the case of works written for patrons—but in other cases it was a function of transmission, or else subservient to performance. The ‘oral’ texts of the performer’s notebook were of a different class from the ‘literary’ texts of the pothi or grantha.38

(p.17) Writing, Scripts and Scribes After Timur’s incursion that precipitated the splintering of the Delhi Sultanate, ‘new public arenas emerged when state-created revenue and judicial bureaucracies extended their reach into village society’ in the fifteenth, sixteenth, and seventeenth centuries (Chapter 4, p. 128). In his contribution to this volume, Richard Eaton usefully compares the different options chosen by the sultans of Bengal and of Bijapur and Golkonda with respect to the language of administration, and points out the different languages that were used for stone inscriptions (on religious versus non-religious buildings, see Chapter 7), for court treatises on governance, for coins, for court literature, and, most significantly, for the administration. It was the much greater availability of paper in the fifteenth century and choice of the vernacular for revenue and judicial bureaucracy, he argues in the case of the Bahmanis, that brought ‘Marathi-

Page 10 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction speaking peoples of all classes into closer communion with the state and with one another through the medium of written Marathi’ (p. 120).39 It is harder to make a comparable argument for the sultanates of north India— Sharqi Jaunpur, Lodi Delhi, Khalji Malwa, Sher Shah’s short-lived empire— because of the almost complete lack of reliably datable contemporary documents. A handful of surviving documents (more are said to lie in private hands) and a few references in contemporary chronicles suggest that Persian was not the only language of sultanate governance. At least at the district (pargana) level in Delhi, (p.18) some amount of bilingualism and biscriptualism was in place, with two karkuns (writers)—one to write in Hindavi and another to write in Persian — appointed for every pargana. In the north, records were transcribed in two languages and scripts—Persian in ta‘liq and Hindavi in kaithi (see Figure 1.1). In Malwa and Gujarat, though, the nagari script was used.40 We also hear that in the time of Sikandar Lodi (r. 1489–1517), Hindus ‘learned to read and write the Persian script, which had not been common among them until then’.41 Muslim officials are known to have used the vernacular for public announcements or decrees. A tax remission decree from 1513 under the reign of mahārājādhirāja rājaśrī Sulitāna Mahamūda Sāhi bina Nāsīra Sāhi [rājye] (Mahmud Khalji II, r. 1511–31) of Mandu shows Malu[k] Khan, the muqta‘ (rights-holder) and local officials using the vernacular in the nagari script to communicate a benevolent decision to the local population: Siddhiḥ samvat 1570 satarā varshe māgha vadi 13 Soma dine mahārājādhirāja rājaśrī Sulitāna Mahamūda Sāhi bina Nāsīra Sāhi rājye asau Damauva nagare śrī mahāshāṇa Ājama Malū Shām (Khan) biṇa Malū Shām (Khan) mukte varttate tat-samayai dāmabijāī va maḍavā va dāī va darajī ai rakamau ju dama[ḍ]ā lāgate mukte mījī va vahadārāṇa hara berisa sālīnā le to mumāphuki ai chhoḍe ju koi isa barisa va isa desa thī inha maha [le]- (p.19) hi damaḍā pai[kā] mā[m]gai leï su apaṇa dīṇa thī be[j]āḍha hoï. Musalamānu hoï damaḍā leï tisahi suvara kī sauṁhā Hindu hoï leï tisahi gāï kī sauṁha. Pravānigī Malika Seshaṇa (Shaykh) Hasaṇa Shāṁ (Khān) [Nirabadāchha Mau] koṭhavālu Sonīpahaju Gopāla Sha(Kha)lachi-[pura-vare śubhaṁ bha]vatu.

Success! In the Samvat year 1570, on Monday, the 13th day of the dark (fortnight) of Magha, during the reign of the great king, the illustrious Sultan Mahmud Shah, son of Nasir Shah, in the town of Damauva (Damoh), while the muqta‘ grant of the great Khan Ajam Maluk Khan, son of Maluk Page 11 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction Khan, exists, the mukta grantee [and home farmers?] who take every year the annual fees levied on account of seed-loan, marriage booth, midwife and tailor should remit them according to this. Whoever demands these fees from this year and this country is to become an outcaste from his religion. If he be a Muslim and take the fees, to him be the curse of a pig. If he be a Hindu and take (the fees), to him (be) the imprecation/curse of (killing) a cow. By permission of malik Sheikh Hasan Khan [of Nirabadachha Mau?], and kotwal Sonipahaju Gopala of Khalachi-pura. Let (success) attend.42 We may note the presence of Arabic and Persian nouns and expressions related to administration (muqta, mumaphik for mu’āfik, parvāngī, and of course, sulṭān), but also to common life (daraji for darzī, dīn for religious community). In nearby Gujarat, the Sanskrit model letters and documents from the 1470s in the Lekhapaddhati suggest that some form of Sanskrit chancellery was still practised, or at least learnt.43 The currency of Sanskrit, for inscriptions, texts, and letter-writing seems linked to the revival of courtly practices by small chiefs. Competence in Sanskrit was prized; in (p.20) addition to the travelling poet Gangadhara discussed by Kapadia, there were other Sanskrit poets active in Gujarat including Udayaraja, the author of a long poem dedicated to Sultan Mahmud Begada, and the authors of numerous Sanskrit prashastis including the long Dohad inscription described by Sheikh. Sanskrit inscriptions were commissioned in other sultanates too; among the names of commissioners and scribes, we encounter Hindus, Muslims, and Jains.44 This multilingualism and multi-scriptualism is reflected in manuscripts from this period. Although very few manuscripts from the fifteenth century have survived, Eloïse Brac de la Perrière has noted the great diversity of scripts used for Arabic and Persian books in this period, from the bihari used largely for Qur’ans (see her contribution) to the varieties of naskh, including what Losty has called ‘naskhi-diwani’, an intermediary form between riqa‘ and ta‘liq that was probably used by the Tughluq chancellery, and nasta‘liq, of which we see the first examples in the early fifteenth century.45 The similarity in iconography between certain Jain illustrated manuscripts and illustrated Persian codexes from this period is another instance of the more intense cultural connections within the sultanates’ polities and cities, and of the beginning of a layered but interconnected taste for ‘the art of the book’ that was to develop much more widely in the Mughal period.46 Equally striking is the fact that in this period some of the Sufi Hindavi romances (Candāyan, Mirigāvatī) were occasionally produced (p.21) as illustrated codexes, that is, as precious objects, in the same form as Persian illustrated books—sure evidence of the bilingual culture of sultanate north India.47 In fact, the multiple scripts and formats in which these texts circulated are the only Page 12 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction evidence we have for their circulation in different contexts, given the absence of colophons in many manuscript copies. They were produced in unadorned nasta‘liq for use in the Sufi khanqah, illustrated in naskhi-diwani for local Persianate elites, unadorned in kaithi probably for the use of storytellers or of lesser patrons, and illustrated in kaithi for local Rajput/merchant elites.48 Evidence about scribes and scribal groups in the fifteenth century would have to be collated from the colophons of individual manuscripts and inscriptions, a task that so far has been undertaken only for Jain manuscripts.49 As the designation for professional scribes and scribal groups, Kayasthas (thus kaithi) appears often. Direct evidence about Brahmins and their occupations in north India in this period is strikingly scarce when we consider the evidence that Kayasthas, Khatris, and Brahmins formed the majority of Persian scribes and administrators in the Mughal period. However, this very fact suggests that they were (p.22) probably among those Hindus who learnt Persian from Sikandar Lodi’s time, as noted above.50 What lessons can we draw for literary history from the persistent biscriptualism and multilingualism of the long fifteenth century?51 First, as we have already noted above, while traditionally for Hindi and Urdu literary histories the script of a text has been a primary indicator of where and to whom the text ‘belongs’, we have seen that the choice of script for Hindavi texts like the Sufi romances depended on the scribe and the person commissioning the copy. Script was an indicator of circulation rather than of the intrinsic nature of the text. At the same time in a context where so many texts (poems, songs, tales) were routinely read out and recited, the script in which a text or genre was transcribed and copied cannot be taken as an indicator of the limits of its audience, especially in the case of vernacular texts. Furthermore, only a comparative, multilingual approach can help us recognize the dynamics of individuals and genres within the religious public sphere and arrest the hallowed tradition of placing and explaining texts and authors on the basis of their putative religious identity.52 For north India in this period, it is vital to decouple language from script.

Regional Polities What was the political context of fifteenth-century north India? Since standard histories, as we have seen, view this period as the ‘twilight’ (p.23)

Page 13 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (p.24) of the Sultanate and as a period of fissiparous tendencies and decline, a historical reappraisal seems timely.53 As we have seen, the Delhi sultanate was devastated by Timur’s invasion in 1398 and reduced to a small regional kingdom. The invasion was only a catalyst, however, since the real authority of the sultans had declined in late years. As regional governors became wealthier and more autonomous in the latter half of the fourteenth century, Delhi’s central resources of men, horses, elephants, weaponry, and gold, shrank. Since the time of Firuz Shah (d. 1388), agricultural revenues and the size of the army had steadily dwindled —Firuz’s successor, Mahmud Shah (r. 1394–1412), met Timur’s forces with 10,000 horsemen, 20,000 footsoldiers, and 120 elephants—a quarter of imperial armies in previous decades.54 The invasion was followed by a decade of conflict between members of the Tughluq ruling family and Figure 1.1 Biscriptual (Persian ta‘liq and prominent courtiers, each of kaithi) document from the Lodi period, whom canvassed the powerful early 16th century. Reproduced from M. regional governors for support. Mohammad Shafi, ‘Three Old But Timur strengthened the hands Documents’, Proceedings of the Idara-i of certain factions: as Simon Digby Maarif-i Islamia, Lahore, 1936, 281–5. points out, Khizr Khan, the former governor of Multan who became the first ‘Sayyid’ ruler in 1414, had submitted to Timur and benefited from the contacts and deals he made in Timur’s entourage. Khizr Khan and his successors may have continued to pay submission to Timur’s son Shah Rukh in Herat until the middle of the fifteenth century.55 But the rulers of Delhi for the next half century exerted only limited authority over a small region contiguous with the capital city, provoking the famous remark that the sultan’s sway extended only from Delhi to the suburb of Palam.56

Khizr Khan and his successors relied on the support of the Afghan families then powerful in Punjab and Multan, including the Lodi chieftains who became prominent in the 1440s. By 1451, Bahlul Lodi had taken over Delhi. Bahlul and his successors invited more Afghan (p.25) chiefs to help shore up their rule against the increasingly influential sultans of Jaunpur, but the immigration into Page 14 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction the Doab of these powerful families meant that the Lodis had to share sovereignty with them, including the sultan’s monopoly of the most vital symbol of military might, the elephant.57 The Afghan clans could only forge an uneasy alliance, fraught with struggles for primacy, and eventually another prince of the Timurid lineage was invited to invade north India in 1526. As every South Asian school-child knows, Babur defeated Ibrahim Lodi at Panipat in a triumph for artillery over the foot soldiers and elephants of the Lodi army. But the Afghan families had now put down roots over large parts of north-eastern India, and the Mughal advance was by no means a foregone conclusion. Babur and his son Humayun had to repeatedly battle Afghan interests before Humayun’s decisive return to India in 1555. Decades before Timur, in the middle of the fourteenth century, the Tughluq conquests had reached their limits and the sultans had to face the challenges of scale. Muhammad b. Tughluq died chasing a rebel through the Kachchh desert, and his cousin Firuz was obliged to be constantly vigilant to suppress revolt. Tughluq provinces were ruled by the sultan’s nominees; unlike the Khaljis who relied on relations, military slaves or Iranian/Central Asian courtiers, the Tughluq nominees were often from recently converted peasant or warrior families. As Kumar describes in this volume, the déraciné courtiers and slaves of former times, tied by relations of slavery or service to great lords or recently uprooted from their natal contexts, were gradually developing networks through marriage or alliance and becoming ever more prominent and threatening to the sultans. The regional governorships were greatly prized: the governorship of the wealthy trading region of Gujarat was even put up for auction in the 1370s.58 The sultans attempted to prevent their governors from setting down local roots and getting too comfortable with the locals, but all too often they were powerless to prevent (p.26) the rise of local bases of power. The principality of Khandesh became virtually independent as early as 1382, when Firuz Shah was still alive.59 In Gujarat, too, the last-but-one governor Farhat al-Mulk was deposed for his closeness to the Hindus and was replaced by the loyalist Zafar Khan, the son of a peasant convert in the retinue of Firuz Shah.60 In the years following Timur’s invasion, these regional governors hedged their bets, deciding whom to support in Delhi before striking out on their own. Khwaja Jahan of Jaunpur remained loyal to the Tughluq princes Muhammad Shah and Mahmud Shah; it was only after his death that his adopted son took royal titles. Similarly, Dilawar Khan Ghuri of Malwa may never have officially become sultan. In Gujarat, Zafar Khan supported another prince, Nasir al-Din Nusrat Shah, and took royal titles only in 1407. The political ambiguity of the first decade of the fifteenth century gave way to assertive regional polities as the erstwhile Tughluq governors became regional rulers in alliance with local chieftains.

Page 15 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction The Tughluq empire had facilitated the movement of men, resources, and animals over its extent. While the imperial reach of the fourteenth century was long gone, entrepreneurs were able to travel widely and make long-distance deals. Many of the routes, towns, and agrarian and trade zones that we associate with the Mughals should be linked to fifteenth- or early sixteenth-century innovations. One consequence of the ascendancy of the Afghans meant that the whole of the northwest was restored as an area for the circulation of men and resources from the plains and became a place for military recruitment and naukari.61 With the rise of the regional polities, rival rulers had to struggle for access to war resources such as elephants and horses. Hushang Shah of Malwa travelled disguised as a horse trader to Orissa, offering horses in exchange for valuable elephants (Chapter 2). The Gujarat sultans went to great pains to protect horse merchants; in 1487 the Raja of Sirohi faced the threat of swift retribution after ‘seizing’ a company of merchants bringing 400 Iraqi and Turki horses from Khorasan (p.27) and Iraq.62 Mahmud Begada, the Sultan of Gujarat, reimbursed the merchants from his own treasury and sent a stern order to the raja to return the confiscated property. The raja complied immediately. Each regional kingdom forged close links with groups who could bring them valuable goods or connect them to long-distance trade. In Gujarat, new towns were built with merchant colonies for Hindus and Jains at their heart, surrounded securely by concentric circles of other inhabitants and fort walls.63 The Jains, with their networks of credit and monetary skill, were particularly sought after, and many fifteenth-century kingdoms protected Jains and invited them to settle (Chapter 12). Jains had already held bureaucratic positions in sultanate administrations: Thakkura Pheru worked in the mint of the Khaljis in the early fourteenth century, and Jinaprabha Suri is said to have been invited to meet Muhammad b. Tughluq.64 In the fifteenth and sixteenth centuries, Jain names come up frequently as mantris—courtiers or bureaucrats—in sultanate administrations. Other non-Muslim naukars appear as accountants, secretaries, and administrators. As none of these regional kingdoms had unfettered access to war resources, nor did they have the vast resources of the fourteenth-century Delhi sultanate, each was obliged to cultivate fighting men for their armies. Although the postfifteenth-century culture of naukari—military employment— has traditionally been seen as distinct from the bandagi—military slavery—that had formerly dominated the Delhi army, Sunil Kumar argues in this volume for a longer-term view that does not radically distinguish these two forms of service (Chapter 3). Peasant groups and formerly mobile pastoralists were increasingly drawn into politics. For part of the year they were needed to fight; for the rest, they were required to clear forested and marginal lands, and to increase agricultural productivity. Much of the literature of this period arises from the experience (p. 28) of the newly politicized rural groups of north India as they entered new alliances and relations of servitude. Page 16 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction An important point that Simon Digby makes is that power in this period was multi-locational. He cites the Afghans who maintained bases in various noncontiguous locations in north India while keeping links with their homeland in the Gomal valley. Of the Baghela chieftains too, Digby reads their boasts of having conquered Kashi and other places as ways of declaring their interests or bases in different locations. Even with the more settled sultanates of the period such as those of Malwa, Jaunpur, or Gujarat, kings were obliged to be constantly mobile, and the apparatus of power moved with them from campaign to campaign. Secondly, rulers rarely exerted direct sovereign power over either men or revenues in the manner of the Mughals. Standing armies were rare; fighting men had to be mobilized repeatedly for campaigns. The armed peasants of the countryside that Dirk Kolff has written about were often accessible to rulers or political claimants only through their leaders—whether small landlords or chieftains, or religious ascetic collectives such as the Naths—and thus power was a matter of constantly making deals with those who could supply manpower and resources. The Gujarat sultans were able to partially stabilize the manpower situation by the mid-fifteenth century by offering heritable jagirs and regular cash payments to soldiers, but most other claimants to power had to continually remake agreements. While the Lodi sultans may have been the nominal sovereigns of late fifteenth-century Delhi, their authority rested on remarkably tenuous deals with various Afghan families, peasant warlords, merchants, horse suppliers, and so on. It was only with Sher Shah Sur, who managed to secure valuable alliances both martial and marital, significant wealth, and who regularized the payment of men and branding of horses, that a more stable military enterprise came into being. Until then the successful military entrepreneur was the one who understood the art of deal-making. With constantly evolving chains of authority, people learnt to signal their status and ambitions on fresh terrain. While political boundaries were defined and redefined, they were not always negotiated on religious grounds. In other words, while sectarian and ethnic affiliations did matter in alliance, employment, and promotion, they were not the only or even the most important considerations. As Digby remarks, the Baghelas supplied men to both the Jaunpur sultanate and the Afghans; conversely, (p.29) they took Afghans into their service, as did Rana Sanga of Mewar.65 The Jaunpur sultans, cut off from renewable supplies of horses, were obliged to turn to non-Muslim sources of military manpower. A Vaishnava landholder in north Gujarat boasted of his ships and his respectable Puranic ancestry while noting the Muslim names of his sons in sultanate service.66 Marriage was a way of making alliances and again, some were across ethnic or religious boundaries. The Afghan Sher Shah Sur is believed to have married ‘Guhar Gusain’ and Lad Malika, the widow of a Turki warlord—both marriages brought him the support and military manpower of non-Afghan families.67 The Muslim Kayamkhani family of Fatehpur claimed to have taken brides from Hindu Rathor clans.68 Intriguingly, Hushang Shah, the sultan of Page 17 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction Malwa, had a Jain wife and a practising Jain son who left a long inscription asserting his parentage and his religious piety.69 While pragmatism generally prevailed over dogma in making alliances and deals, we would not like to suggest that ascriptive identities were irrelevant or ‘fluid’ in this period. In fact, belonging and exclusion were carefully calibrated, although ethnic, religious, and status markers did not always carry the social meanings we assign to them today. For example, instances of ‘religious conversion’ are recounted not as ideological or faith transitions but usually as pragmatic shifts of allegiance helped along by prominent Sufis. Several such narratives are connected with Firuz Shah in the late fourteenth century. One such is the story of (p.30) the ‘conversion’ of Karamchand, the ancestor of the Kayamkhani clan of Rajasthan. According to the Kyamkhān rāsa, a seventeenthcentury community history in Braj Bhasha, Firuz Shah and his companion Sayyid Nasir came upon Karamchand asleep in the forest. They decided that the boy had miraculous powers and that he would eventually become a ‘Turk’, a Muslim. The boy, renamed Kayam Khan, joined the sultan’s entourage. He prepared to enter Islam under the supervision of Sayyid Nasir but worried that his family would lose their social standing and ability to make good marriage alliances. He was only convinced when Sayyid Nasir assured him that future kings would marry their daughters into his family.70 After becoming a Muslim, he received land and goods from the sultan, thus establishing his place in the landed hierarchy of Rajasthan. Another version of the Kayamkhani story makes no mention of conversion. Here, the young Chauhan boy was brought up by Sayyid Nasir and later employed by the sultan Bahlul Lodi.71 Similarly, the ancestors of the Gujarat sultans were modest militarized peasants who converted to Islam after joining Firuz’s retinue, this time at the behest of the Sufi Jalal al-Din Bukhari, also known as Makhdum-i Jahaniyan Jahangasht.72 The sixteenth-century chieftain Silhadi became a Muslim under intense pressure. In Sikandar’s seventeenth-century telling of the tale, Bahadur Shah, the sultan of Gujarat, declared, ‘This wretch keeps Muslim women in his house … I will never let him go alive unless he becomes a Musalman’.73 After Silhadi was cornered by Bahadur’s forces at Raisen, he finally agreed to ‘enter Islam’. We hear also that the young son of the deposed Chauhan ruler of Champaner, eventually defeated by Mahmud Begada, was brought up as a Muslim, eventually rising to a high position as Qiwam al-Mulk Sarang.74 There is little evidence of large-scale conversion to Islam in this period or for ‘forced’ or incentivized conversion. The very fact that the instances recounted above are discrete narratives in the sources suggests that they were remarkable occasions. (p.31) We might even consider the possibility, with Dirk Kolff, that ‘Afghan’ and ‘Rajput’ were not exclusive birth-identity markers in this period. Beyond the basic fact of confessional identification, there was considerable scope for interpretation. The Kayamkhanis of Talbot’s account continued to be very much Page 18 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction part of the Rajput world even after their ‘conversion’, and their Cauhan ‘got’, gotra or lineage continued to be vital to their dealings and status. On the other hand, Kolff suggests that just as low-status Bhils and Minas were adopted into the evolving Rajput hierarchy as junior clans,75 militarized peasants who ‘entered’ Islam often became known as Pathan, or Afghan.76 Thus, ‘Rajput’ or ‘Pathan’ could at this time be military identities rather than ‘ethnic or genealogical denotations’; those who secured employment within one military tradition or another would adopt the vocabulary of upward mobility and allegiance that being ‘Rajput’ or ‘Afghan’ connoted. Accordingly, what might appear to us as religious conversion or ethnic confusion was ‘often a device to register either recruitment or professional success whether military or otherwise’.77 In many texts there is considerable anxiety around the control of women. A key part of Silhadi’s transgression was the fact that he had Muslim women in his harem. In this deal-making society, women wielded considerable clout by virtue of their natal links, and there are several examples of women who did not convert to the dominant religious or cultural ethos of their husbands’ families. Hushang Shah’s Jain wife is one such instance. Another classic case is that of Mira, the devotee of Krishna who refused to abandon her devotion and accept her marital family’s Shakta orientation. Women were often the instigators of religious change; many of the Sants counted women as their primary devotees, who then took the new message to their husbands and families. Sants and religious practitioners who promised fertility were also popular with women. In the clan-based peasant and military families of north India, women had considerable agency, partly due to the importance of their natal families in marriage/military alliances. This was true of Afghans too: Bibi (p.32) Fath Malika and Lad Malika were both successful leaders in Sher Shah’s times. While the ‘unstable and segmented warrior polity’ of the long fifteenth century afforded opportunities for women (including the camp-following women of the early Mughals78), the rise of later genealogy- and honour-based polities relegated them to seclusion in harems. Beyond the warrior world, too, women had considerable buying power. The donations of pious women—Muslim, Hindu, and Jain—to build or refurbish religious structures are regularly found on inscriptions. There are even inscriptions in which no male figure is mentioned.79 In this era of entrepreneurship, exalted ancestry was not a guarantor of success, but for most it was important to acknowledge family and lineage, even when those were modest or made up. Got or gotra, lineage, or kula, clan, were invoked often, considerably more than varna or jati. In Holi’s Sanskrit inscription of 1424, Sahi Alambhaka (Alp Khan alias Hushang Shah) is said to belong to the Gauri (Ghuri) kula.80 When Bibi Ayisha, a daughter of Bahlul Lodi, and Bibi Muradi Khatun, her sister-in-law, had a well built and commissioned a bilingual Persian–Sanskrit inscription in 1517, their family names appear as gotras in the Page 19 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction Sanskrit part: Ayisha of the Serati (Sherani) gotra and Muradi of the Sarwani.81 Raidhu’s prashastis also record the ‘caste’ and gotra of his Jain merchant patrons (Chapter 12). There are also frequent mentions of occupational mobility which would be precluded by a varna-based notion of caste. Thus, Bhima Kavi says that his father was a Kayastha or scribe who was also a kubera or merchant.82 In the Mansa inscription from sixteenth-century north Gujarat, a merchant claims inheritance in a Rajput lineage while signalling his role as sultanate bureaucrat.83 The Afghans who came to India ‘became merchants’, accruing considerable wealth through the sale of horses and other goods. The honour codes of the Afghans, the ethos of naukari and overlordship, the developing lineage hierarchy of the Rajputs, and the evolving refigurings of classical warrior (p.33) mythology—these were potent codes of prestige and upward mobility available to transitioning groups in north India. For the emerging elites of the time, rootedness was the exception rather than the rule; the reality of displacement and the possibilities of reinvention are features of much of the literature of the time.

Geographies in Literature In addition to the mutability of occupation and identity and the low importance of certified lineage, the frequency of travel and dislocation should form the backdrop to our understanding of the literary genres of the time. In the Vīsaḷadevarāsa, written likely in late fifteenth-century Marwar by the poet Nalha, the king, stung by his wife’s taunts, leaves his country to take up service as a courtier (uḷagāṇau), handing the kingdom over to his nephews.84 Opportunities for service—or what Dirk Kolff has called naukari—were abundant at the time, and many men took the decision to relinquish the autonomy of a peasant or pastoralist life and travel forth in search of employment, leaving behind women and households. Viraha—separation or yearning—is the dominant sentiment of much of the literature of the time. Visaladeva’s wife Rajmati angrily spurns an elderly woman’s suggestion that she should take a ‘friend’ in her husband’s absence (vv. 77–8), but the verse brings out an underlying theme in many of the kathas, anxiety about retaining the fidelity of women. A similar anxiety is expressed in accounts of jauhar—the self-immolation by Rajput women when defeat is certain or when kinsmen are too far away to help.85 Merchants, too, travelled in search of opportunity. De Clercq’s study of colophons reveals how Jain merchants in Gwalior had close family and business connections with other Digambara families in Delhi, Hisar, Kurukshetra, ‘Lahadapura’, and Mandu, crossing political boundaries with seeming ease. Marriages were arranged over long distances and both ascetic and lay religious leaders travelled widely to reach their followers. Spiritual geographies did not always overlap with the political. When the Afghan soldier Dattu Sarwani travelled around north India in the (p.34) 1530s in the service of different armies, his pir, the Sufi master Shaykh ‘Abd al-Quddus Page 20 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction Gangohi, appeared regularly in his dreams to offer him counsel, thus creating a powerful realm—wilaya—within which Dattu felt protected. The dream-pir appeared to warn Dattu against local Sufis such as Shaykh Ahmad of Mandu, guided him to seek the help of the pirs of Gujarat (anecdote 97), and even kept him safe from a sanyasi offering to help the Afghan cause. In the latter dream, Dattu was attracted by the sanyasi who ‘talked well about God being one’, but his pir was at hand to enforce religious boundaries (anecdote 93). He appeared in a dream instructing Dattu to stay away and even driving the sanyasi away with his slipper.86 While Dattu criss-crossed political boundaries and allegiances, his faith in his pir remained constant, whether he met him in person, as at Agra, or experienced his protection while asleep. He travelled not just through political territories but through the domains of powerful religious figures, brother-Sufis to his own master, that were as real to him as kingdoms. In other texts, epic-puranic geographies combine with terrestrial, local ones. In Bhima Kavi’s Ḍaṅgvai kathā (1493), the magical mare describes a classical but very familiar geography of north India in her wanderings through Jambudvipa des: she roams the forests of Dandakal and Kedali, the mountains of Binjha/ Vindhya and Vindhyadhar, Kannauj des, the country of the Kurus, Kamani, Banga and the mountains of Tilangi, Kashmir, Jalandhar, Chaupar and Naipar/Nepal, before stopping at a ‘forest near Kashi’.87 When Visaladeva marries the princess of Malwa, his inlaws gift him the kingdom of Sapadalaksa, Nagarachala, the lake of Sambhar, Toda, Tonk, Kudala, Bundi, Mandor, and in addition, Saurashtra and Gujarat, with the sea.88 This almost believable dowry of contiguous places indicates what a small Marwari chieftain and his poet might have considered a sufficiently grand empire. For the Sanskrit poets of mid-century Junagadh and Champaner, their skill lies in their ‘intricate interaction between local and cosmopolitan geography, between real and Puranic topographies and between local and transregional geopolitics’ (Chapter 8, p. 238), as they trace fabulous places of (p.35) worship alongside detailed route directions, news of disorder in other kingdoms compared to peace at home, and the virtues and drawbacks of princesses from all over the subcontinent. Patrons wanted to show off their influence and investments, how they had brought prosperity and opportunity to their territories, and how they were aware of their world beyond. Poets, similarly, wanted to show off their mastery of classical tropes while customizing their compositions to best please their patrons and display their grasp of specific political realities. Now that the little kingdoms were the loci of the celestial courts, myth had to descend to the qasba, the fort, and the countryside. Representations of landscape and terrain in the texts and paintings of the time cleave closely to the familiar and the believable while stretching the imagination towards fabulous realms. Where were these compositions produced, recited, and enacted? The fifteenth century saw power moving to a new set of cities and fort-towns such as Gwalior, Chanderi, Mandu, Ahmadabad, Champaner, Jaunpur, Kalpi, Gaur, Pandua, and Page 21 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction Hampi.89 With the spotlight away from Delhi, writers, entrepreneurs, artisans, and religious specialists populated the newly popular towns. The warlords who were their customers were not always city-dwellers and often did not have the resources to build great cities. This is perhaps why the city occupies a place of anxiety and desire in many narratives; the patrons ‘may not have had the resources to build such royal cities, but they could do the next best thing, perhaps: patronise poets who could describe such splendour and the political status that it implied’ (Chapter 9, p. 262). The increased opportunities for scribes and writers in these towns suggest that much of the writing, especially by the bureaucrats, prashasti writers and scribes, was done there. But many members of the new elite lived in mud forts, or on the hoof with temporary or shifting ‘courts’; for the locales of many performances of our texts we should envisage modest surroundings.

(p.36) The Religious Public Sphere How Islamic were the Sultanates? Conventionally, in histories of medieval India the chapter on religion begins with the rulers’ supposed religious policy, gleaned either from their statements or from the evaluation of contemporary historians, along two axes, one that sees the ruler pulled in different directions by the (strict, conservative, elite) ‘ulama and the (liberal and popular) Sufis, and the other that pitches the Muslim ruler and his Muslim support base against the Hindu populace and their religious sentiments and practices. How inappropriate, simplistic, and misleading such top-down approach is is thrown into stronger relief in the fifteenth century, when suddenly a much larger number and varied cast of non-Muslim voices and characters appear on the north Indian stage, apparently insensible of living under ‘Muslim rule’. No case is more striking in this respect than that of Braj. After noting Sikandar Lodi’s policy of having mosques built opposite the ghats at Mathura and prohibiting Hindus from bathing in the Yamuna (whether he destroyed any temple there is unsure), Alan Entwistle noted, ‘Ironically, it was during the reign of Sikandar Lodi, a staunch oppressor of Hinduism, that propagators of the emotional variety of devotion to Krishna came in search of the sacred places of Braj’—Nimbarka, Vallabha, and Chaitanya.90 Nor are the archives of religious groups—so important and informative for this period—to be completely relied upon, given the striking selective silences about groups that must have inhabited the same social space. The approach of this book is bottom-up—we begin with the texts and the voices of Jains, Sants, Sufis, and devotees and we try to work out the context in which they co-existed, interacted with the local and regional power holders, and developed their own spaces. Already Nizam al-Din Awliya had encouraged his disciples, many of whom came to his khanqah in Delhi from all over the country, to go and establish their presence in other regions.91 By the fifteenth century, networks of Chishti, Qadiri, Page 22 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction Suhrawardi, and Kubrawi shaykhs and (p.37) their disciples, as well as qalandars and majzubs, criss-crossed north India from Uchch-Multan to Bengal to the Deccan.92 Sufi presence was particularly dense in the eastern regions, as recorded by ‘Abd al-Haqq Dihlawi in his comprehensive tazkira Akhbār al-Akhyār (1642). Thus, beside Delhi, Thanesar and Panipat in north-western India, Sufi centres dotted Bengal (Pandua, Lakhnauti, and Sonargaon), Bihar (Maner, Purnea), the kingdom of Jaunpur, to which Ibrahim Shah Sharqi invited formidable scholars and Sufis like Qazi ‘Abd al-Muqtadir and Qazi Shihab al-Din from Delhi,93 and Avadh (Manikpur, Kara, Rudauli, Kichaucha, Khairabad, Makanpur, Kannauj). Even the small and short-lived kingdom of Kalpi and the nearby ‘Iraj fort sought to attract Sufis, as did the Khaljis of Malwa. We can call this a network since biographical notes show that most Sufis travelled back and forth, either in their close environs or across these regions, typically when they were looking for a spiritual guide or when looking for patronage, and their successors often moved and established themselves in nearby towns.94 Only very few Indian Sufis in this period travelled further afield in other parts of the Islamic world,95 while a few Sufis continued to come to India from central Asia, like Shah ‘Abd Allah, who in the early fifteenth century introduced the Shattari path to India. His imposing royal appearance and cortege, and his challenge to local Sufis to either teach him or be taught by him, led to confrontations in Manikpur, Jaunpur, and Bengal, until the Khalji sultan invited him (p.38) to settle in Mandu where he died in 1485. His disciples established Shattari centres in Bengal, Jaunpur, Gujarat, and of course Gwalior (the seat of Muhammad Ghaus).96 Timur’s invasion of Delhi prompted a further movement of Sufis and scholars southwards and eastwards. Simon Digby recalls Khwaja Muhammad Gisudaraz’s quick decision to leave Delhi with his entourage, and his subsequent peregrinations to Gwalior, Gujarat, and finally the Bahmani kingdom. Others who left Delhi at the time were Shaykh Abu’l Fath, who went to Jaunpur, Mawlana Khwajagi, who left for Kalpi already before the invasion, and the young Ahmad ‘Abd al-Haqq (see Chapter 14).97 Sufis settled in cities and garrison towns along trade and military routes, or in village outposts in less inhabited areas. In his evocation of Mathura, Malik Muhammad Jayasi gives us a sense of how such towns were settled, including his own qasba of Jayas, where a dargah of Ashraf Jahangir Simnani is still standing. The fort is built first, surrounded by a moat. Inside it, the ruler builds his palace with several courtyards and ranis from several countries to enhance his fame. Outside it, tall houses, walled gardens and pavilions are built to accommodate the growing population. Chieftains from neighbouring areas also come to pay their respects, while traders come attracted by good roads, water supplies, and the wealth of chiefs and soldiers. Once healthy trade is established, religious mendicants also start flocking in, and various types of wandering performers and religious specialists ply their trades. They sing and Page 23 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction tell stories to pass the time and entertain the people in the fort, who in turn give handsome rewards.98 It is interesting that Jayasi should mention trade in precious objects, for according to the biographical information we have, many Sufis came from merchant families or had merchants among their local disciples, and Jaunpur was indeed a centre for the trade in precious stones.99 (p.39) Irfan Habib also notes that the new technology of distillation was employed by this time for both liquor and perfume.100 A seventeenth-century history of a local Chishti lineage recalls that one ancestor, called Pir Buddhan and said to be the spiritual guide of Sultan Husayn Sharqi (r. 1458–79), shared with the Sultan a love of music but also of precious perfumes and scented wood, which he would send to the Sultan from his large estate in Rapri (near Chunar).101 All the rulers of this time—the Sharqis of Jaunpur, the Lodis, the Khaljis of Malwa, and the Ilyas Shahis and Husayn Shahis of Bengal, the Sher Shahis— seem to have had close personal relations with Sufi shaykhs and qalandars, and anecdotes of predictions, dreams, supernatural interventions or refusals to do so dot both histories and Sufi tazkiras.102 Such close alliances could and did create enmities when new rulers came into power, as was the case with Babur’s predilection for Naqshbandis over Chishtis, for example, or with Sher Shah’s reprisal against Muhammad Ghaus.103 Several Sufi poets in particular occupied a position that was courtly and spiritual at the same time. Shaykh Jamali, for example, became close to Sultan Sikandar Lodi (r. 1489–1517) at the suggestion of his pir, Shaykh Sama’ al-Din, who had been close to both Bahlul and Sikandar Lodi. He was consequently pushed aside by Sultan Ibrahim Lodi (and in fact wrote an elegy for Miyan Bhuwa, another courtier executed by the Sultan), and welcomed Babur’s arrival with a qasida and followed Humayun on his campaign to Gujarat, where he died. Of his two sons, one became a poet at the court of Sher Shah and Islam Shah, while (p.40) the other, Shaykh Gada’i, remained loyal to the Mughals and removed himself from Delhi for the duration of Sher Shah’s rule. He later became the powerful Sadr-i Sudur, the controller of land grants and stipends for religious purposes, under Akbar, while at the same time continuing to hold sama‘ sessions and write spiritual poetry.104 This dual courtly and spiritual character is at the heart of the striking verse love stories by Avadhi Sufi poets (see Chapter 10). In Behl’s words, this was ‘a powerful Indo-Islamic literary tradition that circulated in courts, bazaars, shrines, and private salons throughout the Sultanate period and in subsequent centuries’ (p. 275). In his essay in this volume, Aditya Behl unravels the multiple layers of signification (bhav) in the story that a listener could draw from the elements, episodes, and characters of the story, guided by an expert storyteller or a spiritual guide.

Page 24 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction Barring a few cosmopolitan Sufis such as Ashraf Jahangir Simnani and Shah ‘Abd Allah who established the Shattari silsila in India, this was a world of provincial Sufis in the sense that Stefano Pellò gives to the term. Although they mostly wrote in Persian and in the classical genres of Sufi religious discourse, they drew upon an earlier canon of Persian spiritual poetry (Rumi, ‘Attar, Nizami) rather than on current trends of the Turko-Persian Timurid world.105 Also, despite their adherence to the protocol of writing in Persian, they also show unmistakeable traces of vernacular use, both in their everyday practices and in their poetic and musical tastes (see Orsini in this volume). So while Sufi sources only mention yogis and not Sants or bhagats, the striking commonalities in their vernacular compositions in this period suggest that they shared a common language and some common spiritual vocabulary, even if they inflected the language and vocabulary differently—how far they shared audiences in north Indian towns remains a matter of speculation. Conversely, Nanak’s early journey and encounters as presented in the Purātan janam-sākhī closely map those of Sufi biographies, suggesting a common patterning of spiritual life.106 (p.41) With the establishment of Sufi centres and their increasing imbrication in local economies in this period, the era of the ghazi or warrior pir can be said to have come to an end. The last of the ghazis in Gujarat was Latif Khan Dawar al-Mulk, remembered as victorious over the Rajputs of north Gujarat (whose descendants are now devotees of his shrine, as with Ghazi Miyan of Bahraich).107 While the narratives associated with the silsila-based pirs were of settled life, established wilaya, and increasing influence in their neighbourhoods, the sagas of non-affiliated pirs also speak of settlement, cultivation, curing cows, and fertility.108 In the context of Gujarat, Samira Sheikh has argued that the fifteenth century saw ‘the making of a stable polity based on the settling of pastoralist clans and their participation in sultanate politics, [and] the rise of religious activity as a lucrative economic sphere’.109 To some extent, this was the case throughout north India. As we have seen, each of the sultanates depended for their survival on local resources of manpower and weaponry. Each was thus the location for political experiments in recruitment. In Gujarat, Mahmud Begada and his son Muzaffar were able to win the support of the major religious figures of the region: these included Shaykh Ahmad Khattu, the Bukhari lineage of Ahmadabad, and the Isma‘ili pir Imam Shah. While in the case of north India polities were less stable, the undeniable activity of religious figures and the emergence of new religious sects points to a similar ‘religious marketplace’, with Vaishnava shrines in Braj, Bhakti gatherings of songs, preaching, and storytelling (satsang), Sufi shrines, and festivals. Religious groupings had now become lucrative for practitioners—witness Kabir’s (p.42) railing against the corruption of priests—and were equally the source of military and ideological support for rulers.

Page 25 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction The new sampradayas of this period—the Nanak- and Dadu-panth, Ramanandis, Gaudiya and Pushti marg, and silsilas—have been seen as developments with no connection to politics. Yet the rise of Vaishnavism is deeply connected with the workings of the north Indian sultanates. As Heidi Pauwels has pointed out, it cannot be coincidence that the ‘recovery’ of Braj as a site for pilgrimage and as the locus of new Vaishnava activity occurred during times of intense political upheaval in the early sixteenth century.110 At one level, pilgrimage was facilitated by the increased security of roads under the Lodi and Sur administrations. Later, during Akbar’s reign, rulers such as Madhukar Shah began to employ the vocabulary of Vaishnava bhakti as a signifier of upward mobility for the status-hungry mobile peasants they were trying to recruit. But even in the late fifteenth century, the groups newly enriched or energized by sultanate polities were reformulating devotional expression, group identity, and networks of pilgrimage. Since the thirteenth century, royal temples had become associated with Hindu kingship and were no longer being built in sultanate territories. Jain sites continued, by and large, to be maintained, refurbished, and patronized by merchants or those in sultanate employment. For other non-Muslim groups, Vaishnavism offered a devotional vocabulary that did not pose the political threat of royal Shaivism or goddess worship. The ‘rediscovery’ of Braj by followers of Vallabha and Chaitanya, among others, connected most of north India into increasingly dense networks of trade and pilgrimage.111 The new Vaishnavisms—with their reliance on the Bhāgavata purāṇa, avatara theory, and intense proselytizing—offered a mutable vocabulary to which many groups responded. The vartas of the Pushtimarg tradition record scores of tales of conversion by agricultural, mercantile, Rajput, and even Muslim groups. The Gaudiya tradition incorporated a similar range too. Even Rajput groups (p.43) formerly associated with goddess or Shiva worship began to link up their genealogies with Krishna (see, for example, Kanhadade, the fifteenth-century ruler of Jalor, whose name is equivalent to Krishna).112 Vaishnavism offered a relatively open vocabulary for sectarian interpretation and investment. The Krishna narratives of the Bhāgavata purāṇa were one set that was creatively interpreted in the sampradaya traditions. For Rajputs and Rajputizing groups, the epic versions of Krishna were more relevant for their evolving genealogical ideologies. Thus Rajputs and Ahirs elaborated versions of the Krishna story, especially parts that recall the exploits of the Pandavas from the Mahābhārata.113 Occupational castes such as the Bairagis also took to Vaishnava narratives of this kind. Vaishnavism was even incorporated within the narratives of mythological pirs, such as Satya Pir, who appears in Vaishnava guise as Satyanarayana.114 The cultural and political inroads of Vaishnavism did not go unresisted: one of the most famous narratives is that of the Krishnaite

Page 26 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction poet Mira, who faced the anger of her Shakta inlaws for her devotion to Krishna, but persisted nevertheless. In rehabilitating the long fifteenth century, this volume has its limitations. It is concerned primarily with the literatures and cultures of north and central India. While there are two chapters on Gujarat, there are none on Jaunpur or Bengal. Both were the seat of important regional sultanates, both saw the flowering of Sufi silsilas and the new Vaishnavism, and both were part of the evolving, linked, literary culture of what we have called Hindavi. Sindh, Kashmir, and Punjab receive hardly any attention either. Richard M. Eaton’s paper on bureaucracy and language is on the Deccan, but there are no other papers on the south. Here, the justification has been the linguistic distinctiveness of southern literary cultures. The Naths and other ascetic groupings that appear often in our period do not receive much attention. The Sants, similarly, vital to the articulation (p. 44) and dissemination of literature in this period, are only in the background of some of the papers here. In spite of these constraints, we believe that a literary and multilingual approach like that of this volume can help bring together what are now separate strands and provide a more connected linguistic, social, and political history of the period, and we hope this volume will act as a spur to further investigations of the period and to more comparative studies. Notes:

(1) On British colonialism, Sheldon Pollock makes a similar point, ‘[W]e cannot know how colonialism changed South Asia if we do not know what was there to be changed’; Sheldon Pollock (ed.), Forms of Knowledge in Early Modern Asia: Explorations in the Intellectual History of India and Tibet, 1500–1800, Durham: Duke University Press, 2011, p. 1. (2) In Malwa, Shihab Hakim wrote Ma’ās̱ir-i Maḥmūd Shāhī (completed 1467–8); six historical works in Persian survive from the Gujarat sultanate (see S. Sheikh, Forging a Region: Sultans, Traders and Pilgrims in Gujarat 1200–1500, New Delhi: Oxford University Press, 2010, pp. 11–12 for details); and for Kalpi, Muhammad Bihamad Khani’s Tā’rīkh-i-Muḥammadī (1438–9), translated by Muhammad Zaki, Bombay: Asia Publishing House, 1972, is an important record. (3) These Sants and bhagats (devotees) are often described in terms of a Bhakti ‘movement’ and a kind of ‘social revolution’, though usually unsupported by socio-historical dimensions and contextualization; for a recent book that seeks to address this question seriously, see Purushottam Agrawal, Akath kahānī prem kī: Kabīr kī kavitā aur unkā samay (Love’s Unspeakable Tale: Kabir’s Poetry and His Times), New Delhi: Rajkamal Prakashan, 2009.

Page 27 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (4) The earliest dated manuscript that contains song-verses by Kabir, Surdas, and the like, Pada Sūradāsajī kā (ed. Gopal Narayan Bahura, with an essay by Ken Bryant, Jaipur: Maharaja Sawai Man Singh II Museum, 1984) is dated 1582, while the Ādi Granth compiled by Guru Arjan is the earliest systematic collection (dated 1604, though possibly earlier versions are found in the Goindwal and Harsahai manuscripts of the early 1570s); see G.S. Mann, The Making of Sikh Scriptures, New Delhi: Oxford University Press, 2001; see also J.S. Deol, ‘Text and Lineage in Early Sikh History: Issues in the Study of the Adi Granth’, Bulletin of the School of Oriental and African Studies, 64(1), 2001: 34–58. (5) Biographies, lists, and dictionaries and lists of model devotees and poetsaints were composed in the seventeenth century within the Vallabha sampraday (e.g., Caurāsī Vaiṣṇavan kī vārtā), the Dadu panth (Raghavdas’s Bhaktamāl), the Ramanandi sampraday (Nabhadas’s Bhaktamāl), and the Sikh panth (Janamsākhīs); for an overview, see W.E. Callewaert and R. Snell (eds), According to Tradition: Hagiographical Writing in India, Wiesbaden: Harrassowitz, 1980. (6) Francesca Orsini (ed.), Before the Divide: Hindi and Urdu Literary Culture, New Delhi: Orient Blackswan, 2010, Introduction. (7) By far the most sophisticated account of South Asian literary history, the essays in Sheldon Pollock’s Literary Cultures in History: Reconstructions from South Asia (Berkeley: University of California Press, 2003a), each deal with one language, and multilingualism is dealt with mostly as diglossia in the relationship between Sanskrit and an Indian vernacular language. (8) See, for example, ‘Abd al-Qadir Bada’uni, Muntakhab al-tavārīkh, W.H. Lowe (ed.) and W. Haig (trans.), Calcutta: Asiatic Society of Bengal, 1898–1925, vol. 3, p. 238ff. (9) See María Angeles Gallego, ‘The Languages of Medieval Iberia and Their Religious Dimension’, Medieval Encounters, 9(1), 2003: 107–39. (10) We consider Hindavi here as synonymous with bhakha. Though modern scholarship distinguishes between western and eastern Hindi, and between Avadhi, Braj Bhasha, Khari Boli (Hindi and Urdu), and so on, it is our contention —supported by the wide circulation of texts like the ‘Avadhi’ Candāyan in Delhi and of Kabir’s poems or Gvaliyari dhrupad all over North India—that vernacular (or bhakha, ‘language’, as they are called in vernacular sources) literary forms travelled easily and widely at this time within a unified language domain, and that script was a function of written transmission and not intrinsic to a language, at least in this case. True, terms like ‘eastern’ (purbi) and ‘of Gwalior’ (gvaliyari) were also sometimes used in this period, but it was only at the end of sixteenth century that Braj Bhasha emerged as a separate, specific, (partly) codified

Page 28 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction literary vernacular; see Allison Busch, Poetry of Kings: The Classical Hindi Literature of Mughal India, New York: Oxford University Press, 2011b. (11) For references to Afghans’ poor command of Persian, see I.H. Siddiqui, Waqi‘at-e-Mushtaqui of Shaikh Rizq Ullah Mushtaqui: A Source of Information on the Life and Conditions in the Pre-Mughal India, New Delhi: Indian Council of Historical Research and Northern Book Centre, 1993, p. 9; for the ‘spoken Persian’ hymns of the early Sikh Gurus, see C.S. Shackle, ‘Approaches to the Persian Loans in the Ādi Granth’, Bulletin of the School of Oriental and African Studies, 41(1), 1978: 73–96. (12) Malik Muhammad Jayasi, Padmāvat: Malik Muhammad Jāyasī kṛt mahākāvya, Vasudev Sharan Agraval (ed.), Chirgaon: Sahitya Sadan, 1998 [1956], pp. 499, 1–4, 6–7. (13) Jayasi, Padmāvat (1998), p. 527. A vernacular tale composed for Baghela patrons from 1493 who were familiar with, as Simon Digby puts it, ‘the governmental framework of Muslim power in the Gangetic plain’ shows a comparatively high number of Perso-Arabic words related to the military: jin/zin (reins), samser (sword), tirandāz (archer), taslīm (subordination), asrar/iṣrar; see Bhima Kavi’s Ḍaṅgvai kathā, S. Misra (ed.), Allahabad: Hindi Sahitya Sammelan, 1966; S. Digby, ‘Two Captains of the Jawnpur Sultanate’, in Circumambulations in South Asian History: Essays in Honour of Dirk H.A. Kolff, J. Gommans and O. Prakash (eds), Leiden: Brill, 2003, p. 165. (14) The first manuals for learning Persian in India were written in Sanskrit by Jains; see S.R. Sarma, ‘Sanskrit Manuals for Learning Persian’, in Adab Shenasi, A.D. Safavi (ed.), Aligarh: Aligarh Muslim University, 1996, pp. 1–12. According to tradition, Guru Nanak himself got some Persian (‘Torki’) education through a maulvi and there are a few Persian and mixed-language compositions in the Ādi Granth; W.H. McLeod, Early Sikh Tradition: A Study of the Janam-sākhis, Oxford: Clarendon Press, 1980, and Shackle (1978), pp. 82–3. (15) A land-grant from a village near the qasba of Sandila (now dist. Hardoi) to a local shaykh from Sultan Ibrahim II, son of Sikandar Lodi, dated 927 H (1520), records the deed in Persian ta‘liq and in Hindi kaithi script below, and there is a similar (though illegible in print) grant from Sher Shah’s time; M. Mohammad Shafi, ‘Three Old Documents’, in Proceedings of the Idara-i Maarif-i Islamia, Lahore, 1936, pp. 281–5. (See Figure 1.1). (16) See H.N. Dvivedi, Mān Simh aur Mān-kutūhal, Gwalior: Vidya Mandir Prakashan, 1956 reprint. (17) After the Ghunyat al-munya (1374–5), commissioned by the governor of Gujarat Malik Shams al-Din Ibrahim Hasan Abu Khan [Rajā], the Laḥjat-i Sikandar Shāhī was composed in our period by Hammad Yahya al-Kabuli during Page 29 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction the reign of Sikandar Lodi (r. 1489–1517); see F.N. Delvoye, ‘Les chants Dhrupad en langue Braj des poètes-musiciens de l’Inde Moghole’, in Littératures médiévales de l’Inde du Nord. Contributions de Charlotte Vaudeville et de ses élèves, Françoise Mallison (ed.), Paris: Publications de l’École française d’Extrême-Orient, 1991, pp. 139–85. See also Madhu Trivedi’s The Emergence of the Hindustani Tradition: Music, Dance, and Drama in North India, 13th to 19th Centuries, New Delhi: Three Essays Collective, 2012. (18) C. Minkowski, ‘King David in Oudh: A Bible Story in Sanskrit and the Just King at an Afghan Court’, Inaugural Lecture for the Boden Professorship, University of Oxford, 7 March 2006, http://users.ox.ac.uk/~ball2185/ Minkowski.Inaugural.pdf (accessed 14 June 2014). (19) C.D. Dalal and G.K. Shrigondekar (eds), Lekhapaddhati, Baroda: Central Library, 1925, Preface, p. vii. There are several earlier precedents for such reverse Sanskritization of Apabhramsha words; see R. Salomon, ‘The Ukti-vyaktiprakaraṇa as a Manual of Spoken Sanskrit’, Indo-Iranian Journal, 24(1), 1982: 13–25. (20) For example, the Sanskrit kavya on Sultan Mahmud Begada; see A. Kapadia, ‘The Last Cakravartin? The Gujarat Sultan as “Universal King” in Fifteenth Century Sanskrit Poetry’, The Medieval History Journal, 16(1), 2013: 63–88. (21) See Allison Busch, ‘Culture from the Cowherd’s Mountain: Gwalior and the Early History of Courtly Brajbhasha’, paper presented at the conference ‘After Timur Came’, SOAS, London, June 2007. (22) Another case could be Jaunpur, a courtly environment in which both romances and songs in the vernacular were patronized as well as Persian poetry and scholarship; for romances, see Behl (Chapter 10 in this volume) and his translation of Madhumālatī (with S. Weightman), Madhumālatī: An Indian Sufi Romance, Oxford: Oxford University Press, 2000; for music, see the mention in Digby (Chapter 2 in this volume) as well as Allyn Miner’s paper ‘Ragas and Raginis, Sufis and Sants: Music in North India in the Early Sixteenth Century’, forthcoming in F. Orsini and K. Schofield (eds), Tellings and Texts: Singing, Story-telling and Performance in North India; for Persian poetry at Jaunpur, see the Dastūr al-shu‘ara’ otherwise known as the ‘Jaunpur Anthology’, Ms. Or. 4110; for a short description of its contents, see C. Rieu, Supplement to the Catalogue of the Persian Manuscripts in the British Museum, London: British Museum, 1895, pp. 232–3. (23) Sheldon I. Pollock, The Language of the Gods in the World of Men: Sanskrit, Culture, and Power in Premodern India, Berkeley: University of California Press, 2006. (24) Pollock (2006), p. 337. Page 30 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (25) Pollock (2006), p. 318. (26) Pollock (2006), p. 298. (27) And prashasti, or ‘workly’ political discourse in inscriptions. (28) Pollock (2006), p. 299. (29) At the end of a very articulate and convincing plea for the importance and radical innovation of writing for literature, Pollock comes to some rather startling conclusions: Only authors of written work are included in the canons included in ethnohistorical accounts of literature; the oral poets stands entirely outside of history … Such oral culture is not only unknowable in its historicity, it is excluded from the literary history made by committing texts to writing … It is no redundancy to say that a literary work does not exist until it becomes literate. (2006, pp. 317–18) This seems hard to adhere to in light of the robust oral literary life of Kabir, Surdas, Mira, and so on. Elsewhere he acknowledges that ‘the ongoing interaction between the oral and the literate constitutes one of the most remarkable and unique features of Indian literary culture. If oral compositions could be literized, literized compositions could also return to oral circulation, and the interplay between oral and literate composition and transcription could become dizzyingly complex.’ (2006, p. 316) (30) See the examples in Shackle (1978), pp. 73–96. (31) This is the kind of literary culture analysed by Allison Busch in her Poetry of Kings (2011b); but to say that it was a ‘singularly influential form of culture that occupied the entire conceptual domain of aestheticized language use’ (Pollock [2006], p. 322) seems unwarranted. (32) See Agrawal (2009) and C.L. Novetzke, ‘Bhakti and its Public’, International Journal of Hindu Studies, 11(3), 2007: 255. (33) To avoid confusing this early definition of the north Indian vernacular with modern Hindi (Khari Boli), we will use Hindavi, as previously indicated. (34) See in contrast the eighteenth-century writer Anandram Mukhlis’s comment on the ‘sweetness of the purabi tongue’ when he heard it recited by his servant; S. Phukan, ‘“Through Throats Where Many Rivers Meet”: The Ecology of Hindi in the Persian Imagination’, Indian Economic and Social History Review, 38(1), 2001: 35.

Page 31 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (35) If not grammars per se (until the Persian Tuḥfat al-Hind, 1675). See Busch (2011b), who also takes up the suggestion that this codification (or choice of name?) was partly influenced by the adoption of Braj Bhasha by Bengali Gaudiya Vaishnavas in their mixed poetic language of Braj Buli (mixed Braj Bhasha and Bengali). (36) The songs collected in his Kṛṣṇa gītāvalī (ca. 1590), Gītāvalī and the poems in the Vinaya patrikā, Kavitāvalī, and many of the verses in the Dohāvalī; see R.S. McGregor, Hindi Literature: From the Beginnings to the Nineteenth Century, Wiesbaden: Harrassowitz, 1984, pp. 114–57. (37) See C. Vaudeville, Kabir, Oxford: Clarendon Press, 1974. For word substitution by singers, W. Callewaert and M. Lath, ‘Musicians and Scribes’, in The Hindi Padāvalī of Nāmdev, Delhi: Motilal Banarsidass, 1989, pp. 55–117. (38) Christian Novetzke’s important insight on Marathi texts and composers’ notebooks seems very relevant to the kind of manuscripts available for fifteenthcentury literature. See C.L. Novetzke, ‘Note to Self: What Marathi Kirtankars’ Notebooks Suggest about Literacy, Performance, and the Travelling Performer in Pre-Colonial Maharashtra’, forthcoming in Tellings and Texts: Singing, Storytelling and Performance in North India, Orsini and Schofield (eds); see also Religion and Public Memory: A Cultural History of Saint Namdev in India, New York: Columbia University Press, 2008. For North India, see Callewaert and Lath (1989). (39) As regards the introduction of paper and paper manufacture in India, the information is patchy. P.K. Gode argues that paper was introduced into India from China around 1000 and was manufactured in Delhi by the mid-fourteenth century. See Gode, ‘Migration of Paper from China to India—A.D. 105–1500’, in Studies in Indian Cultural History, vol. 3, Poona: Bhandarkar Oriental Research Institute, 1969, pp. 1–12. Brac de la Perrière quotes Barani’s observation that fourteenth-century booksellers took unsold books back to papermakers (kāghaziān), who would wash and reuse the paper. A fifteenth-century Iranian treatise mentions paper from Bengal among the kinds of papers available (though the reading might be wrong), while a contemporary tazkira of poets from Kashmir mentions papermaking in the region. By the fifteenth century, almost all Jain manuscripts from Gujarat were copied on paper; E. Brac de la Perrière, L’art du livre dans l’Inde des sultanats, Paris: PUP, 2008, pp. 96–7. See also Chapter 4 in this volume (p. 121n23). (40) Reference is usually made to ‘Abbas Sarwani, who says that Sher Shah appointed for every pargana one shiqdar (governor), one ‘amil (revenue collector), one fotahdar (treasurer), and two karkuns (writers), one for Hindavi and the other for Persian; see ‘Abbas Khan Sarwani, The Tā’rīkh-i Sher Shāhī, vol. 1, ed., S.M. Imamuddin, Dacca: University of Dacca, 1964, p. 210. See also Page 32 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction K.R. Qanungo, Sher Shah, Calcutta: Kar, Majumdar, 1921, p. 351. The document reproduced in Figure 1.1 shows that this bi-scriptual practice was in use under the Lodis as well; see Shafi (1936), pp. 281–5. See also M. Momin, The Chancellery and Persian Epistolography under the Mughals, from Bábur to Sháh Jahán, 1526–1658, Calcutta: Iran Society, 1971, p. 28. (41) ‘Kāfirān bakhwāndan o–neveshtan-i khaṭ-i fārsī ki tā ān zamān dar miyān-i īshān ma‘mūl nabud, pardākhtand’, in Nizam al-Din Ahmad’s Ṭabaqat-i Akbarshāhī, quoted in ‘Sikandar Lodhī aur uske ‘ahad ke ba‘ż fārsī muṣannifin’, Oriental College Magazine, 32, 1932: 29. (42) Text, transcription, and translation (slightly modified) in R.B. Hiralal, ‘Damoh Hindi Inscription of Mahmud Shah II of Malwa: (Vikrama) Samvat 1570’, Epigraphia Indica 7 (1920), pp. 291–3. Similar inscriptions from Gujarat are discussed in Chapter 7 in this volume. (43) P. Prasad, Lekhapaddhati: Documents of State and Everyday Life from Ancient and Early Medieval Gujarat, New Delhi, Oxford University Press, 2007. Two of the manuscripts are dated 1475/6 and 1464, and the significant variation between the successive manuscripts (including the insertion of new mode letters, e.g., Nos. 35, p. 116; 46, p. 135; 73, p. 187; 78, p. 192) suggests that the copyists made changes according to current usage. For letters dated 1475, see pp. 70, 75, 87, 110, 111, 114, 116. (44) At least in Delhi, Malwa, and Gujarat, Jains functioned as ministers or senior bureaucrats. As Eva De Clercq’s contribution shows, the biographies of Jain patrons in Apabhramsha texts reveal a network of families and associates between Delhi (Yoginipura) and Gwalior as well. The persistence of Sanskrit can also be traced to the much understudied presence and agency of Jain merchants, some of whom commissioned texts and biographies in ‘Jain’ Sanskrit. (45) Reference in Brac de la Perrière (2008), p. 127ff., especially pp. 140, 143–4. (46) See, for example, B.N. Goswamy, A Jainesque Sultanate Shahnama and the Context of Pre-Mughal Painting in India, Zürich: Museum Rietberg, 1988. For evidence of Jain illustrated (and non-illustrated) manuscripts copied in Delhi (Yoginipura) from the fourteenth to the early sixteenth centuries, see B.N. Goswamy, ‘In the Sultan’s Shadow: Pre-Mughal Painting in and around Delhi’, in Delhi Through the Ages, R.E. Frykenberg (ed.), Delhi: Oxford University Press, pp. 137–9. The colophon of one Ādipurāṇa dated 1404 (VS 1461) says: ‘Here, today, in prosperous Yoginipur, where many resplendent feudal chiefs preside. During the reign of Sultan Sri Muhammad Shah (Nasir al-Din Muhammad Tughluq?).’ Other Jain texts in Apabhramsha composed in Delhi include a Pāsṇāh cariu, by Sridhar Ayarval (Agraval), a Śāntināth cariu by one Mahindu in 1530;

Page 33 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction see H.V. Kochhar, Apabhraṃśa-sāhitya, Delhi: Hindi Anusandhan Parishad, 1959, pp. 210–45. (47) See Brac de la Perrière (2008). (48) See, for example, the illustrated copy of the Mirigāvatī in kaithi script now in the Bharat Kala Bhavan, Banaras (accession number 7742–91), which, in absence of a colophon, Karl Khandalavala dates on iconographic grounds to 1525–70; K. Khandalavala, ‘The Mṛgāvat of Bharat Kala Bhavan: As a Social Document and its Date and Provenance’, in Chhavi: Golden Jubilee Volume, A. Krishna (ed.), Banaras: Bharat Kala Bhavan, 1971, pp. 30–2. If we hypothesize that kaithi manuscripts of these texts were copied for local Rajput or merchant elites, then this would be a very early evidence of investment in an illustrated codex, given that the illustrated Bhāgavata purāṇa folios dated to the early sixteenth century ‘from the Delhi-Agra region’ are in Sanskrit, and that illustrated copies of Keshavdas’s poems are significantly later. (49) See P.J. Shastri (ed.), Jain granth praśasti saṅgrah, Delhi: Vidya Sewa Mandir, 1954–63. (50) Rajeev Kinra notes that Chandar Bhan’s father Dharam Das, a Panjabi Brahmin from Lahore, had ‘mastered Persian well before any of Akbar and Todar Mal’s revised administrative and educational politics took effect’; R. Kinra, ‘Secretary-Poets in Mughal India and the Ethos of Persian: The Case of Chandar Bhān Brahman’, PhD thesis, University of Chicago, 2008, vol. 1, p. 21; for Hindus as scribes in the Mughal administration, see M. Alam, ‘The Culture and Politics of Persian in Precolonial Hindustan’, in Pollock (2003a); and M. Alam and S. Subrahmanyam, ‘The Making of a Munshi’, Comparative Studies of South Asia, Africa and the Middle East, 24(2), 2004: 61–72. (51) Sumit Guha has argued that it had its roots in the professional specialization of scribes. Sumit Guha, ‘Mārgī, Deśī and Yāvanī: High Language and Ethnic Speech in Maharashtra’, in Mārga: Ways of Liberation, Empowerment, and Social Change in Maharashtra, M. Naito, I. Shima, H. Kotani (eds), New Delhi: Manohar, 2008, p. 133. (52) For a powerful critique of this tendency, see Agrawal (2009). (53) See K.S. Lal, Twilight of the Sultanate: A Political, Social and Cultural History of the Sultanate of Delhi from the Invasion of Timur to the Conquest of Babur 1398–1526, Delhi: Munshiram Manoharlal, 1980, which remains the most comprehensive history of the period. (54) Peter Jackson, The Delhi Sultanate: A Political and Military History, Cambridge: Cambridge University Press, 1999, pp. 314, 316–17. (55) Jackson (1999), p. 322. Page 34 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (56) This is cited by Lal (1980), p. 124, n. 64, citing Ahmad Yadgar about the last Sayyid ruler ‘Ala’ al-Din ‘Alam Shah. The remark reappeared in the eighteenth century when it was applied to the Mughal Shah ‘Alam II (r. 1759–1806). (57) Jackson (1999), p. 324. (58) Shams al-Din Damaghani offered a higher bid for the governorship of Gujarat than the incumbent could offer. In addition to the normal tribute, he promised an additional payment of 100 elephants, 200 Arab horses, and 400 Hindu and Abyssinian slaves. S.C. Misra, Rise of Muslim Power in Gujarat, London: Asia Publishing House, 1982 [1963], p. 130. (59) P. Hardy, ‘Fārūḳids’, in Encyclopaedia of Islam, P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel and W.P. Heinrichs (eds), Leiden: E.J. Brill, second edition. (60) Misra (1982), pp. 143–4. (61) See Dirk Kolff, Naukar, Rajput, Sepoy: The Ethnohistory of the Military Labour Market in Hindustan, 1450–1850, Cambridge: Cambridge University Press, 1990. (62) Sikandar b. Muhammad, Mir’āt-i Sikandarī, S.C. Misra and M.L. Rahman (eds), Baroda: M.S. University, 1961, p. 144. (63) V.S. Pramar, ‘The Effects of Trade and Urbanization on the Architecture of Gujarat’, in Studies in Trade and Urbanization in Western India, V.K. Chavda (ed.), Baroda: M.S. University, 1985, p. 86. (64) See S.R. Sarma, Ṭhakkura Phurū’s Rayaṇaparikkhā: A Medieval Prakrit Text on Gemmology, Aligarh: Viveka, 1984. (65) Kolff (1990), p. 57; S. Digby, ‘Dreams and Reminiscences of Dattu Sarvani a Sixteenth Century Indo-Afghan Soldier (Part 1)’, Indian Economic and Social History Review, 2(1), 1964: 64; Ni‘mat Allah, History of the Afghans, London: Oriental Translation Committee, 1829–36, p. 164n. (66) See Chapter 7 in this volume. (67) Kolff (1990), p. 56, citing Qanungo, Sher Shah, pp. 66–75. (68) C. Talbot, ‘Becoming Turk the Rajput Way: Conversion and Identity in an Indian Warrior Narrative’, Modern Asian Studies, 43(1), 2009: 225–6, citing the Kyamkhān rāsa. The alliance between the Kayamkhanis and Rathors persisted into the seventeenth century: when the Mughal emperor Jahangir ordered the Kayamkhani chief to proceed against a Rathor, Dalpat, the latter invoked their

Page 35 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction long alliance and succeeded in getting the Kayamkhani to ignore the imperial order; Talbot (2009): 241. (69) Pushpa Prasad, Sanskrit Inscriptions of the Delhi Sultanate, 1191–1526, Delhi: Oxford University Press, 1990, pp. 183–4. (70) Talbot (2009), p. 4. (71) Talbot (2009), pp. 5, 6. (72) Sikandar (1961), p. 3. (73) Sikandar (1961), p. 284. See Chapter 9 in this volume. (74) Sikandar (1961), p. 253. (75) B.D. Chattopadhyaya, ‘Origin of the Rajputs: The Political, Economic and Social Processes in Early Medieval Rajasthan’, in The Making of Early Medieval India, New Delhi: Oxford University Press, 1993, pp. 57–88. (76) Kolff (1990), p. 57. (77) Kolff (1990), p. 58. (78) Ruby Lal, Domesticity and Power in the Early Mughal World, Cambridge: Cambridge University Press, 2005, Chapter 2. (79) See Chapter 2 in this volume. (80) Prasad (1990), pp. 183–99. (81) Prasad (1990), pp. 37–40. (82) Bhima Kavi (1966). (83) See Chapter 7 in this volume. (84) Narapati Nalha, The Visaladevarasa, J.D. Smith (trans.), Cambridge, New York: Cambridge University Press, 1976, v. 36. (85) See Sreenivasan’s account of jauhar in Raisen in 1532 on pp. 253–4 in this volume. (86) Digby (1964), pp. 65, 71–2. (87) Bhima Kavi (1966); see F. Orsini, ‘Travelling Tales’, in F. Orsini and K. Schofield (eds), Tellings and Texts, forthcoming. (88) Nalha, Vīsaladevarāsa, vv. 20, 21.

Page 36 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (89) Ahmadabad and Champaner were purpose-built in the fifteenth century while several others were of thirteenth- or fourteenth-century origin, considerably expanded or fortified in the fifteenth century. On Chanderi, see Gérard Fussman and K.L. Sharma, Naissance et déclin d’une qasba: Chanderi du Xe et XVIIIe siècle, Paris: Publications de l’Institut de Civilisation Indienne du Collège de France, 2003. (90) A.W. Entwistle, Braj: Centre of Krishna Pilgrimage, Groningen: Egbert Forsten, 1987, pp. 135–6. (91) On Chishti dispersal in the fifteenth century, see Carl W. Ernst and Bruce B. Lawrence, Sufi Martyrs of Love: The Chishti Order in South Asia and Beyond, Basingstoke: Palgrave Macmillan, 2002, Chapter 5. (92) See S.A.A. Rizvi, A History of Sufism in India, New Delhi: Munshiram Manoharlal, 1978, vol.1 for a comprehensive list; Ashraf Jahangir Simnani alone established a network of centres. (93) Among them, Shaykh Abu’l Fath, Shaykh Ahmad ‘Isa, Shaykh Baha’ al-Din and his son Shaykh Adhan Jaunpuri; Rizvi (1978), pp. 262–4. (94) For example, Shaykh Qutb-i ‘Alam disciple Shaykh Husayn Dhukhaposh moved from Pandua to Purnea, where he established his own khanqah; Rizvi (1978), p. 260. (95) One exception is Shaykh Jamali of Delhi who embarked on a long journey in the reign of Bahlul Lodi; after the Hajj he travelled to the Maghreb, Yemen, Turkey, Syria, Iraq, Iran, Khorasan, and visited Sri Lanka on his way back to Delhi. As Rizvi notes, ‘He was keenly interested in collecting stories about Indian Sufis who had lived or travelled in various parts of the Islamic world’; Rizvi (1978), pp. 286–7. (96) Rizvi (1978), pp. 152–6. (97) Rizvi (1978), pp. 247, 262, 273–4. (98) Malik Muhammad Jayasi, Kanhāvat [1540], S.S. Pathak (ed.), Allahabad: Sahitya Bhavan, 1981, stanzas 11–12, pp. 10–11. (99) For example, Shaykh Jamali belonged to the Kanbo Sunni merchants who, according to Rizvi, rose to considerable prominence during the reign of the Lodis; Rizvi (1978), p. 286. Several traders crop up in the stories relating to the biography of Shaykh Ahmad ‘Abd al-Haqq of Rudauli, whose most trusted disciple, Shaykh Bakhtiyar, had been the servant of a jewel merchant in Jaunpur; ‘Abd al-Quddus Gangohi, Anwār al-‘uyūn (1484), with Urdu tr., Lucknow: Matba-i Mujtaba’i, 1909, pp. 74 ff.

Page 37 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (100) I. Habib, Medieval India: A Study in Civilization, New Delhi: National Book Trust, 2007, p. 67. (101) See Hafiz Mahmud Sherani’s summary and translation into Urdu of Shaykh ‘Ala’ al-Din Barnawi’s Chishtiya Bihishtiya (1655); ‘Makhdūm Shaikh Bahā’ al-Dīn Barnāwī’, Oriental College Magazine, 1, August 1927: 41–58. (102) For an account and discussion of such dreams and predictions, see Digby (1964). (103) See Rizvi (1978), p. 157. (104) Apparently, he revoked the grants previously given to a number of ‘ulama and shaykhs who had supported the Afghans; Rizvi (1978), p. 288. (105) Barring Shaykh Jamali, who proudly recalled his conversations with Jami in Herat; Rizvi (1978), p. 287. (106) See W.H. McLeod, Gurū Nānak and the Sikh Religion, New Delhi: Oxford University Press, 1986. The double succession of kin (to the gaddi) and spiritual ‘disciples’ at least for the first gurus, also reproduces Sufi succession, while the practices around the holy book in the gurudwara (the chadar, the flywhisk, the chanting) are strongly reminiscent of practices at a Sufi dargah. For an analysis of the Sikh Gurus’ use of Persian vocabulary, see Shackle (1978). (107) Sheikh (2010), p. 156. On Ghazi Miyan, see Shahid Amin, ‘On Retelling the Muslim Conquest’, in History and the Present, P. Chatterjee and A. Ghosh (eds), Delhi: Permanent Black, 2002, pp. 19–32. (108) Such as some of the narratives of Satya Pir in Bengal. See Tony K. Stewart, ‘Alternate Structures of Authority: Satya Pīr on the Frontiers of Bengal’, in Beyond Turk and Hindu: Rethinking Religious Identities in Islamicate South Asia, David Gilmartin and Bruce B. Lawrence (eds), New Delhi: India Research Press, 2002, pp. 21–54. (109) Sheikh (2010), p. 172. (110) Heidi Pauwels, ‘The Saint, the Warlord, and the Emperor: Discourses of Braj Bhakti and Bundela Loyalty’, Journal of the Economic and Social History of the Orient, 52(2), 2009: 187–228. (111) C. Vaudeville, ‘Braj Lost and Found’, in Myths, Saints and Legends of Medieval India, Vasudha Dalmia (ed.), Delhi: Oxford University Press, 1999, pp. 47–71. (112) Padmanabha, Kānhaḍade prabandha, K.B. Vyas (ed.), Jaipur: Rajasthan Puratattva Mandir, 1953. Page 38 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Introduction (113) See Stuart Blackburn, Peter J. Claus, Joyce B. Flueckiger, and Susan S. Wadley, Oral Epics in India, Berkeley: University of California Press, 1989; Alf Hiltebeitel, Rethinking India’s Oral and Classical Epics: Draupadi among Rajputs, Muslims and Dalits, Chicago: University of Chicago Press, 1999. (114) See Stewart (2002).

Access brought to you by:

Page 39 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

After Timur Left North India in the Fifteenth Century* Simon Digby

DOI:10.1093/acprof:oso/9780199450664.003.0002

Abstract and Keywords This chapter depicts the political landscape of north India in the wake of Timur’s invasion. Going against the popular notion that Timur’s appearance was an aberration with little long-term effect, the essay contends that there were long connections of patronage and kinship that preceded and followed the invasion as Sultanate lieutenants set up rule in areas they had been appointed to govern. The chapter is particularly illuminating on the kin and political networks of the Afghans in north India, drawing attention to the multilocational nature of power in this period. Keywords:   Sultanate, Timur, Afghans, Sufis, Tughluqs, horse trade, Baghelas

AFTER ATTENDING THE ‘URS OF Nizam al-Din Awliya in Delhi, I came across an Urdu translation of his malfuzat or conversations, the Fawā’id al-fu’ād. Here was, I realized, a source for social and economic relations in the sultanate period, though it was clearly a literature to be interpreted with considerable caution.1 By comparison with the Fawā’id al-fu’ād, in the malfuzat of Sayyid Muhammad Gesudaraz (d. 1422) one sees considerable cooperation between the sultan and the great shaykh for the preservation of the community. For example, they contain a detailed account of a message sent by ‘Ala’ al-Din Khalji to Nizam al-Din saying that the Mongols were coming and urging him to get his men inside the city, which Nizam al-Din did. An assembly of mawlanas and ‘ulama and (p.48) Sufis—all sat on the rooftop of the jama‘atkhana praying for the sultan’s victory, and disciples in the sultan’s army brought tidings of how the battle was Page 1 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left going. Such direct dealings between Sufi shaykhs and sultans are usually not entered into the official record. In a similar way, any kind of commercial activities are not directly mentioned, and one must read against the grain in order to find them.2 In order to make sense of north India after Timur left, one needs some knowledge of affairs in Transoxiana (Turkestan) and in Persia, of the disintegration that was taking place there and of the reintegration that took place with Timur’s slow rise to power over amirs of non-Chingizid descent of comparable status, a process more or less completed by 1370.3 Timur was a very able tactician. He prepared advance news and information services from almost every quarter of the Eurasian world and calculated his chances carefully. His campaign in India and conquest of Delhi in 1398–9 was not an attempt at integrated conquest, nor was it an attempt at plunder and loot. Rather, his Indian campaign was conducted in order to secure an influence over important adjacent territories so that no larger state formations could emerge in that direction. In this respect, it is comparable to his campaign in Anatolia against the rising power of the Ottomans. If we look at the way Timur’s two campaigns were conducted—the Indian campaign and the Ottoman campaigns three years later in 1402—certain common features arise. First, the imperative to smash central authority. Second, the establishment of a series of subsidiary alliances within the country, both at the time of the invasion, when Timur waited long enough to win over neighbouring chieftains and men of power, and afterwards. According to this strategy, one must attack the most dangerous enemy, but other chieftains and powerful neighbours may be more usefully maintained in power and induced to acknowledge one’s suzerainty and pay one as much as seems useful. (p.49) This is particularly true of the way in which Timur behaved towards the sultan of Kashmir, Sikandar Butshikan, the idol-breaker (r. 1389–1413).4 Timur demanded from the sultan less tribute than his assessors had determined and also took one of his sons—the future Sultan Zayn al-‘Abidin—as hostage to Samarkand. This is of some importance, since those who went to Samarkand, the new capital built by Timur, very often profited from the contacts they made with Timurid society and from what they saw of industrial techniques. Frontier chieftains had a tendency to join the Mongol force—of which only a very small proportion was in fact Mongol—and we find that a Khokar chieftain, Khizr Khan, who was sent to Timur as an ambassador and negotiator for tribute from the most adjacent area, the Punjab, ultimately became the power-holder in Delhi, thanks to the contacts he had acquired.5 This state of affairs can be compared with what happened in Ottoman Turkey when Bayazid the Thunderbolt was on the verge of taking Constantinople after defeating and completely wiping out the large Christian alliance force at Nicolopolis in 1396 in one of the most notable victories of the century. The Page 2 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left flower of the French nobility, apart from twenty people exchanged for the maximum sum of money, were slaughtered in front of Bayazid, while the King of Hungary barely managed to escape. Yet, in 1402, Timur defeated Bayazid in a brilliant campaign, and literally wiped out the Ottoman force. Here again Timur reduced the frontier chieftains whom Bayazid and his immediate predecessors had eliminated on the eastern borders of Anatolia, and reinstated their small local dynasties. Undoubtedly, the consequence of Timur’s campaign was that the Ottoman capture of Constantinople was postponed by fifty years, for it took the Ottomans half a century to recover. (p.50) At this time, the Ottomans were a rising power, whereas the sultans of Delhi were a declining power, partly due to the end of their resources. The dethesaurization of the Indian subcontinent by the looting of the Delhi sultans had reached its climax when their power stretched to Kanyakumari and there was nowhere else to acquire the huge stores of money by which their vast armies were maintained. The reign of Firuz Shah Tughluq, in the second half of the fourteenth century, is one of gradual loss of power over the provinces. It is worth noting as a general feature of what happened after Timur’s campaign to India that in no case in the areas under sultanate control did power pass to a native of the area. Everywhere power passed on to the sons of those who had been appointed as the commanders of the area. Leaving aside Delhi for the moment, it was a slave-administrator of Firuz Shah Tughluq from the south Andhra coast who became the ruler of Jaunpur. In fact, there are traces of this connection in what we know of the patronage of the sultans of Jaunpur, in that they imported South Indian musicians to Jaunpur and practised the South Indian style of music.6 In Khandesh we find again the local representative of the Sultanate becoming the ruler at the time of its collapse, and the same in Gujarat and in the small state of Kalinjar, which enjoyed an existence of half a century. In all these areas we find former officers of the greater Sultanate becoming local rulers. Let me give a brief sketch of the powers competing against one another in fifteenth-century India. At the extreme west there was the (p.51) Afghan presence. The Afghans were multi-locational and had several power bases; this was in fact a feature of the politics in this time of disunity. After the Sayyid dynasty, who had descended from Khizr Khan, a local representative of Timur and the Timurid interest, the tribes of Afghan warriors who took over Delhi in the middle of the fifteenth century all came from a small area in the North-West Frontier near the Gomal river and the Suleiman mountains. For reasons not entirely clear, possibly the disturbances in Kabul, this extremely difficult pass which requires one to descend in a single file had become the main centre of the horse caravans descending from one terminus in the Bukhara area to the other terminus, the great Indian plains. Horses were the most important military animals and they were in the hands of the people who lived at a frontier town at Page 3 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left the foot of the pass called Daraband (known under various spellings, meaning ‘the pass close’). I saw the last survivors of these horse caravans, great processions with thousands of men, with women and children marching beside their Bactrian camels, when I was a young man, though by that time the route had changed to the Khyber further north and they mainly engaged in the trade of dry fruit from Gilgit. They were still a formidable armed entity on the march, with very small boys carrying rifles, women unveiled and sometimes also armed —altogether several thousand people. The caravans going into the north Indian plains in the fourteenth and fifteenth centuries had to be sufficiently powerful in their arms and their fighting tactics to be able to defeat any local ruler who would rather confiscate their horses and other goods than pay for them. There is an instance recorded in the fifteenth-century Sufi tazkira of ‘Abd al-Quddus’s predecessor, Shaykh Ahmad ‘Abd al-Haqq of Rudauli (d. 1434),7 in which the local ruler tried to lay his hands on the caravans and the caravans had to give fight.8 We can find this kind of migratory activity in some of the other emerging regional sultanates too. In the accounts of the sultans of Malwa, a smaller entity further south from Jaunpur and sandwiched between Jaunpur, Gujarat, and the rising power of the Ranas of Mewar, we find that when attacked by his (p.52) neighbours in Gujarat, the Sultan himself went off on a great trading expedition to acquire horses: In the year 825 A.H. (1421) … Sulṭān Hūshang selected one thousand horsemen out of his army; and in the garb of merchants advanced towards Jājnagar. He took some silver gray and iron gray horses which the Rāy of Jājnagar was very fond of, and some other kinds of merchandise, which the people of that country took with pleasure. His object in taking this journey was this, that in exchange for the horses, and the other merchandise, he would select some elephants, and take (or buy) them. So that by means of their strength, he should be able to have his revenge against Aḥmad Shāh.9 When the Sultan reached the outskirts of the town, he invited the raja of Jajnagar (that is, of Orissa) to inspect his wares in return for elephants. The raja came, accompanied by five hundred men, but his forty elephants ran wild during a sudden thunderstorm and ruined Hushang Shah’s goods. In the skirmish that followed, the raja was taken hostage and was released in return for seventy-five elephants. The Sultan thus got all the elephants, kept his own horses, and then took his way home. These large, armed caravans find parallel also in the Deccan, with horses brought by the Arab traders.10 The Afghan tribes were of recent formation and in many cases quite recently Islamized. Using the same techniques that Richard Eaton has used with regard to the names of the people coming to the shrine of Pakpattan, one normally finds entirely non-Islamic names in the pedigrees of three generations earlier.11 Afghan codes of honour and the cohesive forces of behaviour, the so-called Page 4 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left pakhtunwali, were very much (p.53) in evidence, and in fact it was the social obligation of Afghans to take revenge that was the ruin of the Lodi dynasty. Within the confederation of competing groups that we call the Lodi Sultanate, it became very difficult for the sultan to put to death an heir of one of the tribes without the automatic rebellion of a portion of his helpers. At the time of Sikandar Lodi (r. 1489–1517) these conflicts resurfaced, this time between two groups of descendants of holy men—the Farmulis, who were keepers of a shrine on the horse-trade route from Ghazni, and the Sarwanis, who were also descended of a holy family—and the other groups of Lodis, that is, the Shahu Khel Lodis of the Punjab and the more ‘Afghan’ Afghans. A minor incident could and did become a major cause of disintegration: during a game of polo an amir happened to hit another on the head, allegedly accidentally, and injured him. The other amir replied by hitting him on the head deliberately. When the Sultan went to the playing ground, a follower of one party rushed out and tried to assault the other party in the Sultan’s presence and was severely beaten. Thereupon a confederacy of twenty-two amirs decided to depose the Sultan in favour of his younger brother. The younger brother, a fifteen-year-old boy, ran away and told his concerned mother the plot. The Sarwanis, who had participated in the fracas on the polo field, were exiled rapidly, and their leadership, consisting of four brothers who were sons of the chieftain who had been largely instrumental in putting Sikandar on the throne, left Delhi.12 The Sarwanis first took refuge with the Tomar ruler of Gwalior, and it became an immediate concern to Sikandar Lodi that they should go on somewhere else.13 Next they descended to the court of the sultan of Malwa, where other influential people were to be found at different times (p.54) in the early sixteenth century, notably the great captains of the Jaunpur Sultanate.14 And then, in the midst of the Malwa-Gujarat conflicts, the Sarwanis went on to the Sultans of Gujarat. They reappeared in Delhi thirteen years later in the time of Ibrahim Lodi (r. 1517–26), when the entire balance of power collapsed and other groups became disaffected. The mechanism whereby the leaders of a tribe, namely the Sarwani brothers, could survive for thirteen years at a remote distance from their tribal hosts and be reaccepted and promptly reassume command of it is hard to understand. But the Sarwanis did so and, again in opposition to the other groups of amirs, supported the last ruler, Ibrahim Lodi. By that time other groups, like the Lodis of the Punjab, were going to the court of Babur, amir of Kabul, to invite him to invade. Thus, the particular social structure of the Afghans is of considerable importance when considering the long period of the Lodi dynasty, from about 1456 to 1527. The long and bitter campaigns between the sultans of Delhi and of Jaunpur on the battleground of the Doab between Delhi and Jaunpur, in what is now the Ayodhya-Faizabad area, and in which Afghan war bands fought on each side, provides the background to Aditya Behl’s essay on the Mirigāvatī (1505). Eventually victory came in the 1480s to Bahlul Lodi, though the Sharqi dynasty Page 5 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left survived, still fighting, and Sultan Husayn Sharqi still carried on in Kolgaon at the frontier of Bengal. Bahlul Lodi’s victory brought a consolidation of power in Delhi once more. Two smaller sultanates, Kalpi and Khandesh, were entities created by Timur’s invasion. Khandesh survived thanks to its submission to the powerful successor state in Gujarat, of which it became a subsidiary ally. In the case of Kalpi—for which we have the tarikh of Muhammad Bihamad Khani15—its suppression was due to the combined efforts of its two powerful neighbours, the rulers of Jaunpur and of Malwa. This was a time when denouncing your opponent for being dependent on the military services of the leaders of Hindu peasant soldier groups was a political ploy. While Nasir Khan, the ruler of Kalpi, was condemned (p.55) for allowing the marriage of Muslim women to Hindu notables of his court, he himself wrote to the sultan of Malwa to denounce Ibrahim Shah Sharqi of Jaunpur for employing Hindu muqaddams (chiefs), of making ‘Musalmans captive’ and of looting and carrying away their property ‘as if it had been an infidel land (dār alḥarb)’.16 This brings us to the issue of military manpower, the military market, and the relationship between Hindu chiefs and the sultans. Although Jaunpur had an adequate supply of military elephants from south of the Ganges at this time, their access to the caravans of military horses was declining. The sultans of Jaunpur became increasingly dependent on militant groups of high status, mainly foot soldiers. The particular names that are mentioned are those of the muqaddams of Gahora and Baksar (Buxar),17 the ancestors of those Baghela rajas of Bhatta (afterwards Rewa) and of the Ujjainiyas of Baksar who in later times were such great recruiters for all armies, down to the British period.18 We thus see an anticipation of many of the relationships of the Rajputs of Rajasthan: the Baghela rulers in the fifteenth century were in a subordinate relationship to the Jaunpur sultan but were also a Hindu warrior state, able to calculate how to maintain balances among the competing Muslim rulers of the flatlands. In relation to their neighbours, the rulers of Bhatta pursued a policy of survival and of self-strengthening that involved allying against the weaker party in various coalitions with extreme skill. The Baghelas were also multi-locational, for it was necessary to have control over a distant, inaccessible fortress in difficult terrain such as that Bandhogarh but also over a fertile area in order to provide steady sustenance and to support the population of the fortress—theirs was on the southern side of the river at the Sangam at Arail, opposite Prayag (afterwards Allahabad). Moreover, according to a Kabir-panthi tradition, the raja of Bhatta, Bir Singh Dev, had a haveli in Jaunpur in the early sixteenth century, thus maintaining a presence outside the areas strictly under his control. I have suggested that the Baghelas’ late-sixteenth century Sanskrit historical (p.56) kavya which claims that they conquered Prayag, Varanasi, and so forth, should be interpreted as meaning that they maintained an establishment, a palace, in these various places.19 It is also clear that they used any kind of support which might lend them some religious prestige. Besides the ‘Brahminical’ kavya, they Page 6 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left patronized the Kabir-panthis,20 and a leading follower of Kabir established a math in Bandhogarh itself. Possibly following the disintegration of the Baghela state, the Kabir-panthis proceeded onwards towards Chhattisgarh, with the historical consequence that there are nowadays more followers of Kabir in the Chhattisgarh region than in the Gangetic plain. The raja of Bhatta also provided temporary refuge to the Afghan hosts who were defeated by Babur in Panipat and left their place in Delhi.21 Thanks to Dattu Sarwani’s recollections, we know that after this defeat the Sarwanis crossed the river and the Gangetic plain with their sheep and their animals and took refuge in the dominions of the raja of Bhatta, who assigned villages to them. It was from that distant point that they set out to fight against the Mughals, both in Avadh and in northern Bihar and southwards towards Mandu and Gujarat. It is quite clear that the Sarwani Afghans afterwards lived for about thirty years in what was called Balapat; their previous Indian stronghold had been in Sambhal, not very far from Sirhind, and they retained property there. During Sher Shah’s rule, some of them unwillingly moved back to Sambhal and became administrators in the alliance of Sher Shah. As is clear from the narration of another member of the family, ‘Abbas Sarwani, old contacts were not neglected, either, and Sarwanis were still present in the original homeland by the Gomal river.22 The contact with Roh and the Afghan hills was maintained also by the Lodis. For example, Alp Khan Lodi’s father sent him off with horses (p.57) telling him that he could either sell the tribal horses or fight with them for power. This anecdote appears again in the story, which I do not disbelieve, of the rise of Bahlul Lodi. It is known that Bahlul Lodi had relations in the sultanate service even in Firuz Shah Tughluq’s time, but in the early stages of his takeover from the Sayyid dynasty, he appeared with a string of horses and was assigned a large district rather than money. The multi-locational nature of power appears, therefore, a characteristic of this disintegrating but enterprising society. There are other examples of people who tried to establish control over different areas some hundreds of kilometres apart. Probably, as a consequence of Timur’s attempt at gaining control of Khorasan in 1383, a group called the Sayyids of Bayhaq appeared on the scene. They were related to the Kubrawi Hamadanis of the circle of ‘Ala’ al-Din Simnani and established themselves both at Huttalan, a high territory upon the affluents of the Oxus on the route from Bukhara to Balkh, and also in Kashmir. These Bayhaqi Sayyids were welcomed in Delhi when they appeared just a few years before Timur’s sack of Delhi and were assigned land in the Doab not far from Kannauj, at a place called Jareja. This arrangement was not good enough for them, however, and they looked for other opportunities. They seriously considered reinvading their homeland of Bayhaq through southern Afghanistan but managed to establish themselves at Naushahra, one of the major southern Page 7 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left passes of Kashmir 50 miles east of Peshawar. They then appeared in the Valley of Kashmir and became one of the quarrelling factions marking the entire independent sultanate of Kashmir (one of them even sat on the throne for three months). What is significant is that they tried all these avenues and maintained, I believe, an interest in the Doab down to later times. One consequence of this period of disorder was the perishing of historical material from Delhi. It is remarkable, after the comparative stability of the greater Delhi Sultanate, how many manuscripts seem to fail to reach the next period of stability and collection after 1570, when Akbar regained control of the northern area of the Delhi Sultanate. Among the many things that perished were a number of important manuscripts. No manuscript survives with the final chapter of the history of the majority of Firuz Shah’s reign, ‘Afif’s Ta’rīkh-i Firūz Shāhī.23 (p.58) Similarly, none of the six manuscripts of the great Persianscript Hindi poem Candāyan (1379), one of the most important and popular cultural products of the sultanates, has reached us with the conclusion, which points to a period of destruction. The versified Persian translation of the Hindi Candāyan that ‘Abd al-Quddus wrote at the time of Bahlul Lodi’s invasion of the eastern Doab, presumably the concluding one of 1482, was destroyed in the fighting.24 While there is no trace of Lodi histories written by contemporary historians, we know that there were histories of Jaunpur that were lost in the sack of Jaunpur.25 From Sirhindi’s Tā’rīkh-i Mubārakshāhī, which ends in 1434, we know the events of the Sayyids’ dynasty but not its conclusion; no one apparently took up the writing of their history. We have a parallel account in Muhammad Bihamad Khani’s history of Kalpi, but that comes to an end in 1440. There is no continuation of the history of Delhi in this period until historians tried to pick up the tradition when Akbar embarked on his vast historiographical enterprise in the 1570s. Afghan traditions were preserved largely in the memoirs of an aged mullah in Delhi, Mushtaqi, a collection of largely disorderly narratives. Otherwise we have the politically motivated series of productions during Akbar’s reign.26 There is one exception, an independent work by Muhammad Kabir, Fasāna-i Shāhān-i Hind, which is interesting as it represents the traditions of the embattled Afghans in Bihar still fighting the Mughals in the beginning of the seventeenth century.27 All the other Afghan histories are from the sixteenth century and are derived from Mushtaqi. Mushtaqi also began a fascinating corpus of folk tales and stories of the Afghans in which I have tried to trace the stemma, the subsequent changes in the corpus, and the variations between the different authors, and from which one can derive some vivid images of social life in (p.59) this period.28 Just to give one example, from Sikandar Lodi’s lightning campaign that led to the conquest of Jaunpur in 1479, we have the story of a cowherd, an Ahir, in love with the bania’s daughter. After she was carried off during the sacking of the village of Nimkar on the Gomti, not far from Jaunpur, the cowherd went in search of her. As the story appears in Mushtaqi, after two years the cowherd saw his beloved serving her Page 8 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left Afghan master and husband. He was badly beaten up but was eventually taken into the household as a useful manager of animals and became a trusted servant of the Afghan. During another campaign, the cowherd was given instructions to collect the Afghan’s baggage and forward it along with his wife to the frontline, but he managed to divert the baggage in one direction and the wife in another. While the wife rested, the Ahir, who was dressed as a yogi or a darvesh in other versions, was recaptured by the Afghan and strung upside down from a tree while the wife and the Afghan went to sleep. Some snake poison fell into the bowl they both drank from and the Afghan was struck down. The ahir now told the woman that she had become his property! The story was told to Mushtaqi by Khwaja Khan, an old warrior who had participated in the campaigns. Khwaja Khan was travelling on a road with a little hut with an awning beside it, much like a dhaba. A severe rainstorm broke out and Khwaja Khan asked the husband and wife in the hut to feed him, and the story emerged gradually from the conversation with husband and wife. Notes:

(*) This is a slightly edited transcript of the keynote address Simon Digby gave at the conference ‘After Timur Came’. The editors have added references wherever possible, but have maintained the rhetoric of the address. (1) For a close analysis of topics that are not included in Sufi records, including financial matters and an intense relationship with the rulers whom one is supposed to avoid, see Riazul Islam, Sufism in South Asia: Impact on Fourteenth Century Muslim Society, Karachi: Oxford University Press, 2002. (2) For example, the influence of the Suhrawardi shaykhs in this period may well be related to commercial enterprises on the trade route to Transoxiana. It is hardly by chance that many of their followers were based in the then frontier town of Kara on the Ganges, at the end of this trade route; Simon Digby, ‘Before Timur Came: Provincialization of the Delhi Sultanate through the Fourteenth Century’, Journal of the Economic and Social History of the Orient, 47(3), 2004: 314–18. (3) See B.F. Manz, The Rise and Fall of Tamerlane, Cambridge: Cambridge University Press, 1999. (4) See Simon Digby, ‘Export Industries and Handicraft Production under the Sultans of Kashmir’, Indian Economic and Social History Review, 44 (4), 2007b: 407–23. K.N. Pandit, Bahāristān-i-shāhī: A Chronicle of Mediaeval Kashmir, Calcutta: Firma KLM, 1991, ch. 3, which, however, minimizes the impact of Timur’s requests; see also S. Digby, ‘Between Ancient and Modern Kashmir: The Rule and Role of Sultans and Sufis’, in The Arts of Kashmir, P. Pal (ed.), New York: Asia Society, 2007a, pp. 114–25.

Page 9 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left (5) Peter Jackson, The Delhi Sultanate: A Political and Military History, Cambridge: Cambridge University Press, 1999, pp. 318–19. (6) The evidence of this is found in an extensive Sufi tazkira by ‘Ala’ al-Din Barnawi called Chishtiyya-i Bihishtiyya (1655). At a later date these musicians passed into the service of a family of Sufi pirs who were consolidating their authority in two sites of North India, and these musicians survived into the sixteenth century; see M.H. Sherani, ‘Makhdūm Shaikh Bahā’ al-Dīn Barnāwī’, part 1, Oriental College Magazine, August 1927: 41–58. Also see Katherine Butler Brown, ‘The Origins and Early Development of Khayal’, in Hindustani Music: Thirteenth to Twentieth Centuries, Joep Bor, Françoise ‘Nalini’ Delvoye, Jane Harvey, and Emmy te Nijenhuis (eds), New Delhi: Manohar, 2010, pp. 175– 8. [Editors’ note: We have found no evidence that the first sultan of Jaunpur was from Andhra; also, Katherine Butler Brown (now Schofield) believes that the ‘Deccani’ music at the Jaunpur court is likely to have followed the north Indian system. (Personal communication, 13 August 2014)] (7) Shaykh ‘Abd al-Haqq was also important in the rise of the Hindi literature of that area, see Chapter 14 in this volume. (8) ‘Abd al-Quddus Gangohi, Anwār al-‘uyūn, Lucknow: Matba-i Mujtaba’i, 1909, pp. 88–9. (9) Nizam al-Din Ahmad, The Ṭabaqāt-i-Akbarī of Khwājah Nizāmuddīn Aḥmad, translated by B. De, Calcutta: The Asiatic Society, 1927–9, vol. 3, pp. 475–7. (10) Jean Deloche has published some murals found in the temple in Rameshwaram showing the Arab traders and the horses descending and going up the rocky hills. J. Deloche, ‘Etudes sur la circulation en Inde: VII. Konkan warships of the xlth–xvth centuries as represented on memorial stones’, Bulletin de l’Ecole française d’Extrême-Orient, 76(1), 1987: 165–84, fig. 12. (11) Richard M. Eaton, ‘The Political and Religious Authority of the Shrine of Baba Farid in Pakpattan, Punjab’, in Moral Conduct and Authority: The Place of Adab in South Asian Islam, Barbara Metcalf (ed.), Berkeley: University of California Press, 1984, pp. 333–56. (12) These are portions of the story of Dattu Sarwani whose significance I did not realize at the time; S. Digby, ‘Dreams and Reminiscences of Dattu Sarvani, a Sixteenth Century Indo-Afghan Soldier’, Indian Economic and Social History Review, 2(1), 1964: 52–80, and 2(2) 1964: 178–94. [Editors’ note: This anecdote is not to be found in the article.] (13) Illustrated and illuminated manuscripts from this period are of difficult attribution, but it is worth noting that among the khatt-i bihari Qur’ans, which obviously had roots in the Delhi Sultanate, the earliest one comes from Gwalior Page 10 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left at a time when a great many of the notables of the Delhi Sultanate were regrouping at Gwalior; see Chapter 11 in this volume. (14) See S. Digby, ‘Two Captains of the Jawnpur Sultanate’, in Circumambulations in South Asian History: Essays in Honour of Dirk H.A. Kolff, O. Prakash and J. Gommans (eds), Leiden: E.J. Brill Publishers, 2003, pp. 159–78. (15) Muhammad Bihamad Khani, Tā’rīkh-i-Muḥammadī (1438–9), Muhammad Zaki (trans.), Bombay: Asia Publishing House, 1972. (16) ‘Ali ibn Mahmud Kirmani, Ma’ās̱ir-i Maḥmūdshāhī, ed. N. Ansari, Delhi: Danishgah-i Dihli, 1968, p. 58; see also Digby (2003), p. 174. (17) Digby (2003), p. 174. (18) This is the material surveyed in detail by Dirk Kolff, Naukar, Rajput and Sepoy: The Ethnohistory of the Military Labour Market in Hindustan, 1450– 1850, Cambridge: Cambridge University Press, 1990. (19) H. Shastri, The Baghela Dynasty of Rewa, Calcutta: Government of India, Central Publication Branch, 1925. (20) According to tradition, Bir Singh Dev Baghela was close to Kabir and involved in the division of Kabir’s mortal remains; D. Lorenzen, Kabir Legends and Ananta-das’s Kabir Parachai, New York: State University of New York Press, 1991, pp. 14–15 and 105ff. (21) See also Digby (1964). (22) ‘Abbas Khan Sarwani, The Tā’rīkh-i Sher Shāhī of ‘Abbās Khān Sarwānī, ed. S.M. Imamuddin, Dacca: University of Dacca, 1964, pp. 165–6. (23) There are five or six manuscripts probably from a single stemma but we lack the conclusion. (24) See Laṭā’if-i Quddūsī, which preserves a few fragments; see also S. Digby, ‘Abd al-Quddus Gangohi (1456–1537 A.D.): The Personality and Attitudes of a Medieval Indian Sufi’, Medieval India: A Miscellany, Aligarh: Aligarh Muslim University, 1975, vol. 3, pp.1–66. (25) See the Introduction. (26) On this topic see also S. Digby, ‘The Indo-Persian Historiography of the Lodi Sultans’, in Les Sources et le Temps, F. Grimal (ed.), Pondichéry: Institut Français de Pondichéry, 2001, pp. 243–64. (27) See Digby (2001), pp. 262–4.

Page 11 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

After Timur Left (28) Rizq Allah Mushtaqi, Waqi‘at-e-Mushtaqui of Shaikh Rizq Ullah Mushtaqui: A Source of Information on the Life and Conditions in the Pre-Mughal India, Iqtidar Husain Siddiqui (trans. and ed.), New Delhi: Indian Council of Historical Research and Northern Book Centre, 1993.

Access brought to you by:

Page 12 of 12

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Bandagī and Naukarī Studying Transitions in Political Culture and Service under the North Indian Sultanates, Thirteenth–Sixteenth Centuries* Sunil Kumar

DOI:10.1093/acprof:oso/9780199450664.003.0003

Abstract and Keywords This chapter offers a long-term history of relations of military servitude in north and eastern India from the thirteenth to the sixteenth centuries, offering correctives and elaborations of D.H.A. Kolff’s reading of the fifteenth-century military labour market. The chapter argues that we should pay attention to the political trajectories of déracinés, those who were slaves or not born to privilege, as they emerged from natal alienation and began to establish links with new and sometimes non-Muslim allies. This was a time when texts for aspiring elites came to be produced, such as Vidyapati’s Puruṣa parīkṣā, and groups such as the Afghans negotiated transitions between slavery and military service. By foregrounding continuities between the pre- and post-Timur periods, the chapter calls attention to emic terminology and its subtle transformations. Keywords:   Sultanate, Delhi, Vidyapati, slavery, servitude (naukari), Maithili, Afghans

HISTORIES OF THE DELHI SULTANATE are usually organized into rather simple binaries: years of centralized governance of the Sultans of Delhi are measured against years of decentralized rule under the ‘regional’ Sultanates. Historians frequently arrange their narratives to describe the origin–apogee– decline career graph of the Sultanate where ‘stasis and decline’ is usually the chapter before the epilogue on the years after the (p.61) Timurid invasion (1398/99) and the fragmentation of the Tughluqid dominion.1 The inordinate persistence of this representation is surprising, not least of all because it is in Page 1 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī the face of some recent scholarship on the fifteenth and sixteenth centuries, not least of all by Dirk Kolff and Simon Digby.2 Dirk Kolff’s observations on the culture of service, naukari, the search for military employment compensated by a salary and/or other social and political rewards, was placed as a distinguishing feature of the fifteenth and sixteenth centuries, setting this period apart from the one prior to Timur’s invasion. In Kolff’s analysis, neither the decline of the Sultanate nor Timur were primary contributors to the making of this new political culture. Instead naukari was closely linked to the military labour market in the subcontinent where the opportunities for, and availability of, armed personnel had grown exponentially through the fifteenth into the sixteenth centuries. It was this world, rather than that of the Delhi Sultans, that had an abiding impact on the Mughals. (p.62) Despite Kolff’s innovative rereading of the post-Tughluq and pre-Mughal centuries (1414–1556), his work also begged the question: what was the prehistory of this culture? From what elements was it constituted? What range of personnel and ties did political service encompass prior to the fifteenth century? How did this alter and in what arenas? In other words, even as Kolff elaborated on the quality of naukari, the history of this culture remained murky. Notably, in his analysis it was never clear why the Khaljis or the Tughluqs did not tap into the military labour market in the same way as the later Afghans or the Jaunpur, Malwa, and Gujarat regimes. Were the differences merely conjunctural and contingent, or were they a consequence of larger structural changes in representation and interpersonal relationships through the fourteenth century, or a mixture of all these factors? In as much as conventional historiography considers the fifteenth and sixteenth centuries as substantive departures from a preceding period, this chapter extends the provocative question posed by the editors in the introduction to this volume and asks, ‘How well do we know the history of the thirteenth and fourteenth centuries that we should consider the fifteenth and sixteenth centuries so different?’ There are considerable similarities in both periods: the armed peasantry were also recruited in huge numbers by the Delhi Sultans (especially ‘Ala’ al-Din Khalaji and Muhammad Shah Tughluq),3 and immigrant personnel, occasionally of very humble origins (including the Afghans) were integral to Sultanate armies in the thirteenth and fourteenth centuries, some going on to become governors of provinces and rulers of Delhi.4 (p.63) Although Kolff’s understanding of naukari was further inflected by the notion of ethnogenesis, where groups developed their identities through shared service, domicile, language, literature, and memories, these processes were, in themselves, not unique to the cultural world of the fifteenth century warband.5 Many of the similarities and differences between the period before and after Timur’s invasion have not been systematically explored and the vectors of analysis are still based on impressionistic evidence concerning scale of military and political formations. Perhaps most critically, there has been no comparison Page 2 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī of the ways in which regimes and their participants were discussed in the literary materials of the time, or of the ways in which they constructed ideas of service, loyalty, and leadership and their material dimensions—what aspects continued, what changed, and how are they manifest through the thirteenth into the fifteenth–sixteenth centuries? This is actually quite astonishing because most historians have commented on shifts in the nature of the literary materials available to the historians from the fifteenth century. As the editors of the volume have noted, the epoch of the great histories of Delhi, the tawarikh of Juzjani, Barani, ‘Isami had tapered to an end, there were few contemporaneous Persian histories on the fifteenth and sixteenth centuries and far more that were retrospective accounts produced in the Mughal chanceries or in the courts of latter-day Rajput princes. There were also histories produced in a variety of vernacular languages following literary conventions borrowed from Sanskrit and an assortment of Bhakti and Sufiinflected styles. But the sheer difference in the literary materials, the cognitive worlds that they mapped, and their retrospective projections of social and material conditions into an earlier period left historians trained in reading the grand narratives of the Persian tawarikh of the Delhi Sultans (p.64) or the Mughal period with no ready-made templates that they could readily follow.6 Since these literary materials were very different from each other in their readings of history, their literary styles, and authorial intent, their interpretation needed, as Carlo Ginzburg reminded historians in a different context, a great deal of caution. As Ginzburg noted, ‘The fashionable injunction to study reality as a text should be supplemented by the awareness that no text can be understood without a reference to extra-textual realities’.7 In other words, from the perspective of my argument, since the literary materials before and after Timur’s invasions are so different, and the evidence they proffer so incommensurate, any comparative exercise of the cultural and political worlds of the two periods has to have some degree of structural correspondence. I keep that in mind as I try to disaggregate the elements that constituted political culture and service at differing moments of time and investigate three relatively precise subjects: the recruitment of changing kinds of personnel and what this might tell us regarding expectations and ambitions of masters and servants; the ways in which political entitlement related to the social and cultural profiles of individuals and groups and might serve to alter it; and the manner in which Persian and other writers framed and commented on departures from idealized norms of political conduct. Because of the varied modes of literary presentation, I have tried to be as precise as possible in presenting the contexts in which different personnel constructed a range of service relationships through the long period of the thirteenth–sixteenth centuries, paying close attention to the ways in which the literary materials commented on this subject over time.

Page 3 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī In my analysis of political culture and service, I use the two terms bandagī and naukari hermeneutically to analyse the relationships of the different kinds of personnel recruited by the Delhi Sultans and other patrons between the thirteenth and sixteenth centuries. The adjective (p.65) bandagi is derived from the noun banda, the Persian term for military slave (plural: bandagān), and describes political relationships that were strongly touched by traditions of servitude present in the practice of military slavery.8 I use the term to discuss political relationships amongst personnel who were not always, strictly speaking, ‘unfree’. While these personnel could have been free, their relationships were also strongly touched with the paradigms of service associated with bandagi. On the other hand, the adjective naukari as a means to understand political culture may appear as more of a neologism. In its Persian usage the noun naukar carried some of the meanings present in its antecedent Mongol form, nökör (singular)/nököd (plural)—personal retainer, loyal friend, comrade in arms, bodyguard—and within the limited context of a dyadic relationship with a master, its meaning was very close to banda-i khāṣṣ. In its original Mongol sense, the nököd were free and honourable servants, who had voluntarily accepted service with a great lord; it had none of the pejorative meanings associated with slavery.9 In the pre-Timurid years of the Sultanate I have found no evidence in (p.66) the Persian chronicles of the usage of the term naukar; it was, however, in circulation in the Afghan records of the sixteenth century, but not, as far as I know, in its adjectival form, naukari. Even if the term naukari was absent in the pre-Timurid centuries, I am interested in researching the elements that constituted relationships with naukars (or naukari, as I refer to it) and here, happily, there is considerable information. I place naukari in a dialectical relationship with bandagi, with whom it can be usefully juxtaposed for purposes of elucidating the different kinds of political relationships and the tensions apparent in their representations. In this context I found the diachronic mode particularly useful in avoiding what might appear as a bandagi/naukari binary. While the elite personnel encompassed in these terms were always in flux and their relationships and identities shaped by a variety of material factors, they were not the authors of their histories. Moving through time has the advantage of careful contextualization to enable an inflected analysis of how Persian and other chroniclers responded to Sultanate personnel with shared or contrasting backgrounds and conditions of service. It allows me the opportunity to note continuities and transitions by paying attention to the ways in which particular characteristics of these personnel and their contexts were elided or foregrounded. Cohering political relationships around terms such as bandagi and naukari, therefore, serves as points of entry through which contingent relationships and their representations can be further interrogated to finesse particularities, especially over time. Towards that end, the following section studies the meaning of bandagi from the thirteenth to the sixteenth century in somewhat greater detail. Page 4 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī The Value of Déracinés: Bandagi in the Pre-Mughal Period For the longest time historians suggested that the usage of military slaves was a condition unique to the thirteenth century ‘Turkish’ Sultanate regime. The evidence contradicts this conclusion; the deployment of military slaves continued well into the Lodi regime in the sixteenth century and beyond. Although historians find it easy to recall Qutb al-Din Ai-Beg, Iltutmish, and Balban as examples of important and powerful slaves in the thirteenth century, we need to add to this list of significant examples of military slaves in the service of Delhi Sultans the eunuch (p.67) Malik Kafur Hazar Dinari, the Khalji general who led ‘Ala’ al-Din’s armies into the Deccan and managed court politics in Delhi at the end of his master’s reign;10 Khusraw Khan Barwari (r. 1320), the Khalji slave commander who deposed his master, became Sultan of Delhi, and had cordial relations with the Sufi Shaykh Nizam al-Din Awliya;11 the Tughluq slave ‘Imad al-Mulk Bashir Sultani who was Firuz Tughluq’s commander of armies;12 and Malik Sarwar Khwaja Jahan, a eunuch and Firuz Tughluq’s commander, who went on to become Sultan of Jaunpur (1394–9) and founded the Sharqi dynasty (1394–1457).13 We also need to take into account the cadre of military personnel present during the period of Lodi (1451–1526) and Sur rule (1540–55), contingents that were described as khassa khayl. These contingents were retained by the Sultans and his nobles, amidst which were a significant number of slaves.14 It is not the persistence of slave retinues through this period that is actually of such great significance, but its attendant implication: through these centuries there was also considerable structural stability in the modes of selecting and socializing personnel into a culture of political service which I refer to as bandagi. This was evident in the ways in which common slaves were transformed into bandagan-i khass. (p.68) As in the thirteenth century, so too in the fifteenth, masters chose their bandagan-i khass from their personal slaves that they had usually possessed for a considerable period of time, and who had responded favourably to the attentive fosterage (parwarish) and training (tarbiyat) imparted to them. For example, the details of the slave of Shams al-Din Iltutmish in the early thirteenth century, Taj al-Din Sanjar Kazlak Khan, harmonizes very well with the information on ‘Imad al-Mulk Bashir Sultani, the slave of Firuz Shah Tughluq from the late fourteenth century. Iltutmish (1210– 36) purchased Taj al-Din before he had seized the throne when the slave was still a youth. He was reared in the household of the monarch together with his master’s eldest son and was elevated slowly from domestic responsibilities such as a cup-bearer (chashnīgīr), to become, eventually, the commander of Uchch, one of the important settlements in the north-west, a position he continued to occupy till his death.15 A century and a half later, ‘Afif provided a biography of ‘Imad al-Mulk, a valued slave (ghulām-i khāṣṣ) of Firuz Shah (1351–88), in the possession of the monarch prior to his accession. He had served the Sultan’s family earlier, perhaps as a part of his mother or wife’s domestic staff. This Page 5 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī probably provided the setting for his proximity to the monarch; ‘Afif noted that ‘Imad al-Mulk had access to the Sultan at all times, within and outside the palace. The chronicler was also very precise in mentioning that the slave was the first person to be honoured with a title on his master’s accession and he remained as the commander of Firuz Shah’s army (sar-i lashkar) for the duration of his lifetime.16 The examples of Taj al-Din and ‘Imad al-Mulk elaborate on the dyadic nature of the relationship established between the master and the slave, a relationship summarized by Juzjani in the thirteenth century as one where the slave ‘was extremely close’ (qurbat tamām dāsht) to his master.17 Clearly what changed over time was the ethnic and racial profile of slaves (Turkish slaves giving way to Ethiopian or slaves from the subcontinent), not the dependence of the several monarchs on deracinated (p.69) military personnel. And this continuation, not the shifts in the ethnicity or the size of slave retinues, marked a critical structural characteristic of Sultanate regime.18 The inability to appreciate the full impact of bandagi in Sultanate regimes is a product of a rather narrow equation of the culture of service that might be touched with aspects of military slavery with the jural condition of unfreedom. To begin with, in Islamic law the military slave continued as a dependant of his erstwhile master as his mawālī even after manumission (and/or conversion). That the bandagan-i khass were legally free, having been manumitted sometime during their career, did not in itself alter their bandagi relations with their master/ patron; their jural condition was, therefore, immaterial in this context.19 What is more germane to note is that, whether freed or not, the bandagan set a normative standard of an idealized loyal servant (p.70) and the Sultans sought other bodies of military personnel who could most closely approximate their special déraciné qualities.20 They found these personnel amongst other marginal groups who were equally alienated from the networks of power and, hence, dependent upon their master in ways similar to that of the bandagan. Marginality carried with it the same baggage of ostracism from elite social circles that Delhi Sultans had sought to use to gain the dependence of their favoured slaves. Their intent was captured by Barani in the middle of the fourteenth century when he explained that authoritarian monarchs like Balban did not trust ambitious individuals who believed they had a prior claim to status or authority. Such individuals, he noted, inevitably wanted to associate in kingship, shurakā-i mulk. Instead, Barani observed, Balban favoured individuals who were entirely his creatures and relied on his support and protection, ḥimāyat-i Balbanī, to gain status.21 As Persian chroniclers noticed, these concerns led the Delhi Sultans to patronize some exceptional bodies of people who were unfailingly described as social outcasts. Thus, when Afghans were recruited by Sultan Balban, Amir Khusraw commented:

Page 6 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī In this (?) fortress live the Afghans—nay man-slaying demons, for even the demons groan in fright at their shouts. Their heads are like big sacks of straw, their beards like the combs of the weaver, long-legged as the stork but more ferocious than the eagle, their heads lowered like that of the owl of the wilderness. Their voices hoarse and shrill like that of a jack-daw, their mouths open like a shark. Their tongue is blunt like a home-made arrow, and flings stones like the sling of a battering ram. Well has a wise man said that when speech was sent to men from the sky, the Afghans got the last and least share of it.22 (p.71) Similarly, at the end of the thirteenth century, Mongol contingents were recruited by Sultan Kayqubad (1287–90), Jalal al-Din Khalji (1290–6), and ‘Ala’ al-Din Khalji (1296–1316). Barani’s snobbery was evident in his sneering reference to them as ‘new-Muslims’, nau-Musalmān. These were not universally shared opinions—certainly the favour shown by the Delhi monarchs was in marked contrast to the hideous images of the Mongols constructed by the Persian litterateurs. ‘Isami noted: Their features were repulsive and their language rude and impolite…. Their countenance, their hair and their nature were repugnant and their sweat smelt foul … Their hands were ghost-like and their feet highly ominous … They had flat noses and protruding faces. They bore an angry and fatal look. Their eyes were so small that they were completely hidden under their long drawn-out brows. Everyone felt dismayed on seeing them. Their bodies were scarred by the bites of the lice that they carried on themselves, each louse feeding itself on their blood.23 Sultanate litterateurs were very harsh in condemning monarchs who recruited military personnel who were the ‘lowest and basest of the low and base born’ (siflatarīn wa raẕālatarīn-i siflagān wa raẕalagān); they brought ruin to the realm.24 But the marginal status of these newly empowered groups in Sultanate society was exactly what made recruits like the Afghans, the Mongols, and other frontier groups valuable to their patrons. Like the bandagan these ‘base born’ (bad aslī wa nā kas) were also divorced from their natal contexts, raised to high office in Sultanate society where their political position, wealth, and power were completely incongruent to their social status. The disjunctions between their political and social positions only served to underline their dependence on their master’s dispensation of favours. But they were free and sometimes held high positions in the service of the Delhi monarchs. This elicited complex responses from the Persian chroniclers and it might be (p.72) useful to appreciate their inflected responses by looking at the case of the Bhattis from Firuz Tughluq’s reign (1351–88) somewhat more carefully.

Page 7 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī The Bhattis were a pastoral group settled in the western Punjab, minimally supervised by the governors of the Delhi Sultans.25 In ‘Ala’ al-Din Khalji’s reign (d. 1316) their areas came under the supervision of Ghazi Malik (the future Sultan Ghiyas al-Din Tughluq) and his brother, Sipahsalar Rajab, who were appointed as commanders to the territories of the city of Dipalpur. Ghazi Malik wanted to get his brother married to one of the daughters of the Rais of Dipalpur (az dukhtar-i rā’ī az rāyān Dībālpūr), and when informed that the daughter of Raja Ranmal Bhatti was particularly eligible, he made the proposal to her father.26 When Raja Ranmal scorned the offer, Ghazi Malik intervened in his territories with extortionate demands. Faced with brutal punishment and economic disaster, Raja Ranmal’s daughter persuaded her father to accept the marriage and save his subjects. Although the Bhatti princess was honourably received in Ghazi Malik’s family, she did convert and was renamed Bibi Kadbanu, and her son, Firuz Tughluq, eventually became the Sultan of Delhi. Other than the episode of renaming/conversion, ‘Afif underlines the princess’s sense of natal alienation when he reports her request to her father to accept the marriage in the larger good and consider her abducted by the Mongols in one of their raids.27 ‘Afif did not disclose why Ghazi Malik was looking for a bride specifically amongst the daughters of the chieftains of Dipalpur, but it could have been an attempt to consolidate his position in the region; the social profile of the Bhatti pastoralist followers of the Raja was similar to the (p.73) other ‘rustic’ frontiersmen that comprised Ghazi Malik’s retinue.28 But we have no evidence of any Bhatti recruitment by Ghazi Malik or the later Tughluqs; instead their resistance to the Sultanate seems to have continued, and some decades afterwards Muhammad Tughluq campaigned against them.29 The marriage with Malik Rajab, however, must have meant that some members of the bride’s family resided with her in Delhi. There is a reference to one relative, the proverbial maternal uncle (the nīyā > māmūn/māmā), Rai Bhiru/Pheru Bhatti, who joined his nephew’s service and, as his title indicates, had obviously not converted to Islam. In the early years of Firuz Tughluq’s reign, Rai Pheru was instrumental in helping his nephew neutralize a conspiracy hatched by the late Sultan Muhammad Tughluq’s sister.30 While not much more is known about the role of Rai Pheru in the politics of the capital, Firuz Tughluq invested considerable resources to insulate Delhi and Haryana from the depredations of frontier pastoral elements like the Bhattis.31 Rai Pheru, however, was distinct from the larger body of the Bhatti clans; he was a part of Firuz Tughluq’s personal retinue. While he retained some elements of his old social identity, there was also the loss of his old social context and, hence, honour; he was natally alienated.

Page 8 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī Persian chroniclers could be extremely judicious in their criticism of the great lords of the realm. They may have been regarded as déracinés and social menials, but these individuals were also very well connected and possessed considerable power. Many bandagan-i khass were provided with patently false genealogies and were differentiated from the riff-raff, newly purchased slaves who were summarily dismissed as frivolous and purchased with coins (harzgān wa diram-i kharīdgān).32 Political success certainly made Amir Khusraw revise his opinions of his patrons: he was harshly critical of the Qara’una Mongols, but some years afterwards produced a fulsome eulogy when one of them, Ghiyas alDin Tughluq, (p.74) morphed into a Delhi Sultan.33 So how did transitions in the fortunes of their deracinated protagonists reflect on the manner in which Persian chroniclers understood bandagi? The bandagan certainly tried to alter their déraciné and abject social status by establishing large households, accumulating sizeable armed retinues, and breaking out of restrictive endogamous relationships by marrying women above their status.34 Describing these developments in the last quarter of the thirteenth century, Barani felt the need to coin the term mawlāzādgān, sons of freed slaves, to note the presence of this new social group in Sultanate polity. With his aristocratic prejudices, Barani used the term derisively to remind his readers of the ‘real’ social origins of some Sultanate elites. Their numbers had increased enormously by Sultan Kayqubad’s reign (d. 1290), and Barani noted their search for alliance with Mongol converts (naw-Musalman) who had just migrated to the Sultanate—one body of social menials gravitating into alliance with another, in search of roots as it were, and leverage against entrenched elites.35 The déracinés were seeking to alter their condition of natal alienation and social death, the features that Orlando Patterson argued were integral in the constitution of slavery.36 (p.75) Like most chroniclers, Barani was very judicious in whom he chose to insult: he never disclosed to his readers, for example, that Amir Khusraw was one such mawlāzāda. As it happens, the great Persian poet and close friend of the Sufi saint Nizam al-Din Awliya was not only the son of a Shamsi slave, Malik Saif al-Din Lachin, but his mother was also the daughter of a slave. The son of such a slave couple—and to be fair, they may have been slaves but they were politically very well connected—chose to give up a possible career in the army, got an education instead, and became a poet and the quintessential courtly naukar.37 Nor was Amir Khusraw the exception; through the thirteenth-century families of slave origin increasingly established large homesteads that Barani describes as khaylkhāna, a term that would remain relatively stable as a description for an elite military commander’s household well into the sixteenth century.38 The efforts of the bandagan at greater social and political emancipation in Sultanate society had a decidedly material character. Although I will discuss this in greater detail a little later in the chapter, in the following section, I would like to study a different body of Sultanate servants—people of Page 9 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī high status by virtue of birth or possession of knowledge and skills that were esteemed and in short supply. Their social profiles were mixed but still very different from the déracinés that we have just studied. How were they placed in thirteenth–fourteenth Sultanate society and what conditions of service did they inhabit?

Service with Status and Honour: Negotiating Naukari If naukari has to be distinguished from bandagi, the Sultanate personnel that we would have to study would be those not considered déraciné. These people would be recent émigrés and they would also carry with them their genealogies and titles and other modes of proclaiming their different skills—their signatures of great social worth in their new homes. (p.76) As a prosopographical survey of early Sultanate elites brings out, however, Ghurid aristocrats served in north India to make quick fortunes, but until the 1220s few chose to stay on; life was violent and uncertain in the frontier marches, comforts were few, and greater ambitions could be fulfilled in the courtly politics of the Shansabanid dominions in Afghanistan.39 Only the humble stayed on in the insecure north Indian marches to carve their fortunes. This changed in and around the 1220s, after the full-fledged campaigns of the Mongols in Transoxiana, eastern Iran, and Afghanistan when a larger number of immigrants of elite social backgrounds fled to north India. These émigrés were received honourably, but as Sultan Iltutmish asserted the authority of Delhi over north India, he was extremely selective in his dispensation of favours; very few free amirs of high birth were favoured with independent command.40 At least as far as naukari for military commanders was concerned, until the Timurid invasion, the careful discrimination of the Delhi Sultans privileged the social menials, not the aristocrat émigrés.41 This selective discrimination against immigrant aristocrats was in marked contrast to the favour shown to the literati: jurists, secretaries, accountants, scholars, poets—those who were adept in Persian diplomatics and ideally, some level of Arabic scholastics, the ‘people of the pen’ (ahl-i qalam) more generically. These men had a portfolio of skills that were rare and therefore in very high demand; the management of large households, leave aside ambitious state formation, could not budge without their support. The Mongol invasions had left their urban networks in disarray in eastern Iran and with their patrons either dispersed or killed, the ahl-i qalam migrated to the subcontinent in large numbers. (p.77) They were in search of naukari and carried the ambitions of reproducing a Persianate cultural world in north India that was stable and familiar. Their past service with hostile regimes notwithstanding, it was their skills that immediately secured their favour with successive Delhi monarchs and a variety of other warlords.42

Page 10 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī Interpreting these bodies of people primarily as collaborators, the much criticized worldly ‘ulama’-i duniyā43 loses sight of their pietistic motivations, the important role played by the ahl-i qalam in manufacturing the idea of the Muslim community in the subcontinent, or their huge contribution in its reproduction which won them great respect. Placing them within ties of naukari restores to them a level of autonomy available to free individuals, especially those who possessed knowledge that was in great demand. The social stature and portfolio of skills of these naukars provided them with greater options which they used to negotiate conditions of service commensurate with their social status. This was evident in the way members of the Kirmani family plotted their careers. One of the members of the Kirmani family, Amir Khwurd, was the author of the biographical encyclopaedia of the Chishti saints, Siyar al-Awliyā’ (completed ca. 1360s).44 The Kirmanis were one of Delhi’s important and wealthy fourteenth-century Sayyid families—beyond the family’s close association with the Sufi saints Baba Farid and Nizam al-Din Awliya, they participated in a network of relationships with other Sayyids, ‘ulama, Sufis, merchants, the Delhi Sultans and their secretaries.45 One of their (p.78) members, Kamal al-Din Ahmad served Sultan Muhammad Tughluq in his Telangana army and was later appointed mushīr (counsellor) to the monarch and given the title malik-i mu‘azzam (the great malik).46 Muhammad Tughluq’s vizier Khwaja Jahan approached Kamal a-Din’s brother, Sayyid Husayn, to accompany him to Daulatabad. Sayyid Husayn was agreeable but placed two conditions before the vizier: first, he would retain his sayyid and Sufi dress (libās-i siyādat wa ahl-i tasawwuf bar man muqarrar bāshad)’ and second, ‘they would not require him to take up any employment (bahech shughl-i mu‘aiyan mashghūl nagardānand)’.47 The association with the Sayyid was important enough for the vizier and he accepted these stipulations. Other than networking with the Sultan and his administrators, the Kirmani Sayyids were also scrupulous in doing all they could to leverage their position within the Chishti mystical order at a time when the fraternity was expanding through a dispersal of its primary exponents in north India and the Deccan. Rather than facing anonymity as a mere Chishti adherent, a member of the Kirmani family, Nur al-Din Mubarak, travelled twice to Chisht in Afghanistan, home of the great ancestors of the Chishti mystic order and received the cloak of discipleship from Khwaja Qutb al-Din, the descendant of the primordial ‘Khwajas of Chisht’.48 This was a rare honour and within the circles of Chishti disciples raised the status of Nur al-Din Mubarak and the rest of the Kirmani family above all their contemporaries.49 Even as the Kirmanis accepted naukari, they enhanced their accomplishments to ensure that they were never completely hegemonized by any one relationship. The sum of their variegated relationships reflected the great social value they possessed, a public reputation that allowed

Page 11 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī them to negotiate favourable terms providing them unparalleled freedom and initiative. Not all naukars were so fortunate, and from the mid-thirteenth century, we can also notice the presence of exceptionally skilled but a socially humbler body of Muslims seeking service in Sultanate administration. Many of them were recent converts to Islam; some were mawlazadas (p.79) and certainly all were dependants (mawali) of their patrons.50 Barani provides us with an early picture of the arrival of these people, individuals who, despite their lowly social status, possessed knowledge of chancery protocol and secretarial skills that was fetching them naukari. Since they lacked birthright and standing in Muslim society, however, Barani showered them with abuse, something which he strategized quite admirably—he stepped around the powerful who could determine his own prospects, but condemned others.51 In his Ta’rīkh, Barani provided only the most cursory recollection of one of the most powerful families in the Tughluq history: the father and son, who served as viziers of Firuz Tughluq, both titled Khan-i Jahan.52 Barani’s contemporary, ‘Afif, provided a biography of the vizier and clarified that the father, Kannu, was from Telangana where he was a trusted courtier of the local Raja/Rai.53 He was captured during the Tughluq campaigns in the Deccan, transported to Delhi together with his master, and became a Muslim at the hand of Muhammad Tughluq (and was hence his mawali). Eventually he was given the title ‘Qiwam al-Mulk’ and became the deputy vizier of the Sultan. Kannu’s rise within the Tughluqid chancery was astonishing (and equally Barani’s silence on the matter), especially if one notes ‘Afif’s remark that the father ‘was not educated and could not write, yet he was very wise’. This was probably not strictly true: ‘Afif also mentions that the deputy vizier signed orders as ‘Maqbul, the slave (banda) of Muhammad [Tughluq]’.54 What ‘Afif meant was that Kannu’s administrative experience (p.80) notwithstanding, at the time of his deployment, he lacked the writing skills in Persian required of secretaries serving the Sultanate. Also, as a new convert (and in all likelihood a manumitted slave), Kannu lacked the social background or network that could provide him with status and was dependent upon his master for all favours. But not for long: the new convert went on to become the vizier of the realm in Firuz Tughluq’s reign and as his career graph would suggest, his deficiencies in education and social recognition were very quickly remedied: ‘Afif’s biography draws particular attention of the reader to the vizier’s meticulous record keeping.55 With power and prestigious secretarial skills, it was also not long before the vizier possessed a huge circle of supplicants; like many of his peers who were military commanders, this self-effacing banda was also no longer natally alienated.

Page 12 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī Nor was Kannu/Khan-i Jahan unique in the fourteenth century; there were other non-Muslim literati whose suspect Persian skills, not to speak of their infidel backgrounds (at least in the eyes of Barani), did not hamper the flow of Sultanate patronage. There was Ratan the fiscal administrator of Sindh, who was ‘skilled in calculation and writing’; Bhiran, the auditor (mutasarrif) at Gulbarga; Samara the governor of Telangana; and Dhara the deputy vizier at Daulatabad in 1344.56 None of them could have held positions of public responsibility described by Barani unless they were able to convince masters and subordinates, colleagues and supplicants that they were adept in diplomatics, accountancy, and mathematics (ḥisāb and siyāq).57 Their weaknesses in Persian must have also been quickly overcome otherwise they would not have held such high offices in the Tughluq administration. Officers such as Ratan, Bhiran, Samara, or Dhara were not resident in Delhi, nor were they as powerful as Kannu/Khan-i Jahan (whose influence would have been integral in any possible political rehabilitation of the disgraced Barani—hence the author’s politic silence), and Barani was brutal in his summation of their qualities. (p.81) Barani introduced the theme of moral corruption in politics, a consequence of misdirected royal patronage, in the very beginning of his history, in Balban’s reign. In the narrative of that monarch’s reign, Barani insinuated the story of Kamal Mahiyar, the son of a Hindu ghulam who was recommended to Balban for appointment as the accountant (khwaja) of Amroha.58 Although in his narrative, Balban was not swayed by this recommendation, Barani wanted his readers to know that this Sultan was the paradigmatic exception; later Sultans lacked Balban’s acuity and good sense. Kamal Mahiyar, for example, figured in his successor’s list of grandees.59 The favouritism shown to personnel of ‘poor lineage and background’ (kam aslī wa kam bizā’at), according to Barani, had reached its culmination in Muhammad Tughluq’s reign. And indeed it had, but this was not because of any deficiency displayed by the later Delhi Sultans. To the contrary: by the middle of the fourteenth century, despite the continuing immigration and favour shown to learned émigrés, the portfolio of skills that set members of the Persianate chancery apart was no longer the exclusive preserve of the distinguished secretarial families of the traditionally conceived Persianate ecumene. Although neither Barani nor ‘Afif wanted to get into the question of how these base Hindus came to possess knowledge of Persian epistolary traditions, mathematics, and accountancy, even a cursory review of Sanskrit land grants from an earlier period will underline the great technical expertise in land management and general administration possessed by local administrators.60 What is worth appreciating then is the adroitness with which the likes of Kannu transported these managerial skills from one cultural milieu to another.

Page 13 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (p.82) It is critical to keep in mind that references to these groups of Sultanate administrators commence from Balban’s reign and increase exponentially in the fourteenth century as the Sultanate armies cleared larger areas of the jungle tracts in the Gangetic plains, pushed into the areas of Rohilkhand, Bundelkhand, and Baghelkhand, and commenced their march through Rajasthan, towards Gujarat and into the Deccan. ‘Isami and Barani note the presence of local guides, interpreters, and itinerant provisioners who ferried supplies to the army; shadowy, ubiquitous personnel without whom the Sultanate could not survive.61 The conjuncture between the settlement of new agricultural tracts from the decades of the thirteenth century and the appearance of the ahl-i qalam of a different ilk (the Ratans, Bhirans, Samaras, Dharas, and Kannus) in Sultanate histories is not coincidental. For the Delhi Sultans these scribal groups, familiar with local conditions and languages with their distinct histories, proved to be valuable naukars. These were certainly the groups who would have the diglossic skills to produce the multilingual inscriptions and documents discussed in this volume in the articles by Sheikh and Eaton. The Delhi Sultans always in search for secretaries were delighted to recruit such dependent and able personnel. They were bandagan-like creatures (according to ‘Afif, a point that Kannu did not mind acknowledging), but they possessed skills and access to knowledge systems that elicited patronage, altering the sense of ‘employment’ into something honourable, a recognition that only particularly qualified individuals achieved for managing the public affairs of the community. And concurrently, we only need to pay attention to Barani to appreciate the tensions and hostility that the arrival of these people created within entrenched secretarial circles, a competition that the Sultans and other warlords used to their advantage. Any opportunity to gain advantage over secretarial groups was important for the military commanders. The ability of these groups (p.83) to entrench their positions in Sultanate society meant that changes in political dispensation sometimes also witnessed a great purge of the administrative staff.62 These were unusual occasions and were often carefully reported by chroniclers. But more generally, the ahl-i qalam showed great social resilience and their families remained important over generations in Sultanate society. For example, the great thirteenth-century chronicler and jurist, Minhaj-i Siraj Juzjani’s daughter’s son, Qazi Sadr al-Din ‘Arif, was married to the sister of Muhammad Tughluq. We have had a chance to notice their son earlier in the paper when he was involved in a coup to depose Firuz Tughluq.63 But what was the trajectory/fate of the bandagan and other déracinés in Sultanate society in the processes of reproduction of status and authority in Sultanate society? Did they remain natally alienated?

Establishing Homesteads, Alliances, and Natality: Transiting from Bandagi In Sultanate historiography there has been little interest in the study of the social rooting of Sultanate elites and its possible impact on the regime’s territorial dimensions.64 Instead, it is the impermanent nature (p.84) of Page 14 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī territorial command that has been the subject of scholastic comment. Sultanate historiography has placed a huge emphasis on the iqta‘’ as a means of introducing centralized governance—taxation, policing, curbing refractory chieftains—through the temporary ‘posting’ of Sultanate military commanders in various territories.65 The evidence does not bear out this understanding of early thirteenth-century governance: senior Sultanate military commanders—the bandagan-i khass—were seldom shifted from their commands, especially from the more important provinces. As soon as they were adjudged reliable administrators, Iltutmish seemed to have valued their experience and left them in situ in their various appointments.66 Although originally a strategic decision with special regard to the bandagan-i khass, by the middle of the thirteenth century, this special dispensation had become a more universal fait accompli—most military commanders used their iqta‘s (or seized the iqta‘s of others) as safe sanctuaries to challenge not just Delhi but all other competitors. Note the instances of Kabir Khan Ayaz in Sindh and Ikhtiyar al-Din Yuzbeg in Lakhnauti, who transformed their commands into independent Sultanates, and Balban with his headquarters in the Siwalikh region, Rayhan in Bahraich, and Qutlugh Khan in Avadh.67 Juzjani notes that although Balban spent most of his time occupied with the affairs of Delhi, he never lost sight of his investment in his iqta‘ where every town, qasba, and fort was managed by his administrators and servants (muta‘alliqān wa khuddām).68 Nor was he unique: when Rayhan and Qutlugh Khan evicted Balban from Delhi in 1253, the military commander retired (p.85) to his territory in the Siwalikh. Rayhan and Qutlugh Khan replaced Balban’s cohort in the capital with their own confederates, one of whom, Shams al-Din Bahraichi, was appointed the chief jurist of the realm.69 As the nisba in his name informs us, Shams al-Din belonged to Bahraich, a town associated with the early career of the two commanders, and the town to which Rayhan retired when forced to leave Delhi in the following year. Qutlugh Khan resided in nearby Avadh but shifted to Bahraich on Rayhan’s death. Already in the late thirteenth century these military commanders were ‘bi-locational’ in ways that Simon Digby described fifteenthcentury Afghans in his essay in this volume (Chapter 2). They had multiple residences and investments in social and political networks in different locales which sustained them during periods of conflict. The nature and resilience of these networks become clearer in Juzjani’s account of the conflict between Qutlugh Khan and Balban, where he describes the allies that supported the Bahraichi regime against Delhi. Qutlugh Khan’s allies had their bases in the mawās (plural, mawāsāt), a term that literally means ‘sanctuary’, in this case a refuge for local chieftains from Sultanate armies—an oddly empathic choice of term from the usually combative Delhi-centric Sultanate records.70 In the early fourteenth century, Ibn Battuta described the mawas as places where local chieftains

Page 15 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī fortif[ied] themselves in mountains, in rocky, uneven and rugged places as well as in bamboo groves. In India the bamboo is … big … its several (p. 86) parts so intertwined that even fire cannot affect them … The infidels live in those forests which serve them as ramparts, inside which are their cattle and their crops. There is also water for them within … they cannot be subdued except by powerful armies, who enter those forests and cut down the bamboos with specially prepared instruments.71 In his study of the Baghela kingdom of the fifteenth and sixteenth centuries, Simon Digby had clarified the ‘bi-axial’ nature of these kingdoms and their two relatively distinct ecological zones.72 The armed, strategic centre of the realm was usually located in hilly redoubts, protected by a mix of ravines, defiles, jungle tracts, and impenetrable thickets. This was intimately connected to an alternate agrarian and riverine zone that sustained agriculture and trade, providing these chieftaincies with considerable economic resources and material resilience. The Sultanate garrison towns of Avadh and Bahraich were surrounded by these mawas lands where Qutlugh Khan reportedly had a network of allies spread over a vast distance along the Himalayan terai and the lower plains, from Tirhut/ Mithila in the East to Sirmur in the West. Juzjani notes that when Qutlugh Khan appeared in Sirmur, he ‘had rights (ḥuqūq-i s̱ābit dāsht) on all these people. Wherever he (Qutlugh Khan) went they provided assistance because of his past claims (bajihat-i ḥuqūq-i mātaqaddum) and their hope for the [positive] conclusion of his affairs’.73 With reference to Rana Ribal (or Ranbal), the ruler of Sirmur, Juzjani reported of this great ruler of the Hindus that it was ‘the custom (p.87) amongst these people to offer sanctuary (‘ādat-i ān jamā‘at muḥāfażat-i multāji’ān būdī)’.74 Notably when Balban marched against Qutlugh Khan, he also had to vanquish that military commander’s allies. As it happened, both Qutlugh Khan and the Raja of Sirmur evaded Delhi’s army and fled into the jungles of the mawas. At that time Delhi’s retribution was directed against the stockade capitals of the Rajas, their commercial marts (bazārgāh), and the riverine and agricultural tracts—the economic spine of these territories had to be broken. The mutually supportive alliances between the local Raja and his ally and patron, the local Malik, placed the iqta‘ of Avadh within a regional matrix well outside the orbit of Delhi, giving the banda very, very dangerous roots. Persian chroniclers chose not to detail these ‘rebellions’, and yet the evidence at hand reflects remarkable consistency in the description of the imagined dangers to the Sultanate. In the last decades of the thirteenth century when Barani described Balban’s efforts to suppress Tughril (ca 1280–1), his recalcitrant slave governor of Lakhnauti, he used the phrase mutawaṭṭinān ān diyār, local residents or natives, to describe the people who supported the rebel.75 Certainly Page 16 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī from a Delhi-centric perspective it might be worth querying if the consolidation of the iqta‘ through accommodative alliances with local chieftains also carried with it the fear that the local Muslim governor had also ‘gone native’; mutawattin, from the Arabic root watan, homeland, would then have been an apposite term for the Persian chronicler to describe Tughril in Lakhnauti. Juzjani certainly described the earlier cited example of Balban’s campaigns against Qutlugh Khan and his allies as crusades: ‘where no king had ever reached and where no army of Islam (lashkar-i Islām) had ever gained victory’.76 Juzjani deliberately excised Qutlugh Khan’s prior presence in these regions since he was no longer a part of the Delhi Sultanate or the Muslim millat. Like Tughril, he was also at one with the natives.77 (p.88) That, of course, was the crux of the problem for Persian chroniclers: what kind of relationships should Sultanate commanders and Sultans have with the mutawattinan, the ‘natives’? The dangerous moment in this world occurred when the banda transited from natal alienation and established roots that brought him into close proximity with the infidel. This instance was used by the great Persian tawarikh-writers of the thirteenth and fourteenth centuries as an opportunity to insert significant didactic warnings in their narratives.78 As is evident from the way one of the last of these Delhi-centric histories, the Ta’rīkh-i Mubārak Shāhī of Yahya Sirhindi (completed 1434), embellished Barani’s narrative of Tughril’s rebellion in Lakhnauti, these insertions could be unexpectedly multileveled.79 A large part of Sirhindi’s narrative does not depart significantly from Barani’s plot, but for the insertion of a long story of Balban’s alliance with Raja Dhanuj of an unnamed region (and not Raja Bhoj of Sonargaon as in Barani). This was an unusual discursive interpellation for the normally taciturn Sirhindi, and its main focus was the protocol surrounding the meeting between the two monarchs. Sirhindi has Raja Dhanuj demand that the Delhi Sultan display appropriate honour by standing up to receive him. ‘But how’, Sirhindi had Balban anxiously ask his readers, ‘could a Muslim ruler show respect to an infidel?’80 The resolution to this problem, Sirhindi reported, was to have Balban sit on the throne with a falcon on his arm, and to stand to release the bird as Raja Dhanuj approached the throne. The expectations of the Raja were met with Balban standing to receive him, and the honour of the Delhi Sultan saved, even if by a contrivance. Although we may focus on this act of contrivance and the author’s unusual efforts at dissimulation, we should not miss Sirhindi’s communication (p.89) of relief at the staging of a successful stratagem that saved the dignity of a Muslim monarch. Written in the early fifteenth century, when the late Tughluqs and Sayyid monarchs were compromising with a range of military commanders, the story had a didactic message for Sirhindi’s elite readership regarding appropriate courtly protocol. Sirhindi did not want to alienate his readers and Page 17 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī was, therefore, very diplomatic in his summation of the political conditions in the Sultanate. Later, and in a different context, the sixteenth-century author, Bada’uni, showed no such reticence. He noted the consequences of bad governance on the late fourteenth century Tughluq Sultan(s): ḥukm-i khudawand-i ‘ālam az Dihlī tā Pālam, the writ of the Ruler of the World extends from Delhi to Palam.81 We can read the narratives of Persian chroniclers far more skilfully if we can comprehend the range of subjects that made them reticent to divulge the conditions of their age. Disaggregating this information so that the historical processes through which military commanders and their allies strove for autonomy brings out the thick complexity folded within the monolithic state—it serves to recapture the sublimated ethnographies of different personnel and the contexts in which they established relationships. While the authoritarian features of bandagi might have remained very attractive for the purposes of state formation, we have to recognize how its paradigms also erased the social history of very powerful political actors; the contradictions in providing political initiative to military commanders while denying them social rights was unsustainable in the long run. Since most scholars have not focused on the ways in which thirteenth- and fourteenth-century Sultanate records sought to erase or sublimate these changing political relationships, historians have remained trapped in a rhetoric that makes these centuries appear starkly distinct from the following years and in the reproduction of their silences. As we will see in the next section, however, the transitions from the thirteenth and fourteenth centuries into the fifteenth were a lot more complex than the formulaic ideas that ‘rupture’ or ‘change and continuity’ might suggest.

(p.90) Contextualizing the Fifteenth Century: Bandagi, Naukari, and Service As we turn to the north Indian Sultanates in the fifteenth century, it is perhaps mundane but still essential to recognize that these regimes did not emerge out of an inspiration to create alternate, structurally distinct political formations. The military and civil personnel that served the Sultanates of Jaunpur, Kalpi, Malwa, and Gujarat in particular had learnt strategies of war, alliance formation, administration, and revenue collection as commanders and governors of the Tughluq Sultans. This circulation of personnel, skills and knowledge systems, not to speak of diglossic skills, had occurred over a long period of time and had encompassed a diverse body of military commanders, secretaries, and other administrators. The Delhi Sultanate of the thirteenth and fourteenth centuries had always been in flux and its iqta‘ holders had periodically turned to neighbouring rulers as allies to consolidate their individual realms. The support from neighbouring Rajas that local commanders had counted on during their conflict with Delhi in the thirteenth and fourteenth centuries was an aspect of statecraft manifest in the fifteenth century, for example, in the relations that the Jaunpur Sultanate established with the Baghela Rajas of Bhatta (modern Rewa). Page 18 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī In 1482 the Baghela Raja Bhaidachandra provided money, horses, and elephants to Sultan Husayn Sharqi in his conflict with the Lodis of Delhi; in 1491 Bhaidachandra seized the Lodi governor of Kara as a part of a larger uprising of local rulers in support of the Sharqis; and in 1494, in the midst of Lodi aggression, Bhaidachandra’s son Lakshmichand invited Sultan Husayn to Jaunpur when he sensed Lodi’s weakness.82 How did these people communicate with each other and in what registers? If in the thirteenth and fourteenth centuries the Persian chroniclers of the Delhi Sultans had railed against the alliances of their Maliks (p.91) with Ranas, the fifteenth- and the sixteenth-century chronicles chose to ignore how the kafirān might be befriended by their own rulers; they criticized instead the alliances of neighbouring monarchs with the infidels. The Ma’ās̱ir-i Maḥmūdshāhī, a history of the Malwa Sultans by Mahmud Kirmani, reports the accusation of the Kalpi monarchs about the conduct of the Sharqi Sultan of Jaunpur who, during his campaign in 1443–4, had made Muslims ‘captive and had looted and carried away their property as if it had been an infidel land (dār al-ḥarb). A group of Hindus were attached to him, like the muqaddams (chieftains) of Gahora and Bakhesar’.83 In the same vein, the raconteur of the Lodi and Sur dynasty, Rizq Allah Mushtaqi (d. 1581–2) recorded the support given to the Jaunpur Sultanate by the Bachgoti rulers/chieftains with great hostility. Mushtaqi noted that when Sultan Sikandar (1489–1517) heard of the rebellion of Rai Jodha in the Jaunpur region, he explained to Sultan Ibrahim that his military actions were directed against the kafiran. Sikandar invited his fellow Muslim monarchs to reconsider his foolish alliance with this infidel whom he regarded as his servant (chākar).84 And this is the same Sultan Sikandar who eventually won over the Baghelas in 1494 and in 1498 commanded his ally, Raja Salivahan, to send one of his daughters to him as a bride!85 We can expect selective amnesia from dynastic histories that eulogistically record the activities of their own Sultans. But since the Persianate world of the fifteenth century did not produce a general history of the north Indian Sultanates or a normative text on statecraft, we lack an (p.92) overarching perspective to assess how strategic Sultanate alliances with the mutawattinan could have been a product of a different sensibility regarding political formations, a cumulative result of developments through the thirteenth and into the fifteenth century.86 This lacuna in the Persian archive can be usefully breached if we turned to a Sanskrit normative text, the Puruṣa parīkṣā, written in the early fifteenth century by Vidyapati, a Brahmin litterateur and courtier of the Rajas Kirti and Shiv Singh of the Oinivar dynasty who ruled over Mithila/ Tirhut.87 The rulers of Mithila were close allies of the Sharqis and had for many centuries been associated with the history of the Delhi, Bengal, and the Jaunpur Sultanate —we briefly noticed its ruler as an ally of Qutlugh Khan in his conflict with Page 19 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī Balban. Associations of this nature had not really threatened the autonomy of its Karnata dynasty in the thirteenth century, or the Oinivars that had supplanted it in the middle of the fourteenth century. Oinivar rule was briefly interrupted by a military adventurer, Malik Arsalan, in the late fourteenth century. But with the support of the Sharqi Sultan Ibrahim (1402–40), its rulers Kirti and Shiv Singh managed to re-establish Oinivar rule as allies of Jaunpur.88 Vidyapati was personally quite familiar with the Sharqis of Jaunpur; he had travelled to the city with his patrons, and described it in another text, the Kīrtilatā. Vidyapati was a prolific author, and other than his haunting Bhakti songs in Maithili, his Puruṣa parīkṣā became an extremely popular text.89 As is often the case with normative literary materials, the Puruṣa parīkṣā was also carefully grounded in the history of north India. But (p.93) this text is particularly significant and unusual for its choice of actors to illustrate ethical and moral principles. Its protagonists can be broadly differentiated into two groups, a division that can be illustrated from a consideration of the first four stories itself, where the fascinating mix of important historical protagonists (Vikramaditya, Sultan ‘Ala’ al-Din Khalji, Raja Jaichand, and Sultan Muhammad [Tughluq]) are separated from the rest of the socially and politically subordinate actors in the story—juxtapositions of status and class, but not of ethnicity or religious denomination. In an interesting narrative move, Vidyapati endows the second, subordinate group of characters with a heroic nature and uses them to provide moral lessons to the more powerful and famous. These distinctions mirror Vidyapati’s explanation for the writing of the book in the first place: it was arranged as a series of anecdotes narrated by the Brahmin Subudhi to help King Paravara discover the qualities of a ‘real man’ who could be an appropriate groom for his daughter. The deciphering of such ideals of masculinity in the Kaliyuga, as the king/father observed, required wisdom unique for its age. Three stories in the Puruṣa parīkṣā deal directly with the Delhi Sultans. The fourth story concerns Muhammad Tughluq, who had the two princes Narasimhadeo and Chachikadeo in his service. The former was a prince of the Karnata lineage (the old rulers of Mithila), the latter a Chauhan. In this story, the Muslim ruler was attacked by an unnamed ‘Kāfira’ Raja—a startling use of Islamic terminology in a Sanskrit text to define a non-Muslim king. Facing certain defeat, Vidyapati has Sultan Muhammad address his troops, ‘Oh, my Princes, champions of my army, is there none amongst you who by the might of his arm cannot for a short moment bring to stand my troops, routed as they are by the enemy?’90 On hearing their king’s plea, the two princes immediately rush into battle. Narasimhadeo kills ‘Kafira Raja’ with an arrow and is himself grievously injured. Chachikadeo, who was right behind, decapitates ‘Kafira Raja’ and brings the head to Sultan Muhammad. While Vidyapati has the Muslim king praise both heroes, the valour shown by the princes in the service of their ruler appears as a secondary virtue to the honesty they displayed in not exaggerating (p.94) their brave accomplishments. Even when they had the opportunity to Page 20 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī gain fame, each sublimated their own achievement and praised the other’s valour. There is complete normalcy in the narrative of the two Kshatriya princes serving a Muslim ruler in combating a third, non-Muslim antagonist. In deliberately referring to him as ‘Kafira Raja’, Vidyapati distances this nonbeliever from the shared world of Sultan Muhammad and the Princes—this was a stratified community of master and honoured naukars where membership was not contingent on their being Muslims. In this shared terrain, the Muslim Sultan was the leader of the natives. The forty-first story in the Puruṣa parīkṣā is an equally complex story concerning Sultan Shihab al-Din/Mu‘izz al-Din Ghuri of Delhi and Raja Jaichand of Kannauj.91 It also includes the latter’s beautiful and fickle wife and two Brahmin ministers, Chaturbhuj and Vidyadhara, ministers of Mu‘izz al-Din and Jaichand respectively. The story begins with Sultan Mu‘izz al-Din’s frustrated efforts to conquer Kannauj. Deciding finally to turn to stratagem rather than to arms, he sends Chaturbhuj to Kannauj to cuckold Jaichand’s wife. She will influence her love-besotted husband, Raja Jaichand, and alienate him from his wise minister Vidyadhara. The deceit works and the wise minister is marginalized. This gives Sultan Mu‘izz al-Din the opportunity to seize Kannauj. The story concludes with Vidyadhara’s valiant sacrifice in the service of his master, and Mu‘izz al-Din’s execution of Jaichand’s fickle wife. Although intended to serve as a moral about the dangers of greed and infatuation, the story’s subtext is about the two loyal Brahmin ministers who used all their skills in the service of their respective masters. Although this story, like the others, revolves around historical figures, there is nothing to refer readers to the historical conflict between Mu‘izz al-Din Ghuri and Jaichand or their differing religious affinities. Instead, the two rulers appear as well-matched heroic figures, largely because both have ministers who are very similar in social background, intelligence, sense of honour, and singleminded loyalty. The conclusion to the subtext of this story was that great disaster befalls a realm when its Brahmin minister, someone like Vidyadhara (and Vidyapati?), is no longer around to manage its affairs. Since both realms shared personnel (p.95) of a similar social profile, the reader has little but the greedy queen to differentiate between them. The other stories in Puruṣa parīkṣā also have a fairly precise geographical spread—from Gaur in the East to Ujjain and into Dwarka in the West, spanning the Gangetic plain. Vidyapati deployed his intimate knowledge of the history of the subcontinent quite deliberately to narrativize ethical principles of governance within a spatial and temporal past that had an obvious resonance to fifteenth-century elite participants. This was a text written by an extremely erudite, cosmopolitan scholar who was comfortable with his own Brahminical location and had internalized the knowledge of other cultures and histories to use them creatively in his own work. His Brahminical and Sanskrit training also included within its precincts the Sultanate world, and service to a yavana master

Page 21 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī posed no ethical or caste problems to his protagonists who were presented as exemplars of Kshatriya and Brahminical virtue.92 But perhaps his second story is the most significant amongst those referring to the Delhi Sultans. One of its dominant personalities is the Delhi monarch, Sultan ‘Ala’ al-Din Khalji, but there are also others, the real subjects of his story: Raja Hammir Deo, the ruler of Ranthambhor, his wives, and the renegade military commander Malik Mahim, who had fled from the Delhi Sultan’s wrath and received sanctuary from the Raja.93 The story details messages between the two monarchs, the bloody battles, and the extent to which Raja Hammir Deo and his wives remained true to the virtues of compassionate hospitality to their Muslim guest, and eventually gave up their lives valorously in his defence. This is not a novel story: it shares some features with Nayachandra Suri’s well-known Hammiramahākāvya (ca. 1401).94 Both share a Sanskrit provenance (p.96) and a similar set of actors, but differ vastly in their emplotment and their contextual events and locales. The truncated story in the Puruṣa parīkṣā is actually far closer to the episode of ‘Ala’ al-Din’s conquest of Ranthambhor narrated in ‘Isami’s Futūḥ al-salāṭīn, where the conflict originates because Rai Hammir gave sanctuary to two Mongol Maliks fleeing from ‘Ala’ al-Din.95 Despite the different actors, both texts carry similar kinds of diplomatic exchanges between the rulers, debate the question of honourable hospitality, and correspond in attributing the bloody conclusion to the siege to Raja Hammir’s stubbornness (a good or bad trait depending on the author’s position). Unlike the Hammīramahākāvya, the Puruṣa parīkṣā and the Futūḥ al-salāṭīn are not in the least concerned with Hammir’s wives as trophies of war. Although this is a significant exclusion and strengthens the Persianate link to the Sanskrit story, it hardly proves the direction of transmission. But that is belying the point: the interesting feature of the Puruṣa parīkṣā is not its passive reception of information from one text or another, but the choices made by the author. Vidyapati broke away completely from the Hammiramahākāvya and the Pṛthvīrāj rāso cycle of literature, or that of Kirmani and Mushtaqi. When he created a fictional world, it seemed to carry selective ideas from ‘Isami’s narrative, but also departed significantly in details and completely in what the reader should imbibe from it. For Vidyapati’s didacticism to be effective—which it was, judging from the wide circulation of its manuscripts—his messages had to be located in an imaginary world, but one that was socially and culturally cognizable to its readers. We can appreciate, however, how novel this textualized world might have appeared to readers of thirteenth- or fourteenth-century literary materials. Although very much a part of the thirteenth-century political experience, the earlier discussed Rajas of Sirmur or Sonargaon were never models of naukars, or paradigms of hospitality in the literary records of that time. And had we only read Barani’s invective towards low-born administrators and were (p.97) unaware of Kannu’s service with Firuz Tughluq, the presence of Brahmin viziers serving Sultans in the Page 22 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī fifteenth century would appear absurd. For Vidyapati, however, these details were utterly commonplace; his focus was not on the actors but on the morals the readers could derive from their activities. As an insight into the cultural sphere of a greater Jaunpur region, the Puruṣa parīkṣā provides us with an unparalleled insight into how a range of interpersonal relationships could be hypothesized for elites as if they were not just normal, but ‘rules for [heroic] men’.96 While Vidyapati was relatively unique in his historical insights on the ecumene of the north Indian fifteenth century, the narrower perspectives of the Persian narratives produced for the Afghans also contained a commentary on the changes that their world was experiencing, particularly on their understanding of bandagi and naukari in the context of Lodi and Sur state formations. All of this literature is retrospective; it was produced during the rule of the Mughals and carries an important assertion of Afghan history and ethnicity when there was considerable fear of its loss. Their narratives often telescope changes that occurred in their clan organizations—even in their sense of a composite Afghan identity—and have to be carefully disaggregated.

Afghans between Bandagi and Naukari At their time of migration and establishment of power under the Lodis (1451– 1526), the Afghans were organized in kindred groups which were relatively nucleated but usually presented as agnatic bloodlines.97 These were centred on the family of a chieftain, and included assorted fictive kinsmen and were described as khayls or birādarīs. Political fortunes expanded and contracted this group and there was always the possibility (p.98) of confederates striking off on their own and seeking neo-eponymous identities. Authority was quite diffuse within the early warband, especially during the reign of Bahlul Lodi (1451–89), but sharpened considerably in the following decades and especially in the years of Ibrahim Lodi (1517–26) and Sher Shah’s rule (1540–5). Despite the gradual intrusion of a variety of institutional checks on clan autonomy and privilege, authority was often mediated through sophisticated cultural codes where individual rights, traditions of honour, and prestige were subjects of great concern. These rights had material dimensions—cattle, home, land, women, family—as well as more abstract, subjective terms of references as well, such as protocols of social interaction, language, humour, and so on. The processes of kin/clan formations and establishment of durable regimes placed Afghan practices and their traditions within a larger Persianate and north Indian domain that created, as one historian put it, new ‘cosmopolitan contradictions’.98 Persian litterateurs who reported on the Afghans were cognizant of both, that is, of the history of the Sultanate prior to the Lodi Sultans and of the ways in which the new rulers sought to insinuate themselves within these domains without the loss of their own identities.

Page 23 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī These transitions are rather usefully captured in the sixteenth-century account of the Lodi seizure of power in the Wāqi‘āt-i Mushtāqī. Here the author narrates how Bahlul Lodi’s Afghan contingent strategized living up to their by now well enshrined humble reputations and assuaged the concerns of their patron by a display of rustic innocence. Mushtaqi narrates how the Afghans decided to ‘appear like simpletons (tajāhul numāyand) and behave foolishly so that [the vizier of the Sayyid monarch] may consider us sympathetically as simple people, with no ambitions of greatness’. Indeed the behaviour of Bahlul’s soldiers shocked the vizier and he asked, ‘What kind of people (chih qisam mardum) are these?’ And Bahlul replied, ‘They are savages (waḥshīyān-and). They know how to eat and die; nothing more’.99 The Afghans gained proximity to their patron by recourse to exactly those unflattering qualities ascribed to them by the early Persian litterateurs (p.99) whom we have already studied. In this narrative, the author has both his Afghan protagonists and the vizier of the Sayyids recognize that these were the qualities that Sultanate prejudices associated with bandagi. But the two protagonists in this tale were seizing on these unflattering characteristics to achieve dissimilar goals. Mushtaqi explains the intentions of the Afghans quite lucidly. Rather than accept the ties of bandagi and the processes of gradual bonding and acculturation that would deracinate them but provide political privilege, Mushtaqi has the Afghans protest at their exclusion from the private meeting between the vizier and Bahlul Lodi. ‘We are not the domestic servants (chākar)’ of Ballu, they protested, and that they were as much the naukars of the Sayyid vizier, and should, therefore, receive equal preference from the vizier. In the face of this logic (and their rambunctious behaviour), the vizier capitulated, the Afghans were allowed into his private quarters where, as the conclusion to the grand strategy, he was imprisoned. The term chakar literally means servant, but note the author’s distinguishing it from naukar, and later from birādar (brother, and by extension, a member of the kindred biradari). Nor is it ever confused with banda, the ordinary slave. The distinction is one of hierarchy based upon levels of social entitlement, where brothers and naukars weigh in with more influence than respectively the chakar or the banda.100 Since these terms were never randomly used, attention to their systematic usage is very useful in order to access and contextualize underlying social and political hierarchies. (p.100) Mushtaqi’s text was produced in the sixteenth century during the period of Mughal rule, and its retrospective gaze at the early history of the Afghans communicated a complex mix of history (sometimes from multiple perspectives) and a nostalgic didacticism of a courtier romanticizing the good old days.101 He had Bahlul caricature his Afghan contingents as savages, an assessment that coincided with that of the Sayyids’ vizier. In this Mushtaqi also had Bahlul suggest to the vizier that, though an Afghan himself, he was quite different in status and cultural upbringing. The lingering echo of what this difference might imply was lost in the unravelling conspiracy to depose the Page 24 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī vizier, but not before Mushtaqi managed to communicate his nostalgia for a time when state formation had not yet introduced elements of stratification in Afghan society, when the Afghans could still refer endearingly to their chieftain as ‘Ballu’. The logic that animated naukari was voiced again by the Farmuli chieftain Khani Khanan at the death of Bahlul and the accession of Sikandar Lodi in 1489.102 At that time, ‘Isa Khan Lodi and some members of the Sahu Khayl clan were manoeuvring to keep Sikandar Lodi out of succession to the throne. Sikandar’s mother intervened to plead her son’s case. She was insulted: her humble social origins (she was, reportedly, a goldsmith’s daughter) were invoked to discredit her son, and quoting from the Panchatantra, ‘Isa Khan reminded his audience that ‘An ape can’t do the work of a carpenter’. Khan-i Khanan Farmuli protested against this insult and was chastened by ‘Isa Khan to remember his station, since ‘he was no more than a naukar, he should not interfere in the affairs of the royal family’. ‘Isa Khan may as well have substituted (p.101) the phrase bandai khass or chakar for naukar here, and for a member of as important a family as the Farmulis this was a clear put down. Khan-i Khanan was not one to let such an insult pass and his retort clarified his right to intervene: ‘I am the [exclusive] naukar of Padshah Sikandar and not the naukar of anyone else’. He was participating in the deliberations as a servant and friend of the (future) king. The subjective assessment of social entitlement meant that naukari was always susceptible to lapsing into bandagi in this world, which is why I think it impossible to gather its meanings in as ahistorical and culturally neutered a term as ‘service’. When Khan-i Khanan Farmuli and Bahlul’s Afghan retinue described themselves as naukars and not chakar or, hypothetically speaking, bandagan, they were sensitive to the meanings of honour and status present in one set of relationships but absent in the latter. But equally, we need to remember that those dispensing patronage— the vizier of the Sayyids (in the case of Bahlul’s Afghan retinue) and ‘Isa Khan Lodi (in the case of Khan-i Khanan Farmuli)—were trying to diminish the sense of naukari by ascribing to it qualities that were perilously close to bandagi. Afghan personnel were present in large numbers in the Sharqi Sultanate as well, and sixteenth-century Afghan chroniclers privileged Afghan ethnicity between these dispersed groups serving rival regimes. People of the same ethnic background, it was suggested, had a common matrix through which they comprehended ties of service. When an embassy from the besieged residents of Delhi tried to get the great Sharqi military commander, the Afghan Mubarak Khan Lodi, to defect or at least to relent his siege, Mushtaqi described how the Delhi envoy nonchalantly probed the military commander about the nature of his relationships (nisbat) with the two monarchs: his non-Afghan Sharqi master and Bahlul Lodi. This was a leading question, and Mushtaqi used it as an opportunity to distinguish between ‘service as employment’ (as Dirk Kolff might understand Page 25 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī it), and ‘service with honour and social entitlement’. Mushtaqi had Mubarak Khan’s reply admitting his difference in status in the two realms: he was a mere servant of the Sharqi Sultan (chakar) but the brother (biradar) of Bahlul Lodi.103 (p.102) Slightly later in time, there are clear references to very influential bandagan and mawlazadas within Sher Shah’s dispensation. Malik Sukha was one of his old slaves, and his two sons were both titled Khawass Khan—the ‘Special Khans’, a title which in its absence of a nisba was appropriately indicative of their natal alienation. The younger performed the tasks of the older on his death—a seamless continuity that fused the personae of the individual slaves.104 But other than at the point of his appointment by Sher Shah, an author like Sarwani never mentions the slave status of the junior Khawass Khan ever again. Khawass Khan rose to great heights of trust and favour in Sher Shah’s reign and was involved in campaigns into Bengal, Rajasthan, and against the Mughals.105 His obvious importance in Sher Shah’s dispensation cannot be disputed and Mushtaqi, in fact, devotes a long section to this courtier in his text. Here he discourses on the military commander’s great authority in Sher Shah and Islam Shah’s reigns, his courtliness, his hospitality, and his alienation from Islam Shah, one of his master’s sons.106 Mushtaqi provides absolutely no information that would divulge that Khawass Khan was a mawlazada, nor would it have served his narrative in any way because Mushtaqi’s protagonist carried no vestiges of bandagi anymore. This curious inhibition to identify those who were the special slaves of their masters (bandagan-i khass) persists in one of Mushtaqi’s rare account of a banda, Malik Suman Zabardin, an important commander of the Jaunpur Sultan Mahmud Sharqi (1440–57).107 There is actually very little in Mushtaqi’s text that reveals that he was a slave. The first hint we get of his jural status is when Mushtaqi ambiguously mentioned that earlier Malik Suman ‘lived’ with (literally, ‘was with [mībūdī]’) Malik Fath Khan Harevi, who apparently cared very deeply for him. He goes (p.103) on to add, but with no greater clarity, that on one occasion Mahmud Sharqi wilfully demanded Malik Suman from Fath Khan as a gift. Fath Khan was devastated at the imminent loss of his prized servant, when the slave asked, ‘who am I that for my sake you should so grieve?’108 Malik Suman went on to reassure Fath Khan that while serving the monarch, he would still remain ‘sincere and loyal (muḥibb o mukhliṣ)’ to him because his ‘friendship and dependence (ittiḥād o i’tiqād)’ was to such an extent that his corpse would be first brought to Fath Khan’s house.109 Mushtaqi narrated the story to underline Malik Suman’s exceptional sentiments of love for his first master. This was not unusual in itself for military slaves; Juzjani had provided several instances where the Shamsi slaves fondly remembered the care that their first masters had lavished upon them. In Mushtaqi’s story, however, the deep sentimental attachment between the slave and the master rooted him. Apparently the slave was not moved by the status of Page 26 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī the Sultan, his new master and, in a rare instance where the unfree were given a voice, Mushtaqi has Malik Suman declare that his home was with Fath Khan, his first master. As in the case of Khawass Khan, Mushtaqi’s dispersed information on Malik Suman’s life congeals into an account of an individual of great personal honour and integrity, an attribute that both protagonists carried into their public service.110 In this account, Malik Suman and Khawass Khan could have been Afghan naukars; they could not be biradar (brother, clansmen) since they were not members of the Afghan khayls. Nor would Mushtaqi have described them as bandagan (not even bandagan-i khass), or chakar, for the former terms’ associations with natal alienation and the latter’s with a humiliating service without honour could not communicate the exalted position that he felt the two Maliks occupied in their separate regimes. If we recall the historical dialectic between the two terms bandagi and naukari, we can appreciate Mushtaqi’s efforts at ambiguity; it actually served the purpose of communicating their status quite precisely. During Sher Shah’s rule in the sixteenth century, the social contexts and hence the sense of prestige of Afghans of high status and ancient lineages was under stress. It made people like ‘Abbas Khan Sarwani’s (p.104) venerated grandfather, Shaykh Bayazid, very ‘anxious for his honour (tā‘żīm-i khud) because … Sher Khan had changed the custom (qā‘ida) of honouring the Afghan nobles’. But Sher Khan showed discrimination in the display of his authority, and when Shaykh Bayazid had an audience with the monarch, he ‘was extremely happy that the king observed the old traditions of my ancestors that prevailed in the Afghan community (qā‘ida-i qadīm buzurgān-i man ki dar qaum-i Afghān būd nigāh dāsht)’.111 More than at any other time, it made a difference to participants if their interpersonal relationships, and hence the terms of their service, to powerful patrons were not reduced from naukari to bandagi.

Conclusion When historians struggle to represent continuity and change over any length of time, their sense of historical time leads them to organize their narratives into time periods that carry shared qualities, variously interpreted. In the context of the pre-Mughal past, a uniquely statist reading has led to the circulation of terms such as ‘Regional Sultanates’ in the historiography of the subcontinent’s medieval period to contrast the fifteenth and sixteenth centuries from the centuries of dominance by the (great) Delhi Sultans. This difference was then extended into other areas where the fifteenth century was interpreted as a period of diminutive cultural production—in the areas of arts, literature, architecture, and so on. The chapters in this book challenge these assumptions as they also foreground the ability of local political formations to emerge as centres of patronage and cultural efflorescence with a transregional impact. These arguments, however, still leave the thirteenth through the sixteenth centuries divided into the worlds of the Delhi Sultanate and the Regional

Page 27 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī monarchs, confirmed in the argument that these centuries also witness a shift from the dominance of the Persianate to vernacular cultural realms. My essay has taken a slightly different approach to periodization. It deliberately conflates the pre-Timurid and the post-Timurid centuries in its analysis of political culture and relationships, linking these with processes of state formation and resistance that were not dramatically (p.105) different over the passage of these centuries. There was considerable continuity in the composition of political elites and their subordinates, and the kind of political relations that they structured to entrench their respective positions. I found it useful to use the terms bandagi and naukari to bring out the politics of this time period and the great deal of tension within this world, as different participants competed with each other and defied the relations of service imposed on them. The terms bandagi and naukari were etic categories that allowed me access to emic understandings of status and autonomy as they were (or equally important, were not) discussed in the literary materials of the time. ‘Honourable service’ was inflected by a variety of factors that were structural—class, generational location, knowledge, skills—and a variety of contingencies: the question of whose servant/slave, for example, could throw an individual’s status up for debate. These etic categories were not very distant from emic ones regarding social entitlement and political status and remained extremely subjective and rhetorical in their modes of presentation in the Persian records of the thirteenth through the sixteenth century—one man’s naukar appeared like another one’s chakar/banda over and over again. Or equally, understanding the contingencies that might frame the representation of elites gives us an insight into why Kannu/ Khan-i Jahan, the vizier of Firuz Tughluq, was not tarred with the same brush as a whole host of other Sultanate administrators with a similar social profile. While my analysis was context-specific and diachronic, I juxtaposed bandagi and naukari as the extreme book ends of a sliding scale, within which the qualities of service and social entitlements could vary. The usage of these terms could therefore be assessed over time and from multiple points of view. Rather than an abrupt transition, my work argues for an increasing sense of social entitlement amongst political and social elites through this period. The bandagan or other social menials and émigrés did not remain deracinated forever. As they established roots, they made claims in the political realm that were commensurate to their sense of social status. And as the domain of bandagi shrank, some of its ideas shifted to the sense of chakari. But it is the boundaries of naukari that expanded, and even as it enfolded ideas of selfless loyalty from bandagi, it was also touched by the new sensibilities of biradari and qaum. From this perspective, rather than a study of abrupt departures, I consider the thirteenth–sixteenth centuries as a (p.106) period of assimilation of new experiences without the germination of new political structures.

Page 28 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī For the interim, the idea of naukari expanded to include new associations, and so did the idea of bandagi. The latter occurred largely under the influence of mystical pietistic movements that gained ground from Nizam al-Din Awliya’s death (1325), where the idea of bandagi came to circulate within a new, larger subcontinental environment and within a more disciplined set of signifiers.112 In his Fawā’id al-fu’ād, Nizam al-Din’s disciple Amir Hasan Sijzi already uses banda in a novel, distinct mode. He describes himself in the introduction as ‘banda’i gunāhgār, the abject sinner’ who had the benefit of kissing the feet of the monarch of the world.113 When Barani used the term in his Fatāwā-yi jahāndārī, he was picking up another discrete context in which bandagi could be used. He reminded his readers that Adam and his descendents were created for the service of God (barāy bandagī-yi ḥaẓrat-i ṣamadiyat), an abject condition that kings may not readily accept.114 This was in contrast to other pietistic individuals: roughly contemporaneous with Barani we notice banda used creatively in the title ascribed to the Chishti Sufi pir, Sayyid Gisudaraz Bandanawaz (d. 1422). By the time of his death, however, bandagi no longer had just a Sufi provenance, and Mushtaqi notes its transition as a title attached to two ‘alims from the period of the early Lodis: Bandagi Mi’an Qadan Danishmand and Bandagi Malik al-‘Ulama Mi’an ‘Abd Allah Ajudhani.115 In the sixteenth century, when it was no longer honourable to follow bandagi in the service of another mortal, its virtues of self-abnegation and selfless service were ideal in the (p.107) khidmat of God. Its possibilities, of course, were explored by Akbar, and by Sarwani writing in that monarch’s reign (ca. 1570s–80s): As the subjects of God (bandagān-i ḥaqq) are all under your command You also submit (bandagī kun) to God and carry out His order; For every king who girded his loins for the obedience of God People also girded readily their loins for his service (khidmatash).116

In our search for conjunctures we can sometimes underappreciate the cumulative impact of experience. We need to keep in mind that the struggle to expand the domain of naukari was carried out over four centuries in the face of entrenched opposition that would wish to incarcerate autonomy and the search for social entitlement within the circuits of bandagi. As Ali Anooshahr has recently pointed out, Mushtaqi’s text stands at a cusp in historiographical representation: it discusses the agency for historical change available to individuals in a manner quite distinct from the fatalism present in the great Sultanate tawarikh of an earlier generation of authors like ‘Isami.117 In contrast to taqdīr, predestination, which dominates in ‘Isami, Mushtaqi speaks of tadbīr, individual strategy and human agency. The combination of implications present in one instance when Mushtaqi uses this term is particularly striking. While on campaign Mushtaqi has Khawass Khan explain to his wonder-struck comrades that his efficient management of supplies was ‘not [because of any] karāmat but my tadbīr’.118 If we recall, Khawass Khan was the mawlazada of Sher Shah who Page 29 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī by dint of his service had become such a powerful, important commander that Mushtaqi did not think it essential to mention his slave origins. There was nothing fortuitous or accidental in slaves transiting from bandagi to naukari; it was, as in Khawass Khan’s case, by personal volition, tadbir. And the distance travelled between the litterateurs of the thirteenth and fourteenth centuries and Mushtaqi lay in the latter’s ability to make that philosophic distinction and credit an individual such as Khawass Khan with that human agency. Again, the glimpses of these shifts visible in Mushtaqi were more carefully enunciated in Akbar’s reign in the writings (p.108) of his chronicler Nizam al-Din. In this sense, Mushtaqi’s observations on the changes that were emerging towards the end of the Delhi Sultanate were the beginnings of a new world, enough for us to make out in it the outlines of an early modernity. But to be able to comprehend how this was possible and their complex interlinkages, we will have to recognize the need to connect the thirteenth with the sixteenth century more systematically and meaningfully than we have done so far. Notes:

(*) This chapter has profited from the critical interventions of Samira Sheikh, Francesca Orsini, Munis Faruqi, Tanika Sarkar, William Pinch, Pankaj Jha, and Ali Anooshahr. I am grateful to Francoise ‘Nalini’ Delvoye for the opportunity to first think about the subject in my lectures at the EPHE in 2006. Anjali Kumar, as always, was a part of its writing. This would have been a lesser work without Sikandar Kumar’s perspicacious engagements with its arguments. I am grateful to Ali Anooshahr and Pankaj Jha for making the texts of Mushtaqi and Vidyapati available to me. (1) Note Peter Jackson, The Delhi Sultanate: A Political and Military History, Cambridge: Cambridge University Press, 1999, whose penultimate chapter, ‘Stasis and Decline: Firuz Shah and His Successors’, is followed by ‘Epilogue: c. 1400–1526’. (2) Dirk Kolff, Naukar, Rajput and Sepoy: The Ethnohistory of the Military Labour Market in Hindustan, 1450–1850, Cambridge: Cambridge University Press, 1990; Simon Digby, War Horse and Elephant in the Delhi Sultanate: A Study of Military Supplies, Karachi: Orient Monographs, 1971; Digby, ‘Anecdotes of a Provincial Sufi of the Delhi Sultanate: Khwāja Gurg of Kara’, Iran, 32, 1994: 99– 109; Digby, ‘Before Timur Came: Provincialization of the Delhi Sultanate through the Fourteenth Century’, Journal of the Economic and Social History of the Orient, 47(3), 2004: 298–356; Digby, ‘Two Captains of the Jawnpur Sultanate’, in Circumambulations in South Asian History: Essays in Honour of Dirk H.A. Kolff, Jos Gommans and Om Prakash (eds), Brill: Leiden, 2003, pp. 159–78; Digby, ‘The Indo-Persian Historiography of the Lodi Sultans’, in Les Sources et le temps, F. Grimal (ed.), Pondichéry: l’École française d’Extrême-Orient, 2001, pp. 243–61; Digby, ‘‘Abd al-Quddus Gangohi (1456–1537 A.D.): The Personality and Attitudes of a Medieval Indian Sufi’, Medieval India: A Miscellany, Bombay: Asia Page 30 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī Publishing House, 1975, vol. 3, pp. 1–66; Digby, ‘Dreams and Reminiscences of Dattu Sarvani, a Sixteenth Century Indo-Afghan Soldier’, Indian Economic and Social Economic and Social History Review, 2(1), 1965: 52–80, and 2(2): 178–94. Digby’s contributions are considered slightly later in the chapter. (3) For details on the huge standing army of the two monarchs see Peter Jackson, ‘The Problems of a Vast Military Encampment’, in Delhi Through the Ages, R.E. Frykenberg (ed.), Delhi: Oxford University Press, 1986, pp. 18–33, and more tangentially Irfan Habib, ‘The Price Regulations of ‘Alā’uddīn Ḵẖaljī—A Defence of Ẓiā ’Baranī,’ Indian Economic and Social History Review, 21(4), 1984: 393– 414. (4) All of the founding dynasts of the Khalaji, Tughluq, Sayyid, and Lodi regimes were immigrants to the subcontinent and originally frontier commanders. For a more detailed response of the Persian literati to these groups of people, see Sunil Kumar, ‘The Ignored Elites: Turks, Mongols and a Persian Secretarial Class in the early Delhi Sultanate’, Modern Asian Studies, 43(1), 2009: 45–77, and Kumar, ‘Courts, Capitals and Kingship: Delhi and its Sultans in the Thirteenth and Fourteenth Centuries CE’, in Court Cultures in the Muslim World: Seventh to Nineteenth Centuries, Albrecht Fuess and Jan Peter Hartung (eds), London: SOAS/Routledge Studies on the Middle East, 2011, pp. 123–48. (5) See, for example, the early work of B.D. Chattopadhyaya, ‘Origin of the Rajputs: the Political, Economic and Social Processes in Early Medieval Rajasthan’, Indian Historical Review, 3(1), 1976: 59–82, oddly missing in Kolff’s bibliography. And from the adjoining region of Transoxiana and Afghanistan, but considering historical processes that have a bearing on South Asia, see Jean Aubin, ‘L’ethnogénèse des Qaraunas’, Turcica, 1, 1969: 65–94. (6) For an important overview on the sources of this period and the historiographical issues they raise for the historian, see Francesca Orsini, ‘How to Do Multilingual Literary History? Lessons from Fifteenth and Sixteenth Century North India’, Indian Economic and Social History Review, 49(2), 2012: 226–46. (7) Carlo Ginzburg, ‘Checking the Evidence: The Judge and the Historian’, Critical Inquiry, 18(1), 1991: 84. (8) While the term bandagī has not been used by historians (as yet) to unravel aspects of pre-modern political culture, the term itself is not a neologism. See, for example, Ziya al-Din Barani, Fatāwā-yi jahāndārī, A. Salim Khan (ed.), Lahore: Idarah-i Tahqiqat-i Pakistan wa Intishgah-i Punjab, 1972, p. 333. The term seems to have had considerable currency as a title in the fifteenth and sixteenth centuries. For a notable example, consider the title of Khizr Khan (r. 1414–21), the founder of the Sayyid dynasty. See Yahya Sirhindi, Ta’rīkh-i Mubārak Shāhī, M. Hidayat Hosain (ed.), Calcutta: Bibliotheca Indica, 1931, p. Page 31 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī 181 and the title of the monarch ‘Bandagī rāyāt-i a‘lā’ Khizr Khān’. This would translate literally as ‘In the Service of the Exalted Banners’. Sirhindi’s anecdotes on the monarch’s Sayyid status would suggest a reference to ‘service to the banners of the Prophet’. But there is enough ambiguity to allow readers to consider a possible genuflection towards Timur: see Sirhindi (1931), p. 182, and the title of the monarch stated as bandagī’i bandagān rāyāt-i a‘lā’ and the conclusion to this chapter for a further discussion of post-Sultanate usages of bandagi. (9) For a useful discussion of the Mongol nökör, see V. Vladimirtsov, Le Regime social des Mongols, M. Carsow (trans.), Paris: A. Maisonneuve, 1948, pp. 110– 30; Igor de Rachewiltz, The Secret History of the Mongols: A Mongolian Epic Chronicle of the Thirteenth Century, Leiden: E.J. Brill, 2004, vol. 1, pp. 256–7; and Gerhard Doerfer, Türkische und Mongolische Elemente im Neupersischen, Weisbaden: Franz Steiner Verlag, 1963, vol. 1, pp. 521–6. (10) On Malik Kafur Hazar Dinari, see the synopsis of Peter Jackson (1999), pp. 174–7, 206–7. (11) Jackson (1999), pp. 158, 177. (12) Shams-i Siraj ‘Afif, Ta’rīkh-i Fīrūz Shāhī, Maulavi Vilayat Husain (ed.), Calcutta: Bibliotheca Indica, 1888–91, pp. 436–45. (13) For his appointment during the reign of the Tughluq monarch Sultan Mahmud Shah II (1393–5), see Sirhindi (1931), pp. 156–7. (14) For an account of the khassa khayl, see I.H. Siddiqui, Some Aspects of Afghan Despotism in India, Aligarh: Three Men Publication, 1969, pp. 111–17 and Siddiqui, ‘The Army of the Afghan Kings in India’, Islamic Culture, 39, 1965: 223–7. The khassa khayl contingents seem to have been very similar to the thirteenth- and the fourteenth-century qalb retinue—the central contingent of the army that surrounded the ruler at moments of conflict and stayed with her/ him in the capital. The qalb contained a significant number of the ruler’s bandagan as well; see Sunil Kumar, Emergence of the Delhi Sultanate, Delhi: Permanent Black, 2007, pp. 78, 132, 258–9, 321. For the recruitment of slaves and their sons by Sher Shah, see also Kolff (1990), pp. 66–7. (15) Minhaj-i Siraj Juzjani, Ṭabaqāt-i Nāṣirī, Abdul Hay Habibi (ed.), Kabul: Anjuman-i Tarikh-i Afghanistan, 1963–4, vol. 2, p. 4. (16) ‘Afif (1888–91), pp. 435–7. (17) Juzjani had used this phrase in the context of another slave of Iltutmish, Hindu Khan. See Juzjani (1963–4), vol. 2, p. 19.

Page 32 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (18) This would contradict much of the historiography on the Delhi Sultanate where the reign of ‘Ala’ al-Din Khalji is posited as a ‘revolution’ because of the transitions in the racial and jural profiles of the personnel: Turkish slaves giving way to nobles who were free and non-Turks. For the primary exponents of this view, see Mohammad Habib, ‘Introduction to Elliot and Dowson’s History of India vol. II’, in Politics and Society during the Early Medieval Period, Khaliq A. Nizami (ed.), Delhi: People’s Publishing House, 1974, vol. 1, pp. 33–110 (especially pp. 103–10 for the changes that occurred in the ‘Indo-Turkish Slave Bureaucracy of the Thirteenth Century’) and Irfan Habib, ‘Barani’s Theory of the History of the Delhi Sultanate’, Indian Historical Review, 7, 1981: 99–115 (especially, pp. 108–15). Peter Jackson (1999), pp. 82–5, 171–7, makes a similar assertion even though he notes the continued deployment of military slaves by fourteenth-century Delhi Sultans. He leaves unexplained how the ‘emergence of a new elite’ (p. 171) in his work implied any structural transitions in the way authority was dispensed by the Khalji or the Tughluq monarchs. (19) For a valuable discussion of the ties between the freed slave and his patron, see Paul G. Forand, ‘The Relation of the Slave and Client to the Master or Patron in Medieval Islam’, International Journal of Middle East Studies, 2(1), 1971: 59– 66 and Roy P. Mottahedeh, Loyalty and Leadership in an Early Islamic Society, New Jersey: Princeton University Press, 1980, pp. 84–9. For its manifestation in Egypt, see David Ayalon, ‘Studies in al-Jabartī I: Notes on the Transformation of Mamluk Society in Egypt under the Ottomans’, Journal of the Economic and Social History of the Orient, 3(3), 1960: 283–8. For the early Delhi Sultanate, see Kumar (2007), pp. 94–7, 123–4, and 132. (20) Here the advice of the Seljuq vizier Nizam al-Mulk (d. 1092) was particularly apposite: ‘A slave who one has brought up and promoted, must be looked after, for it needs a whole lifetime and good luck to find a worthy and experienced slave.’ See Nizam al-Mulk, Siyāsat nāma, translated by H. Darke, The Book of Government or Rules for Kings, London: Routledge and Kegan Paul, 1978 reprint, p. 117. (21) Ziya’ al-Din Barani, Ta’rīkh-i Fīrūz Shāhī, S.A. Khan (ed.), Calcutta: Bibliotheca Indica, 1860–2, p. 65; Shaykh Abdur Rashid (ed.), Aligarh: Department of History, 1957, p. 77. (22) Amir Khusraw, Tuḥfat al-ṣighār, IOL Persian Ms 412, fol. 50 seq., cited by Wahid Mirza, The Life and Works of Amir Khusrau, Delhi: Idarah-i Adabiyat-i Delli, 1974 reprint, pp. 51–2. (23) ‘Abd al-Malik ‘Isami, Futūḥ al-salāṭīn, A.S. Usha (ed.), Madras, University of Madras, 1940, p. 456; A.M. Husain (trans.), New York, Asia Publishing House, 1977, vol. 2, pp. 427–8. I have used Husain’s translation.

Page 33 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (24) This was Barani’s description of the kinds of people patronized by Sultan Muhammad Tughluq. See Barani, Ta’rīkh-i Fīrūz Shāhī, Khan (ed.), 1860–2, p. 505. Khan’s edition has a misprint: razālatarīn and razālagān should be read as raẕālatarīn and raẕālagān. (25) Note the critical intervention of S.H. Hodivala, Studies in Indo-Muslim History: A Critical Commentary on Elliot and Dowson’s History of India, Bombay: Popular Book Depot, 1939, vol. 1, pp. 305–10. On p. 305, Hodivala clarifies that the territory (talwandī) of the Bhattis was fragmented and centred around ‘a camp made by a ring of ox-wagons set close together’ near a water source. (26) ‘Afif (1888–91), pp. 36–7. (27) ‘Afif (1888–91), pp. 37–9. ‘Afif narrates the princess’s remark in the first person but he has clearly not taken the subject’s point of view, where instead of Mongol, Turk/Turushka would have been more likely the term used by the princess. By deliberately using the term ‘Mongol’, ‘Afif appropriated the woman within Sultanate prejudices and could glorify her readiness to sacrifice herself. All of this was important for ‘Afif because of its rather serious implications for his protagonist, Sultan Firuz Tughluq, the Bhatti princess’ son. (28) For further details, see Kumar (2009). (29) Jackson (1999), p. 128. (30) ‘Afif (1888–91), pp. 103–4. (31) See also Jos Gommans, ‘The Silent Frontier of South Asia, c. A.D. 1100– 1800’, Journal of World History, 9(1), 1998: 17–23. (32) Barani, Ta’rīkh-i Fīrūz Shāhī, S.A. Khan (ed.), p. 37 (For Balban as a descendent of Afrasiyab) and p. 27 for the ‘frivolous slaves’; Ta’rīkh-i Fīrūz Shāhī, S.A. Rashid (ed.), pp. 33, 44. (33) Note Amir Khusraw’s description of the Qara’una Mongol who had captured him in 1285: ‘he sat on his horse like a leopard on a hill. His open mouth smelt like an armpit, whiskers fell from his chin like pubic hair’. See Amir Khusraw, Wasaṭ al-‘ayāt, cited in ‘Abd al-Qadir Bada’uni, Muntakhab al-tawārīkh, Maulavi Ahmad Shah (ed.), Calcutta: Asiatic Society of Bengal, 1868, vol. 1, p. 153. Compare with his Tughluq nāma, composed circa 1320 where Ghiyas al-Din Tughluq is eulogized as Ghazi Malik and ‘Saviour of Islam’. See Amir Khusraw, Tughluq nāma, ed. Sayyid Hashmi Faridabadi, Aurangabad: Urdu Publishing House, 1933. (34) For further details on asymmetrical matrimonial relationships by Shamsi slaves, see Sunil Kumar, ‘The Woman in the Ḥisāb of Men: Sultan Raziyya and Early Sultanate Society’, forthcoming A. Page 34 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (35) On Kamal Mahiyar, see Barani (1860–2), pp. 36–9, Rashid (ed.), pp. 43–6, and on mawlazadgan in Kayqubad’s reign and their relations with the nawMusalman, Barani (1860–2), p. 134, Rashid (ed.), p. 155. (36) Orlando Patterson, Slavery and Social Death: A Comparative Study, Cambridge: Harvard University Press, 1982. For a useful revision of some of his arguments, see Indrani Chatterjee and Richard Eaton (eds), Slavery and South Asian History, Bloomington: Indiana University Press, 2006. (37) For Amir Khusraw’s social background, see Wahid Mirza, Life and Works of Amir Khusrau, Delhi: Idarah-i Adabiyat-i Delli, 1974, and the more recent Sunil Sharma, Amir Khusrau: The Poet of Sultans and Sufis, Oxford: Oneworld Publications, 2005. (38) Barani, (1860–2), p. 134, Rashid (ed.), p. 155. Rashid’s reading differs from Khan’s but is clearer. (39) For further details, see Kumar (2007), pp. 65–78. (40) Kumar (2007), pp. 146–51. (41) For details, see Kumar (2009): 45–77; Kumar, ‘Service, Status and Military Slavery in the Delhi Sultanate of the Thirteenth and Early Fourteenth Centuries’, in Richard Eaton and Indrani Chatterjee (2006), pp. 83–114, and Kumar (2011), pp. 123–48. The evidence cited here would contradict Irfan Habib’s assumptions that the early fourteenth century marked a ‘plebianisation of the nobility’ when the ‘sluggish nobility resting on its hereditary claims’ could not keep pace with the dynamic expansion of the political sphere and complex economic life of the Sultanate under the great Khalji and Tughluq monarchs. See Habib (1981), p. 109. (42) Notably, all the great litterateurs of the early thirteenth century—‘Awfi, alKufi, Juzjani—were first in the service of Qubacha, Sultan of Sindh, and on his defeat and death transited seamlessly to the service of Iltutmish. For details, see Kumar (2009), pp. 212–37. (43) See Kumar (2007), pp. 219n61 and 223–4, for the historiographical discussion of the terms ‘ulamā’-i ākhirat and ‘ulamā’-i duniyā and their contrasting juxtaposition by Nizam al-Din Awliya and Ziya al-Din Barani. (44) Amir Khwurd, Siyar al-awliyā’, Sayyid Mahdi Ghuri (ed.), Lahore: Markaz-i Tahqiqat-i Farsi Iran wa Pakistan, no. 23, Mu’assi-yi Intisharat-i Islami, 1978. (45) For a useful analysis of the Kirmani family’s network, see Jyoti Gulati, ‘Stature, Social Relations and the Piety-Minded: Reading Amir Khwurd’s Siyar al-Auliya’, MPhil. dissertation, Delhi University, Department of History, 2005,

Page 35 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī charts 1 and 2. The details on the Kirmanis below come from her unpublished dissertation, pp. 64–105. (46) Khwurd (1978), pp. 224–5. (47) Khwurd (1978), p. 228. (48) Khwurd (1978), p. 221. (49) Khwurd (1978), pp. 223–4. (50) For the technical position of the freed slave and convert in Islamic law, note the earlier discussion. See also Joseph Schacht, Origins of Muhammadan Jurisprudence, Oxford: Clarendon Press, 1975, p. 161; and A.J. Wensinck, I. Goldziher, and P. Crone, ‘Mawla’; R. Brunschvig, ‘‘Abd’; and D. Sourdel, C.E. Bosworth, and Peter Hardy, ‘Ghulam’, in Encyclopaedia of Islam, P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, W.P. Heinrichs (eds), Leiden: E.J. Brill, second edition, 1956–. (51) For Barani’s harsh criticism of administrators of low status in Muhammad Tughluq’s reign, see Barani, (1860–2), pp. 504–5. (52) Barani (1860–2), pp. 454, 512, 578–9. (53) For the biography of the two Khan-i Jahans, see ‘Afif (1888–91), pp. 394–430, and on his origins, see p. 394. (54) ‘Afif (1888–91), p. 395. (55) ‘Afif (1888–91), p. 397. (56) See Barani (1860–2), pp. 504–5 and Jackson (1999), pp. 185–6. (57) For an extremely valuable summation of the personal, social, and educational abilities that secretaries in the Sultanate might have been expected to possess, see Nizami ‘Aruzi Samarqandi, Chahār maqālah, Mohammad Qazvini (ed.), revised by Mohammad Moin, Teheran: Zawwar Bookshop, 1955–7; Edward G. Browne (trans.), London, E.J.W. Gibb Memorial Series, 1978 reprint. (58) On Kamal Mahiyar, see Barani, Ta’rīkh-i Fīrūz Shāhī, Khan (ed.), pp. 36–7, Rashid (ed.), pp. 42–3, and more generally for the incident, Khan (ed.) pp. 36–9, Rashid (ed.) pp. 43–6. (59) Barani, Ta’rīkh-i Fīrūz Shāhī, Khan (ed.), p. 126, ed. Rashid, p. 147. (60) There is a huge literature on this subject and personal favorites would include Y. Subbarayulu, Political Geography of the Chola Country, Madras: State Department of Archaeology, Government of Tamilnadu, 1973, and his South Page 36 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī India under the Cholas, Delhi: Oxford University Press, 2012. Closer in time to the Sultanate, see the details in Pushpa Prasad, Sanskrit Inscriptions of Delhi Sultanate, 1191–1526, Delhi: Oxford University Press, 1990, and Samira Sheikh, Forging a Region: Sultans, Traders and Pilgrims in Gujarat, 1200–1500, Delhi: Oxford University Press, 2010. (61) Unfortunately, this important subject has not been systematically studied. For very interesting references to guides and interpreters—turjumān—during Khalji campaigns into Ma‘bar, see ‘Isami, Futūḥ al-salāṭīn, pp. 296–7. See also Barani, Ta’rīkh-i Fīrūz Shāhī, Khan (ed.), p. 89, Rashid (ed.), p. 104, for Balban’s campaign against Tughril and the help he solicited from itinerant provenders, the baqqāl, to pin down the location of the fleeing rebel. (62) See, for example, in the reign of Sultan Rukn al-Din (1236); Juzjani, Ṭabaqāti Nāṣirī, vol. 1, 1963–, p. 456. (63) Hodivala (1939), p. 309. See the earlier-mentioned episode of the Bhattis for a reference to his role in the attempted coup against Sultan Firuz Tughluq. (64) Interestingly, this is in contrast to the historiography of subcontinental ‘regions’ where there is considerable interest in the establishment of Muslim societies, especially around networks established by Sufis and other pietistic figures. See Richard M. Eaton, Sufis of Bijapur, 1300–1700: Social Roles of Sufis in Medieval India, Princeton: Princeton University Press, 1978; Eaton, The Rise of Islam and the Bengal Frontier, 1204–1760, Delhi: Oxford University Press, 1993; Carl Ernst, Eternal Garden: Mysticism, History, and Politics at a South Asian Sufi Center, Albany: State University of New York Press, 1992; Jyoti Gulati Balachandran, ‘Texts, Tombs and Memory: The Migration, Settlement and Formation of a Learned Muslim Community in Fifteenth Century Gujarat.’ PhD dissertation, Department of History, University of California at Los Angeles, 2012, and Nile Green, Making Space: Sufis and Settlers in Early Modern India, Delhi: Oxford University Press, 2012. (65) See W.H. Moreland, The Agrarian System of Moslem India, Delhi: Oriental Books Reprint Corporation, 1968 reprint, pp. 216–23, Irfan Habib, ‘An Economic History of the Delhi Sultanate—An Essay in Interpretation’, Indian Historical Review, 4(2), 1978: 287–303, Habib, ‘iqta’, in The Cambridge Economic History of India, vol. 1, c. 1200–1750, Tapan Raychaudhuri and Irfan Habib (eds), Cambridge: Cambridge University Press, 1982b reprint, pp. 68–75. (66) Sunil Kumar, ‘When Slaves Were Nobles: The Shamsi Bandagan in the Early Delhi Sultanate’, Studies in History, 10(1), 1994: 23–52 and Kumar (2007), pp. 167–75. Note also the extensive comment on Sultanate historiography in these publications. (67) For a detailed discussion, see Kumar (2007), pp. 266–86, 295–8. Page 37 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (68) Juzjani (1963–4), vol. 2, p. 61. (69) Juzjani (1963–4), vol. 2, p. 487, and for their machinations in the capital, vol. 2, p. 492 and vol. 2, p. 74. (70) See Juzjani (1963–4), vol. 2, p. 71, and the description of the mawas faced by Ulugh Khan/Balban on his campaign against Qutlugh Khan: ‘… chūn tafarruq bar īshān rāh yāft wa jangalhā-yī Hindustān gashn wa mazā’īq-i lūrhā wa iltifāq-i ashjār-i bisyār …’ (when as their paths were separated [because] of the dense forests of Hindustan and the torrential gorges and the extremely dense woods). Hodivala explained this term to mean ‘a tract or district which was a sort of sanctuary or place of refuge on account of the physical features which made it a natural fastness … [where] … indigenes had retreated ….’ See Hodivala (1939), pp. 226–9. More recently, historians have ignored this meaning and describe the mawas as ‘rebellious areas’. Irfan Habib, ‘Slavery’, in The Cambridge Economic History of India, vol. 1, c. 1200–1750, Tapan Raychaudhuri and Irfan Habib (eds), Cambridge: University Press, 1982c reprint, p. 90. (71) Ibn Battuta, The Reḥla of Ibn Baṭṭūṭa, Mahdi Husain (trans.), Baroda: Oriental Institute, 1976, p. 124. (72) In the last years of his life, Simon Digby went back to study the history of the Afghans. Amongst his unpublished writings on the subject was a detailed history of the Baghela dynasty of modern-day Rewa and their relations with the Sharqis, Lodis, Surs, and the Mughals. He had provisionally titled the paper that I am referring to here as ‘The Kingdom of Bhatta and Muslim India’ (unpublished manuscript-A) and some of the ideas therein are touched upon in his essay included in this volume (Chapter 2), and in Digby (2001) and (2003). (73) Juzjani (1963–4), vol. 2, p. 72. (74) Juzjani (1963–4). On Raja Ribal/Ranbal’s status amongst the local rulers, see Juzjani (1963–4), p. 73. (75) Barani (1860–2), p. 84, Rashid (ed.), p. 98. (76) Juzjani (1963–4), vol. 1, p. 491. (77) This was also Firishta’s sentiment regarding Farhat al-Mulk, the latefourteenth-century Tughluq governor of Gujarat: he ‘became desirous of establishing his independence … and [promoted] rather … than suppressed the worship of idols’. See Sheikh (2010), p. 219n1. (78) Barani harangued the Delhi Sultans about compromising with infidels, his authorial voice speaking through different protagonists. Note, for example, Jalal al-Din Khalji’s ruminations in Ziya al-Din Barani, Sahīfa-i na‘t-i Muḥammadī, Page 38 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī cited by Muzaffar Alam, The Languages of Political Islam c. 1200–1800, Delhi: Permanent Black, 2004, p. 83. (79) On Tughril’s rebellion, see Barani (1860–2), pp. 81–91, Rashid (ed.), pp. 99– 108, and Sirhindi (1931), p. 42. (80) Sirhindi (1931), p. 42. This is an idiomatic translation of the text: Sulṭān mutafakkir shud keh ūlū ‘l-amr-ra ta‘żīm kāfirī wajib nabāshad. (81) Bada’uni (1868), p. 266. Bada’uni was communicating his opinion of conditions in Delhi in 1394, when the different cities of Delhi hosted two Tughluq sultans: Sultan Mahmud and Nasir al-Din Nusrat Shah. (82) For the Baghelas, see Nirodbhusan Roy, Niamatullah’s History of the Afghans, Part 1—Lodi Period, Translated with Variorum Notes, Calcutta: Santiniketan Press, 1958, pp. 195–202; Akhtar Husain Nizami, ‘The Baghela Dynasty of Rewa’, Proceedings of the Indian History Congress, 9, 1946: 242–5, S.H. Askari, ‘Bihar in the Time of the Last Two Lodi Sultans of Delhi’, The Journal of the Bihar Research Society, 41(3), 1955: 357–76, and Digby (unpublished manuscript-A). (83) Mahmud Kirmani, Ma’ās̱ir-i Maḥmūdshāhī, Nurul Hasan Ansari (ed.), Delhi: Jamal Printing Press, 1968, p. 59. The text was completed ca. 872H/1467–68 in the court of the Malwa Sultan Mahmud Khalji (1436–69). For a contrasting complaint from the Sharqis regarding the Kalpi Sultan’s actions against Muslims, see Nizam al-Din Ahmad, Ṭabaqāt-i Akbarī, B. De (trans.), Calcutta: Asiatic Society, 1927–9 (1996 reprint), vol. 3, p. 454. For a full discussion of this campaign, see also Digby (unpublished manuscript-A), pp. 9–10. (84) Rizq Allah Mushtaqi, Wāqi‘āt-i Mushtāqī, Iqtidar Husain Siddiqui and Waqarul Hasan Siddiqi, Rampur: Reza Library, 2002, p. 28. On Mushtaqi’s contrasting usage of chakar and naukar, see the discussion below. (85) See references to the Baghelas above for details. (86) Mushtaqi’s Wāqi‘āt had a brief mention on the Malwa Sultanate and the early Mughals, but this was not a chronicle of the reigns of these dynasts, but anecdotes of curious events that occurred during their rule. (87) Vidyapati, Puruṣa parīkṣā, edited and translated into Maithili by Surendar Jha ‘Suman’, Patna: Maithili Academy, 1988, second edition, lists forty-eight stories while the English translation by George A. Grierson, London: Royal Asiatic Society, 1935, has forty-four. The difference seems to lie in the subdivisions introduced by Surendar Jha that divided some stories into two.

Page 39 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (88) For details on Mithila and Vidyapati, see Pankaj K. Jha, ‘Articulating Power in 15th Century Mithila: Conceptions of Gender, Caste and Other Sources of Authority in Vidyapati’s Compositions’, Nehru Memorial Library, Working Papers, forthcoming A. (89) Pankaj K. Jha, ‘Reading Vidyapati: Language, Literature and Cultural Values in 15th Century North Bihar’, PhD thesis, Department of History, Delhi University, forthcoming B. (90) Vidyapati (1988), p. 30 and (1935), p. 19. I have used Grierson’s translation with a slight emendation. (91) Vidyapati (1988), pp. 218–31 and (1935), pp. 163–70. (92) The classic work of Ronald Inden, Marriage and Rank in Bengali Culture: A History of Caste and Clan in Middle Period Bengal, Berkeley: University of California Press, 1976, is very suggestive on how caste rankings, service, and marriage may have been reconceptualized in the thirteenth century. His study pertains to the Bengal region whose cultural ambits would spill into Mithila. The Puruṣa parīkṣā locates several stories in Gaur and Rarh and has Raja Lakshmana Sena as a protagonist in more than one of them. (93) Vidyapati (1988), pp. 14–21 and (1935), pp. 9–18. (94) Nayachandra Suri, Hammīramahākāvya, English translation by N.J. Kirtane, Bombay: Education Society’s Press, 1879. (95) ‘Isami, Futūḥ al-salāṭīn, Usha (ed.), pp. 271–7, Husain (trans.), vol. 2, pp. 445–51. Note that Barani’s and Amir Khusraw’s brief accounts of ‘Ala’ al-Din’s campaign against Ranthambhor do not draw any attention to Rai Hammir or the episode of sanctuary to Sultanate renegades. See Barani, Ta’rīkh-i Fīrūz Shāhī, Khan (ed.), pp. 272, 276–7, and Amir Khusraw, Khaẓā’in al-futūḥ, Muhammad Wahid Mirza (ed.), Calcutta: Asiatic Society, 1953, pp. 50–4. (96) Space forbids a consideration of Vidyapati’s Likhanāvalī, Pankaj K. Jha (trans.), forthcoming. This is a writing manual of model letters and documents that administrators of different ranks could follow. The closest parallel to this text would be within the insha tradition but it differs from this genre in that it contains only model letters and no original documents. (97) For an early history of kindred groups in the Delhi Sultanate, see Sunil Kumar, ‘Mooring the Household: Problems in Writing a History of Early Sultanate Elites’, in Looking Within/Looking Without: Pre-colonial Households, Kumkum Roy and Nandita Sahai (eds) (forthcoming B). (98) Green (2012), p. 70. There is considerable recent work on the Lodis and the Surs and much of it is now conveniently referenced in Green (2012). Page 40 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (99) Mushtaqi (2002), pp. 5–6. (100) Mushtaqi (2002), p. 6. The term certainly had a long history of usage in Rajasthani history, where its meanings in the seventeenth and eighteenth centuries differed from those implied by Mushtaqi and other Afghan chroniclers. For the Rajput Marwari usage of the term, see Norman Ziegler, ‘Action, Power and Service in Rajasthani Culture: Social History of the Rajputs of Middle Period Rajasthan’, PhD dissertation, University of Chicago, 1973, and Ziegler, ‘Rajput Loyalties during the Mughal Period’, in Kingship and Loyalty in South Asia, J.F. Richards (ed.), Delhi: Oxford University Press, 1998 reprint, pp. 256–8, 268. (101) For an extremely useful reading of Mushtaqi, see Ali Anooshahr, ‘Author of One’s Fate: Fatalism and Agency in Indo-Persian Histories’, Indian Economic and Social History Review, 49(2), 2012: 197–224. And for a general review of sources on the Lodis, see Digby (2001). (102) While most Afghan sources gloss over the details of the succession, they do mention the support that Sikandar received from the Farmulis at his accession. The details of the dispute over the succession are, however, carried in the seventeenth-century text of Muhammad Qasim Firishta, Gulshan-i Ibrāhīmī, John Briggs (trans.), Delhi: Low Price Publications, 1990 reprint, vol. 1, pp. 328–9. This incident is narrated with slight variations in the different manuscripts and editions of Firishta. The reference to the Panchatantra, for example, is present in Briggs and Digby (2001), pp. 255–8 which I have used. It is absent in the Ta’rīkhi Firishta, edited by Muhammad Riza Nasiri, Teheran: Anjuman-i Asar wa Mufakhir-i Farhangi, 1967, vol. 1, pp. 590–1. I am grateful to Ali Anooshahr for supplying a copy of this passage. (103) Mushtaqi (2002), p. 8. (104) For details, see ‘Abbas Khan Sarwani, Ta’rīkh-i Sher Shāhī, S.M. Imamuddin (ed.), Dacca: University of Dacca, 1964, vol. 1, pp. 51, 103, 107. (105) Sarwani (1964), pp. 51, 107, 215–16. (106) Mushtaqi (2002), pp. 146–53. (107) Mushtaqi (2002), p. 13. (108) Mushtaqi (2002), p. 14. The ambiguity in the sentence should not be missed. In a text that is laced with miraculous happenings and a proximate mystical sense of God’s presence, this sentiment stands as a recognition of the worthlessness of transient human life. Read more prosaically and materially, it reaffirmed the worthlessness of a slave who was but a commodity. (109) Mushtaqi (2002), p. 14.

Page 41 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Bandagī and Naukarī (110) For references to Malik Suman see Mushtaqi (2002), p. 14. (111) Sarwani (1964), pp. 166–7. (112) This is well brought out in Amir Khwurd’s Siyar al-awliyā’. The foundational work for a careful historicization of the spread of Sufis hospices, ideas, and institutional organization in the fourteenth and fifteenth centuries has been done by Simon Digby, ‘The Sufi Shaikh as a Source of Authority in Medieval India’, Puruṣārtha (Islam and Society in South Asia) 9, 1986: 57–77, Digby (2004), Digby (1965), Digby (1975). (113) Amir Hasan Sijzi, Fawā’id al-fu’ād, Khwaja Hasan Thani Nizami Dihlawi (ed.), Delhi: Urdu Academy, 1990, p. 1. (114) Barani (1972), p. 333. (115) Mushtaqi (2002), pp. 10, 20. (116) Sarwani (1964), vol. 1, p. 205, and vol. 2, Imamuddin (trans.), p. 158. I have followed Imamuddin’s translation. (117) Anooshahr (2012). (118) Anooshahr (2012), p. 214, Mushtaqi (2002), p. 147.

Access brought to you by:

Page 42 of 42

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

The Rise of Written Vernaculars The Deccan, 1450–1650 Richard Eaton

DOI:10.1093/acprof:oso/9780199450664.003.0004

Abstract and Keywords This chapter traces processes of vernacularization that can be extended to north India as well. It sees the emergence of vernacular languages, in this case Marathi, as emerging through significant linguistic and legal interventions. These include the naming of the language or its region, the use of the language to record legal transactions or conflicts, and the use of the language for poetry or prose. While the latter intervention is the one usually considered in histories of literary languages, the chapter argues that legal texts in the vernaculars— thousands of which are found in Marathi—helped forge the discursive community of Marathi speakers. This process was helped along by the state’s institutional and conflict solving processes as well as the increased availability, from the fourteenth century onwards, of paper. Keywords:   Deccan, vernaculars, Persian, Bengal, Marathi, Dakani/Urdu, Bahmanis, Bijapur, paper

SCHOLARS HAVE RECENTLY SHOWN MUCH interest in the origins of vernacular literatures in South Asian history, especially since Sheldon Pollock’s formulation of the notion of a ‘Sanskrit cosmopolis’, spanning much of southern Asia from roughly 300 to 1300, followed by what he calls a ‘vernacular millennium’.1 Such a formulation raises issues central to the concerns of social historians. How can one explain the transition from Pollock’s ‘Sanskrit cosmopolis’ to the ‘vernacular millennium’? What mechanisms were at work? Was the transition initiated ‘from above’, as a result of patronage by Indian courts? Did it arise ‘from below’, as a result of religious poets addressing the Page 1 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars common people in already existing spoken vernaculars? Or was something entirely different going on? It makes a great deal of difference whether a given text is, for example, a stone (p.112) inscription rich in factual information, an expository treatise read by a tiny circle of courtly elites, or a religious tract intended for wide consumption. This is because what numismatists would call the ‘velocity’ of a text’s circulation—that is, the number of hands through which it passes—can indicate the presence of a self-aware, integrated speech community. Such circulation can also contribute to the making of that community, in the same way that the circulation of common coinage can contribute to the making of a self-conscious political community. In north India, the splintering of the sprawling Delhi sultanate in the early fifteenth century, catalyzed by Timur’s 1398 sack of Delhi, had long-term linguistic consequences. Ever since the eleventh century, modern Persian had been the prestige language of power in the eastern Islamic world, even though most rulers in that area were ethnic Turks. And when Timur’s descendant Babur launched the Mughal empire in the early sixteenth century, Persian returned as the language of imperial power. Muzaffar Alam has argued that the Mughals had no choice but to continue the legacy of patronizing official Persian, and not just because of that language’s great prestige, or because of Akbar’s unusual interest in promoting socio-cultural contacts with Iran. More importantly, the Mughals aspired to evolve a political culture that could arch over the diverse ethnic identities of their millions of subjects.2 Persian emerged as the ideal imperial language precisely because it was native to no single region in India and could therefore preside supremely and with equanimity over the empire’s ethnically diverse subregions. The Delhi sultans had made the same linguistic choice, for largely the same reason. By this logic, the weakening of the Delhi sultanate in the fifteenth century—and also the Mughal empire in the eighteenth —posed a challenge to the hegemony of Persian while making possible the emergence of regional vernacular languages and literatures patronized by new, but smaller, regional states. Such a pattern is seen in fifteenth-century Bengal, where by the mid-fourteenth century, some fifty years before Timur’s invasion of north India, a dynasty of regional sultans had already asserted its independence from Delhi. These rulers thus had a head start in abandoning imperial Persian and in merging their identity with that of their subjects in the (p.113) Bengal delta. Having travelled through the interior of Bengal in 1406, the Chinese traveller Ma Huan noted that vernacular Bengali was ‘the language in universal use’ at the court of the sultans in Gaur.3 By the second half of the fifteenth century, that court was patronizing the production of a range of religious works in Bengali, including translations from the Sanskrit such as the Mahābhārata and Rāmāyaṇa epics, and also popular Vaishnava devotional works such as Maladhara Basu’s Śrī Kṛṣṇa-Vijaya or Yasoraj Khan’s Kṛṣṇa–maṅgala.4 There are several reasons the Bengal sultanate sought to indigenize its cultural identity, reflected in its coinage and Page 2 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars architecture as much as in its literary patronage. For one thing, violent rebellions led by some of the kingdom’s elite Hindu officers between 1410 and 1417 induced the sultans to reach over the heads of those officers and seek support among commoners. For another, patronizing Persian was ruled out for both political and logistical reasons. Politically, Persian was too closely identified with the Bengal rulers’ former overlords in Delhi, from whom they strenuously endeavoured to distance themselves. At the same time, Bengal’s prolonged political and commercial isolation from north India, not to say from the Islamic lands further west, deprived the delta of a regular infusion of Persian or Persianized immigrants who might otherwise have preserved the region’s tenuous links to the prestige language of the eastern Islamic world. As a result, the Bengal sultans of the fifteenth and sixteenth centuries fused vernacular language and literature with their sovereign domain, effectively transforming the lower Ganges delta into a single ethno-political subregion.5 Not all Bengali literature, however, was initiated by the royal court at Gaur. Religious texts such as Krishnadas’s Caitānya caritāmṛta (ca. 1600), the most important hagiography of the Bhakti poet Chaitanya, were composed far from the court’s orbit of patronage. Yet the socio-linguistic consequences of the circulation of such works were just as (p.114) profound as were those of texts patronized at the royal court. As Sudipta Kaviraj observes, the composition of the Caitānya caritāmṛta—its dual role as biography and treatise, its incorporation of both Sanskrit and Bengali, high and low— helps us understand what medieval authors were attempting to achieve by writing in Bangla. Every time a religious movement had to widen its circle of followers, it had recourse to this linguistic technique. Thus, the historical process by which Bengalis became a people in a linguistic sense must be related to these periodic extensions, these successive ‘democratizing’ movements of religious ideas.6 In short, the emergence of a self-aware Bengali speech community in the fifteenth and sixteenth centuries was associated with initiatives emanating both from and beyond the royal court. What patterns are found if we look south, to the Deccan plateau? Already by the eleventh century, states in the western Deccan had been using Marathi to record official transactions. By the late thirteenth century the court of the Yadava kings (1175–1313) was producing Marathi political treatises that betrayed, at least among their courtly patrons, a consciousness of a Marathi-speaking territory, even of Marathi political territory.7 Marathi devotional literature also appeared in the late thirteenth century, composed by members of two religious communities in Maharashtra, the Mahanubhavas and the Varkaris. The founder of the former, Chakradhara, not only wrote in Marathi; he advised his followers to stay in Maharashtra and to avoid going to the Kannada or Telugu regions. And Page 3 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars the first great Varkari poet, Jnanesvar, boasted that ‘although my language may be Marathi, it will surpass nectar’.8 All this points to a popular identification of vernacular speech with a particular region, and also with that region’s people. When the seventeenth-century Bhakti poet Tukaram wrote of the interface between deity and devotee, he did not emphasize the encounter between the individual and God, but rather, the encounter between God and the (p.115) collective, the community of devotees—more precisely, the community of Marathi-speaking devotees.9 The poet was thus giving voice to a deep-rooted collective identity among Marathi-speakers, and particularly among nonBrahmins of Maharashtra. Dilip Chitre has gone further and suggested that Tukaram and other Varkari poets ‘helped to bind the Marathas together against the Mughals on the basis not of any religious ideology but of a territorial cultural identity’.10 Is it that simple? Can we indeed attribute the emergence of Marathi vernacular literature—indeed, of a politically mobilized territorial cultural identity, as Chitre would have it—to a four centuries’ tradition of poet-saints writing Marathi poems in praise of the immensely popular deity Vithoba? To evaluate the merit of such an argument, we need to situate the emergence of written Marathi in the broader context of language use in the Deccan as a whole. In particular, we should examine the division of labour respecting written materials during the period when most of the Deccan was ruled by the Bahmani sultans (1347–1518) and that dynasty’s successor states, notably Ahmadnagar, Bijapur, and Golkonda. All these rulers affirmed their claims to sovereign authority by issuing coins and inscriptions recorded not in any Indian vernacular, but in the classical languages of the Perso-Islamic world. Coins were inscribed mainly in Arabic, the sacred language of the Islamic world community, following formulaic conventions inherited (p.116) from centuries of Islamic numismatic tradition. Inscriptions on mosques or Muslim shrines also normally appeared in Arabic. On the other hand, inscriptions placed on non-religious monuments, being more informative than formulaic, were generally recorded in the prestige court language, Persian. Significantly, the epigraphic evidence left by the sultanates does not reveal the sort of linguistic bifurcation that was typical of earlier dynasties of the Deccan, in which an inscription’s expressive or eulogistic portion (praśasti) typically appeared in Sanskrit, the cosmopolitan ‘language of the gods’, while the operative, or ‘business’ portion appeared in the vernacular. Rather, most of the Deccan sultanates’ inscriptions were composed entirely in Persian; only a handful were recorded in Persian and one of the local vernaculars.11 One might expect that the Bahmanis, being descended from rebels against the hegemonic rule of the Delhi sultanate, would have invested themselves in the vernacular cultures of their subject population, as did their contemporaries in Bengal. But unlike the Bengal sultanate, the territory the Bahmanis ruled over comprised speakers of not one but three major vernaculars—Marathi, Kannada, and Telugu. Unable to identify their regime with any single regional language, Page 4 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars the Bahmanis had two options. Either they could retain Persian as a transregional official language as did their parent state, the Delhi sultanate. Or they could patronize Dakani, the vernacular spoken by that section of the ruling class—the so-called ‘Deccanis’—that was descended from north Indian Muslim immigrants who had colonized the upper Deccan under Tughluq authority in the early fourteenth century, and who in the middle of that century had thrown off their allegiance to Delhi. In fact the Bahmanis, and more prominently their two principal successor states, Bijapur and Golkonda, chose both options. In terms of its social base and reach, Dakani can be compared with Urdu in north India. It was the vernacular spoken mainly by Deccan-born Muslims who clustered around regional urban centres associated (p.117) with the administration of the Bahmanis and their successor states.12 When Dakani literature first appeared in the fifteenth century, it was initially called ‘Hindavi’ or ‘Hindi’, owing to its remembered association with the Delhi sultanate’s colonial connection with the Deccan between 1327 and 1347, when thousands of migrants had moved down from the North, or ‘Hind’. The language’s high level of Panjabi vocabulary alone points to this northern connection. By the seventeenth century, however, authors had begun referring to the language as ‘Dakani’, a term reflecting the new point of geographical reference, and the new spirit of cultural independence, of the language’s native speakers. The deep social antagonism between native Deccanis and foreign-born immigrants from the Middle East found its linguistic and literary counterpart in the appearance, and also the patronage, of Dakani literature. By the sixteenth and especially the seventeenth century, Dakani poets were confidently composing poetry in their own vernacular without seeking ratification from outside authority. As Shamsur Rahman Faruqi remarked, they felt no need to ‘genuflect before the ancients, Sanskrit and Perso-Arabic’. Writing in the 1640s, the poet San‘ati Bijapuri summed it up: ‘Dakhani comes easy to one who doesn’t have Persian. For it has the content of Sanskrit, but with a flavour of ease.’13 Dakani poets like San‘ati or Nusrati Bijapuri were supported by enlightened royal patrons like Sultan Muhammad Quli Qutb Shah (1580–1611) of Golkonda and Sultan Ibrahim ‘Adil Shah II (1580–1618) of Bijapur, both of whom invested themselves deeply in classical Indian and local cultures. They also wrote Dakani verse themselves. (p.118) Sultan Ibrahim’s outstanding literary achievement was his Kitāb-i nauras, an essay on Indian aesthetics set to prescribed musical ragas. Sultan Muhammad, like other Dakani poets, even adopted the Indian trope of an author speaking in the voice of a woman awaiting the favours of her male lover, as in the love of Radha for Krishna.14 Beyond the court, Sufi shrines also emerged as centres of Dakani literary production. Like many such shrines elsewhere in South Asia, those in the Deccan sought to preserve the remembered sayings and deeds of spiritually powerful shaykhs, which prompted disciples to record the biographies (tazkirat) and conversations (malfuzat) of such shaykhs. In addition, Sufis themselves wrote works on mystical subjects.15 Page 5 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars Prominent genres of this non-courtly Dakani literature included the chakki-nama or charkha-nama, which, sung by non-elite women while engaged in domestic chores like grinding meal or spinning thread, used simple language and imagery to transpose abstract ideas about the Sufi’s path to spiritual enlightenment into every-day guidelines for a pious life.16 However, since Dakani was not widely spoken beyond predominantly Muslim urban areas, it could not have served as a vehicle for proselytizing Islam among the masses of Kannada-, Telugu-, or Marathi-speakers. Indeed, as Sumit Guha has suggested, it seems likely that Dakani was the only vernacular that Sufis and their immediate socio-religious circles knew.17 If Dakani was cultivated by the sultanates’ ruling class for aesthetic and expressive reasons, and by Sufi centres and their networks for devotional purposes, official patronage of Persian did not diminish in the Deccan as it did in Bengal. For unlike Bengal, the Deccan plateau—especially the western plateau, with its access to the Arabian Sea—enjoyed close maritime contact with the Persian Gulf, which permitted the Bahmanis to recruit Iranian literati and military men into their (p.119) ruling class. The steady influx of so-called ‘Westerners’ (Pers. gharbiān) between the fourteenth and seventeenth centuries thus kept alive the use and patronage of Persian as the state’s language of power. The influx also effectively split the Bahmani ruling class between a Dakani-speaking old guard descended from original colonizers from north India, and newcomers recruited straight from the Persian-speaking world. Indeed, this division precipitated the ultimate demise of the Bahmani state itself, as fierce rivalries between Deccanis and Westerners periodically degenerated into open civil war. Even the break-up of the Bahmani state in the early sixteenth century did not end these internal tensions, which were simply inherited by each of the Bahmanis’ several successor states—the most prominent being the Nizam Shahi kingdom of Ahmadnagar, the ‘Adil Shahi kings of Bijapur, and the Qutb Shahi kings of Golkonda. The fragmentation of the large, transregional Bahmani kingdom into several successor-states had important linguistic consequences. First, the borders of the easternmost successor-state, Golkonda, perfectly coincided with Telangana and the Andhra coast, which was predominantly Telugu-speaking. This made it relatively easy for the Qutb Shahi sultans of Golkonda to conceive of their political realm as a Telugu realm. Consequently, they vigorously patronized the production of Telugu literature, just as Bengali sultans patronized the production of Bengali literature. Indeed, K. Lakshmi Ranjanam described the Qutb Shahi kings as ‘virtually … Telugu Sultans’.18 Sultan Ibrahim Qutb Shah (1550–80), known in Telugu sources as ‘Ibharāma cakravartī’, was so thoroughly steeped in Telugu aesthetics that he would sit, ‘floating on waves of bliss’, as one court poet put it, listening to the Mahābhārata recited to him not in Sanskrit, but in the Telugu translation begun in the eleventh century by the poet Nannaya.19 Not only did Ibrahim himself patronize works of Telugu poetry, so did his Muslim Page 6 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars noble Amin Khan, and more importantly, his son and successor, Muhammad Quli Qutb Shah (1580–1612). In doing so, these men were following a time-honoured (p.120) tradition in Andhra, where local elites had patronized the production of Telugu poetry ever since the time of the Kakatiya monarchs in the mid-twelfth to the early fourteenth century.20 On the other hand, the ‘Adil Shahi sultans of Bijapur, whose sovereign domain embraced a Kannada-speaking south and a Marathi-speaking west, are not known to have patronized literature in either language. Nor is there evidence that the Nizam Shahi sultans of Ahmadnagar patronized vernacular literature, despite the fact that their territorial domain was completely nested within the Marathi-speaking western Deccan, as Golkonda was within the Telugu-speaking eastern Deccan. The likely explanation for this is geographical. Having ready access to the Arabian Sea and Persian Gulf through ports on the Konkan coast, in particular Chaul, the sultans of Ahmadnagar received a steady stream of Iranian nobles and Persian literati who monopolized literary patronage under the Nizam Shahi sultans.21 Nonetheless, by the seventeenth century peoples of the Marathi-speaking plateau, or Desh, had become bound together, as Chitre noted, ‘on the basis not of any religious ideology but of a territorial cultural identity’.22 In explaining how this happened, it seems to me that two factors—one administrative, one technological—might be taken into account. New governing institutions and new technologies of communication introduced by the ‘Adil Shahi rulers of Bijapur had the effect of bringing Marathi-speaking peoples of all classes into closer communion with the state and with one another through the medium of written Marathi. Even though these rulers did not patronize Marathi literature, they made the crucial decision to conduct the routine business of government at the local level—collecting revenue and administering (p.121) justice—in written Marathi, not Dakani or Persian. Moreover, it was conducted through the revolutionary medium of paper, which greatly increased the velocity of movement of records through the hands of various levels of local functionaries.23 This move to the bureaucratic vernacular was triggered in part by the discovery in both Golkonda and Bijapur that the most efficient way of managing the judicial and revenue branches of government was to employ those same classes of skilled, literate administrators—mostly Brahmins—that for centuries had served ruling authorities of the Deccan. In Golkonda, the growing importance of this class of clerks is seen in that kingdom’s increasing reliance on vernacular, bureaucratic Telugu in its administration. In the sixteenth century, Qutb Shahi royal edicts had been issued from Golkonda in Persian only. By the early seventeenth century, they were often bilingual, in Persian and Telugu. And by the end of the seventeenth century, such edicts were issued entirely in Telugu, with brief Persian summaries appearing only on their reverse sides. At local

Page 7 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars levels, meanwhile, revenue papers were prepared largely in Telugu by Niyogi Brahmins expert in clerical and administrative skills.24 (p.122) In Bijapur, the switch to the use of Marathi in the local government came abruptly in 1535 with the accession of Ibrahim ‘Adil Shah I, who appears to have made this change as much for political as for pragmatic reasons. As a member of Bijapur’s Deccani class, Ibrahim sought to overturn the pro-Persian, pro-Iranian ideological orientation of his Westerner predecessors. Accordingly, he initiated a top-down political-cultural revolution that aimed at indigenizing his regime. At court he replaced Shi‘i rites with Sunni rites, since most Deccanis were descended from Sunni immigrants from north India, whereas most Westerners were Shi‘i. He forbade the wearing of hats in the style of the Safavid Iranian court. He expelled from his nobility all but 400 foreign-born immigrants, replacing them with native-born Deccanis and slaves from East Africa. And, most relevant for our purposes, he ordered that all public revenue accounts, formerly recorded in Persian, be kept in Marathi and placed under Brahmin management.25 The state’s judicial records were also maintained in vernacular languages. In the early 1960s Hiroshi Fukazawa studied a large collection of ‘Adil Shahi local records that were published by the Bharat Itihas Sanshodhak Mandal in Pune. Of these records, which date to the sixteenth and seventeenth centuries, only farmans, or royal decrees issued from the court, were written in Persian. All other documents were in Marathi. These included khurdkhats (orders issued by crown bureaucrats or ‘fief-holders’), misalis (letters issued by mid-level bureaucrats or police chiefs), qaul-namas (letters of assurance issued by any higher authority, from the sultan down to a desai), and mahzars (decisions issued by judicial assemblies).26 Writing in the early 1600s, Bijapur’s court historian Muhammad Qasim Firishta noted that one consequence of this sweeping policy was (p.123) that it gave Brahmins enormous influence in governmental affairs.27 While this was no doubt true, the policy seems to have had an even more significant consequence that Firishta would not have noticed. Inasmuch as the litigation of disputes in nearly any literate society tends to churn up evidence and leave ‘paper trails’, a close look at contemporary court testimony reveals a great deal about a government’s methods of resolving conflicts. It also furnishes a worm’s-eye view of social relations at the lowest level, pointing to one of the central concerns of the present volume, namely the consolidation of self-conscious language communities defined by region. It is true that pre-Bahmani courts such as the Rashtrakutas, the Kalyana Chalukyas, and the Yadavas had patronized Marathi and Kannada inscriptions and expository treatises.28 Given the technology of communication in those eras, however, such texts would have had little circulation beyond elite groups. But in the early sixteenth century, the deployment of these same vernaculars at the grass-roots level for the purpose of carrying out Bijapur’s revenue and judicial functions, and doing so through the

Page 8 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars revolutionary medium of paper, marked an important breakthrough in the progressive vernacularization of the Deccan plateau. We can see something of this breakthrough by taking a close look at a judicial case litigated in 1611 in the heart of the Desh, or Marathi-speaking portion of the plateau governed by Bijapur.29 This particular case involved a dispute between two kin-groups: a family of hereditary village headmen (patil), represented in court by a plaintiff (p.124) named Narasoji; and a family of Muslim merchants and moneylenders represented by a defendant named Bapaji. It also involved the shrine of a local Sufi saint, Pir Jalal al-Din, and its own family of hereditary attendants (mujawirs). The dispute began over the use of offerings made to this shrine, it evolved into a murder case, and it ended in a bitter struggle over which family—that of Narasoji or that of Bapaji—should rightfully hold the office of headman of a village in modern Satara District. Each stage in the resolution of the dispute, as it was litigated across rural Maharashtra, sheds light on social, political, and communal relations at the village level, as well as on villagers’ relations with the wider world. But most importantly, the case reveals the evolution of a community that defined itself not by caste, not by class, and not by religion, but by a shared language, a shared history, and a shared conception of civil society. The dispute began sometime in the late 1500s over the disposition of offerings made to the tomb of one Pir Jalal al-Din, whose shrine in Satara District was venerated by both Muslims and Hindus. By custom, Hindus placed offerings for Hindu mendicants on one side of the tomb, and Muslims put offerings for Muslim holy men on the other. But at some point, local Muslims persuaded servants of the shrine to discontinue the practice, evidently in a manner that would channel all the offerings to themselves. When the village headman ordered a return to the earlier practice, he and one of his sons were murdered by servants of the shrine. This act was avenged when another of the headman’s sons, on succeeding to his father’s post, murdered three servants of the shrine. What is fascinating is how quickly the government of Sultan Ibrahim ‘Adil Shah II involved itself in the case. This happened when a prosperous trader and moneylender in the village, who was related to one of the shrine’s murder victims, sought justice by going to the district headquarters. There he had an interview with the region’s deputy governor (na’ib), ‘Abd Allah Husayn, who governed the district on behalf of the ‘Adil Shahi government, and who happened to be related to the moneylender by marriage. The upshot of their meeting was that the governor issued a warrant against the new village headman. When the latter failed to answer it, ‘Abd Allah charged the headman with murdering the three shrine attendants, fined him 4,500 gold coins (huns), and ordered him kept in custody for three years. The man’s home was also looted, with the governor’s apparent connivance.

Page 9 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars (p.125) The headman paid part of the fine imposed on him, but then absconded without paying the balance. When deputy governor ‘Abd Allah Husayn heard about this—and here we reach the crux of the matter—he transferred the rights of village headman from the absconding murderer to the same moneylender who had brought the complaint to his attention, and who was a relation of his. But since this man was after all a moneylender, he had enemies. In fact, indebted villagers joined forces and managed to drive him out of town altogether. The office of headman now passed to the moneylender’s son, Bapaji. At this point Narasoji, the son of the former headman who had murdered the servants of the shrine and later absconded, appealed directly to Sultan Ibrahim ‘Adil Shah II for justice. Specifically, he sought to have the office restored to his own family, that is, to himself. To resolve the dispute, the sultan assigned the case to one ‘Anbar Khan, a state-appointed judge, who ordered Narasoji to pay the unpaid portion of the fine that had been levied on his father, after which he, Narasoji, would receive the office of village headman. A council of arbiters in Satara District also upheld the judgement that the office of headman should be transferred from Bapaji to Narasoji. But Bapaji, complaining that these arbiters were biased against him, arranged to have the case referred for final arbitration to a dharma-sabha, or an assembly of learned Brahmins, in distant Paithan, some 200 miles north of the village in question. This request was made most likely some time in 1610. Bapaji had requested the change of venue since the dharma-sabha of Paithan was especially renowned for its fairness. Upon reaching that centre of Brahmin learning, the contending parties were taken to a holy place (dharma-cauthara) intended for such judicial assemblies. There Narasoji and Bapaji presented their cases before a council of Brahmin judges (dharmadhikaris), learned in the dharma-shastras. The judges first scrutinized all the documents sent up from earlier litigation. These included the orders of deputy governor ‘Abd Allah Husayn, the ruling of ‘Anbar Khan, and the findings of the district arbiters, together with some royal edicts related to the case. Plainly, several levels of judicial review and appeal were operating within the sovereign domains of the ‘Adil Shahi state. Then the trial began. Since it was not the custom that plaintiffs or defendants have their cases argued by advocates, Bapaji and Narasoji had to plead their own cases before the council of Brahmins. The high point of the trial came when the defendant Bapaji (p.126) called on the plaintiff Narasoji to perform an ordeal. ‘If you succeed in the ordeal’, said Bapaji, ‘you owe me nothing and I owe you nothing.’ At this point, the judges cited the Mitākṣara, a Sanskrit legal text written by the eleventh-century jurist Vijnanesvara, to the effect that anyone who demanded an ordeal when reliable testimony was available should be executed. Addressing Bapaji, the panel of Brahmin judges said,

Page 10 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars We conclude that you have caused much trouble to Narasoji and his family, and that you should be executed. But since you are Muslim, and the present government [of Bijapur] is a Muslim government, you will be pardoned. However, we find that you have no valid claim to the position of village headman. And you have no right to recover money owed by your father’s debtors. Nor do you have any right to receive compensation for the murder of your relatives. On 14 January 1611, a mahzar, that is, a legal document attested by witnesses, was drawn up in Marathi, duly signed, and handed to Narasoji. Pronouncing that Bapaji had no valid claims to the village headship, the document finally transferred the office to the plaintiff, Narasoji. What can we make of this case? In the first place, the case reveals the depth to which the authority of the Bijapur sultan extended in the Deccan’s rural society. Even a lowly village headman—or in this case, a man who made claim to the office—could lodge a complaint directly with the supreme sovereign of the realm. And to this appeal the sultan did indeed respond, through his representative ‘Anbar Khan, whose ruling would have restored the headship to the plaintiff. All of this challenges the Orientalist stereotype of Indo-Muslim sultanates as remote, lofty superstructures that sat uneasily atop autonomous and self-governing village communities. Clearly, the state penetrated the very lowest levels of Deccani society. Second, in the end, it was a council of learned Brahmins, the dharma-sabha, that, at the request of the Muslim defendant, was called upon to adjudicate the matter, and whose judgement was deemed to be final. This occurred only after the dispute had been heard and ruled upon by several different authorities: the sultan’s deputy governor (na’ib), the sultan’s personal representative, and the district council (pargana sabha). It is also significant that the dharma-sabha of Paithan settled the case on the basis of Hindu dharma-shastras, to which all parties including the Muslim Bapaji consented, and more specifically, on the basis of a law code that dated to the time of the imperial Kalyana Chalukyas, who (p.127) hadn’t ruled over this region since the twelfth century. The only part of the council’s judgement on which it did not have jurisdiction was the matter of capital punishment. Otherwise, the judges’ finding on the central issue of the village headship served to confirm the judgements of all the earlier levels of adjudication. In effect, the offices of the deputy governor and district council functioned as de facto lower courts, inasmuch as their decisions could be appealed to the council of Brahmins whose judgements were considered final on all matters excepting capital punishment for a Muslim. Finally, one must consider the long-term socio-cultural effect of thousands of documented cases, such as that of Narasoji and Bapaji, being argued across the Marathi-speaking domains of the Bijapur sultanate. Involved here were repeated Page 11 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars face-to-face disputations, conducted in vernacular Marathi, in which any single case might be taken from court to court, from locale to locale across the Desh, and from level to level along the socio-political spectrum—in each instance involving diverse peoples of various castes, classes, and religions.30 The very process of engaging in public disputation served to connect villagers with each other, even when they were gripped in bitter contention, and to reinforce a consensus on how conflicts should be resolved. Importantly, this process had the deeper effect of forging a larger, discursive community around vernacular Marathi both as spoken in the courts and as recorded by them.31

(p.128) Conclusion In the final analysis, vernacularization may be understood as a gradual process along a timeline punctuated by a series of significant interventions—for example, the first appearance of the name of a vernacular tongue or the region in which it is spoken, a state’s first use of the vernacular to record some official transaction on a copperplate or stone slab, the first use of the vernacular language in expository prose or creative poetry, and the first use of a written vernacular to record face-to-face conflict resolution among members of non-elite groups. In forging a discursive community that cut across the entire gamut of village society, the last stage constituted an important breakthrough. This would suggest that Maharashtra’s four centuries’ tradition of Varkari devotional poetry, about which so much has been written, was not the only force that helped shape a community of self-conscious Marathi speakers, or that promoted writing in that language. Thousands of litigated cases like that of Narasoji and Bapaji had also contributed to those outcomes. Two factors, one political and the other technological, had served to create the public space within which such disputations could legitimately occur and conflicts be adjudicated. One was the state, which had built the institutional frameworks for the resolution of conflicts in the people’s own language. The other was the technology of paper-making, which the Bahmani sultans had introduced to the Deccan in the fourteenth century and which diffused throughout the region in the succeeding centuries. The new technology greatly facilitated the growth and reach of the sultanates’ revenue and judicial bureaucracies, while at the same time enhancing the social prominence of those Brahmin writing communities that staffed them. Indeed, the ordinary words for ‘paper’ and ‘pen’ in the vernacular languages of the Deccan—in fact, throughout South Asia—are derived from two Perso-Arabic words, kaghaz and qalam. This suggests not only the new technology’s path of transmission from Central Asia to the Deccan, but also the profound impact that these exogenous technologies of writing had had on the peoples of the plateau. To conclude, after Timur came—that is, in the fifteenth, sixteenth, and seventeenth centuries—new public arenas emerged when state-created revenue and judicial bureaucracies extended their reach into village society. In Bijapur Page 12 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars and Golkonda, they did this using the (p.129) vernacular languages, mediated and facilitated by the diffusion of new technologies of writing—the paper and pen. Conducted and recorded in the vernaculars, the myriad face-to-face encounters that occurred in such arenas slowly but inexorably knit together new discursive communities. Such communities could, in turn, be transformed either into ethno-religious solidarities such as the Varkari movement based on the pilgrimage centre at Pandharpur, or into ethno-political solidarities such as the seventeenth-century Maratha kingdom. Thanks to a process rooted in the judicial and revenue institutions of this Indo-Muslim state, the groundwork for a gifted poet like Tukaram—or a gifted politician like Shivaji—had already been established. Notes:

(1) Sheldon Pollock, ‘India in the Vernacular Millennium: Literary Culture and Polity, 1000–1500’, Daedalus 127(3), 1998: 41–74; ‘The Cosmopolitan Vernacular’, Journal of Asian Studies, 57(1), 1998: 6–37; and The Language of the Gods in the World of Men: Sanskrit, Culture, and Power in Premodern India, Berkeley: University of California Press, 2006. (2) Muzaffar Alam, The Languages of Political Islam: India 1200–1800, Delhi: Permanent Black, 2004, p. 134. (3) W.W. Rockhill, ‘Notes on the Relations of Trade of China with the Eastern Archipelago and the Coast of the Indian Ocean during the Fourteenth Century’, T’oung Pao, 16(2), 1915: 43. (4) Niharranjan Ray, ‘Mediaeval Bengali Culture’, Visva-Bharati Quarterly, 11(2), Aug.–Oct. 1945: 54. (5) Richard M. Eaton, The Rise of Islam and the Bengal Frontier, 1204–1760, Berkeley: University of California Press, 1993, pp. 63–70. (6) Sudipta Kaviraj, ‘The Two Histories of Literary Culture in Bengal’, in Literary Cultures in History: Reconstructions from South Asia, S. Pollock (ed.), Berkeley: University of California Press, 2003a, p. 520. (7) Pollock (2006), pp. 289–90. (8) Pollock (2006), pp. 382–3. (9) Note Tukaram’s use of plural pronouns: God ignores our hierarchies, Showing loyalty only to devotion. They’ve organized a game on the river sands, The Vaishnavas are dancing, ho! Pride and wrath they trample underneath their feet, Page 13 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars At one another’s feet they’re falling, ho! We servants of Vishnu are softer than wax, But hard enough to shatter a thunderbolt. Dead but alive, awake while asleep, Whatever is asked of us we go on giving.

Gail Omvedt and Bharat Patankar (trans.), The Songs of Tukoba, Delhi: Manohar, 2012, abhangs nos 2820, 189, 987. (10) Dilip Chitre, Says Tuka: Selected Poems of Tukaram, New Delhi: Penguin Books, 1991, p. xvii. (11) See Stan Goron and J.P. Goenka, The Coins of the Indian Sultanates, New Delhi: Munshiram Manoharlal, 2001, pp. 312–42; Z.A. Desai, A Topographical List of Arabic, Persian and Urdu Inscriptions of South India, New Delhi: Indian Council of Historical Research, 1989; Z.A. Desai, Arabic, Persian and Urdu Inscriptions of West India, a Topographical List, New Delhi: Sundeep Prakashan, 1999. (12) See Sumit Guha, ‘Transitions and Translations: Regional Power and Vernacular Identity in the Dakhan, 1500–1800’, Comparative Studies of South Asia, Africa, and the Middle East 24(2), 2004: 25–6. (13) It is notable that the poet situates Dakani with respect to the two transregional, cosmopolitan traditions of his day—Persian and Sanskrit—and not with respect to Marathi, Kannada, or Telugu. Shamsur Rahman Faruqi, ‘A Long History of Urdu Literary Culture, Part I: Naming and Placing a Literary Culture’, in Pollock (2003a), pp. 824, 831, 836–7; Shamsur Rahman Faruqi, Early Urdu Literary Culture and History, New Delhi: Oxford University Press, 2001, pp. 95– 104. See also Simon Digby, ‘Before Timur Came: Provincialization of the Delhi Sultanate through the Fourteenth Century’, Journal of the Economic and Social History of the Orient 47(3), 2004: 333–6. (14) D.J. Matthews, ‘Eighty Years of Dakani Scholarship’, Annual of Urdu Studies, 8, 1993: 104; Carla Petievich, ‘The Feminine and Cultural Syncretism in Early Dakani Poetry’, Annual of Urdu Studies, 8, 1993: 120–1. (15) Faruqi in Pollock (2003a), pp. 838–9; Matthews (1993): 92. In Bijapur, pioneers of Dakani literature included Chishti shaykhs such as Shah Miranji Shams al-‘Ushshaq (d. 1499) and his son Burhan al-Din Janam (d. 1597). (16) Richard M. Eaton, Sufis of Bijapur, 1300–1700: Social Roles of Sufis in Medieval India, Princeton: Princeton University Press, 1978, pp. 136–42, 155– 64. (17) Guha (2004): 25.

Page 14 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars (18) K. Lakshmi Ranjanam, ‘Telugu’, in History of Medieval Deccan (1295–1724), H.K. Sherwani and P.M. Joshi (eds), Hyderabad: Government of Andhra Pradesh, 1974, vol. 2, p. 147. (19) Phillip B. Wagoner, ‘Emperor Ibharama and the Telugu World of Qutb Shahi Hyderabad’, Paper delivered at the Institute of Fine Arts, New York University, 10 March 2004. (20) K. Lakshmi Ranjanam (1974), pp. 147–61, 161–3. In their inscriptions, the Vengi Chalukyas of the Andhra coast began replacing Sanskrit with Telugu for political discourse as early as the early tenth century. Court poets began composing kavya in Telugu from around 1050. See Pollock (2003a), p. 289. (21) Notable men of letters from Iran included Shah Tahir (d. 1549), Ghulam ‘Ali Hindu Shah Astarabadi, ‘Ali Niwadi (d. 1567), Sayyid Qasim Arsalan (d. 1606), Malik Qummi (d. 1616), Mawlana Haydar Zuhuri (d. 1616), Khwaja Ahmad Fani, Shah Fath Allah Shirazi, and Mirza Hayati (d. 1643). See Radhey Shyam, The Kingdom of Ahmadnagar, Delhi: Motilal Banarsidass, 1966, pp. 378–90. (22) Chitre (1991), p. xvii. (23) Persianized Turks had introduced paper-making technology from Central Asia to north India in the thirteenth century. By the fifteenth century, it had diffused to the Deccan, where the use of paper gradually displaced the palm leaf. The earliest surviving paper documents pertaining to local administration at Golkonda or Bijapur date to the sixteenth century. The link between paper production and Indo-Muslim governance in the Deccan is suggested by an ancient town called Kaghdhipura (‘paper-town’), located very near Daulatabad (‘administration-town’), the oldest centre of Indo-Muslim governance in the Deccan. See P.K. Gode, ‘Migration of Paper from China to India, A.D. 105 to 1500’, in Paper Making (as a Cottage Industry), K.B. Joshi (ed.), Wardha, India: V.L. Mehta, 1947, pp. 198–214; Jeremiah P. Losty, The Art of the Book in India, London: British Library, 1982, pp. 10–12; Jonathan M. Bloom, Paper before Print: the History and Impact of Paper in the Islamic World, New Haven: Yale University Press, 2001, pp. 41–2; Nile Green, Indian Sufism since the Seventeenth Century: Saints, Books and Empires in the Muslim Deccan. New York: Routledge, 2006, p. 65. (24) Muzaffar Alam, ‘The Culture and Politics of Persian in Precolonial Hindustan’, in Pollock (2003a), p. 157. (25) Muhammad Qasim Firishta, History of the Rise of the Mahomedan Power in India, John Briggs (trans.), 1829, repr. Calcutta: Editions Indian, 1966, vol. 3, pp. 47–8. Muhammad Qasim Firishta, Tā’rīkh-i Firishta, Lucknow: Neval Kishor, 1864–5, vol. 2, p. 27. Firishta called the new language ‘Hindavi’, the term Westerners like Firishta used in referring to any Indian vernacular. Since this Page 15 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Rise of Written Vernaculars term could have applied to Kannada, the indigenous vernacular of the kingdom’s southern tracts, one may presume that local records in Kannada also appeared at this time. As yet, though, no such records have turned up. (26) Hiroshi Fukazawa, The Medieval Deccan: Peasants, Social Systems and States, Sixteenth to Eighteenth Centuries, Delhi: Oxford University Press, 1991, p. 5. (27) Briggs (1966), vol. 3, pp. 47–8. Text: ‘daftar-i fārsī bar ṭaraf sākhta hindawī kard, wa bahamana-rā ṣāḥib-i daḵẖḻ gardānīda, jami‘-i davābiṭ-i Ibrāhīm ‘Ādil Shāh-rā bar ham zad’; Firishta (1864–65), p. 227. (28) As we have seen, both Marathi and Kannada were used for documentary purposes as early as the tenth century. For expressive purposes, Kannada was used from the eleventh century, and Maratha from the early fourteenth century; see Pollock (2006), pp. 288–9. (29) See Graham Smith and J. Duncan M. Derrett, ‘Hindu Judicial Administration in Pre-British Times and its Lesson for Today’, Journal of the American Oriental Society 95(3), 1975: 417–23. The original Marathi document is cited in V.K. Rajwade, Marāṭhyāñcya itihāsaci sādhane 15, Bharat Itihas Sanshodhan Mandal, 1912, pp. 22–8 (no. 6). (30) Noting that they combined both state officials and local officers and landholders, V.T. Gune remarked that Brahmin-led councils issuing mahzars represented ‘an excellent mixture of the Muslim and indigenous institutions of local administration … All problems of local interest, whether administrative, social, or religious, were treated as a subject for Mahzar.’ V.T. Gune, Judicial System of the Marathas, Poona: Deccan College Postgraduate and Research Institute, 1953, p. 139. (31) One should not conclude, however, that such a process occurred only among Marathi-speakers. What we know about Narasoji and Bapaji could be due to an accident of the preservation of Marathi documents; such documents in Telugu and Kannada have yet not turned up. But one may assume that, just as the sultan of Bijapur presided over an orderly system of dispute management operating in Marathi, neighbouring sultans would have managed similar institutions using the vernacular tongues of the lands they ruled. As a consequence, discursive communities of Telugu- and Kannada-speakers likely evolved along a similar path as occurred in the case of Marathi-speakers.

Access brought to you by:

Page 16 of 16

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Turki and Hindavi in the World of Persian Fourteenth- and Fifteenth-Century Dictionaries Dilorom Karomat

DOI:10.1093/acprof:oso/9780199450664.003.0005

Abstract and Keywords This chapter offers primary material from a number of Persian dictionaries produced in India for the benefit of non-Persian-speaking readers in the fifteenth and the early sixteenth centuries. Compiled in the regional sultanates, such dictionaries demonstrate the growing use of Persian in India. The chapter contains abundant quotations, especially on musical terminology, and an analysis of the use of Turki and Hindavi in such dictionaries. The chapter suggests that such dictionaries helped members of a linguistically heterogenous elite with interests in trade and worldly pursuits and with cultural aspirations that looked towards Persia and Central Asia. Keywords:   Persian, Turki, Hindi/Hindavi, dictionaries, lexicons, music

FOR ABOUT A THOUSAND YEARS, Persian was a multicultural language within different empires and regions and served as a language of political administration, commerce, and literature. Over a very wide cultural area, Persian acted also as the intermediary between a superordinate tongue, Arabic, and the various Iranian, Turkic, and Indic vernaculars in the domains of religion and science.1 Given these two features, it is unsurprising that lexicography became one of the most important fields of literary endeavour in Persian literary culture, beginning in the eleventh century and up to the present day, given that Persian (p.131) language and literature were acquired and cultivated by people whose first language was not Persian, and that people within the textual world of Persian regularly encountered terms coming from a wide range of languages, Page 1 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian both archaic and current. Persian dictionaries compiled in India were not bilingual in the modern sense of having one language for the lemmata and one for the explanations (e.g., English-Persian or Persian-English), yet they were sufficiently multilingual in their lemmata and in the synonyms and terms used in the explanations to provide us with a precious guide to the multilingual world they served. Persian dictionaries include two different kinds, the lughat and the farhang. A lughat, which can be glossed as ‘book of language’, is a simple dictionary in which the words and their equivalents in one or several languages are given, sometimes as many as seven or more!2 A farhang instead is a compilation of knowledge and culture and of the resources of the literary language. It is an ‘explanatory dictionary’ of Persian vocabulary. The large number of such explanatory dictionaries in Persian bears witness to the great demand for these scholarly works, which contain valuable material for the historical lexicology of the Persian language and Iranian dialectology, the history of classical literature and textual criticism, and also for ethnography, historical geography, and cultural history.3 (p.132) As Baevskii noted, Arabic lexicography came into existence, ostensibly at least, as a tool for reading religious texts, the Qur’an in primis. By contrast, the impetus for Persian farhangs came from the need to appreciate poetry.4 Reading Persian literary texts was often impeded by the occurrence of obsolete and archaic words introduced by earlier poets, of numerous names of historical figures, legendary and mythological personages, of geographical locations, of terms from different traditional sciences, of curiosities pertaining to foreign cultures, and of words used by poets in unfamiliar dialectal or allegorical senses. Moreover, the practice of writing poetry and ornate prose required authors to be fluent in the use of the vast vocabulary of literary Persian. Thus, farhangs were valuable guides for poetry readers and writers. They also, by necessity, involved certain elements of linguistics. Lexicographers not only called attention to phonology in order to establish the correct pronunciation of a word, they also examined individual lexical and grammatical formatives, singled out synonymic and homographic groups, and traced the etymology of individual words. Each farhang provided its own range of vocabulary that required definition and would not simply copy or repeat what was in other dictionaries.5 The early period of Persian lexicography in India, from the thirteenth to fifteenth centuries, played a significant part in the history of Persian lexicography in general, and many of the technical principles introduced by the authors of the first Persian farhangs in India were widely adopted, and elaborated by later Persian lexicographers in India, in the sixteenth to nineteenth centuries. The early dictionaries were characterized by a gradual and significant increase in scope—with foreign words, dialectal and colloquial lexis, phraseology, and in particular the ordering of words by initial and the systematic ordering of the Page 2 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian following letters, culminating in general alphabetical reference books that are standard today.6 (p.133) This essay deals with farhangs written in India in the fourteenth and fifteenth centuries, and in particular with the clues that the presence and range of Turkic and Indian words can give us about the production and reproduction of multilingual north India at the time.7 These dictionaries were compiled in the various locales of the north Indian Sultanates—Delhi, Jaunpur, Bihar/Bengal, and Malwa—but as each successive dictionary was aware and made use of the ones before, they were clearly part of a well-connected world. The dictionaries in question are the Dastūr al-afāżil fi’l-lughāt al-fażā’il (1342) by Hajib-i Khayrat Rafi Dihlawi (whose patron was Shams al-Din Muhammad Jajniri);8 the Adāt alfużalā, compiled in 1419 by Qazi Badr al-Din Muhammad Dihlawi called ‘Dharwal’;9 the Farhang-i zafāngūyā u jahānpūyā, compiled in (p.134) 1433 by Badr al-Din Ibrahim, who in 1409 or 1419 had set out from Jaunpur in order to ‘kiss the threshhold’ of the illustrious prince Qadar Khan b. Dilawar Khan, the founder of the Ghurid dynasty in Malwa;10 the Mujmal al-‘ajam (also called Farhang-i ‘Asimī, 1493), compiled by ‘Asim Shu’ayb ‘Abdusi for the ruler of Berar, son of the vizier ‘Ali Akbar Dadu Khan ibn ‘Imad al-Mulk;11 and the Sharafnāmayi Munyārī (or Manyārī or Manerī), also called Farhang-i Ibrāhīmī, compiled in 1473–4 by Ibrahim Qiwam al-Din Faruqi, a native of Bihar, under the rule of Sultan Abu’l Muzaffar Barbak Shah of Bengal (r. 1457 to 1474), though dedicated to the celebrated Sufi saint Sharaf al-Din Ahmad b. Yahya Maneri (d. either 1371 or 1381). The author was also a poet and used his own verses to introduce the work and each letter chapter (bab), in this case with verses that ended with the letter of that particular bab.12 (p.135) These farhangs were not only explanatory and defining dictionaries,13 but also functioned simultaneously as encyclopedias,14 dictionaries of (p.136) rhyme,15 of synonyms,16 and antonyms.17 Furthermore, they served as a kind of commentary on major literary works, especially Firdawsi’s Shāhnāma, since Persian medieval manuscripts were not accompanied by notes or comments. It was the farhang that fulfilled this role.18 In his pioneering work on Early Persian Lexicography: Farhangs of the Eleventh to the Fifteenth Centuries, the Russian scholar Solomon I. Baevskii identified the following characteristic principles of dictionaries compiled in this period: 1. Ordering of the vocabulary in alphabetical order by initial. The Dastūr al-afāżil, Farhang-i zafāngūyā, Adāt al-fużalā, Mujmal al-‘ajam, and Sharafnāma-yi Munyārī are thus organized. 2. Within the ordering by initial, consecutive alphabetical ordering is predominantly by second letter. The third step in the structural ordering of entries, by final letter, was pioneered by the Adāt al-fużalā.

Page 3 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian 3. Implementation of a grammatical classification of lexis: in the farhang Zafāngūyā, infinitives feature in a special section, and in both Zafāngūyā and Adāt al-fużalā metaphorical and idiomatic expressions (p.137) are included within a special section, either an addendum or a separate chapter. 4. Gradual increase in the number of entries. 5. Alternative arrangement of Persian defining dictionaries on a thematic principle (e.g., Fakhr-i Qawwās and part two of Baḥr al-fażā’il). 6. Compilation of partly multilingual dictionaries (Zafāngūyā, Baḥr alfażā’il, Sharafnāma-yi Munyārī).19 The Adāt al-fużalā can also be mentioned here, though the Turkic words are given without any special mention. As a rule, the text in the Persian dictionaries were written continuously, on one line without distinguishing columns. The entry words were either underlined or written in red ink to differentiate them from the rest of the text written in black. The text of an entry could be very brief, often containing just two words—the entry word and its one-word explanation or equivalent. But more often entries were quite extensive: they included not only an explanation of the given word but, for example in the case of a noun, also the description of the object as well as of the material the object was made of, and how it functioned.20 In some entries authors specified whether such an object existed in India or not, a testimony that those dictionaries were written for Indian readers.21 Often the explanation of an entry was supported by a poetic quotation with an instance of how the entry word was used. One example from Sharafnāma-yi Munyārī: Āqsunqur, with quiescent qāf and zamma on sin and the second qāf, is a white hunting bird. Turks also give such names to their slaves. [This is followed by a verse by Khaqani.]22 Sharafnāma, 2006, vol. 1, p. 57 (p.138) The skill of ‘sprinkling’ one’s speech and writing with poetic quotations was considered a crucial component of medieval Persian education. As is widely known, the lack of diacritics in Persian manuscripts can cause a certain amount of difficulty in reading and understanding the text for people whose first language is not Persian. In the dictionaries, the earliest known means for ensuring the correct reading and pronunciation of the words explained were the auxiliary upper- and underline marks, the so-called ḥarakāt, borrowed from Arabic.23 The Persian farhang, thus, played an important role in fixing, correcting, maintaining, and preserving the norms of literary pronunciation (and in providing a historical account of phonological shift and variation).24 In the farhangs an entry word was repeated over and over again in a rhyme, so its metre and the metre of the neighbouring rhymes would confirm Page 4 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian the accurate reading of a given word. It was a method of documenting pronunciation through ‘corroborative quotation’ using well-known Persian verses that was and remains a very valuable tool for the study of the language. (p.139) This kind of poetic quotation was applied in rare circumstances also to Turkic words that had been assimilated into Farsi, such as in the term Āqsunqur mentioned above. In the only copy of Miftāḥ al-fużalā, preserved in the British Library, words are presented with two different aids for pronunciation, diacritical marks (harakat) and grammatical terms, though rarely in the same entry. It is worth mentioning that diacritics were used not only for Persian, Arabic, or Turkic languages. For instance, in the Miftāḥ al-fużalā and in some of copies of Adāt al-fużalā, diacritics were applied to equivalent words in Indian languages, and not just for vowels but for consonants as well. (see Plate 5.1)26 This supports the hypothesis that such dictionaries served a twofold purpose: as Persian textbooks for Indian learners and as manuals of Hindavi for Persian speakers.25 From the earliest examples, it became a stable practice to begin a farhang with an introduction (muqaddima or dibacha), in which the author, after the traditional invocations and tribute to his patron, usually laid out the motives that had brought him to this effort, indicated his goal, the languages he knew, and the texts he had consulted.27 Typically, he would say that a group of friends (or sometimes pupils, or sons) had addressed him with a request to create a textbook or a kind of educational aid that would help them appreciate the text of a poetical work (see also Chapter 6 in this volume).28 Among the poems (p.140) frequently mentioned were the Shāhnāma of Firdawsi and Iskandar-nāma of Nizami.29 Muqaddimas often contained partial criticisms of previous lexicographers, alongside with names of the dictionaries that the compiler had used and an account and evaluation of his experience. Thus, the author of the Dastūr al-afāżil claimed: I translated some Arabic and Persian dictionaries … and I included in my dictionary words of Arabic, Turkic, Mongolian, Pahlavi, Persian, Afghan [Pashto], Jewish [Judaeo-Persian?], Christian [Aramaic?]; the tongues of the Magians, Syrians, philosophers, and Tajiks; Hebrew; words from the dialects of Rayy, Hijaz, and Transoxiana, poetical idioms from every city, scholarly coinages, and popular sayings.30 Only a few lines long at first, these introductions sometimes attained the size of a complete article that discussed the lexico-grammatical outlines of Persian and of rare Turkic languages, as in the Sharafnāma-yi Munyarī.31 Grammatical elements began to appear consistently in farhangs only in the seventeenth century,32 while Persian-language grammars of (p.141) Persian were written at a much later stage, the sixteenth to eighteenth century (see Chapter 6).

Page 5 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian What was in the works of the classics of Persian poetry that went beyond the comprehension of readers in north India and required definition, explanation, and comment? The evidence from the farhangs suggests the following categories of words:33 1. proper names of historical, mythical, as well as literary characters of poems. For example: ‘Zaratusht, Zardusht, Zaradtusht, Zarātusht and Dharaduhsht: all five words are the name of the wise man who falsely claimed to be a prophet (paighambar). He founded the religion of the Magis, the fire-worshippers … and wrote books of fire-worship like the Zand, Pāzand, Zandastā and so on.’34 2. ethnic names of peoples and tribes, for example: ‘Khifchāq: the name of a desert in which nomadic Turks live’.35 3. geographical names of countries, regions, towns, rivers and mountains: ‘Warzrūd: name of Māwarā’an-nahr (Transoxiana)’.36 (p.142) 4. local names of flora and fauna, for example, ‘Ushturghār: a plant Indians call jānwāsah.37 ‘Darakht sunbah: a black bee which drills a hole in the wood of a tree or a bamboo. In Hindavi they call it bhōnrā’.38 5. specificities of foreign cultures and religions (e.g., Christians, Zoroastrians, Buddhists, etc.): ‘Ustā [= Avesta]: with zamma and fatḥa, the name of a book of the Magis on the injunctions of the religion of fireworship, the commentary to the Zand, the composition of Ibrahim Zaratusht. Also pronounced satā, with omitted alif/hamza. They worship it. And with kasra it means ‘termination’ (followed by a verse of Khaqani).39 6. terms and definitions related to the sciences of astronomy, mathematics, palmistry, chemistry, music, and so on, as well as to tools, implements and toys: for example, ‘Ābān: the period in which the Sun remains in Cancer that Persians call the ‘month of āyān’ and Indians call kātig or aghan’;40 ‘Uqlīdis: with zamma on hamza [alif] and kasra on dāl, the title of a book on the science of philosophy [hikmat] named after its author. One hears it from Shihab al-Din Hakim Kirmani and from Shaykh Wahidi with kasra on alif and fatḥa on dāl, and they say that the meaning of this book is geometry …’;41 ‘khūn-i Siyāvushān: the name of a medicine which in Arabic is called dam al-akhwīn and in Hindavi pāndurūt’;42 ‘ijtimā’: the name of the fourteenth figure in the science of geomancy’;43 ‘bādfira: a whirligig with which children play, a small piece of wood with a thread in the middle: children twist and whirl it. In Hindavi they call it phirkī. Children run and it spins when they run.’44 In his monograph, Baevskii outlined the significant features of each farhang under study here.45 In this essay, instead, I will focus on (p.143) Indo-Persian farhangs as documents of the intermediary place of Persian within multilingual north India. I will, therefore, focus on the presence of Turki, the language of much of the Sultanate military elite, and of Hindavi, the local language that Page 6 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian lexicographers customarily referred to, and which they and their readers must have known, though it normally remained outside the world of written Persian (see Chapter 14). In order to push a little further Baevskii’s argument that farhangs provide a precious source for the history of culture, we shall see how much of the Turki language was preserved and considered significant in fifteenth-century north India, and what aspects of Turkic and Indian culture (including material culture) Indo-Persian lexicographers deemed worthy of notice.

Turki in Indo-Persian Dictionaries The Turks, who were divided into different tribes who spoke at least thirty-five languages and inhabited a vast territory spanning different cultural zones, not only contributed to the development of Persian literature but also played a considerable role in spreading the Persian and Turkic languages. Turkic languages were not only spoken by soldiers and used at court or by the aristocracy. Although many Turkic poets were bilingual and actually wrote more in Persian than in Turki, some Turkic languages, like Chaghatai with ‘Ali Shir Nava’i in Khorasan and Azerbaijani with Fuzuli in Baghdad, became poetic languages themselves.46 In case of Indian dictionaries from the fourteenth and fifteenth centuries, the term ‘Turki’ usually refers to Chaghatai (also called Eastern Turki, Jagatai, Jaghatai), a Turkic literary language of Central Asia of the fourteenth to sixteenth centuries, named after Genghis Khan’s second son who ruled over this territory. I have chosen the term ‘Turki’ here in line with common Persian usage, though we have seen above that lexicographers at times specified which Turki tribe (and language) they were referring to. (p.144) Turkic lexicography dates back to at least the eleventh century, with the remarkable Dīwān lughāt al-Turk by Mahmud Kashghari.47 It is unlikely, however, that the farhang compilers in India knew of this Turkic work, or of the Muḥākamat al-lughatayn (The trial of two languages, 1499) written by their contemporary ‘Ali Shir Nava’i.48 Persian lexicographers in India referred to the Turki that was developed and used in India by different Turkic tribes such as the Chaghatai, the Aq Qoyunlu, the Barlas, or the Qanqali. Dictionaries compiled until the sixteenth century contain words of Central Asian (Chaghatai) Turki, and later an admixture of Chaghatai and Osmanli. This Turki was gradually assimilated into the local languages by the end of the eighteenth century.49 Turki terms featured in Indo-Persian dictionaries from the start, first with a few terms in the Dastūr al-afāżil (1342) and the Adāt al-fużalā (1419), and then with as many as 603 dedicated word entries and a complete separate section (bakhsh) in Badr al-Din Ibrahim’s Zafāngūyā (1433, according to the Patna manuscript).50 The Zafāngūyā seems (p.145) to have been the only dictionary in which foreign words and dialects were collected in special chapters (named Page 7 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian bakhshs). Other dictionaries, following the pattern of the Sharafnāma-yi Munyarī, placed Turkic and Arabic words at the end of each letter chapter.51 Indo-Persian lexicographers thus considered Turkic words an integral part of the stock of words that they sought to explain, but while later lexicographers were aware of their predecessors, they often chose to include different terms, or to give different spellings. If we take the Turkic words from two sample chapters (‘alif-alif’ and ‘alif-te’) of the Zafāngūyā and Sharafnāma, for example, we notice that the author of Sharafnāma defined Turkic words that were not mentioned in Zafāngūyā, omitted others that had been mentioned, and occasionally gave different spellings.52 The difference between the Zafāngūyā and Sharafnāma lies not only in the choice of words and their spellings but also in that the author of Sharafnāma usefully described how a word should be pronounced: Aksur: with vowel sound ‘a’ on the first and vowel sound ‘u’(zamma) on the third letter, without father and without mother. (Sharafnāma, 2006, vol. 1, p. 57)53 In the earliest Indo-Persian dictionaries we cannot discern an orientation towards Turkic lexicography, by which I mean the desire to include not just words of Turkic origin that were already integrated and used in Persian. This orientation emerges with the Adāt al-fużalā and the Sharafnāma, which distinguished between the two categories and (p.146) sometimes note that ‘this word is Turkic’. Many dictionaries included Turkic equivalents, which the author prefaced with the phrase ba-turkī ‘in Turkic’.54 A possible reason behind this orientation is that the Sharafnāma was compiled in Bihar, where many of the Sultanate settlers lived and were called ‘Turks’, and where even today one can find many Turkic words in the local dialects. Although we do not know much about the author of Adāt al-fużalā, the unusual abundance in his dictionary of toponyms from Transoxiana, what is today South Tajikistan, suggests that he or his family were probably originally from that region. As already noted, whereas lexicographers often presented Indian equivalents for Persian words they never did so for Turkic word entries, though in rare instances we find such equivalents in entries of Turkic words assimilated into Farsi that bore no indication that they were ‘Turki’. Among the considerable number of words included in the farhangs that mention the Turkic origin of some objects (or describe the objects brought by the Turks) or have Turkic equivalents were, as we shall see, several musical terms such as sabazghuv, surghin, surnāi, and so on (see section below). The number of Turki word entries increased over time, not only due to the general increase in dictionary entries but also to a growing interest in preserving Turki.55

Page 8 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian In early farhangs the definitions were usually extremely brief, often consisting of a single word.56 But the descriptions of their pronunciation and the diacritics that showed how to read a word are of great value (p.147) in tracking the phonetic changes that took place, though this is an area that has hardly been investigated yet. One occasionally finds a curious interpretation of Turkic terms in India, for example ikdish. Scholars like Baevskii and Kapranov have read this word as yākdish.57 But thanks to the descriptive method of pronunciation in the Sharafnāma and Adāt al-fużalā, we know that it should be read as ikdish. Moreover, while in Central Asia ikdish refers to a horse bred from Arab and Turkic stock, in India the term ikdish expanded to refer to a person of mixed Hindu and Turk parentage. Here we can see how the definition of a term changed in the Indian context and was ‘Indianized’: Ikdish [‫]ﺍﮐﺪﺵ‬: so named a mixed child, that is, from two kinds [jins] which merged into Turki. (Zafāngūyā, 1997, vol. 2, fifth bakhsh, p. 98) Ikdish: with vowel sound ‘i’ on first and third letters, that is a Turk whose father or mother is Hindu or a horse which is from one side Turkish and from other Arab’ [there follow verses from Nizami’s Khusraw Shīrīn and by Mas‘ud Sa‘d Salman] … Īkdish [ekdish, ‫]ﯾﮑﺪﺵ‬: the same as Ikdish. (Sharafnāma, 2006, vol. 1, p. 66) Ikdīsh [‫]ﺍﮐﺪﯾﺶ‬: that is a Turk who is Indian from the mother’s side and Turk from the father’s side and that is a horse … (Adāt al-fużalā, Or. 265, f. 6b) In this example, phonetic variants are added to the entry word, but in other cases even dialectal synonyms are represented, as in the following case: Ātash zanah [= touchwood]: a tool [made] from steel, which struck on flint will start a fire, and in Turki it is called Chaqmākh [‫]ﭼﻘﻤﺎﺥ‬, Chaqmāq [‫]ﭼﻘﻤﺎﻕ‬ and also Chaqmaq [‫… ]ﭼﻘﻤﻖ‬ (Sharafnāma, 2006, vol. 1, p. 105) Every dictionary contained a certain quantity of synonymic lexis or synonymic combinations, which could be easily found within the alphabetically organized dictionary. In some cases, as with bughrā, the word (p.148) was mentioned in both the Persian and Turki sections, since it was of Turki origin but assimilated into Farsi: Bughrā: with zamma, it is a famous dish and it is also called pūrak;

Page 9 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian ‘Al-Turki’ Bughrā [‫]ﺑﻐﺮﺍ‬: with zamma, the name of a ruler of Khwarazm. Also a crane which flies ahead of the others. A type of dish, and also a male camel. (Sharafnāma, 2006, vol. 1, p. 132) If Indo-Persian lexicographers were following tradition in the way they explained and expounded on Persian terms, the authors of the Zafāngūyā and Sharafnāma invented their own ways for representing Turkic words, with extremely brief definitions. Another important point in the context of the topic of this essay is that the Sharafnāma-yi Munyarī also contains information about Turki grammar in a separate chapter on ‘letters, meanings and [grammatical] rules’ (Bāb-i ḥuruf, alma‘ānī wa al-qawā‘id). This is very significant and speaks both of the lexicographer’s pride and care in displaying his knowledge of Turki in this Persian text, and also of the perceived interest this information would have for his prospective readers: Section in [letter] mim: ‘… and Turks also add it to the end of a word, which is to augment for themselves, for example: ughil = a son, ughlim = my son… Section in [letter] yā: …and when Turks [wish to show] the connection between one thing to another [they] append [this letter] to the end [of a word], like: bāsh [bash] = head, bāshī [bashi] = his/her head … Section in [letters] L with R: when Turks want to make the plural of a word they add ‘lar’ to the end of a word, for example: āt = horse, ātlar = horses … Section in [letters] Ch with yā: when at the end of a Turki word you add chi, the creator of that thing is intended, as in: yāy [yā] = bow, yāychī [yāchī] = bowmaker … Section in [letters] L with Gh: when in the words of Turks you add ‘ligh’ [lugh] at the end of a word, it belongs to someone or something, as in: āt = horse, atligh [atlugh] = with [on] a horse … Section in S with Z: when Turks add ‘siz’ to the end of word, this word is absent, like: yārmaq [yārmāq] = wire, yārmaqsiz = without wire … Section in N with G: when they add [these two letters] at the end of Turki words, then this thing is being linked to another that is present, like: qul = slave, servant; quling = your slave, your servant …

Page 10 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (p.149) Section in Q with H and G with H: whenever [these letters] are added at the end of Turki words, they mean [the definite object], as in ton [tōn] = garment, tonqah and tongah = the garment.58 The author also specifies at the end which variety of Turki he refers to: ‘In this chapter several grammatical rules of the Turki language were quoted with an explanation of Turki words according to the speech of Qanqali.’59 That Turki was given such an important place as one of the major languages is unsurprising, given the fact that most of the Sultanate rulers and elites were in fact Turks. Yet its presence has often been neglected due to the prominence of Persian. The range of Turki words in Indo-Persian dictionaries of the fourteenth and fifteenth centuries is indeed vast. Dictionaries included words of different tribes and covered all areas of life, suggesting that they were meant to help in understanding terms occurring in literary texts, as well as providing useful terms for merchants, travellers, local inhabitants, and newcomers. Unlike Persian words, Turkic words were not commented upon with poetical examples (except those words which had been assimilated into Persian at an early stage), and unlike Hindavi, they were not used as support to explain a lemma. The structural difference between Turki and Hindavi in relation to Persian in the dictionaries is that Turki appears as a language which needed to be commented or translated into Persian, whereas Hindavi worked as an aid to understanding both Persian and rare Turkic words. In the context of a volume on fifteenth-century literary culture in north India, what is of particular interest is the extent of multilingualism in relation to terms connected with different domains of culture (including music, clothing, and food), science (particularly astronomy and botany), as well as geographical and ethnographic knowledge. Indo-Persian dictionaries can be fruitful sources in this enquiry and give a concrete sense of the kind of knowledge that was proffered, required, and transmitted. In the rest of this essay I shall, therefore, explore these different areas, noting in particular the increase in Indian words and their distribution across the various domains. I will conclude with a section (p.150) on musical terms, since this was an area in which we find a particularly high number of Turki and Hindavi terms. This evidence suggests that, first, a number of Turki professional musicians and musical genres were active, or actively remembered, in this period; second, that the Persian-educated Turki elites were particularly interested in Indian music and actively drew connections between Turki and Indian musical instruments and concepts. This confirms the understanding of this period of music historians, for whom dictionaries can provide particularly concrete and detailed evidence. But let us turn first to Hindavi words.

Page 11 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Hindavi in Indo-Persian Dictionaries Within Persian dictionaries, Hindavi words occurred predominantly as synonyms in the explanation of lemmata, though occasionally they also appeared as lemmata themselves, following the usual typology of proper names, toponyms, and culturally specific terms. Though the presence of Indic (Hindi, Hindavi) equivalents is peculiar to lughat and farhangs composed in India, it is perfectly in line with the pre-Indian tradition of Persian lexicography that included local languages and dialects in the text of the dictionary entries. Hindavi was often simply added to other language equivalents:60 Robāh-i Turkī [= porcupine]: with vāv-i farsī, it is a name of a crawler that has very sharp and long quills like a grown spindle on its back. When a dog follows it, it scatters its quills. The quills leap out from its back and injure the dog like arrows. Arabs call it qatfad, Indians [ahl-i Hind] sīh [seh], and in the language of the Deccan they call it sārsag. (Adāt al-fużalā, Or. 1262, f. 38b).61 Lexicographers occasionally gave two or more Indian equivalents, showing an awareness of the existence of several languages within north (p.151) India, or that different words were used by specific caste and religious groups: Sag angūr: the grapes of the desert called lahsūrah. Some [say] that sag angūr is the plant which in Arabic is called ‘anab al-tha’alab and in Hindavi kānū [‫ ]ﮐﺎﻧﻮ‬and kanwāyī [‫ﮐﻨﻮﺍﯼ‬ ‫ﹰ‬ ]. Some say that in Hindavi it is called pūthī [‫]ﭘﻮﺗﻬﯽ‬, while others call it amlawarah [‫]ﺍﻣﻠﻮﺭﻩ‬. (Zafāngūyā, 1997, vol. 2, second bakhsh, p. 31) Basāk: a chaplet of flowers that [some] Indians call sehra and other Indians call maur [‫]ﻣﻮﺭ‬. (Adāt al-fużalā, Or. 1262, f. 17a) The presence of Indian equivalents shows not only that that the authors of these dictionaries knew the local languages, but that they considered the Indian vernaculars part of their linguistic system of knowledge. Persian dictionaries aimed at and took pride in giving a complete definition for each word, and Hindavi was necessary within this scheme. At the same time, this multilingualism of principle made the dictionaries also pragmatically useful and equally comprehensible for Indian, Persian, and Turkish readers. In the early dictionaries we do not find Turkic and Indic equivalents in the same entry, evidence of the relative prominence accorded to Persian as the language of power and administration, but there are many examples of Arabic and Indic equivalents in the same entry. Manuscripts copied by different scribes through the ages and in different locations present differences not only in terms of the Page 12 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian lemmata and definitions, but also in the number of Indic equivalents. For example, the two copies of Adāt al-fużalā preserved in the British Library give a different number of Indic equivalents.62 One explanation could be that one copy was made for Indian readers and the other for Iranian or Central Asian readers. Comparative research on dictionary entries shows that over time Indic equivalents gradually increased in number. For example, the total (p.152) number of Indic equivalents in the Zafāngūyā is ninety-two,63 whereas the later Ghiyās al-lughāt (1887) includes about one thousand. Moreover, in the Zafāngūyā half of these Indic equivalents refer to medicines and medicinal plants, which give us a clue to the dictionary’s specific focus and to the interest in Indian medicine in the Persian world. Occasionally, the compiler did not give the meaning of a common word in Persian, but only the Indic equivalent, as in this example: Bād bīzan: in Hindavi it is called pakah [pankah] (Zafāngūyā, 1997, second bakhsh, vol. 2, p. 8) Several definitions remark upon the fact that a word was borrowed from Hindavi into Persian, as in the case of the term bīlak: Bīlak: with Persian yā, it is an arrow [dart] for hunting and this is an Indian word that was used in Persian. It is also pronounced as fīlak.64 (Sharafnāma, 2006, vol. 1, p. 176) The thematic range of Indic words in Indo-Persian dictionaries of the fourteenth and fifteenth centuries includes, as already mentioned, names of people, groups, and places; culturally specific items and kinship terms, words related to time, astronomy, medicine, botany, and agriculture; terms for implements and crafts, arms, toys, clothing; and, as we shall see, music. The categories are roughly the same as those of the dictionaries themselves, but with a noticeable focus on everyday terms and items, on the Indian branches of scientific knowledge (‘ilm) that Indo-Persian elites were most interested in—astronomy, botany, gemology, gardens, warfare, and music—as well as on items of international trade (clothes, spices, precious stones).65 Let us now consider a few examples from each category. (p.153) Proper names included historical characters that appeared in historical works. Note, for example, that Ram is not glossed as the name of a god or king, though the definition ‘gentle and obedient’ suggests Ram’s character: Jaypāl [‫]ﺟﹻﹻﯿﭙﺎﻝ‬: with vowel ‘a’ and Persian yā, is the name of the ruler (pādshāh) of Lahore who had three hundred elephants and thirty thousand horsemen; the royal kettledrum that Indians [Hinduvān] call singhnād Page 13 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian [‫‘ ﺳﻨﮑﻬﻨﺎﺩ‬lion’s roar’; interestingly, no Persian gloss is given] was established by him. In the end he was captured alive naked by order of Sultan Mahmud Sabuktigin. (Sharafnāma, 2006, vol. 1, p. 330) Jaysing: with Persian kāf, it is a name of the Rāi of Gujarat who first claimed to be an Rāi and ruler (pādshāh) in the country (mulk) of India. (Adāt al-fużalā, Or. 262, f. 21b) Rām: gentle, obedient [compliant], a common name which is found in the land (vilāyat) of India. (Adāt al-fużalā, Or. 1262, f. 37a) Toponyms of cities, regions, rivers, and mountains also tended to be similar to those found in historical texts and relevant to Indo-Persian elite readers: Sōmnāth [‫]ﺳﻮﻣﻨﺎﺓ‬: with Persian vāv and delayed ‘mīm’, it is the name of an idol-temple (but-khana) that was in the land (zamīn) of Gujarat. Sultan Mahmud Sabuktigin destroyed it. [followed by a verse from the Bustān …] (Sharafnāma, Add. 7678, f. 235a) Sind [‫]ﺳﻨﺪ‬: with vowel sound ‘i’ … also the well-known name of the river [which is] the borderline between Hindustan and Khorasan [followed by a verse from the Shāhnāmā]. (Sharafnāma, Add. 7678, f. 137b) Bihār [‫]ﺑﻬﺎﺭ‬: a (Buddhist) idol-temple in Turkestan; a city in India; [bahar ‘spring’] a season. (Dastūr al-afāżil, in Baevskii, 2007, p. 80) (p.154) Khaybar: the name of a fortification (hisār) which the Amir alMu’minin ‘Ali Karim Allah Vajh [= Muhammad] conquered [in 629], and some say that it is a fortification in Gujarat, which Indians call Jūnāgar and Karnār. (Adāt al-fużalā, Or. 1262, f. 24b) Danīr [Daner?]: the name of a city from the land of Hindustan. (Sharafnāma, 2006, vol. 1, p. 434) Comparatively, few terms relating to Indian religions and worship appear, and only among synonyms: Page 14 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Kharzahrah: a small tree, whose flowers Hindus/Indians worship (but parastī kunand), and Indians call it kīnū [‫]ﮐﯿﻨﻮ‬. If a horse or a donkey eats it they die, and it is called kīz. (Adāt al-fużalā, Or. 1262, f. 30b)66 Tarsā: a sect of fire-worshippers; the Arabs call them nasrāni and Indians call them muni [? ‫]ﻣﻦ‬ (Adāt al-fużalā, Or. 1262, f. 22b) A more substantial set of Indian words appears as synonyms for names of precious stones and for astronomical terms, including months, planets, zodiac signs, where the text also becomes more explicatory: Almās: a gem, in Hindavi they call it hīrā. (Zafāngūyā, 1997, fifth bakhsh, vol. 2, p. 98)67 Zāvūsh: the name of a star, a star/planet in the sixth heaven, which Arabs call Mushtari and Indians call Biraspat [‫]ﺑﺮﺳﭙﺖ‬. (Adāt al-fużalā, Or. 1262, f. 39a) Āb: … also the period when Sun is in the house of Leo [the fifth solar month of the Iranian calendar], which the people of Rūm called ‘the month of Āb (Ābmāh)’ and consider one month. Winds blow without profit. Indians call it Bhādon. (Sharafnāma, 2006, vol. 1, p. 30) Āzār: This is the period when the Sun dwells in the house of Pisces, which the people of Rūm consider as one month and call it ‘the month of Āzār’; Indians call it Chait. (Sharafnāma, 2006, vol. 1, p. 47) (p.155) Another set of Indic terms that appear in the dictionaries deals with medicinal substances, medicinal plants, with plants and grains in general—even very common ones—or with parts of the body: Zurunbād: a medicine which in Arabic is called rijl al-jarrad and in Hindavi kachūr.68 (Zafāngūyā, 1997, first bakhsh) Zāg: [vitriol] a drug which in India is called kasīs [‫]ﮐﺴﯿﺲ‬. (Baḥr al-fażā’il fi manāfi‘ al-afāżil, fol. 117a, in Baevskii, 2007, p. 107) Page 15 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Bahman: with fatḥa, the name of a medicine (dārū) of which there are two kinds—red and white—and which is used for expelling wind, fats and for potency. In India they call it iskandah [‫]ﺍﺳﮑﻨﺪﻩ‬. (Sharafnāma, 2006, vol. 1, p. 198) Āstarang [= ginseng]: with delayed ‘sīn’, a plant in the regions of China which grows in the image of men … It is found in some books on medicine that in India they call it lakhimnān [?‫]ﻟﮑﻬﻤﻨﺎﻥ‬. The lakhimnān plant was seen twice by the compiler of this book. It does not have the quality [?] that every one who digs it out dies. The root of this plant is in the image of a man. It grows in the province of Bihar and in valleys at the feet of mountains. (Sharafnāma, 2006, vol. 1, p. 76) Baqam: a red wood used by dyers for colouring clothes red. In Hindavi they call it bagam [bakam?], and also call it ‘artīnā.69 (Zafāngūyā, 1997, fifth bakhsh, vol. 2, p. 102) Hanā: with fatḥa, well-known, that is, hana-i zīn, and with kasra, that is, hinā with that tincture they dye hair red, Indians call it mehndī. (Sharafnāma, Add. 7678, f. 155a) Banūmāsh: with zamma, the name of a grain which Indians call mung. (Sharafnāma, 2006, vol. 1, p. 168) Mūza: well-known fruit, called in Arabic talh and in Hindavi kēla. (Zafāngūyā, first bakhsh, in Baevskii, 2007, pp. 96–7) Chārmaghz: with Persian ‘jim’ [that is, ch], a walnut which Indians call akhrōt. (Adāt al-fużalā, Or. 1262) (p.156) Ispinākh: with kasra on Persian ‘be’ [that is, p], a vegetable which in India they call pālak. (Sharafnāma, 2006, vol. 1, p. 40) Baqlat al-himqā: a purslane also named baqlat al-miyār, baqlat-i Zuharā. In Hindavi they call it lōng. (Zafāngūyā, 1997, fourth bakhsh, vol. 2, p. 65)

Page 16 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Pāshna: with Persian ‘be’, the back of the foot, which in Hindavi is called ērī. (Adāt al-fużalā, Or. 1262, f. 13a) Dahan mulāzhah: with Persian ‘zā’ [that is, zh], a dangling [bit of] flesh in the mouth. In Hindavi it is named gākalā [= gullet]. (Miftāḥ al-fużalā, Or. 3299) Note that in the case of parts of the body only the pronunciation and the Hindavi equivalent are given, possibly an indication that these were explanations aimed at Hindavi-speaking Indian readers. The other categories under which numerous Hindavi terms appear are animals, implements, including cooking implements and foodstuff (the latter in line with the special interest demonstrated in the Ni‘matnāma, written in Malwa.)70 Material and items of clothing, by contrast, are few. Animals include: Panj pāyāh: a crab, which in Hindi is called kēkra.71 (Zafāngūyā, 1997, second bakhsh, vol. 2, p. 14) Dulphīn: a water animal which has no eyes and a thin neck. It lives in dark, salt water. It has big teeth. In Hindavi they call it būlū. (Zafāngūyā, 1997, fifth bakhsh, vol. 2, p. 108) Shabtāb: a little worm like a spark of shining fire. Also called shab chirāghak, ‘arūsak, and kāghinah. In India they call it jugnī. (Sharafnāma, Add. 7678, f. 262a) Chughd: with Persian ‘jim’, a bird of bad omen which brings out news at night and becomes blind during daytime. Indians call it ullū. Some call it būm. (Adāt al-fużalā, Or. 1262, f. 26b) (p.157) Rasū: a reptile and an enemy of snakes. Indians call it neval. (Adāt al-fużalā, Or. 1262, f. 37b) Ghuram: a wild ewe, some call it mountain ewe, and a sheep that children ride. Indians call it ērka. (Miftāḥ al-fużalā, Or. 3299) Talī: an implement of the barber, which inhabitants of India name bhāndī. Page 17 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (Adāt al-fużalā, Or. 1262, f. 23b) Akkār: in the vazn [rhyming with] of khummār, a gardener whom Indians name kōerī. (Sharafnāma, 2006, vol. 1, p. 55) Anbur: with fatḥa on the first letter and zamma on the third, an iron tool with which blacksmiths clutch the hot iron. In Arabic they call it kalūb and in India they call it sandāsī.72 (Sharafnāma, 2006, vol. 1, p. 56) Kalāba: a wooden implement [= reel] for rolling/wrapping yarn and silk. In Hindavi they call it paretī. (Miftāḥ al-fużala, Or. 3299) Ardan: a thing like a skimmer which has a lot of holes. Sweetsellers clarify sugar and ghee with it. Indians call it pūta.73 (Adāt al-fużalā, Or. 265, f. 4a) Vanang: a rope, one end of which is tied up in one place and the other somewhere else, laid on bunches of wet grapes to make them dry; also called āvang. Indians call it bilkinī. (Miftāḥ al-fużalā, Or. 3299, f. 294a) Pisht: with kasra on Persian ‘be’, flour from roasted barley which Indians call sāttū. (Adāt al-fużalā, Or. 1262, f. 17a) (p.158) Chuvkhā: with Persian ‘jim’, a woollen robe and a kind of dress of Christians that is short. In India it is worn by yogis. Indians call it kanthā.74 (Adāt al-fużalā, Or. 265, f. 21a) Shārah [‫]ﺷﺎﺭﻩ‬: thin red cloth which is wrapped around a candle so that the air cannot blow it out. Also special cloth worn by the inhabitants of India [verse from the Shāhnāma follows]. (Sharafnāma, Add. 7678, f. 276a) Bādāmah: a kind of inferior silk which the inhabitants of India call pūta. (Adāt al-fużalā, Or. 1262, f. 13a)

Page 18 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Persian dictionaries were really polyglot dictionaries that included, beside Hindavi, Arabic, Turkic, Syrian, Greek, and Latin and sometimes even Pashto equivalents (all in Arabic transcription).75 As such, they were the result both of the aspiration of Indo-Persian lexicographers to explain precisely the meanings of words and expressions, and of the complex linguistic and ethno-social situation in Sultanate India, where individuals and groups of different backgrounds had come together and shared social practices. Note, for example, the high number of synonyms for polygamy: Anbāgh: with fatḥa, when two women are married to one man. Each woman becomes anbāgh to the other; also called sanānj, banj, vasnī. Indians call it sōgan [‫]ﺳﻮﮐﻦ‬. (Sharafnāma, 2006, vol.1, p. 68)76 The main feature of the written Persian dictionaries written in India is that all of them, as a rule, contained Hindavi words. Hindavi words occurred predominantly as synonyms—occasionally more than one—in the explanation of lemmata, though occasionally they also appeared as lemmata themselves, alongside the usual typology of proper names, toponyms, and culturally specific terms mentioned above. Comparative research on dictionary entries shows that Indic equivalents gradually increased in number over time, and that occasionally the compiler began (p.159) to only cite the Indic equivalent rather than give the meaning of the word in Persian, suggesting a definitely bilingual readership. As we have seen, the thematic range of Indic words in Indo-Persian dictionaries of the fourteenth and fifteenth centuries included words related to time, astronomy, medicine, botany, and agriculture, terms for implements and crafts, arms, toys, clothing, and, as we shall see, music.

Multilingual Musical Knowledge The multilingual context of north India as well as the presence of social and professional groups of musicians, entertainers, and patrons who had their own linguistic and lexical distinctiveness and repertoire of styles found reflection in the farhangs. For music historians, in fact, farhangs constitute a most valuable and often unique source, especially when descriptions are accompanied by visual depictions, as in the British Library manuscript of the only copy of Miftāḥ al-fużalā. In the field of music in particular, and of entertainment in general, we find a heightened multilingualism. In other words, we find many precise terms for musical instruments and genres of Central Asian/Turkic origin, for varieties of oral entertainment like riddles and enigmas, and for groups of professional musicians—many with Arabic and Indian equivalents given.77 This plurality shows not only the lexicographers’ pride in detailing even obscure or obsolete musical terms, but also a tendency towards setting up equivalents, especially between Central Asian and Indian musical instruments and forms.

Page 19 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Literary musical names find their way into farhangs, as in the case of the most famous musicians of the Iranian (partly Turanian) world, Khusraw Parvez’s musician Barbad and the beautiful heavenly musician Zuhra (= Venus): Bārbad [33]: delayed ‘rā’, it is the name of a musician of Khusraw [Parvez] who sang a rhymed song and that song is called khusrawānī; it is also the first letter of the alphabet of the Persian language; [followed by a verse by Salman].78 (Sharafnāma, 2006, vol. 1, p. 142) (p.160) Zāwar: … with fatḥa on the third letter, the planet in the third heaven which is ascribed to the fifth region, its home is the house (burj) of Taurus and Gemini. She is called a musician of the heavens and she is also called Bīdikht and Nahīd. Astrologers call [the planet] Sa‘d aṣūsh. In Arabic she is called Zuhra and so it is written in the Sharafnāma. (Adāt al-fużalā, Or. 1262, f. 39a)79 But dictionaries also paid attention to contemporary terms for musicians: Paikūb: a dancing girl whom Indians call pātar/pātur. (Adāt al-fużalā, Or. 1262, f. 11a) Indo-Persian lexicographers detailed a few names of ancient tunes,80 and also of Turki ones, especially in the seventh bakhsh of the Zafāngūyā that was dedicated to Turki terms, where generic Turki terms for song, singing, and so on, are given: Īr: a song. (Zafāngūyā, 1997, seventh bakhsh, vol. 2, p. 146)81 (p.161) Ūn: a voice (āvāz). (Zafāngūyā, 1997, seventh bakhsh, vol. 2, p. 151) In some entries of the Miftāḥ al-fużalā, Indian musical terms are given without any indication of the language of origin—in this case possibly the Indic ‘dhun’? Duna: with zamma on ‘dāl’, a tune. (Miftāḥ al-fużalā, Or.3299) Most definitions for musical terms in Indo-Persian dictionaries (over 300 in total) are of musical instruments. Many of them are very simple and short. For example:

Page 20 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Gāvdum: a karnāi, that is, a būq, also called small būq, which is like a cowtail. (Zafāngūyā, 1997, second bakhsh, vol. 2, p. 39) Sitār: with kasra, the name of a musical instrument that is also called sitāra and sehtār. (Sharafnāma, Add. 7678, f. 239b) Occasionally, lexicographers provided descriptions, sometimes very detailed ones, of a particular musical instrument, with information about its function, appearance, structure, parts. They also often described the material the instrument was made of and how it was played: Khar-i-rabāb and kharrak-i rubāb: a wooden tool that is [placed] on the belly of the rabāb and [instruments] like it. One pulls the strings over it [followed by a verse by Sipahani]. (Sharafnāma, vol. 1, p. 377) Arghanūn: an instrument of the Rumis [that is, Byzantine/Greeks], [also] the family of musical instruments such as Rabāb, Barbat, Chang, Tanbūr; also pronounced Arghun. (Zafāngūyā, sixth bakhsh: Rumi and Yunani, 1997, vol. 2, p. 139)82 (p.162) Several terms relating to musical instruments are very rare and cannot be found anywhere apart from Turkic dictionaries written in India in the eighteenth and nineteenth centuries.83 Even in contemporary Khorasan, poets writing in Turki like ‘Ali Shir Nava’i used popular Persian equivalents for those instruments rather than the ones given below: Aylaq/Ablaq: kamāncha [a string/violin-like instrument].84 (Zafāngūyā, seventh bakhsh, 1997, vol. 2, p. 148) Oūparim/Oubarim: chang, a [stringed] musical instrument.85 (Zafāngūyā, seventh bakhsh, 1997, vol. 2, p. 151) Chughrah: a musical instrument that is called chārtārah. (Zafāngūyā, seventh bakhsh, 1997, vol. 2, p. 161) (p.163) Other Turkic musical instruments given in Zafāngūyā, instead, were as popular then as they are now: Sīrghū: a windpipe.86 Page 21 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (Zafāngūyā, seventh bakhsh, 1997, vol. 2, p. 164) Sarghīn: with fatḥa and kasra, a Turkic flute (windpipe) that is also called surnāy and surnī [verse follows].87 (Sharafnāma, Add. 7678, f. 252a) The chaghāna, which in ‘Ali Shir Nava’i’s Turki poetry is often accompanied by a qanun, and in some modern dictionaries is referred to as both a bowed and drummed instrument, seems to refer to different instruments in Indo-Persian dictionaries. Most of them show two kinds of instruments, one stringed and played with a bow (as pictured in the Miftāḥ al-fużalā), and the other like a cymbal: Chaghānah: with Persian ‘jim’, a kind of instrument which musicians play with bow [while they sing a song—Or. 265]. (Adāt al-fużalā, Or. 1262) According to the Miftāḥ al-fużalā, the chaghānah was like a small rabab, played with a bow, while the Sharafnāma compared it to the Indian surmandal.88 Interestingly, already at this time some authors were attempting to compare Persian and Turkic musical instruments with their nearest Indian equivalents: Karnāy: a būq that in Hindavi is called bher.89 (Zafāngūyā, 1997, second bakhsh, vol. 2, p. 38) (p.164) Often the same Indian equivalent was given for several instruments: bher, for example, was given as equivalent not only for the būq and the karnāy, but also for the nāqūr—in other words, for all kinds of windpipes.90 This trend towards setting up equivalence between Persian, Arabic, and Turki terms and Hindavi ones extended to other kinds of entertainment.91 This was, of course, within the very nature of multilingual Persian dictionaries that encompassed different cultural and historical zones. But it may also indicate a coming together of musical styles in the cosmopolitan and mixed musical culture of Sultanate north India that music historians have already remarked upon.

Conclusions What is the evidence about the readership of these dictionaries? Apart from the lexicographers themselves, internal clues suggest a readership of poets and literate elites who needed and wanted to read Persian works, which themselves contained and assimilated words from many languages. At the same time, the attention towards everyday useful words and objects, and to specific vocabulary related to food, plants, arms, music, astronomy, and so on, in Arabic, Turki, and Hindavi suggests a linguistically heterogenous court and literate society that Page 22 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian was interested in both trade and in a wide range of worldly pursuits, and that was keenly aware of the Central Asian world. What about the physical clues of the dictionaries as objects? Unlike the monumental dictionaries of the Mughal period,92 some of the Sultanate dictionaries, while carefully (p.165) calligraphed and sometimes illustrated (like the monumental Miftāḥ al-fużalā), were pocket- or small-sized, indicating a practical use. Dictionaries fulfilled several different functions: they were aids for Persian literature and culture, manuals for widening one’s knowledge of Hindavi, and also perhaps useful books for merchants. While the early Persian dictionaries written in Sultanate India prepared the ground for the large multilingual dictionaries written in the Mughal period, the Turki section of the Zafāngūyā and the repertoire of Turki lemmata in the other early farhangs became the basis for Turkic dictionaries written in India in later centuries, which contained neither Eastern nor Western Turkic but ‘Indian Turki’. Thus, Turki dictionaries written in India in the eighteenth and nineteenth centuries include words that are not found in Western or Eastern Turkic languages. Other words, like ikdish mentioned above, acquired new, specifically Indian, meanings; the pronunciation also changed for many words, and diacritical marks show that they came closer to Indian phonology. Overall, we can speak of a gradual process of assimilation. Indo-Persian multilingual dictionaries, thus, show us several distinct phenomena and processes. First, they show a keen and continuing awareness of both written and oral multilingualism, both in India and more generally in the Persianate world they drew upon—and within Persian itself. Second, they set up equivalences between Persian (and Arabic), Turki, and Hindavi at all levels, from material culture to astronomy. Third, they reveal an increased tendency to rely upon Hindavi as the language for explanations, thus suggesting that this was the language readers were familiar with. Fourth and final, they preserved Turkic linguistic consciousness, albeit in a way that reveals a gradual tendency towards the creation of a specific ‘Indian Turki’. Notes:

(1) Amir Khusraw Dihlawi wrote in his masnavi Nūh Sipihr (The Nine Skies, 1318): Next comes the Persian of the Persian people. It is a very sweet language with its centre around Shiraz. The Persian spread from there and became a repository of learning of the world. It shone over every city like [the] moon. Next comes Turkī, which was the language of the Turks of Ataz and of the Tribes of Qanqalī, Auyāghar, Īratī, and Ghuzz. It originated in the plains of Khafchāq (Qipchāq) and Yamak. Thence it spread to other countries, like the salt of India. Though other languages (of the world) are also very

Page 23 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian beautiful, they are neither as sweet or enlightening (as Arabic, Persian, and Turki). Translated by R. Nath and Faiyaz ‘Gwaliari’, in India as Seen by Amir Khusrau (in 1318 A.D.), Jaipur: Historical Research Documentation Programme, 1981, p. 72, lines 33–7 ff. He also wrote: ‘Persian is famous for its sweetness. There is the Persian saying: “Arabic is knowledge and learning, Turki is Art [though mastery is a more correct translation], Persian is sweet, all other languages are bad and worth nothing.”’ Nath and Faiyaz (1981), p. 77. In Central Asia there is a popular saying: ‘Arabic is the language of knowledge, Persian is the language of poetry, Turkic is the language of war (warriors).’ Already in the thirteenth century Arabic, Persian, and Turkic were the literary languages of the Muslim world. (2) For examples, the Turki sections in Persian farhangs can be called lughat because they do not include explanations, only equivalents. (3) See S.I. Baevskii’s fundamental research on early Persian dictionaries originally published in Russian in 1989, and in English translation by J.R. Perry as Early Persian Lexicography: Farhangs of the Eleventh to the Fifteenth Centuries, Folkestone: Global Orient, 2007. (4) Baevskii (2007), p. 128. (5) Baevskii (2007), especially pp. 173–4 for this point. (6) Persian farhangs from the fifteenth century onwards are characterized by the increasing volume of their vocabulary: while, for example, the Lughat-i furs had 1,700 lemmata and Mi‘yār-i Jamālī had 1,600, the Farhang-i zafāngūyā contained approximately 5,170 lemmas, the sixteenth-century dictionary Tuḥfat al-sa‘ādat about 14,000 lemmas, and the Burhān-i qāṭi‘ (Decisive argument, 1652) 20,000. (7) The first Persian dictionary compiled in India, the Farhang-i-Qawwās (also called Farhangnāma, Farhang-i Fakhr-i Qawwās, Farhang-i panjbakhshi, Farhang-i Shāhnāma) written by the poet Fakhr al-Din Mubarak Shah Qawwas Ghaznavi between the end of the thirteenth century and the beginning of the fourteenth, falls outside the purview of this volume, though it was used by later lexicographers such as the authors of the Dastūr al-afāżil, Adāt al-fużalā, Zafāngūyā, Baḥr al-fażā’il, Sharafnāma-yi Munyārī, Mu’ayyid al-fużalā, Farhangi-Jahāngīrī, and so on. The Farhang-i Qawwās is preserved in a single manuscript at the Asiatic Society, Kolkata, and was described by W. Ivanow (Concise Descriptive Catalogue of the Persian Manuscripts in the Curzon Collection, Asiatic Society of Bengal, Calcutta: Baptist Mission Press, 1926) and edited by Nazir Ahmad, Tehran: Bungah-i Tarjumah wa Nashr-i Kitab, 1974; see Baevskii (2007), pp. 71–7. Page 24 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (8) Dastūr al-afāżil fi’l-lughāt al-fażā’il, Asiatic Society, Kolkata, Ms. 517; see W.I. Ivanow (1924), p. 676; edited by Nazir Ahmad, Tehran: Bunyad-i Farhang-i Iran, 1973; see Baevskii (2007), pp. 77–83. It was used by the authors of Adāt alfużalā, Farhang-i zafāngūyā, Baḥr al-fażā’il, Sharafnāma-yi Munyārī, Tuḥfat alsa‘ādat (1510), Mu’ayyid al-fużalā (1519), Madār al-afāżil (1592), Farhang-i Jahāngīrī (1609). (9) Adāt al-fużalā, British Library, Mss. Or.1262 (copy dated 1691) and Or.265 (n.d.), fols. 2–60; C. Rieu, Catalogue of the Persian Manuscripts in the British Museum, London: British Museum, vol. 2, 1881, pp. 491–3; see also Baevskii (2007), pp. 87–94. It was used by the authors of the Sharafnāma-yi Munyārī, Mu’ayyid al-fużalā, and Madār al-afāżil. (10) Nazir Ahmad, Farhang-i zafāngūyā u jahānpūyā of Badr al-Din Ibrahim (1433), Khuda Bakhsh Library Journal, 55–6 (1990 and 1997): 1–173; see Baevskii (2007), pp. 94–104. According to Baevskii (2007, p. 99), this ‘became the authoritative lexicographical work’. It consists of seven parts (bakhsh), including Pahlavi, Dari, Arabic, Rumi, and Turki words. (11) Mujmal al-‘Ajam, British Library, Or. 265, fols. 62–161 (copied approx. seventeenth century); see Rieu (1881), p. 493. Baevskii notes that [n]o information about the author, ‘Āsim Šu’ayb ‘Abdūsī, has survived. Like other authors of medieval farhangs, Āsim Šu’ayb states in his preface that he compiled his dictionary at the request of friends who were having difficulty reading poetical texts in which they came upon unfamiliar words ‘in Persian, Pahlavi, Rūmī, Nabāṭī (i.e. Aramaic), and Turkī’. The Mujmal al-‘Ajam consists of two parts (qism): Part One defines simple words, and Part Two defines compound words. Each part is ordered alphabetically by initial (bab), then by final. The gloss is typically illustrated with numerous verse quotations, with no indication, however, of the names of the poets. For a number of entry words, the compiler of the farhang furnishes Indic equivalents. Baevskii (2007), pp. 113–14. (12) Baevskii notes that the Sharafnāma-yi Munyārī contains ‘… a lengthy author’s preface (muqaddima) of a lexico-grammatical character’ where the author explains the use of some suffixes and enclitics. The dictionary contains thirty-one letter chapters (bab), ordered alphabetically by initial, and further subdivided by final letter (faṣl). The glosses define general Persian vocabulary chosen from the works of early Persian language poets, and includes many names. There are also a few Arabic words, unlabelled. At the end of some of the subdivisions, the lexicographer glosses Turkish words, labeled as Turki. Indic equivalents are occasionally given in the text of the definition. The

Page 25 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian pronunciation of entry words (for only some of the vocabulary) is provided in longhand. Baevskii (2007), p. 111. He also remarks that the ‘definitions of a considerable portion of vocabulary entries are illustrated by supportive verses’, either of ‘classical’ Persian poets (Unsuri, Sana’i, Anvari, Khaqani, Sa‘di, and Hafiz) or of his own; every bab is prefaced by a qasida of his own composition (p. 113). There are several manuscripts of this work: for example, British Library, Add. 7678 (copied approx. seventeenth century); Rieu (1881), pp. 492–3. Another dictionary mentioned by Baevskii that I have not been able to see is the Baḥr al-fażā’il fi manāfi‘ al-afāżil (The sea of scholarship for the benefit of scholars), compiled in India in either 1433 (according to Nazir Ahmad, reading 837H) or 1336–7 (according to Ivanow, reading 737H) by Muhammad ibn Qawwam ibn Rustam ibn Ahmad Balkhi. Balkhi also wrote a commentary (sharh) on Nizami’s masnavi Makhzan al-asrār around 1393. His dictionary gives definitions of mainly Persian but also Pahlavi, Arabic, and Greek words. Baevskii notes that so much of the vocabulary is nonPersian that the work might well be classified as a multilingual or an ArabicPersian dictionary (p. 105). (13) An example from Zafāngūyā (fifth bakhsh: Miscellaneous words of Arabic, Nabatean [i.e. Aramaic], and Dari, vol. 2, pp. 125–6): Quqnus [= phoenix]: with zamma [i.e. ‘u’] on qāf and sometimes spoken with qāf and vāv [as] Qaqnūs, and sometimes [it is pronounced] with fatḥa [i.e. short ‘a’] on fā [as] Qāfanus, that is, an animal with a pleasant voice, named ‘Thousand Voices’, and it has come to us that it does not have a companion and gives birth once every thousand years; when its time comes [to die], it starts singing in so many variously coloured notes (tunes) and becomes inebriated and dances, to the point when fire lights from its body and it begins to burn, and from its ashes an egg emerges, and from this egg another identical animal rises with the blessing [order] of God; wise men have taken the knowledge of music from it. The same definition is given in Mu’ayyid al-fużalā, whereas the Ghiyās al-lughāt (1887) maintains that it is a Greek word (in Arabic transcription) and says that when the Quqnus becomes very old … It starts singing in different tunes, one of which is named in India ‘Dīpak’, and from that wood near it, it spontaneously ignites and the bird is burnt to ashes. (14) One example from the Sharafnāma-yi Munyārī: Baḥrām: with fatḥa; the name of the Padishah of Iran—some of whose qualities have been told in the word ‘Baḥrām Gōr’ in Chapter ‘Ra’; and also the name of a major general of Hormuz bin Noushirvan nicknamed Bahram-i Chōbīn …; and [it is] the name of the son of Goudarz, the Hero of Keykavus; and also the name of a planet on the fifth heaven which is a Page 26 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian guardian of the heavens; it has a house in the signs Aries and Scorpio. Sharafnāma-yi Manyarī [or Munyarī] yā Farhang-i Ibrahimī, Hakima Dabiran (ed.), Tehran: Pizhuhishgah-i ‘Ulum-i Insani va Mutala‘at-i Farhangi, 2006, vol. 1, p. 186. (15) For example, ‘Banj: rhyming with ganj, the same as batanj, thus in the Lisān al-shu‘arā’. Sharafnāma (2006), p.139. (16) From Adāt al-fużalā: ‘Aywān: roof, terrace, a high place for sitting’; ‘Khān: a room/house (khāna), a caravanserai, the name of the ruler of Samarqand, the ruler of Turkistan’. (17) From Zafāngūyā: ‘Āshnā (= friend): is the opposite of stranger’; ‘Zinda (= alive): is the opposite of dead’; ‘Tang (= narrow): is the opposite of wide’. (18) Fakhr-i Qawwas composed his dictionary, the Farhang-i-Qawwās, mainly for use in reading the Shāhnāma. For this reason the sources sometimes refer to it as the Farhang-i Shāhnāma. It should be emphasized here that these were the considerations that defined the essence of medieval Persian lexicography for centuries, as the following definition shows: ‘Azhdahā, Azhdahā: with Persian “zh”, male snake, big and very famous, also called Azhdar and Barghimān; in Arabic [Tazi] Tinnīn and Thu‘bān; and in the Shāhnāma it means the most powerful Sultans like Zahhak and Afrasyāb; and it is also the name of a science which has an image [figure] of an enormous snake’. Sharafnāma (2006), vol. 1, p. 26. Or: ‘Tūr: the land of Turān; Hakim Firdawsi has said: [verse follows]’; Adāt al-fużalā, Or.1262. (19) Baevskii (2007), p. 115. (20) For this reason, they have been used by historians of science in India; Irfan Habib, for example, notes that the illustration in the Miftāḥ al-fużalā provides the earliest evidence of the use of the Persian wheel in India. I. Habib, Medieval India: A Study in Civilization, New Delhi: National Book Trust, 2007, p. 67. (21) ‘Bāqlah: this is a grain which is not grown in India’. Sharafnāma (2006), vol. 1, p. 204. ‘Shāh: well-known, a broad avenue … and the name of an animal which lives in India’. Sharafnāma, Add. 7678, British Library. (22) About the poets whose verses are cited in these dictionaries, Baevskii notes that the author of the Fakhr-i Qawwās, besides Firdawsi, quoted from IndoPersian poets Mas‘ud Sa‘d Salman, Taj-i Riza, ‘Awfi, Shihab-i Mahmara, and Nasiri. Baevskii (2007), p. 73. He remarks that the Dānishnāma-yi Qadar Khān (Book of knowledge [in honor] of Qadar Khan) by Furughi quotes more than seventy-two poets of the tenth to fifteenth centuries from Transoxiana and Khorasan (and Mas‘ud Sa‘d Salman and Amir Khusraw from India [p. 86]), and that many of the verses attributed to a mysterious ‘Buzurgī’ (i.e. a great/old Page 27 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian poet) were in reality the creation of many famous poets of different epochs and diverse literary circles (pp. 168–70). See also Chapter 6 in this volume. (23) Diacritics (harakat) for short Arabic short vowels include fatḥa, a short diagonal line placed above the letter, for a short ‘a’, the kasra, a short diagonal line below the letter, for the short ‘i’, and the zamma, a small curl (vāv) above the letter, for the short ‘u’. (24) The question of correct pronunciation increased in importance for Indian speakers of Persian, especially from the seventeenth century, when in Iran the so-called majhul vowels (i.e. ê and ô) began to collapse with î and û, whereas in India and Central Asia the old values were preserved. In 1708 Amin Allah wrote a pronunciation dictionary in verse, the Dafi’ al-lughāt or Defender from errors, on the correct pronunciation of common Persian words that were often mispronounced; C.A. Storey, Persian Literature: A Bio-Bibliographical Survey, vol. 3, part 1, London: Royal Asiatic Society, 1984, p. 37. In early Persian dictionaries, harakat were also written for rare words: e.g. ‘Bayāganam: with Persian kāf [i.e. gāf], it is barkunam’; Adāt al-fużalā, Or.1262. (25) For another example, ‘Azukh: a piece of flesh that grows on the nose of men and [is] black on top; some cut it, and Arabs call it sūlūl and Indians name it massa’. Adāt al-fużalā, Or.1262; see also J.T. Platts, A Dictionary of Urdū, Classical Hindī, and English, New Delhi: Munshiram Manoharlal, 1997, p. 1029. The harakat here is on the letters mim and alif, and a tashdid on sin. (26) ‘Korasht: with Persian vāv, daymīn (‫ )? ﹶﺩﯾﻤﹻﯿﻦ‬wood with which children play, and in Hindavi it is called danda muvahi; also spoken with Persian kāf’ [i.e. gorasht]’. Muhammad ibn Da’ud ibn Muhammad ibn Mahmud Shadiabadi, Miftāḥ al-fużalā, British Library, Or. 3299, f. 220b (only known copy); C. Rieu, Supplementary Catalogue of the Persian Manuscripts in the British Museum. London: British Museum, 1895, p. 116; see also Platts (1997), p. 567. (27) See Baevskii (2007), pp. 128, 132. (28) The author of the Farhang-i Qawwās mentioned that his friends desired to fully understand the Shāhnāma but were unable to do so. They asked him: ‘Collect [words] of Persian and Pahlavi all together, and compile a dictionary so that everyone may benefit from these languages.’ Accordingly, the author says, ‘I resolved to bring together all the dictionaries. First of all, [I took] the Shāh nāma, which is a king among books, and read it from beginning to end. I jotted down the Pahlavi words, each one of them on a separate sheet of paper, then divided them into bakhsh, gūna, and baḥr’. Farhang-i Qawwās, ed. N. Ahmad, Tehran: Bungah-i Tarjumah va Nashr-i Kitab, 1974, pp.1–3; my translation, but see also Baevskii (2007), p. 132. (29) Baevskii (2007), p. 135. Page 28 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (30) Dastūr al-āfāżil (1973), p. 57, translated by Baevskii (2007), p. 81. See also the Zafāngūyā, in which Badr al-Din writes: I accepted their request. Taking up my pen and without regard to fatigue, I collected word after word. It was my firm intention to consolidate all the dictionaries and to bring together all the scattered words. And all the words that there are in Arabic, Persian, Pahlavi, Dari, Yunani [Greek], Turkish, and other tongues, I examined individually, and one by one put them in a particular place. All that I was able to discover, all that I heard and saw, I put in order; each vocabulary I compiled separately, dividing the farhang into bakhsh, gūna, and baḥr. And when I was arranging them, I found the most appropriate place for each word. I aimed at brevity and not verbosity, at utility and not diversion. Zafāngūyā (1989), vol. 1, pp. 143–6. (31) Baevskii (2007), p. 111. (32) For example, Farhang-i Ja‘farī (1630–1), the Lughat-i furs, the Farhang-i zafāngūyā, the Siḥāḥ al-furs; Baevskii remarks that ‘attempts at grammatical explanations of the material, though quite cautious and rare, are evident in all the Persian farhangs of the eleventh to the fifteenth centuries. However, the lexicographer’s level of interest in the grammatical aspects is far from uniform’ (2007, p. 151). See also his ‘Pervye opyty grammaticheskogo osmysleniia iazyka farsi (XI–XV vv.)’ [The first attempts at a grammatical conception of Persian], PPiPIK.XVIII Godichnaia nauchnaia sessiya LO IVAN SSSR, Part 2, Moscow, 1985, pp. 115–18. (33) Baevskii also noted it (2007), pp. 129–31. (34) Adāt al-fużalā, Or. 1262. (35) Zafāngūyā, fifth bakhsh, vol. 2, p. 106. See also Adāt al-fużalā, Or.1262: ‘Khafchāq: with Persian ‘jim’, a desert [wilderness] which is named Dasht-iKhifchāq and is the origin of the name of the Turks named khifchāqīs’; the Dīwān lughāt al-Turk (11th century) gives: ‘Qifjāq: 1. a tribe of Turks; 2. the name of a place near Kashghar’. Mahmud Kashghari, Dīwān lughāt al-Turk, Kazakh edition in one volume, Russian translation with introduction by Z.A. Auezova, Almaty: Daik Press, 2005, p. 440. (36) Zafāngūyā, 1997, second bakhsh, vol. 2, p. 44. See also ‘Badakhshān: the name of a province’ (Adāt al-fużalā, Or. 265); ‘Bukhārā: the name of city, one of the famous cities in the Upper Region [i.e. Transoxiana]’ (Adāt al-fużalā, Or. 1262); ‘Bukhārā: with zamma, the name of a great city, there is not a better place like it in the world’; Sharafnāma-yi Munyarī (2006), vol. 1, p. 126. (37) Zafāngūyā (1997), second bakhsh, vol. 2, p. 3. (38) Zafāngūyā (1997), second bakhsh, vol. 2, p. 26. Page 29 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (39) Sharafnāma (2006), vol. 1, p. 27. (40) Sharafnāma (2006), vol. 1, p. 84. (41) Sharafnāma (2006), vol. 1, p. 64. (42) Zafāngūyā (1997), second bakhsh, vol. 2, p. 21. (43) Sharafnāma (2006), vol. 1, p. 67. (44) Zafāngūyā, vol. 2, second bakhsh, p. 9. (45) Baevskii (2007), Chapter 4. (46) Muhammad ibn Sulayman, who used the pen name Fuzuli (1498–1556) wrote verses with equal ease in Arabic, Persian, and Azeri Turki. Many of his poetic and prose works are preserved in Indian libraries such as the Asiatic Society, Kolkata, Khuda Bakhsh Oriental Public Library, Rampur Raza Library, and the Osmania University Library. (47) Kashghari, Dīwān lughāt al-Turk, Russian trans. Z.A. Auezova, Almaty: DajkPress, 2005. (48) In the Muḥākamat al-lughatayn, ‘Ali Shir Nava’i wrote about the richness of the Chaghatai language. Nava’i also compared Chaghatai Turkic and Persian and argued that Chaghatai was a very useful language in many ways that poets could successfully use it for poetry alongside Persian; ‘Ali Shir Nava’i, Muḥākamat, R. Devereux (ed. and trans.), Leiden: Brill, 1966. (49) It is worth noting that, apart from Babur’s memoirs and his Diwan, Turki literature written in India still remains uninvestigated. Though my personal experience is that there are not many Turki manuscripts extant in Indian libraries, even those have not been catalogued. Many manuscripts surviving in a single copy are often not dated and this makes it hard to determine the context of their composition. For instance, the dictionary Āmad-nāma-i Turkī preserved in the Rampur Raza library appears to be very old (late-fifteenth or earlysixteenth century), but we cannot tell exactly when this dictionary was written. Benedek Peri of Budapest University, and Uzbek scholars like Aftandil Erkinov and N. Nizamiddinov are currently working on Turki in Mughal times. (50) For Turki lemmata in the Dastūr al-afāżil (occasionally labelled by region) and Adāt al-fużalā, see Baevskii (2007), pp. 78–9 and 92–3. Muzaffar Alam’s claim that ‘[i]t was only by the late fifteenth century that separate sections for Turkish words in Persian began to appear, as in Shaikh Ibrahim Qavvam Faruqi’s Sharaf-namah-i-Manerī (1472)’. ‘The Culture and Politics of Persian in

Page 30 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian Precolonial Hindustan’, in Literary Cultures in History, S. Pollock (ed.), Berkeley: University of California Press, 2003a, p. 149 stands to be corrected. (51) See Baevskii (2007), p. 111. Arabic, Turkic, Chinese, and Indic words that had become fully assimilated into Persian were often not labelled etymologically though. (52) That the author of Sharafnāma used the Zafāngūyā for his al-Turkī section is clear from the fact that he mentioned it as a source, though he did not reproduce the section entirely. (53) Cf. ‫ ﹸﺍﮐﺴﹹﺰ‬uksuz: ‘orphan, confused’. The initial form of this [word] is öksuz [‘muddle-headed’], which is formed from ök ‘intellect, comprehension’ Dīwān lughāt at-Turk (2005), vol. 3, p. 127. (54) For example, ‘Fallā: yoghurt or sour cream made from milk: in Turki they call it qaymāgh’. Majmū‘at al-furs, cited in Baevskii (2007), p. 65. (55) In fact, Turki-Farsi dictionaries were compiled from the seventeenth century onwards, such as after Fazl-allah Khan Barlas’s Farhang-i zabān-i turkī, Khwaja Amir’s Tālīf-i al-amīr (Rampur Raza Library, autograph Ms. 2501, 31 fols.), the Risāla-i lughāt-i turkī by Mirza Qatil (Khuda Bakhsh Oriental Public Library, Patna [India], No. Acc. 1934, 18 fols.); see D. Karomat, ‘Turkish Manuscript Dictionaries in India’, in Nażr-e Khudā Bakhsh, Patna: Khuda Bakhsh Fellows, 2001, pp. 115–26 (in Urdu). (56) For example, ‘Ay [‫]ﺍﯼ‬: moon [mahtāb]’ (Zafāngūyā, 1997, seventh bakhsh, vol. 2, p. 145); or ‘Āy [‫]ﺁﯼ‬: moon [mah]’ (Sharafnāma [2006], vol. 1, p. 120); or ‘Bitik [‫]ﺑﺘﮏ‬: [hand-writing, deed]’ (Zafāngūyā [1997], vol. 2, seventh bakhsh, p. 89). (57) See V.A. Kapranov, Tadzhiksko-Persidskaia leksikografiia v Indii XVI-XIX vv. (Tajik-Persian lexicography in India in the sixteenth to nineteenth centuries), Dushanbe: ‘Donish’, 1987. (58) Sharafnāma (2006), vol. 1, pp. 14–19. (59) Sharafnāma (2006), vol. 1, p. 19. (60) Even dictionaries written in India gave examples of the Persian dialects of Bukhara, Shiraz, Nishapur, Farghana, Transoxiana, Gilan, Azerbaijan, and so on. For example, ‘Dādar [= dodar]: in the language of Transoxiana it is a brother’ [Dastūr al-afāżil]; ‘Dādar: in the language of Transoxiana it is a brother. In Khorasan it is [pronounced] with fatḥa on ‘dāl’ and some people pronounce it as wādar’. Adāt al-fuzalā, Or. 1262, f. 24b. (61) ‘A porcupine; a hedgehog (= sihi, q.v.)’. Platts (1997), p. 715. Page 31 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (62) There are more in Or.1262 than in Or. 265. For example, ‘Dasmah: a kind of grain’ (Adāt al-fużalā, Or. 265, f. 25b) and ‘Dasmah: a kind of grain called shakhil, Indians call it arhar’ (Adāt al-fużalā, Or. 1262, f. 34b). (63) To be exact, 133 in the copy of the Khuda Bakhsh Library. See Ahmad (1990, 1997); ninety-two in the second to sixth bakhshs; the third (Pahlavi and Dari) and seventh (Turki) bakhshs do not have any Indic equivalents at all. (64) According to the Madār al-afāżil, ‘it is a kind of arrow for hunting’; ‘Fīlak is a Badakhshanian arrow’ (Lughāt-i furs); Platts defines it thus: ‘A small mattock or hoe, a spade; belok, s.f. The iron point of an arrow’ (1997), p. 210. (65) For a discussion of the kinds of knowledge that Indo-Persian elites needed to know, and their interest in the Indian equivalents, see Emma J. Flatt, ‘Courtly Culture in the Indo-Persian States of the Medieval Deccan, 1450–1600’, PhD thesis, SOAS, University of London, 2009; for music, see Katherine Butler Schofield, ‘The Mughal rasika’, forthcoming in Tellings and Texts: Music, StoryTelling and Performance in South Asia, F. Orsini and K. Schofield (eds). (66) Also in Adāt al-fużalā, Or. 265; the Sharafnāma calls it kanīr [‫]ﮐﻨﯿﺮ‬. (67) The other copy of the Adāt al-fużalā (Or. 265, f. 6b) and Sharafnāma do not give the Indic equivalent. (68) Also in Miftāḥ al-fużalā, Or. 3299. ‘The plant or root sedoary, Curcuma serumbet’. Platts (1997), p. 819. (69) ‘bakkam, s.m. Sappan or red wood (= bagam, q.v.)’. Platts (1997), p. 100. (70) See Norah Titley (trans.), The Ni‘matnāma Manuscript of the Sultans of Mandu: The Sultan’s Book of Delights, London: Routledge, 2005. (71) See also Adāt al-fużalā (Or.1262, f.18b): ‘Panj pāyak: both “be” are Persian, an animal that lives in the water, but if it falls onto the ground it can also walk with its own legs. Arabs call it saratān, Persians ghaltānag and Indians call it kēkra.’ (72) Cf. ‘H. sadāsī, H. sandsā s.f…. A pair of pincers or nippers, small tongs, forceps (see sanrsī)’. Platts (1997), p. 683. See also ‘Sindān: one of the tools of blacksmiths with which they hit the iron. Indians call it nihāyī (= anvil).’ Sharafnāma, Add.7678, f.252b. (73) Puvta in Adāt al-fużalā, Or.1262. See also ‘Chūbah: with Persian “be”, a wooden tool with which one makes bread, in India they call it bēlan’. Sharafnāma (2006), vol. 1, p. 359. And ‘Duwshah: a vessel with which one milks, and in Hindavi it is called duhnī’. Adāt al-fużalā, Or.1262, f.36a.

Page 32 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (74) ‘S. kanthā, s.f. (m.?) A patched cloth or garment, a covering of rags and patches (such as is worn by jogīs and faqīrs; syn. gudrī); a beggar’s wallet and band’. Platts (1997), p. 851. (75) For example, ‘Khwāre [‫]ﺧﻮﺍﺭﻩ‬: in the Afghan language, the word for food’. Adāt al-fużalā, Or. 1262. (76) Madār al-afāżil gives both sōgan [‫ ]ﺳﻮﮐﻦ‬and saut [‫]ﺳﻮﺕ‬, Adāt al-fużalā (Or. 1262) saukan. (77) For example, ‘Rasan bāz: a clan of jugglers who walk the tightrope’. Adāt alfużalā, Or. 1262, f.37b; ‘Indians call them bīdnī’. Adāt al-Fużalā, Or. 265, f.27b. (78) Barbad Marvazi (sixth or or seventh century) was the musician of the Sassanid king Khusraw II Parvez (590–628), and his name appears in all fifteenth-century dictionaries, together with melodies that were supposedly invented by him: for example, ‘Shādravān: flatweave blanket which is hung over a doorway; a wardrobe; name of a melody’, Dastūr al-afāżil. ‘Khusrawānī: a kind of rhymed songs that Barbad, the musician of King Khusraw, sang’; Sharafnāma (2006), vol. 1, p. 415, and Sharafnāma, Add. 7678; and ‘Rāh-i Khusrāwānī: a rhythmic song invented by Barbad, the musician of Parvez, and which he called Khusrawānī’; Sharafnāma (2006), vol. 1, p. 497. See also Zafāngūyā (1997), second bakhsh, vol. 2, p. 6; the term occurs also in Miftāḥ al-fużalā, Adāt alfużalā (Or. 1262 and Or. 265) See also N.N. Negmatov (ed.), Borbad: His Epoch and Cultural Traditions, Dushanbe: ‘Donish’, 1989 (in Tajik); and A. Radjabov, Az tarikhi afkori musiqii Tojik (asrhoi XII-XV), Dushanbe: ‘Donish’, 1989 (in Tajik). (79) ‘Zuhra (=Venus): … and with zamma, it is also the planet which is a musician of the Heavens, [verse]’; Sharafnāma (2006), vol. 1, p. 530. (80) ‘Ārāyish-i Khūrshīd: the name of a melody and tune’; Sharafnāma (2006), vol. 1, p. 41; ‘Bād-i Nawrōz: well known and also the name of a melody and tune’; Adāt al-fużalā, Or. 1262; Sharafnāma, vol. 1, p. 158. (81) Also declined as ‘īrān: he/she is singing a song’, and ‘īrlādī: he/she had sung a song’; Zafāngūyā, seventh bakhsh (1997), vol. 2, pp. 151, 153. For a detailed comparison of this term in various Turkic languages, see E.N. Nadzhip, Istorikosravnitelnyi slovar tiurkskikh iazykov 14 v., na materiale ‘Khosrau Shirin’ Kutba [The Historical Comparative Dictionary of Turkic Languages of the 15th Century, based on Kutba’s Khosrau and Shirin], Moscow: Glav. red. vostochnoi lit-ry, 1979, vol. I, pp. 417–18; and my own article ‘O tiurkskih muzykalnyh terminah v istochnikah XI-XIX vekov’ [About Turkic Musical Terms in Manuscripts of 11th-19th Centuries], published in O’zbekistonda Ijtimoiy Fanlar, 2000(1): 66– 70.

Page 33 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (82) This word is given in the Zafāngūyā in the special bakhsh dedicated to Greek words. Greek culture and language were assimilated into Iranian and Indian culture in two waves: first with the Graeco-Bactrian kingdom, covering the Central Asian regions of Bactria and Sogdiana from 250 to 125 BCE, and the Indo-Greek expansion into northern India from 180 BCE to 10 CE. The second wave came after the Arab conquest through the Arabic language and often in Arabic form. With the term ‘Rum’ lexicographers indicated the Byzantine Empire, and the category ‘Rūmī’ included words borrowed from Western languages. In his introduction to the Farhang-i zafāngūyā, Baevskii noted: It is very interesting how the selection of words was done, a selection that includes a range of terms and symbols related with Christianity which were in greatest common use in Persian literature (prose and poetry). One can suppose that the material for the author of this dictionary came to a considerable extent from works of the famous poet Khaqani (1120–99), in whose poetry one frequently encounters motifs and terms related to Christianity. Badr al-Din Ibrahim, Farhang-i zafanguia va dzhakhanpuia (Slovar’ govoriaschii i mir izuchaiushchii). Faksimile rukopisi, izdanie teksta, vvedenie, spisok tolkuiemykh slov, prilozhenia S.I. Baevskogo, S.I. Baevskii (ed.), Moscow: Nauka, 1974, pp. 59–60 (my translation from Russian). (83) In particular, the Turkic dictionary Tālīf-i al-Amīr by Khwaja Amir (approximately eighteenth century). (84) The same definition is given in Mu’ayyid al-fużalā. Tālīf-i al-Amīr by Khwaja Amir and the Risāla-i lughat-i Turkī by Mirza Qatil both have amlaq. (85) The same in Mu’ayyid; Sharafnāma has Ōbarim; Tālīf-i al-Amīr and Risāla-i lughat-i Turkī have Ōba’rim = chang (= harp). See my article ‘O tiurkskih muzykalnyh terminah’ (2000): 68. Nadzhip’s article cites a similar word with the same meaning ‘Üjrim—harp, lyre’, with the remark that this word is not found in historical sources, dictionaries, or modern languages. E. Nadzhip, ‘Tiurkskii iazyk deliiskogo sultanata XIV veka’, The Journal of Soviet Turcology, 3 (1982): 74. (86) The term is also found in Mu’ayyid as sitrghu; in Sharafnāmā as sirghū; in Miftāḥ al-fużalā as ‘sabazghav: a Turkic windpipe which is made from reed’. Sibizghu is a shepherd’s reed, which, with slight variations, is popular in many Turkic countries under different names: in Bashkir as kuray, chībīzga, sībīzga; in Karachaevo-Cherkessiya as sibizga; among Kumiks and in Dagestan as zībīzgī; in Uzbekistan as sibizik; in Kazakhstan as sībīzgī, and so on. (87) ‘Sarghīn: a Turkic flute, also pronounced with fatḥa and kasra on “sīn”’. Miftāḥ al-fużala, Or.3299.

Page 34 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Turki and Hindavi in the World of Persian (88) Sharafnāma (2006), vol. 1, p. 358. (89) The reverse equivalence appears in the fifth bakhsh, dedicated to miscellaneous words of Arabic, Nabatean (i.e., Aramaic) and ‘Ajami origin: ‘Būq: karnāy, that is, bher’; Zafāngūyā, fifth bakhsh (1997), vol. 2, p. 102. (90) ‘Nāqūr: a big flute that in Hindavi is called bher’. Zafāngūyā, sixth bakhsh: Roman and Greek (1997), vol. 2, p. 143. (91) ‘Chīstān: with Persian “jim”, a riddle. Arabs call it ekrah [?] and Indians pahelī’. Adāt al-fużalā, Or. 1262. ‘Churpak: with zamma, a speech (sukhan) which one says against another, and something asked out of exaggeration that in Arabic is called laghz and abdah and Indians call pahēli [verse by Zahir]’. Sharafnāma (2006), vol. 1, p. 350; Add. 7678. See also ‘Bardak: with fatḥa, a story, and an exaggerated question that is also named churpak, chistān, and girdak. In Arabic it is called laghz and in Hindi pahēlī’. Sharafnāma (2006), vol. 1, p. 173. See also Madār al-afāżil. (92) See, for example, D. Guizzo, I tre classici della lessicografia persiana d’epoca moghul: Farhang-i gahängïrï, Burkän-i qätì, Farhang-i rasïdï, Venice: Cafoscarina, 2002. Um fugiatquam est velitassinia nihil molores susam fa

Access brought to you by:

Page 35 of 35

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis?

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Local Lexis? Provincializing Persian in Fifteenth-Century North India Stefano Pellò

DOI:10.1093/acprof:oso/9780199450664.003.0006

Abstract and Keywords The fifteenth century has generally been dismissed as a period when Persian literature in India declined. This chapter argues that a number of Persian texts were produced in this period but have been overlooked because of the great cultural flowering of the Central Asian Timurid courts at the same time. Persian, the chapter claims, was ‘provincialized’ in north India becoming rooted and adapted to the local environment. In contrast to the standard meaning of ‘provincial’ as unsophisticated or limited, the author suggests that ‘provincial’ Indian Persian represented linguistic appropriation in philological and ideological terms, thus laying the foundation for Persian to later become a truly South Asian language. The chapter points particularly to the proliferation of lexicons and interest in philology as evidence of intense yet localized connection to the Persian cosmopolis. Keywords:   Persian, Central Asia, lexicons, dictionaries, Turki, Sultanate

‫ﺍﮔﺮﭼﻪ ﺑﻮﺩﻡ ﺍﺯ ﺩﻫﻠﯽ ﺑﺴﯽ ﺩﻭﺭ‬ ‫ﺩﻟﻢ ﻣﯽ ﯾﺎﻓﺖ ﺍﺯ ﺣﺐ ﺍﻟﻮﻃﻦ ﻧﻮﺭ‬ ‫ﺯ ﺑﻌﺪ ﻣﮑﻪ ﺳﯿﺮﻡ ﺩﺭ ﻋﺠﻢ ﺑﻮﺩ‬ ‫ﻭﻟﯽ ﺑﯽ ﻫﻨﺪ ﺧﺎﻃﺮ ﻣﯽ ﻧﯿﺎﺳﻮﺩ‬ ‫ﺯ ﻫﻨﺪﻭﺳﺘﺎﻥ ﺍﮔﺮﭼﻪ ﺩﻭﺭ ﺑﻮﺩﻡ‬

Page 1 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? ‫ﭼﻮ ﻃﻮﻃﯽ ﺩﺭ ﻗﻔﺺ ﻣﻬﺠﻮﺭ ﺑﻮﺩﻡ‬ ‫ﺧﺮﺍﺳﺎﻥ ﮔﺮ ﺑﯿﺎﺿﯽ ﺩﺍﺷﺖ ﭘﺮ ﻧﻮﺭ‬ ‫ﺳﻮﺍﺩ ﺍﻋﻀﻢ ﺁﻣﺪ ﻫﻨﺪ ﻣﻌﻤﻮﺭ‬ ‫ﻣﻦ ﺁﻥ ﺧﻀﺮﻡ ﮐﻪ ﺍﺯ ﻓﻀﻞ ﺍٰﻟﻬﯽ‬ ‫ﮐﺸﯿﺪﻡ ﺁﺏ ﺣﯿﻮﺍﻥ ﺍﺯ ﺳﯿﺎﻫﯽ‬ Though I was so far away from Delhi the love for my homeland lit my heart. After Mecca my path was towards Persia, but my mind could not rest without India. And though I had parted from India I was like a parrot in a cage, lamenting separation. If Khorasan has a whiteness full of light well, prosperous India’s blackness has no equals: I am that Khizr who, blessed with divine grace, in blackness was able to find the water of life.1

(p.167) THE FIFTEENTH CENTURY HAS BEEN sometimes described by literary historians as a period of slowdown, if not of crisis, for Persian literary culture in India. According to Husam al-Din Rashidi, the editor of Jamali Dihlawi’s poem Mihr u Māh (completed in 1499), for instance, the period between the death of Firuz Shah Tughluq and Akbar’s accession to the throne was characterized by a striking drought in the production of Persian literary works. He goes as far as saying: ‘Not a single important personality appeared on the scene in this period, nor was a single useful and valuable book composed’.2 Judgements such as this are based mainly on the limited production of courtly poetry, and sometimes they stem from a still quite widespread perception of Persian as the literary language of a monodimensional ‘high culture’. By contrast, this essay argues that while the Central Asian Timurid courts undoubtedly formed the centre of the Persian and Persianate literary world in the fifteenth century, during the same period in north India Persian was actively ‘provincialized’ in ways that rooted it further in the Indian socio-cultural landscape. As a distinct, individualized part of a wider entity, a province is usually understood as being relatively far from a centre, yet not separate from it— exactly how the fifteenth-century Indo-Persian literary space of north India stood with respect to the literary culture of the Timurid renaissance. So, while we usually deem ‘provincial’—as opposed to ‘metropolitan’, ‘international’, or ‘avant-garde’—cultural products and intellectual approaches that are unsophisticated and limited in perspective, here I use the term to describe the Page 2 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? relationship of complementarity and the various degrees of interaction that existed between local and transregional trends within Persianate Eurasia, of which north India was an essential part. In fact, it is only by looking more closely at the autonomous and yet nonseparate formative and transformative processes in the deeply connected ‘provinces’ of the Persian cosmopolis that we can begin to understand its internal functioning. For the Persian cosmopolis, I argue here and elsewhere, was a fragmented, polycentric space, in which the (p.168) cosmopolitan was continuously (re-)localized and vice-versa.3 Gernot Windfuhr and John R. Perry have used the adjective ‘colonial’ to describe the varieties of Persian used outside the Persian-speaking areas, especially South Asian Persian.4 Here I suggest that in cases such as that of post-Timurid north India it might be more useful to use the term ‘provincial’ instead—stripped of its derogatory meaning of course—not least to avoid the teleological echo of ‘colonial’ English. In the fragmented linguistic and political panorama of north India after Timur came there was no Persian ‘colonialism’. Rather, what we see is a trend towards regional linguistic appropriation—in philological as well as in ideological terms— which consciously kept in view, exploited, interpreted, and re-shaped international trends. This ‘provincial’ linguistic appropriation laid the sociolinguistic foundation that allowed Persian to become, within a couple of centuries, ‘a South Asian language for all intents and purposes’.5 One of the international trends that north Indian Indo-Persian intellectuals picked on and extended was the keen Timurid interest in the philological reconstruction and consolidation of the Persian classical tradition.6 This was closely connected to a new recognition (p.169) of, and emphasis upon, the role of multilingualism.7 The result in ‘provincial’ north India was a substantial output, if not of ‘great poetry’, of lexicographic works that aided poetic education and understanding, complemented by an understanding of Persian as a hyper-language, one that included Turki, Hindi, as well as other tongues. To speak of a ‘crisis’ of Persian in India seems out of place in these circumstances. Against this more promising backdrop, we need to begin to ask what actually happened to Persian—as a literary culture—in fifteenth-century north India, and how we can properly identify and describe the ‘provincial’ trends just mentioned. In order to do so, we need to adopt an approach that goes beyond the traditional evaluation of the langue of Persian literature in India—including the amount and typology of poetic production—and instead to address its parole, that is, the actual role Persian played and the actual ways it worked in this alleged ‘century without poetry’, to use the expression Benedetto Croce provocatively employed to describe the Italian literary scene in the hundred years following Petrarca’s death.8

Page 3 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? Here I will attempt to sketch an interpretative framework—without any claim to finality—by looking at some aspects of lexicography, no doubt our most important source for observing the ‘life’ of Persian literary culture and provincial trends in fifteenth-century north India (see also Chapter 5). As scholars like Baevskij, Naqavi, and Dabirsiyaqi have long shown, an unprecedented lexicographic production characterizes north Indian Persian literary culture during the period between (p.170) Timur and Babur.9 This unparalleled fortune of Persian lexicography in South Asia might be connected, not so paradoxically, with the fragmentation of north India into several independent states, which appears to have implied a need for several Indian Muslim courts to reassert (or renovate), through the output of normative texts, their literary–linguistic status within the Persianate cosmopolis. In fact, the high degree of mobility among Indian lexicographers of Persian, who often moved away from their birthplace and from one court to another to seek employment as language experts, might signal the need by certain courts to establish a new, and localized, linguistic authority—the case of Malwa especially claims further scholarly attention. To give a couple of examples, the Adāt al-fużalā was written in 1419 by a Delhiborn man of letters, Qazi Badr al-Din Dharwal Dihlawi, who before compiling his dictionary moved from the court where he was employed, that is Jaunpur, to the court of Qadar Khan, the younger brother of Hushang Shah, the second Ghurid sultan of Malwa;10 the Sharafnāma-yi Munyarī by Ibrahim Qawam Faruqi was compiled in Bihar for Rukn al-Din Barbak Shah (1460–74) of the Ilyas-Shahi dynasty of Bengal, after an unsuccessful period of wanderings that had brought the author to Delhi in search of spiritual and material patronage.11 Beyond these stray examples, we should start asking if and how this wide-ranging movement of language experts (who often came from older cultural centres such as Delhi and moved towards other courtly spaces) actually contributed to a wider and deeper dissemination of Persian linguistic culture within the north Indian ‘province’—thus (p.171) laying, as I have already suggested, the basic fundamental conditions, especially in terms of linguistic ideology, for the Mughal construction of north India as a global centre of Persian textual production. While expressing specific linguistic views, as we shall see, fifteenth-century north Indian lexicographers also disseminated the knowledge of a set of older, authoritative texts that they acknowledged as sources and that established a tradition, within the already described context of the Timurid trend toward the (re-)construction of the literary and scholarly canon. For instance, the introduction of the Adāt al-fużalā—written, as we saw, in Malwa in 1419— mentions directly in its introduction having used as many as seven different lexicographic sources for completing the work, namely, the Lughat-i furs by Asadi Tusi, the oldest surviving dictionary in Persian, the Farhang-i Qawwās, the Lisān al-shu‘ara, and the Dastūr al-afāżil (all composed in India in the fourteenth

Page 4 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? century),12 as well as the lost Farhang-i Firdawsī, Fawā’id-i Burhānī, and Risālat al-naṣīr.13 About a hundred years later, the author of the Tuḥfat al-sa‘ādat (completed in 1510 for a patron connected with Sikandar Lodi’s court) mentioned in his introduction no less than twenty-eight lexicographic sources, fourteen of which were Persian and the remaining seven Arabic.14 This not only suggests a parallel growth in textual production and intertextual exchange, but also signals a diffusion and use of multiple lexicographic books that we still need to properly understand. In other words, what was the need to use so many dictionaries, and why were they updated so often? This lexicographic intertextuality and the quick spread of authoritative works within the north Indian provincial space is referred to, for instance, in the introduction of the Mu’ayyid al-fużalā by Shaykh Muhammad Lad Dihlawi (probably completed in 1519): What urged me to devote myself to the compilation of this book was the fact that the Sharafnāma, though a recent (muta’akhkhir) and comprehensive (p.172) (jāmi‘) work, is not at all an exhaustive text and there is a clear need for other sources where words can be searched, like especially the Qunyat al-ṭālibīn […], which is more or less equivalent to the Sharafnāma as for the number of listed words. But since the Qunyat alṭālibīn does not record what had been recorded in the Sharafnāma unless another meaning is given, this writer decided to prepare a text that is intended to be a complete and useful compendium of these and other sources.15 According to the author, then, the motivation behind the composition of the book was the necessity to complete another lexicographic work written in India no more than fifty years earlier, despite the fact that the Sharafnāma-yi Munyarī had apparently acquired a great deal of authority among the court literati of Hindustan. This unprecedented need for compiling connected garlands of new authoritative tools for dealing with Persian language urges us, again, to question the traditional literary–historical views of ‘crisis’ versus ‘splendour’ and to explore new theoretical frames for understanding what we can tentatively describe here as a project of mimesis and (re-)appropriation. While post-Timur northern Indian courts were no longer among the main centres for Persian literary production, they adopted what I call here a provincial strategy of remaking key ‘cosmopolitan’ forms of knowledge—in this case, philology. Persian philology was a ‘cosmopolitan’ form of knowledge also because it was directly connected to the linguistic disciplines of the exemplary—in terms of cultural identification—Arabic textual past. We still don’t know if and how much these texts actually circulated beyond courtly milieux, and precisely what their actual use was in the constitution of the local, or provincial, cosmopolitanism we are trying to describe. We do know, Page 5 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? however, that some of these works played an important role in the subsequent formation of the Safavid-Mughal transregional koiné pointing to a wider textual circulation among the provinces of the vast Persianate world, and showing the complementarity between the twin processes of provincialization and internationalization. A systematic textual comparison has shown, for instance, how much the later Farhang-i Mirzā Ibrāhīm (written in Safavid Iran by (p.173) Mirza Ibrahim ibn Shah Husayn Isfahani around 1568) draws upon, despite never openly mentioning it, the Sharafnāma, a text that includes, as we will see in this chapter and in Dilorom Karomat’s (Chapter 5), a certain amount of ‘local’, that is, Hindi, lexicon.16 Moreover, many of the dictionaries written in this period in north India became the fundamental sources for the three great dictionaries written in seventeenth-century India, that is, the Farhang-i Jahāngīrī, the Burhān-i Qāṭi‘, and the Farhang-i Rashīdī.17 And it cannot be by chance that in the middle of the eighteenth century, Siraj al-Din ‘Ali Khan Arzu still mentioned a ‘provincial’ work like the Sharafnāma-yi Munyarī as an authority in his most ambitious work, the Mus̱mir, a text that subtly celebrates the ability of the Persian language to include local lexicon.18 Thus, the multiplication of active lexicographic laboratories across several centres that characterized the provincialization of Persian in fifteenth-century India seems to have favoured the transregionalization and the deepening of the study of Persian as a linguistic and literary culture in the following centuries. Strictly related to the issue of the provincial and transregional circulation of these dictionaries is that of their connection with poetry. It is a well-known fact that Persian lexicography was born as an ancillary discipline to poetry and maintained this characteristic for a long time. Identifying what kind of poetry was quoted, discussed, or even just referred to in the dictionaries will help us investigate which poetry was popular and circulated, and by which means, in an apparently poetry-obsessed century without poetry. (p.174) The first impression one gets from a brief overview is that, in sharp contrast to the paucity (both in quantity and in quality) of direct verse production, the philological study and the reading and commenting of Persian poetry were an essential feature of the cultural milieu of fifteenth-century northern India, a widespread cultural practice and a courtly instrument of selfrepresentation and legitimation. The introduction of the Tuḥfat al-sa‘ādat is eloquent in this respect. About the motivations for writing the book, Mahmud ibn Shaykh Ziya al-Din Muhammad writes: Since the children of the times, be they religious authorities (‘ulamā), erudite people (fużalā), princes (umarā) or sultans, invariably have a great taste for poetry, and I myself, after having built the foundation of the poetic art, have spent many years […] in the practice of writing verses and doing justice to the different forms of Persian poetry, as ghazal, quatrain, qaṣīda

Page 6 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? and mas̱nawī, [….] my friends asked me to compile a dictionary which was to be a summary of the oldest ones.19 The lexicographer here refers to his regular practice of poetry-making in the main forms of Persian verse to satisfy the ‘great taste’ for Persian poetry in a milieu that comprised military, courtly, scholarly, and religious elites. Some authors went into considerable detail about what was taught and studied in the various provincial centres of Indo-Persian learning, giving an impression that a structural need existed for instruments that would help complete and consolidate a pre-defined cursus honorum of the provincial literati’s poetic education. A good example is the introduction to the coeval Mu’ayyid al-fużalā, which helps us in identifying which authors were mainly read at the latefifteenth and early-sixteenth century Delhi court. Shaykh Muhammad Lad leaves no space for doubt when he writes about the aim of his work: This text is useful and complete enough to read and teach Firdawsi’s Shāh nāma, Nizami’s Khamsa, Sana’i’s Sitta and the diwans of Khaqani, Anwari, ‘Asjadi, Hafiz, Salman, Sa‘di, [Amir] Khusraw and others.20 (p.175) The didactic aim is evident not only in the explicit reference to teaching (the lexicon, thus, primarily acquires the features of a school instrument) but also in the choice of a limited set of extremely authoritative and scholastic mutaqaddim or ‘old’ authors. Persian poetic education, in other words, seems to consist more of learning of the fundamental linguistic and literary code than of entering into a dialogue with contemporary verse—not a single Timurid author is mentioned here. In other cases, certain features of these dictionaries make them look like technical manuals or lexical repertoires for composing poetry. The last section of the Sharafnāma-yi Munyarī, for instance, consists of a bāb al-buḥūr or ‘Chapter on Metres’, where the reader finds what appears to be a ready-to-use compendium of Persian poetic metres.21 Though it does not contain verse specimens, the Adāt al-fużalā is another interesting example: the book is divided into two parts, and the second is specifically devoted to compounds and expressions that are employed (and, evidently, employable) in poetry. In the introduction, moreover, the author states clearly that, after having progressed on the path of poetry and learnt the ‘skills necessary to this activity’ (luzūm-i īn kār) thanks to his masters, he wanted to collect in his own dictionary the words that the other ‘travellers on this road’ (sālikān-i īn rāh) might need.22 Both books were written, as we saw, at provincial courts distant from Delhi, namely the Ilyas-Shahi of Bengal and the Ghuri of Malwa. We may also recall, in this context, the so-called Jaunpur Anthology, actually a voluminous poetic manual called Dastūr al-shu‘arā, composed for the Sultan of Jaunpur Mubarak Shah in 1400–1.23 Although the overall picture of Persian poetic production and Page 7 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? reception in fifteenth-century India remains unclear, these diverse specimens suggest a dominant trend towards the necessity of learning and teaching (or, in other terms, acquiring and prescribing) the basics of poetry. The systematic study of Persian poetic commentaries—a discipine closely connected to lexicography—in north India in this period would be extremely useful in this respect. The centrality given to the correct understanding and interpretation of (p.176) Persian poetry—and, possibly, to the didactic sphere, since at least some of these lexicographers could easily have also been teachers of Persian— transpires not only in the direct or indirect references to poetry that we find within the dictionaries, but also from the other works of at least some of the lexicographers mentioned here. The author of the Baḥr al-fażā’il fī manāfi‘ alafāżil, for instance, is known to have written commentaries on Nizami Ganjawi’s long romances Makhzan al-asrār and Iskandarnāma, while Muhammad ibn Da’ud Shadiabadi, the compiler of the Miftāḥ al-fużalā, composed guides for reading the great qasida-writers of the Seljuqid era, Anwari and Khaqani.24 While Persian retained its prestige in the various smaller political–geographical domains in India, the responses of local lexicographers to multilingualism show that a fundamental need was felt to re-interpret and re-define linguistic and literary values and roles in Persianate north India after Timur. This was another ‘provincial’ trend that connected north India to trends in the wider cosmopolis of Timurid Central Asia. Persian’s relation with Turki is a case in point and played a complementary role in both Central Asia and India. But lexicographers also appear to have been working to equip Persian to cope directly with the local milieux, especially for its interaction with the sphere of Hindavi, which marked the need and will to deal with provincial specificities. It is interesting to read what the dictionaries say about the language (or cluster of languages) they are dealing with. Persian is presented as a composite expressive whole—a sort of hyperlanguage—made up of several different and recognizable languages. This, for instance, is what the Farhang-i zafāngūyā u jahānpūyā (completed before 1433 by Badr al-Din Ibrahim) says on the subject: after mentioning the need of bringing order in lexicography, ‘in which all the languages have been so far treated as if they were the same thing’ (that is, the origin of foreign loanwords recorded in Persian lexicons was generally not specified), the author informs the reader that some of his friends, whom he describes as literati always absorbed in reading poetry and using vocabularies to understand it, asked him to solve the problem, and he responded in the following manner: I accepted their request and took the pen and closed on myself the door of affliction and connected word to word and kept my mind on and (p.177) concentrated my passion to put together the dictionaries and collect the scattered words, and to record and put in the right place whatever was

Page 8 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? there from the Arabic, Persian, Pahlavi, Dari, Greek, Turkic, and other languages.25 Philology, I argue, thus becomes a way to ‘conquer the world’ (the dictionary is described, in the title itself, by the terms zafāngūyā ‘polyglot’ and jahānpūyā ‘world-traveller’, or ‘experienced’—in a word, ‘cosmopolitan’). Education in the Persian language also involves acknowledging the inclusion, within Persian, of the main languages of the world, both ancient and modern. Analysing the Persian language, then, means learning the code of the wider linguistic cosmopolis, and the absence of Hindi/Hindavi among the languages mentioned might be, in this context, indirectly significant of the provincial approach we are dealing with: ‘local’ north India is the place from which the world is observed, not itself an object of observation.26 In other words, it is not the province that gets philologized, but rather philology needs primarily to be taken and explained to the province, which creates itself as such by marginalizing its own local linguistic features—that is, Hindavi—and separating it from the main linguistic koiné. Yet, insisting on multilingualism, even without mentioning Hindavi, can also be seen as a way to leave open the possibility of including Hindavi as well. A comparable method, for instance, will be exploited by Arzu three centuries later.27 The awareness of literary Persian as a composite expressive medium was not, of course, new to Persian dictionaries. In the Dastūr al-afāżil, (p.178) completed in 1341–2 by Hajib Khayrat Dihlawi at the Tughluq court, we find an even wider view of Persian as a composite, multicultural, and ‘impure’ language: […] and I read and found very useful the compositions by the poets of Khorasan and India, mentioning whose names would take a long time here, and I collected words from Arabic, Turkic, Mongolian, Pahlavi, Parsi, Afghani, Judaic, Christian [Syriac?], the language of the Magians and Syrians, the language of philosophy, Tajik, Hebrew, the language of Rayy and of Hijaz and the lands of Transoxiana, and from the expressions of the poets of every city.28 The nod towards the ‘inclusive’ features of Persian in fourteenth-century Delhi, then still one of the main centres of the global Persian-writing community, mainly works as a guarantee for the completeness of one’s lexicographic research. The Zafāngūyā, on the contrary, operates through distinction and separation, its aim being precisely to show the etymological origins of the Persian words it records, analysing multilingualism rather than celebrating it. The introduction itself, significantly enough, does not contain any loanwords, the stress being on the effort to identify, isolate, and classify the various elements and choose among them, or even to choose not to use them, as the author seems

Page 9 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? to prefer, perhaps in order to signal, through the exhibition of ‘purism’, a particularly high linguistic consciousness and skill.29 Multilingualism is, of course, the key issue when dealing with the provincialization of Persian in north India after Timur, especially in view of the layered relationship between Persian on the one side, and with Turki (already both a local and transregional language) and Hindi (a local language, but projected through Persian into the wider world) on the other side. Without going into the details of the ecology of Turki words in our dictionaries, which is discussed in Chapter 5, I will concentrate here on the ‘local’ dynamics of Persian as a multilingual container in (p.179) fifteenth-century India by briefly highlighting a few aspects of the use of Hindavi vocabulary in these dictionaries. By doing so, I hope to provide a few pointers towards the more systematic research on the modes of formation of a provincial self-identification for Persian in South Asia. A better understanding of the role Hindi played in fifteenthcentury lexicons–which are fundamentally conscious reflections on, and representations of, a linguistic standard—is useful not only in order to assess the space of Persian in post-Timur multilingual India, but also to place the debates on the peculiarities and values of the so-called ‘Indian Persian’ that characterize the mature phase of Indo-Persian literary culture within a longer perspective.30 The use of Hindavi words in Indo-Persian lexicons is not, it must be underlined, a completely new feature of the post-Timur age. According to Nazir Ahmad, the Farhang-i Qawwās, the oldest surviving Persian dictionary written in South Asia, contained (only) eight Hindi terms used as equivalents of the same number of Persian words.31 A trend towards including more and more words of Indian origin can be easily observed even before the fifteenth century: the Farhang-i lisān al-shu‘arā, most probably composed around 1378 at the Delhi court (and surely before 1419), already contains several Hindi terms that are used to better explain a select vocabulary, mainly comprising non-literary terms relating to everyday life, such as bhelī (‘bran’), given as equivalent of tagazhdāna, gophan (‘sling’), occurring in the lemma dedicated to falākhan, or gudgudī (‘tickle’) to explain ghilghilīch.32 The last example is especially interesting, since the kind of explanation given clearly shows that a knowledge of Hindi was required in the reader, and that the use of Hindi was not merely an addition: ghilghilīch: […] it is that thing which in hindawī is called gudgudī, when you touch the armpit with the finger.33 (p.180) The employment of Hindi words to explain Persian terms became a constant and more decisive feature of fifteenth-century Indo-Persian lexicography. Hindi terms are commonly found with remarkable frequency not only in the dictionaries produced at the regional courts but also in those written during the Lodi renaissance at Delhi: they are equally represented in the Miftāḥ al-fużalā (written, as we have seen, in Malwa in 1468) and in the Mu’ayyid alPage 10 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? fużalā, composed in Delhi in 1519.34 In the first two chapters (letter alif and letter ba) of the Sharafnāma-yi Munyarī, one finds as many as fifty-six Hindi words defined as such (thirty in the chapter devoted to words beginning with alif, and twenty-six in the chapter of letter ba), plus some other Hindi terms listed as lemmas or occurring in verses.35 As for the typology of the Hindi vocabulary contained in this dictionary, though we cannot speak of a definite change compared to, for example, the Lisān alshu‘arā, we can note an interesting widening of the semantic spheres (see also Chapter 5 in this volume). In fact, besides the usual words relating to animals (e.g., gharghat, ‘chameleon’, under the voice āftābgardak36), crops and vegetables (such as palak, ‘spinach’, given as an equivalent of Persian ispinākh37), tools (phāla, ‘ploughshare’, provided as an explanation of āhanjuft38), and so on, we also find a noteworthy attention to the objects of scientific learning (such as chūmak, ‘magnet’, under the voice āhanrubā39), calendrical terms (e.g., bhādon, given as the Indian equivalent of the Syriac month āb), and also a few generic terms which can also belong to the Persian poetic idiom. For instance, when dealing with the Persian word āsīb, ‘blow’ or ‘misfortune’, Faruqi choose to provide also the Hindi equivalent dhakka.40 As Dilorom Karomat also notes, it is especially noteworthy that the Sharafnāma at times provides Hindi equivalents for terms belonging to the technical vocabulary of rhetorics and poetic science, as in the case of (p.181) bardak, ‘riddle’, explained with the Hindi pahelī alongside the technical Arabic lughz.41 When providing the Hindi equivalents of Persian terms, the Sharafnāma generally uses the formula ‘in India they say (hind-ash gūyand/khwānand)’, while the term hindī is relatively seldom used for the name of the language. In some cases, as we saw, Hindi terms are provided as lemmas, as in the case of anguz, the iron hook with which elephants are driven. Here, significantly, the author quotes a Persian poetic example for this word that contains another Hindi term, mahāwat, ‘elephant driver’: tu gū’ī ki ṭūr-ast u mūsā mahāwat ba jāy-i ‘aṣā anguz-ī mārpaykar.42 You would say (the elephant) is Mount Sinai and the elephant driver is Moses, instead of the pastoral staff (he has) a snake-like iron hook.

In other cases, the Indian origin of a word, or its status as a Hindi loanword in Persian, are explicitly mentioned. Read, for instance, the definition of the word belak: belak, with the Persian yā, is a kind of arrow used in hunting, and this is a Hindi word employed in Persian. It is called also felak.43

Page 11 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? At times, the Sharafnāma gives the impression that the author is not providing a Hindi equivalent of the explained term but rather explaining how the word in question is pronounced by Indian users of Persian—in other words, the dictionary seems to move in the direction of identifying an Indian Persian. For instance, in the entry devoted to the word baqam (a kind of red wood), the lemma is defined by the formula ‘ma‘rūf’, ‘it is well known’, which is used when the meaning is considered universally clear and no need for explanations is felt. Ibrahim Faruqi, however, adds as a comment that ‘in India they say bakam (hindash bakam gūyand)’.44 As further evidence of the provincializing trends of Persian in fifteenth (p.182) century reflected in Indo-Persian lexicography, I would also emphasize the fact that the word barahman was considered so common by the author of the Sharafnāma that it is defined only by the formula ma‘rūf just mentioned.45 In summary, in many respects fifteenth-century Indo-Persian lexicons can be read as multilingual dictionaries that used Persian as their basic medium while at the same time they provided information on the languages with which Persian at least partly shared its socio-textual spaces—Hindi and Turki, but also, of course, Arabic.46 Especially with respect to the compresence of Arabic and Hindi words used as equivalents of Persian lemmas, it should be underlined that the provincial ‘citizenship’ accorded to Hindi within lexicography seems also to have been activated by the juxtaposition of these two languages: using an Arabic and a Hindi lexical equivalent in the same entry and at the same level would have given the latter language a certain prestige, considering the immense cultural, intellectual, and linguistic weight retained by Arabic in the whole Islamicate world. The same can be inferred in the case of the juxtaposition of Hindi and Turki words, bearing in mind the high position of Chaghatai Turkic occupied in the Timurid world. It was in fact in this period that Hindi began to be consciously considered a part of the Persian literary-linguistic cosmopolis, with its own specific ecological niche. We may consider, for instance, the Baḥr al-fażā’il fi manāfi‘ al-afāżil by Muhammad ibn Qawam Balkhi (compiled in Gujarat in 1433), in which not only Indic words are used to explain Persian lemmas, but there is even a specific section devoted to Hindi words employed in poetry.47 The discussion on linguistic prestige and multilingualism can also be expanded by looking briefly, in conclusion, at the beginning of Persian grammatical descriptions and their possible connections with what we (p.183) have been calling here the provincialization of Persian in north India. Roughly a century before Timur came, Amir Khusraw had written that Persian—the language of the

Page 12 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? cosmopolis that had Delhi as one of its main centres at that time—had no need for grammars: No rhetorician has recorded the rules of the sweet language of the Persians/ […] but I cannot find anybody who is not able to speak it.48 In the fifteenth century, instead, Persian grammar began to be studied and recorded by Persian literati working at provincial courts of northern India. One of the first specimens is the Sharafnāma itself. In a chapter named ‘The chapter on letters (also, technically, “morphemes”) meanings and observations’ (bāb alḥurūf, al-ma‘ānī wa’l-fawā’id), the author deals over fourteen sections (faṣl) with the morphological values of the alif, hamza, bā, and the other letters of Arabic– Persian alphabet; he then explains the formation of the plurals, the substantiation of the verbal stems, the rules determining the use of the kasra-yi iżāfa, and so on.49 Some fifty years later, the author of the Mu’ayyid al-fużalā added an appendix to his dictionary which contained a summary of the norms regarding the numerical value of letters and numbers (abjad) and some rules and dispositions (taṣarrufāt) for the correct use of the Persian language.50 Both these grammatical surveys have a quite clear didactic function, probably pointing towards a new need, in the fragmented context of post-Timur north India, of written sources that would work as reliable schemes of reference. Could this need be in some way related to the lack of living Persian linguistic authorities in the context we are dealing with? I do not have an answer to such questions at present. Still, I would suggest that the study of Persian grammar seems to have been itself a provincial fact: to (p.184) my knowledge, the first complete grammar of Persian written in Persian was compiled in southern China, in the town of Shangdong, by a Hui Muslim in the middle of the seventeenth century,51 while we have to wait until 1846 for the first Persian grammar compiled in Iran.52 Relating, and in fact anticipating, the dynamics of provincialization discussed here, we may also recall Persian grammatical writings in Sanskrit: at least two such works were completed in the second half of the fourteenth century, the Yavananāmamālā (1364) by Vidyanilaya, probably a Jaina working at the court of Firuz Shah Tughluq, and the Pārasīnāmamālā (1365) by Minister Salaksa (another Jaina) at the Gujarati court of king Haribrahma of Ilavarana (modern Idar).53 From our perspective, it is significant that the author of the second work, in Sarma’s words, ‘justifies his composition by saying that proficiency in several languages leads to high honour at royal courts’, citing the case of Varahamihira, who popularized Greek astronomy through his works.54 This tradition continued also after Timur: the Pārasīkābhā-ṣānuśāsana by the Jaina Vikramasimha, for instance, was composed in the late fifteenth–early sixteenth

Page 13 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? centuries, probably in the area of Malwa or Gujarat (where intense Persian lexicographic activity is noticeable around the same period).55 Although still highly tentative, and also openly provocative, the concepts of ‘provincial’ and provincialization discussed in this essay can nevertheless help us find the proper hermeneutical tools to deal with the relatively little known Indo-Persian textual production of the fifteenth century, also beyond lexicography. As I have suggested, the notion of provincial helps us understand the connections between Persian in India (p.185) and the Timurid cosmopolis, the relationship between Hindi and the multilingual background of Persian, and the changing meaning of multilingualism in Persian dictionaries, the particular use and meaning of philology in fifteenth-century north India, and the writing of Sanskrit manuals for learning Persian. In other words, the notion of provincial seems to work for processes taking place simultaneously at the trans-Indian, inter-Indian, and infra-Indian levels, showing the opportunity of analysing such different phenomena as parts of an interconnnected process. ‘Provincial’ may also be used as a conceptual tool with which to deal with other important literary facts that still need to be properly studied and understood, such as the first organized attempts at giving Sanskritic literature a place in the Persian cosmopolis through translations and re-writings (by, e.g., Sultan Zayn al-‘Abidin in Kashmir), or the circulation of Indian writers of Persian from South Asia to Central and West Asia.56 Notes:

(1) Jamali Dihlawi, Mas̱nawī-yi Mihr u Māh, S.H. Rashidi (ed.), Rawalpindi: Iran Pakistan Institute of Persian Studies, 1974, p. 167. (2) S.H. Rashidi, ‘Introduction’, in Jamali (1974), p. 16. A similar judgement can be found in M. Alam, ‘The Culture and Politics of Persian in Precolonial Hindustan’, in Literary Cultures in History, S. Pollock (ed.), Berkeley: University of California Press, 2003a, p. 157. (3) See S. Pellò, ‘Persian as a Passe-partout: The Case of Mirza ‘Abd al-Qadir Bidil and his Hindu Disciples’, in Culture and Circulation: Literature in Motion in Early Modern India, A. Busch and T. de Bruijn (eds), Leiden: Brill, 2014, pp. 21– 46. (4) G. Windfuhr and J.R. Perry, ‘Persian and Tajik’, in The Iranian Languages, G. Windfuhr (ed.), London: Routledge, 2009, p. 418. (5) M. Alam and S. Subrahmanyam, ‘Discovering the Familiar: Notes on the Travel-Account of Anand Ram Mukhis, 1745’, South Asia Research, 16(2), 1996: 132.

Page 14 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? (6) The beginning of substantial poetic tazkira-writing (with model texts such as the Tażkirat al-shu‘arā by Dawlatshah Samarqandi or the Majālis al-nafā’is by ‘Ali Shir Nava’i) as a way to establish the canonical archive and authoritative code of Persian literature is among the most significant aspects of the Timurid age for what concerns us here; see P. Losensky, Welcoming Fighānī: Imitation and Poetic Individuality in the Safavid-Mughal Ghazal, Costa Mesa: Mazda, 1998. Closely connected is the building of a conscious, technical relationship with the textual past of poetic production and rhetorical reflection; see, for example, R. Zipoli, The Technique of the Ğawāb. Replies by Nawā’ī to Ḥāfiẕ and Ğāmī, Venezia: Cafoscarina, 1993, and Maria Simidchieva, ‘Imitation and Innovation in Timurid Poetics: Kashifi’s Badayi’ al-afkar and its Predecessors, al-Mu‘jam and Hada’iq alsihr’, Iranian Studies, 36(4), 2003: 509–30. (7) Think of the complex, extremely productive dynamic of ‘rivalry’ between Persian and Chaghatai Turkic in the Timurid world, exemplified, for instance, by the well-known work by ‘Ali Shir Nava’i, Muḥākamat al-lughatayn or ‘The trial of the two languages’ (1499), where Chaghatai is deemed literarily superior to Persian; ‘Ali Shir Nava’i, Muḥākamat, R. Devereux (ed. and trans.), Leiden: Brill, 1966. (8) See B. Croce, ‘Il secolo senza poesia’, in Poesia popolare e poesia d’arte. Studi sulla poesia italiana dal tre al cinquecento, P. Cudini (ed.), Napoli: Bibliopolis, 1991, pp. 191–216. (9) I am referring to Solomon Baevskij, Rannjaja Persidskaja Leksikografiya XI– XV vv., Moscow: Nauka, 1989; Baevskii, Early Persian Lexicography: Farhangs of the Eleventh to the Fifteenth Century, revised and updated by J.R. Perry, Folkestone: Global Oriental, 2007; Muhammad Dabirsiyaqi, Farhanghā-yī Fārsī va Farhanggūnahā, Tehran: Intisharat-i Isparak, 1368H/1989; Shahryar Naqawi, Farhangniwīsī-yī Fārsī dar Hind u Pākistān, Tehran: Wizarat-i Farhang, 1341H/ 1962. (10) See Naqawi (1962), pp. 59-60. The slightly earlier Danishnāma-yi Qadar Khān (1405), as patent from the title itself, was dedicated to the same prince by a native of the Malwa region, Ashraf ibn Sharaf al-Muzakkar Furughi; see Baevskii (2007), p. 84. (11) See C. Rieu, Catalogue of the Persian Manuscripts in the British Museum, 3 vols, London: British Museum, 1879–1883, vol. 2, p. 492. (12) On the date of the Lisān al-shu‘arā, see P. Orsatti, ‘Il Lisān al-šu‘arā: un lessico persiano d’India (sec. XIV)’, Annali. Istituto Universitario Orientale, 50(2), 1990: 144–5), and Nazir Ahmad (ed.), Farhang-i lisān al-shu‘arā, Delhi: Rayzani-ye farhangi-ye jumhuri-ye islami-ye Iran, 1374H/1995, pp. 1–3.

Page 15 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? (13) See Naqawi (1962), p. 60; Dabirsiyaqi (1989), p. 54; and Baevskii (2007), p. 88. (14) Dabirsiyaqi (1989), pp. 70–1. (15) Naqawi (1962), p. 67. (16) See Ibrahim Qawam Faruqi, Sharafnāma-yi Munyarī, H. Dabiran (ed.), Tehran, 2001, vol. 1, hifdah. (17) The most recent analysis of the sources of these texts is D. Guizzo, I tre classici della lessicografia persiana d’epoca moghul: Farhang-i Ğahāngīrī, Burhān-i Qāṭi‘ e Farhang-i Rašīdī, Venice: Cafoscarina, 2002. (18) See, for instance, the discussion on the Hindi word kotwāl (Siraj al-Din ‘Ali Khan Arzu, Mus̱mir, R. Khatun [ed.], Karachi: The Institute of Central and West Asian Studies, 1991, p. 175); see also S. Pellò, ‘Persiano e Hindi nel Musmir di Sirāj al-Dīn ‘Alī Xān Ārzū’, in L’Onagro Maestro. Miscellanea di fuochi accesi per Gianroberto Scarcia in occasione del suo LXX sadè, R. Favaro, S. Cristoforetti, M. Compareti (eds), Venezia: Cafoscarina, 2004, p. 254. (19) Quoted in Naqawi (1962), pp. 65–66. (20) Quoted in Naqawi, p. 67, emphasis added. (21) See Faruqi (2006), hifdah. (22) See Naqawi (1962), p. 60. (23) British Museum, Ms. Or. 4110; for a short description of its contents, see C. Rieu, Supplement to the Catalogue of the Persian Manuscripts in the British Museum, London: British Museum, 1895, pp. 232–3. (24) See Naqawi (1962), pp. 58, 61; Baevskii (2007), pp. 105, 109. (25) Badr al Din-i Ibrahim, Farhang-i zafanguia va dzhakhanpuia (Slovar’ govoriaschii i mir izuchaiushchii). Faksimile rukopisi. Izdanie teksta, vvedenie, spisok tolkuiemykh slov, prilozhenia S.I. Baevskogo, S.I. Baevskii (ed.), Moscow: Nauka, 1974, pp. 145–6. (26) As a matter of fact, as already noticed by Baevskii, the author of the Zafāngūyā makes an extensive use of Indic equivalents to explain the terms he deals with in the text; Baevskii (2007), pp. 96–7. (27) See, for instance, his observations in the already mentioned Mus̱mir (1991, pp. 160–1) and in the Mawhibat-i ‘użmā (Siraj al-Din ‘Ali Khan Arzu, ‘Aṭiya-yi kubrā wa Mawhibat-i ‘użmā [nukhustīn risālāt ba zabān-i fārsī dar bayān wa ma‘ānī], S. Shamisa (ed.), Tehran: Firdaws, 2002, p. 100). See also my analysis Page 16 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? in S. Pellò, ‘Poeti hindu e circoli intellettuali persiani tra Delhi e Lucknow (1680– 1856): un caso di interazione letteraria’, unpublished PhD dissertation, Rome: University ‘La Sapienza’, 2006, pp. 27–51. (Also see Chapter 5 in this volume.) (28) Quoted in Dabirsiyaqi (1989), p. 44. (29) Badr al-Din-i Ibrahim (1974), pp. 143–6. This reminds closely of the Hellenistic ‘provincial complex’ when it comes to multilingualism: for instance, as noticed by Giusto Traina, who refers to Plutarch, V. Anton. 27, in the context of Ptolemaic Egypt the ability to speak several languages (a field in which Cleopatra was particularly skilful) was always accompanied by a struggle to preserve the purity of the Greek language, purged from barbarisms and ‘macedonisms’; see Giusto Traina, Il complesso di Trimalcione, Venice: Università degli Studi di Venezia, 1991, p. 14. (30) See, for instance, S.R. Faruqi, ‘Unprivileged Power: The Strange Case of Persian (and Urdu) in Nineteenth Century India’, The Annual of Urdu Studies, 13 (1998): 3–30. (31) See Fakhr al-Din Mubarakshah Qawwas, Farhang-i Qawwās, N. Ahmad (ed.), Tehran: Bungah-i Tarjuma wa Nashr-i Kitab, 1974, p. 12. (32) N. Ahmad (1995), pp. 140, 258, 240–1. (33) Ahmad (1995), pp. 240–1. Some comments on the use of Hindavi words in the Lisān al-shu‘arā can be found in Orsatti (1990), pp. 162–3. (34) I am relying on the information given in Naqawi (1962), p. 66 and Qawwas (1974), pp. 9–10. (35) See Faruqi (2006). (36) Faruqi (2006), p. 72. (37) Faruqi (2006), p. 40. (38) Faruqi (2006), p. 35. (39) Faruqi (2006), pp. 25–6. (40) Faruqi (2006), p. 30. (41) Faruqi (2006), p. 173. (42) Faruqi (2006), p. 60. The verse, as stated by the same lexicographer, is taken from the Tāj al-ma’ās̱ir.

Page 17 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? (43) Faruqi (2006), p. 176. The word is probably not Indian in origin (it appears to be formed by Pers. bel, ‘shovel’ + the diminutive suffix –ak), but the testimony is important in itself. (44) Faruqi (2006), p. 185. (45) Faruqi (2006), p. 193. (46) In the first two bābs of the Sharafnāma, for instance, more than a hundred Arabic equivalents are listed; Faruqi (2006). (47) See the brief description in Baevskii (2007), pp. 105–8. See also the observations by S.R. Faruqi, Early Urdu Literary Culture and History, Delhi: Oxford University Press, 2001, p. 76, who relies on M.H. Sherani, Maqālāt-i Sherānī, M.H. Sherani (ed.), Lahore: Majlis Taraqqi-e Adab, 1966, vol. 1, pp. 102–31, 161. (48) Amir Khusraw, Nuh Sipihr, M. Wahid Mirza (ed.), London, New York: Oxford University Press, 1950, pp. 173–4. See the comments in E. Jeremiás, ‘Grammar and Linguistic Consciousness in Persian’, in Proceedings of the Third European Conference of Iranian Studies, C. Melville (ed.), part 2, Wiesbaden: Harrassowitz, 1999, pp. 19–20; and Pellò in Mirza Habib Isfahani, Dabistān-i Pārsī. Una grammatica persiana del XIX secolo, S. Pellò (ed.), Venezia: Cafoscarina, 2003, p. 10. (49) Faruqi (2006), pp. 10–19. (50) Dabirsiyaqi (1989–90), p. 74. (51) The grammar has been published in Iran: Muhammad ibn al-Hakim alZinimi, Minhāj al-ṭalab. Kuhantarīn dastūr-i zabān-i fārsī, M.J. Shari‘at (ed.), Isfahan: Mu’assisa-yi matbu‘at-i mash‘al-i Isfahan, 1982. (52) See Eva Jeremiás, ‘Tradition and Innovation in the Native Grammatical Literature’, Histoire Epistémologie Langage, 15(2), 1993: 51–68. (53) See S.R. Sarma, ‘Sanskrit Manuals for Learning Persian’, in Adab Shenasi, A.D. Safavi (ed.). Aligarh: Aligarh Muslim University, 1993, pp. 2–3. (54) Sarma, p. 3. On the issue of Sanskrit-Persian grammatical writings, see also S.R. Sarma, ‘From Yāvanī to Saṃskṛtam: Sanskrit Writings Inspired by Persian Works’, Studies on the History of Indian Thought, 14 (2002): 84–7. (55) See Sarma (1996), p. 4. (56) I think especially of the literary figure of Jamali Kamboh Dihlawi (d. 1535), the poet laureate at Sikandar Lodi’s court. His poetic personality is determined by his travels in the wider Persian world, especially to the centres of the Timurid Page 18 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Local Lexis? cosmopolis about which he himself tells us in his tazkira of Sufi saints, the Siyar al-‘ārifīn, as well as other works; see Jamali (1974), pp. 64–76, and S.B.F. Husaini, A Critical Study of Indo-Persian Literature during Sayyid and Lodi Period, 1414–1526 A.D., Delhi: M.S. Publications, 1988, pp. 41–69.

Access brought to you by:

Page 19 of 19

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Languages of Public Piety Bilingual Inscriptions from Sultanate Gujarat, c. 1390–1538* Samira Sheikh

DOI:10.1093/acprof:oso/9780199450664.003.0007

Abstract and Keywords Stone inscriptions were a way of signalling land ownership or patronage of monumental structures in premodern Gujarat, as elsewhere in South Asia. They were also a way for rulers or officials to issue edicts. A number of stone inscriptions from Gujarat were written in two or three languages and scripts— generally combinations of Sanskrit or Gujarati and Arabic or Persian. This chapter disaggregates contexts in which bilingual inscriptions were produced, arguing that as they were formal or even legal texts, they were generally written in transregional languages. In formal vocabularies drawn from each language, they represented new ways of representing land relations, hierarchies, and religious affiliation without resorting to literal translation of each other. Keywords:   epigraphs, inscriptions, Gujarat, Sultanate, Sanskrit, Persian, Arabic, water architecture, mosques

STONE-CARVED EPIGRAPHS FROM THE FIFTEENTH century record land rights, the construction of buildings, legal proclamations, and religious donations. They were usually commissioned by patrons to commemorate the building or repair of monumental structures—mosques, fort walls or gates, wells, sarais, temples, and so on.1 Perhaps more directly and explicitly than any other literary genre, epigraphs represent the economic and political relations of the day. Constructing a prestigious building and commemorating it in an epigraph was a political act and (p.187) an expensive one. It signalled the power, wealth, and influence of the builder. Epigraphy gives us an entry into Page 1 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety another kind of text production, one that is meant for long-term and often public display. In their assertion of authority, ownership or philanthropy, the commissioners and authors of epigraphs were generating legal texts. Historians usually use epigraphs as sources for reconstructing social and political history. Although many epigraphs from this period are elaborate, long, and expensively carved, they tend to be mined for the names and dates they contain, while the rest of the text is ignored. But epigraphs were a literary genre too. They were often specially composed by poets or authors, some of whose names we know. Many used regular literary conventions such as those of the Sanskrit prashasti to describe their patrons’ achievements, genealogies, or pious intentions. Epigraphs embody and represent arenas of cultural patronage that may be distinct from the patronage circles for other forms of literature. Epigraphs were—and are—a means for patrons to inscribe themselves upon the landscape. In the fifteenth century, as a new lineage of independent sultans came to power in Gujarat, changes in the region’s politics were reflected in epigraphs. The new political order gave rise to a class of patrons with sufficient wealth and influence to build large structures and to express in epigraphs their diverse linguistic and literary aspirations and interests. Many of these new patrons had prospered through association, direct or indirect, with the fifteenthcentury Gujarat sultanate court. Monumental epigraphy was a form of literary production which was directly related to the new forms of government, land-use, and prosperity in the sultanate. While commercial and domestic buildings must have regularly been built, these were not usually (p.188) accompanied by inscriptions. It was meritorious and prestigious to build forts, walls, gates, mosques, rest-houses, and wells, and many of these structures have stone inscriptions recording the circumstances of their construction. The epigraphic choices made by patrons and builders can be revealing. The size of an inscription can tell us whether it was meant to be seen from afar. The stone used could be costly marble or cheaper local stone. Its location in the building can tell us whether it was intended for private viewing or public display. The quality of calligraphy might reflect the patrons’ proclivities and available talent. From the text of inscriptions, we can make surmises about literacy and literary culture. Some inscriptions were in public places, others in spaces restricted by status or sect. From the placement of an epigraph in a building, we may reflect on the regulation of people’s movements. The remarkably frequent names of poets and masons deepen our knowledge of cultural production and patronage. Epigraphs of fifteenth-century Gujarat point us in the direction of the new spaces for text production that were becoming available in the fifteenth century. They also illuminate the question of language choice. In our period, Persian, Arabic, Sanskrit, and versions of Gujarati were all used for monumental inscriptions. At times more than one language or linguistic convention was used; Page 2 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety thus we have so-called ‘bilingual’ or even ‘trilingual’ inscriptions. Such inscriptions may be particularly revealing: why were two or more languages chosen? With their different scripts, what kind of visual impact might they have had? Such ‘dual’ or ‘multiple’ inscriptions were evidently aimed at different sets of readers, especially as differences in their form and content suggest that they were not exact translations of each other. What religious and legal authority did they invoke? Can we discern patterns in the choice of language for epigraphs? Were there changes in the nature of epigraphic production over the course of the century? What might these changes tell us about language use, religious affiliation, and regional identity under the sultans of Gujarat?

Epigraphic Language and Form From the reigns of Muzaffar I (r. ca. 1398–1411) to the end of the Muzaffarid dynasty in the 1580s, over 300 stone-carved inscriptions (p.189) have survived and been published. The largest number of these are in Persian, some with a quotation from the Qur’an at the outset. A significant number—about 44—are in Arabic, many of which are tombstone inscriptions or donations made by the prosperous Muslim merchants of the Gujarat coastline. About 50 inscriptions are in Sanskrit. From this period, dated inscriptions exclusively in the vernacular— local versions of Gujarati—are fewer. There are 7 such inscriptions, although there may be many more Gujarati inscriptions such as the paliyas commemorating heroes or satis that are as yet unpublished. Making the assumption that those educated in Persian would be able to read at least rudimentary Arabic, I have not considered inscriptions that include both Persian and Arabic as bilingual for the purposes of this study. Similarly, I am assuming here that those who read Sanskrit would also read early Gujarati; inscriptions that include these two languages are not considered bilingual either. Each bilingual inscription discussed below also employs two scripts: versions of Perso-Arabic usually written in naskh, occasionally with elements of tughra, and nagari for Sanskrit and Gujarati. Most bilingual inscriptions are carved into the same piece of stone; a few are on separate pieces. Bilingual inscriptions also use different forms of recording dates. For the Persian and Arabic parts, the dating system is generally according to the Hijri calendar, while the Sanskrit and Gujarati sections use the Vikrama Samvat and Shaka systems. Going by these criteria, there are less than fifteen bilingual inscriptions from Gujarat in our period. The most common are in Persian and Sanskrit. One is in Arabic and Sanskrit, and two inscriptions include text in Arabic, Persian, Sanskrit, and the local vernacular. These language choices were not accidental. I have found no bilingual inscriptions attached exclusively to sectarian structures such as mosques or temples, nor to Jain buildings.2 Most bilingual inscriptions are attached to civic structures such as walls or forts and water structures, especially stepwells. Several record edicts or orders issued by officials for the notice of the local population. Each category will be discussed below. For Page 3 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety context, I have also discussed ‘monolingual’ inscriptions that use phrases or (p. 190) words derived from other languages, as well as selected inscriptions that illuminate the evolving religious and linguistic landscape of sultanate Gujarat.

Pre-sultanate Bilingual Epigraphs The Gujarat sultans and their subjects were not the first in the region to install bilingual inscriptions; the tradition began much earlier. One of the first bilingual inscriptions from Gujarat is the much-discussed Arabic–Sanskrit one from Verawal that dates to 1264.3 The Arabic and Sanskrit texts are on separate stones and the contents of the inscriptions are by no means identical. The Arabic part was carved two months after the Sanskrit. Produced on the orders of a merchant-shipowner from Hormuz, nakhuda Nur al-Din Firuz, it records the building of a mosque, and mentions, among other details, that the local authorities and the guardians of the Somanath temple sold land for the structure. One significant point is that the Sanskrit part begins with an invocation to Visvanatha (om namah śrī viśvanātha), the lord of the world, usually interpreted as a Sanskrit term that would be equivalent to Allah.4 In the vicinity of the Somanath temple, the term Visvanatha would also (p.191) be understood as Shiva, the presiding deity of the ruling Chaulukyas. The inscription also makes use of the terms viśvarūpa (image of the universe), śūnyarūpa (formless), and lakṣālakṣya (visible/invisible), all appropriate ‘translations’ for a formless divinity.5 Bilingual inscriptions continued to be produced in the fourteenth century when Gujarat was ruled tenuously by governors sent out by the sultans of Delhi. One early inscription from eastern Gujarat dates from 1304, after ‘Ala’ al-Din Khalji’s armies had conquered Gujarat, but before sultanate rule had been firmly established.6 The inscription records a land grant made during the rule of Karna Vaghela, a ruler who had been ostensibly defeated and banished by the Khalji army in 1299, but who appears from this inscription to have returned to power in 1304, before he was driven out again. The damaged inscription is carved into a marble tablet, with a cusped arch and a tree motif carved into it. Within the arch frame are twelve lines of Persian in ornamental naskh script, followed by eight lines in Sanskrit. The Persian part relates that during the reign of the just king (ṣadaqāt-i badshāh-i ‘ādil) Ra’i Karan Dev, and the great lords Balchaq and Shadi, Taj al-Din Hasan made a donation of all the territory and proceeds of the village Samba (modern Sampla) for the Jami‘ mosque of distant Cambay. The Sanskrit part below gives the date in the Samvat system, and declares that mahārājadhirāja (king of kings) Ka[rna], malikh śrī Badcak, and śrī Hasan gave the village of Sampa for religious purposes (dharmme) for the Cambay jāme misi (Jami‘ mosque). The Persian part ends with a warning to Muslims to obey the order, and that any who attempted to change it would incur the wrath of God and his prophet, according to Qur’an 2:181.7 The Sanskrit part (p.192) indicates a different Page 4 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety readership: while it is addressed to all the rāṇakas or local landholders of the region, ordering them to honour the endowment, no curses are appended. The inscription is also interesting in showing how Karna Vaghela had allied with Balchaq and Shadi, Mongol chieftains who were renegades from the Khalji army led by Ulugh Khan.8 But little is known about the actual donor, Taj al-Din Hasan; why he sought to harness the revenues of a distant inland village for his mosque; and why he had the deed recorded in two languages.

Bilingual Epigraphs on Civic Structures Several bilingual inscriptions from the period around and after the invasion of Timur in 1398 were associated with civic structures such as fortifications, city walls, or rest-houses. In 1395, the brothers Ya’qub and Musa, officials of the faraway Tughluq ruler of Delhi, reinforced the gates of the port city of Mangrol to protect its inhabitants. Ya‘qub proudly recorded the fact in both Persian and Sanskrit.9 The inscription records that the Tughluq ruler in Delhi was Nusrat Shah (Nasarath in Sanskrit), and that Zafar Khan ruled Gujarat in his name. Nusrat Shah, a grandson of Firuz Shah (r. 1351–88) had been placed on the throne by a court faction. This inscription is one of the very few that testify to Nusrat’s authority.10 We also learn about the structures of authority under Zafar Khan who later ruled as Muzaffar I, the first independent sultan of Gujarat. The Persian part of the inscription explains that Zafar had appointed in every district a commander who would be ‘the overthrower of sedition (fitna) and cherisher of subjects’. One such was our builder Ya‘qub, the governor of Sorath (the Saurashtra peninsula) who worked with his brother Malik Musa, kotwal of the region, to install iron gates and steel chains. (p.193) In both Sanskrit and Persian, the brothers claimed Tamim descent, a term that usually denotes actual or claimed Arab origins.11 The Sanskrit text mentions their father’s name too: Rāya Sultālīyajavayarasī [?]; it is unclear whether this is an indigenized Islamic name or whether it should be read as Raya Multani, or chief of Multan, as Desai does.12 We are also lucky enough to know who composed Malik Ya‘qub’s inscription; his name appears in both Persian and Sanskrit. In Persian, we read that Qadi (Qazi) Badr, son of Zahir, was the author. In Sanskrit, we are told that the Torki (Turkish) script was composed by Kadi Badardin Khoja, son of Jahir. The Sanskrit text gives us the name of the engraver (sūtra[dhāra]) too: Rāṇig, son of Virdhavala.13 Ranig may well have composed the Sanskrit prose section himself; it is a formulaic declaration requiring no particular literary skill. The inscription is not large, nor was it installed in a prominent location. It is an 18″ ×6″ slab of white marble into which are incised twenty-three verses in Persian and six lines of Sanskrit prose. It was set into a small room to the west of the Gadi gate of Mangrol.14 The compositions—both Persian verse and Sanskrit statement—are not particularly elegant, and the calligraphy is functional rather than ornamental. But nevertheless, it was important to the brother officials to Page 5 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety record their achievement in both languages. This would suggest that while they thought of Persian as the language of prestige and poetry, they nevertheless wanted the Sanskrit-reading literati to recognize the key facts of their local authority and contribution: their own names and lineage, the names of the Delhi ruler and his Gujarat deputy, and the names of the author and the engraver. In the surviving bilingual inscriptions from the early fourteenth century, it is more common to find an elaborate Sanskrit section (p.194) accompanied by a rudimentary Persian text. By the late fourteenth century, as in the Mangrol inscription discussed above, the Persian section is longer. An instance of the early style is seen in the Karakhadi inscription from 1340 which has just four lines of Persian set above nine lines of Sanskrit, the latter containing considerably more detail.15 In this text, a village official (mantri) named Mokha Mehta records that he built a well and a mosque on behalf of Malik Muzaffar (Madaphara in Sanskrit), the governor of Baroda. While the calligraphy of the Persian lines is fine Arabic-style naskh, the content does not suggest an educated Persian writer.16 This would suggest that the local engraver was perhaps accustomed to carving gravestone texts in Arabic but did not know Persian. The Persian section ends with the familiar Qur’anic imprecation against anyone who or altered the endowment (Qur’an 2:181). There are several bilingual inscriptions that record civic building activity. An inscription from 1408 records that Mahamalik Phajaral Ahamed (Malik Fazl allah Ahmad Abu Raja in Persian) built a wall around the port city of Verawal. The title ‘mahāmalik’, or great prince, is itself a Sanskrit–Persian compound.17 Both parts of the inscription mention the reigning sultan Muzaffar Shah and include his pre-regnal name, Zafar Khan (Dapharkhān in Sanskrit).18 Later, during the reign of the sultan Qutb al-Din Ahmad, an official named Malik Asad built a fort (Sk. koṭ and Pers. ḥiṣār) around the town (p.195) of Prashnavada in 1457–8. The inscription that gives us this information consists of three lines of Persian in naskh followed by five lines of Sanskrit in the nagari script. The Persian part is not particularly informative other than providing the date, the ruler’s name, and the names of Malik Asad’s father and grandfather.19 The five lines in ‘faulty’ or colloquial Sanskrit give us more. Apart from the names of Asad’s father and grandfather, it relates the names of all the artisans involved in the project: Ṣ[Kh]īmmā, Surā, Ṣ[Kh]alā, Ghīṇā, Suhāmihi, and the scribe, Sal Kāyastha son of Sārangade.20 From the Sanskrit part we hear also that Asad was the governor of Devpatan, better known as Somanath Patan. The rest of the ‘Sanskrit’ is hard to decipher: amārati sāhāṇḍ saravīlavā bīrāsilā, or, alternatively, amārati sāhāṇḍ sarakhīla kabīrāsilā.21

Bilingual Edicts and the Ass Curse From this period, there are at least three official edicts and orders in two languages and scripts. To make the import of the edict entirely clear, two were Page 6 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety followed by obscene engravings featuring animals that graphically depicted the fate that would befall anyone who breached the order.22 A fragmentary and undated inscription in Persian and old Gujarati from the reign of Ahmad Shah (r. 1411–42) relates the name and titles of the king and one Malik Mujahid al-Mulk. It also includes a phrase bringing (p.196) a curse on Hindus and Muslims (‘ahd-i khudā shikasta wa agar hindū yā musalmān bāshad) for the violation of some order or provision, now unfortunately illegible.23 The Gujarati part was shallowly incised and has now deteriorated. A similar badly damaged pillar inscription from Baroda district has text in Persian and Sanskrit and dates from the rule of Muhammad Shah II in 1443. The sun and moon are carved on the top, and at the bottom again are animal figures. Little remains of this edict which appears to abolish certain taxes.24 Another such is a damaged inscription from 1455, during the reign of Qutb al-Din Ahmad II. Incised on a slab of yellow sandstone, the Persian part consists of a royal order instructing the officials of Patan-Sukhnath (Somanath) to put an end to an exploitative local practice. A barely legible word at the end of the line has meant that it is not entirely clear what the practice was, but it appears to have been a levy imposed at the time of the arrival of new officials. Breaching the order would amount to breaking the oath of God and the Prophet for Muslims, and for kafirs, that of Sarsati and Satwanti, ‘whom they worship’.25 The request came from the ‘ulama, the qazis, important people, officials, and the Mahajan, the latter a committee of merchants (p.197) and prominent local citizens. The 18 lines of Persian are followed by 6 lines of Sanskrit, unfortunately too faded to be deciphered.

Stepwell Inscriptions in the Sultanate Water structures—primarily stepwells, but also wells and tanks—were built throughout sultanate Gujarat, almost always by merchants or courtiers. While one might have assumed that rulers would have some hand in the provision of water in an arid region such as Gujarat, it was, in fact, subordinates and merchants who usually claimed responsibility. The accompanying ‘monolingual’ inscriptions—all in either Sanskrit or Persian—give some sense of their motives and show how groups and individuals, often newly enriched by their association with the sultanate, proclaimed their benevolence and loyalty in their stepwell inscriptions. Inscriptions in a single language—Sanskrit, Arabic, or Persian— were most common, although there are several bilingual epigraphs that will be discussed in the next section. Supplying water along the trade routes or in unirrigated villages was a source of political and spiritual merit and donors were anxious to make clear their generous intentions towards all humanity and animals. Malik Haji, a courtier, declared in an Arabic inscription in 1460 that his fine garden-house with its lofty porch, four painted walls, fruit-bearing trees, and a well and a tank had been constructed for the benefit of all men and animals.26 Many stepwell inscriptions Page 7 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety carry such benevolent messages. Water architecture was not entirely distinct from religious architecture; although it served a practical need, it was also a merit-generating act of philanthropy. Although several Muslim donors such as Malik Haji built stepwells, most were associated with Hindu donors and embellished with Hindu iconography. For wealthy Hindus, stepwells were a politically safe way to show off their power and influence in the sultanate order as temples were associated with Hindu kingship and were the target of periodic bouts of iconoclasm.27 While few new Hindu temples were built in this period (p.198) (although patrons did record the renovation of certain old ones), water structures were not considered sites for rebellious consolidation. In spite of the declared hopes of their builders that their wells would benefit all mankind and animals, it is hard to judge how access to stepwells was regulated. Were they indeed meant for all, regardless of community, as Malik Haji’s pious claim would suggest? While it is hard to judge who used stepwells, it has been suggested that the religious nature of stepwells changed from the fourteenth century with Muslim influence. From being sacred public buildings, perhaps, they became private retreats for the elite during the hot summer.28 There are virtually no Jain sponsors of stepwells.29 Later stepwell inscriptions indicate a growing link between the state, local chieftains, and trade.30 Most stepwell inscriptions commissioned by traders do not fail to mention the chieftain in charge of the locality and the reign of the sultan in which the well was built. One such inscription that survives was struck in 1480 in Sanskrit and Gujarati and comes from Gosa in Porbandar taluka.31 The inscription records the building of a stepwell by Munja, son of Sura, of the vania caste, in the reign of Sultan Mahmud and during the time of the Jethva ruler Vikramait. Two years later, four Sanskrit inscriptions were struck in a well in Rampura in Wadhwan district that record the reign of Sultan Mahmud (pātasā śrī Mahimūda), the local chief Rana Vaghji, in addition to the officers in charge of the district—Parmar Lakhdhir, Parmar Hada, and Khānśrī Alukhan. The well was built by Ranibai and Valhade, wives of the merchant Vina of the Shrimala caste and resident of Jhanjhnagar (Jhinjhuvada). Most such inscriptions give us a picture of an evolving adminstrative hierarchy—from Rajput or Muslim official upwards through the local Rajput overlord to the sultan. It was important for the newly wealthy to announce their achievement to the government; (p.199) it was equally important for the government to acknowledge and ratify such proclamations of wealth and civic pride made by merchants and their families. This mode of inscription-making, mostly on the walls of stepwells, continued throughout the reigns of Mahmud and his successors. In 1482, two long and elaborate Sanskrit inscriptions were commissioned by Dhanaka of the Mer (Mehar) community to accompany a stepwell he had built in the port of Cambay. The first inscription begins by praising the reign of the sultan. The second announces that the virtuous Dhanaka built the stepwell for the love of Vishnu Page 8 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety and had it embellished with elephants, fretwork, and niches. We are told that the prashasti was composed by Hala, of the scribal Nagar Brahmin community and that the engravers were Vanak and Khetak. In Dhanaka’s lavish inscription, we see how a member of the Mer community, formerly a militant pastoralist group, emerges as an urban notable during the reign of Mahmud.32 A stepwell inscription in Sanskrit of 1498 from Bhoj in Baroda district gives a list of donors that includes Kaja from the Chahamana lineage, variously described as merchant and king, and other individuals from mercantile castes, their wives and daughters.33 Kaja seems to have been in charge of at least 84 villages that were under the administration of Bhupanarayan, a minister described as a ‘wise trader from the Disaval caste’. This is an example of how a ‘princely’ Rajput chieftain was, by the end of the fifteenth century, a trader with administrative and revenue-collecting powers. While these are not bilingual inscriptions (they are in Sanskrit), they show how Hindu merchants and members of Rajput clans had come by the late fifteenth century to conduct the administrative work of the sultanate, and to embellish their constructions with Sanskrit verbiage. In 1499, one of the most famous stepwells of Gujarat, the Adalaj vav near Ahmadabad, was built under the patronage of Rani Rudadevi, (p.200) wife of a Vaghela chieftain.34 The inscription (mostly in Sanskrit with a few lines in Gujarati) records the building of the well at a cost of five lakh tankas. Significantly, the inscription records the lineages of both her husband’s and father’s families, but does not fail to mention the reign of Mahmud Begada. Members of non-Muslim clans such as the Mers, Chahamana, and Vaghelas record in these inscriptions their role as arbiters of the sultanate hierarchy. Their ability to build major structures testifies to their prosperity in the sultanate economy, then burgeoning with a rapid expansion of trade during Mahmud Begada’s reign. The choice of literary Sanskrit in these late-fifteenthcentury inscriptions suggests that Sanskrit learning was easily available.35 The stepwell inscriptions come from all over Gujarat and show the ubiquity of the process of accommodating erstwhile pastoralist clans into the new economy. The fifteenth century had thus seen the pacification of militant pastoralist groups— the Rajputs of the future—and their settlement into an accommodative triangular relationship between the sultan, the trader, and the chieftain. In certain areas, the roles of merchant and chieftain were hardly distinguishable. The settlement of land was instrumental in the stabilization of that other source of wealth—trade. An inscription struck in 1525 to mark the construction of a stepwell by Dharaji, a notable from Mansa in north Gujarat, provides an idea of the process by which erstwhile pastoralist groups were settling down to a mercantile identity by the fifteenth century.36 Given that the donor Dhara had adult grandchildren in 1525, it can be safely assumed that he belonged to a generation that was active under Sultan Mahmud in the fifteenth century. The inscription, which begins by acknowledging the auspicious reign of Muzaffar Shah II, is partially in Sanskrit verse and the rest follows in Gujarati prose. It Page 9 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety lists the descendants of a junior or allied branch of the erstwhile militarypastoralist family of the Vaghelas who had ruled Gujarat in the thirteenth century. The text is long and well-formulated and indicates a literate culture and audience. (p.201) The composition of this history of a minor clan in Sanskrit verse is a legitimizing gesture by an upwardly mobile clan with a pastoralist past. In a Sanskrit quartet in the middle of the inscription, Dhara claims to be devotee of Hari (Vishnu), a warrior and king. Another verse in Sanskrit indicates that he was also a prosperous merchant of the town. This verse praises the sea and honours Varuna, the sea-god, as the guarantor of his well-being, indicating that his wealth was achieved through the sea-trade.37 However, the family seems to have aligned itself seamlessly with other sectarian groups too. The names of Dhara’s sons from one wife—Mīyā Śrī Phatūlā, Kesnājī, Bhābhujī, Arjanjī, and Bhīmjī—are a combination of Vaishnava and Islamic names. One grandson from Dhara’s daughter Rajbai is named Malik Śrī Nijāmal Malik Savāī, clearly a rendering of a sultanate title. The sons from his other wife have clearly Vaishnava names: Rāmjī, Lakhmanjī, and Nārāyanjī. Given that Dhara is a devotee of Hari and his offspring are named after avataras of Vishnu, it is likely that he was claiming a Vaishnava identity. This did not prevent at least one son and a daughter being allied to a Muslim identity.38 A classicizing impulse was at work here that shows a member of a pastoralist group staking a claim for higher ritual and cultural status as chieftain, merchant, and Vaishnava. However, the route to this status is two-fold: through Dhara’s own success as a merchant with the blessings of Varuna and through the employment of his offspring as servants of the sultans. The inscription brings together two important themes of this period—the stabilization of pastoralist clans under the sultanate—and the stabilization of sectarian identities under formal institutionalized sects like Vaishnavism.39

Bilingual Inscriptions on Water Architecture While most stepwell inscriptions are in either Persian or Sanskrit, or in Old Gujarati, there are several bilingual inscriptions too. One of the first is from Petlad, an Arabic-Sanskrit bilingual record from 1323 in two parts found close to a structure now known as the dargah of Baba Arjun Shah. Both parts of the inscription state that Ghiyas al-Din Tughluq ruled (p.202) Delhi at the time.40 In Arabic, Haji Isma‘il recorded that he had a well built in Petlad for public use near the tomb of a Shaykh. The son of ‘Usman Shirazi and a resident of Cambay, Isma‘il had received a grant of 20 qunthas (kubhas in Sanskrit) of land from the rights-holder or muqta‘ of Petlad to build a well for the benefit of locals and outsiders. Isma‘il’s name is damaged in the Sanskrit part which also suggests that the well was not a new structure but a renovation (jīrṇoddhār). The Sanskrit part explains that the land was then under the sway (sannidhau) of Śrī Arjan Ghori, and Isma‘il had secured planning permission from the divan of Anahillapattan, the capital of sultanate Gujarat, and the official of Petlad, Page 10 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety Badaradin Abubak Ahmad Amir Koh. The Arabic part ends with a plea to future amirs to maintain the structure. In Sathod, also in eastern Gujarat, Dungar, Tikam, and Natha repaired a dilapidated stepwell in the mid-fourteenth century. They had reconstructed the well ‘out of their own money, so that men and animals of the village and those passing by’ could drink its water. Seventeen years later, in 1383, two officials (wajhdars) of the village of Bakrol, Akram and Nizam Aflah Muhammad Shahi, formally attested that 16 bighas of land had been granted to Dungar and Natha for the maintenance of the well. This inscription is unusual in using three languages: fifteen lines of Persian in naskh on top, followed by sections in Old Gujarati and Sanskrit. In the Persian part, they granted possession of the land to Dungar and Natha so that ‘every year from the date of this deed, they and their lessees might sow the Kharif and Rabi crops there…’. The wajhdars swore that if they or any of their agents interfered with the cultivation, their wives would be considered divorced from them, the attestors (zanān-i muqirrān-i maẕkūr az muqirrān-i maẕkūr muṭallaqah bāshand)! In addition they would have breached the pledge to God and the Prophet (‘ahd-i khudā’ī wa rasūl-i khudā’ī shikaste bāshand). As in a formal deed, the attestors declared that the contents were true and that it was signed by their own hands. Additionally, it was witnessed by one Daulatshah. The Persian text is carved in relief. It dominates the deed and forms the legal part of the inscription.41 It is followed by fourteen lines in Old (p.203) Gujarati, incised into the stone, and thus less legible. The Gujarati part was challenging for its editor, who only remarked that it included a list of names preceded by their titles such as raja, thakur, purohit, pandit, and the clan name Vaghela.42 Here Dungar is called Shah Dungarshi, the title shah suggesting that he was a merchant, and the deed declares that the stepwell be enjoyed by the brothers Shah Dungarshi, Kelha, and Delha and their descendants. At the end are seven lines of versified Sanskrit that give the genealogy of the brothers, who built the stepwell for the merit of their late parents. Dungar’s mother Punji is mentioned here as one of the builders. The two official attestors of Bakrol do not appear in this part of the deed. The Old Gujarati section ends with a curse on those who ignore the deed: violators would have committed the four sins of having killed a cow, a Brahmin, a woman, and a child. The Sanskrit part is followed by an obscene carving of a woman having sex with an ass—a common imprecation, as we have seen, in medieval Gujarat. The Sathod inscription is revealing of subtle transitions in the power hierarchy of late fourteenth-century north India. It dates from the period when governors from Delhi still ruled Gujarat. Although the initial part of the Persian section is damaged, and thus likely missing the name of the current ruler (Firuz Shah), it nevertheless shows changes to the local land hierarchy. While Dungar and his brothers had repaired the stepwell at their own expense, they received land for Page 11 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety its maintenance. Dungar also claims the title of shah, or merchant, suggesting that land and mercantile interests were not incompatible, and that merchants were being incorporated into the evolving system or that landholders were taking to mercantile activities and titles. It was local Sultanate officials, wajhdars, who witnessed and signed the deed.43 With its language of signing and witnessing, the stone-carved deed may even be a direct rendering of a manuscript version. According to Desai, the relief-carved naskh of the Persian part ‘betrays strong tendencies of the cursive type of writing which came to be generally employed in deeds and documents at a later period’.44 The brothers took care to make this information available in (p.204) all three languages of governance: in Persian for the notice of sultanate officialdom, in Gujarati for the notice of the local notables, and a genealogy in Sanskrit to claim a respectable lineage. Each segment enjoins compliance with a different threat: the Persianlanguage peril of dissolving marriages and broken Islamic allegiance; the Gujarati sin of killing cow, Brahmin, woman, or child; and finally, the graphic and obscene gadheḍā gāḷ, ass curse, clear without recourse to language. In 1405, Malik Nathu built a stepwell and mosque in Baroda. There were formerly two inscriptions set into the entrance of the well, one bilingual in Persian and Sanskrit, the other exclusively in Persian. Only the latter is now in situ, while the former is set into the entrance of the city’s Jami‘ mosque, likely relocated from the stepwell. The bilingual inscription records the construction of a stepwell by Mira Natha son of Takshara in Sanskrit, Amir Nathu in Persian. The more elaborate Persian epigraph tells us that Nasir al-Daulat wa’l-Din Amir Nathu, the son of Takhir, built this stepwell during the time of the great khan, Zafar Khan, son of Wajih al-Mulk, the muqta‘ or rights-holder of the province of Gujarat. Nathu’s project came about through the good offices of the governor of the east, Malik Adam b. Sulayman.45 By siting his inscription at the entrance of his well, Nathu was signalling to different political constituencies at a time of political uncertainly when the governor of Gujarat, Zafar Khan, had yet to declare himself the first independent sultan. Nathu’s inscriptions, as well as the trilingual Sathod inscription described above, were thus legal texts, imprinting the benevolence and position of their patrons in the chain of authority in the standard languages of governance. While most bilingual stepwell inscriptions come from the fourteenth and early fifteenth centuries, there is one important exception. In 1499, a female courtier named Bai Harir Sultani had a lavish stepwell built at Asarwa, near Ahmadabad. To commemorate her construction, she had inscriptions written in Arabic and Sanskrit, set into opposite walls of the well, facing each other.46 The Sanskrit part, intriguingly, begins (p.205) with an invocation of ‘sṛṣṭi-kartā, lord of water, giver of life, Varuna’.47 The briefer Arabic inscription begins directly, without invocation: ‘This graceful building and noble edifice, with a lofty porch and painted four walls, well and tank were built and fruit-bearing trees planted … for public benefit in the reign of Mahmud Shah’.48 In Sanskrit, Bai Harir Page 12 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety claims to have been a senior courtier—the guardian of the inner quarters—of Sultan Mahmud and to have constructed the well at the expense of 3.29 lakhs for the thirsty humans, animals, birds, trees, and all the 84 lakh creatures. The well was built by Malik Bihamad, obedient to the orders of the Supreme God (parameśvarājñāpālaka), Gajadhara Vaishya, the son of (or architect?) Vira, and four other helpers. This inscription is unusual in several respects. Firstly, while donations and constructions by women were by no means uncommon in sultanate Gujarat, this deed by a single female slave-attendant has no parallel.49 Secondly, it is the only bilingual inscription from the Gujarat Sultanate in which Arabic is the language of choice as opposed to Persian. This would suggest that Bai Harir was making a statement through her choice of language, even though the Arabic part is shorter and less informative than the Sanskrit. Her name means ‘silk’ in Arabic, and the suffix ‘sultani’ suggests that she was a slave-official. Could she have been an Arabic speaker? She is clear in both languages that her considerable expenditure was for the public good.

‘Muslim’ Epigraphs in Sanskrit and Gujarati It would be wrong to assume that Muslims exclusively used Persian and Arabic for their records while non-Muslims employed Sanskrit or Gujarati. In addition to the bilingual inscriptions discussed above, there are several inscriptions associated with mosques and Muslim builders (p.206) that are exclusively in Sanskrit and Gujarati. In 1497, one Sandal Sultani built a stepwell, a mosque, and a hazira near the sultanate capital of Champaner. The accompanying sixteen-line inscription, which uses a mixture of Sanskrit and Old Gujarati phrases with a few Perso-Arabic words thrown in, testifies that he had received a grant of land for religious purposes (dharmārthe).50 While Sandal claims to have built a mosque, the inscription invokes the famous goddess Kalika mata of Champaner, and begins with routine invocations to Ganapati and Sharada. While Sandal may have been a Muslim, his epigraph would suggest that in the matter of seeking spiritual favour, he was hedging his bets. Another inscription from 1499 found in the suburb of Mahmudabad near Ahmadabad is in Sanskrit and includes words in Old Gujarati and Persian.51 Unusually, there is no surviving dedication to a deity; it begins with auspicious astrological signs in Sanskrit. It is now set in a wall of an old sarai, but it was previously set into the wall of a stepwell in Mahmudabad. It records that in Samvat 1522 (1465 CE), during the victorious reign of the Gurjara Patshah Shri Mahimud, Malik Mahimud Nijam, the architect (sūtradhāra) Shangha and the modi Bala together built the city of Mahimudabad. The longevity of the town— 500 years, 7 months, and 13 days—was reckoned according to the sarvaprakāśagrantha (the all-illuminating volume = the Qur’an?). The next phase of building was commenced thirty-one years later, in 1496, and shortly afterwards, Mahamalik Sardul, the son of Durjan of the Nahada clan of Page 13 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety Mandovar near modern Jodhpur, entitled sarkasaduphala (governor of the east?), had the following constructions built (in Gujarati, eṭalī amārati nīpajāvī): the torana (Gujarati for gateway), mashid (mosque), and sarah (Persian for resthouse). He did so on the order of (hakamapharamāṇa = Pers. hukm-farmān), the ruler Mahmud, with the help of village officials, sculptors Sura and Shakarai, Shreshthi Shekha, Doshi Ratna, Parekh Khima, and ‘the people’.52 While most sultanate inscriptions are in Persian, it was not unusual for Sanskrit inscriptions to be commissioned by Muslim officials throughout (p.207) the duration of the sultanate. One of the most elaborate ones is the Dahod inscription of 1488. It is interesting that the first inscription from the Gujarat sultanate that offers official information about the ruler’s conquests and deeds should be in Sanskrit.53 It appears to have been ordered by the courtier ‘Imad alMulk after the construction of Dahod (Dadhipadra, Dohad) fort. It is a relatively large inscription of twenty-two lines, carved on a painted stone slab 3′3″ ×1′7″. The text is a historical one, giving us the deeds and genealogy of the reigning sultan Mahmud Begada. The inscription begins with an invocation to Kashmirvasini devi, the goddess who resides in Kashmir, or Sarasvati. The editor surmises that this was because the Sharada temple in Kashmir was popular throughout India in the fifteenth and sixteenth centuries.54 But Sarasvati appears in other sultanate references too—a play staged at the court of Muzaffar II, starred his favourite courtesan in the role of Sarasvati.55 Part of the narrative in the mahakavya written about Mahmud, the Rājavinoda, is an account of Sarasvati’s visit to Mahmud’s court. The goddess Sarasvati appears to have had a particular appeal for Mahmud Begada and his son Muzaffar.56 In 1538, an official named Malik Rakunal (?Rukn al-) had a dharmashala or resthouse built in the ancient city of Patan at the order of Dariyakhan, a courtier from the time of Pātsāh Mihimūd, or Mahmud III, who ruled between 1536 and 1554. Mahmud was only thirteen at the time of the inscription, and the real power in the sultanate rested with Darya Khan.57 The epigraph that records this matter—the construction of a resting place by a Muslim official on the order of a Muslim courtier (p.208) during the reign of a Muslim sultan—is a white marble tablet with twelve lines of Sanskrit prose. While the language is not flawless, and is ‘loose like that of legal documents’, it is easily comprehensible.58 Four lines of the epigraph shower praise on the young king in classical prashasti tradition. Like Vishnu, he is the one garlanded by Rajyalakshmi, the goddess of (political) fortune; he is the scorcher of enemies; the victor over the darkness of injustice; the protector of his people. Most interestingly, he is compared to alakṣajagadīśvara, the invisible or attributeless lord of the universe, a term used in Sanskrit as the equivalent of the indescribable Allah.59 The Sanskrit or Old Gujarati inscriptions that were commissioned by Muslim courtiers or officials often begin with standard phrases or invocations to what might now be deemed ‘Hindu’ deities. The inscription from Mahmudabad in Page 14 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety 1499 cited above begins with astrological signs and invokes the Sun. The Mansa inscription of Dharaji, with a Muslim-Vaishnava family, invokes Varuna. The Sanskrit part of Bai Harir’s stepwell inscription of Asarva invokes, again, Varuna. Sandal Sultani’s Sanskrit inscription in Champaner, the capital built by Sultan Mahmud Begada, commissioned to commemorate the building of a stepwell and accompanying mosque, begins with an invocation of Ganesha and Sharada, and includes praise of Kalika mata.60 ‘Imad al-Mulk’s elaborate Sanskrit epigraph from Dahod begins with an invocation to ‘Kāśmīrvāsini devi’, the goddess who resides in Kashmir, a deity identified with Sarasvati. Language, in the case of the inscriptions of sultanate Gujarat, did not bear the exclusivist sectarian burden of subsequent times. Inscriptions from Gujarat in the fifteenth century were associated with new forms of governance under the sultanate. In language use and literary convention, they represent the emergence of formal ways of representing land relations, local hierarchies, and religious affiliation. As these are formal, even legal documents, Persian, Arabic, and Sanskrit (p.209) dominate as linguistic choices. However, there is a significant minority of inscriptions that are written in local versions of Gujarati, while others include Gujarati words and phrases. The bilingual inscriptions suggest that their patrons wished to signal towards two separate traditions of formal claim-making, and accordingly, these inscriptions undertake a limited translation of concepts from one vocabulary to another—Arabic to Sanskrit or Sanskrit to Persian, as the case might be. Elsewhere, however, the bilingual inscriptions do not translate each other, but offer information that is significant in the linguistic tradition of choice. What is most striking about the entire range of fifteenth-century inscriptions from Gujarat, however, is not their linguistic choices, but how the sultanate had managed to wield together an overarching ‘Gujarati’ identity for the region, an identity that was almost universally accepted by the composers and patrons of epigraphs. (p.210) Notes:

(*) For points of clarification, I am grateful to Sunil Sharma and the anonymous reviewers; errors are mine alone. (1) The deaths of warriors or satis were also recorded on stone inscriptions, as were those of Muslim notables on gravestones, but these will not be discussed here. On warrior and sati stones from medieval Gujarat, see Eberhard Fischer and Haku Shah, ‘Vetra ne Khambha’: Memorials for the Dead, Wooden Figures and Memorial Slabs of Chodhri, Gamit, and Vasava Tribes, South Gujarat, India, Ahmedabad: Gujarat Vidyapith, 1973; Romila Thapar, ‘Death and the Hero’, in Mortality and Immortality: The Anthropology and Archaeology of Death, S.C. Humphreys and Helen King (eds), London: Academic Press, 1981, pp. 293–315; articles by Sarayu Doshi and Jyotindra Jain in Memorial Stones, S. Settar and G.D. Sontheimer (eds), Dharwad, 1982; and B. Shelat, ‘The Memorial Stone Page 15 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety (Pāḷiyā) Inscriptions: The Cultural Heritage of Gujarat, Saurashtra and Kutch’, in Script and Image: Papers on Art and Epigraphy, Adalbert J. Gail, Gerd J.R. Mevissen, and Richard Salomon (eds), Delhi: Motilal Banarsidass, 2006, pp. 185– 216. On gravestones, see E. Lambourn, ‘Carving and Communities: Marble Carving for Muslim Patrons at Cambay and around the Indian Ocean Rim (Late 13th–Mid-15th Centuries C.E.)’, Ars Orientalis, 34 (2004): 110–35. (2) Although Malik Nathu’s bilingual inscription of 1405 is now set into the Jami‘ mosque of Baroda, it refers to the building of a stepwell and was probably originally located there. See p. 204 for further discussion. (3) There is a substantial literature on this inscription. Discussions of the Sanskrit part can be found in B.D. Chattopadhyaya, Representing the Other? Sanskrit Sources and the Muslims (8th–14th Century), New Delhi: Manohar, 1998, pp. 71–6; R. Chakravarti, ‘Nakhuda Nuruddin Firuz at Somanath: AD 1264’, in Trade and Traders in Early Indian Society, Delhi: Manohar, 2002, pp. 220–42; and Romila Thapar, Somanatha: The Many Voices of a History, London: Verso Books, 2005, pp. 88–102. For a discussion of the Arabic part, see Alka Patel, Building Communities in Gujarāt: Architecture and Society During the Twelfth through Fourteenth Centuries, Leiden, Boston: Brill, 2004, ch. 2 (pp. 52– 4, 73), and ‘Communities in Collaboration: Interpreting the Somanātha-Verāval Inscription of 1264 CE’, in Ancient India and Its Wider World, Carla M. Sinopoli and Grant Parker (eds), Ann Arbor: University of Michigan Press, 2008. The Sanskrit record is published in D.C. Sircar, ‘Veraval Inscription of Chalukya Vaghela Arjuna, 1264 A.D.’, Epigraphia Indica, 35(3), 1961–2: 141–50, and the Arabic in Z.A. Desai, ‘Inscription of 1264 AD from Prabhas Patan’, Epigraphia India: Arabic and Persian Supplement, 1961: 10–15. (4) Chattopadhyaya (1998), pp. 75–6. (5) For arguments to conceptualize Hindu–Muslim encounter as exercises in translation, see Tony K. Stewart, ‘In Search of Equivalence: Conceiving MuslimHindu Encounter Through Translation Theory’, History of Religions, 40(3), 2001: 260–87, and Finbarr B. Flood, Objects of Translation: Material Culture and Medieval ‘Hindu-Muslim’ Encounter, Princeton, NJ: Princeton University Press, 2009. (6) Z.A. Desai, ‘A Persian-Sanskrit Inscription of Karna Deva Vaghela of Gujarat’, Epigraphia Indica: Arabic and Persian Supplement, (hereafter EIAPS) 1975: 13– 25. (7) Qur’an 2:181. And whoso changeth (the will) after he hath heard it—the sin thereof is only upon those who change it. Lo! Allah is Hearer, Knower. (8) Desai (1975): 16, citing ‘Isami, Futūḥ al-salāṭīn, Agra, 1938, pp. 244–5.

Page 16 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety (9) The Persian part is published in Z.A. Desai, ‘Khalji and Tughluq Inscriptions from Gujarat’, EIAPS, 1962: 1–40, 32; the Sanskrit part appeared in D.B. Diskalkar, Inscriptions of Kathiawad (hereafter IK), Poona: reprinted from New Indian Antiquary, 1938–41, no. 55, and is reprinted in H.G. Shastri, Historical Inscriptions of Gujarat (hereafter HIG), vol. 4, The Sultanate Period, Bombay: Forbes Gujarati Sabha, 1979, p. 12. (10) Desai, EIAPS (1962), p. 32. (11) M. Lecker, ‘Tamīm b. Murr (or Tamīm bt. Murr, when the tribe or ḳabīla is referred to)’, in Encyclopaedia of Islam, P. Bearman, Th. Bianquis, C.E. Bosworth, E. van Donzel, and W.P. Heinrichs, (eds), Leiden: E.J. Brill, second edition, online. (12) Desai, EIAPS (1962), p. 34. (13) Shastri, HIG (1979), p. 12. (14) At least it was there in the late 19th century when the compilers of the Corpus Inscriptionum Bhavanagari noticed it first. It has since been moved to Junagadh Museum. Corpus Inscriptionum Bhavnagari: Being a Selection of Arabic and Persian Inscriptions Collected by the Antiquarian Department, Bhavnagar State, 1889, p. 16; Desai, EIAPS (1962), p. 34. (15) The Karakhadi inscription was first published in G. Yazdani and R.G. Gyani, Important Inscriptions from the Baroda State, vol. 2, Muslim Inscriptions, Baroda: Baroda State Press, 1944, p. 1. The Sanskrit part has not been edited since then. The Persian part is discussed by Z.A. Desai in Persian and Arabic Epigraphy of Gujarat, Baroda: M.S. University, 1982, p. 16, and in ‘On Karakhdi Bilingual Inscription of 1340’, Journal of the Baroda Museum and Picture Gallery, 12, 1960, pp. 33–5. Desai offered a corrected reading in Corpus of Persian and Arabic Inscriptions in the Museums of Gujarat, Vadodara (Baroda): Director, Department of Museums, Gujarat State, 1985, pp. 4–6. (16) Desai (1960), pp. 33–5. (17) For other titles used in Sanskrit inscriptions for Muslims, see Chattopadhyaya (1998), pp. 28–60. (18) The Sanskrit part is published in Diskalkar, IK (1938–41), no. 63, and reprinted in Shastri, HIG (1979), no. 6; The Persian part is in Desai, EIAPS (1953–4), p. 64. See also Desai (1982), p. 18. (19) The Persian text is transcribed and discussed in Desai (1985), pp. 40–4.

Page 17 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety (20) The Sanskrit part is published in Diskalkar, IK (1938–41), no. 78; Shastri, HIG (1979), no 9, p. 20. The Persian is in M. Nazim, ‘Inscriptions from the Bombay Presidency’, Epigraphia Indo-Moslemica, 1935–6, p. 48, XXXb. For a list of other readings of this inscription, see Z.A. Desai, Arabic, Persian and Urdu Inscriptions of West India: A Topographical List, New Delhi: Sundeep Prakashan, 1999, no. 1153. (21) Desai reads this phrase as containing the title or name Sarkhail Kabir; Desai (1985), p. 42. (22) An elaborate form of the ass curse can be found in the fourteenth-century Jagaḍucarita, the verse biography in Jain Sanskrit of a Jain merchant from Kachchh. See Sarvananda, The Jagaducharita of Sarvananda. A Historical Romance from Gujarat, G. Bühler (ed.), Wien: Sitzungberichte der Kais. Akademie der Wissenschaften in Wien, 1892, 5:35. The ass curse appears on several inscriptions in the period, including the Sathod inscription discussed on p. 203. (23) Desai (1999), no. 551; Desai (1985), p. 24. (24) Desai (1985), no. 4. See also Z.A. Desai, ‘Inscriptions of the Gujarat Sultans’, EIAPS (1963), pp. 22–3. (25) The text goes thus: ‘uhda-dārān-i kotwālī-yi qaṣba-yi patan sukhnāth bedānīd [bedanand] ke ba-āmadan-i āyinda az khāna b‘ażī khalq allāh kathā mīkashīdand bar īshān ḥaraj wa ẕulm wa ta‘ddī wa f‘il-ī nāmashrū‘ mīshud ba-ittifāq-i ‘ulamā wa qużāt wa mashāhīr wa thānedārān wa sarān-i gurūh [?] wa ‘uhdadārān-i pargana wa saudāgarān wa mahājanān-i maẕkūr f‘il-ī nāmashrū‘ dūr kardand.

(Desai, ‘Inscriptions from the Museum of Antiquities, Junagadh’, EIAPS (1955–6): 95–6.) The critical term is ‘kat-hā mīkashīdand’. The compilers of the Corpus Inscriptionum Bhavanagari translated the term as ‘bedsteads’, the plural of the Hindi khāṭ, suggesting that new officials demanded beds. Although Desai repeated the bedstead translation in 1985, it seems unlikely; Desai (1985), p. 31. (26) Desai (1999), no. 17. (27) On the political patterns of iconoclasm, see Richard M. Eaton, ‘Temple Desecration and Indo-Muslim States’, in Essays on Islam and Indian History, New Delhi: Oxford University Press, 2000, pp. 94–132.

Page 18 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety (28) Jutta Jain-Neubauer, The Stepwells of Gujarat in Art Historical Perspective, Delhi: Abhinav Publications, 1981, p. xv. (29) While donations to Jain temples continued throughout the sultanate period, there are only about ten instances of donations to Hindu temples recorded during our period. (30) On architecture and politics, see Sheikh, Forging a Region, ch. 2. (31) Diskalkar, IK (1938–41), no. 85. (32) Pravin Chandra Parikh and Bharati Shelat, Gujarat-na abhilekho: svadhyay ane samiksha [Inscriptions of Gujarat: Study and Analysis]), 1991, pp. 94–101, Shastri, HIG (1979), nos 14–15. (33) M.N. Katti, ed., Annual Report on Indian Epigraphy for 1986–87, New Delhi: Archaeological Survey of India, 1998, Shastri, HIG (1979), no. 19A. (34) James Burgess and Henry Cousens, Revised List of Antiquarian Remains in the Bombay Presidency. Bombay, 1897, pp. 301–11; Shastri, HIG (1979), pp. 19– 23. (35) See Chapter 8 in this volume. (36) Shastri, HIG (1979), pp. 63–6. (37) The invocation of Varuna recalls the inscription of Bai Harir discussed below. (38) Mīyā Śrī Phatūlā’s wife was Bībī Śrī Lālājī and their sons had Islamic names: Alījī, Rājejī and Cāndajī. Shastri, HIG (1979), pp. 64–6. (39) The inscription and its implications are further discussed in Sheikh, Forging a Region, ch. 5. (40) The Sanskrit part is published in Shastri, HIG (1979), pp. 7–8 and the Persian part in G. Yazdani, ‘Arabic Inscription in the Shrine of Arjun Shah at Petlad (Baroda State)’, Epigraphia Indo-Moslemica, 1915–16: pp. 15–18. (41) The inscription is now in Baroda Museum. For the Persian part, see Desai (1985), no. 2. (42) The Sanskrit and Old Gujarati parts of the inscription are described in B.L. Mankad, ‘A Trilingual Stone Inscription from Sathod’, Bulletin of the Baroda Museum and Picture Gallery, 10–12, pp. 73–6. (43) Desai (1985), p. 9n1. According to Desai, the term means ‘holder of cropshare’ and continues to be used in Gujarat for landowners and jagirdars.

Page 19 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety (44) Desai (1985), p. 9. According to Desai, this is also an early use of the term khaṭ, or letter of the alphabet, for a deed. (45) Desai (1982), p. 18. (46) The Sanskrit part is in Shastri, HIG (1979), no. 21, pp. 63–4. For the Arabic, see H. Blochmann, ‘Eight Arabic and Persian Inscriptions from Ahmadabad’, Indian Antiquary, 4, 1875: 367–8, and M.A. Chaghatai, Muslim Monuments of Ahmedabad through Their Inscriptions, Poona: Deccan College Research Institute, 1942, p. 70. (47) Sṛṣṭi-kartā or creator can be considered an equivalent term for Allah. The term appears also in the Batihagarh inscription of Jallala Khoja from 1328, and in ‘Adil Shah Faruqi’s bilingual Arabic-Sanskrit mosque inscription of 1590 from Burhanpur. See Chattopadhyaya (1998), p. 76n32. In the Bai Harir inscription, however, there seems to be the intriguing possibility of equivalence drawn between Allah and Varuna, the deity of water. (48) Chaghatai (1942), p. 70. (49) For temple and mosque donations by women in sultanate Gujarat, see Sheikh (2009), pp. 84–5. (50) V. Sonavane, ‘Cāmpāner Māṇḍvī vāv-no samvat 1554 (Śaka 1419) no śilālekh’ [Inscription of the Champaner-Mandvi Stepwell Dated Samvat 1554 (Śaka 1419)], Svadhyāya, 9(4), 1972: 433–5. (51) Shastri, HIG (1979), no. 20a, pp. 60–2. (52) Parikh and Shelat (1991), pp. 102–7. (53) H.D. Sankalia, ‘Mahmūd Begaḍā ke samay kā Dohād kā śilālekh (Vi. Sam. 1545; Śāke 1410)’, in Rājavinodamahākāvyam, G.N. Bahura (ed.), Jaipur: Rajasthan Oriental Research Institute, 1956, p. 26. (54) Sankalia (1956), p. 28 fn. (55) See Françoise ‘Nalini’ Delvoye, ‘Indo-Persian Accounts of Music Patronage in the Sultanate of Gujarat’, in Confluence of Cultures: French Contributions to Indo-Persian Studies, M. Alam, Françoise ‘Nalini’ Delvoye, and M. Gaborieau (eds), New Delhi: Manohar, 2000, pp. 253–80. (56) Udayaraja, Mahākavi Udayarāja viracita Rājavinodamahākāvyam or Mahmuda-suratrāṇa-carita, G.N. Bahura (ed.), Jaipur: Rajasthan Oriental Research Institute, 1956. (57) B.J. Sandesara, ‘A Sanskrit Inscription of the Sultān Mahmūd III (1536–1554) of Gujarāt’, Journal of the Oriental Institute, 1(2), 1950: 172–3. Page 20 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Languages of Public Piety (58) Sandesara (1950), p. 173. (59) Sandesara says the word is ‘a literal translation of the ‘invisible God’, namely Allah, and reminds one of the Sanskrit legend on the dirham of Mahmud of Ghazna mentioning alakṣa ekam (‘invisible is one’). B.D. Chattopadhyaya discusses several such Sanskrit ‘equivalents’ for Allah such as those in the 1264 Verawal inscription; Chattopadhyaya (1998), pp. 75–6n32. (60) Sonavane (1972), pp. 433–5.

Access brought to you by:

Page 21 of 21

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Universal Poet, Local Kings Sanskrit, the Rhetoric of Kingship, and Local Kingdoms in Gujarat Aparna Kapadia

DOI:10.1093/acprof:oso/9780199450664.003.0008

Abstract and Keywords A poet from Vijayanagara travelled to Gujarat in the mid-fifteenth century where he produced at least two Sanskrit compositions for local kings as they carved out their political and cultural space in a land dominated by the powerful Gujarat sultans. Although vernacular text production was increasingly the norm in the non-Muslim courts of the fifteenth century, Sanskrit still held considerable cachet for rulers seeking a transregional vocabulary of power, and Gangadhara’s skills as a poet were valued accordingly. In a play written for Gangadasa, the Chauhan ruler of Champaner, and in a praise poem for Mandalika, the ruler of Junagadh, Gangadhara embedded the region’s specificity and the rulers’ aspirations in a universalizing Sanskrit idiom, showing off his skill while fulfilling the needs of his patrons. Keywords:   Sanskrit, Gujarat, Sultanate, kingship, kavya, epic, Rajput

SOMETIME IN THE MID-FIFTEENTH CENTURY, Gangadhara, a poet skilled in Sanskrit and patronized by king Pratapadevaraya of Vijayanagara,1 went on a pilgrimage to Dwarka. From there he proceeded to Ahmadabad to serve the Sultan of Gujarat, Muhammad (r. 1442–51).2 After defeating his rivals at the Sultan’s court, Gangadhara travelled southeast to the small kingdom of Champaner, which was then ruled by the Chauhan chieftain Gangadas. Gangadhara writes of his movements within what we know today as the region of Gujarat in a Sanskrit play entitled Gaṅgadāsapratāpavilāsanāṭakam or ‘the play about the glory of Gangadasa’.3 In this work that follows the conventions of Page 1 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings classical (p.214) drama, the poet narrates the Chauhan king’s campaign against and the subsequent victory over the Sultan of Gujarat. Another work, the Māṇḍalīkanṛpacarita or ‘biography of king Mandalik’, a mahakavya or epic poem eulogizing the Chudasama king Mandalik of Junagadh in the Kathiawar or Saurashtra peninsula, is also attributed to Gangadhara.4 Gangadhara thus appears to have travelled across the entire region, offering his literary skills from court to court, where they certainly seem to have been in demand. The fifteenth century was a period of considerable political dynamism and flux in Gujarat. While the Gujarati sultans fought to establish their superiority over the rival sultans of Malwa and of the Deccan, they also struggled to consolidate their position within the region of Gujarat. In this regard their primary competitors for control over territory and resources were a number of local chieftains and warrior kings such as Gangadas and Mandalik. These chieftains belonged to lineages and groups of diverse origin who, as they gained political ascendancy, sought a legitimate space within the wider brahmanical varna hierarchy by making claims to exalted Kshatriya status. Simultaneously, they negotiated their political positions vis-à-vis the powerful sultans and fought back a host of smaller, local rivals. Several historical accounts in Persian from the Sultanate and the Mughal period bear witness to these rivalries and transformations.5 Besides, a number of courtly narratives, epic poems, and biographies in Sanskrit and Old Gujarati were composed at the small chieftaincies and kingdoms of the region, reflecting their version of these changes.6 By focusing on two such narratives in Sanskrit, the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita, I will explore how the cosmopolitan language of Sanskrit and the correspondingly cosmopolitan idiom of kavya were put at the service of local (p.215) kings in this particular period, when vernacularization was concurrently under way. Over the past few decades, Sheldon Pollock has demonstrated the close relationship between language and political power, specifically focusing on the role of Sanskrit in the royal courts of South and Southeast Asia prior to the first millennium of the Common Era.7 In the post-1000 period, however, Pollock suggests that the prestigious position of Sanskrit was gradually taken over by the different vernaculars that developed from this period onwards. While these vernacular languages showed a keen awareness of regional specificities, they nevertheless drew from the literary tropes offered by Sanskrit, particularly the genre of aestheticized poetry and prose known as kavya. While Pollock’s formulations are extremely significant in the study of literary cultures of the subcontinent, more recent scholarship has shown that Sanskrit was not relegated to the background during the second millennium, but, in fact, came to serve a variety of different functions and was to become one among a number of prestigious language choices in this period.8 In the case of fifteenth-century Gujarat, as this essay will demonstrate, Sanskrit, available at a premium, continued to remain a language of prestige that could be harnessed by local Page 2 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings chieftains seeking to secure their positions within their patrimonies. It was also used extensively in conjunction with the other languages that were in use in the region such as Arabic and Gujarati, particularly in the writing of inscriptions. Further, it is noteworthy that during the fifteenth century and in the centuries that followed, the rajas or chieftains of Gujarat continued to draw prestige (p. 216) for their Kshatriya warrior aspirations through the patronage of Bhats and Charans, the traditional genealogist historians of the region who sang in their own specialized dialects. Several written works such as the Raṇmallachanda, composed in the early fifteenth century by a Brahmin poet, drew from these oral traditions, which were in fact divergent in their approach to their patrons from their Sanskrit kavya-inspired counterparts. The oral narratives of Gujarat are beyond the scope of the present essay. Yet it is important to keep in mind that the Sanskrit works produced in the region were located in a courtly milieu where patronage was available for a variety of languages and genres. Thus, the itinerant poet, Gangadhara, used his literary skills and imagination to fashion and articulate for his local patrons a rhetoric and ideology of kingship that combined idealization with a keen sense of the local political reality. While the older Puranic models of kingship are evoked, particularly through the use of the cosmopolitan Sanskrit kavya tradition by a poet who claimed that he hailed from outside the region, they are, in fact, reconfigured by the social and political realities of the local kingdoms of Gujarat. The same interplay of cosmopolitan and local can be seen in the geographical sensibility at play in the two works, where the key site of the hill fort also received extensive literary treatment in Sanskrit for the first time.

Hill Forts and Local Kingdoms: Champaner and Junagadh Commanding a height of 2,500 ft above the surrounding plains, some 40 km northeast of Baroda, Champaner separates present-day Gujarat from Madhya Pradesh. In the medieval period, it was an important location that gave the rulers of Patan and Ahmadabad access to, as well as protection from, the region of Malwa. The Pavagadh hill at Champaner was also the site of a complex religious landscape, as oral traditions and excavations show. It was a site for the worship of goddess Kali but also appears to have had a number of Shaiva as well as Jain shrines.9 As an active pilgrimage site, it would have been a (p.217) valuable source of revenue for rulers, and excavations have revealed the remains of a large city built in the late fifteenth century by Sultan Mahmud Begada (r. 1458–1511), the most influential of the Gujarati Sultans.10 The fort of Junagadh was also a significant economic and strategic location for exerting control over Gujarat, with its access to other important pilgrimage sites like the Girnar hill, Dwarka, and Somanath. Long before the rule of the sultans, the revenues of these pilgrimage sites, particularly of Somanath, had been a bone of contention between the peninsular chieftains and those ruling in the east from Patan.11

Page 3 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings Hill fortresses like Idar, Champaner, and Junagadh were strategically important to be able to rule over the entire region. But even after ‘Ala’ al-Din Khalji’s conquest in 1297, control over these territories remained difficult as long as Gujarat remained part of the larger empire of the Delhi Sultanate. However, with the gradual decline of the authority of Delhi in the province, powerful local nobles and regional sultans attempted to bring the chieftains (along with their lands and other resources) more effectively under their sway. Even so, the chieftains’ familiarity with the terrain and the protection afforded by the forts they built around their territories made the task of controlling them far from simple. Until Mahmud Begada managed to capture the two forts of Champaner and Junagadh in the late fifteenth century and build new towns there, almost all his predecessors had made unsuccessful attempts to gain (p.218) control over them.12 Sultan Mahmud’s reign marked an important shift in the nature of the polity in Gujarat, integrating older models of alliance politics into the larger Sultanate polity.13 But prior to his takeover, chieftains like the Rathods of Idar, the Chauhans of Champaner, and the Chudasamas of Junagadh remained extremely powerful in their local domains and posed a threat to any form of regional or imperial rule. The produce from these areas, whether from agriculture or from the forest, was also a vital element in the consolidation of a regional-level polity. By controlling the principal hill forts, the chieftains had access to the numerous smaller branches of land-owning lineages, with whom they often shared kinship ties. Control over these men and their forts as a source of military manpower as well as an economic resource was understandably a major concern for the Sultans. Who were these chieftains? While the Sultans of Gujarat (and later also the Mughals) often categorized these chieftains as zamindars, they preferred to style themselves as rajas or kings. In addition, many of them gradually sought to obliterate their obscure origins by setting up courts based on classical Indic models and patronizing poets and Brahmins in their well-defended forts. It is, therefore, reasonable to view these chieftains as continuing the process of local state formation that B.D. Chattopadhyaya described for an earlier period, and which he saw as the major integrating factor in the development of regional political, economic, and socio-cultural trends.14 The vast substratum of chieftains (p.219) that continued to exist in Gujarat during the rule of the regional sultans and the governments that followed are an example of this process, as are the large number of local and sub-regional states that continued to exist and to emerge beside the other larger regional states in Malwa and the Deccan.15 A similar continuity can be detected in the ideological realm, as many of these local chieftains claimed Kshatriya status and commissioned narratives and genealogies that connected them with the ancient royal families of the surya vamsha (solar lineage) and chandra vamsha (lunar lineage). These claims were accompanied by the image of the king as an idealized warrior, performing royal Page 4 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings rituals, and protecting Brahmins, cows, and vassals. The chieftains took on universally recognized titles such as maharaja (great king) and maharajadhiraja (great king among kings), while chakravartin (ruler of the world) signalled a claim to imperial authority. The Chudasamas of Junagadh, the most powerful among the lineages active in Saurashtra, are one important example of this process. Like other lineages, including the Gohils and Jhalas, the Chudasamas had migrated into the region in the early medieval period. They had long been associated with the abhiras or pastoralists sharing close links with the Sammas of Sindh, who were Muslims, as well as with the Jadeja chieftains of Kachchh, until they acquired the principality of Vanthali from the local ruler and subsequently occupied the already fortified city (p.220) of Junagadh.16 From this fort city they were able to control a considerable portion of Saurashtra until the Sultans from the east defeated them in the late fifteenth century. Further, as Samira Sheikh has shown, prior to this defeat the Chudasamas had come to acquire an elaborate court and aspired to a prestigious Sanskritic identity, giving up the more heterodox traditions that the Jain and other chronicles attribute to them.17 While a number of hero-stones or paliyas in the region simply mention the Chudasamas as rulers, inscriptions on temples and stepwells that were patronized by merchants or courtiers and their wives link them to the Puranic moon dynasty; later inscriptions and the Māṇḍalīkanṛpacarita also link them to Krishna’s Yadu lineage, thus legitimately incorporating them into the Vaishnava fold.18 Less is known about the ancestry of the Chauhans of Champaner. A late nineteenth-century tradition claims that they belonged to the Khichi branch of the Chauhan lineage at Ranthambhor and had migrated to Gujarat soon after the conquest of Khichiwada by the armies of ‘Ala’ al-Din Khalji.19 While it is not entirely clear how they acquired the territory of Champaner, a Sanskrit inscription from c. 1469 (VS 1525) that gives a genealogy of nine predecessors of Gangadas indicates their long-standing presence in the region and their claim to Kshatriya status. The inscription is composed in a mix of Sanskrit and Old Gujarati and notes that it has been issued during the victorious reign of the great king (maharaja) Jayasimhadeva, for the benefit of his mother: In the lineage of Prithviraja, the chief of the Chauhans, many kings have ruled (Old Gujarati: ghaṇā rājā hoā). In the family of Hammiradeva, the ornament of his kula, was Raja Shri Ramadeva, Shri Changadeva, Shri Chachimgadeva … Shri Palhanasimha, Shri Jitakarna, Shri Kumpuraula, Shri Virdhavala, Shri Savaraja, Shri Raghavadeva, Shri Trimbakabhupa, Shri Gangarajeshwara. His son is renowned for increasing the religious merit of (p.221) his ancestors, worshipper of Shri Shakti and for being a perpetual bestower of cows and gold (Sanskrit: nitya suvarṇadhenudānakartā), as well as the giver of grants (śāsana) to Brahmins, a donor of elephants, the illustrious king over kings (rājādhīraja) Page 5 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings Shri Jayasimhadeva, in the village of Ayasiamanu he built (this) well for the spiritual benefit of his own mother (nija janani), Shri Phamadevi …20 While Jayasimhadeva, the son of Gangarajeshwara (Gangadas), thus embodies all the qualities of an ideal ruler, it is noteworthy that the inscription links the Chauhans of Champaner to historical kings like Prithviraja and Hammira, who were popular in the wider region of western India, rather than to a divine lineage. The Mir’āt-i Sikandarī (1611) of Sikandar bin Muhammad alias Manjhu depicts the Chauhan rulers of Champaner as actively involved in the politics of the region and as an obstruction to the Sultans’ efforts to consolidate their rule. Around 1416 Trimbakdas, the Raja of Champaner, seems to have formed a confederation with other chieftains of the region such as Raja Punja of Idar and Satarsal of Jhalawar and invited Sultan Hushang Shah of Malwa to invade Ahmadabad while Sultan Ahmad was away from the capital dealing with other rebels.21 Sikandar also writes that when Sultan Muhammad II marched against Raja Gangadas of Champaner (the son of Trimbakdas), the latter was defeated and, after putting up a fight, fled to the upper part of the fort: ‘When the garrison of the castle became strained, the Raja sent ambassadors to Sultān Mehmūd of Māndu offering to pay him a lakh of tānkās for every march he should make to his assistance’,22 an episode alluded to in a somewhat different manner in the Gaṅgadāsapratāpavilāsanāṭakam, as we shall see. (p.222) Drawing on Chattopadhyaya’s analysis of the social processes of local state formation and on ‘Kshatriyaization’, I thus suggest that these processes continued in western India even after 1200 and contributed to defining the manner in which the chieftains chose to portray themselves in the fluctuating politics of fifteenth-century Gujarat. By then the rulers of Junagadh and Champaner had held their patrimonies in the region for several generations and, as inscriptions and textual production show, they could patronize scribes and men of literature to write about their lineages and their rule. These chieftains gradually developed large fortifications that included a wide-ranging population including priests, warriors, traders, and craftsmen,23 and were deeply involved in the politics of the region, since the resources they controlled were valuable to the Sultans, who were trying to consolidate their rule as well as to fight their extra-regional rivals. It is in this context that the choice of Sanskrit and the articulation of kingship within the cosmopolitan idiom of Sanskrit kavya acquire particular interest.

The Narratives The Gaṅgadāsapratāpavilāsanāṭakam exists in a single manuscript.24 On the basis of the script used, its editor has suggested that it may have been copied sometime in the sixteenth century from an earlier (p.223) manuscript.25 The original play, on the other hand, is believed to have been composed much closer to the actual historical event in 1449, possibly between 1450 and 1460. The accuracy of this estimate is supported by the fact that after the reign of Page 6 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings Gangadhara’s Vijayanagara patron, Pratapadevaraya (r. 1426–47), the fortunes of Vijayanagara seem to have declined for almost half a century: it is therefore possible that the poet left this court in search of better prospects in other parts of the subcontinent. The Gaṅgadāsapratāpavilāsanāṭakam is a play in nine acts that makes use of both prose and poetry, and is composed primarily in Sanskrit. The Sanskrit is interspersed with a form of Prakrit, used traditionally in Sanskrit classical drama by the vidushaka or court jester and the female characters. In addition, the soldiers of the Sultan’s army use a language that appears to be some form of Hindavi associated by the poet with Muslim soldiers.26 However, in keeping with conventions, the Sultan speaks Sanskrit like the other prominent male characters in the play. The setting for the performance itself is the festival dedicated to the goddess Mahakali. While the political conflict between the Chauhan king and his rival, the Sultan of Ahmedabad, form the basic core of the narrative, it further follows the conventions of courtly Sanskrit drama with elaborate praises of the palaces and the ruler in addition to containing a variety of emotions or rasas. It also includes another important element of the genre, that of the play within the play. Briefly, the plot of the play is as follows. The Sultan has demanded the Chauhan chieftain’s daughter in marriage, but the latter is unwilling to stake the honour of his lineage by acceding to the request. Meanwhile, news arrives that the Sultan of Mandapa (Mandu) has agreed to assist Champaner against his longstanding rival, the Gujarat Sultan. This brings great joy to all and sundry: elaborate prayers are offered to the (p.224) family goddess and several Vedic rituals are performed. While the festivities continue, a set of actors, who like the poet also hail from the south, perform a play in honour of the chieftain, depicting an affectionate exchange between him and the queen, Pratapadevi, in their youth. While the king and the queen are thus engaged in various festivities, a chamberlain brings the news that one of Gangadas’s generals has arrived at court with the slain heads of some men from the Sultan’s army, the yavanas, as Muslims are often referred to in this period. The battle has begun and a victory has evidently been won, as the Sultan’s attempt at reconciliation in the fourth act also suggests. A message is sent to Gangadas stating that the reason behind the Sultan’s attack on Champaner is that Gangadas has been sheltering in his court certain recalcitrant, trouble-making garasiyas or landholders.27 It would be wise, the message suggests, for Gangadas to accept the Sultan’s suzerainty instead of acting in favour of his enemies. This message is sent to the Champaner court by two Rajput allies of Sultan Muhammad Shah. However, for Gangadas his independence is so precious that he insults his fellow Kshatriyas who have accepted the Sultan’s authority. A confrontation is inevitable. Gangadas now takes up arms himself, and in the ensuing battle the Sultan’s

Page 7 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings forces suffer several reverses that force them to flee. Gangadas decides not to pursue the fleeing army because it would be a dishonourable act. The disheartened Sultan, however, is cheered when one of his Kshatriya allies, Virama,28 presents him with an elaborate map of the Pavachala fort in Champaner. The Sultan decides once again to besiege the fort but again meets with reverses. We are also made privy to the fact that he is worried about a simultaneous attack on his armies by his rival, the Sultan of Mandapa, who we already know is allied with (p.225) Gangadas. The Sultan launches a new strategy of attacking the tribal areas around the Champaner fort which, once sacked, will lead to the falling of the fort ‘like a ripe fruit’.29 Both parties seem to suffer equal reverses and in fact the trouble-making garasiyas sheltered by the chieftain are also killed. However, when the announcement that the Sultan of Mandapa is indeed on the outskirts of Muhammad Shah’s territories and is moving forward with a large army is received, Muhammad Shah deems it wise to give up the siege of Champaner. Gangadas once again does not pursue the retreating army, since his code of honour does not permit him to attack an army that has turned back. In the final act we return to the Champaner court. While the last few folios of the manuscript are missing, the play seems to end with the chieftain and his queen offering prayers to the goddess Mahakali, who in turn grants them her blessings. The Māṇḍalīkanṛpacarita is a Sanskrit epic poem in ten sargas or cantos and 638 verses, composed as a traditional carita or biographical eulogy.30 It narrates the life and exploits of Mandalik, the Chudasama chieftain of Junagadh. The narrative begins with a description of the city of Jirnadurga or Junagadh and its formidable fort, ruled by the Chudasamas. This is followed by a detailed genealogy of the lineage, spanning five generations prior to our hero, Mandalik. Mandalik’s father Mahipala obtained a son after seeking the favours of the deity Radha-Damodara (Krishna). The child, therefore, is associated with Vishnu and is projected as his incarnation as Krishna throughout the narrative. After this genealogy the carita takes us through the childhood of the prince, who grows up to be an extremely religious, intelligent, handsome, and brave youth, who surpasses his teachers in everything. When Mandalik becomes eligible for marriage, a suitable bride is found for him, the daughter of the Gohil chief Arjuna, who has been brought up by her paternal uncle Duda. The Krishna-like Mandalik is then installed as the yuvaraja or crown prince, and under his administration, the city of (p.226) Jirnadurga turns into a utopian place of virtue, prosperity, and happiness. Chiefs from the neighbouring provinces also offer their allegiance; only Sangan, ‘king of the Western Ocean’, remains defiant. This chieftain can be identified as a pirate of the Vadhel clan who took control over parts of the coast of southern Saurashtra.31 Mandalik is able to quickly set Sangan straight, acquiring a rich tribute of gems and horses in the bargain. He

Page 8 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings also defeats and kills Duda, his Gohil father-in-law, as a favour to the yavana king, the Sultan of Ahmedabad. After this victory, Mahipala hands over the kingdom to his son and retires from active political life. Once on the throne, Mandalik requests his minister to find him another suitable wife. The minister presents a list of about fifteen princesses from all over the subcontinent. But the minister feels that the princess most suitable for the young king is Uma, the daughter of the Jhala chieftain Bhima. She is suitable both in terms of her own virtues as well as her lineage. The poet describes the marriage procession and ceremony in great detail. This is followed by a description of Mandalik’s benevolent rule, where it is reiterated that the Jhala and Gohil chieftains serve the king in a subordinate position.32 A considerable portion of the narrative now describes the approach of spring and the king’s romantic dalliances with the queens in the pleasure gardens, in imitation of classical kavya style and Sanskrit court drama. After this interlude, the scene shifts back to the world of military and political activity. Mandalik’s minister informs him that all his neighbouring chieftains have accepted shelter at his feet except Sangan, who is once again challenging his authority. The protracted battle between the two is described in detail and involves Sangan seeking aid from a parasika or Persian chief. Mandalik eventually defeats his enemy and acquires a large booty. The Māṇḍalīkanṛpacarita ends with a further eulogy of Mandalik which compares him with Vishnu’s last incarnation, that is, Kalki, the saviour of the Kali age.

The Rhetoric of Kingship Both the Chudasamas and the Chauhans aspired for higher social and political status in the region, and patronizing Sanskrit works of kavya that (p.227) eulogized them may have been one among many attempts at fulfilling this aspiration. Kavya, Daud Ali has convincingly argued, was crucial to the production and reproduction of courtly culture.33 The transmission of the kavya tradition was accompanied by elaborate gestures and had a ‘performative’ or ‘spectacular’ character to it. Courtly gestures and signs were interpretable by all those, men and women, who were a part of courtly society. This meant that even if many did not understand the language itself, a set of formulaic ideals emerged that unified the audiences within courtly life and also gave them a shared ability to interpret the indicators of this life. Thus, Ali argues, drama and poetry produced in this tradition played an important role in shaping the ideologies and values of the people who were part of the courtly world.34 Moreover, this tradition, associated as it was with elite groups, was aspirational, and in the post-Gupta period, it was emulated and adapted by the numerous small and large courts of India to suit their own particular needs. Even the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita are no static reiterations of formulaic values of kingship, nor are the regional chieftains merely fitted into a pre-existing framework. Instead, both these works actively

Page 9 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings negotiate cosmopolitan ideals of kingship. These, in turn, are reconfigured by the needs of the local polity. The idiomatic language used in the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita is highly stylized and the narratives are located in what appears to be an elaborate courtly setting. Gangadhara endows Gangadas and Mandalik with extravagant palaces and courts within the precincts of their forts. These forts, as we shall see, are embellished with marks of prosperity, including numerous temples dedicated to Puranic deities, lakes and wells overflowing with water, ample food (p.228) provisions, and elaborate weaponry. The protagonists’ kingly duties are also more varied than those of simple warriors: they include maintaining the moral, political, and social order in their kingdoms. It is, in fact, the rule of these virtuous kings that makes these places utopias of prosperity and morality. Already with Mandalik’s consecration as crown prince (abhiṣekamasya yuvarājapade) all the people in his father Mahipala’s kingdom are happy and conduct their duties with utmost honesty. There is no thief in the kingdom, except the great sun who ‘robs the darkness of its treasures’.35 Nobody recites harsh words except the students of the tarkashastra (a branch of the Nyaya school of philosophy), and prince Mandalik himself only speaks sweet words.36 The merchants of the kingdom are skilled and powerful, while the best of the Brahmins are happy and satisfied (3.7); Mandalik’s good qualities prevent the populace from deviating from the path of virtue. These images of the king’s qualities belong to no particular instance in time, but rather draw from the long tradition of aestheticized kavya literature. Using the ornate style of prose and poetry kavya, then, the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita project their patrons as idealized Kshatriya kings. Both Gangadas and Mandalik are aware of this role and constantly reiterate its constituent values. When Duda, the Gohil chief, asks Mandalik to withdraw from the battle and ‘live long to enjoy the pleasures of having a son …’,37 he replies: It is a merchant’s ambition to enjoy the pleasures of a home in the company of a wife and relatives; a king aspires for the higher joys of heaven obtainable by those who die on the battle-field.38 For the same reason, when the Sultan’s army flees the battlefield, Gangadas chooses not to follow them. By allowing his militarily superior (p.229) rivals to run away without an actual fight, he establishes the superiority of Kshatriya values.39 Ideal kings need excellent lineages. As we have seen, the Māṇḍalīkanṛpacarita provides Mandalik with a genealogy that covers five generations before him. His ancestors belong to the lunar lineage or chandra vamsha and to the Yadava kula. All of them were great warriors, subdued neighbouring chieftains, destroyed the yavanas, were extremely virtuous and religious, and were generous to the Brahmins. Mandalik, who is himself a part incarnation of Damodara or Krishna, Page 10 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings shares all these qualities with his ancestors. While the play does not provide a conventional genealogy for Gangadas, he is referred to as the descendent of the great Chauhan Hammira of Ranthambhor more than once and is represented as a virtuous and religious benefactor of the Brahmins. His virtue is even attested by a disembodied voice from the sky: when the Sultan receives the news that the women of his harem have been captured by the Chauhan chieftain, he is angry and alarmed. But a voice (perhaps meant to be divine) reassures him that: These dancing girls were brought before Gangadas while he was sitting with Pratapadevi, Namalladevi and his other queens. The king, who did never cast a glance at other women, was displeased. He gave jewels and ornaments to the girls and returned them safely in palanquins to the Sultan’s camp.40 Both narratives, then, portray the personalities of their protagonists within a formulaic cosmopolitan idiom that later came to be associated with a legitimate Rajput high culture, entirely obscuring the more ambiguous origins of their protagonists. At the same time, these timeless cosmopolitan portrayals of Gangadas and Mandalik are combined with, and located within, the political contexts to which they belong. While at first sight the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita seem to contain most of the essential elements of the courtly kavya tradition, both narratives present an unusually detailed depiction of the political activity surrounding the Champaner (campakapuri) and Junagadh (jirnadurga) kingdoms. Gangadhara mentions specific personal names of military commanders, courtiers, subordinates, and so forth, together with the details of the events that he chooses to portray. (p.230) Thus, while the initial motive for Muhammad Shah’s attack on Champaner appears to be Gangadas’s refusal to give him his daughter in marriage, the real reason is revealed to the audience in a later act of the play. The two ‘Kshatriya’ allies of the Sultan, have written a letter to the Chauhan chieftain, stating: Do not shelter the garāsiyas, who are the enemies of the Sultan and are making trouble in his territories … do not initiate enmity … a clever man knows these times well, this is not the time of the Kshatriyas, it is the Kali age of the yavanas.41 Advising Gangadas to swap the older model of alliance politics for the new Sultanate polity, they counsel him to marry his daughter to the Sultan, wash the Sultan’s feet in submission (thus giving up his own honour and autonomy), and accept his suzerainty instead of challenging him by giving refuge to troublemakers.42

Page 11 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings The tension between the old and new models, articulated in terms of a crisis of the Kshatriya ideal of honour, emerges again at the end of the play. When the battle between the Sultan and Gangadas is at its height, some of the garasiyas are killed by the soldiers of Muhammad Shah.43 The families of the deceased men are desperate with rage, the sons of those brave warriors have left the fort to fight the Sultan while their devoted wives have walked into their funeral pyres as the war trumpets sound in the background.44 Gangadas is disappointed on hearing the news. He articulates in no uncertain terms that the garasiyas were the cause of his rivalry with the Sultan. He is upset that he has not been able to save the lives of those who sought his protection and chides himself for not living up to his Chauhan lineage, which is well known for the granting refuge to those who need it.45 He now forbids his officers from using the services of the remaining garasiyas, for it is his duty as their protector to keep them from danger. In an emotional exchange, the surviving garasiyas declare that they are eager to fight since they have pledged their lives in gratitude to Gangadas. Despite this emotional climax, the death (p.231) of the garasiyas has created a sense of futility around the enmity between the Sultan and Gangadas and reestablished the partial superiority of the Sultan. In the end the poet resolves the matter by shifting the focus on to another field of competition: Muhammad Shah must leave the battlefield because his other major rival, the Sultan of Mandapa, is about to seize Ahmedabad with a large army of 100,000 cavalry, 200,000 foot soldiers, and 1,000 elephants. Sultan Muhammad’s ally Virama further provides a justification for this retreat by pointing out that the protection of one’s own territories is a king’s foremost task.46 A similar tension between alliance politics/tribal alliances and the integration into the Sultanate is expressed in the Māṇḍalīkanṛpacarita as well. The Sultan of Ahmedabad sends an envoy to Mandalik’s father Mahipala, to complain that Duda, the Gohil chieftain who is his son’s father-in-law, is wreaking havoc in the Sultan’s territories.47 The envoy warns the king about the Gohil and his associates, stating that they will disregard their matrimonial ties with him in due course.48 Mahipala reassures the envoy that he considers the Sultan’s enemy as his enemy too. Yet he is also troubled by the thought of fighting his relative in support of the yavana. He notes that, on the one hand, it is not a happy thing to fight the yavanas, who have increased their strength owing to this Kali age. Already the king of the yavanas has deprived several kings of their kingdoms. However, the yavana king has shown no open enmity towards the royal family of the Yadavas (namely, the Chudasamas) and, thus, Mahipala feels it is wise not to voluntarily initiate a hostile situation.49 His minister also counsels him to the same effect: That yavana king, who on the strength of his army of elephants and thousands of horses had conquered the world, has courted your friendship. What greater good and safety do you ask for? It would therefore be best for you to do what is pleasing to him. On the other hand, if I were to Page 12 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings recount the misdeeds of Duda I am afraid that I would incur the prince’s displeasure. These chiefs always seek shelter under you when they are attacked by the (p.232) yavanas and yet claim as their own the lands bordering your kingdom (sīmabhūmimapahṛtya).50 This pragmatic advice, which reminds the king of his unsteady control over his own allies and subordinates, convinces Mandalik. He relinquishes his moral dilemma and himself kills his father-in-law in the interest of Chudasama authority in Saurashtra and of his relationship with the more powerful Sultan. The tension with rival claimants over resources in the region is at play also in the encounters with Sangan, the king of the Western Ocean, who defies the Chudasama claim of complete authority over Saurashtra. At Mandalik’s consecration as the crown prince, the kings of the bordering territories send gifts, accepting his supremacy, and he suitably honours them in turn.51 As we have seen, Sangan disregards the news brought by the Chudasama envoy. Mahipala, though angered, only smiles at this slight, whereas his son Mandalik pledges to fight the insubordinate chief.52 The battle is described in some detail, and at the end the prince manages to break Sangan’s weapon and makes him fall from his horse. Despite the clear advantage he has over his enemy, Mandalik now spares his life (jivanadānadadāmiti, lit. ‘I grant you the boon of life’).53 Instead, he collects a tribute in the form of horses and gems.54 The tension between these two rival claimants to authority in the region does not end here. Sangan reappears later in the narrative, when he demands that the Chudasama chieftain submit to his own authority, blatantly disregarding Mandalik’s previous magnanimous gesture according to the logic of the narrative.55 The subsequent battle is described in even more intricate detail than the earlier one.56 Once again Mandalik wins and is able to acquire large quantities of gold, silver, pearls, and jewels, as well as horses and camels. These are, in turn, distributed among subordinate kings, artisans, and bards.57 (p.233) The narratives thus show an acute awareness of the region’s political processes and realities. In this regard the poet depicts multiple spheres of rivalry and negotiations that the protagonists have to face vis-à-vis the Sultan, with the garasiyas, the Gohil tribal chieftains and the pirate chieftain Sangan as well as the tensions between the Sultans of Gujarat and of Malwa. In these depictions we can read the pressures to forge ties and struggles to establish hierarchy that must have existed between the different players. It is within these multiple spheres of rivalry that the Sanskrit poet constructs idealized Kshatriya personas for his Chauhan and Chudasama patrons. The poet shifts easily back and forth between the universalized, timeless, and idealized realm of kavya and the specificities of the region’s contemporary politics.

Page 13 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings Two more points need to be made. First, the poet provides no background to the various crucial political conflicts in the play and the mahakavya. In the Gaṅgadāsapratāpavilāsanāṭakam, he gives no explanation of the role of the garasiyas or for the long-term rivalry between the Sultans of Gujarat and of Malwa. In a similar way, the intricate details of Mandalik’s relations with the Sultan and with other actors in the region are not supported by any additional information about them. In other words, Gangadhara assumes a familiarity with all these factors on the part of his audience. In order to understand these narratives, one needs prior familiarity with the region’s geography and politics. The events he describes firmly situate Gangadhara’s narratives within their local contexts. Rather than spreading their patron’s glory to other parts of the subcontinent, works of cosmopolitan Sanskrit kavya like the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita would thus have established his fame within his own social and political domains. A second and final point, despite the use of various universalized idioms of kingship, neither king nor Gangadas nor Mandalik aspire to the conquest of all directions or digvijaya, an essential element of a Hindu king’s aspirations to expand his realm. Instead, their aspirations are limited to protecting their sovereignty and status within their own patrimonies and do not even extend to a conquest of the Sultan’s territories. Arguably, this merger of dharmic norms of kingship with their localized manifestations would have made it easier for these chieftains to become accepted as rajas or kings within the areas over which they sought supremacy.

(p.234) Political Geography and the Importance of Place Just as Gangadhara’s Sanskrit narratives strive to portray the intricate details of the region’s political history while projecting his patrons as idealized Kshatriyas, they display an intricate interaction between local and cosmopolitan geographies, between real and Puranic topographies, and between local and transregional geopolitics. Once again, through Sanskrit Gangadhara seems to be evoking both a local and cosmopolitan geography for his local audience. One form of geographical knowledge which Gangadhara depicts in his work is that of the local topography, particularly that of Champaner and its adjoining hill, Pavachala or Pavagadha. The Gaṅgadāsapratāpavilāsanāṭakam begins with the sutradhara extolling the virtues of this hill, which is the abode of Mahakali. It is also the place which Lord Shiva visited after having left his snow-clad mountain in the agony of separation from his beloved. It [the hill] is the support of the weak … it is the place where the residents of all three worlds find friendship … where the earth is pure and radiant, touched by the soft breeze and the skies are bright and clear …58 Later in the play these poetic effusions merge with the strategic requirements of the conflict. When, in the seventh act, the Sultan is disheartened by his losses at Page 14 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings the hands of Gangadas’s army, one of his Rajput allies, Virama, gets him a detailed map of the fort and its surroundings painted on a patta or cloth, in order to facilitate their movements through it.59 The map combines Puranic/sacred geography with exact topography. It shows that on the top of the hill there is Shiva temple made of gold and silver. In the distance, in the north-east, there is a lake named Ramanaganga that was built by Rama (rāmacandreṇa nirmitam).60 The deep lake to the south was created by Sita, while another lake to the (p. 235) west of it is named Bhimagaya and was created by Bhima.61 On its west there is a large lake with white waters (bhūrisudhādhvalavāriṇa) that Gangadas built, which is surrounded by temples of Ganesha, Durga, Dinkara, Kshetrapala, and the Jina.62 The clouds that are ever visible on the top of the hill are the smoke coming from the sacrificial fires. But the fort is also dotted with the dwelling places of the various members of the royal family, and the unfurled victory flags (jayapatākā) are a constant reminder of the ongoing festivities.63 The fort is prosperous beyond belief and replete with food, wealth, and wells. The subordinate kings also live happily within this fort that is forever watched over by the gods. On the south of the king’s own palace are the stables (vājiśālā) and living space for cows and other cattle. The goddess Mahakali is constantly protecting Gangadas from the summit, where she sports with the gods.64 On the left of her temple is the temple of desire-fulfilling Jareshwara. Virama’s elaborate description thus takes the audience (and the Sultan, within the play) through the intricate details of the hill fort’s geography, mixed with references to Puranic mythology. With every particular palace, temple, or water body, the poet is careful not only to describe its location but also to emphasize the prosperity that surrounds it; an important consideration when reciting or performing the play before the audience of royals that may have been attending the celebrations related to the goddess festival. Thus, the Vijayanagara poet’s displayed knowledge of local politics is matched by his displayed knowledge of local geography. The idealization of his protagonists is similarly matched by the idealization of the local landscape. Another form of geography enters the narratives through the poet’s imagination of various links that his patrons in the little kingdoms of Gujarat have with other subcontinental rulers/polities. While Champaner was an important actor within the politics of the emerging Gujarat region, it was much smaller and less powerful than the other regional polities such as Malwa, Vijayanagara, or the kingdoms of the Deccan. Yet, in a particularly interesting act of the Gaṅgadāsapratāpavilāsanāṭakam, the poet appears to equate the little kingdom of his Chauhan patrons with (p.236) the other regional polities of the time. Gangadhara dedicates almost an entire act to the Gangadas’s interaction with his courtiers inside his palace or antahpura. As the king sits on his throne in full courtly regalia, surrounded by musicians, female attendants, and ministers, he receives his envoys or dutas who bring news from every direction. Each envoy has returned with detailed reports after visiting the courts of the king of Page 15 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings Simhaladesha, the Gajapati king of the east, the king of Champaranya in the north, and the Sultan of Delhi. They bring news of utter political mayhem in these places where treason and war have undermined the rulers’ positions. Compared to the chaotic state of affairs everywhere else, Gangadas’s kingdom in the west stands out as a haven of peace and prosperity. Once again, the cosmopolitan political geography that is evoked is a mixture of poetic conventions and political reality, possibly with a good dose of the itinerant poet’s wondrous experiences thrown in, and it is worth describing it in some detail. The act opens with Ranachanga, one of Gangadas’s military commanders, entering the court after having slain the Sultan’s commander Naroj and five thousand of his soldiers. His triumphant entry is followed by the arrival of the envoys, as if to emphasize the vast reach and power of Gangadas’s rule. The envoys narrate their observations in great detail, giving a distinct sense of the nature of each of the places described. Thus, the messenger reporting from the city of the Simhala lord (siṃhaleśa) describes the rather complex process by which precious gems are produced there: there is a lake in that city called mantharavāhinī, in which there are lotuses of a number of hues. When the king bathes in the lake along with his wife, drops of water splash into the flowers.65 These drops of water, along with a number of other factors including the impact of the gem goddess (ratnadevyāha), turn into solid gems of the colour of the lotus in which they were born: red lotuses produces rubies, blue lotuses produce sapphires, and yellow lotuses produce gold-coloured gems; drops which have solidified within lotuses with two or three filaments bring forth cat-eye gems. The courtiers are struck by this unusual phenomenon, but the messenger reassures them that he is merely reporting what he has experienced himself (pratyakṣamanubhūta) rather than (p.237) from interference (anumaneṇa) or through the words of a loved one (āptavacanena).66 Next, king Gangadas enquires about events in the east. Here, the messenger reports, the Lord of the Elephants, Gajapati, has been poisoned by his minister with the intention of usurping the throne. The envoy also describes in some detail the preparatory rituals associated with the pleasures of Lord Jagannatha (though these are, according to him, indeed beyond this world or lokottarameva). The next messenger brings news from the north, from Champaranya, where the wise king has gradually managed to increase the size of his already large army.67 In this region, another kind of precious stone is held significant. This is the shaligrama stone, the black and usually spherical stone considered to be the aniconic form of Vishnu. The next messenger reports about Ḍhillīpuram, where there sultans rule. The line of the sultans is coming to an end, he informs the king. The continued hand-to-hand fighting is causing their end and the goddesses of death (yoginis, a reference to Delhi’s other name as Yoginipura, see Chapter 12 in this volume) are hovering there in groups, eager to drink the enemy’s blood.68 Page 16 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings After these animated reports from the three directions and from Delhi, the king wants to know about the happenings in the west. The messenger reports: … hundreds of kings who have taken shelter under him [Gangadas], along with their sons and grandsons, remain extremely satisfied and happy. Burning with the desire of swallowing the gurjara-maṇḍala at every occasion, mahammad suratrāṇa (Muhammad Sultan) bites his lip [in defeat].69 The Chauhan kingdom is thus seen as equal, if not superior, to a number of other important kings in the subcontinent, including the Sultans of Ahmedabad and Delhi. (p.238) In the Māṇḍalīkanṛpacarita, the interplay between cosmopolitan and local geography is played out distinctly in favour of the local. In two ways: first, although Mandalik is eulogized in the universalizing terms of the ideal, almost god-like kavya king, yet his actual military and political achievements are restricted to the realm of the domestic or local. Despite repeated mention of him and his ancestors as ‘destroyers of the yavanas’, Mandalik only defeats chieftains of local lineages like the Gohils or the Jhalas, or like Sangan, the chieftain from coastal Saurashtra. It is against these chieftains, who share with the Chudasamas the political space of Saurashtra, that Mandalik conducts his military expeditions and gains ritual submission. Second, in the realm of matrimonial alliances, the prominence of the local within the universal becomes even more obvious. Here Mandalik’s minister describes to him all the princesses that Mandalik could potentially choose from once he has been crowned king: [T]he daughter of the king of Simhaladvipa is a padmini, she has lotus-like eyes … [but] she is of low birth70… the daughter of the Karnata king is proficient in playing the vina and in other musical arts, [she] is endowed with all the auspicious bodily marks and has beautiful eyes [but] she is not appropriate because of her dark complexion71 … the daughter of the king of Madhyadesha is proficient in painting [but] her thighs are thickly covered with hair72 … the daughter of the king of Maharashtra is well dressed and has a cuckoo-like voice (but) is much too clever and witty …73 The minister describes the princesses of Trilinga, Kalinga (Orissa), Kanyakubja (Kannauj), Kamarupa (Assam), Gopachala (Gwalior), Medapada (Mewar), and several other kingdoms all over the subcontinent in a similar vein. The criticisms finally end when he comes to the daughter of the Jhala chieftain, Bhima. She is not only beautiful, virtuous, and skilled in every art but she is also from a noble lineage. This princess and her father are both mentioned by name, and we are given details of Bhima’s capital and his current whereabouts. The cosmopolitan poet thus claims for his ‘regional’-level protagonist a position among (p.239) the wider political networks of the subcontinent. In Gangadhara’s imagination Page 17 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings Mandalik obviously has access to the daughters of all these different kings. In the act of rejecting them, he establishes his superiority over their fathers’ kingdoms, while the choice of marrying the Jhala princess confirms his position within the local political scenario. The interplay within the local and the cosmopolitan is also in evidence in the intended and imagined audiences for these works. On the one hand, as I have argued, the thickness of local details and implications in the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita demand a local audience who will understand them without need for further explanation. But Gangadhara brings to his local patrons and local audiences not only a vision of the wider political world, he also imagines an audience that lies beyond the immediate audience comprising the ‘entire group of kings who has come to the shrine of Mahakali to worship the goddess in the autumn season’.74 The last act of the play begins with another travelling poet or vaitalika, who sings the praises of the victorious Gangadas. He is joined by the glory or kirti of the Chauhan king and the infamy or apakirti of the Ahmadabad Sultan, personified as two women, the first being of noble birth. It is noteworthy that the noblewoman kirti speaks in Sanskrit while the vaitalika and apakirti use Prakrit and are therewith assigned a lower status in the play. But the vaitalika’s ability to comprehend Sanskrit also allows him to transmit Gangadhara’s tale in the language of the people. The vaitalika introduces two women to each other, and they soon discover that they share a birthday, that is, they are both born on the day that Muhammad Shah, with his mighty armies, was defeated by Gangadas. Gangadas’s kirti and the Sultan’s apakirti are now eager to travel the world. The bard, whom they consider their brother, has promised to take them from ‘country to country (deśa deśāntaram), island to island (dvīpa dvīpantaram), pilgrimage to pilgrimage (tīrtha tīrthāntaram), city to city (pura purāntaram), royal court to royal court (rājasabhā rājasabhāntaram), from one gathering of noblemen to another (sajjanasabhā sajjanasabhāntaram), forest to forest (vana vanāntaram) …’75 and to any other place beyond these that they may (p.240) wish to go to (yad yad gantumicasi tat tadeva nayiṣye).76 The poet thus imagines the whole world pervaded by the story of Gangadas’s victory and the Sultan’s loss. In bringing the vaitalika together with the personified forms of glory and infamy, he also brings about the merger of the written with the oral. Gangadhara kavi thus hopes for the spread of his written tale through the oral version that the bard or vaitalika will sing on his travels along with his two sisters. It is through the vaitalika, after all, that the Vijayanagara king, Mallikarjuna, learns about Gangadhara’s travels to Champaner and Gangadas’s wealth and generosity.77 Thus, the Sanskrit poet imagines multiple geographies in the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita. However, whether it is the topography or the wider political networks outside and within Page 18 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings Gujarat, the king and his kingdom are primarily situated within the local context. In other words, while the universal values of kingship are evoked, in the poet’s imagined geography they are woven into and reconfigured by the ‘place’ to which these kings belong.

Conclusion In this chapter I have shown how the poet Gangadhara represents his patrons from Champaner and Junagadh in his Sanskrit compositions. In both the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita, the poet projects his patrons as universalized Kshatriya kings, and yet he situates them within their very localized political and geographical context. From this combination of cosmopolitan idiom and local reality emerges a rhetoric of kingship that appears to have been created for the localized milieu of immediate rivals and the local population. The image of the morally superior Kshatriya king makes for an effective foil against the yavana sultan as well as the rival Kshatriya chieftains who may have chosen to support him. Their success within their local political scenario, in turn, establishes their moral and political superiority. It can also be suggested, therefore, that Gangadhara’s compositions belong to the body of ‘regional’ Sanskrit texts that had immense significance despite the emergence of the regional vernaculars. As Yigal (p.241) Bronner and David Shulman have suggested for another context,78 not only did such texts often use a form of Sanskrit that was modified by the grammar of the regional language, they were also shaped by the region’s geography and historical specificities. As works that were firmly set in the local context, the Gaṅgadāsapratāpavilāsanāṭakam and the Māṇḍalīkanṛpacarita would have established their patron’s glory within his own social and political domain, rather than spreading his fame far and wide. In this regard they can be viewed as carrying the aspiration of regional elites to reinforce their political and moral values within the fluid politics of the region. For their composer, kavi Gangadhara, a poet originally hailing from Vijayanagara, who we are told had been travelling from one court to another in Gujarat, it must also have afforded the possibility of traversing several domains in which he could display the versatility of his poetic skills. Notes:

(1) Pratapadevaraya is also referred to as Devaraya II. (2) Gangadhara, Gaṅgadāsapratāpavilāsanāṭakam, Bhogilal J. Sandesara (ed.), Baroda: Oriental Institute, 1973, p. 18. (3) Gangadhara (1973). The title can also be translated as ‘The pleasures of Gangadasa and Pratapa’ as the wife of the chieftain was called Pratapadevi.

Page 19 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings (4) Gangadhara, ‘Māṇḍalīka Mahākavya of Gangādhara Kavi’, Bharatiya Vidya, 15(1), 1954a: 35–57 and ‘Śrī-Gaṅgādharakavi-kṛt-śrī-Maṛḍalīka-mahākāvyam’, Bharatiya Vidya, 15(2), 1954b: 13–40. (5) See, for instance, Sikandar b. Muhammad, Mir’āt-i Sikandarī, S.C. Misra and M.L. Rahman (eds), Baroda: M.S. University, 1961; ‘Abd al-Husayn Tuni, Ta’rikh-i Maḥmūd Shāhī, S.C. Misra (ed.), Baroda: M.S. University, 1988. (6) See my PhD thesis, ‘Text, Power and Kingship in Medieval Gujarat (c.1398– 1511)’, School of Oriental and African Studies, University of London, 2010, for a wider range of these Sanskrit and vernacular texts. (7) For Pollock’s formulations on these issues, see, for instance, Sheldon Pollock, ‘India in the Vernacular Milennium: Literary Culture and Polity, 1000–1500’, Daedalus, 127(3), 1998: 41–74; ‘The Cosmopolitan Vernacular’, The Journal of Asian Studies, 57(1), 1998: 6–37; and The Language of the Gods in the World of Men: Sanskrit, Literature, and Power in Premodern India, Berkeley, Los Angeles, London: University of California Press, 2006. (8) One example of a detailed study on the changing role of languages, Sanskrit, Persian as well as the vernaculars, comes from Sumit Guha’s work on the Deccan. See Sumit Guha, ‘Speaking Historically: The Changing Voices of Historical Narration in Western India, 1400–1900’, The American Historical Review, 109(4), 2004: 1084–1103, and ‘Transitions and Translations: Regional Power and Vernacular Identity in the Dakhan, 1500–1800’, Comparative Studies of South Asia, Africa and the Middle East, 24(2), 2004: 23–31. (9) Even today, Pavagadh remains an active pilgrimage site for the worship of the goddess Kali in Gujarat. A number of traditional garba songs from the region are also dedicated to the Kali who resides at Pavagadh. However, the remains of a Lakulisha-Mahadeva temple, as well as an actively worshipped Jain shrine, are also to be found on the hill. The numerous dargahs and mosques that survive from the medieval city of Champaner-Muhammadabad at the base of the hill further contribute to the complex religious geography of the site. (10) Excavations were first conducted at the site of Champaner in the 1940s by the German scholar Hermann Goetz. Another archaeological project was led by R.N. Mehta of the University of Baroda (1969–75). Mehta’s report on the excavations gives important insights into the region’s history as he combines his archaeological finds, starting from pre-historic times, with literary sources and oral traditions; R.N. Mehta, Champaner: A Medieval Capital, Baroda: Heritage Trust, 1978; and Cāmpāner: ek adhyayan, Baroda: Maharaja Sayajirao Vishwavidyalaya, 1979.

Page 20 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings (11) For an early account of the rivalry between the rulers of Junagadh and Patan, see Merutungacharya, Prabandhacintāmaṇi, Durgashankar Kevalram Shastri (ed.), Bombay: Forbes Gujarati Sabha, 1932. (12) The epithet ‘begada’ comes from the Gujarati term for ‘two forts’ and the Sultan is said to have become well known as this after his conquest of the two important forts of Champaner and Junagadh. (13) See Samira Sheikh, Forging a Region: Sultans, Traders and Pilgrims in Gujarat, 1200–1500, New Delhi: Oxford University Press, 2010, and ‘Alliance, Genealogy and Political Power: The Cuḍāsamās of Junāgaḍh and the Sultans of Gujarat’, Medieval History Journal, 11(1), 2008: 29–61. (14) Chattopadhyaya relates the proliferation of dynasties since around the sixth century of the Common Era to a continuous process of regional and local state formation. He identifies the process of local state formation, the peasantization of tribes and caste formation, and the appropriation and integration of local cults as the three major interconnected processes at work through all the phases of Indian history. His articulation has provided an alternative approach to the study of the period between the end of the Guptas and the beginning of the Delhi Sultanate, which was viewed, until recently, as a period of crisis and decentralization, though since his work ends around 1200, he does not address the contribution of the Islamic states to this process, underscoring instead the role of Brahminical ideology and especially of the Rajput ruling lineages. See B.D. Chattopadhyaya, The Making of Early Medieval India, New Delhi: Oxford University Press, 1994. (15) Hermann Kulke and others have also observed continuities in state formation in Orissa, particularly through the process of cult assimilation; see A. Eschmann, H. Kulke, and G. Tripathi (eds), The Cult of Jagannatha and the Regional Tradition of Orissa, New Delhi: Manohar, 1978. B.D. Chattopadhyaya’s suggestions are further born out in Nandini Sinha-Kapur’s study of state formation in the Guhila kingdom of Mewar, which emerged as a significant regional kingdom after the decline of the Delhi Sultanate; see N. Sinha-Kapur, State Formation in Rajasthan: Mewar During the Seventh–Fifteenth Centuries, New Delhi: Manohar, 2002. (16) See Sheikh (2008), pp. 32–3. (17) Sheikh (2008): p. 36. (18) Sheikh (2008), p. 36–8. (19) J.W. Watson, ‘Historical Sketch of the Hill Fortress of Pāwāgaḍh, in Gujarāt’, Indian Antiquary, 6 (1877): 1. Several lineages of Rajasthan and Gujarat trace their migrations from their original homelands to ‘Ala’ al-Din’s incursions to the Page 21 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings region. However, many of these are later recordings and it is difficult to establish the veracity of this tradition. (20) See Watson (1877), pp. 2–3 for full text of the inscription. (21) Sikandar Manjhu Gujarati, Mirati Sikandari (A Study in the Medieval History of Gujarat), Fazlullah Lutfullah Faridi (trans.), Gurgaon: Vintage Books, 1990 (repr.), pp. 14–15. (22) Sikandar (1990). He continues, ‘Sultān Mehmūd He continues, ‘Sultān Mehmūd without any regard for Islam in his venal greed for money, marched his army to Dohad which town was under Gujarat on the frontier of Malwa. On hearing this, Sultān Muḥammad raised the siege and came to the village of Kothra Saonli, where falling ill, he returned to Ahmedabad and died on the twentieth of Muharram A.H. 855 (CE 1451–1452). (23) These are often referred to in the texts from these courts in formulaic terms as the ‘eighteen varnas’. (24) The manuscript of the Gaṅgadāsapratāpavilāsanāṭakam is in the British Library (Ms. 2388). It seems to be missing a few pages and lacks a colophon with the exact date and place of production; it only tells us: ‘this book belongs to the excellent Vaidya Bhamaji (vaidyavaraśībhāmājīmahattamānam pustakamidam)’ (f. 136). No commentaries on the text have yet been discovered, nor has the text ever been translated, although Sandesara has discussed some portions of it; B.J. Sandesara, ‘Gaṅgadāsapratāpavilāsa by Gaṅgādhara: A Historical Sanskrit Play Depicting the Conflict Between Sultān Muhammad II of Ahmedabad and the King Gaṅgadāsa of Chāmpāner’, Journal of the Oriental Institute, 4, 1953–4: 193–204; and ‘Detailed Description of the Fort of Chāmpāner in the Gaṅgadāsapratāpavilāsa, an Unpublished Sanskrit Play by Gaṅgādhara’, Journal of the Oriental Institute, 17, 1968–9: 45–50. Sandesara subsequently edited the play, which was published by the Oriental Institute in 1973. All translations are mine. (25) Sandesara in Gangadhara (1973), p. ii. (26) I am grateful to Francesca Orsini pointing out the link with Hindavi in this part of the text. One example of their speech is as follows: aṣkaundālam dekhataṅ kimu laḍho chohmi khudālammkā baṅdā tīra kamāṇā lekara hiṅdū diwānā ihāṅ, āyā jāya kahāṅī? tāla pagadoṅ ghālo galāṅ pāgaḍī bistākī kartā khudālam age dartā nahin amhakuṅ. Gangadhara (1973), act 7, p. 53.

Page 22 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings (27) Garasiya (called ‘grāsino rājānaha in the text), a term that came to be used in the period for landholders. It did not specify the ethnic or community origin of the landholders (although Kolis or Bhils were usually singled out). This term could be used for landholders of different levels. (28) From the Persian accounts we know that Muhammad Shah conquered Idar in the year 1441. The Raja of Idar also gave his daughter to the Sultan in acceptance of his suzerainty. The Sultan attacked Bagar in the same year; see E.C. Bayley, The Local Muhammadan Dynasties. Gujarat (History of Gujarat), 2nd ed., New Delhi: S. Chand and Co., 1970. The two allies of the Sultan could perhaps have been the chieftains of these kingdoms. (29) Gangadhara (1973), 8, p. 62. (30) Even though its author provides no information about himself except that he ‘was the conqueror of the poets of the Kali age’, it is indeed quite possible that the itinerant poet from Vijayanagara, who had travelled to the courts of Ahmadabad and Champaner, stopped at the Junagadh court on his way from Dwarka in order to continue his poetic digvijaya, or conquest of the directions; Gangadhara (1973), 1, p. 18. (31) Sheikh (2003), p. 85. (32) The Jhalas and Gohils were locally powerful clans based in Saurashtra. (33) Ali is specifically concerned with the courtly sources of beauty, refinement, and love, which, he points out, were most volubly attested by literary texts that were produced and heard widely at the households of men of rank. These included a wide variety of praise-poems or eulogies, particularly in the form of inscriptions, as well as exchanges of letters, manuals on style and performance like the Nāṭyaśāstra, as well as shorter proverbial verses and stories with morals like the Pañcatantra, and manuals on love and sexuality like the Kāmasūtra; see D. Ali, Courtly Culture and Political Life in Early Medieval India, Cambridge: Cambridge University Press, 2006. (34) Ali (2006), pp. 78–85. (35) Gangadhara (1954a), 3.3, p. 43. (36) Gangadhara (1954a), 3.4. The poet elaborates on this theme: no one in the kingdom talks excessively, except when praying to Lord Purushottama when Mandalik only spoke a few, sweet and clever words (3.5). Nobody tells lies, except the deceitful lover, and if anyone did tell one it would only be for the benefit of others and not with a selfish motive (3.6). (37) H.D. Velankar, ‘Māṇḍalīka’, the Last Great King of Independent Saurāṣṭra’, Bharatiya Vidya, 14, 1953: 45, and Gangadhara (1954a), 3.58, p. 50. Page 23 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings (38) Gangadhara (1954a), 3.63 and Velankar (1953): 45. (39) Gangadhara (1973), 8, p. 71. (40) Gangadhara (1973), 8, p. 71. (41) Gangadhara (1973), 4, p. 40. (42) Gangadhara (1973), 4, p. 40. (43) Gangadhara (1973), 8, p. 67. (44) Gangadhara (1973), 8, pp. 67–8. (45) Gangadhara (1973), 8, p. 68. (46) Gangadhara (1973), 8, p. 70. (47) Velankar (1953): 44. (48) Velankar (1953) and Gangadhara (1954a), 3, pp. 47–51. (49) Gangadhara (1954a), 3.34, 3.35 and Velankar (1953): 44. (50) Velankar (1953), p. 45 and Gangadhara (1954a), 3.40, p. 49, emphasis added. (51) Gangadhara (1954a), 3.10, p. 47. (52) Gangadhara (1954a), 3.13. (53) Gangadhara (1954a), 3.23, p. 48. (54) Gangadhara (1954a), 3.22. (55) Gangadhara (1954b), 8, pp. 28–32. (56) Gangadhara (1954b), 9.10, pp. 34–5. (57) Gangadhara (1954b), 9, pp. 33–7 and Velankar (1953): 51. (58) Gangadhara (1973), 1, pp. 1–2. (59) The source of these details is a well-known figure of such narratives, the treacherous Brahmin: here it is a Brahmin who regularly visits the fort in order to receive the generous donations that Bhamaba, Gangadas’s mother, continuously makes to Brahmins. (60) Gangadhara (1973), 7.7, p. 57. (61) Gangadhara (1973), 7.9, 7.10. Page 24 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Universal Poet, Local Kings (62) Gangadhara (1973), 7.11, 7.12. (63) Gangadhara (1973), 7.14. (64) Gangadhara (1973), 7.30, p. 59. (65) Gangadhara (1973), 4, p. 36. (66) Gangadhara (1973), pp. 36–7 and verse 17, p. 37. (67) The messenger uses an interesting analogy here. He says that the king first gained control over the source of the flow of the water and having freed this flow, he then managed to acquire the ocean, meaning that with the help of a small army he has been able to obtain more soldiers thus enlarging the size of his core army. Gangadhara (1973), 4.22, p. 38. (68) Gangadhara (1973), 4.26, p. 39. (69) Gangadhara (1973), 4, p. 39. (70) Gangadhara (1954a), 4.8. (71) Gangadhara (1954a), 4.9. (72) Gangadhara (1954a), 4.15. (73) Gangadhara (1954a), 4.19. (74) Gangadhara (1973), 1, p. 4. (75) Gangadhara (1973), 9, p. 73. (76) Gangadhara (1973), 9, p. 73. (77) Gangadhara (1973), 2, p. 18. (78) Y. Bronner and D. Shulman, ‘“A Cloud Turned Goose”: Sanskrit in the Vernacular Millennium’, Indian Economic and Social History Review, 43(1), 2006: 1–30.

Access brought to you by:

Page 25 of 25

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550*

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* Ramya Sreenivasan

DOI:10.1093/acprof:oso/9780199450664.003.0009

Abstract and Keywords Between the mid-fourteenth and mid-sixteenth centuries, when Delhi’s power shrank to that of a regional kingdom, a number of chieftains rose to power as rulers of small kingdoms and lineages. Such rulers and military entrepreneurs attempted to join the political elite by patronizing poets and performers whose compositions, in turn, reflect the values and contexts of their patrons. The chapter sees the romance and narrative genre popularized by texts such as the Candāyan, the Chitāī-carita, and the Padmāvat as evidence of the entrepreneurial politics of groups who are otherwise missing from the historiography of the period. Keywords:   Sultanate, Malwa, Gujarat, Rajput, courtliness, masnavi, Sufi, warriors

NORTHERN INDIA EXPERIENCED SIGNIFICANT IMPERIAL expansion in the early decades of the fourteenth century under ‘Ala’ al-Din Khalji and then Muhammad bin Tughluq, and then again in the late sixteenth century under the Mughal emperor Akbar. In the intervening period, provincial governors and regional chiefs asserted themselves against emperors who aspired to centralize authority. Such chiefs drew in turn on the resources of local warlords who provided soldiers to ambitious regional satraps such as the Sultans of Jaunpur, Malwa, and Gujarat, or the (p.243) Rajput chiefs of Rajasthan.1 But while regional rulers in the fifteenth and sixteenth centuries have left behind a rich architectural and inscriptional record, their local allies—men who occupied the continuum between village headman and successful warlord—have been far less Page 1 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* visible in an official archive that was produced and reorganized during successive waves of state-building and resource extraction in northern India. And yet such local chiefs, who often controlled strategically important towns like Dalmau, Sarangpur, and Chanderi in the hinterland, clearly would have aspired to assert their own membership in the broader north Indian political elite. Their limited resources precluded the lavish construction of temples, palaces, or fortresses—indeed, it is not clear whether some of these chiefs’ households could even be considered as ‘courts’. But it was to such aspiring chiefs that several vernacular courtly narratives—including Mawlana Da’ud’s Candāyan, Narayandas’s Chitāī-carita, and Malik Muhammad Jayasi’s Padmāvat—were addressed between the late fourteenth and mid-sixteenth centuries in northern India. From the evidence of such literary narratives, it is apparent that one of the strategies that such rural gentry and local warlords in the hinterland adopted to assert their elite status was the patronage of poets and performers. In the same period, tales about heroic warriors killed in battle defending their land, cattle, and honour also began to circulate as oral epics in northern India. Thus, it was in the fourteenth and fifteenth centuries that legends about warriors such as Lorik and Pabuji began to circulate in parts of modern Uttar Pradesh and Bihar, and in modern Rajasthan respectively.2 Both the specialist performers and the audiences for such tales about the often miraculous exploits of warrior heroes, and their (p.244) death in battle against an invariably unjust overlord, were typically from non-elite groups, such as the peasant-pastoralist Ahirs for the tales about Lorik, and the low-status Nayaks and pastoralist Rebaris who are the most numerous devotees of the warrior-turned-god Pabuji. I argue in this chapter that in narratives like the Candāyan (c. 1379), the Chitāīcarita (c. 1526), and the Padmāvat (c. 1540), all composed in local variants of the Hindavi vernacular, we see the socio-political world of the oral epics— articulating the political assumptions and expectations of its non-elite performers and audiences—finding patronage at the marginal ‘courts’ of aspiring ‘elites’ in the period between the late fourteenth and mid-sixteenth centuries. I argue that by adapting the received conventions of canonical classical genres like the Sanskrit mahakavya (courtly epic) and the Persian masnavi (romance), poets like Da’ud, Narayandas, and Jayasi were essentially addressing the particular political concerns and constraints of their patrons— rural gentry and local warlords in the hinterland.3 Mawlana Da’ud’s Candāyan was composed in Dalmau (in modern Uttar Pradesh) in 1379, Narayandas’s Chitāī-carita in Sarangpur (in modern Madhya Pradesh) circa 1520, and Malik Muhammad Jayasi’s Padmāvat in the Rae Bareli area (in modern Uttar Pradesh) in 1540. All three narratives were composed in local forms of Hindavi rather than the classical Sanskrit or Persian. While excerpts were often recited, they were clearly transmitted primarily as written manuscripts meant to be read. Two—the Candāyan and the Padmāvat—were composed by poets who were Sufi initiates, while less is known about Page 2 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* Narayandas who composed the Chitāī-carita.4 All three narratives consciously adapted classical literary models to local contexts, though in the instance of the Candāyan and Padmāvat, their adaptations of canonical Sufi doctrine and narratives have been the primary focus of scholarly enquiry.5 (p.245) The classical models for the Candāyan, the Chitāī-carita and the Padmāvat—in the masnavi and mahakavya genres—typically dealt not with ambitious warriors but with great kings from illustrious lineages. Between the late fourteenth and mid-sixteenth centuries, however, in the absence of successful imperial authority, regional state-building projects proliferated. That is to say, the absence of a successful and durable centralizing regime in this period encouraged the emergence of the kinds of local warlords described by Dirk Kolff: military entrepreneurs with an ability to recruit peasant and pastoralist soldiers seasonally.6 In this context, poets seeking patronage from such local patrons articulated the latter’s distinctive concerns in novel adaptations of received classical genres. Thus, the new, vernacular masnavi or kavya of this period celebrated the great warrior rather than the distinguished king from an ancient and exalted lineage. By the 1570s, the political context had changed decisively with the emerging new imperium under the Mughal emperor Akbar. In this altered context, the vernacular tales about ambitious warriors that had emerged in the fifteenth century no longer found aristocratic patronage. While such tales continued to circulate in the ‘folk’ and ‘oral epic’ domains, in courtly contexts we see instead a resurgence of more canonically oriented poetic narratives (both Persianate and Indic).7 These included narratives about kings from putative royal (p.246) lineages of asserted antiquity, the only kinds of ‘kings’ that Akbar and his Mughal successors would negotiate with, as well as conduct manuals about cultivating masculinity and/or aesthetic refinement befitting normative courtiers and rulers.8 In the hinterlands beyond emergent regional capitals between the fourteenth and sixteenth centuries, by contrast, vernacular narratives from marginal courts —such as the Candāyan, the Chitāī-carita, and the Padmāvat—all focus on a valiant warrior afflicted by desire who embarks on a quest to win (or regain) his beloved. The quest typically involves some austerities and renunciation, and/or initiation into ascetic discipline by a prominent master belonging to the Shaiva ascetic order of the Naths. Nath yogic powers then enable the warriorprotagonist to obtain (or regain) his beloved. In parallel, these warriorprotagonists have to negotiate the thorny question of a mutually beneficial relationship with an ever-looming overlord. The sticking point in such negotiations is typically the status of the warrior’s beloved, desired equally by the overlord for her reputed beauty. The resulting conflict has a different outcome in each of these narratives.

Page 3 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* The local chiefs to whom such narratives were addressed between the late fourteenth and mid-sixteenth centuries did not build grand new temples or forts, as I said. Their control over existing fortresses was precarious, and they did not have substantial patrimonial lands. Nor have they left behind an inscriptional record of gifts to individuals, religious groups, or shrines. In the absence of adequate resources to assert their claims to kingship in the conventionally accepted idiom of their time, such local chiefs and warlords adopted two other practices as significant markers of their rank: first, the scale of their households and their (p.247) households and their control over its women, and second, the patronage of poets, scholars, and performers. Both in the trajectories of such chiefs themselves as well as in the narratives addressed to them, these practices constituted the key markers of their elite status. The literary history of South Asia, including the more historically sensitive revisionist version that has emerged in recent years, has focused on genre primarily as a universe of literary rules, only broadly susceptible to significant historical changes such as imperial expansion or the emergence of regional polities.9 In focusing on new vernacular adaptations of courtly genres in the long fifteenth century, I hope to show how the changing sociology of patronage—the emergence of petty chiefs wielding significant political power—was articulated much more closely than the scholarship has probed yet, by poets making significant revisions to the classical models that they had inherited and were deeply familiar with. Furthermore, literary and narrative sources constitute our primary sources for reconstructing a history of courtly cultures in the South Asian context. The new vernacular tales about ambitious warriors that emerged between the late fourteenth and mid-sixteenth centuries can, therefore, also tell us a lot about the character of nascent or aspiring courts. Daud Ali has explored how normative courtly practices structured the political and affective relationships of a new aristocratic society in the early medieval period, and even helped to constitute normative selfhood for its aspiring rulers and aristocrats.10 From the vantage (p.248) point of marginal courts in the hinterland between the fourteenth and sixteenth centuries, however, narratives like the Candāyan, Chitāī-carita, and Padmāvat suggest that courtliness appears to have been a threshold of normative practices that aspiring elites among the rural gentry and local warlords of northern India crossed only partially. This chapter begins by describing the historical context of fifteenth-century northern India, especially the continuing contests over resources that characterized qasba towns like Dalmau and Sarangpur where tales about ambitious warriors were composed. There follows a discussion of these three tales of love in which I explore how they adapted inherited, classical conventions in articulating the concerns typical of their local patrons. In the alternate narrative model that these tales articulate, a successful warrior is expected to successfully negotiate a favourable relationship with the imperial overlord. He is also expected to cultivate himself through the pursuit of ascetic discipline. And Page 4 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* he is expected to maintain his household on a certain scale, in terms of both size and conspicuous consumption. In short, both in these narratives and within the historical context in which they were located, the warrior’s relationship with his overlord and his control over ‘his’ women were crucially constitutive of his political authority in both his household and in the wider community.

Hinterland Politics and Marginal ‘Courts’ in Northern India C. 1350–1550 The scholarship on the Avadhi Sufi tales of love such as those composed by Da’ud and Jayasi has reconstructed a primarily Sufi cultural and political landscape in which such narratives were composed and circulated. In contrast, the scholarship on poets adapting Indic themes such as Vishnudas (see Chapter 13 in this volume) or on the later, seventeenth-century riti poets, has reconstructed a primarily Rajput political and cultural landscape in which such poets functioned.11 While the immediate (p.249) contexts signalled within such narratives may indeed have been Sufi or Rajput respectively, when we consider the evidence available about places like Dalmau and Jayas (both in Rae Bareli district today), and about Sarangpur, Chanderi, and Raisen—the locations where Hindavi tales about ambitious warriors emerged—we get a far better sense of the complex relationships between local communities, men of religion, and local chiefs, patrons, and overlords.12 Mawlana Da’ud composed the Candāyan between 1371 and 1379 in the town of Dalmau, on the banks of the Ganges. The town, just south of Rae Bareli today, seems to have been the site of local skirmishes in the late fourteenth century. In his invocation, Da’ud mentions that Firuz Shah (Tughluq) was the Sultan of Delhi at the time, and that Jauna Shah was his renowned wazir. He also mentions a Malik Mubarak, the son of the valiant Malik Bayan, as the local chief (miru). From contemporary sources, we know that parts of the region that is now Uttar Pradesh remained restive through the fourteenth and fifteenth centuries. We are told by the fifteenth-century chronicler Yahya ibn Ahmad Sirhindi that in 1394 the rais and zamindars of Dalmau had stopped their payments, and that the Hindu chiefs in the area had looted and destroyed forts and ravaged the countryside. Malik Sarwar, who founded the Sharqi Sultanate in Jaunpur (to the east of Dalmau) in 1394, was the trusted lieutenant who had been sent out from Delhi by Firuz Shah Tughluq to crush the ‘eastern rebels’ that same year. We know that he attacked local chiefs in Dalmau, Avadh, and Qannauj.13 As for Mawlana Da’ud himself, as Simon Digby argues: His way of living and livelihood was not that of a recognized Sufi holy man. He had professed discipleship to a great figure in the Chishti lineage … [and] had received miscellaneous teaching in Shaikh Nasir al-Din’s dargah. This would (p.250) entitle him to be called Maulana … Dalmau was a military checkpost of importance, and given the suggested profession of

Page 5 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* older members of his family, it is likely that he was a soldier or garrisonofficial there …14 A Persian-language ethical manual from late-fourteenth-century Avadh exhorts Muslim peasants in the countryside to defend their lands against ‘Kafirs … at such time as you can see that the Kafirs having been making a general disturbance’.15 A couple of decades later, the prominent Sufi Sayyid Ashraf Jahangir Simnani advised Hushang Shah Khalji of Malwa about the conditions for waging war (jihad) against infidels: while jihad was mandatory for all Muslims when confronted with an invasion by external infidels, it was optional in the instance of an internal rebellion by kafirs.16 That such counsel was given freely to both rulers and commoners in the normative discourses of the time suggests that contests over control of land and resources continued through the fifteenth and early sixteenth centuries in the territories of the Delhi, Jaunpur, and Malwa sultanates. Accommodation and coexistence were also discernible in this north Indian hinterland. Local legend connects Malik Muhammad Jayasi who composed the Padmāvat (c. 1540) with the presumably Rajput Raja of Amethi. Whether under the control of local Rajputs or Afghans, the area continued to be a centre of Chishti Sufi activity throughout the sixteenth and seventeenth centuries. In the early sixteenth century, Shaykh Nizam al-Din was appointed the spiritual successor (khalifa) to Shaykh Ma‘ruf Jaunpuri in Amethi, and continued living there.17 In the invocation to the Padmāvat, Jayasi also connects himself with the discipline lineage of Sayyid Ashraf Jahangir Simnani (see above), and with Shaykh Burhan al-Din of Kalpi. Much more information is available about the patron to whom Narayandas addressed his Chitāī-carita in the 1520s—Silhadi Purabiya, the Rajput chieftain who controlled the towns and hinterland of (p.251) Sarangpur, Raisen, Chanderi, and Bhilsa in northern Malwa for about two decades. Sarangpur (about 300 miles to the south of Dalmau), where the Chitāī-carita was composed or recited in the 1520s, is believed to have been founded in the fourteenth century by Shaykh Sarang, a disciple of Shaykh Qiwam al-Din, who in turn had been initiated into the Suhrawardiyya order by the prominent Sayyid Jalal al-Din Makhdum Jahaniyan in Uchch (in modern Sind). Shaykh Sarang was an important functionary in the administration of Firuz Shah Tughluq.18 A 1493 inscription in Arabic from Sarangpur records the construction of a madrasa there in the reign of Ghiyas Shah Khalji of Malwa.19 Shah Manjhan Shattari (the author of another Sufi tale of love, the Madhumālatī, c. 1545) moved to the town in 1553. ‘A large number of scholars began to attend his Sarangpur seminary and khanqah and the town began to vie with Shiraz as a prominent centre of learning’. Before moving to Sarangpur, Manjhan had been at Raisen, held by Silhadi and his sons before its conquest by Sher Shah Sur in 1543. The presence of a sizeable Muslim community in Raisen is suggested by the fact that Sher Page 6 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* Shah invited Manjhan to move to Raisen ‘and to assume the post of Shaikhu’lIslam’ there. Manjhan established his own khanqah in the town, and moved to Sarangpur only when the Rajputs reconquered Raisen in 1553.20 The third garrison town and its hinterland that Silhadi controlled in northern and northeastern Malwa in the 1520s was Chanderi, where soldiers and settlers from Delhi, many of whom owed allegiance to the Chishti Shaykh Nizam al-Din, had founded a new settlement in the early fourteenth century. Archaeological evidence from the town suggests that at the end of the fifteenth century the town had a total population of about 30,000, with a permanent garrison of between 3,000 and 5,000.21 As the ample inscriptional record shows, a Sufi (p. 252) tomb, several stepwells, several mosques with attached gardens, as well as a congregational mosque were built throughout the course of the fifteenth century. These buildings were sponsored by prominent noblemen, local notables, even a pair of local grocers and the wives of a local Sufi Shaykh.22 And as Simon Digby has pointed out, in the late fifteenth century one dominant group of the inhabitants of the town were the Shaikhzadas, who were the descendants of Sufi settlers … They held assignments of the agricultural land around the walled town, which yielded harvests that contributed to sustain the local urban population. Further evidence … suggests that the Chanderi Shaikhzadas, like other rural qasbacommunities of Muslims, were trained in the use of arms and had their own supply of horses.23 The abundant inscriptional record at Chanderi dates to the fifteenth century under the Khalji Sultans of Malwa, and from the second half of the sixteenth century onwards under the Mughal Emperors. It is noteworthy that no inscriptions have been discovered from the 1520s when Silhadi Purabiya controlled the town—a fact that suggests the more fleeting and limited nature of his rule there. Silhadi’s career (as reconstructed by Dirk Kolff) conveys the same sense of alltoo-evanescent resources and status. Silhadi was born in the village of Suhjana, about 12 miles southeast of Gwalior—we know because Babur visited the place in 1528, curious about where Silhadi had been born. It is not clear whether he was a Tomar, though modern historians have generally assumed that he was.24 He controlled parts of northern Malwa in the second and third decades of the sixteenth century: by 1520 Rana Sanga of Mewar had married his daughter to (p.253) Silhadi’s son, and by 1529 Silhadi’s growing power in Malwa and intrigue within the Khalji court may have led to the Malwa Sultan instigating an attempt on his life. Silhadi fled, first to Mewar and then to the court of Bahadur Shah Gujarati, who clearly wished to retain his support. The Gujarat Sultan first honoured Silhadi with a gift of 700 robes of gold and 70 horses, and then with an additional gift of 30 elephants and 1,500 robes. The scale of these gifts is an Page 7 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* indicator of Silhadi’s perceived rank at the Gujarat court. In 1531 Silhadi aided Bahadur Shah during the latter’s victorious campaign against the Malwa Sultan; in return he was given rights to revenue from the sarkars of Ujjain and Sarangpur, and the strategic fort of Raisen. However, the question of his appropriate rank among the amirs at the Gujarat court remained unresolved. As tensions between Silhadi and Bahadur Shah continued, the latter laid siege to Raisen, which was held by Silhadi’s brother Lakshman Sen. Under pressure, Silhadi seems to have converted to Islam.25 The seventeenth-century Mir’āt-i Sikandarī partly ascribes the tensions between Silhadi and Bahadur Shah to the status of Muslim women alleged to have been part of the Rajput’s household. When Bahadur Shah hears the allegation, he is irate: ‘This wretch keeps Muhammadan women in his house, and by the holy Law of the Prophet he deserves death. I will never let him go alive unless he becomes a Musalman’.26 By 1531, Silhadi had lost Ujjain and Sarangpur to Bahadur Shah. In the final assault on the fort of Raisen, the Mir’āt-i Sikandarī records: The seven hundred wives of Silhadi, and the wives and daughters of the other Rajputs and the daughter of the Rana, the wife of Bhupat, threw themselves together into the flames and were reduced to ashes … It is related that Durga [the chief wife of Silhadi] forced all the Musalman women kept by Silahdi to enter the pile, and they were all reduced to ashes except one whom Heaven saved from this Nimrodian holocaust and who came forth alive from the flames.27 Hariharnivas Dvivedi, the prominent local historian of Gwalior, surmises that the Pir Salah al-Din, at whose grave inside the Raisen fort (p.254) both Hindus and Muslims worshipped in the 1950s, was Silhadi, buried there by Bahadur Shah.28 By the mid-sixteenth century, it would seem from the careers of Silhadi’s sons, imperial overlords were far less willing to negotiate with the local chiefs who had wielded considerable power in the course of the fifteenth century. Silhadi’s son Bhupat (who was also Rana Sanga’s son-in-law), who had survived the jauhar (mass immolation) of 1532, joined the service of the same Bahadur Shah (his father’s friend-turned-foe) for three years, but may have betrayed him in 1535 during Humayun’s siege of Mandu. By 1537 Bahadur Shah had died and Malwa had a new ruler, Mallu Khan or Qadir Shah, who again looked for support to the Rajput and Muslim amirs of the old Khalji regime. Bhupat Rai and Puran Mal—Silhadi’s sons—then returned to Malwa from Mewar, to be given control of Raisen once again. Bhupat died soon after, while Puran Mal remained dominant in Raisen and Chanderi until the Sur campaigns of the early 1540s.29 By 1543, the Raisen Rajput lineage may have come to an end. ‘Abbas Khan Sarwani records how Sher Shah Sur captured Puran Mal’s daughter and three of his nephews (brother’s sons) in 1543. The Sur ruler then ‘gave away the daughter of Puran Mal to a rope-dancer in order to make her dance in the market and Page 8 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* ordered the aforesaid three boys to be castrated in order to put an end to the line of the oppressor’.30 Such brutality reminds us of Akbar’s complete destruction of the fort of Chitor in 1569. To return to Silhadi, it is clear that chroniclers in rival courts—in Malwa, Gujarat, and the Mughal court—paid considerable attention to him. None of these sources regarded him as a king, however. Silhadi left no monuments behind, no new temples, and only one known inscription in the Raisen fort, dating to 1526. This presents a striking contrast to both Rana Kumbha in Chitor half a century earlier, and Man Singh Kachwaha in Amber half a century later, both of whom were newly ascendant regional chiefs proclaiming the extent of their authority (p.255) through multiple, monumental building projects, and through an abundant inscriptional record.31 Nor did Silhadi build any new forts like Kumbha had done. Moreover, his control over the territories of Raisen, Chanderi, Sarangpur, and Bhilsa in Malwa was relatively short-lived, as we have seen above. If we consider Malik Mubarak of Dalmau in 1379, Silhadi Purabiya in Malwa in the 1520s, and the unnamed Rajput Raja of Amethi in 1540, it becomes clear that we are dealing with rural gentry and local warlords with limited resources. The perennial vulnerability of such patrons—both to challenges from within the locality and to pressure from an aspiring overlord— can be gleaned from the narratives that emerged from their ‘courts’.

Classical Genres Adapted at Hinterland Courts, 1350–1550 Modern scholars have paid much attention to Da’ud’s Candāyan and Jayasi’s Padmāvat as Sufi narratives.32 Here I am concerned instead with the treatment of the warrior-protagonists and their social and political relationships in these narratives and in Narayandas’s Chitāī-carita. The representation of warriors in narratives composed at marginal courts in this period, I argue, was shaped by both the conventions of inherited, courtly genres, as well as by the political horizons of the patrons for such narratives—patrons such as the unnamed Raja of Amethi associated with Jayasi, and Silhadi Purabiya in Malwa. Having discussed the (p.256) political circumstances of such patrons between the late fourteenth and mid-sixteenth centuries in the preceding section, I now explore the conventions of courtly genres as articulated and adapted in these narratives. While Da’ud, Narayandas, and Jayasi reveal a deep familiarity with the standard topoi of courtliness from canonical texts, they modify the plots of their narratives and the trajectories of their protagonists in order to reflect the political constraints and opportunities faced by such warriors in this period. In what follows, I provide brief summaries of the plots of these three narratives before demonstrating their familiarity with the literary conventions of canonical courtly genres. In Mawlana Da’ud’s Candāyan, the beautiful Chanda, who belongs to the most beautiful and accomplished category of padmini women, is the daughter of Mahar Sahadev who rules over the splendid city of Govar. When she is four years Page 9 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* old, she is married to Siuhar Bavan, the son of Jait. The match is appropriate because Jait is of prominent rank, and is also accomplished and virtuous; he is respected throughout the land (jāti karama guna āgara desa māna sabha loga [36]).33 The young Chanda moves to her father-in-law’s household. Twelve years pass, but her husband Bavan is short, squint-eyed, and filthy from having never washed his hands, feet, or face. Worse, they do not sleep together (au tehi upara saṅgi na sovā [43]). The bitterly disappointed Chanda sends word of her plight to her father and moves back to her natal home, disregarding her mother-in-law’s pleas. A mendicant musician (originally from Ujjain where Bikamajit rules) comes to Govar and begs for alms; he catches a glimpse of Chanda as she is standing at her palace window and is stricken with a fever [54]. Recovering, he leaves Govar and reaches Rajpur. To its king Rupchand the musician describes Chanda’s beauty in an elaborate catalogue of her attributes (sarāpā/nakha-śikha varṇan), and the king of Rajpur decides at once to attack Govar. Approached by emissaries of Mahar Sahadev, Rupchand demands the hand of Chanda in marriage. Sahadev’s emissary remonstrates with Rupchand: such a demand does not befit a king who should regard the daughter of another man as his mother. Besides, Chanda has already been married to another man; Sahadev’s men are ready to (p.257) die and have their city (gāũ) burned down, but they will not surrender Chanda [96]. While many of Sahadev’s advisors are disheartened, the warrior Lorik kills Rupchand’s foremost warrior, Bantha, just as Raghava [Rama] had earlier rescued the gods and men [133]. Lorik is summoned to the palace in Govar by a grateful Mahar; meanwhile, Chanda is curious about the warrior who has saved Govar. When she catches a glimpse of him, she becomes restless with desire. She asks her father Mahar to invite all his subjects to a great feast of thanksgiving for his victory over Rupchand. When Lorik catches a glimpse of her at the window during this feast, he faints. Upon recovering, he pleads with Chanda’s servant Biraspati to arrange a meeting with his beloved. When the servant tells him that this is impossible (for he is a lowly warrior and Chanda is a king’s daughter), Lorik becomes a follower of the path of Gorakh—a Nath yogi— practising austerities for a whole year at a temple. He sings songs of love, takes to eating the wild fruit of the forest, and meditates upon his essential truth, Chanda [164]. The servant Biraspati eventually arranges for the lovers to meet on a dark night in Chanda’s bedroom. Rumours of their relationship cause fisticuffs between Lorik’s irate wife Maina and Chanda. Besieged by the scandal, Chanda gives Lorik an ultimatum and they elope, pursued unsuccessfully by Bavan, Chanda’s husband. The scorned Bavan now curses Chanda that she will be bitten by a snake. Two snakebites and two miraculous cures later, Lorik and Chanda settle in the city of Hardi Patan where they find refuge. Their idyll is, however, Page 10 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* disrupted by Maina’s grief at being separated from Lorik. Upon receiving Maina’s message, Lorik he sets off for Govar with Chanda. Upon his return he finds that Chanda’s father Mahar Sahadev has killed his brother and seized his cattle and lands, reducing his mother Kholin and his wife Maina to penury. Da’ud’s narrative ends abruptly at this point. In some modern versions of the oral epic Lorikāyan, which diverges from Da’ud’s narrative in many respects, Lorik returns home and dies in battle confronting his many enemies.34 (p.258) In Narayandas’s Chitāī-carita, Chhitai is the daughter of Ramadev, king of Devagiri. Confronted by Alauddin (‘Ala’ al-Din) Khalji’s army, Ramadev agrees to accept the Sultan as his overlord, and spends three years at the Sultan’s court in Delhi where he is treated with the same honour as his new overlord’s most eminent ministers.35 When he returns to Devagiri three years later to get his daughter Chhitai married, he asks Alauddin for a painter from the imperial court to help adorn the palace for the impending festivities. The painter catches a glimpse of Chhitai and is smitten with her beauty; he returns to Delhi after Chhitai’s marriage to the prince Saunrasi of Dhorasamudra [Dwarasamudra]. Saunrasi goes hunting daily in the forests near Devagiri, disregarding the apprehensions of his father-in-law who reminds him that hunting brought about the downfall of many kings in the past, including Pandu and Dasharatha. One day, Saunrasi follows a doe to Bhartrhari’s ashram; the sage is disturbed from his deep penance and tries to rescue the doe but fails. Angered, he curses Saunrasi that he will lose his wife to another man. At key points, the narrative reiterates that the subsequent course of events is determined by this curse, and therefore by Chhitai’s and Saunrasi’s destinies. Meanwhile, the painter’s praise of Chhitai has kindled Alauddin’s desire. The Sultan marches south and lays siege to the fort of Devagiri. As the siege continues, Alauddin asks Raghav Chetan (who had assisted him in his earlier conquest of Chitor) for help. Raghav Chetan goes to Ramadev with Alauddin’s demand to surrender his daughter and the fort. Of course, the king of Devagiri refuses. Meanwhile, the curious Alauddin has disguised himself as an ordinary soldier and accompanied Raghav into the fort. He is recognized by Chhitai’s servant Mainareha but pleads with her to let him escape from the fort. In return for his freedom he promises to lift his siege, withdraw from Devagiri, and abandon his pursuit of Chhitai. Mainareha does release the Sultan and he withdraws, true to his word. A grateful Ramadev wishes to reward the servant when a jealous minister intercedes and asks her to prove her (p.259) power over Alauddin by getting him to return. The irate Mainareha obliges, and Alauddin lays siege to Devagiri all over again. Ultimately, Chhitai is captured by the Sultan while worshipping at a Shiva temple some distance from the fort. She is taken back to Delhi but addresses Alauddin as her father. The Sultan accepts

Page 11 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* defeat in his quest but does not send his new ‘daughter’ back to Devagiri for fear of ridicule. Saunrasi hears of his wife’s capture and becomes a mendicant. He learns the discipline of jog from his guru Chandranath, and learns the art of enchanting men and beasts with his music. He seeks word of Chhitai as a wandering mendicant and ultimately hears at Jatashankar (where yogis congregate) that she is in Delhi. He proceeds to Alauddin’s court, wins the Sultan over with his music, and asks for Chhitai in reward. Alauddin pleads with Chhitai to surrender to the mendicant and preserve the Sultan’s word. She reveals that this mendicant is her husband in disguise. The triumphant Saunrasi returns to Dwarasamudra with his rescued wife, and with the additional gift from the Sultan of the revenues of Gujarat. Malik Muhammad Jayasi’s Padmāvat (c. 1540) describes the love between Ratansen, ruler of Chitor, and Padumavati, the princess of Singhaldvip.36 Ratansen hears of her beauty from her former companion, the parrot Hiraman. Disregarding the protests of his mother and his first wife Nagmati, he at once dons the robes of an ascetic and embarks on a quest to obtain Padumavati. He tries different strategies, first undertaking penance in a temple to Shiva and then launching an attack on the Singhal fort during which he is captured. As he is about to be executed, his identity is revealed and he is rewarded with the hand of the princess in marriage. The warrior, thus, returns to Chitor with his princess and with many wondrous gifts. The journey home involves several further adventures, including a shipwreck. When the Brahmin Raghav Chetan abuses his magical powers to deceive the king of Chitor, he is expelled. Padumavati seeks to placate the humiliated Raghav with the gift of her priceless bangle, but the vengeful Brahmin goes to the court of Alauddin in Delhi and describes her beauty to the Sultan. Alauddin lays siege to Chitor and demands the surrender of (p.260) the queen; Ratansen refuses but offers to pay tribute instead. Alauddin suggests fresh terms and enters the fort, tricks Ratansen and captures him, and returns with the captive king to Delhi. The brave warriors of Chitor, Gora, and Badal, launch a rescue, disguised as Padumavati and her companions. Meanwhile Devapal, ruler of neighbouring Kumbhalmer, offers to marry Padumavati in Ratansen’s absence. She refuses and relates the insult to Ratansen when he returns from Delhi. The latter sets off to punish Devapal, promising to return before Alauddin’s forces reach Chitor; however, both are killed in the ensuing single combat. Ratansen’s virtuous wives, Nagmati and Padumavati, become sati. When Alauddin’s army arrives, the women commit jauhar and the defenders of Chitor die on the battlefield. Alauddin acquires an empty fortress, cheated of victory even as Chitor ‘becomes Islam’.

Page 12 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* We know that the Candāyan and the Padmāvat were both read as Sufi masnavis by informed audiences in the sixteenth and seventeenth centuries.37 The masnavi, ‘the primary subject of which … [was] not heroism and valor but love’, had become a major genre of courtly poetry in the Persianate world by the mideleventh century.38 Its being embedded in the courtly world meant that even if, unlike the epic, it was not concerned with feats of individual heroism for their own sake, martial prowess and valour still remained key attributes of many masnavi protagonists and revealed normative traits of their character.39 The Candāyan and the Padmāvat deploy many of the characteristic topoi of the masnavi such as a prologue (praising God the Creator, his Prophet Muhammad, and the first four Caliphs), and an elaborate, conventionalized catalogue of the heroine’s attributes as an image of divine beauty. Both narratives also (p.261) provide glosses explaining the spiritual symbolism of the protagonist’s quest for his beloved as the soul’s quest for union with the divine. In addition, however, Da’ud’s and Jayasi’s narratives invoke poetic concepts and literary conventions ultimately derived from the Sanskrit tradition, thus sharing common ground with Narayandas’s Chitāī-carita as much in this respect as in their contexts of patronage and composition. These conventions include the aesthetics of rasa (‘flavour’, also dominant ‘flavour’ or emotion conveyed or intended to be conveyed by a narrative),40 as well as conventions derived ultimately from the Sanskrit mahakavya (courtly epic). More than half a millennium before Da’ud composed his Candāyan, those conventions included the hero’s noble birth, as well as descriptions of cities, oceans, mountains, seasons, the rising of the sun and the moon, playing in pleasure-parks and in water, drinking parties and the delights of love-making, the separation of lovers, weddings, the birth of a son, councils of war, spies, military expeditions, battles, and the victory of the hero.41 The attributes expected of the hero were further elaborated in the ninth century, by which time he was expected to be ‘completely possessed of the three royal abilities, having every excellence, with all his subjects devoted to him, and desirous of conquest …’42 The presence of these conventions in the Candāyan, the Chitāī-carita, and the Padmāvat indicates the deep familiarity of vernacular Hindavi poets in the hinterland between the fourteenth and sixteenth centuries with the topoi of classical (and canonical) literary genres. Thus, all three narratives carry detailed and conventionalized descriptions of the splendours of the capital city—its numerous tanks and lakes with their abundant waterfowl, its fragrant trees with the cooling breeze wafting through them, its busy and colourful markets, and its wealthy merchants (p.262) (Candāyan [18–30], Chitāī-carita [488–506], and Padmāvat [27–44]).43 In all three narratives the splendour of the city is clearly Page 13 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* meant to signal the status of its ruler, the father of the woman who will become the protagonist’s beloved. Thus, it is Govar, Devagiri, and Singhaldvip that are described within the conventions of mahakavya; all three narratives mention only cursorily several other cities ruled by other kings—Kalinga and Hardi Patan in the Candāyan, Dhorasamudra and Dhili in the Chitāī-carita, and Dhili in the Padmāvat. What is worth remarking is that all three narratives were composed and/or recited for patrons in fortress and market towns in the hinterland; neither such patrons nor the towns they controlled were likely to have had the resources necessary for such lavish urban architecture as the narratives describe. In other words, there is a conspicuous gap between the idealized royal city that the narratives describe, and the more modest habitations of their patrons between the fourteenth and mid-sixteenth centuries. This foregrounds the normative function of such courtly tropes when invoked for local chiefs in towns like Dalmau, Sarangpur, and Amethi. Such men may not have had the resources to build royal cities, but they could do the next best thing, perhaps: patronize poets who could describe such splendour and the political status that it implied. Equally conventionalized and, I would argue, performing the same kind of function for their audiences of patrons and potential patrons—that is, offering normative horizons of appropriate courtly behaviour—are the elaborate descriptions of lovemaking between the protagonist and his beloved in the Candāyan [212–25], Chitāī-carita [198–207], and Padmāvat [316–20] respectively.

Warrior-tales in Hinterland Courts, 1350–1550 In his discussion of the connections between lineage formation and the emergence of courtly manners in early medieval India (c. fifth to twelfth centuries), Daud Ali has pointed out how the emergence of an (p.263) aristocratic courtly culture in that period was marked by ‘a great emphasis on the regulation of the relationships among the state offices … between the king and his “princes”, “allied”, and “rival” nobles—that is, men who already possessed or could potentially possess their own household-retinues or territories’. Ali also argues that ‘adoption of Sanskrit public language by nascent lineages entailed more than simply participating in a self-consciously cosmopolitan discourse; it also meant the adoption of social values embodied in this discourse and their implications for everyday discourse’.44 In this context, it is worth comparing the vernacular ‘courtly’ tales about warriors of northern India in the long fifteenth century with the Sanskrit mahakavya texts from Gujarat discussed by Aparna Kapadia in this volume. As Kapadia’s discussion demonstrates, the latter narratives display the concern that Ali describes, to articulate and regulate every political relationship around the monarch. In contrast, the vernacular ‘courtly’ narratives of the north Indian hinterland evoke a marginal court whose hierarchies reflect those of the rural gentry and local warlord worlds within which such a court was located.

Page 14 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* Thus, the Candāyan, Chitāī-carita, and Padmāvat narrate within a courtly frame tales of warriors closer in context to what scholars have labelled as ‘martial oral epics’—oral epics that ‘turn on the themes of revenge, regaining lost land, or restoring lost rights’,45 and circulate in a non-elite domain—performed by nonelite performers for non-elite audiences such as the peasant-pastoralist Ahirs for the cycle of Lorik legends. Numerous traces of such non-elite actors and their horizons of expectations are visible in the Candāyan, Chitāī-carita, and Padmāvat, and their presence in these narratives functions to modify the literary and social conventions of more entrenched courtly contexts. Thus, none of the three protagonists is endowed with an ancient and distinguished lineage. While two— Saunrasi in the Chitāī-carita and Ratansen in the Padmāvat—are princes, the third, the Candāyan’s Lorik is merely a renowned warrior, and when an angry king seizes his livestock in his absence, Lorik’s family is reduced to penury. In a (p.264) period when other courtly narratives in Sanskrit like Nayachandra Suri’s Hammīramahākāvya (fifteenth century), composed at the Tomar court in Gwalior, continued to invoke Puranic ancestry for their protagonists as well as their real-life patrons, the absence of such exalted lineage for the protagonists of the vernacular ‘courtly’ narratives of the hinterland is striking. Equally significant is the fact that, in the world of these narratives, such absence of lineage is not an obstacle to upward mobility. Further, the warrior associates of the protagonists are of equally modest origins. The warriors and military commanders defending the fort of Devagiri in the Chitāī-carita are referred to by their first names alone; their lineages are unknown. The minister who advises Ramadev is merely Pipa the Minister (Pipa parigahi). In the Candāyan, Lorik has fifty-two unnamed followers. In the Padmāvat, Gora and Badal are brave warriors, again of unknown lineage. And on Ratansen’s epic journay to Singhaldvip, 16,000 nameless followers, a band of ash-smeared warrior ascetics, accompany him. This is a world quite different from the elaborately hierarchical and formalized aristocratic culture described by Ali and Kapadia. Equally striking is the prominent role played by female servants: Biraspati, Chanda’s companion in the Candāyan, and Mainareha, Chhitai’s companion in the Chitāī-carita. In the Padmāvat, the Brahmin woman sent by Devapal to woo Padumavati on his behalf is similarly vividly sketched. It is Biraspati who first extols the triumphant warrior Lorik to Chanda; and it is she who controls access to Chanda and helps Lorik to be united with his beloved. The power wielded by these female servants is depicted most strikingly in the Chitāī-carita. Here Mainareha captures the Sultan, extracts a promise from him that he will withdraw from Devagiri, and then frees him. Remarkably, the Sultan is true to his word. It is an arrogant minister who is brash enough to doubt the servant’s abilities and challenges her to prove her power over Alauddin. And, of course, he is proved wrong. The power wielded by such servants would be entirely unimaginable in the more hierarchical world of the mahakavya; it was much Page 15 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* more familiar in the world of the oral epics, of the peasant-pastoralist communities from which soldiers were recruited, and of the qasba towns that rural gentry and warlords from such backgrounds controlled. It would seem, then, that the most courtly aspects of the Candāyan, Chitāī-carita, and Padmāvat are their classically conventional discussions of erotics and of royal architecture. In the socio-political relationships that they describe, Da’ud, Narayandas, and (p.265) Jayasi portray the world of the hinterland that they would have been familiar with, and in which their patrons were located. Not only do the protagonists of these courtly warrior-tales operate in sociopolitical contexts where exalted lineage is unimportant, their expectations also depart from the normative aspirations of mahakavya protagonists. As mentioned above, in the world of the Sanskrit mahakavya, worthy protagonists were expected to be ‘desirous of conquest’. In the Candāyan, Chitāī-carita, and Padmāvat what the protagonists desire most is not imperial conquest but union with a beloved. There are several issues to be considered here. Vishwambhar Sharan Pathak pointed out four decades ago that the Sanskrit mahakavya routinely encodes imperial expansion in the stock trope of the victorious protagonist consolidating his conquest through marriage ‘to a beautiful princess symbolizing the goddess of Royal Fortune’ (rājyaśrī).46 In keeping with such conventions, the beautiful princess in the Candāyan, Padmāvat, and Chitāī-carita is the daughter of a powerful king. In the Sufi Candāyan and Padmāvat, in addition, the beloved Chanda and Padumavati also symbolize the perfection of the divine, so that the protagonist’s quest for his beloved is an allegory for the soul’s spiritual quest. The courtly substratum in all three narratives reveals itself, however, in the female protagonist’s skill in the erotic arts and in the eighty-four positions of lovemaking. At the same time, the political context of marginal courts between the fourteenth and sixteenth centuries significantly modifies these classical courtly conventions. In order to obtain or regain the beloved, Lorik, Saunrasi, and Ratansen embark not on a campaign of military conquest but on a personal quest through renunciation and initiation into Nath ascetic practices. It is in disguise as bearded, ash-smeared, penniless Nath ascetics that Ratansen and Saunrasi win and regain Padumavati and Chhitai respectively. Lorik, too, moves closer to his goal of union with Chanda only after a year’s austerities as a Nath ascetic. While I return to the use of Nath practices in these narratives below, it is pertinent to note here that at the moment when the protagonist obtains or regains his beloved, the Candāyan, Chitāī-carita, and Padmāvat depict a context of male hypergamy, with the hero marrying above his own social rank. (p.266) The marriages themselves, whether between a warrior and a princess as in the Candāyan or between a prince and a princess as in the Chitāī-carita and the Padmāvat, do not bring two lineages or clans and their resources together. Instead, they are configured as the triumph of a love transcending socio-political considerations and hierarchies. For the protagonist in these narratives, who Page 16 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* retains traces of a humbler, oral epic context, being a prince does not translate into soldiers and armies at his command. Equally, the fact of his being a warrior or a prince does not bring with it the kind of rank and resources that will enable him to obtain his beloved smoothly in an arranged alliance. Instead, in another trope revealing the assumptions of this hybrid warrior-courtly world, obtaining the beloved requires initiation as a Nath ascetic. Saunrasi in the Chitāī-carita may be the son of the (unnamed) king of Dhorasamudra, but in order to regain his beloved from Alauddin Khalji, he must be initiated as a Nath ascetic, and learn Nath skills that bend men to his will. Similarly, Ratansen might be the king of Chitor, but if he is to obtain the princess of Singhaldvip, he must renounce the world and become an ascetic, and launch his attack on her father’s fortress as an ash-smeared ascetic with his 16,000 ascetic followers. Similarly, in the Candāyan, if the triumphant warrior Lorik is to obtain his beloved Chanda, he must become a Nath ascetic and practice austerities at a Shiva temple for a whole year before Chanda comes to worship there with her companions. The familiarity of medieval and early modern Sufis with the hatha yoga techniques and cosmology of the Naths, indicated in texts such as ‘Abd alQuddus Gangohi’s Rushdnama in the fifteenth century (see Chapter 14 in this volume), has been widely noted. Similarly, the appropriations in the Sufi Candāyan, and especially the Padmāvat, of Nath physiology to represent the soul’s spiritual quest have also been commented on.47 (p.267) Instead of rehearsing that familiar ground, I am concerned here with other aspects of the presence of Nath yogis in these tales about ambitious warriors. For one, tales about Nath figures—especially Bhartrhari and Raja Gopi Chand—were performed in the same sorts of non-elite contexts in which the oral epics about heroic warriors circulated.48 Even more relevant is the fact that armed bands of Nath ascetics are known to have been active in northern India by the sixteenth century at least, as apparent from the Emperor Akbar’s encounter with them.49 When we consider this historical context, we begin to sense how the transformation of the questing lover into a Nath renunciant in the Candāyan, Chitāī-carita, and Padmāvat might have taken on additional resonance for the rural gentry and local warlord patrons of these narratives between the fourteenth and sixteenth centuries. In the narratives, warriors and princes with uncertain resources in terms of men and money win their objective—their beloved—by disciplining their bodies and minds through Nath initiation and practice. In the Padmāvat, such is the power of Ratansen’s Nath discipline that Padumavati is filled with desire even before she ever sets eyes on him [168–9]. The Chitāī-carita spells out the power that Saunrasi has acquired from his Nath guru Chandragiri—he has mastered a music that bends the minds of its listeners—not only men but also the beasts in the forest [798–802]. While the treatment of music in the Chitāī-carita deserves fuller investigation than is possible here, this power proves remarkably potent in helping Saunrasi achieve and even surpass his goals. Not only does he regain his Page 17 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* wife from the Sultan, her honour and status intact, he also wins the governorship of the entire province of Gujarat. His practice of Nath discipline has, thus, catapulted him into the role of trusted vassal of his new overlord. I would suggest that the Candāyan, Chitāī-carita, and Padmāvat point to the promise of Nath yogic discipline for rural gentry and local warlords with modest resources—a promise of physical and spiritual perfectibility that will (p.268) improve the chances of success, with the potential bonus of miraculous powers. In this sense, for all their familiarity with Nath esoteric doctrine, I would argue that the Candāyan, Chitāī-carita, and Padmāvat represent an outsider’s perspective on the Nath worldview. Viewed through this prism, Nath skills are a valuable asset in the pursuit of one’s political and military goals as a petty chief or warlord. Finally, I turn to the issue of the warrior/prince’s status in relation to his overlord. In the mahakavya context, marriage to the daughter of a powerful ruler marked the crowning acknowledgement of the prince’s status as a sovereign; in the world of the masnavi, union with the beloved marked the culmination of the protagonist’s quest. In the world of these vernacular courtly narratives between the fourteenth and sixteenth centuries, those characteristic classical tropes are modified in ways that, I would suggest, mirror the constraints and predicaments of their patrons. Thus, in the Candāyan, Chitāīcarita, and Padmāvat, marriage with a powerful ruler’s daughter guarantees neither political success nor military triumph. Indeed, while Lorik must defend his livestock and his lands from Chanda’s angry father for having eloped with her (at her urging) in the Candāyan, Saunrasi and Ratansen in the Chitāī-carita and Padmāvat must defend their wives from rivals and overlords. Thus, marriage with the beloved in these narratives does not consolidate imperial expansion and military victory as it does in the mahakavya world. The women in these narratives may still embody political resources, and obtaining them constitutes a point of arrival for the warrior/prince protagonists. However, in this vernacular world of marginal courts, marriage to such women also brings with it new vulnerabilities and risks. In this political world, it is not only overlords who threaten the wives in the warrior/prince’s household. In the Padmāvat, when Ratansen is captured by the Sultan and taken to Delhi, the neighbouring ruler Devapal sends an emissary to Padumavati, asking her to marry him. In the Chitāī-carita the threat to the princess comes from the overlord, Alauddin Khalji of Delhi. As I have outlined above, the period between the fourteenth and sixteenth centuries was marked by intense political rivalries and frequent conflicts between the sultans of Gujarat, Malwa, Delhi, and Jaunpur, as well as the emergent Rajput rulers of Rajasthan. Several scholars have noted the role of marriage as an important mechanism for negotiating and confirming military alliances during this period. Both the instance of Silhadi Purabiya’s career in the early (p.269) sixteenth century and the treatment of these issues in the Chitāī-carita that was composed Page 18 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* or recited at his court suggest that it was not merely the wives of warriors/ princes who were at risk, but women as a whole in the households of chiefs and potential vassals of any overlord. Thus, in a subplot in the Candāyan, the king Rupchand of Rajpur hears of Chanda’s beauty from a mendicant poet and attacks Govar in order to obtain her. His demand for Chanda is rejected, though the ruler of Govar does not object to paying other kinds of tribute to Rupchand. The warrior Lorik defends Govar and Chanda against Rupchand’s demands. In the Padmāvat too, Ratansen is quite willing to accept the Sultan Alauddin as his overlord; it is his wife that he will not surrender. As the Chitāī-carita suggests, some women could be gifted to the overlord, but at the vassal’s initiative rather than at the overlord’s demand. At the start of the Chitāī-carita, after Ramadev of Devagiri has accepted Alauddin as overlord and has been given the appropriate rank in the imperial court, his minister warns him against revealing to the Sultan that he has a daughter, since Alauddin would promptly demand that she be surrendered to him. In the Chitāī-carita this has nothing to do with Alauddin’s character, but is formulated in more general terms as one of the inevitable consequences of vassalage. In other words, the relationship between overlord and vassal is construed in terms of access to, and control over, women in the vassal’s household. The resonances with Silhadi’s own household are striking, where chroniclers almost a century later are scrupulous to recount that upon Silhadi’s defeat at Raisen in 1532, his queen Durgavati forced the Muslim women in the household to join in the mass immolation rather than surrender them to Bahadur Shah Gujarati. In the Chitāī-carita, Ramadev gifts female slaves to Alauddin but is outraged at the suggestion that he might be called upon to surrender his daughter. Such a possibility militates against his friendship with Alauddin, a friendship that both he and the Sultan invoke repeatedly. As he states early in the narrative, friendship, enmity, and marriage are all conducted only among equals: byāha bayara mitriyā pramāna, e tina chāhiyi āpa samāna [150]. For Saunrasi as well, the honourable recovery of his wife is the precondition for accepting Alauddin as his overlord. Narayandas resolves these conflicts in such a way that the honour of everyone involved—the warrior/prince, his wife, and the overlord—is preserved. It remains significant, however, that in all the three narratives—the (p.270) Candāyan, Chitāī-carita, and Padmāvat—the warrior/prince has no objection in principle to accepting an overlord. The question that these narratives pose is about the appropriate terms of this relationship from the warrior’s perspective, and the latter’s ability to negotiate those terms successfully. In the uncertain political climate between the fourteenth and mid-sixteenth centuries, success in these negotiations was by no means guaranteed. Thus, Lorik must fight the father of his beloved who is also his king—the latter has had his brother killed, his lands Page 19 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* and livestock confiscated, and has reduced Lorik’s mother to poverty. Ratansen must avenge Devapal’s insult to his wife and to himself at the same time as Alauddin’s imperial forces are on their way to lay siege to Chitor yet again, demanding submission and the surrender of Padumavati to the overlord. Far from the triumphant conquests and celebrations that mark many classical mahakavya narratives, in the vernacular world of marginal, hinterland ‘courts’, the success of the warrior/prince protagonist is uncertain. Love might triumph, and it does in the Candāyan, Chitāī-carita, and Padmāvat. But it is a love that might just have to transcend death, as in the Candāyan and Padmāvat—a trope ideal for Sufi appropriations. The resurgence of imperial authority under Akbar in the later decades of the sixteenth century saw a corresponding shift in the fortunes of both such patrons and of the vernacular narratives addressed to them. As is apparent from the fate of Silhadi Purabiya and his sons, the space of (relative) autonomy for such rural gentry and local warlords in the hinterland of a rapidly expanding Mughal empire had begun to shrink rapidly even by the time of Sher Shah Sur’s imperial campaigns in the 1540s. Akbar’s well-known and well-publicized preference for dealing with ‘kings’ from distinguished lineages has been tied to the proliferation, among the Rajput chiefs of Rajasthan, of genealogies that established precisely such antiquity and uninterrupted descent.50 Beyond Mughal intervention, trajectories of clan and lineage formation among the Rajputs equally contributed to the emergence of a socio-political hierarchy in which men like Silhadi Purabiya could no longer wield the (p.271) same kind of power as they had half a century earlier.51 Scholars have recently begun pointing towards norms of courtly behaviour, aesthetic refinement, and ‘manipulation[s] of bodily and gender identity’, as significant elements of Mughal imperial strategy under Akbar and his successors.52 In such a context, it can be shown that the kavya and masnavi narratives, which found patronage and sponsorship at the new, more self-consciously courtly contexts of northern India, reveal a resurgence of the social assumptions of those classical genres as well, including a return to the norms of kingship as defined by classical literary conventions.

Conclusion The study of courts and courtly narratives has been based, understandably, on those courts and rulers which were successful enough to leave behind a material record. Such rulers would have already attained a certain threshold in terms of their durable capacity to extract resources and to hand down that capacity to their heirs, and they signalled this capacity through the building of forts, palaces, and temples, and through an inscriptional record. In this chapter, I have focused instead on local elites, whether landed or military entrepreneurs, who maintained elaborate households and, as I have suggested here, aspired to membership within the political elite through their patronage of poets and performers, and through negotiating mutually favourable relationships with Page 20 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* their overlords. These were men who had moved up beyond the level (p.272) of village headmen, but had not amassed sufficient resources to leave behind either an inscriptional or an architectural record. The vernacular narratives from the north Indian hinterland between the fourteenth and mid-sixteenth centuries emerged from and were addressed to this stratum of patrons, men who have largely been missing in the historiography. In the absence of a durable imperial power in north India between the midfourteenth and mid-sixteenth centuries, such men could rise to wield significant power through their ability to offer local resources to regional chiefs and kings. I argue that in the Candāyan, Chitāī-carita, and Padmāvat we see the entrepreneurial politics of this period reflected and articulated in novel adaptations of received classical genres. For such patrons, narratives about great kings from exalted lineages had far less resonance than tales about ambitious warriors who disciplined their bodies and minds and were successful in their quest to marry the king’s daughter. The warriors in such narratives also aspired to negotiate successful relationships of vassalage with the political overlord of the day, through which they could retain control over the women in their households and also win suitable rank at the overlord’s court. This was clearly not the pattern for other, more lineage- and classically oriented realms and rulers in the same period. Nor did the avenues for such entrepreneurial politics continue to expand once Akbar consolidated imperial power once again in the second half of the sixteenth century. As the space for such local ‘big men’ shrank, the vernacular tales about warriors migrated back into a ‘folk’, oral epic realm, where they have survived into the present. Viewed from this perspective, the period between the mid-fourteenth and mid-sixteenth centuries marks a distinctive moment when such tales of warriors—whether told within a Sufi frame or not—found patronage at marginal courts in the north Indian hinterland. Notes:

(*) I am grateful to Indrani Chatterjee, Sumit Guha, Francesca Orsini, Kumkum Sangari, Jack Hawley, and Cynthia Talbot for feedback on earlier drafts. Rebecca French patiently read multiple drafts of this version and helped me to hone the argument. Sumit Guha, Francesca Orsini, Aditya Behl, and the Dean, College of Arts and Sciences, University at Buffalo, graciously provided support for travel to conferences at Rutgers University, Leiden University, SOAS, and the University of Pennsylvania, where I benefited from the comments of audiences. (1) As Dirk Kolff brilliantly showed in his Naukar, Rajput, Sepoy: The Ethnohistory of the Military Labour Market in Hindustan 1450–1850, Cambridge: Cambridge University Press, 1990. (2) ‘Knowledge of “Lorik dances” … is documented from Mithila c. 1325 in Jyotirisvar’s Varṇaratnākara …’; R.S. McGregor, ‘The Progress of Hindi, Part I: The Development of a Transregional Idiom’, in Literary Cultures in History: Page 21 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* Reconstructions from South Asia, Sheldon Pollock (ed.), Berkeley: University of California Press, 2003a, p. 915n3. For an inscription dating to 1458 at the oldest temple to Pabuji in Kolu, see Janet Kamphorst, In Praise of Death: History and Poetry in Medieval Marwar (South Asia), PhD dissertation, University of Leiden, 2008, p. 171. (3) Narratives in more canonical genres continued to be composed at more established courts throughout this period. See Chapter 8 in this volume for the Sanskrit texts produced at the Rajput courts of Gujarat. (4) See Hariharnivas Dvivedi’s discussion of the authorship of the Chitāī-carita, ‘Chitāī-carita ke racayitā’, in Chitāī-carita, H. Dvivedi, and Agarchand Nahta (eds), Gwalior: Vidyamandir Prakashan, 1960, pp. 12–26. (5) Prominent examples include Shyam Manohar Pandey, Madhyayugīn premākhyān, Allahabad: Lokbharati Prakashan, 1982; Pandey, Hindī aur fārsī sūfī kāvya, Allahabad: Sahitya Bhavan, 1989; Aditya Behl, Love’s Subtle Magic: An Indian Islamic Literary Tradition 1379–1545, Wendy Doniger (ed.), New York: Oxford University Press, 2012; and Thomas de Bruijn, ‘The Ruby Hidden in the Dust: A Study of the Poetics of Malik Muhammad Jayasi’s Padmāvat’, PhD dissertation, University of Leiden, 1996. (6) Kolff (1990), pp. 32–116. (7) Prominent oral epic heroes who had already become the foci of cults were incorporated into dynastic accounts throughout the early modern period. See, for example, the account of Pabuji in Muhata Nainasi’s mid-seventeenth-century Khyāt. Nainasi, Munhatā Nainasī rī khyāt, Badariprasad Sakariya (ed.), Jodhpur: Rajasthan Oriental Research Institute, 1993, vol. 3, pp. 58–79. See also John D. Smith’s discussion of the historical evidence available about Pabuji in The Epic of Pābūji: A Study, Transcription and Translation, Cambridge: Cambridge University Press, 1991, pp. 71–88. To my knowledge, however, such oral epic heroes did not become a focus on their own for full-fledged kavya or masnavi narratives at prominent royal courts. (8) See the discussion of norms of courtliness and masculinity elaborated during Jahangir’s reign in Muhammad Baqir Najm-i Sani’s Mau‘iẓah-i Jahāngiri (completed c. 1612), in Rosalind O’Hanlon, ‘Manliness and Imperial Service in Mughal North India’, Journal of the Economic and Social History of the Orient, 42(1), 1999: 47–93. For instances of the renewed salience of classical, Puranic models of kingship among Akbar’s Rajput associates, see Amrit Rai’s Mānacarita (composed c. 1585), and Narottama Kavi’s Mānacaritra rāso (composed between 1604 and 1614), both composed under Kachwaha Rajput patronage at Amber, in Gopal Narayan Bahura (ed. and annot.) Mānacaritāvali, Jaipur: Maharaja Sawai Mansingh Sangrahalaya, 1990.

Page 22 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* (9) Sheldon Pollock, Literary Cultures in History: Reconstructions from South Asia, Berkeley: University of California Press, 2003a; Pollock, The Language of the Gods in the World of Men: Sanskrit, Culture, and Power in Premodern India, Berkeley: University of California Press, 2006; Yigal Bronner and David Shulman, ‘A Cloud Turned Goose: Sanskrit in the Vernacular Millennium’, Indian Economic and Social History Review, 43(1), 2006: 1–30. Narratives overtly concerned with the past have been subjected to more socio-historical scrutiny, however. Notable recent works include Velcheru Narayana Rao, David Shulman, and Sanjay Subrahmanyam, Textures of Time: Writing History in South India 1600–1800, Delhi: Permanent Black and New York: The Other Press, 2003; Prachi Deshpande, Creative Pasts: Historical Memory and Identity in Western India, 1700–1960, Delhi: Permanent Black and New York: Columbia University Press, 2007. (10) Daud Ali, Courtly Culture and Political Life in Early Medieval India, Cambridge: Cambridge University Press, 2004. (11) See, for example, Allison Busch, ‘The Anxiety of Innovation: The Practice of Literary Science in the Hindi/Riti Tradition’, Comparative Studies of South Asia, Africa and the Middle East, 24(2), 2004: 45–59; Busch, ‘Literary Responses to the Mughal Imperium: The Historical Poems of Keśavdās’, South Asia Research, 25(1), 2005: 31–54. (12) Scholars have only recently begun to explore these more complex literary articulations of the political relationships between Mughals and Rajputs. See Cynthia Talbot, ‘Becoming Turk the Rajput Way: Conversion and Identity in an Indian Warrior Narrative’, Modern Asian Studies, 43(1), 2009: 211–43; and Allison Busch, ‘Hidden in Plain View: Brajbhasha Poets at the Mughal Court’, Modern Asian Studies, 44(2), 2010: 267–309. (13) Cited in Mian Muhammad Saeed, The Sharqi Sultanate of Jaunpur: A Political and Cultural History, Karachi: University of Karachi, 1972, pp. 12–13, 27–30. (14) Simon Digby, ‘Before Timur Came: Provincialization of the Delhi Sultanate through the Fourteenth Century’, Journal of the Economic and Social History of the Orient, 47(3), 2004: 345. (15) Yusuf Gada’s Tuḥfat al-Naṣā’iḥ, cited in Digby (2004): 306–7. (16) S.A.A. Rizvi, A History of Sufism, New Delhi: Munshiram Manoharlal, 1978, vol. 1, p. 268. (17) Rizvi (1978), vol. 2, p. 287. (18) Rizvi (1978), vol. 1, p. 273. Page 23 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* (19) The madrasa was built by Malik al-Sharq Badarat Khalji, when Khan-i-A‘zam Rustam Khan was the muqti‘ of Sarangpur. Epigraphia Indica, Arabic and Persian Supplement, 1964 and 1965, p. 72. (20) Rizvi (1978), vol. 2, p. 434. (21) Gerard Fussman and K.L. Sharma, Chanderi: Naissance et declin d’une qasba, Paris: Publications de l’Institut de Civilization Indienne, College de France, 1999–2003, vol. 2, p. 301. (22) Epigraphia Indica, Arabic and Persian Supplement, 1964 and 1965, New Delhi: Archaeological Survey of India, rpt. 1987, pp. 52–76. (23) Digby (2004): 313. (24) H. Beveridge accepts James Tod’s account in Annals and Antiquities of Rajasthan (1829–32). Beveridge, ‘Some Account of Silhadī, Otherwise Ṣalāhu-dDīn (Saladin), A Rājpūt Renegade, and of the History of Gujarat entitled the “Mirāt Sikandarī”’, The Asiatic Review, 7 (1915): 383. Tod’s account is the earliest that I have been able to find of Silhadi being a Tomar Rajput, that is, a member of the clan that ruled Gwalior in the fifteenth and early sixteenth centuries. Contemporary sixteenth-century sources are silent on the issue of Silhadi’s clan affiliation. (25) Sikandar, Mir’āt-i Sikandarī or The Mirror of Sikandar, by Sikandar, the son of Muhammad, alias Manjhu, Gujarati, Fazlullah Lutfullah Faridi (trans.), Gurgaon: Vintage Books, 1990 (repr.), p. 172. (26) Sikandar (1990), p. 170. (27) Sikandar (1990), p. 175. (28) Hariharnivas Dvivedi, Tomārõ kā itihās (A History of the Tomars), Gwalior: Vidyamandir Prakashan, 1976, vol. 2, p. 229. (29) Kolff (1990), pp. 102–3. (30) Kolff (1990), p. 110; ‘Abbas Khan Sarwani, Tarikh-i Shershahi, Brahmadeva Prasad Ambashthya (trans.), Patna: K.P. Jayaswal Research Institute, 1974, p. 609. (31) For the construction projects and inscriptions of Rana Kumbha in fifteenthcentury Mewar, see Upendra Nath Day, Mewar under Maharana Kumbha, 1433– 1468 A.D., New Delhi: Rajesh Publications, 1978, pp. 91–127, 155–84; and Ram Vallabh Somani, Maharana Kumbha and his Times, Jaipur: Jaipur Publishing House, 1995, pp. 182–219, 243–73. For the architectural patronage of Man Singh Kachwaha in the late sixteenth and early seventeenth centuries, see Page 24 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* Catherine Asher, ‘Mughal Sub-imperial Patronage: The Architecture of Raja Man Singh’, in The Powers of Art: Patronage in Indian Culture, Barbara Stoler Miller (ed.), New Delhi: Oxford University Press, 1992, pp. 183–201. (32) See footnote 6 above; also John Millis, ‘Malik Muhammad Jayasi: Allegory and Symbolism in his “Padmāvat”’, PhD dissertation, University of Chicago, 1984; and Shantanu Phukan, ‘Through a Persian Prism: Hindi and Padmāvat in the Mughal Imagination’, PhD dissertation, University of Chicago, 2000. (33) Mataprasad Gupta (ed.), Candāyan: Dāūd-viracit pratham hindī sūfī premkāvya, Varanasi: Vishwavidyalaya Prakashan (1967) 1996. All citations are from this critical edition. Stanza numbers are indicated in square brackets. (34) Shyam Manohar Pandey, The Hindi Oral Epic Lorikayan, Allahabad: Sahitya Bhawan, 1987. For a list of differences between Da’ud’s Sufi allegory and modern, oral epic versions, see pp. 52–74. For another version, see Pandey, The Hindi Oral Epic Canaini, Allahabad: Sahitya Bhawan, 1982. (35) Hariharnivas Dvivedi and Agarchand Nahta (eds), Chitāī-carita, Gwalior: Vidyamandir Prakashan, 1960. This is a more complete edition than the one edited by Mataprasad Gupta, Kashi: Nagari Pracharini Sabha, 1958. All citations are from Dvivedi edition; square brackets indicate stanza numbers. (36) Mataprasad Gupta (ed.), Jāyasī granthāvalī, Allahabad: Hindustani Akademi, 1952. All citations are from this edition. Stanza numbers are indicated in square brackets. (37) For Bada’uni’s description of the Candāyan as a Hindi masnavi, see Muntakhab ut-Tawarikh by Abdul Qadir Badauni (1898), Delhi: Atlantic Publishers and Distributors, 1990, volume 1, p. 333. For readers in the seventeenth and eighteenth centuries interpreting the Padmāvat as a masnavi, see Behl (2012) and Phukan (2000). (38) Verse romances with epic, heroic, or romantic content that took their name from the rhyming distich meter that they typically used. Annemarie Schimmel, Islam in the Indian Subcontinent, Leiden: E.J. Brill, 1980, p. 269. (39) Julie Scott Meisami, Medieval Persian Court Poetry, Princeton: Princeton University Press, 1987, pp. 81, 98–9. (40) See Aditya Behl’s discussion of the distinctively Sufi adaptations of Sanskritic rasa aesthetics in the Avadhi Sufi romances in Behl (2012). (41) Dandin in the seventh-century Kāvyadarśa, translated and cited in Indira Viswanathan Peterson, Design and Rhetoric in a Sanskrit Court Epic: the Kirātārjunīya of Bhāravi, Albany: State University of New York Press, 2003, p. 8.

Page 25 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* (42) Rudrata, translated and cited in David Smith, Ratnākara’s Haravijaya: An Introduction to the Sanskrit Court Epic, Delhi: Oxford University Press, 1985, p. 29. (43) While the vocabulary used by Da’ud and Jayasi in these descriptions is entirely Indic, it must be noted that there is a comparable topos of praise of the city in the Persian tradition as well. See Sunil Sharma, ‘The City of Beauties in Indo-Persian Poetic Landscape’, Comparative Studies of South Asia, Africa, and the Middle East, 24(2), 2004: 73–81. (44) Daud Ali, ‘Violence, Courtly Manners and Lineage Formation in Early Medieval India’, Social Scientist, 35 (9/10), 2007: 9, 11. (45) Stuart H. Blackburn, Peter J. Claus, Joyce B. Flueckiger, and Susan S. Wadley (eds), Oral Epics in India, Berkeley: University of California Press, 1989, ‘Introduction’, p. 5. (46) Vishwambhar Sharan Pathak, Ancient Historians of India: A Study in Historical Biographies, New York: Asia Publishing House, 1966, p. 27. (47) For Sufi familiarity with Naths, their texts and their techniques, see Rizvi (1978), vol. 1, pp. 331–49; Simon Digby, Wonder-Tales of South Asia: Translated from Hindi, Urdu, Nepali and Persian, Jersey: Orient Monographs, 2000, pp. 221–33; and Carl W. Ernst, ‘The Islamization of Yoga in the “Amrtakunda” Translations’, Journal of the Royal Asiatic Society, 13(2), 2003: 199–226. For Jayasi’s appropriations of Nath symbolism, see Rizvi (1978) vol. 1, pp. 368–9. David Gordon White, The Alchemical Body: Siddha Traditions in Medieval India, Chicago: University of Chicago Press, 1996, pp. 260–2; and de Bruijn (1996), pp. 91–3. (48) For an ethnographic account from the late twentieth century of the social world of the Naths and their legends, see Ann Grodzins Gold, A Carnival of Parting: The Tales of King Bharthari and King Gopi Chand as Sung and Told by Madhu Natisar Nath of Ghatiyali, Rajasthan, Berkeley: University of California Press, 1992. (49) William R. Pinch, Warrior Ascetics and Indian Empires, Cambridge: Cambridge University Press, 2006, pp. 28–32. (50) Norman P. Ziegler, ‘Marvari Historical Chronicles: Sources for the Social and Cultural History of Rajasthan’, Indian Economic and Social History Review, 13(2), 1976: 233–4. (51) For a lineage-oriented refashioning of Rajput legends by the later seventeenth century, see Cynthia Talbot, ‘The Mewar Court’s Construction of History’, in Kingdom of the Sun: Indian Court and Village Art from the Princely

Page 26 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Warrior-tales at Hinterland Courts in North India, c. 1370–1550* State of Mewar, Joanna Williams (ed.), San Francisco: Asian Art Museum, 2007, pp. 13–33. (52) For norms of comportment at the Mughal imperial court, see J.F. Richards, ‘The Formulation of Imperial Authority under Akbar and Jahangir’, in Kingship and Authority in South Asia, J.F. Richards (ed.), Delhi: Oxford University Press, 1998 (1978), pp. 285–326; for a new courtly aesthetic of literary refinement, see Busch (2004); for Mughal norms of masculinity, see Rosalind O’Hanlon, ‘Kingdom, Household and Body: History, Gender and Imperial Service under Akbar’, Modern Asian Studies, 41(5), 2007: 889–923; the quotation here is from p. 922.

Access brought to you by:

Page 27 of 27

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Emotion and Meaning in Mirigāvatī Strategies of Spiritual Signification in Hindavi Sufi Romances* Aditya Behl

DOI:10.1093/acprof:oso/9780199450664.003.0010

Abstract and Keywords The Hindavi Sufi text Mirigāvatī was translated by the author of this chapter, and this chapter contains an analysis of its literary and cultural strands. The chapter shows how the Sufi author, Qutban Suhravardi, used rasa poetics to elaborate his vocabulary of sublimation in this romance written in provincial Jaunpur in 1503. Through an analysis of Qutban’s poetical strategies, the chapter shows how this Hindavi romance offers a richly layered spiritual language built into an evocative love and adventure story. It suggests that the Mirigāvatī, like others of the prema-kahani (love-story) genre created by Indo-Muslim poets from the late fourteenth century, could have worked as ‘blueprints for spiritual education’ in the sultanates, thus integrating even distant provinces into a larger Islamic conversation. Keywords:   Sultanate, Jaunpur, Sufi, Hindi, Hindavi, narrative, music, romance

Introduction: Sufism in Hindavi Texts MIRIGĀVATĪ OR ‘THE MAGIC DOE’ is the work of Shaykh Qutban Suhrawardi, the Indian Sufi master who was also an expert poet and storyteller attached to the glittering court-in-exile of Sultan Husayn Shah Sharqi of Jaunpur. Composed in 1503 as an introduction to mystical practice for disciples, this powerful Hindavi or early Hindi Sufi romance is a richly layered and sophisticated text, simultaneously a spiritual enigma and (p.274) an exciting love-story full of adventures. The Mirigāvatī is a story that draws freely on the large pool of Indian, Islamic, and European narrative motifs in its distinctive telling of a Page 1 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī mystical quest and its resolution. Adventures from the Odyssey and the voyages of Sindbad the Sailor—sea voyages, encounters with monstrous serpents, damsels in distress, flying demons and cannibals in caves, among others— surface in Qutban’s rollicking tale, marking it as first-rate entertainment for its time and, in private sessions in Sufi shrines, a narrative that shaped the interior journey for novices. A study of the Mirigāvatī reveals the precise mechanism and workings of spiritual signification in a major tradition of world and Indian literature, the Hindavi romances composed by Chishti1 Sufis in the eastern part of the Ganga-Jamuna doab. Let me begin with a discussion of the history of Sufism in India during this period, and let me introduce that discussion with a personal memory, of my first exposure to Sufism. Just after my freshman year in college, I went to Ajmer Sharif to see the shrine, motivated mainly by curiosity about the civilization that had produced the elegant Urdu poetry with which I had become fascinated. It was a rich visit, including a meeting with the senior Sufi master at the shrine and an invitation to a private or band sama‘ mehfil at the Chilla Usmani, a hilltop shrine built around a cave where legend had it that the Pir of Shaykh Mu‘in alDin Chishti had engaged in a spiritual retreat for forty days. On Thursday night there was a public majlis in the courtyard of the main shrine. The qawwals were seated directly facing the door to the sanctuary of the tomb, and we, the audience, formed a hollow square by seating ourselves on either side of them. We waited for the formal head of the order, the sajjada nashin, who would direct the performance in the capacity of mir-i majlis. There was a stir at the arched gateway to the compound, and a small procession made its way into the courtyard against the dimly lit backdrop of the grand high mosques built by successive Mughal (p.275) emperors after Akbar’s legendary pilgrimage on foot to this ancient site of grace. A torchbearer ran ahead of the sajjada nashin with a flaming torch, lighting his way, and another one followed in his retinue. The detail that struck me forcibly then, and has stayed in my memory ever since, is that the spiritual head of the premier Sufi shrine in India was dressed from head to foot in saffron robes, with a saffron-coloured turban on his head. How odd, I thought; I wonder what he’s doing in the clothes of a sanyasi or yogi? It seemed a convention from the oldest layer of Sufi activity on the subcontinent, indicating some deep connection with indigenous forms of renunciation, but how could that be? I did not know it then, but in a sense my intellectual career has been devoted to trying to explain this odd fact—why should the Sufi dress as a yogi? Years later, as if by coincidence, I was to come across this yogic disguise again, in the texts of the Hindavi Sufi romances, also a product of the early encounter of the Sufis with the Indian religious landscape. The Mirigāvatī of Shaykh Qutban Suhrawardi is a well-crafted instance of a powerful Indo-Islamic literary tradition that circulated in courts, bazaars, shrines, and private salons throughout the Sultanate period and in subsequent centuries. The genre, called prema-kahani or ‘love-story’ in Hindavi, was created Page 2 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī by Muslim poets in the late fourteenth century, a product of cultural and religious encounter in the garrison towns of the eastern provinces of the Delhi Sultanate. Written and performed from then until the nineteenth century, the Hindavi Sufi romances mark the inauguration of a major new literary and devotional culture in a local language. The poets were members of the Persianspeaking courtly elite of the Delhi Sultanate and of the regional sultanates that followed in its wake. In 1896, Sir George Abraham Grierson, with Pandit Sudhakar Dvivedi, began to publish the text of Malik Muhammad Jayasi’s Padmāvat (c. 1540), a romance of the genre. In the decades that followed, five major romances were made available to the Hindi-speaking world: Mawlana Da’ud’s Cāndāyan (1379), written in the heyday of the centralized Delhi Sultanate; several works by Malik Muhammad Jayasi and Shaykh Manjhan Shattari; and Qutban Suhrawardi’s Mirigāvatī (1503). To begin with the historical placement of the Sultanate of Jaunpur: the invasion of Timur in 1398 was a fatal blow to the centralized Tughluq state that had evolved up until the end of the fourteenth century. After the eclipse of Tughluq power in the fifteenth century, the Sayyid (p.276) and Lodi sultans of Delhi could not assume even a titular sovereignty over the rulers, landholders, and iqta‘-dars of the different regions of northern, central, and eastern India. Preeminent among the sultanates of the east was the Sharqi kingdom of Jaunpur, founded in the 1390s by Malik Sarwar, a eunuch or khwajah-sara in the service of Sultan Firuz Shah Tughluq. Appointed custodian of the town of Jaunpur, he played an important part in the succession disputes following Firuz Shah’s death in 1388 and eventually consolidated his own position as ruler of Jaunpur. His successors through an adoptive son, Malik Mubarak, strengthened the realm and ruled the region as independent sultans till the accession in 1458 of Husayn Shah Sharqi, who involved himself in a protracted and ill-fated struggle to conquer Delhi from the Lodi sultans Bahlul (r. 1451–89) and Sikandar (r. 1489– 1517).2 Husayn Shah Sharqi was dethroned by Bahlul Lodi’s capture of Jaunpur in 1483, and fled to a small enclave around the town of Chunar in Bihar. Despite repeated campaigns against Sikandar Lodi, Bahlul’s successor, Husayn Shah, could not dislodge the Lodi forces from Jaunpur. He was later given Colgong or Kahalganv in Bhagalpur district in Bihar by the Bengal sultan ‘Ala’ al-Din Husayn Shah, and he had coins bearing his name issued until his death in 1505.3 Husayn Shah was a cultured ruler, and although he had suffered a political reverse that he failed to overcome for the rest of his life, his court-in-exile was part of the constellation of regional cultures that flourished in the various Indian sultanates that followed in the wake of the more centralized Delhi Sultanate. The coming into prominence of regional sultanates such as Malwa, Gujarat, Jaunpur, Bengal, and the Deccani states led to the development of the new courtly artistic, literary, and performative cultures that this book seeks to delineate. These new kingdoms (p.277) invented, as part of their project of stateformation and legitimation, regional artistic and literary styles out of the Page 3 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī Islamicate culture of the Delhi Sultanate and local languages and aesthetic media. Husayn Shah Sharqi himself was a poet and a noted patron of the distinctive Sharqi style of architecture. He was also an accomplished musician, credited with the creation of Raga Jaunpuri, the various Syams, and four different versions of the morning Raga Todi in north Indian classical music.4 Qutban, a poet attached to Husayn Shah’s court in exile, dedicated the Mirigāvatī to him in 1503, and in order to please this cultivated patron, he included a miniature rāga-mālā as part of the description of a courtly assembly in his romance. Little else is known about Qutban except that he was a disciple of Shaykh Buddhan Suhrawardi, whom he mentions as his spiritual preceptor in the prologue to the Mirigāvatī. The identity of Shaykh Buddhan, whose name means simply ‘the eldest one’, is a matter of some controversy, for there are several individuals with this rather generic name. S.A.A. Rizvi notes that Shaykh Buddhan was ‘the disciple of Shaikh Muhammad Isa Taj of Jaunpur. Although Shaikh Isa Taj was a distinguished Chishti, Shaikh Buddhan seems to have been initiated into both the Chishtiyya and Suhrawardiyya orders’.5 S.M. Pandey has suggested that Qutban was affiliated to a Chishti Shaykh Buddhan, a ‘great musician, [who] lived during the time of Husayn Shah Sharqi in Barnawa in Meerut district near Delhi’.6 Barnawa’s proximity to Delhi places the town in the direct path of the invading armies of the Lodis and the Sharqis, far too dangerous a place to set up a hospice. Additionally, Shaykh Buddhan Chishti (p.278) (d. 1497) is known to have had musical contests with his patron Husayn Shah, who lived in Bihar rather than next to the Lodi sultans in Delhi. Shaykh Buddhan’s solely Chishti affiliation, moreover, makes Qutban’s discipleship with him an unlikely possibility.7 There is one other possibility, a Shaykh Shams al-Haqq, known as Buddhan, the grandson of Shaykh Sadr al-Din ‘Chiragh-i Hind’, one of the major figures in the Suhrawardiyya silsila (genealogical chain of traditional masters) in Jaunpur. This Shaykh Buddhan was close to Sultan Ibrahim Shah Sharqi (d. 1440), who is reputed to have had a spiritual retreat built for him, the site of an annual fair (now close to the Central Post Office in the city of Jaunpur).8 It is impossible to decide definitively amongst these ascriptions, but we will return to the specific terms in which Qutban praises his pir in the prologue, for they have a bearing on our reading of the romance, Mirigāvatī. The Sultanate period was marked by very complex historical processes: the constitution of a range of courtly polities and devotional communities, and the formation of the new Indo-Aryan languages, the linguistic and literary ancestors of all the modern north Indian languages. While the Sufi mystical agenda can be understood in terms common all over the Islamic world, its regional articulations reinvent Islam in locally comprehensible vocabulary. The implications of Hindavi Sufi poetry for the history of devotional movements (bhakti) in India are (p.279) far-reaching; for example, for the Sufi poets demonstrate a transfer from Sanskritic aesthetics to devotionalism among Islamic groups predating the Page 4 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī sixteenth-century ‘Hindu rebirth’ of devotionalism in northern India. Rather than any simple syncretism or separatism, the Mirigāvatī is the product of a dialogue amongst the major indigenous devotional and literary movements of its time. Let me conclude this section by returning to the example with which I began, of the jogiya bhesa or yogic disguise of the seeker. Why was it necessary to the Sufi pirs of the Sultanate period to construct this elaborate masquerade? Why could they simply not present a Sufi seeker or salik as the hero of their narratives of the sequential revelation of the absolute? The reasons are to be found in the complex situation of cultural encounter of which the Sufis are part, and are both external and internal to the silsilas. Externally, the yogic disguise marks the competitive appropriation and assimilation of yogic techniques and even sites that are corroborated by numerous episodes in the Persian sources, in which the Sufi tames a yogi through a contest of miraculous powers, often resulting in the yogi’s conversion to Islam and acceptance of the authority of the Shaykh. Further, the episodes explicitly mention territorial take-overs of particular sites from these vanquished yogis. This indicates not just a land-grab, but also an assimilation of techniques, imagery, and symbolism from the Nath panth. Sufi silsilas were deeply and intimately involved with political matters since their inception, and their spread to India followed similar patterns of the integration of spirituality into the world.9 Internally, aside from the competition between silsilas for disciples, political patronage, and areas of spiritual authority (vilayat), there was also the relationship of the Sufis with other sources of legitimate power in their society, which in the case of Islam included both scriptural and political powers. With both these groups, assuming the Indic disguise allowed the Sufis to mount an internal critique of Islam through cultural role-playing and the elaborate masking of their agenda in Indic terms. Thus, in the second half of the (p.280) Padmāvat, the Sufis are the Rajputs, the natives of the land defending it against the wicked and violent Turks (the same group to which the author, Jayasi, himself belongs) led by Sultan ‘Ala’ al-Din Khalji. In a sense, the Sufi love-stories are a call to reckoning to those who carried out that conquest, an invitation to deal with psychological consequences of war, violence, and the baser impulses to which the human race is prone. The Sufi poets’ Hindavi quests can best be understood as interior voyages with the goal of spiritual transformation, and the narrative grammar of this quest demonstrates the reinvention of an Islamic genre in an Indian landscape. At the beginning of the romance, the Prince (Rajkunvar) is out hunting in a forest with a company of nobles. He sees in the distance a glimmering seven-coloured doe, and follows her. Qutban’s poem takes its title from this narrative trigger for the Prince’s interior journey: the doe, called Mirigavati (lit. ‘doe-woman’), a magical creature whose presence signifies divinity in an as yet mysterious form. She has the powers of shape shifting, flying, making herself invisible or large or small. The elusive doe awakens in the Prince a desire that overpowers his mind and Page 5 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī heart and does not allow him any peace. This narrative motif echoes Marica in the Rāmāyaṇa, the golden hind that lured Rama away while Ravana abducted Sita, and this Hindu motif shows how entwined the Sufi poets were with the nonSufi local traditions, both religious and literary. The vision of the magic doe triggers the plot’s impulse towards consummation through a series of episodes that delay the satisfaction of desire. The moment of desire is the initial arousal in the series of deferrals that are implied by the Sufi idea of ordinate love, in which each object of desire is loved for the sake of one higher to it, all the way up to God.

Critical Approaches to the Text The Hindavi Sufi romances belong to a cultural world remote from us in its moorings, its politics, its aesthetic canons, and spiritual practices. Several assumptions are common to the modern criticism on the genre: that these texts constitute a single coherent genre modelled on Persian masnavis or verse romances, that they express a syncretic mingling of elements between Indian and Islamic poetic and religious traditions, and that they are somehow ‘allegorical’ (sanketātmak) in equating human and divine love. With regard to the texts themselves, we find that scholars have concentrated on classifying the genre within the premodern history of the canons of modern standard Hindi and finding some (p.281) sort of schema to explain its spiritual meaning. Critics have adopted a number of schemata to explicate the surviving Hindavi Sufi romances. These strategies include interpreting the texts as examples of vernacular populism,10 of love for a formless divinity (nirguna prema-marga) and simultaneously of ‘un-Indian’ sensuality,11 of premarital love,12 of the indigenization of Persian romances,13 of the adaptation of the Alexanderromance into Indian literature,14 of allegory understood as a point-to-point correspondence between two levels of meaning,15 of numerical symmetry,16 of ‘image-ism’ (pratibimba-vada),17 of religious syncretism18 and its opposite,19 and of the deployment of multiple metaphoric systems towards a moral and didactic purpose.20 (p.282) But this way of framing the question of spiritual signification is itself part of the problem, for it implies a flat, point-to-point correspondence between two levels of meaning, the narrative and the spiritual. Once mapped, the text is explained away, the meaning decoded, and we have yet another object of Indological study analysed into its component parts, in T.S. Eliot’s phrase, ‘like a patient etherised upon a table’. Compounding the problem is the supposedly ‘formulaic’ nature of the narratives, which commonly use the formula of a hero awakened by a dream or vision or verbal description of a beautiful woman (in the case of Mirigāvatī, by the magic doe-woman), impelled to don a yogic disguise to go on a quest, often abandoning his wife (who represents the world), attaining the object of his quest and coming back to make peace between his two wives, who represent the demands of the spiritual and the material worlds respectively. Thus, it is commonly understood that their poems somehow equate Page 6 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī human and divine love, but the precise mechanism for spiritual signification has remained elusive, as also its links to the surrounding mystical culture. No key or allegorical scheme quite fits the lock, and all attempts at mapping point-to-point correspondences between levels of meaning in the narratives or between text and context have failed. With such a polyphony of critical voices, how is one to make sense of the Hindavi Sufi romances? We need to pay attention to the transfer of generic forms and deployment of aesthetic concepts, to the hierarchies of response that attend the reception of these romances in courtly and shrine settings, and to the richly symbolic imagery crafted by these poets. Given this tangle of issues, historical and conceptual, how can one understand how the texts might have functioned within their contexts of production and reception? The Hindavi Sufi romances were sung aloud at evening parties or courtly assemblies, but most ideally in private sessions in Sufi shrines. Here the novice was exposed to poetry and music in order to direct him towards realizing divinity through his own inner journey. The romances were enmeshed in complex protocols of reading aloud, singing, and recitation in different contexts. Outside the private instruction given within a Sufi shrine, these works were a major performance tradition in royal and aristocratic courts and storytelling salons hosted by connoisseurs. The notion of flexibility of meaning deployed by the Sufi poets (more on which below) allowed these works to circulate widely, and contemporary and later commentators often (p.283) note the power and lush appeal of the poetry. The relation of these texts to this performance context requires an understanding of the coding and narrative structure of these romances, in order to demonstrate the precise mechanism of spiritual signification. The absence of a directly stated principle of allegory is due to the presence of a teaching shaykh in the context of the closed shrine audience, who would comment upon the story and guide the initiates along the path if and when they were ready. The coincidence of plot and journey, inner and outer, is of course a widespread phenomenon in world literature. There is a complex relationship between the infinite capacity of the storyteller to spin out a tale in episodes, its episodicity, and his ability to curtail this limitless expansion to carry out a distinct literary and spiritual agenda. Critics of the romance genre have remarked on how the genre does and undoes itself, that it is ‘a form that simultaneously quests for and postpones a particular end, objective, or object’.21 Working from this insight, Barbara Fuchs has developed a view of romance as strategy, and I find this a productive model for thinking about how the axiomatically archetypal, the formulaic, can also be historical. Fuchs uses the notion to examine fruitfully the dialectical interplay between history and literary form, showing how the genre encapsulates different sorts of strategies for the articulation and deferment of desire, and its ultimate overdetermined satisfaction. It is a commonplace of romance criticism, following Northrop Frye, that romances are archetypal and that these formulae are about idealization and wish-fulfilment, but even Page 7 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī formulaic narratives are constructed, by historical agents for particular purposes that carry the markings of class, gender, and ideological agenda. Among the plural and diverse literary traditions of the Indian subcontinent, the romance stands at the juncture of several overlapping repertoires of generic forms, skilfully sketched out recently by Francesca Orsini:22 (a) the Sanskritic repertoire, focused on the notion of kama (desire, erotic love, pleasure) and exemplified in amatory manuals, plays, and erotic lyrics; (b) the oral repertoire, in the forms of songs, tales, and (p.284) epics told and retold, containing basic psychological and social tensions and providing much material for literary and ‘high-culture’ renditions; (c) the Perso-Arabic repertoire, embodied in the adaptation of notions like muhabbat (love, affection) and ‘ishq (excessive passion) bordering on madness (junun) into the South Asian cultural lexicon, culminating perhaps most popularly in the Urdu ghazal; and (d) devotional literature, in the form of bhakti songs of yearning, religiosity, and protest, and the hagiographic family trees that were developed as an indigenous historiography to account for the extraordinary spread of bhakti from region to region like, in the memorable phrase of A.K. Ramanujan, a ‘lit fuse’.23 To this extremely useful inventory I would juxtapose another, overlapping categorical imperative, of the different ‘matters’ of story that go into the making of narratives. In the preliminary categorization of Suniti Kumar Chatterji, and drawing on the ‘matters’ of the medieval European world, these are threefold: the matters respectively of Ancient India, of the particular area or province, and of the Islamic world.24 We may disaggregate these larger ‘matters’ into smaller, more useful ones, tale-cycles or ‘narrative clusters’ such as the ‘matter’ of yoga and the Gorakhnath panth, or the tale of Lorik, Chanda, and Maina. Poets and storytellers took up all or part of each of these ‘matters’ as their subject, and used any or all of the concepts and vocabulary of the four repertoires (the Sanskritic, the oral, the Perso-Arabic, and/or the devotional) to elaborate on their theme, in a variety of genres. The matter of Lorik and Chanda, or Ratansen and Padmavati, was told in multiple regional languages with varying emphases, valences, and episodes, but was obviously dependent on the same core narrative cluster. The tellings and retellings of larger stories, and the permutations and combinations of narrative motifs, were potentially infinite, but each telling happened in time and space, in particular performance contexts, and it is these (p.285) specific circumstances that allow us to make sense of what stories might have meant to the people who told and listened to them.

Meaning and Narrative in the Mirigāvatī By examining particular forms and landscapes and then embedding them within the poets’ articulations of aesthetics and theology, we can begin to understand how these romances might have worked as blueprints for a spiritual education in the north Indian provincial sultanates. The first poets of these romances, the Chishti Sufis, were powerful figures in the cultural and religious life of the Delhi Page 8 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī Sultanate. At the core of their mystical activity and contemplation was the transcendental monotheism of belief in Allah, the invisible centre, cause, and creator of the visible world. They tried to resolve the intractable dilemma posed by a transcendental godhead within a material world. Since this strictly monotheistic, invisible God created the physical, sensible world and yet was apart from the world, how could a seeker have access to divine presence? A careful reading of the text of the Mirigāvatī demonstrates how built forms and landscape descriptions were used to shape the hearer’s consciousness in order to give narrative shape to the process of interior voyaging. As already hinted at, the simultaneous revealing and concealing of the evidence is directly related to the presence of a teaching shaykh who could explicate the secret clues in these romances to guide the spiritual quests of his disciples. After the doe has lured the Prince to a magic lake deep in the forest, it vanishes in its waters. Although the Prince jumps in, she is gone and he is left lamenting. He will not return to court, and sits by the lake under a tree that glitters like a royal canopy, meditating on the vision he has seen, weeping inconsolably like a spring cloud. When his companions return to court and inform the Prince’s father, the entire town comes out to the forest to reason with the Prince, but he will not return. He asks the King to build him a seven-levelled red-and-gold palace around the shining lake. Craftsmen, painters, architects, and goldsmiths arrive to construct the fantastic gilded palace (Stanza 36). This motif serves as a perfect iconic emblem for the poetics of sublimation that Qutban weaves into his romance. The seven-levelled picture-pavilion—the number is significant—serves to excite and assuage his desire by turns. The building is topped with a chaukhandi, a tomb ornament consisting of (p.286) a miniature dome resting on four small arches designed to shade the tomb.25 The craftsmen decorate the palace with painted scenes from the epics and romances (and Qutban refers to a wide range of Indic story literature), as well as pictures of the magic doe that had so afflicted their Prince. The sight of this object of desire would make him weep, then collect himself, for she was also his life’s support. He inhabits this palace in this state of unconsummated desire, longing to see the magic doe again. This significant building encodes several levels of textual and narrative reference. The iconic description of this building, its structure, ornament, and form serves to hide in plain sight the structural principle of allegory and to encode a summary or icon of the Prince’s subsequent quest and the Sufi poetics of sublimation. Like the four-sided palace, the story has four major sections, and the first side of the building, the initial quadrant of the text, treats the meeting of the subject and object of desire and the need for the quest. The seven-levelled palace by the lake finds an echo in the seven ordeals that the Prince, now turned into a yogi, has to overcome in his quest for Mirigavati after they are separated later in the tale. These involve: (a) seeking the guidance of a Page 9 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī wandering ascetic, who sends him off on an ocean voyage; (b) surviving a monthlong battle with the waves; (c) landing on a shore preyed on by a man-eating sea-serpent, and surviving, like Sindbad the Sailor; (d) rescuing a Princess, Rupmini, from a seven-headed demon; (e) marrying the Princess under duress, but living chastely within marriage; (f) being guest to a cannibal herdsman, whom he blinds in an episode curiously reminiscent of Odysseus and the Cyclops; and (g) passing a night in deadly danger in a palace full of the illusions of sensuality. These ordeals are designed to introduce the novice to a path of Sufi practice, somewhat like the seven valleys of ‘Attar’s Conference of the Birds, for they begin with initiation and then illustrate the mastering of a baser impulse through various Sufi techniques and under the guidance of a Sufi master. The seven steps suggest both the steps of the palace and the ordeals on the path towards spiritual perfection, each of which involves the conquest of a base emotion. Each is designed to test the Prince in some particular Sufi virtue such as chastity, trust in God, or the power of rigorous meditation and zikr, remembrance of Allah, in accordance (p.287) with the strong Suhrawardi emphasis on different sorts of zikr and also submission to Allah’s will after human effort has proved useless. The poet thus links the quest of the romance, the Prince’s search for the magic doe, with the Sufi path of true love and ascetic purification. When the Prince has successfully completed these by using prayer, fasting, abstinence, and other Sufi techniques, he can move on to the next part of his quest. Finally, the Prince reaches and enters Kanchanpur, the City of Gold, singing of his pain in separation, viraha, and accompanying himself on his stringed kingari instrument. The Queen (Mirigavati’s mother) hears of this mysterious yogi, and summons him to her palace. Here the poet tips off the attentive listener to his encoding of spiritual purpose in a single short couplet: ‘He leapt across the seven steps/All seven had separate meanings’.26 Thus, the seven steps of the palace gate echo the seven ordeals as well as the seven storeys of the palace by the lake in which the Prince longed for Mirigavati before he set out on his quest. After climbing the seven steps of the palace in Kanchanpur, the Prince answers the questions posed by Mirigavati and her friends, discards the yogic disguise, and is joined in mystical union with her. Since she represents innate divinity, what he awakens to in this episode is the mystical identity-in-difference of the lover and the beloved, ‘ashiq and ma‘shuq. The path involves more than this, however, for the bridal night (suhag rat) is followed by a musical assembly in court, with the Prince enthroned as the principal patron of the performance. The poet includes a miniature raga-mala in verse, one of the earliest in a regional language, indicating that the novices of this order went through a programme of musical purification and sublimation in the City of Gold. Gold itself is here

Page 10 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī symbolic of alchemical transmutation, for it is the one metal that does not tarnish. Sufi romances are composed enigmatically because of a fundamental problem or enigma with which their spiritual users had to grapple. The enigma, how to understand the relation between human and divine love, was a powerful one for the Chishti Sufis who created this literary tradition. How could a seeker trapped in this material world, subject to human desires and impulses, have access to divine presence? In order (p.288) to achieve closeness to Allah, the Chishtis followed ascetic regimes of spiritual self-fashioning, which involved a hard regimen of spiritual exercises, fasting, and extra prayers. Asceticism, in the world of the Sufis, was frequently yoked to erotic language and imagery. Eroticism was understood as a rung on the ladder of transformation that links worldly and spiritual love. Powerful genres of Sufi poetry and music were among the performative traditions that presented and resolved this enigma through lyric, narrative, and symbolic forms. These forms often contained esoteric codes or elaborate designs that shaped the spiritual transformation of initiated novices under the guidance of a teaching Sufi master. We have in the Mirigāvatī a text that, in a Borges-like twist, resists its own reading or the revelation of textual meaning until one allows the operation on oneself of the allegorical and didactic purpose of the text, which is to work the process of meaning-making on the reader/listener, meaning here defined as a flexible construct that allows one constantly to bring in the fleeting ‘scent of the invisible world’ or to learn some lesson in self-mastery and/or spiritual technique. The text is aware of misguided attempts at classification and mapping, and resists them successfully, instead positing its own rhetorical enterprise as a form into which the reader/listener is invited, imprinting this form on the auditor’s consciousness. Only if one enters into the virtual universe of the text do the purposes and outlines of the text become clear. The poet elaborates on chosen themes with all the sophistication of the Sanskrit rhetoric that had come to him as part of the Islamic translatio studii from classical India, a process parallel to the translations from the Greek that were to give Europe so much of its received canon of philosophy and medicine. Thus, it obscures all attempts to map it because one element or another of the ideal scheme will be out of sync. Frequently, it undercuts or extends the poetic logic of the techniques inherited from classical Sanskrit to treat new themes, for example Qutban’s yoga rasa, part of the triad of effects he tries to evoke for his audience in the service of Sufi sublimation, the others being shringara and vira (love and heroism). Reading, or more properly listening, is a process that engages the senses and implies a different relationship to form and embodiment than the one that we moderns, or in some cases, postmoderns, understand. The process involves, first, allowing one’s consciousness to be imprinted with the summary form or audible icon, and then allowing (p.289) the poet/performer to work on all one’s Page 11 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī senses through the poetic content, as well as the resources of melody and metre, to achieve the professed or unstated goal of the text. Thus, a love scene will describe only those only those flowers that excite desire and so on. We cannot resolve our interpretative confusions without understanding literary meaning as the poets of the genre defined it. However, meaning does not inhere merely in the declarative utterance, but more fundamentally in how the audience understands the utterance and is transformed by it. Meaning, in the Hindavi Sufi romances, is multiple and flexible, and its proper determination is in part a contextual matter, in at least two senses. First, meaning is contextual within the text, as the logic of the particular episode or passage dictates. Second, meaning is contextual in that performances of these poems could be either entertainment or courtly poetry or a blueprint for the serious business of enlightenment, depending on the context of reception. The notion of contextual meaning fits well with the little that we know about the performance and reception of narrative poetry in this period, that is, it was primarily a recited and sung tradition, performed differently and with varied intents in courts, salons, public spaces, and shrines and devotional settings. More crucially, enabling this interpretative flexibility is the principle of polysemy in reading aloud, which is basic to the genre.27 Further, the heroes of these lovestories are always educated and tested in it, underscoring the didactic nature of the genre. In itself, this model of reading involves reading aloud by an educated and literate person who has the ability to explicate the text for the audience. Given this mode of aural reception, and the different contexts of performance (court, salon, and shrine), in the ‘closed’ or private performances in Sufi shrines, certain forms were iconic in two senses: (a) in imprinting the consciousness of the hearer, using archetypal motifs such as cities, castles, palaces, and gardens; and (b) in using these forms simultaneously to conceal and, gradually, through narrative, to reveal the allegorical structural principle of the genre. Obviously, we do not have access to those private sessions of spiritual instruction, but the Sufi shaykhs concealed clues to this process in the coding of particular episodes and the overall form and design of their narratives.

(p.290) Rupa in Sufi Romances The Sufi shaykhs worked out their spiritual purposes through a sustained reflection on and use of material form, signalled in the texts by the Hindavi word rupa, ‘form’. This word is generally paired with nama, the name of a concept or divinity. Thus, we find in the Upanishads the nama as the carefully guarded name with which one conjures up a deity or sacred reality, and the rupa as its visible appearance if the mantric procedures are carried out successfully. Both nama and rupa have been widely used in Indian traditions to articulate quite different modes of the relation between mind, body, and consciousness, as well as to reflect on the shape and form of the cosmos itself and its relationship to the embodied human being. Since the Sufis were working in a cultural and Page 12 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī geographical landscape in which other groups—including the Nath panthi yogis, nirguni poets such as Kabir, and the different Vaishnava sampradayas—were articulating their distinct visions of spirituality, it is not surprising that they should use much of the same ‘technology’ and symbolism that is found in these traditions in the making. The two most important such languages or codes of asceticism and devotion that the Sufis drew on were the imagery and concepts employed also by the Nath panth and the worshippers of the god Krishna. In terms of Indic theories of knowledge and aesthetics, they were following what Ananda Coomaraswamy, citing Dignaga and the Buddhist uses of nama and rupa, draws attention to as basic both to the notion of pramana, truth, and to artistic production of any sort: ‘Indian theories of knowledge regard as the source of truth not empirical perception (pratyakṣa) but an inwardly known model (antarjñeya-rūpa) “which at the same time gives form to knowledge and is the cause of knowledge”, it being only required that such knowledge shall not contradict experience’.28 The poet or sculptor or musician goes through a process of visualization and sadhana to arrive at this form, called dhyana. The perfection or imperfection of the realized form depends on the quality of the samadhi or concentration of the artist, musician, or poet, and the flow of rasa, aesthetic feeling, dependent both on the skill of the creator and the state of consciousness of the listening audience. (p.291) If a form is obscure to the hearer, it means that the consciousness still needs cleansing and polishing. Therefore, it is not surprising that a process of cleansing and purification, as it is habitual to undergo in a spiritual regimen, should be encoded in a summary or iconic form whose pattern only gradually reveals itself through the proportions of the finished, larger work. A concrete example of a Sufi use of rupa or form to mark the shape of the spiritual quest is the symbolism of the palace in the Mirigāvatī, to which I have already referred above. The building is topped with a chaukhandi, a four-cornered ornament. During this period, one type of tomb was called a chaukhandi, a four-cornered decorative shikhara with steps like a stupa, as in Sindh and western India.29 Qutban’s use of the term suggests that the end of the quest will be death, or, in the narrative code of the genre, mystical annihilation or fana. There are three, or rather four, other Sufi uses of rupa. The first has to do with embodiment, rupa as human form, the physical body and its more refined sheaths or layers that constitute the subtle body. This use draws in part on yogic symbolism, as in the landscape of Singhala-dipa, the isle of Singhala that is the stage for the elaboration of the mystical vision of Malik Muhammad Jayasi in the Padmāvat. Since the Padmāvat illuminates a much more thorough-going appropriation of yogic technique than earlier works in the tradition such as the Mirigāvatī, it is worth examining in this context.

Page 13 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī Jayasi describes at the beginning of his romance the paradise-like island of Singhala-dipa, the home of the Princess Padmavati, the symbolic landscape on which the seeker’s inner journey is played out. At the centre of this island is the city and fort of Singhala: A moat stretches out around it, vertiginous. One cannot look over the precipice; it makes the thighs tremble! An abyss so deep, it defies the gaze, makes one fearful of falling into the seven circles of hell. It has nine crooked gateways, nine levels, and whoever climbs up to the ninth one escapes into the cosmos within, Brahman’s egg.30 (p.292) The sun and moon always avoid that fort, else their chariots would crash into dust. Nine gates it has, made of adamant, with a thousand foot-soldiers at each. Five captains of the guard make their circuit. The gates shake at the tread of their feet. … The castle has nine stories and nine gates, each with its doors of adamant. Its ascent has four stages. If one climbs with truth, one arrives!31

Jayasi uses the figure of the body as a city with nine gates, a reference to the nine openings of the body: the mouth, eyes, ears, nostrils, and the organs of excretion and reproduction. The tenth door is the secret opening (brahmarandhra) between and above the eyes in the subtle body, through which the practitioner can enter the microcosmic universe within. Jayasi’s description of the nine-storied city of Singhala is both a description of the subtle body and of a lofty fortress. The five captains suggest the five senses that guard the body, the sensorium that guards the human body and governs the ingress and egress of sensation. The four stages suggest four maqams, the four stages of the Sufi path, shari‘at (following the law), tariqat (the Sufi way), ma‘rifat (gnosis), and haqiqat (realizing the truth), and undoubtedly also refer to a symbolic structure within the larger narrative. Jayasi’s spatialization of the subtle body in the technical language of yoga creates the effect of the internalization of vision through the tropes of a built and embodied landscape. The senses lead us to the second Sufi use of rupa. As guards to the fortress of the body, they have a very ancient provenance, as references in the Upanishads and the Gita reveal. The yogi of stable insight (sthita-prajna) withdraws into the nine-gated fortress of the body. As Krishna explains to Arjuna, if one dwells on the objects of the senses, their vishayas, one is trapped by attachment to them (moha) and cannot free oneself from the cycle of rebirth and death. In Patanjali’s Yoga-sūtra, the yogi withdraws his senses like a tortoise retracting its limbs into its shell in order to achieve inner illumination. Among the Sufis, by contrast, the Page 14 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī senses function as a way of drawing the self out of itself in order to go within, using in this case the icon or rupa of the quest. Since desire is fundamental to the human condition, the Sufis use an object (p.293) of desire to arouse the seeker, to unlock his subtle body, then sublimate that aroused longing through the stages of ordinate love to purify the self and achieve the balance of elements that is fundamental to their view of human physiology and psychology. It is only through the senses that the invisible can be grasped, but how does one draw them out to go within? For the Sufis, love is the answer, in the sense of ‘ishq or passionate love, love that exceeds all bounds, draws the self out of itself, exemplified by Majnun in the Perso-Arabic literary tradition. And here one comes to the third sense of rupa, beauty or elegance of form, which is used in the Hindavi Sufi romances as a calque for the Arabic jamal, beauty or grace. At the beginning of the Mirigāvatī, one year after the Prince’s first sighting of her, Mirigavati appears at the magic lake with her handmaidens. The sarapa or head-to-foot description of Mirigavati is a generic set-piece. From the parting in her hair, described as a line of cranes against a dark monsoon cloud, to her cruel black-tipped breasts and her golden limbs dusted with vermillion, love’s inventory becomes in this Sufi context a description of the fierce (jalali) and gentle (jamali) attributes of her body and its physiological effects on the Prince. In a similar scene early in Jayasi’s Padmāvat, Padmavati and her lovely girlfriends go to play in the Manasarodaka, the magic lake on the island of Singhala, also the inner lake of the mind. The girls decide to play at diving for pebbles. One of the girls does not understand the game (and here Jayasi gets in a suggestion that the true name of the game is love), and loses her necklace in the lake. She is in despair because she cannot return home without her precious ornament. All the girls begin to search the lake, but it is hopeless— where are they to find the priceless jewel? At this point the lake itself intervenes: The lake said, ‘I’ve gained what I desired! The philosophers’ stone of beauty has touched me ‘til here, I have become pure from contact with your feet. I have become beautiful at the sight of beauty! A sandal-scented breeze came to my body! I have become cool, my burning is quenched. I do not know who brought this breeze. My condition is purity, my sins are lost!’ At once he gave up the necklace. As the girls got it, the moon smiled …32

(p.294) In this beautiful passage, beauty is transformative, it purifies. On its touch, the lake of the mind is cooled, as if by a sandal-scented breeze, and beauty provides the means also for the cleansing of sins and the attainment of purity. The lake itself offers the heroine’s girlfriend her necklace back, the

Page 15 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī implication here being that once beauty lodges in the mind, its transformative power is miraculous. And this brings us to the fourth and final use of rupa, its cosmic significance as divine beauty that is refracted through the veils of materiality to become apprehensible to our senses. Beauty (Ar. jamal) is one of the ninety-nine attributes of Allah, and as the well-known hadith qudsi or divine utterance revealed to Muhammad has it, ‘I was a hidden treasure, and longed to be known. So I created the world in order that I may be known’. As the second verse of the prologue to the Madhumālatī puts it: In every state the Supreme Lord is One, a single form in many guises (bhesa). In heaven, earth, and hell, wherever space extends, the Lord rejoices in multiplicity of form. … Hidden, He is manifest everywhere. Formless, He is the many-formed Lord.33

Here the poet plays cunningly on rupa, form or beauty, and the notion of bhesa, disguise, to suggest both God’s essence and the many guises that divine essence takes in the world. The commonly used image, drawn from the Qur’an, is of light refracted through many veils that are only gradually lifted from the seeker’s vision. Essence is both multi-formed and formless, the divine play of wahdat and kasrat, unity and multiplicity, which is a basic assumption of Sufi cosmology.

Emotion and Meaning in the Mirigāvatī In a narrative genre like the Hindavi Sufi romance, we must think of story as strategy, of episodes as archetypal in giving us access to basic human tensions and emotions, as well as historical in that they let us explore the historical worlds the narrative world inhabits. There is (p.295) a language of emotion that is implicit in any tale of desire aroused, deferred, and satisfied, and this notion, called bhava in the Sanskritic repertoire of love, allows the author to articulate his unique, yet predictable (for his audience) agenda. In the Mirigāvatī, Qutban’s rendering of emotion and its reshaping is coded with the valences of a Sanskritic technology of sublimation, pressed into the service of a distinctively Sufi agenda, structuring the quest in very specific ways, as we have seen. Even after the Prince has reached full unity with Mirigavati and, with her help, tamed the demon of the carnal soul (nafs-i ammara), he has to return to the world, the domain of his first wife, Rupmini. As in other Sufi romances, when the two wives meet, they fight, and the Prince has to calm them down by assuring them of his enduring love for them both, sleeping with both of them in turn to satisfy their jealous desire. Here, the poet uses the gender politics of the harem to indicate that only the seeker who can balance the spiritual and the material can live happily in this world. The placement in the story of the abandoned wife’s suffering indicates that the seeker has to reintegrate himself into the Page 16 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī rhythms of this world after attaining his spiritual object, balancing the spiritual quest with the demands of the world. This turning point suggests also the homologies of seasonal time with the elements, the cycle of nature constituting the context within which the embodied human being has to operate. Qutban’s romance shows us the Prince’s awakening to inner divinity by controlling his baser impulses, his successful struggle with his carnal soul, and his eventual reconciliation with his first wife and reintegration with the world.34 (p.296) To go back to the steps of the palace gate, Qutban refers to them with the suggestive word bhava, which can signify ‘being’, ‘meaning’, or ‘emotion’. The word is also used in literary criticism to signify the feeling or emotion that is the basis of the rasa that permeates a particular passage, poem, or play. The usage is significant, both historically and conceptually. In a historical sense, the Sufi use of bhava and rasa allows us to add specificity to the genealogical statement of legitimation found in texts like the Bhāgavata-māhātmya (1:48–50), that bhakti was born in the South, grew old in Gujarat and Maharashtra, had her limbs ‘riven by schismatics’, and was then reborn as a beautiful young woman in Vrindavan in the sixteenth century.35 Conceptually, the major move that allows the sixteenth-century Vaishnava sampradayas to anchor their theology of devotional feeling is to use the Sanskritic theory of rasa and bhava, substituting in pride of place bhakti-rasa instead of shringara, the erotic mood of classical poetry. The systematic articulation may be seen in schematic texts like the Bhaktirasāmṛtasindhu from the 1540s, but what is more important is that the theology is used to justify the devotional practices, developed earlier, of the Vaishnava groups. The Sufis were competitors and conversation partners on this scene of historical interaction. Qutban’s Mirigāvatī is the perfect instance of the Sufis’ use of the Sanskritic technology of the sublimation of desire into divine love in order to express their own ideology of Islamic monotheism. The inner journey is depicted as a yogic quest, denoting the historical interaction with the Nath panth. Its poetic logic of sublimation using the language of bhava, emotion, and rasa, the juice or essence of aesthetic and devotional experience, evolved in dialogue with the practices of the sects devoted to the worship of Krishna, who may themselves (p.297) have owed something to the ladder of ordinate love that shapes the Sufi path of asceticism. The Hindavi Sufi poets use both these local languages of asceticism and sublimation, to express in an Indian language their distinctive message of the love of Allah. Qutban’s appropriation of bhava is both competitive and practical, as we can see if we look at the episodes on the Prince’s quest. I will resist the temptation to map any particular Suhrawardi schemata onto the text, because this is the rock on which much of the criticism of the genre has foundered. The journey involves spiritual process, and in it characters can frequently shift roles.

Page 17 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī Here we should consider Qutban’s use of language. If we go back to his praise of his controversial pir, Shaykh Buddhan Suhrawardi, we find the following lines: Our pir is the greatest of all! The one to whom he shows the path, reaches his goal in just one second! If someone knows enough to walk along the path that he has shown to the world, he reaches his goal in an instant, if he holds fast to the path of true feeling. (Mirigāvatī 6.4–6)

The operative phrase here is the last, which reads in the original ‘jau satabhāvahi soi’. Here sata-bhava, the path of ‘true feeling’, conceals a clever pun, for sata is also the contraction usual in prefixed compounds for ‘seven’. Qutban plays on this usage, which seems innocent at first encounter, drawing the listener into various seemingly random ordeals, in some of which he extols the value of sata-bhava (such as in the herdsman’s cave) and only reveals the numbered structure of the code to us at the end, when the Prince steps over Mirigavati’s threshold. Although they enjoy a wedding night together, in which the flow of rasa is graphically depicted as the erotic culmination of the Prince’s quest, it is significant that there is no actual wedding ceremony with its own sapta-padi, its seven steps around a sacred fire. The Prince has completed his mystical ‘seven steps’ and does not need to go through the external ritual. It is also perhaps significant that the entry into the palace occurs at almost the exact numerical halfway point in the narrative. Qutban brings his romance to a conclusion by noting that the rasas of his story have been joga, singara, and vira, asceticism, love, and valour (Mirigāvatī 426.2). These too are signified in the iconic form of (p.298) seven-levelled palace, for the wall-paintings depict scenes of love (Rama and Sita, Krishna and his gopis), valour (Angada in Lanka and Bhima’s killing of Kichaka and Duhshasana), and asceticism (Bhartrihari and Pingala), and, of course, the magic doe, desire for whose form impels the hero on his interior journey. No detail is accidental, and the story is certainly not a random series of episodes strung together, nor just ‘formulaic’. Rather, Qutban’s narrative design reveals a creative and early use of Indic categories to present a pattern of mystical sublimation that is inherently processual, in two senses: in the apprehension of the hidden meanings of the symbolism and imagery (the bhava of each passage) and in the application of these bhavas to the raw material of feelings and drives that impel humans in their journey through the material world. In a sense, each episode asks the listener with heart (sa-hridaya) to understand the nature of that particular emotion and to turn it around, moving on to the next stage when a particular lesson is learnt. The Sufi master who would have guided the disciple’s progress would reveal the code gradually, but the overall pattern is already imprinted on the murid’s consciousness in the opening verses of the love-story. A parallel art Page 18 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī form, also dependent on the sense of hearing, is music, in which the musician also ‘draws the face’ of the raga in the introductory alap (mukhra or shakl khinchna in Hindi/Urdu), in a sense building a melodic house whose outlines are filled out with ornament as the raga unfolds in its vistar. Thus, the arts in this period worked through the education of the five senses, and poets, artists, and musicians frequently played with the sensory coding of their art-forms, creating intricate patterns to deliver the flow of rasa to the educated consciousness. If one looks at the texts of the other Hindavi Sufi romances, one finds there, too, iconic descriptions of buildings and landscapes that refer in summary fashion to the narrative of spiritual process that is to unfold, using magic lakes, cities, castles, gardens, picture-pavilions, and so on, to encode spatially the interior journeys of the different Sufi traditions that were active in the period. The Sufi use of rasa poetics as an integral part of their technology of sublimation shows their level of indigenization, while the distinctive Sufi slant, the Persianate model of the masnavi, shows the integration of even a provincial sultanate like Jaunpur into the larger Dar al-Islam. Notes:

(*) Editors’ note: Aditya Behl was working on a translation of Mirigāvatī when he presented his paper on ‘The Questing Self: Emotion and Structure in Qutban’s Mirigāvatī’ at the SOAS conference in 2007. This has now been published as The Magic Doe: Quṭban Suhravardī’s Mirigāvatī, Wendy Doniger (ed.), New York: Oxford University Press, 2012. What follows is a shorter version of the introduction to his translation prepared by his literary executor, Professor Wendy Doniger, which incorporates parts of ‘The Questing Self’ as well as of another unpublished essay, ‘On Rūpa: Form, Embodiment and Technique in the Hindavi Sufi Romances’, which he gave as a public lecture. The reader might also benefit from consulting another essay, ‘The Magic Doe: Desire and Narrative in a Hindavi Sufi Romance, circa 1503’, in India’s Islamic Traditions: 711–1750, Richard M. Eaton (ed.), New Delhi: Oxford University Press, 2003, pp. 180–208. (1) Khwaja Mu‘in al-Din (or Moin al-Din) Chishti is said to have brought the Chishti Sufi order to India late in the twelfth century. (2) For a more detailed history of the period, see M. Habib and K.A. Nizami, A Comprehensive History of India, Volume V: The Delhi Sultanate (A.D. 1206– 1526), New Delhi: People’s Publishing House, 2nd ed. 1993, pp. 630–732. For a detailed history of the Sharqi kingdom, see M.M. Saeed, The Sharqi Sultanate of Jaunpur: A Political and Cultural History, Karachi: University of Karachi, 1972. (3) Saeed (1972), p. 111. See also S.H. Askari, ‘Qutban’s Mrigavat: A Unique Ms. in Persian Script’, Journal of the Bihar Research Society, 41(4), 1955: 457–8; Askari, ‘Bihar under Later Tughlaqs and Sharqis’, in his Medieval Bihar: Page 19 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī Sultanate and Mughal Period, Patna: Khuda Bakhsh Oriental Public Library, 1990, pp. 22–31, and D.F. Plukker, ‘The Miragāvatī of Kutubana’, PhD dissertation, University of Amsterdam, 1981, p. xviii, n. 4. (4) For details see Saeed (1972), pp. 111–12 and 206–7, and A. Halim, ‘History of the Growth and Development of North-Indian Music During Sayyid-Lodi Period’, Journal of the Asiatic Society of Pakistan, 1(1), 1956: 46–64. Also see Katherine Butler Brown, ‘The Origins and Early Development of Khayal’, in Hindustani Music: Thirteenth to Twentieth Centuries, Joep Bor, Françoise ‘Nalini’ Delvoye, Jane Harvey, and Emmy te Nijenhuis (eds), New Delhi: Manohar, 2010, pp. 159– 94. (5) S.A.A. Rizvi, A History of Sufism in India, Delhi: Munshiram Manoharlal, 1978, vol. 1, p. 367. (6) S.M. Pandey, ‘Kutuban’s Miragāvatī: Its Content and Interpretation’, in Devotional Literature in South Asia: Current Research, 1985–1988, R.S. McGregor (ed.), Cambridge: Cambridge University Press, 1992, p. 180. Editors’ note: One possibility is that S.M. Pandey misinterpreted Hafiz MahmudSherani’s Urdu treatment of the Persian Chistiyya Bihishtiyya, a mid-seventeenth-century account of this family of Sufis. Sherani, ‘Makhdūm Shaikh Bahā’ al-Dīn Barnāwī’, Oriental College Magazine, part 1 (August): 41–58: the Chistiyya Bihishtiyya says that the family first settled in Barnawa, but then in the late fourteenth century one of them, a Shaykh Nasir al-Din, moved east and settled in the Shaykhpura at Rapri and then, fleeing Timur, at a place near Chunar, where the family was supported by the local Afghan amir. Pir Buddhan was his eldest son, and he was the one who had relations with Sultan Husayn Shah Sharqi. (7) M.M. Saeed suggests that another possibility for Qutban’s teacher is the Mahdavi Shaykh Burhan al-Din Ansari of Kalpi (d. 1562–3). Shaykh Burhan alDin was also a Hindavi poet and instructed Malik Muhammad Jayasi, the author of the Padmāvat. However, in view of Qutban’s mention of Shaykh Buddhan Suhrawardi in the text, the ascription cannot stand. See Saeed (1972), p. 200. (8) M.M. Saeed, Tażkira mashā’ikh-i Shirāz-i Hind (Jaunpur), Lahore: Islamic Book Publishers, 1985, pp. 255–6. (9) For a good short history of the important doctrines and political involvements of the order in Baghdad, Multan and Uchch, see Qamar-ul Huda, Striving for Divine Union: Spiritual Exercises for Suhrawardī Sūfis, London: Routledge Curzon, 2000. For an account of their activities in Avadh, see also Rizvi (1978), and Saeed (1972) on Jaunpur. (10) G.A. Grierson, The Modern Vernacular Literature of Hindustan, Calcutta: Asiatic Society, 1889, pp. xviii, 18.

Page 20 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī (11) See the discussion of Ramchandra Shukla’s Hindī sāhitya kā itihās in Ganapatichandra Gupta, ‘Premākhyān-kāvya’, in Nagendra and Sureshchandra Gupta (eds), Hindī sāhitya kā itihās, Noida: Mayur Paperbacks, 1991, pp. 144– 50. (12) Gupta, ‘Premākhyān-kāvya’, p. 144, and Parashuram Chaturvedi, Bhāratīya premākhyān, Allahabad: Bharati Bhandar, 1985. (13) S.M. Pandey, Madhyayugīn premākhyān, Allahabad: Lokabharati Prakashan, 1982, and R.S. McGregor, Hindi Literature from its Beginnings to the Nineteenth Century, Wiesbaden: Harrassowitz, 1984, p. 10. (14) Peter Gaeffke, ‘Alexander in Avadhī and Dakkinī Maṭḥnawīs’, Journal of the American Oriental Society, 109, 1989: 527–32. (15) Pandey (1992), pp. 186–7. (16) S.C.R. Weightman, ‘Symmetry and Symbolism in Shaikh Manjhan’s Madhumālatī’, in The Indian Narrative: Perspectives and Patterns, Christopher Shackle and Rupert Snell (eds), Wiesbaden: Harrassowitz, 1992a, pp. 208–9. (17) V.S. Agraval, ‘Introduction’, in Malik Muhammad Jayasi, Padmāvat: Malik Muhammad Jāyasī kṛta mahākāvya (mūla aur sanjīvinī vyākhyā), V.S. Agraval (ed.), Chirganv, Jhansi: Sahitya Sadan, 1998, p. 57. References to the text are to this edition. (18) John Millis, Malik Muhammad Jāyasī: Allegory and Religious Symbolism in his Padmāvat’, Ph D dissertation, University of Chicago, 1984, p. 108. (19) Shantanu Phukan, ‘The Lady of the Lotus of Gnosis: Muhammad Jayasi’s Padmavati’, unpublished paper, 2–3. I am grateful to Shantanu Phukan for sharing his unpublished work with me. (20) Thomas de Bruijn, ‘The Ruby Hidden in the Dust: A Study of the Poetics of Malik Muḥammad Jāyasī’s Padmāvat’, PhD dissertation, University of Leiden, 2012, pp. 104–6. (21) Patricia Parker, Inescapable Romance: Studies in the Poetics of a Mode, Princeton: Princeton University Press, 1979, p. 4, cited in Barbara Fuchs, Romance, New York: Routledge, 2004. (22) Francesca Orsini, ‘Introduction’, in Love in South Asia: A Cultural History, F. Orsini (ed.), Cambridge: Cambridge University Press, 2006, pp. 4–27. (23) A.K. Ramanujan, ‘Introduction’, in his Speaking of Śiva, Harmondsworth: Penguin, 1973, p. 40.

Page 21 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī (24) Suniti Kumar Chatterji, The Various ‘Matters’ in New or Modern Indian Literature and the Romances of Mediaeval Bengal (Gauda Banga Ramya Katha), Calcutta: The Asiatic Society, 1983, pp. 10–11. He adds a fourth ‘matter’, of the modern world, to cover the literature of the colonial period, but that is not relevant for our purposes here. (25) See fn. 29 later. (26) Qutban Suhrawardi, The Magic Doe: Qut̤ban Suhravardī’s Mirigāvatī: A New Translation, Aditya Behl (trans.), Wendy Doniger (ed.), New York: Oxford University Press, 2012, pp. 118, 211.3. (27) Cf. Jayasi (1998), 108.6, eka eka bola aratha caugunā. (28) Ananda K. Coomaraswamy, The Transformation of Nature in Art, New York: Dover Publications, (1934)1956, pp. 15–16. (29) See Salome Zajadacz-Hastenrath, Chaukhandi Tombs: Funerary Art in Sind and Baluchistan, Oxford: Oxford University Press, 2003. (30) Jayasi (1998), 40.3–5, p. 40. (31) Jayasi (1998), 41.1–3, 8, p. 41. (32) Jayasi (1998), 65.1–5, p. 65. (33) Madhumālatī, 2.1–2, 5; see Mir Sayyid Manjhan, Madhumālatī: An Indian Sufi Romance, S. Weightman and A. Behl (trans.), Oxford: Oxford University Press, 2000, p. 3. (34) Once this resolution is achieved, the narrative universe is folded back into itself: a hunter comes to the Prince because a tiger has spread terror in the jungle by gouging out the swellings on the foreheads of all the elephants that live there. The Prince goes to the jungle and is soon locked in deadly combat with the tiger. The Prince fights the tiger and deals him a death-blow, but is wounded mortally in combat. Meanwhile, the tigress gives birth to a tiger-cub, and the surviving elephants rush into the forest to exact their revenge. But the little cub subdues the elephants and eats the pearls out of their foreheads. A complex symbolism is encoded here: in Indic animal-lore, the tiger is the hereditary enemy of the elephants, in whose forehead swellings are found the rare ‘elephant pearls’ (gaja-moti), symbolizing mystical awareness and insight (ma‘rifat); here they are the pearls of mystical gnosis. The tiger is the force of desire, represented here as a predator who kills spiritual insight. Desire, the poet suggests, is the life force; you can try to kill it, but in the process you die yourself and desire is always reborn triumphant. The implication is two-fold: desire is basic to life, is the life force that impels even God’s creation of this world. If you kill desire, you die, but the tiger-cub of desire lives on and, unless Page 22 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Emotion and Meaning in Mirigāvatī properly tamed (i.e., sublimated towards divinity), can devour the pearls of mystical illumination. (35) John S. Hawley, ‘The Bhakti Movement—Says Who?’ Special issue of the International Journal of Hindu Studies, 11(3), 2007: 209–25, as well as my own response to that issue, ‘Presence and Absence in Bhakti: An Afterword’, pp. 319– 24, for examples of connections between Islamic and Indic uses of images and narrative motifs.

Access brought to you by:

Page 23 of 23

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

The Art of the Book in India under the Sultanates Éloïse Brac de la Perrière

DOI:10.1093/acprof:oso/9780199450664.003.0011

Abstract and Keywords This chapter offers a tentative classification of the genres of illustrated book production under the north Indian sultanates of the fifteenth and the early sixteenth centuries. Through an analysis of the extant codexes, the chapter proposes five categories: the first one, of manuscripts linked by archaizing Persian painting styles whose texts belong to classical or Indo-Persian literature; the second, of manuscripts produced by artists who had worked previously in non-Muslim contexts; the third, containing paintings with contemporary Persian themes and texts; and the fourth being renderings of the Candāyan. The fifth group consists of manuscripts of illuminated Qur’ans. The chapter finds that while its classification found evidence of archaizing tendencies and a range of subject choices, more work needs to be done to classify these manuscripts and engage in stylistic analysis of illustration and illumination. Keywords:   Sultanate, painting, Qur’an, Persian literature, Indo-Persian, Jain, illumination

SEVERAL ORDERS OF PROBLEMS DOG the study of illustrated books in Sultanate India. The first set of problems relates to the paucity, complexity, and lateness of the available sources. While the architectural achievements of the Delhi and regional Sultanates are still standing and impress with their monumentality and aesthetic accomplishments, no dated illuminated manuscript is known prior to the Gwalior Qur’an of 801H/1399 (Plate 11.1)—despite the fact that the Delhi Sultanate had been established two centuries earlier and despite textual evidence dating back to the early fourteenth century that attests that Page 1 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates Indo-Islamic rulers patronized book production. Paradoxically, though almost nothing remains of the frescoes that decorated Sultanate buildings and the overwhelming majority of Sultanate paintings are to be found in manuscripts, textual sources from the period allude many times to the (now vanished) monumental decorations whereas they never mention the arts of the book.1 (p.302) And yet, despite this apparent lack of interest, it seems difficult to deny that the illustration of manuscripts was important to Indo-Islamic elites during the Sultanate period. To give one example, the number of illustrations in each work is high: often more than fifty and sometimes nearly a hundred, with the Nujūm al-‘ulūm (1570) from Bijapur an extreme case with 876 miniatures, more than one per page.2 Even when only a few folios of a manuscript are extant, certain deductions may still be drawn about the overall volume. For instance, from the remaining thirty-one illustrated folios of Amir Khusraw’s Khamsa (Plate 11.2), it is possible to state that the whole manuscript consisted about 300 folios and was probably highly illustrated.3 In the same way, the Berlin Ḥamzanāma contains 199 miniatures, and the Candāyan from the same collection, 140 illustrations per 141 folios. In fact, all the Candāyans were widely illustrated, with paintings that occupy the full page and illustrate the narrative very precisely. Two hundred and eighty-five pages out of 318 folios of the Manchester Candāyan (Plate 11.11)—the most complete copy at our disposal—are painted: this means almost every other page. (p.303) As we shall see, illustrations seem to have been more valued than illumination.4 However, given that most of the illustrated manuscripts have been dismembered and scattered in collections all over the world and often only a few folios are extant, it is extremely difficult to undertake a comparative analysis of how the subjects were treated and of ‘illustrative cycles’, by which I mean the episodes selected for illustration and the frequency of the illustrations in the manuscript, that is to say the specific rhythm with which an image appears in the text. The second order of problems relates to the attribution of the manuscripts to a particular period, location, and context of production and patronage. Of the corpus of over a hundred documents attributed, whether rightly or wrongly, to the various Indian Sultanates, fewer than fifty illuminated works can be ascribed with certainty to this period, a period that covered over three centuries and a territory whose frontiers stretched from Kashmir to the Deccan, from Bengal to Gujarat.5 Only seven manuscripts are dated or located by their colophon, though in the other cases there are incontestable elements that enable us to guarantee that these manuscripts were of Indo-Islamic origin, be it attested (p.304) attribution or some convincing graphic, iconographic, or technical feature. Yet their production context is still for the most part enigmatic, and the lack of references on this point is difficult to comprehend, given that the period concerned is relatively late in the history of Islamic manuscripts. Take the case Page 2 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates of the Gwalior Qur’an (Plate 11.1) of 801H/1399.6 Gwalior is well known in this period as a Tomar court and a Jain centre (see Chapters 12 and 13 in this volume), but how did this richly illuminate Qur’an come to be produced here? The fact that many Sufis fleeing south when Timur’s army attacked and sacked the Tughluq capital city in 1399 stopped there briefly cannot be reason enough to explain how such a work appeared in such a place (despite Simon Digby’s contention).7 The richness of this Qur’an necessarily implies cooperation among several established artists and craftsmen specializing in the art of the book: painters, calligraphers, paper-makers, and so on. Yet nothing in the sources allows us to believe that Gwalior was a centre for this kind of artistic production at any precise time. The second dated manuscript—the British Library ‘Jaunpur Anthology’, whose internal name is Dastūr al-shu‘ara’ (A Manual for Poets)—is probably an autograph manuscript and its foreword refers explicitly to Mubarak Shah, an early fifteenth-century ruler of Jaunpur (Figure 11.1).8 If its origin is less mysterious than that of the Gwalior Qur’an, since it is connected to a sultanate whose patronage is attested, the richness of its quite elaborate decorations is nevertheless inferior to the decorations (p.305)

Page 3 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (p.306) of the Qur’an. But it also confirms that stylistic forms found in later works—such as some of the bihari Qur’ans—existed well before. It constitutes, therefore, an important stylistic link between those later works and the Gwalior Qur’an. The other rare mentions of commissioners or artists in illustrated manuscripts are much later: two manuscripts attributed to the sultanates of Malwa and Bengal in the first half of the sixteenth century,9 and two more to Golkonda in the second half of that century.10

In the same way, the invaluable clues offered by seals or second-hand notations added by librarians or owners are rare in Sultanate books. This could be for a large part a result of the very scattered situation of the material: the folios bearing this information, usually located at the beginning or at the end of the book, are almost always missing because of their very vulnerable place in the

Figure 11.1 Jaunpur Anthology, Dastūr al-shu‘ara, beginning of the fifteenth

manuscript. At the same time, it bears pointing out that no

century (British Library, London, Or 4110, f.153 recto).

systematic study based on the existing data has been carried out. Textual indications concerning patronage and artists are also always scattered and elliptical: they inform only in an indirect manner. We find mention of some books imported from Iran or Egypt for rulers or important citizens who were bibliophiles, or else of the monetary value of a book, more precisely its costliness.11 Another clue is the fact that some sources insist regularly, even if discreetly, on the Persian origin of the craftsmen: this tells us that this specific origin was considered a warranty for the artistic quality of the work in preMughal India and acted as a badge of excellence similar to the one carried by Chinese painters in Persian literature.12 We, therefore, need to read between the lines in order to discover exact indications concerning the production context of the illustrated manuscripts.

Page 4 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates Given the great heterogeneity of the material, and the lack of details about attribution, date, and context discussed so far, classification (p.307) becomes especially important in order to map the corpus and begin to explore the art of the Indo-Islamic book in Sultanate India. In this chapter I will propose my own classification of the existing corpus into five groups. After making some general points about the literary genres that were illustrated, I will tease out the peculiar stylistic features of each group. A stylistic classification is one way art historians have to make links between undated and non-localized manuscripts and to assume that production centres existed. A stylistic classification also aims to gather patterns and heritages in an artistic context. I will then offer some general remarks about illumination and illustration in Sultanate books in a comparative perspective, and about their evolution over the whole period.

Classifications Given the paucity of studies on the art of the book in the Indian Sultanates, it is hardly surprising that suggestions on how to classify them have also been few. Norah Titley wavered between grouping the texts on the basis of Persian stylistic features in the paintings or on the basis of the area to which the manuscripts putatively belonged.13 The heterogeneity of a corpus drawing upon very different artistic traditions as well as the lack of unquestionable attributions revealed all the problems with such an attempt. The classification proposed by J.P. Losty in the context of the beautiful British Library exhibition entitled ‘The Art of the Book in India’ (1982) is more satisfying.14 This classification was essentially based on stylistic criteria and divided the manuscripts into three groups with the first group itself divided into two subgroups. Losty first singled out the manuscripts related to different Persian traditions, traditions sometimes chronologically quite distant from one another. His second group consisted of manuscripts that drew explicitly upon the Indian context, even if their style also included some elements belonging to the Islamic repertory of the thirteenth and fourteenth centuries. His third group comprised two illustrated copies of the Candāyan (Plates 11.6 and 11.7), an Avadhi poem written in the Perso-Arabic script that may be dated to between the first quarter (p.308) and the third quarter of the sixteenth century, thus to the last period of the sultanates’ production.15

Page 5 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates Almost thirty years later and with a larger corpus of manuscripts, I would like to propose a new classification into five groups. My classification is also mainly based on stylistic criteria (e.g., Group 2), but not exclusively so. Other criteria are codicological or textual: for example, all the manuscripts of Group 3 are complete (or partly), whereas those of Groups 1 and 2 are dispersed. Qur’ans cannot be grouped together with the other manuscripts because they could not be illustrated. Their artistic problematic is different because it only concerns the field of illumination. 1. Those manuscripts that may be dated to the fifteenth century and whose paintings present a style close to that of the Persian miniatures dating from the thirteenth and the first half of the fourteenth centuries, and also, to a lesser degree, to the style of the Mamluk contemporary or later manuscripts.16 2. Manuscripts that may be dated to the fifteenth century, whose paintings share the same archaizing references with the former group but also show a great number of non-Islamic Indian features, as much as in the iconography as in the style. 3. Manuscripts that may be dated to the very end of the fifteenth century and the beginning of the sixteenth century, inspired by Timurid or Aq Qoyunlu Persian patterns, or, less frequently, by Safavid ones.

Page 6 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates

Plate 5.1 ‘Korasht’, children playing, Miftāḥ al-fużalā, by Muhammad b. Da’ud b. Muhammad b. Mahmud Shadiabadi. The author here indicates the Indian ‘ड’ (retroflex ‘d’) in the word dandā muvahī with three points under the letter. (British Library, London, Or. 3299, f. 220 verso).

Page 7 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates

Plate 11.1 Qur’an manuscript, completed 801H/1399 in Gwalior Fort. (Aga Khan collection, Acc. No. AKM00281/ex Ms. 32).

Plate 11.2 ‘The tomb of Alexander the Great’, page from a Khamsa of Amir Khusraw Dihlawi, first quarter of the fifteenth century (Aga Khan collection, Acc. No AKM00015/ex INDM001B).

Page 8 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates

Plate 11.3 One page from a Shāhnāma, first half of the fifteenth century (Bharat Kala Bhavan, Varanasi, Acc. No. 9940).

Plate 11.4 ‘Rustam kills the White Dīv’, page from a Shāhnāma, first half of the fifteenth century. (Museum Rietberg, Zurich, Acc. No. RIV 962, Gift Rietberg Society © Photograph: Rainer Wolfsberger).

Page 9 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates

Plate 11.5 ‘Ghiyas Shah seated in a garden pavilion, being offered dishes by female attendants’, page from Ni‘matnāma, between 1495 and 1505 (British Library, London, IO Islamic 149/ Ethé 2775, f.40 verso).

Page 10 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates

Plate 11.6 Candāyan, sixteenth century (Chhatrapati Shivaji Maharaj Vastu Sangrahalaya, Mumbai, Acc. No. 57.I/ I-68, f. 22). (Courtesy of the Trustees, Chhatrapati Shivaji Maharaj Vastu Sangrahalaya [formerly Prince of Wales Museum of Western India], Mumbai, India).

Page 11 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (p.309) 4. Manuscripts that may be dated between 1520 and 1570, characterized by the fusion of elements descending from both Persian and Indian repertories to shape an entirely original style. 5. The complete Qur’ans that belong to the period. Let me now first present the corpus before pointing out the stylistic features of each group.

The First Group This group gathers two codexes (one is incomplete), and also some scattered sheets. All the texts belong either to classical Persian literature (Nizami, Firdawsi, and Rumi) or to Indo-Persian poets (Amir Khusraw). Two or maybe three of the six works that constitute this group (if we consider an

Plate 11.7 Candāyan, Malwa, early sixteenth century (John Rylands University Library, Manchester, Acc. No. Hindustani Ms.1, f. 146 recto. Copyright of the University of Manchester).

isolated page attributed until now to an Iskandarnāma whose text cannot be verified because the painting was separately cut out and then glued on a support) belong to the Khamsa of Nizami. These manuscripts are linked together by the style of their paintings, tied to older and archaizing Persian traditions, though not by their epigraphic or codicological features (format, sizes, composition), which vary considerably. This group includes 1. thirty-four folios belonging to a Khamsa of Amir Khusraw Dihlawi, scattered between various private and public collections (Plate 11.2); 2. seven folios of a Shāhnāma,17 Bharat Kala Bhavan, Varanasi (9940, 9941, 9942, 9943, 9944, 9945, 9946) (Plate 11.3);18 (p.310) 3. Khamsa of Nizami, Biblioteca dell’Accademia Nazionale dei Lincei, Roma (Ms Or. 17);19

Page 12 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates 4. Khamsa of Nizami partially scattered before 1965, but whose quite considerable main part (271 ff.) was presented by Sam Fogg’s gallery in September 2000;20 5. two folios originating from a Masnavī by Rumi: the first is located at the Institut du Monde Arabe in Paris (AI 87-16)21 and the second in a nonlocalized collection;22 and 6. one painting set on a support, perhaps an Iskandarnāma extract, private collection (Berlin).23

The Second Group This group, more disparate than the first one, brings together what appear, on stylistic grounds, to have been the first attempts by artists who had worked in a non-Muslim Indian context to illustrate Islamic manuscripts. They illustrate some Persian, Indo-Persian, or vernacular literary texts. The manuscripts are as follows: 7. About twenty folios of a Shāhnāma, scattered between many collections (Plate 11.4)24 (p.311) 8. A Ḥamzanāma, Staatsbibliothek Preussischer Kulturbesitz, Berlin (Ms. Or. fol. 4181) 9. One folio of a Ḥamzanāma, N.C. Mehta Museum, Ahmadabad (no. 355) 10. Two folios of an Iskandarnāma, N.C. Mehta Museum, Ahmadabad (nos. 6 and 7) 11. One Sikandarnāma (Figure 11.2), now missing25 12. Six folios of a Candāyan, Bharat Kala Bhavan, Varanasi (inv. 230) (Figure 11.3) 13. A second and thicker Candāyan, Staatsbibliothek Preussischer Kulturbesitz, Berlin (Ms. Or. fol. 3014)26 14. Twenty-four folios of a third Candāyan, scattered between the museums of Lahore, Karachi, and Chandigarh27

The Third Group The fourteen manuscripts displayed in this group are largely inspired by contemporary or slightly earlier Persian patterns, barring the (p.312)

Page 13 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (p.313) (p.314) ‘Ajā’ib alṢanā’i‘.28 If the paintings vary according to the different styles each manuscript adopted, the tight link that binds them to Iranian artistic production from the second half of the fifteenth and the beginning of the sixteenth century unifies them in a same group. Unlike the preceding group, none of the manuscripts of this group has been disassembled and scattered.29 As usual, some manuscripts are incomplete. Some of them are dated and located, others only dated. Only six of them do not present any precise dating or location—a very low number in the context of this study—though pictorial signs and information concerning the texts confirm they do belong to the Indian Sultanates. The oldest manuscripts belonging to this

Figure 11.2 One page from a missing Sikandarnāma, fifteenth century.

group are as follows:30

15. Ni‘matnāma, India Office Library, London, Persian Ms 149; between 1495 and 1505 (Plate 11.5).31 16. Bustān, National Museum, New Delhi, 48.6/4; between 1500 and 1502;32 17. Miftāḥ al-fużalā’, British Library, London, Or. 3299; between 1490 and 1500 (see also Chapter 5 in this volume) 18. ‘Ajā’ib al-Ṣanā’i‘, British Library, London, Or. 13718; 4 shavvāl 914H/1509 19. Sharafnāma, British Library, London, Or.13836; 938H/1531–32 All the other manuscripts are later than 1550; they can be considered as the last remnants

Figure 11.3 One page from a Candāyan, fifteenth century, (Bharat Kala Bhavan, Varanasi, Acc. No. 230).

Page 14 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates of the art of the book in the Sultanates and the first examples of the art of the book in the Deccan: 20. Ta’rīf-i Ḥusayn Shāhī, Poona, Bharata Itihasa Samshodhaka Mandala; between 1565 and 1567 (p.315) 21. Zakhīr-i Khwārizmshāhī, Chester Beatty Library, Dublin, Ind. MS 30; 980H/1572 22. Sindbadnāma, India Office Library, London, Persian Ms 3214 23. Anwār-i Suhāylī, Victoria and Albert Museum, London, I.S. (1-126)-1962; 990H/1582 24. Khusraw-u Shīrīn, Khuda Bakhsh Library, Bankipore; Golkonda, 976H/ 1568–69 25. Nujūm al-‘ulūm, Chester Beatty Library, Dublin, Ind. Ms 2; 978H/1570 26. Jawāhir al-mūsiqāt-i Muḥammadī, British Library, London, Or. 12857 27. Yūsuf-u Zulaykha, Bibliothèque nationale de France, Paris, Supplément persan 1919 (Figure 11.4)

The Fourth Group This group is essentially constituted by two illustrated copies of the same work, the Candāyan.33 Their paintings share a few common points with the manuscripts of the preceding group but their style shows a perfect and harmonious assimilation of the various elements. Among all the Sultanate paintings, this fourth group is probably the most original and most fully formed. The stylistic homogeneity may perhaps be linked to the context of production, though there is no evidence to confirm it. The group comprises: 28. Candāyan, Chhatrapati Shivaji Maharaj Vastu Sangrahalaya, Mumbai, Acc. 57.I/I-68 (Plate 11.6).34 29. Candāyan, John Rylands University Library, Manchester, Hindustani MS.1 (Plate 11.7)35 (p.316)

Page 15 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (p.317) The Fifth Group Qur’anic manuscripts or isolated pages from Qur’ans constitute a large percentage of illuminated manuscripts remaining from the Indian Sultanates. They still crop up quite regularly on the art market or in small libraries and the stores of little-known museums. That is why I will not try to give an exhaustive list of them, but consider part of a sample I have presented in an earlier study.36 This sample is a great resource, for it includes representative and original specimens that deserve to be analysed. In addition to these Qur’ans, this group also includes the Jaunpur Anthology kept at the British Library, which is, as we have seen, illuminated in a similar style and datable to the littledocumented beginning of the fifteenth century. This manuscript will allow me to

Figure 11.4 Yusuf-u Zulaykha, text copied in 950H/1539. by Shah Mahmud Nayshaburi, paintings probably second half of the sixteenth century (Bibliothèque nationale de France, Paris, Supplément persan 1919, f. 17 recto).

infer the dating and location of other manuscripts of this group for which we still only have very few clues. I’ll come back to this fifth group later in this chapter. 30. Qur’an, Sadruddin Aga Khan collection, Geneva, AKM00281 (ex Ms 32) (Plate 11.1) 31. Qur’an, Bijapur Museum, 912 32. Qur’an, Nasser D. Khalili collection, QUR 237 33. Qur’an (two folios), Nasser D. Khalili collection, QUR 602 34. Qur’an Christie’s ‘1’, 1/05/2001, no. 21 35. Qur’an Sam Fogg, ‘Islamic Manuscripts’, 2000, no. 14 36. Qur’an Taipei Museum37 37. Qur’an, Bibliothèque nationale de France, Paris, Arabe 7260 (Figure 11.5) 38. Qur’an, Keir collection, VII.4238 Page 16 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates 39. Qur’an, British Library, London, Add. 5548–51 (p.318) 40. Qur’an, Christie’s ‘2’, 1/5/2001, no. 19 41. Qur’an, British Library, London, Add. 18163 42. Jaunpur Anthology, British Library, London, Or. 4110 (Figure 11.1)

Illustrated Manuscripts and Literary Tastes Before coming to the stylistic characteristics of each group of manuscripts, let us briefly note what literary tastes and choices they reveal. Though all we have is a limited corpus, the first thing to note is that, with the exception of the Candāyan and the Qur’ans, it wholly consists of texts either originally written in Persian or of translations into Persian. It is mostly lyrical or epic narratives, in prose and in verse, that are chosen for illustration, whether the work is Persian, Indo-Persian, or in a vernacular language. There are also a few manuscripts with a scientific or more generally didactic aim that seem to require illustration, though they are rarer and only appear in the third group, chronologically later. The evidence of illustrated manuscripts thus supports Stefano Pellò’s argument in this volume (Chapter 6), that in the fifteenth century Persian literary culture in India was ‘provincialized’, by which he means that it was believed to be a distinct province of the wider Persian world. He points out three features, all represented in the corpus: (a) the local reproduction of earlier Persian classics (rather than contemporary poets); (b) a distinct linguistic and multilingual consciousness, represented by the production of dictionaries (also the subject of Chapter 5); (c) a definite multilingualism that can be also found elsewhere in the Persian oecumene at this time. Thus, most of the works illustrated are the Persian classics: Firdawsi’s Shāhnāma and Nizami’s Khamsa appear repeatedly in the second group but are also represented by two manuscripts in the third group.39 We should also notice that the Ḥamzānāma, a dastan that combined adventure, valour, romance, and magic and that became extremely popular in the Mughal period, already appears twice in our corpus.40 Lastly, as (p.319) many as five copies of an original vernacular work such as the Candāyan point to the continuing importance and popularity of this text throughout the period of the Sultanates, well after its composition in 1379. They also point to the wide courtly (and urban?) circulation of this Sufi take on a folk epic about the adventures and tribulations of two lovers (Figure 11.3, Plates 11.6, 11.7).41 Apart from those five manuscripts, the third group is the only one to present original and rare texts that were written or translated specifically in sultanate India. The variety of their topics should be underlined: they include a book entirely dedicated to the many pleasures of cooking, with recipes and indications about the preparation of specific products such as aphrodisiacs, of which the Ni‘matnāma from Mandu is the only known (Plate 11.5);42 a dictionary of rare words found in Persian literature, the Miftāḥ al-fużalā’, written in 873H/1468–9 Page 17 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates by Muhammad b. Da’ud b. Muhammad b. Mahmud Shadiabadi at the order of the sultan of Malwa, Nasir Shah Khalji (r. 1500–11) (see Chapter 5);43 and another manuscript from Mandu’s princely workshop: Muhammad Shadiabadi’s Persian translation of al-Jazari’s Kitāb fī ma’rifat al-ḥiyal al-handasiyya (Book of Knowledge of Ingenious Mechanical Devices)44 entitled ‘Ajā’ib al-Ṣanā’i‘ (The Wonders of Arts/Crafts), whose manuscript in the corpus is probably an autograph copy. A later series of works belongs to courts in the Deccan and testifies to the range of interests cultivated at these courts. Among them the Ta’rīf-i Ḥusayn Shāhī consists of, as the title says, narratives of court life and military campaigns during the reign of Husayn Nizam Shah, the ruler of Ahmadnagar (1554–65); a later manuscript entitled Zakhīr-i Khwārizmshāhī (The Khwarizmshah’s Treasures [of Medicine]) is a copy (p.320) of a medical encyclopaedia that was written in Persian by Zayn al-Din Abu Ibrahim Isma‘il al-Jurjani in the first half of the twelfth century;45 the Sindbadnāma is, as the title says, an adventurous tale of Sindbad’s voyages—this version of that very rare work is probably unique. The same is true also of the anonymous Nujūm al-‘ulūm (The Stars of the Sciences), a book of astronomy, astrology, and magic.46 Finally, we have a treatise on Indian music, known as Jawāhir al-mūsiqāt-i Muḥammadī (The Jewels of Music), written in Persian by a certain Shaykh ‘Abd al-Karim al-Jawnpuri. The illustrated Anvār-i Suhaylī (The Lights of Canopus) in this corpus, Husayn Va’iz al-Kashifi’s version of the Kalila-wa Dimna by Nasr Allah that became so spectacularly popular in Mughal times, is of special interest because it was produced only a few decades after it was composed and constitutes one of its oldest copies.47 As we noted at the beginning, most of the manuscripts produced in Sultanate India contained a high number of illustrations. That the majority of them were narratives may be the first explanation why images are so numerous—tales, romances, or epic poems have twists and turns and richness of action easily provide matter for illustrations. We also need to bear in mind that the social life of these manuscripts, their audience and circulation, is still an open question. We know that in medieval times a lot of illustrated manuscripts were kept in libraries and probably never read. It is therefore important to consider the illustrated book as an artistic work, an object. The illustrated manuscript could be considered like an external marker of wealth. At the same time, it should be pointed out that the fact that a manuscript was more highly illustrated does not necessarily mean that it was of a better quality. For example, the Berlin Ḥamzanāma, which was not commissioned by a prince and was probably made for a rich townsman, (p.321) contains 189 miniatures over 353 folios (i.e., 53.5 per cent of the folios are illustrated), whereas the Ni‘matnāma (Plate 11.5), whose origin in Mandu’s royal workshop is attested in the text, only contains fifty miniatures over 196 folios (25.5 per cent). Thus, the Page 18 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates number of illustrations in a manuscript is not automatically linked to the status of the patron. Nor does the number of illustrations depend a priori either on a specific school or on a specific group of manuscripts: the three least illustrated manuscripts in the whole corpus, that is to say Yūsuf-u Zulaykhā, Sharafnāma, and Ta’rīf-i ḥusayn Shāhī,48 were realized at very different times and in very different styles. The profusion of images cannot be systematically linked to the nature of the text, either. If such abundance of illustrations is easily explained for a dictionary like the Miftāḥ al-fużalā’, where the small vignettes complement the definition given in the text, other examples are more problematic. For instance, a cook-book like the Ni‘matnāma does not deal with a subject that lends itself particularly well to illustration. Nevertheless, the manuscript displays about fifty different representations of the ruler accompanied by women preparing various dishes. The second point to make is that unlike the copies of famous texts like the Shāhnāma or the Khamsa, for which there existed a precise iconographic tradition, some of the manuscripts of the third group deal with subjects in which the didactic function prevails over the narrative one and for which such a tradition did not exist. In fact, several of these works are rare, if not unique. Moreover, the fact that some episodes from works such as Amir Khusraw’s famous Khamsa that were seldom or ever illustrated in Persian manuscripts in Iran and Central Asia were illustrated in these Sultanate manuscripts shows the artistic autonomy of the sultanates.49 To conclude, the illustrative richness of the sultanate manuscripts should therefore be linked more to a taste particular to this period than to patronage or to the nature of the illustrated work, even if the latter was undoubtedly a factor. One interesting example of the importance given to the paintings in these preMughal manuscripts is that of the Roma (p.322) Khamsa, where images were even glued onto the text at the expense of the planned page layout and to the text itself.50

Illustrations in Sultanate Books Broadly speaking, paintings of the Sultanate period reveal a particular Indian manner in which some elements of Islamic iconography were perfectly assimilated while the overall style (or manner) was retained.51 In other words, the ‘Islamic manner’ was not misunderstood but voluntarily set aside. Moreover, Sultanate paintings are nourished both by contemporary Islamic and Jain artistic schools as well as by more archaic styles, as we shall see. The heterogeneity of these paintings, and that of the corpus as a whole, makes establishing a typology all the more complex, though the occasional repetition of certain stylistic combinations makes it possible to gather several manuscripts into a single group. However, the same production centre—for example Mandu at the cusp of

Page 19 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates the fifteenth and sixteenth centuries—and probably a same workshop could generate very dissimilar works. The illustrations from the first group resound with styles that were current in Iran or the Near East during the thirteenth and fourteenth centuries, while those from the third group draw largely upon the Persian schools of the fifteenth and sixteenth centuries. This means that some of the Sultanate manuscripts from the first half of the fifteenth century display some very old features while others adopted the current styles of the Persian schools quickly and effectively, with a chronological gap of rarely more than thirty years. This divergence was therefore probably a consequence of specific tastes and a certain conservatism rather than lack of innovation or contact. Among the archaizing features of the paintings in Groups 1 and 2 we can identify, first of all, the horizontal format, which had been (p.323) generally abandoned in Iran from the second half of the fourteenth century. This format is particularly noticeable in Amir Khusraw’s Khamsa: all the paintings, without exception, occupy the full width of the textual surface (jadwal) but never go beyond a third of its height. Another archaizing feature relates to the internal composition of the picture on a unique ground plane, without any horizon. The background of the illustration is spread with a monochromatic colour, usually red. The robust characters, with round heads drawn without any proportion to the rest of the small and stocky bodies, are nearly as high as the buildings. In Amir Khusraw’s Khamsa, as in other archaizing manuscripts, the characters wear a tiraz, a band of textile with a line of Arabic inscription, embroidered or painted, on the upper sleeves of a robe that adorned the dresses of honour worn by the high dignitaries or the sovereigns; some figures are surrounded by a clouded halo.52 In fact, the closeness between the Amir Khusraw paintings (Plate 11.2) and the Persian paintings dating from the first half of the fourteenth century, added to the very archaic script, allow us to postulate an early date for this manuscript, possibly around 1390 and thus contemporary to the Gwalior Qur’an, whose illumination is also partly inspired by the Persian manuscripts dated to between 1300 and 1350, as we shall see. Amir Khusraw’s Khamsa would then represent one of the stepping stones towards the art of the IndoIslamic book of the first half of the fifteenth century. The remains of the Rome Khamsa and the Banaras Shāhnāma (Plate 11.3) are perhaps also early, painted at the end of the fourteenth century. If this is true, it may signify that some of these manuscripts were executed before the Timurid sack of Delhi, and perhaps under the last Tughluq sultans.53 To go back to the general archaizing features of the manuscripts of Groups 1 and 2, the outdoor buildings have no façade, no volume, and are reduced to their main features: thus an arcade figures as a mosque, or a dome as a mausoleum. Indoor architecture is also reduced to a minimum: at best, the scene takes place under the broken or sometimes poly-foiled arch framing the illustration. The Page 20 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates empty spaces in (p.324) the composition are occupied by large flowered plants or small shrubs. The decoration is always minimal, reduced to a tree-plant, a circumvoluting cloud in the Chinese way, or sometimes rocks or plants on the ground. Finally, the general predilection for a bright clashing palette should also be remarked. Two particular issues arise in the context of Sultanate paintings. The first concerns the links between the art of the book in the Sultanates and the art of the book under the Mamluks. These may have occurred in two different ways. First, the Sultanate paintings datable to the fifteenth century are closely linked to the painting tradition of the Il-khanid period, just like Mamluk manuscripts. It is possible that groups of artists may have joined the Indian Sultanates or the Mamluk empire after having worked under Il-khanid patronage. Moreover, the strong commercial links between the Middle East and India lead us to assume that India was an outlet for the Middle East export of manuscripts, just as at the end of the fifteenth century and the beginning of the sixteenth it became an outlet for Shiraz manuscripts. By contrast, the Turkmen iconographical elements that appear in the Deccani paintings from the last quarter of the sixteenth century (Group 3) represent a special case: these elements had probably become stereotyped patterns that had been used in India for decades, probably inherited from the Indo-Islamic pictorial legacy rather than directly from contemporary Persian sources. A small part of the Sultanate iconography in Groups 1 to 4 evokes a more specifically Indian prosaic reality: tools such as cooking pots, clothes such as odhni and jama, elements of architecture such as chhatri, baoli, and chhajja are recurrent examples. The painters were thus able to reproduce various elements from their daily life into their works. These non-Islamic pictorial features raise the issue of the connection with the painting tradition of Jain manuscripts. Since Jain works have a colophon that mentions the date and the place of execution, it is possible to draw a direct parallel between the production of the two manuscript traditions, Jain and Islamic. This helps to refine our analysis, to suggest new dating and location by cross-checking, and finally to better conceptualize the originality of both. It appears that stylistic changes occurred in Jain manuscript painting in the middle of the fourteenth century. These changes have been attributed to the use of a new material—paper—that provided greater freedom of expression to the painters, but are more likely to have been (p.325) the result of influence from Islamic art. On the one hand, Indo-Islamic art seems to have had an impact on several Indian artistic traditions.54 On the other hand, Jains, settled in great numbers in Gujarat—a crucial point for trade between India and the rest of the world for centuries—might have had access to overseas objets d’art.55 Thus, Jain

Page 21 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates art was influenced by Islamic art in two ways: by the developing Indo-Islamic art and through contact with Muslim countries trading with India.56 One of the main sources for the renewal of Jain painting in the fifteenth century seems in fact to have been Persian art from the first half of the fourteenth century. In other words, the same archaizing features displayed by the Sultanate paintings of Groups 1 and 2 can be found in Jain paintings.57 This conservatism may have been the result of common (p.326) models or a convergence of taste. It is more than likely, though at present impossible to prove, that the same artists worked for Jain patrons and for the Sultans and their amirs/courtiers. The iconography changes but not the style, and we have some Islamic manuscripts painted in a ‘Jain style’.58 It is thus possible to postulate that Indo-Islamic paintings inherited archaizing elements that were shared by both Jain and IndoIslamic painting via Indian—most probably Jain—art. Observing carefully the Shāhnāma (Plate 11.4) or the Ḥamzanāma from the second group it is difficult to doubt such a hypothesis. Even a very common feature of Islamic iconography— that of the marvellous bestiary—is here reinterpreted, even re-elaborated, because of the complete lack of models.59 In terms of composition, Sultanate paintings provide a good example of the specificities of pictorial production in pre-Mughal India. Despite the variety in composition, they share some common rules. These rules were not formulated explicitly, nor were compositions modular as in some Persian or Mughal schools. Yet these common general principles can nevertheless be elucidated by an empirical observation of the corpus. Three different types of composition can be discerned: the first type concerns all the paintings of Groups 1 and 2 (except for the three Candāyans, by far the simplest). Of the eleven manuscripts, nine have an oblong or square format. This first type is based on the two main principles of axial symmetry and compartmentalization of the painting space. The space remains flat and linear, with a well-marked rectangular and ‘smashed’ feature. A second type of composition, always with a vertical format, is found in the third group of manuscripts. Given the complexity and variety of manuscripts in this group, it is difficult to reach any general conclusion about this type. The third type of composition consists of a vertical format divided into superposed registers, as found in the paintings of the five Candāyan manuscripts (Figure 11.3, Plates 11.6, 11.7). The third type of composition is more interesting, since the very different styles of the five copies of the Candāyan share defining principles of composition which cannot be explained by chance but rather (p.327) by an illustrative tradition. All the paintings, without exception, are closely linked to the nature of the text and occupy the whole page—the text appears only on the verso. The division of the space into registers may be explained as the attempt to follow the rule of illustrating the text even when the painter does not have to compose with the usual space on the page dedicated to his copy. By ascribing specific spaces to Page 22 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates particular characters, it becomes possible to render in a single picture events occurring in different places and at different times—the same principle as that of comic books. The more sophisticated Mumbai and Manchester Candāyans present a composition based on three recurrent rules (Plates 11.6 and 11.7): (a) superposition of registers; (b) tripartite division of space; and (c) ‘axiality’. In the paintings of the Mumbai manuscript, the vertical tripartite division appears more visibly than the composition on a central axis, while the principles of symmetry and ‘axiality’ are pronounced in the paintings on the Mumbai pages, and less noticeable in the Manchester manuscript. For example, the horizon is used as an iconographic convention simply to define the higher register—the third register—and to build the picture on this principle. In the same way as the horizon, architectural features are used regularly to provide the main lines of composition, and they highlight its rigour and repetitiveness. These principles, shared within each group, reveal a cohesion that would otherwise be difficult to notice at first glance. They highlight the particular features of Indo-Islamic production: powerful composition on an axis, more than in other Islamic paintings, and division of the picture into superimposed registers or horizontal compartments. If we take this into consideration, it becomes unsurprising to discover that these fundamental principles re-emerge in some of the paintings of the third group even though they appear to be closer to contemporary Persian models. We should nevertheless notice that these principles mainly concern the most ‘Indianized’ works of the group.

Illumination As far as the ornamentation of the Sultanate manuscripts is concerned, it is fair to say that for a large part it is limited and of quite poor quality. Many manuscripts have no ornamentation at all or only display a distinctive opening ornamentation known as sarlawh, meaning an illuminated ribbon that runs through the first lines of the text. The beautiful (p.328) developments in manuscript ornamentation that took place in Iran from the fourteenth century onwards have no parallel in the Indian codex. Paper decoration is almost nonexistent at this stage—no dyeing, no pages flecked with gold, no paper decorated with stencils—in other words, no work aimed at embellishing the virgin page. It is not that paper decoration was unknown in South Asia: J.P. Losty mentions a Nepalese manuscript dating from 1105 that uses yellow paper,60 and several manuscripts from the twelfth and thirteenth centuries, also from Nepal, were copied on black-blue paper.61 We could, then, have expected to find more frequently the same type of embellishing in the Sultanate manuscripts, but this is not so. It should also be pointed out that none of the illuminated manuscripts in the corpus is earlier than 1398, that is to say prior to the Gwalior Qur’an (Plate 11.1) —almost two centuries of the history of Indo-Islamic illumination thus remain to be discovered. Nevertheless, the examples present in the corpus—mainly from the fifteenth century—highlight the originality of Indo-Islamic illumination in this Page 23 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates period. Illumination sometimes drew upon older reference styles instead of on elements current in the Persian art of the fifteenth century, such as those created by the Timurid and Turkmen schools, whose figurative paintings were after all one of the main sources of inspiration for Indo-Islamic art from the end of the fifteenth century. The illumination in most manuscripts of the first group— such as the Amir Khusraw’s Khamsa and the Qur’ans—is instead based on earlier patterns originating from various repertories (mainly Mamluk, Il-khanid and Inju), though they always appear combined in very original ways. The only manuscripts to display a style of ornamentation that can be linked to the Persian manuscripts of the fifteenth century and to manuscripts of the sixteenth century, that is, to their contemporaries, are those of the third group. This suggests that the Malwa Sultan cultivated direct links with Shiraz and perhaps with Herat. (p.329) One group of manuscripts where illumination was constantly employed was, not surprisingly, the Qur’ans. Once again, Qur’an manuscript decorations may be usefully divided into three distinctive groups so as to highlight stylistic similarities. The first two groups only include manuscripts copied in the bihari script. The first group consists of only two works—the Aga Khan Collection Qur’an (dated 801H/1399, Plate 11.1) and the Bijapur Museum Qur’an (dated 888H/1483)—which differ in ornamentation from of the other Qur’ans written in bihari. Not only are their patterns different, the ornamentation is also richer and wider, with profuse, singular, and particularly well-executed decorations. A hundred years apart, the ornamentation of these two Qur’ans is not inspired by the same iconographic repertories, though. The illumination of the Aga Khan Qur’an is based on a mixture of ornamentations coming from Iran and maybe from the Near East during the first half of the fourteenth century, while the Bijapur Museum manuscript shows some more specifically Indian compositions and patterns. The second group comprises the rest of the Qur’ans written in bihari (nos 32–9 in my corpus). They all are constructed along a quite sober model and show a specific and repetitive ornamentation that is distinctive of the pre-Mughal India palette and predilection for circular figures, patterns, and frames.62 Of the three manuscripts comprising the third group (nos 41–2), one is in fact not a Qur’an but a worldly book, the British Library Jaunpur Anthology (no. 42, Figure 11.1). Although these three books present rather dissimilar illumination, they nevertheless have several points in common with the preceding groups. In fact, the Anthology’s illumination shares similarities not only with the Aga Khan Qur’an but also with the Qur’ans written in bihari and displays a low level of finishing that is typical of the second group. So though this manuscript is very close in time to the Aga Khan Qur’an, it could constitute the missing link between it (and all the production that immediately (p.330) precedes it) and the fifteenth-century Qur’ans written in bihari of the second group. In other

Page 24 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates words, the Jaunpur Anthology could testify to the persistent nature of the specific ornamental style of the Sultanate books.63 The Qur’ans of the first two groups were all copied in the bihari script. This cursive calligraphy is peculiar to the Indian sultanates and might have appeared in the Jaunpur region, though the denomination bihari itself probably appeared later.64 The bihari script might be compared to the scripts used in chancelleries, like the one appearing on a farman by Muhammad b. Tughluq,65 though this was favoured for copying religious texts and no example of it is known in a worldly context. In fact, while the use of bihari does not always imply stylistic similitude in the illumination of the various manuscripts, we may postulate that the majority of the second group of Qur’ans used the same repertoire. Specific illuminating and page settings are used almost systematically in (p.331) this type of manuscript. In fact, this group of Qur’ans of ‘ordinary make’ shows remarkable coherence and homogeneity when compared to the rest of the Qur’an production. The very strong stylistic links between their illuminations, beside the fact that they use the same script and a similar page setting, raise a number of questions: was there a large-scale Indian commercial production of Qur’ans? Were these Qur’ans executed by and for particular religious institutions—a hypothesis that could explain the conservatism of their formal elements?66 How long did this production last, since the Mughal Qur’ans are substantially different? Should we infer, since the oldest specimens are by far the finest, that there was a decline in illumination? Or was their quality linked to the social status of their patrons and/or recipients? I would argue that the uneven quality of the Qur’ans in bihari is to be attributed to stylistic evolution during this period rather than to a change in the type of production from ‘courtly’ to ‘urban’. For example, the Qur’an Bnf Arabe 7260 (Figure 11.5) must have been executed in the middle of the sixteenth century, whereas the Soustiel-David/Taipei Qur’an is more easily datable to the end of the fifteenth century. As a matter of fact, Arabe 7260 shows some features of an exhausted, less refined style, and the peculiar elements of the Qur’ans written in bihari appear stereotyped. A possible entry into this question is offered by the fact that these Qur’ans also present many ornamental similarities with a group of talismanic tunics, also ascribed to the Indian sultanates.67 A Qur’an, copied in small letters (ghubari) on a roll made of textile for talismanic use, ornamented with decoration identical to that of the tunic and bearing in chronogram the date 798H/1395, might confirm the earlier dating.68 The Qur’ans written in bihari, like the tunics, were produced in large numbers (though not as large as the tunics) on a constant model over a long period. Did the context of production, the commissioners and the craftsmen who executed these two products have the same origin? Many are the questions arising from this aspect of book production and its uses: should we

Page 25 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates consider the possibility of a talismanic function, especially given the fact that divination books (falnama) are attached to several (p.332) (p.333) Qur’ans written in bihari as well as to the Gwalior Qur’an, while they were otherwise uncommon in the Muslim world before the sixteenth century or even later? Or were those Qur’ans meant for a more official and exoteric use while, by contrast, the tunics involved an esoteric reading and perhaps the existence of magical practices in some orthodox circles?

Another peculiarity of the Sultanate production is the Qur’ans with ‘contemporary composition’, in whose margins we find words from the principal text but with different declensions, which imply the knowledge of different canonical readings of the Qur’anic text. This supports my supposition that these manuscripts may have been copied in religious centres, including schools where these

Figure 11.5 Qur’an, sixteenth century (Bibliothèque nationale de France, Paris, Acc. No. Arabe 7260, f. 46 recto).

different canonical readings (qira’a) were taught.69 It would be necessary to interrogate the links between such productions and Sufi circles, which as we know were well rooted in the Indian world at that time. The careful recovery and analysis of the Qur’ans from the Sultanates, already begun but far from complete, has suggested some lines of enquiry that are interesting and promising. As we have seen, grouping together all Qur’ans with less elaborate illuminating, of ‘ordinary make’, raises its own set of questions relating to their production and use. We may also consider the Qur’ans with the most elaborate decorations as a separate group: first of all, the Gwalior Qur’an (Plate 11.1), then a monumental Qur’an without location or date,70 a two-volume Qur’an, also without location or date (collection Nasser D. Khalili, QUR237),71 and lastly a Qur’an (dated 1483) now kept at the Bijapur Archaeological Museum.72 Even if the dispersion of the material makes research particularly

Page 26 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates challenging, it is now imperative to study systematically and exhaustively the Qur’ans concerned. Finally, these Qur’ans raise art historical questions of their own. Though it is impossible to attest the direct sources of inspirations for (p.334) manuscripts like the Gwalior Qur’an, it is interesting to compare its date of execution with the Persian or Mamluk ones executed in the first half of the fourteenth century. A Persian Qur’an kept in the Khalili collection (QUR 18) shares several common elements with it and is datable with certainty to between 1336 and 1337, while the Indian Qur’an is dated 1399. Two hypotheses offer themselves here: either the Gwalior manuscript is a late specimen of the fusion of outlandish styles used in India for decades and partially integrated into the indigenous artistic vocabulary; or, on the contrary, it might be one of the first examples of a specific Indian school based on Islamic models dating from around 1350. The survival of older elements might have been important in India at this precise time of great political flux. Nevertheless, the Gwalior Qur’an remains a special case deserving a complete study, which goes beyond the scope of this chapter.

Some Mapping and Conclusions The study of the illustrated manuscripts of the period of the Sultanates raises a series of questions that have so far remained unanswered. By maintaining earlier styles—unless it was a return to these styles—these paintings appear archaizing and still enigmatic. Equally, it remains difficult to establish the date of a codex with any certainty even after very careful study. This is made all the more difficult by the heterogeneity of the corpus and the lack of evidence that would help identify the sites of production in order to address to the difficult question of patronage. Nevertheless, I would like here to summarize our information about the different Sultanates before offering some general remarks on the evolution of Sultanate paintings and on possible pathways for future investigation. The Delhi Sultanate: Even if Delhi is not mentioned in any colophon of any Islamic manuscript,73 we must suppose that at least one, and perhaps more, manuscript workshop (kitab-khana) existed because of the central role Delhi maintained until the end of the fourteenth century. The Sultanate of Malwa: The Khalji Sultanate of Malwa is the only attested example of a princely kitab-khana in the pre-Mughal period. (p.335) The presence of early creative activity regarding the art of the book in the history of this Sultanate is attested by the Islamic manuscripts produced in Mandu that are dated or datable to between the end of the fifteenth and the beginning of the sixteenth century, as well as by a Kalpasūtra dated 1439.74 The Sultanate of Jaunpur: As suggested in the foreword of the British Library Anthology (Figure 11.1), this manuscript was probably executed in Jaunpur at the beginning of the fifteenth century. A Kalpasūtra executed in Jaunpur in 1465 Page 27 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates attests to the existence of the production of Jain illustrated manuscripts in this region during the second half of the fifteenth century.75 It is more than likely that the production of illuminated books under the Sharqis (1394–1479) was not limited to the Jain community. The Jaunpur Sultanate reached its peak during the thirty-eight years of Shams al-Din Ibrahim’s reign (1402–40) and was famous for having attracted within its frontiers both local and distant artists and scholars.76 Moreover, the whole of Bihar had probably been an active centre of manuscript production for a long time. It is indeed possible that there was no gap between the last Pala ruler and the first Muslim rulers in this respect—a hypothesis supported by the persistence of several pre-Islamic Indian elements in the iconography. All the Jain and Islamic manuscripts that can be ascribed to this region display links with the Persian painting of the thirteenth and early fourteenth centuries. This point also suggests a long pictorial tradition. Consequently, we may wonder whether the manuscripts from Groups 1 and 2— which display both strong Jain features and many elements of archaizing Persian painting—also came from this region. The Sultanate of Bengal: As the British Library Sharafnāma indicates—the Bengali Sultan is named as its recipient—there was at least one place of production of illustrated manuscripts in Bengal.77 This part of India, no doubt, played a role in the production of manuscripts from at least the fifteenth century, and perhaps until Bengal was annexed in 1539 by the Afghan Sher Shah Sur, who used it as a base to push Humayun’s (p.336) armies out of India. At the end of the fifteenth century, the Bengal Sultan annexed neighbouring Bihar and gave asylum to Husayn Shah from Jaunpur after he was defeated by the Lodis. One can imagine that the two regions influenced each other, and that artists from Jaunpur found refuge in Bengal along with their rulers, just as they did in Mandu after the Lodis took over the Sharqi sultanate. However, given the limited number of elements in our possession, it is still difficult to establish stylistic comparisons between the production of illustrated books in Bengal and in Bihar. The Sultanate of Gujarat: While there is ample evidence that the Gujarat sultans had a library, commissioned books and translations, and had formal links with the Mamluks, it appears that they did not patronize the art of the book.78 Although one Jain illustrated manuscript can be ascribed with certainty to Gujarat—the Vasanta vilāsa, executed in 1451–2 in Ahmadabad—there is no Islamic illustrated manuscript produced in that Sultanate.79 However, because of the location and trade relationships of the Sultanate with the rest of the Islamic world, it is possible that books were imported and others executed on the basis of imported models.

Page 28 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates The Sultanates of the Deccan: The first known examples of illustrated manuscripts ascribed to the Deccan date back to the end of the sixteenth century, when three distinct places of production existed in Ahmadnagar, Golkonda, and Bijapur. Some schools, such as that of Bijapur, had a very particular style before mingling with the dominating Mughal influences in the seventeenth century.80 As for the Bahmanid kingdom (1347–1527) and its capital cities of Gulbarga and Bidar, (p.337) textual documents such as Abu al-Da‘i Ya‘qub b. Hasan b. Shaykh Siraj al-Husayni al-Shirazi’s treatise on calligraphy, the Tuḥfat al-muḥibbīn, executed in 858H/1454 in Muhammadabad (Bidar), attest to the importance of princely patronage.81 This treatise is all the more important since it is the only known technical text about the arts written in preMughal India. For our purposes, this text is far more interesting for its context of redaction than for its content, since the content concerns Iran and not India.82 The Farsi translation of Ibn Badis’s ‘Umdat al-kuttāb—probably the oldest known copy of this treatise until now—is another example of the interest of the Bahmanid sultans in calligraphy. The introduction is dedicated to the sultan Mahmud Shah Bahmani (1482–1518).83 As this brief survey shows, many aspects regarding the art of the book in Sultanate India still await clarification. The production context of these manuscripts is still for the most part enigmatic, despite their relative ‘lateness’ in the context of the history of Islamic manuscripts. As we saw, the rare colophons are equally late, while seals or second-hand notations (p.338) added by librarians or owners are rare. What we are left with is the stylistic analysis of the paintings and calligraphies decorating the books, which undoubtedly has much to offer as an analytical tool. Although the miniatures and the illumination do not rely on the same stylistic sources, this chapter has shown that we can still draw a parallel between their respective evolutions. Even if illumination appears to have been used much more sparingly than illumination in Sultanate manuscripts and to have relied much less on Indian repertoires, illustration and illumination can be usefully contrasted and jointly considered in order to generate research questions and trace their historical developments. Both illumination and illustration went through two similar consecutive phases. In the first period, datable between the fourteenth and fifteenth centuries, some ‘patchworks’ were produced, that is, works that clearly show influences from very different styles at the same time. In the second period that coincides with the sixteenth century, these influences fused together, and the first signs of a specific and independent style emerged. Some ‘series’ illustrate this point: the Qur’ans of ‘ordinary make’ written in bihari script are a good example of illumination, while as far as miniatures are concerned, it is impossible not to think of the various Candāyans. Even if they markedly differ in style among themselves during the fifteenth century, they constitute a very coherent and remarkably closed group in the following century.

Page 29 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates It would, of course, be reductive, not to say dangerous, to conclude a diachronic pictorial analysis finishing with the suggestion that there was a linear and qualitative evolution between the works. In fact, it rather seems that the art of the Islamic book in India encountered several difficulties for a long time before it found its own, accomplished style. Many reasons could be given for this: from the orders that patrons gave according to their tastes, possibly influenced by fashions in India or elsewhere, to the rendering of those tastes by the artists depending on their origins—foreign or indigenous—and education. The integration of Persian masters into the Mughal imperial workshops early on marked the decline of the previous styles. This allows me to suggest, though not to prove, that we consider the later Candāyans as the most mature stylistic realization of Sultanate paintings. These forms survived only in minor and provincial works before they quickly vanished in the seventeenth century. This was the swansong of the art of the Sultanate book. Notes:

(1) These mentions were gathered and translated by Simon Digby in a very complete paper published in 1967, ‘The Literary Evidence for Painting in the Delhi Sultanate’, Bulletin of the American Academy of Benares, 1, 1967, pp. 47– 8. M.A. Chaghataï had written a monograph (Painting During the Sultanate Period (C.E. 712–1575), Lahore: Kitab Khana-i Nauras, 1963) on the same subject a few years earlier, but his translations are sometimes a little approximate. B.N. Goswamy’s study ‘In the Sultan’s Shadow: Pre-Mughal Painting in and around Delhi’ (in Delhi Through the Ages: Essays in Urban History, Culture and Society, R.E. Frykenberg (ed.), Delhi: Oxford University Press, 1986, pp. 129–42) has brought forth some additional material. (2) Corpus no. 25, third group. For an analysis of this text, see Emma J. Flatt, ‘Courtly Culture in the Indo-Persian States of the Medieval Deccan, 1450–1600’, PhD thesis, School of Oriental and African Studies, University of London, 2009. (3) É. Brac de la Perrière, ‘Les Manuscrits à peintures dans l’Inde des sultanats: l’exemple de la Khamsa dispersée d’Amir Khosrow Dehlavi, c. 1450’, Arts Asiatiques, 56, 2001: 35–6. Since this paper, J. Seyller has published a new list of the illustrated folios from that specific manuscript and added a new page that was not mentioned in my paper. See ‘Alexander’s men bearing his coffin’ (Aga Khan collection, AKM00015/ex-Ind. M.1b) in J. Seyller, Pearls of the Parrot of India: The Walters Art Museum Khamsa of Amīr Khusraw of Delhi, Baltimore: The Walters Art Museum, 2001, pp. 145–6. (4) Illumination is the decoration of the book, its ornamentation. It is unrelated to the text, for example, consisting of geometric or vegetal patterns on the frontispiece. The miniatures or illustrations illustrate the text, directly or indirectly.

Page 30 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (5) É. Brac de la Perrière, L’Art du livre dans l’Inde des sultanats, Paris: Presses de l’Université Paris-Sorbonne, 2008, pp. 297–308. As far as the Deccan is concerned, production stretches from the second half of the sixteenth century to the eighteenth century and only briefly coincides with our period. We should therefore consider only those Deccani manuscripts whose paintings contain elements that can be found in works produced in the Sultanates of central and northern India during the first half of the sixteenth century, in particular Malwa. The other Deccani manuscripts should properly be considered in relation to the arts of the Mughal book. The most ancient illustrated manuscript from the Deccan is said to be a Ta‘rīf-i Ḥusayn Shāhī, which could be dated from 1565 to 1567, located in Ahmadnagar, and presently kept at Bharat Itihas Sanshodhak Mandal (Poona). Ahmadnagar was taken by the Mughals in 1600, whereas Golkonda and Bijapur remained independent till 1686–7. M. Zebrowski, Deccani Painting, London: Sotheby Publications, 1983; ‘Transformations in Seventeenth Century Deccani Painting at Bijapur’, Chhavi, 2, 1981: 170–81; and also ‘Painting’, in Islamic Heritage of the Deccan, G. Michell (ed.), Mumbai: Marg Publications, 1986, pp. 92–109. (6) This Qur’an is well known to specialists and has been published in various works, and is now the subject of a French research programme at the CNRS (UMR 8173 Medieval Islam research laboratory) under my direction. The name of the Gwalior fortress (qal‘e Kalyûr) is clearly mentioned in the final colophon, where the name of the copyist, Mahmud Sha‘ban, is also given. Éloïse Brac de la Perrière, Frantz Chaigne, and Mathilde Cruvelier, ‘The Qur’an of Gwalior, Kaleidoscope of the Arts of the Book’, in Treasures of the Aga Khan Museum: Arts of the Book and Calligraphy, Istanbul: Sabanci University, Sakip Sabanci Museum, 2010, pp. 114–23. (7) S. Digby, ‘Before Timur Came: Provincialization of the Delhi Sultanate through the Fourteenth Century’, Journal of the Economic and Social History of the Orient, 47(3), 2004: 326–7. (8) A second-hand annotation dated 935H/1529 gives us the terminus ante quem. J.P. Losty, The Art of the Book in India, London: British Library, 1982, no. 19, p. 56. (9) The two manuscripts are a Būstān kept at the National Museum in New Delhi (48.6/4) and a Sharafnāma kept at the British Library (Or.13836); see Brac de la Perrière (2008), p. 280, no. 17 and p. 282, no. 20. (10) Zakhīr-i Khwārizmshāhī, Chester Beatty Library, Ind. Ms. 30; Khusraw-u Shīrīn, Khuda Bakhsh Library (no. 499). See Brac de la Perrière (2008), p. 283 no. 22 and p. 285 no. 25. (11) Brac de la Perrière (2008), pp. 37–38.

Page 31 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (12) Brac de la Perrière (2008), pp. 48–9. (13) N. Titley, Persian Miniature Painting and its Influence on the Arts of Turkey and India, London: British Library, 1983, pp. 161–85. (14) Losty (1982), pp. 41–54. (15) Though the Candāyan is a poem in Avadhi, and not in Arabic or Persian, the fact that it was written in the Arabic script allows us to include its manuscripts in the corpus of Islamic art, for this is the usual criterion in the codicological field, see F. Déroche, Manuel de codicologie des manuscrits en caractères arabes, Paris: Bibliothèque nationale de France, 2000. For considerations of script in relation to production and circulation, see the Introduction to this volume. (16) I have deliberately set aside the British Library’s Shāhnāma (Or. 1403) that displays features characteristic of Mozaffarid paintings from northern Iran; see Ivan Stchoukine, ‘Origine indienne d’un manuscrit persan achevé en 844 A.H.’, Syria, Paris: Librairie orientaliste Paul Geuthner, 46, 1969, pp. 105–14; N. Titley (1983), pp. 164–6; K. Ådahl, A Khamsa of Nizāmī of 1439: Origin of the Miniatures: A Presentation and Analysis, Upsal: Almqvist & Wiksell International, 1981. (17) Its origin needs to be confirmed, because the folios were cut out and glued onto a new support. The paintings by themselves do not offer enough evidence to identify the text. (18) Some of them are published in K. Khandalavala and M. Chandra, New Documents of Indian Painting: A Reappraisal, Mumbai: Board of Trustees of the Prince of Wales Museum of Western India, 1969; M. Chandra, Studies in Early Indian Painting, Mumbai: Asia Publishing House, 1970, fig. 45, Pl. X; N. Nath and K. Khandalavala, ‘Illustrated Islamic Manuscripts’, in An Age of Splendour: Islamic Art in India, S. Doshi and K. Khandalavala (eds), Mumbai: Marg Publications, 1983, no. 11; Islamic Heritage of India: Special Exhibition to Honour the 1400th Anniversary Celebration of the Hijra Era, New Delhi: National Museum, 1981, no. 129, p. 30. (19) Previously classified under the inventory number Caetani 36; see A.M. Piemontese, Catalogo dei manoscritti persiani conservati nelle biblioteche d’Italia, Roma: Instituto Poligrafico e Zecca dello Stato, 1989, pp. 255–6. (20) Sam Fogg, Islamic Manuscripts no. 22, September 2000, lot 28; S.C. Welch and M.C. Beach, Gods, Thrones and Peacocks: Northern Indian Painting from Two Traditions, Fifteenth to Nineteenth Centuries, New York: The Asia Society, 1965, fig. 2; Khandalavala and Chandra (1969), p. 47. Two other folios from that manuscript are kept in the Keir Collection, while two others previously Page 32 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates belonging to E. Binney are located in the San Diego Museum, and a last one is at the Victoria and Albert Museum. (21) Paris-Drouot, 5 May 1987, no. 36. (22) Christie’s, London, 1 May 2001, no. 87. (23) S. Czuma, Indian Art from the George P. Bickford Collection, Cleveland: The Cleveland Museum of Art, 1975, no. 43; Sotheby’s, Indian, Tibetan, Nepalese Thai, Khmer and Javanese Art, 20-21/09/1985, no. 447; C.-P. Haase, J. Kröger, and U. Lienert (eds), Oriental Splendour: Islamic Art from German Private Collections, Hamburg: Temmen Publications, 1993, no. 93, p. 253. (24) On this scattered Shāhnāma, see B.N. Goswamy, A Jainesque Sultanate Shahnama and the Context of Pre-Mughal Painting in India, Zurich: Museum Rietberg, 1988. When it appeared on the art market in summer 1987, the manuscript had been disassembled in order to sell the folios separately. Seven folios remain in the Museum Rietberg in Zurich; four in Berlin, Museum für Indische Kunst; two others are located in Copenhagen’s David Collection; two others in the Musée Guimet, Paris. The rest of the folios are scattered in different private collections. (25) This manuscript, originally containing twenty-nine pictures, would have been entrusted to the Prince of Wales Museum (now the Chhatrapati Shivaji Maharaj Vastu Sangrahalaya of Mumbai) in the 1960s; because the Museum could not afford to photograph it, the manuscript then vanished without having been completely photographed (K. Khandalavala and M. Chandra [1969], p. 47 and Khandalavala and Chandra, ‘A Ms. of the Sikandar Nāma’, Bulletin of the Prince of Wales Museum, 7, 1959–62: 31). Since all the information concerning this manuscript is drawn from the articles published by K. Khandalavala, I prefer to keep the title he gave to this specific manuscript, Sikandarnāma, whereas the common transcription Iskandarnāma is used for the other manuscripts. (26) According to P.L. Gupta this codex, now comprising 141 folios, originally comprised four times more; see P.L. Gupta, ‘The Berlin Chandāyan Codex’, Chhavi, 2, 1981: 305. (27) Basil Gray offers a complete review of work on that manuscript in ‘The Lahore Laur-Chandā Pages Thirty Years After’, Chhavi, 2, 1981: 5–9. (28) The paintings of this text are very similar to the Mamluk copies of the same text. The manuscript was produced in a Mandu workshop on 4 shavvāl 914H/ 1509, and this element undoubtedly links it to this third group. (29) This point should invite us to investigate more closely the nature of the book bindings of all the pre-Mughal manuscripts, an issue I discuss in Brac de la Perrière (2008), pp. 109–12. Page 33 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (30) I only indicate the dates that cannot be doubted. (31) Norah Titley (trans.), The Ni‘matnāma Manuscript of the Sultans of Mandu: The Sultan’s Book of Delights, London: Routledge, 2005. (32) The folio 229 bears the date of an inspection note ‘rabī‘ al-thānī 908H/1502’. (33) As shown by an isolated page kept at the National Museum in New Delhi, some other manuscripts that could be assimilated to this group have probably existed: see Brac de la Perrière (2008), pp. 78–81. (34) Sixty-eight folios detached from the first codex are kept at Mumbai’s Chhatrapati Shivaji Maharaj Vastu Sangrahalaya in Mumbai (formerly the Prince of Wales Museum). Some other pages are scattered among different museums— for instance, the Virginia Museum of Fine Arts in Richmond and San Francisco’s Asian Art Museum—and different private collections (Nasser D. Khalili); they regularly appear on the art market. (35) Even if about a quarter of the text is missing, this copy, with 318 folios, presents the most complete version of the text. Only thirty-three folios from the whole manuscript display the text both on recto and verso. Even if its style is less rich and more rigid than the preceding copy, this manuscript nevertheless has great artistic value. (36) Brac de la Perrière (2008), p. 82. For a complete bibliography on the Qur’ans, see pp. 289–96. The Keir Collection Qur’an is the only one not to appear in this list. (37) This Qur’an, which I designated as Soustiel-David’s Qur’an in Brac de la Perrière (2008), has been bought by the Taipei Museum (information given by the Sam Fogg gallery). The inventory number is unknown. (38) I would like to express here my thanks to Christiane Gruber for having informed me of the existence of this Qur’an. (39) These manuscripts are the Sharafnāma, first part of the Iskandarnāma in the Nizami’s Khamsa, relating the victories of Alexander the Great, and Hatifi Dakkani’s Khusraw wa Shīrīn. (40) Ms Or. fol 4181, Staatsbibliothek Preussischer Kulturbesitz in Berlin and no. 35, N.C. Mehta Museum, Ahmadabad. (41) On the Candāyan, see Naseem Hines, Maulana Daud’s Candayan: A Critical Study, New Delhi: Manohar, 2009, and the numerous works by Shyam Manohar Pandey.

Page 34 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (42) Because the first four folios of this manuscript were added later, the title attributed to the manuscript is the one inscribed on folio 162 b. (43) From the name of the city of Mandu, since Shādiābād means the ‘city of joy’. (44) This is a technical manual written by al-Jazari, a late twelfth-century engineer; it was edited and translated by D.R. Hill as The Book of Knowledge of Ingenious Mechanical Devices, Dordrecht: Reidel, 1974. (45) On this text, see L. Leach, Mughal and Other Indian Paintings from the Chester Beatty Library, London: Scorpion Cavendish, 1995, pp. 889–90. (46) According to J.P. Losty, only two copies of this work still exist: this one of the corpus and another dating from the seventeenth century and kept in the same collection. Losty (1982), p. 72. For a description and discussion of its contents, see Flatt (2009). (47) Even if the date of the manuscript was rewritten, it remains acceptable. See J. Seyller, ‘Painter’s Directions in Early Indian Painting’, Artibus Asiae, 59(3/4), 2000: 303–4. (48) Although this manuscript is not complete, it seems improbable that it originally contained many more than twelve illustrations. (49) Brac de la Perrière (2001), pp. 35–6 and Brac de la Perrière (2008), pp. 173– 4. (50) Small illustrations were later added to this complete manuscript of 327 folios. Several virgin spaces indicate that this manuscript was supposed to be illustrated with paintings of a larger format. Some illustrations that can be indubitably linked to pre-Mughal India by the style, have been sometimes directly glued onto the text; hidden, this one has been carefully recopied in the margin. (51) This iconography finds its origins in the Arabic world (Egypt and Syria) and the Iranian world jointly, so it would be reductive to use the ‘Persianate’ in this context. (52) See the reproductions in Brac de la Perrière (2001), no. 1 (London, British Library, Add. Or. 4685) and 8 (Chicago, The Art Institute of Chicago, 1962–64). (53) The historical context, the place of the Delhi Sultanate and the importance of Tughluq patronage in the fourteenth century support this hypothesis. (54) S. Doshi, ‘Islamic Elements in Jain Manuscript Illustration’, in An Age of Splendour: Islamic Art in India, S. Doshi and K. Khandalavala (eds), Mumbai: Marg Publications, 1983, pp. 114–21. Page 35 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (55) Gujarat, under Muslim domination since the beginning of the fourteenth century, had been a crucial point for trade between India and the rest of the world since it was integrated into Ashoka’s empire in the third century BCE. On the history of Gujarat under Muslim domination, see S.C. Misra, The Rise of Muslim Power in Gujarat—A History of Gujarat from 1298 to 1442, London: Asia Publishing House, 1963; and J. Chaube, History of Gujarat Kingdom, 1458–1537, New Delhi: Munshiram Manoharlal, 1973. See also S. Sheikh, Forging a Region: Sultans, Traders, and Pilgrims in Gujarat, 1200–1500, New Delhi: Oxford University Press, 2010. On the Muslim communities of Gujarat before the Sultanate, see Z.A. Desai, ‘Muslims in the 13th Century Gujarat, as Known from Arabic Inscriptions’, Journal of the Oriental Institute, Baroda,10(4), June 1961: 351–64. (56) É. Brac de la Perrière, ‘Du Caire à Mandu: transmission et circulation des modèles dans l’Inde des sultanats’, in Écrits et culture en Asie centrale et dans le monde turco-iranien, Xe-XIXe siècles/Writing and Culture in Central Asia and in the Turko-Iranian World, 10th–19th c., F. Richard and M. Szuppe (eds), Paris: Cahiers de Studia Iranica, 2009a. (57) The very few researchers who have examined these problems, such as H. Goetz, have provided only insufficient explanations. See H. Goetz, ‘Vestiges of Muslim Painting under the Sultâns of Gujarât (15th–16th centuries A.D.)’, Journal of the Gujarat Research Society, Bombay, 16(3), July 1954: 216, and ‘Decline and Rebirth of Medieval Indian Art, Marg 4(2), 1950: 36–48. (58) See Goswamy (1988). (59) For other manuscripts, including those in the first group, the question is more complex: the elements shared with the Jain manuscripts could have had a common source but then have been reinterpreted differently depending on whether the production context was Islamic or Jain. (60) Goswamy (1988), p. 35. (61) All three manuscripts are kept at Calcutta’s Ashutosh Museum; two are reproduced in J. Trier, Ancient Paper of Nepal, Copenhagen: The Royal Library, 1972, no. 118 and 119. The first is dated N.S. 340/1220 and the second one dates from the thirteenth century. Cf. manuscript Or. 13971 kept at British Library, dated 1185, reproduced in Losty (1982), pl. VI, no. 11. (62) Brac de la Perrière (2008), pp. 162–5. On this group of Qur’an manuscripts, see É. Brac de la Perrière, ‘Bihârî et naskhî-dîwânî: remarques sur deux calligraphies de l’Inde des sultanats’, in Écriture, calligraphie et peinture, A.L. Udovitch and H. Touati (eds), Studia Islamica, 96, 2003: 81–93. See also my ‘Les

Page 36 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates tuniques talismaniques indiennes d’époque pré-moghole et moghole à la lumière d’un group de Corans en écriture bihâri’, Journal Asiatique, 1, 2009b: 57–81. (63) The two other manuscripts belonging to the third group raise quite different issues. The first one, which appeared on the art market on 1 May 2001 (Christie’s, 1 May 2001, no. 19), is medium sized, in muhaqqaq script. It is quite richly illuminated by patterns realized with evident dexterity. Its palette is based essentially on gold, red, black, and royal blue; the titles are penned in a nice white thuluth. The second one, kept at the British Library (Add. 18163), is copied in a script close to both thuluth and muhaqqaq. A seal certifies that it belonged to the Sultan of Gujarat, Mahmud Shah Begada (1488). This work is an important recomposed book, with some parts added in the nineteenth century; its ornamental programme is unfinished, as testified by the sketched decorations that remain empty of painting; only folios ff. 394a–490a in the whole codex show illuminating that can be ascribed to the Sultanate period. To see some reproductions of this Qur’an’s illuminated pages, see M. Lings and Y.H. Safadi, The Qur’ân, London: British Library/The World of Islam Publishing Company, 1976, p. 83, and Losty (1982), p. 38. These last two examples can be considered marginal and offer very few elements for an overall analysis. (64) For a description of the bihari script, see S. Blair, Islamic Calligraphy, Edinburgh: Edinburgh University Press, 2006, pp. 386–7. (65) S. Digby, ‘The Mughals and the Jogis of Jakhbar. By B.N. Goswamy and J.S. Grewal’, Journal of the Royal Asiatic Society of Great Britain and Ireland, 3–4, 1968: 196; also see B.W. Robinson, E.J. Grube, G.M. Meredith-Owens, and R.W. Skelton, Islamic Painting and the Arts of the Book: The Keir Collection, London: Faber and Faber Limited, 1976, pp. 283–4, and Losty (1982), no. 17. (66) Brac de la Perrière (2003). (67) Brac de la Perrière (2009b), pp. 57–81. (68) Brac de la Perrière (2009b), p. 73. (69) Brac de la Perrière (2003). (70) Sam Fogg, Islamic manuscripts, September 2001, no. 14. (71) D. James, After Timur: Qur’ans of the 15th and 16th Centuries. The Nasser D. Khalili Collection of Islamic Art, London: Azimuth for the Nour Foundation, 3, 1992, no. 28. (72) S.C. Welch, India: Art and Culture, 1300–1900, New York/Middletown: The Metropolitan Museum of Art/Grantha, in association with Mapin and PretelVerlag, 1985, no. 71.

Page 37 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates (73) Palam, near Delhi, appears in the colophon of a Mahāpurāṇa dated 1540; Khandalavala and Chandra (1969): 69–70; see also Chapter 12 in this volume. (74) Khandalavala and Chandra (1969): 17 ff.; Losty (1982), no. 28, p. 60. (75) Losty (1982), pp. 23 ff. (76) M. Muhammed Saeed, The Sultanate of Jaunpur: A Political and Cultural History, Karachi: University of Karachi, 1972. (77) Brac de la Perrière (2008), p. 282. (78) For a full list of Arabic works commissioned by, translated, or composed on the orders of the Gujarat sultans, see M.A. Quraishi, Muslim Education and Learning in Gujarat, Baroda: Maharaja Sayajirao University of Baroda, 1972. Several Mamluk scholars visited Gujarat, especially muhaddithin and faqihs; see Sheikh (2010), pp. 205–6. While the British Library Qur’an Add. 18163 does not indicate the place of production, it confirms that Sultan Mahmud Shah Begada (1458–1511), whose seal appears on folio 338, owned a library. (79) Held at the Washington Freer Gallery of Art (F1392.24W). See Norman Brown, The Vasanta Vilâsa: A Poem of the Spring Festival in Old Gujarâtî Accompanied by Sanskrit and Prakrit Stanzas and Illustrated with Miniature Paintings, New Haven (Connecticut): American Oriental Society, 1962. (80) Zebrowski (1983), pp. 122–38. (81) Ya‘qub b. Hasan b. Shaykh Siraj Shirazi, Tuḥfat al-muḥibbīn dar ā’in-i khushnivīsī wa laṭā’if-i ma‘nawī-yi ān, Suppl. Persan 386, Paris: BnF, 190 folios, dated to the late fifteenth–early sixteenth centuries. One version of this work has been edited by Iraj Afshar and Karamat Ra‘na Husayni and supervised by M.T. Danish-Pajuh, Tehran: Noqti/Miras-i Maktub, 1376H/1997. Also see M.T. DanishPajuh, ‘Sar-guzasht-i nāmahā-yi khushniwīsān wa hunarmandān’, Hunar wa Mardum (Tehran), 4, 1348H/1969–70: 34 and A. Munzavi, Fihrist-i nuskhihā-yi khatti-yi fārsī: Catalogue of Persian Manuscripts, Tehran, vol. 3, 1350H/1971, no. 18101, p. 1907. (82) F. Richard, ‘Nasr al-Soltâni, Nasir al-din, Mozahheb et la bibliothèque d’Ebrâhim Soltân à Shirâz’, Studia Iranica, 30(1), 2001: 87–104. The text is dedicated to prince Habib al-Din Muhib Allah b. Burhan al-Din Khalil Allah b. Nur al-Din Ni‘mat Allah Kirmani, the son-in-law of the Bahmani Sultan and grandson of Shah Ni‘mat Allah’s grandson, the famous Qadiri mystic. In the foreword the author, a Sufi from Iran working under the patronage of Indian rulers, specifies that he was a disciple of the calligrapher Sadr al-Din Ruzbihan Khattat Shirazi’s disciple; he also mentions the important part he played when the madrasa Dar al-Safa was founded in Shiraz; see also S.A.A. Rizvi, A History

Page 38 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

The Art of the Book in India under the Sultanates of Sufism in India, New Delhi: Munshiram Manoharlal Publishers, vol. 2, 1978, p. 57. (83) The India Office copy (Ms 1348) is dated 1601 and the translation was produced a century earlier, as stated in the Foreword.

Access brought to you by:

Page 39 of 39

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India

After Timur Left: Culture and Circulation in Fifteenth-Century North India Francesca Orsini and Samira Sheikh

Print publication date: 2014 Print ISBN-13: 9780199450664 Published to Oxford Scholarship Online: January 2015 DOI: 10.1093/acprof:oso/9780199450664.001.0001

Apabhramsha as a Literary Medium in Fifteenth-Century North India Eva De Clercq

DOI:10.1093/acprof:oso/9780199450664.003.0012

Abstract and Keywords The Jain poet Raidhu wrote profusely in the Apabhramsha language for his Digambara Jain patrons in Gwalior. The chapter explores why Raidhu chose to use Apabhramsha at a time (the mid-fifteenth century) when Hindi/Hindavi had begun to emerge as a literary language in the region. It draws upon his long praise texts (prashastis) to draw a vivid picture of the wealthy, transregional Jain community of northern and central India. As Raidhu’s writings established Gwalior as a sacred pilgrimage centre, they also deployed the strategic, cosmopolitan transregionality of Apabhramsha to carve out his Jain patrons’ transregional identity. Keywords:   Jain, Gwalior, Sultanate, Delhi, Apabhramsha, patronage

THIS BOOK RECOGNIZES THE FIFTEENTH century as a period of important innovations in regional cultures, after the political fragmentation of the Delhi sultanate and the establishment of regional sultanates and kingdoms from the fourteenth century. In literature we witness the growing use of vernacular languages and the rise of new genres. Nevertheless, literary production in the ‘standard’ languages of kavya, Sanskrit, Prakrit, and Apabhramsha, continued in and beyond this century (see also Chapter 8).1 Apabhramsha (Apabhraṃśa), though originally an ecumenical language, had over the centuries become closely associated with Digambara Jainism, with famous poets such as Pushpadanta and Svayambhudeva. In fact, all Apabhramsha literary production in the (p.340) fifteenth century seems to consist of Digambara Jain works. Not Page 1 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India all Jain literary production was in Apabhramsha, though. Over the centuries, Jains used Sanskrit, Prakrit, Apabhramsha, and vernacular languages for multiple literary genres without any rigid division of labour.2 Although some authors composed in several languages, most of them appear to have had a preference for just one. Most Prakrit and Apabhramsha compositions can be considered to be multilingual, in the sense that they sometimes include verses in other languages, especially Sanskrit in the initial eulogies (prashastis), but at times also in the main part of the text. Although, according to literary histories, by the fifteenth century Apabhramsha literature had long passed its zenith, the number of Digambara poets who chose Apabhramsha as a literary medium remained considerable, and paradoxically the most prolific Apabhramsha author in history, Raidhu, lived in fifteenthcentury Gwalior, in an environment where the vernacular was unquestionably developing (see Chapter 13).3 In the first part of this chapter, I present an overview of north Indian literary production in Apabhramsha in this period, using prashastis in order to map authors, patrons, genres and places of composition.4 As we shall see, contrary to modern histories that tell us that Jain merchants and religious professionals ‘fled’ from ‘Muslim oppression’, Jains were actually active and often prominent actors in the economic and political life of the north Indian sultanates, and exercised both religious and cultural patronage.5 In the second part of this chapter I focus (p.341) on Raidhu. Rather than on the poetic quality of his works, I concentrate again on his prashastis, since his accounts of the patronage under which he composed his work throw light on the question why one would still choose Apabhramsha as a literary medium in the fifteenth century.

Apabhramsha and Jain Literary Activity in the Fifteenth Century By far the favourite style of Apabhramsha literature was the sandhibandha, usually described as Apabhramsha kavya in matra metres and divided into sandhis (narrative units).6 Since its subject was often the biography of a mythological or worldly hero (cariu < Old Indo-Aryan carita or purana), the sandhibandha has also been called the Apabhramsha epic.7 The Jain Puranas offered a counter-tradition to the growing popularity of ‘Hindu’ Puranas and were composed from about the seventh century onwards.8 They narrated the stories of one or several Jain shalakapurushas or mahapurushas, ‘great heroes’, who included the twenty-four Tirthankaras, twelve Chakravartins, and nine sets of a Baladeva, Vasudeva, and Prativasudeva, totalling sixty-three Puranic heroes.9 The legends of these heroes and the stories surrounding them formed a more or less coherent view of world history and were favourite subjects in Digambara narrative literature. Indeed, in terms of characterization and setting there seems to have been no strict differentiation (p.342) between puranic and non-puranic narratives. The most popular topics of Apabhramsha sandhibandhas in the fifteenth century were biographies of Jain Tirthankaras or Jain versions of the Mahābhārata (called Pāṇḍavapurāṇa or Harivaṃśapurāṇa) which included Page 2 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India the story of the twenty-second Tirthankara Nemi, who, according to Jain puranic history, was Krishna’s nephew. Another popular genre in this period was that of the vratakatha, which illustrated with a narrative the implementation, purpose and benefits of different vows. The formal characteristics of sandhibandhas10—matra metres, the presence of rhyming padas, the individual stanzas (kadavaka) divided into a main body and closing verse—suggest a performative context.11 These are the same formal characteristics found in Avadhi narrative poems, from Da’ud’s Candāyan (1379) onwards (see Chapter 10). Two things are worth mentioning at this point. First, within the multilingual world of Digambara Jain literary culture that comprised Prakrit, Sanskrit, and the vernacular (bhasha), at least in current use Apabhramsha seems to have become assimilated with Prakrit, possibly signalling its status as a ‘Jain thing’.12 Second, sandhibandhas seem to have become the Apabhramsha style, capable of encompassing most, if not all, the material to be composed in that language. Both cases indicate a certain (p.343) streamlining and division of labour within the multilingual Jain literary culture of the time. In the fifteenth century poets of Apabhramsha seem to have been primarily Bhattarakas and Panditas. Bhattarakas were religious leaders (usually translated as pontiffs) of the Digambara monastic lineages. Contemporary authors such as Raidhu describe them as monks (muni, risi, jati), though they were clothed and allowed to possess land and other material wealth prohibited to regular monks. As such, they appear to have held a position intermediate between monks and laymen and provided service to the lay community, for example, by consecrating temples and Jina images or arranging patronage for the copying of manuscripts and the composition of religious texts. They also seem to have been in contact with local rulers on behalf of the Digambara community.13 A Jain Pandita was a type of lay religious professional who, in addition to his work in the temple, could work as a ‘contract poet’ or religious entrepreneur. Raidhu was a Pandita, and from the dedications in his work we can deduce that he wrote for patrons associated with different Bhattaraka lineages, and that he was commissioned to write by a patron or instigated the writing himself as a meritorious act. Most notably, he was the pratishthacharya, the ‘master of consecration’, for a very large number of Jina images on the walls of Gwalior fort—indeed these are most visible remnants of his activities.14 Gwalior (Gopachal), Mandu, Delhi (Yoginipura), Indur, and Hisar-Firoz are among the places mentioned in the prashastis of fifteenth-century Apabhramsha works either as the place of composition or the place of origin of their patrons— thus spanning Tomar, Khalji, Sayyid, and later Lodi domains. For example, the well-known Bhattaraka Yashahkirti of the Kashtasamgha Mathuragaccha Pushkaragana, younger brother and successor to Gunakirti, appears to have resided mainly in Gwalior. Raidhu frequently describes him as one of his personal spiritual guides. (p.344) Two manuscript colophons mention that they Page 3 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India were copied at his command in Gwalior Fort (Gopacaladurga) in 1430, and his name furthermore figures on some of the Jina images there.15 Yashahkirti’s fame rests on his having completed the Riṭṭhaṇemicariu that had been left unfinished by Svayambhudeva and his son Tribhuvana.16 He also composed at least two of his own Apabhramsha poems in places other than Gwalior. In 1440 he wrote a Pāṇḍavapurāṇa commissioned by an Agraval named Hemaraja who was a minister of the Sayyid sultan Mubarak Khan (suratana mumarakha), better known as Mubarak Shah, who ruled from Delhi between 1421 and 1434.17 In the lengthy dedication of the work we learn that Hemaraja’s family was originally from Yoginipura (joyanipura), Delhi, until his father Vilha moved to Navagrama (navagava), a place probably near Delhi as it was apparently ruled by the Sayyids, and that Hemaraja had previously built a Jina temple.18 In 1443 Yashahkirti completed a Harivaṃśapurāṇa, with a briefer version of the Mahābhārata and Nemi story, in Indur (imdaüra), under the rule of Jalal Khan (jalalakhana).19 The patron was another Agraval named Diudha.20 Gunabhadra was a Bhattaraka of the same monastic lineage who had succeeded Yashahkirti’s successor. Like his predecessors, he seems to have been primarily active in Gwalior, as can be gathered from several manuscripts copied and Jina images installed at his command between (p.345) 1453 and 1533.21 The fifteen ‘vow stories’ (vratakathas) he wrote to illustrate the purpose and benefits of various vows contain very brief prashastis, none of which bears a date and only one a location, Gopagiri (Gwalior). Several compositions do not refer to any patron either.22 Shrutakirti, a Bhattaraka of the Jerahata branch of the Balatkaragana in the line of Tribhuvanakirti and Devendrakirti,23 composed at least five Apabhramsha works in 1495–6, including a long Harivaṃśapurāṇa.24 Though he names several important Jain merchants and describes their major feats, he does not mention any patrons for his works. Nonetheless, he is important for our purposes because he writes that he composed his works in the Neminatha temple of his branch in Mandavagarha (Mandu) in Malwa, under the rule of Sultan Ghayasuddin (Ghiyas al-Din) Shah (r. 1469–1500), thus testifying to Jain presence and literary production in the Khalji capital.25 Other place-names are more difficult to locate. Around 1450 a Pandita Tejapala from a Khandelavala family from the village Vasavapura composed a Saṃbhavacariu narrating the biography of the third Tirthankara in the city of Siripahu (Shripatha), ruled at that time by one Daudashaha from the Aühadda dynasty, identified as Da’ud Shah Auhadi.26 (p.346) By contrast, the family of the patron of Vijayasimha’s Ajitapurāṇa (1448), a biography of the second Tirthankara Ajita, is described at length at the beginning and at the end of the work; the patron was Pandita Devapala, a Khandelavala whose family stemmed from Vanippura (Vanipura or Vanikpura), while of the poet we only know that he belonged to a Padmavati-Puravada family from Merupura and lived ‘in the house Page 4 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India of king Karamasimha’.27 Finally, we have no information about the place where Asavala, the author of a 1422 Pāsaṇāhacariu in thirteen sandhis, lived or about his other works, but we know that his patron, Lona, was from a Yadu family and lived in Karhal, which was ruled at the time by the Chauhan king Bhojaraja and his son Samsarachandra.28 The image that we gain from the information contained in the prashastis, then, is of merchant families that were part of a transregional network and well integrated into the urban economy of north India, settled as they were both in the capitals of the regional kingdoms (Mandu, Gwalior, Yoginipura/Delhi) as well as in smaller towns and hill-forts, just as they would later flock to Agra when it became the Mughal capital. Gwalior, as we shall see, was a particularly dense centre of Jain religious and literary activity, and it was here that Raidhu, in Phyllis Granoff’s words, attempted to turn Gopachala into a tirtha, a sacred space, especially through his sculptural but also through his literary production.29

(p.347) Raidhu The most significant Apabhramsha author of the fifteenth century, both because of the volume of his work and because of the position he held in local society, was Raidhu.30 He was born in a Jain family of the Padmavati-Puravada caste, probably in or near Gwalior. His grandfather Devaraja and his father Harisimha both bore the honourable title of sanghadhipa, a recognition bestowed on laymen for substantial services rendered to the Jain religion, such as by the endowment and organization of a pilgrimage. Raidhu himself appears to have been a layman and functioned as a Pandita for the Bhattarakas of the Kashtasangha Mathuragaccha Pushkaragana in Gwalior. Some of the accompanying inscriptions on the Jina images sculpted in the walls of Gwalior fort and references in Raidhu’s prashastis reveal him to have been the main (p. 348) pratishthacharya, ‘master of consecration’, of these images, which made Gwalior a place worth visiting for Jain pilgrims.31 Raidhu was also the author of about thirty works.32 Not unusually for a Jain, he was multilingual poet: though he composed mainly in Apabhramsha, some of his works are in Prakrit and some are designated as ‘Hindi’.33 Following a practice that went back to the earliest Apabhramsha poets, Raidhu’s Apabhramsha works regularly contained Prakrit and Sanskrit verses in language-specific metres, particularly in the prashasti sections. Most of Raidhu’s available Apabhramsha compositions are sandhibandhas narrating the lives of Jain Puranic or worldly heroes, with topics taken from what can be considered as the ‘traditional’ Jain narrative material, already treated by several authors in the past. The first Puranic work he composed was probably the Pāsaṇāhacariu, a biography of the twenty-third Tirthankara Parshva.34 Raidhu’s Ṇemicariu, also called Ṇemipurāṇa, narrates the life of the twenty-second Tirthankara Nemi within the larger frame of the Jain Harivamsha story, which includes the biographies of Page 5 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India Krishna and Balarama, the Mahābhārata and a Jain version of the Bṛhatkathā transformed into the Vasudevahiṇḍi.35 Noteworthy in his Sammaïjiṇacariu, the Digambara biography of the last Tirthankara Mahavira, is the attention Raidhu (p.349) gives to the events following Mahavira’s death in the tenth sandhi (which also circulated separately under the name of Bhadrabāhu-CāṇakyaCandragupta-kathānaka), foregrounding the Digambara preoccupation with the Kali age, in which the unified community split into different sects.36 Raidhu’s fourth biography of a Jina is the Santiṇāhacariu, a biography of the fifth Chakravartin and sixteenth Tirthankara Shanti, which was illustrated. The Sukkosalacariu, dated 1439 (VS 1496), tells the story of Sukaushala, a ruler of the Ikshvaku dynasty and descendant of the first Jina Rishabha, illustrating the tension between kingship and the Jain ideal of renunciation, and the ultimate superiority of the latter.37 The story of Rama, Lakshmana, and Ravana, known among Jains as the eighth Baladeva, Vasudeva, and Prativasudeva, is told in the Balahaddacariu, also known as Paümacariu.38 Finally, the Mehesaracariu tells the story of Megheshvara, born Jayakumara, the head of the army of the first Chakravartin Bharata, and his marriage to princess Sulochana. Raidhu’s Puranas thus follow a century-old tradition and contain little innovation but often give an abridged version of the story, especially in treating the Jain Rāmāyaṇa and Harivaṃśa/Mahābhārata. Raidhu’s other sandhibandhas do not focus on the mahapurushas and yet the world which their protagonists inhabit often appears quite similar to the Jain puranic setting inhabited by superhuman beings such as Vidyadharas. For example, the third sandhi of the Sammattaguṇa-ṇihāṇakavva, a collection of moral stories to illustrate the ‘eight parts’ of the ‘correct view’ (samyagdarśana) of Jainism, tells of the hardships of Anantamati, who at a young age had taken a vow of celibacy. In the story, Anantamati is kidnapped by a Vidyadhara, who saw her and was amazed by her beauty as he flew by in his vimāna while she was playing (p.350) with her girlfriends. Fearful of his Vidyadhari wife, the Vidyadhara abandons Anantamati in the forest, where she is saved from a tribal who wished to make her his wife by a Vyantara goddess.39 There thus seems to be no strict differentiation between the Puranic and non-Puranic Jain narrative genre. The Sirivālacariu, for instance, which tells the story of king Shripala and his queen Mainasundari, the pair who initiated Siddhachakra worship, is set in a Purana-like narrative frame of a dialogue between Indrabhuti Gautama and king Shrenika, who asks to be told about Siddhachakra worship.40 In these narratives, kings and merchants are models of ideal behaviour, their adventures presented in order to illustrate the workings of karma and transmigration of the soul. Thus, the Jasaharacariu tells the popular Jain tale of Yashodhara and his many, often gruesome, rebirths as an illustration of the workings of karma. The Dhaṇṇakumāracariu deals with the adventures of the merchant’s son Dhanyakumara, who generated wealth for others wherever he went. His auspicious situation is explained as the karmic consequence of his Page 6 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India actions in previous existences, in particular the donation of food to an ascetic.41 The Jīmaṃdharacariu clarifies the benefits of upholding the sixteen kāraṇa bhāvanas, the sixteen behavioural principles, through which one can generate an influx of karma that will cause a soul to be born a Tirthankara in a future life, through the stories of the previous births of prince Jimandhara.42 The Puṇṇāsavakahākosu and Sāvayacariu are collections of moral stories illustrating central Digambara doctrinal precepts, such as fasting, the giving of alms, and the namaskāra mantra.43 Except for the heroes of the Puranas, who are generally situated in the tirtha of (p.351) a particular Tirthankara and conform to Jain traditional chronology, none of Raidhu’s narratives are framed specifically in time. Their spatial settings are generally cities such as Rajagriha, Ujjayini, Ayodhya, and so on, commonly known as the places of action from classical Indian literature. Raidhu hereby positions himself in the continuous line of (Jain) storytelling, often said to go back to the Jinas themselves, with narratives about an age that was not yet as tainted by the calamities of the Kali age, in which he considered himself to be living. Apart from these narrative sandhibandhas, Raidhu further composed four doctrinal works in Apabhramsha, some of which are also styled as sandhibandhas.44 His Prakrit compositions and his one work designated as Hindi also discuss the Digambara doctrine.45 Despite his Prakrit and Hindi compositions, Raidhu was first and foremost a poet of Apabhramsha, and of Jain narrative literature in sandhibandhas in particular. Despite the fact that he was living in an age and area of innovation, his choice to rework material that was probably well-known among the Digambaras of the time with few or no changes shows that Raidhu consciously adhered to what he considered a timeless tradition. I will come back to the importance of ‘tradition’ for Raidhu and his patrons. Aside from being recognized as the most prolific Apabhramsha author in history, Raidhu is moreover considered a good poet, although not of the same renown as the greatest poets of Apabhramsha, Svayambhudeva and Pushpadanta. On the whole, his compositions are much more condensed than those of his illustrious forerunners, leaving less opportunity for the vivid descriptions that Apabhramsha epics are well known for. Nevertheless, his work does contain a number of passages where rasa is effectively evoked, for instance, in the description of Sukaushala’s love play with his wives: With wantonness and laughter, delightful in the pleasure of love, their hearts inflamed by mutual looks, (p.352) they drew many amorous gestures. They attacked [Sukaushala] with sharp arrows of sideway glances. They half showed their breasts from under their veil and made passionate, sweet sounds. Afflicted with love, the charming ladies grew giddy and with their form stole their lord’s heart. Page 7 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India They gave him betel with pride and respect, and passionately kissed him. The young women trembled in their first union and wriggled in the intense embrace of his arms. In the struggle of love-making they groaned with rapture, like humming bees lusting after nectar. They cried loudly as he wounded them with his nails, [but soon] returned to the cage of their lord’s arms. They imprisoned each other in love play and uttered words arousing passion. With their fragrant ointments they perfumed [him] and, overcome with love, rejoined him as soon as they were apart [from him]. These manifold plays and pleasures, divine enjoyments, [Sukaushala] experienced together with the young women. It was as if he was the full embodiment of Indra in heaven! Who can describe the power of the penance [from his previous life]? (Sukkosalacariu 4.3.1–12)

In this passage Raidhu evokes the erotic sentiment, shringara-rasa, by suggestively focusing on some of the physical effects and sounds which he underlines by using onomatopoeia: the ladies grew giddy (tharaharaṃti), wriggled (kasamasaṃti), groaned (huṃkāra diṃti), and shrieked (sikkāravaṃti). The simile comparing their sideway glances to sharp arrows also hints at the physical effect, namely that of sudden strike, causing (in this case, pleasurable) pain. The power of suggestion is further enhanced by some explicit details of the love play, again while concealing it from full view: how the women half reveal their breasts to him, how they kiss him when giving him betel, and so on. But the most striking effect of this passage lies in the final words of the last verse: after building up the erotic sentiment, up to a comparison of Sukaushala with Indra and his suggested love play with the Apsaras, the poet brings his audience back to the central theme of the narrative, namely that of renunciation. The eroticism of the previous verses is expressed as a result of the karmic fruits of Sukaushala’s penance in his previous lives, linking the erotic sentiment with that of renunciation (vairagya). Moreover, the words hint at the fact (p.353) that Sukaushala in this life too is destined to renounce the world, despite the efforts of his mother to make him enjoy worldly pleasures.

Raidhu’s Prashastis and Patronage While Raidhu’s doctrinal works, whether in Prakrit, Apabhramsha, or Hindi, contain hardly any information on their patronage and the circumstances in which they were written, his Apabhramsha sandhibandhas frame the narratives with lengthy prashastis that illustrate the context of patronage.46 In general, prashastis in sandhibandhas are longer and more detailed than those of other compositions. Raidhu’s prashastis typically begin with an invocation to the Jinas, Sarasvati, and the Ganadharas, who transmitted the Jain doctrine after the death of the Jinas, followed by the name and praises of a Bhattaraka lineage. Since in Page 8 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India his various works Raidhu lists no less than four seemingly independent lineages, and it is doubtful that he was a follower of all four—especially since they belong to two different, and often conflicting, monastic groups—it seems likely that the lineage mentioned in a particular composition is the one with which the patron and his family were associated.47 After thus sketching the poem’s religious affiliations, Raidhu generally goes on to describe the city where the patronage came to be. These descriptions appear to conform to classical kavya standards rather than offering a realistic representation of a particular place. A typical example is Raidhu’s description of Gopachala (Gwalior) with its contemporary ruler of the Tomar dynasty, Dungar Singh, in the Mehesaracariu: Here is the beautiful city named Gopācala, treasure of gems, famous on earth, source of happiness, indeed like a peak of the Vaitāḍhya [mountains].48 (p.354) It contains people like the ocean contains gems. Like Indra, it removes the danger of enemies. The golden brilliance of the reputation of its people is [tested] on a touchstone, like [the brilliance] of the crown of the beautiful Lady Earth. Like a soldier it is sheltered by forests and woods. Like a dancer it brings joy to the eyes. The river Sauvarṇarekhā reigns and carries water like the mouth of good people [carries saliva]. Its lofty white rampart is like the continuous fame of the Tomar kings. Here the breathtaking market road shines, rich indeed in the abundance of all objects [sold there]. It flashes with the lustre of excellent gold and gems, as if a rainbow spread out on the earth’s surface. Here people delight in helping [others], ever virtuous and affluent in riches and grain. Here [lives] a king, a mine of virtues, of supreme fame, who brings distress to the families of enemies, named Śrī Duṅgarendra, [like] the sun, subdued [only] by his own glory. (Mehesaracariu 1.3.11– 1.4.9)

In two cases, cities other than Gwalior appear to have been the place of composition. In the Puṇṇāsavakahākosu (1.3–4) Raidhu describes as the home of his patron Chandavada on the banks of the Yamuna, then ruled by the Chahuvani (Chauhan) king Payavarudda (Prataparudra), son of Ramaïndu (Ramendu/ Ramachandra). Chandavada was closely associated with the Digambara community and several Jains are said to have held high ministerial functions at the Chauhan court there and had numerous temples built and statues installed from the thirteenth century onwards.49 In the Jasaharacariu, the place where the patron lived, Lahadapura, ruled by Sulitana Sahi and his son Isappha, is named only at the end. Raidhu briefly depicts the city as being ‘adorned with groves, parks and riches, where mango trees are high and always bear fruit, with houses Page 9 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India made of timber, and where, in every house, modest (p.355) folk worship the Jina, the sun of omniscience’.50 The more substantial part of the prashasti at the beginning describes the patron, his family and how the work’s commission came to be. That patrons were crucial for Raidhu’s literary production is further evidenced not only by the customary mention of their names in the Prakrit or Sanskrit colophons of most of the sandhis but also by the insertion of a namanka or namamudra in the final verse (ghatta) of the last stanza of a sandhi in some compositions.51 The Apabhramsha namanka, a characteristic stylistic device of the sandhibandha style, is a pun on the name, usually of the author but here also of the patron, for instance, Tosaü, the patron of the Sammaïjiṇacariu:

iya ṇisuṇivi devī

suratiya-sevī

citti pahiṭṭhiya gaya sa-hari

ñhiya tosu lahaṃtī

pulaü vahaṃtī

raï dharaṃti jina-dhammi bhari

(p.356) Upon hearing this, the queen, served by the wives of the gods, was delighted at heart and went to her chambers. As she stayed there, she became satisfied (toṣa + labh., lit. ‘receiving satisfaction’), the hair on her body bristling, and found pleasure (rati + dhṛ.) in the teachings of the Jina, as she thought about it.52 Raidhu’s prashastis highlight three types of information about the patrons: the place of origin of their ancestors, the meritorious religious deeds done by the men and women of the family, and their good relations to the local court. In most of Raidhu’s sandhibandhas the patron’s family lineage is given not once but twice in two prashastis, one at the beginning either by Raidhu or by the mediator who suggested the patron, and a longer one at the end. The lineages generally begin by mentioning the caste and gotra of the patron, in most cases an Agraval, and go back several generations, often to ancestors who lived elsewhere. They therefore act as equivalents to the Charans’ genealogies of Rajput patrons. In four out of six cases, the place of origin mentioned is Yoginipura (joinapura, joyanipura), in another it is Hisar-Firoz (hisara-peroju) and once Jhunujhunupura.53 No information is given as to the reason for their migration, from which we may deduce that the move was in the ordinary order of merchants’ patterns of movement, that is, to follow business opportunities. (p.357) Apart from describing them all as ideal Jain laymen and women, Raidhu sometimes mentions the specific acts of religious merit which enhanced their prestige in the Digambara community, from which we get a picture of intense Jain activity in this period. Of Nemidasa, the patron of the Puṇṇāsavakahākosu, for instance, Raidhu reports that he ‘had several Jina statues installed and consecrated’.54 He is also said to have ‘built a temple to the Page 10 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India Jina’.55 From Nemidasa’s own words we learn that Raidhu performed the consecration of the images and had advised the patron to build the temple (Puṇṇāsavakahākosu 1.6.10–11). Raidhu frequently calls Nemidasa, as well as significant number of other Jain laymen, ‘lord of the community’ (samghahiva, or variant samghahi, samghavi, samghaï < sanghadhipa). Nemidasa’s younger brother is called the ‘sanghadhipa of Girnar’, suggesting that he organized and provided financial support for a pilgrimage to Girnar in Gujarat.56 Similarly, of some patrons who had one or more Jina images built and consecrated, it is said that they ‘procured (lit. bound) the gotra of a Tirthankara’, which means that through their auspicious act, they provided themselves with the kind of karma which would make them reborn in a future existence as Tirthankaras.57 Raidhu also takes care to mention his patrons’ frequent association with the local court. Thus, Nemidasa is said not only to have been ‘honoured by king Prataparudra’, the Chauhan ruler of Chandavada, but was also a ‘supreme master in the employment of the king’.58 The Tomar king Dungar Singh, too, appears to have had a number of Digambaras in his employ: Seu Sahu, for instance, father of the patron of the Sāvayacariu, (p.358) is called the ‘head of king Dungar’s treasury’, and three more people are mentioned as advisors in Dungar’s court.59 The cordial relations between the court of Gwalior and the Digambara community are well illustrated by an episode in the Sammattaguṇaṇihāṇakavva (1.9–11), where Raidhu’s patron Kamalasiha narrates how one day he went to the palace to get approval for his plan to erect an image of Rishabha in the walls of the fort. Dungar Singh invited Kamalasiha inside for a personal conversation and, impressed by Kamalasiha’s piety, he not only consented to erecting the image but instructed his personal servants to provide Kamalasiha with anything he might need (Sammattaguṇaṇihāṇakavva 1.11.14–21). The Tomars, like other rulers, were keen to secure the presence of flourishing merchants in their city and to maintain good relations with them. Allowing religious sponsorship was one way of doing so. What about the merchants themselves? From Raidhu’s prashastis it appears that the circumstances of patronage varied, suggesting a kind of Jain literaryreligious market in which Bhattarakas acted as mediators, taking the initiative and suggesting a specific topic to the poet—praising him in the prashasti for his poetic skills that will ‘bring auspiciousness and gladden the minds of the people’—and then suggesting an appropriate patron when the poet laments the absence of an audience (soyāra) to support his poetry.60 Twice Raidhu is approached by mediators other than Bhattarakas, while on two occasions the initiative comes from Raidhu himself, not without some clever display of rhetoric.61 In the (p.359) Jīmaṃdharacariu (1.3) Raidhu recalls how he one day heard some people discussing two of his compositions that had been commissioned by the same patron, Kumthudasa. As he reflected on these words, he felt these two compositions were like a crown and one earring for Kumthudasa and that the other earring was still lacking. He then approached Page 11 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India Kumthudasa, who gladly accepted to act as a patron for the Jīmaṃdharacariu (1.4). Monetary reward is not explicitly mentioned; rather the currency is that of fame for the poet and merit for the patron. So while the patron and his family are praised at length in the prashasti, as we have seen, in this description of the transaction it is the poet who needs to be reassured that the patron will be suitable to spread the poet’s fame. As Raidhu remarks in the Sukkosalacariu (1.4.1–3): Without an audience a book does not excel (…). Without it, how will it spread among the people? Without an audience, it will not become famous. Without an audience, even the voice of the Jinas cannot go forth, nor can their principles be expounded.62 The Bhattaraka then suggests a patron, whose family he describes at length and convinces Raidhu that this man will ‘spread this work’.63 Though none of the prashastis state explicitly how the patron will fulfil this responsibility, what is possibly meant by this is sponsoring the performance of the poem to larger audiences or arranging for the manuscript to be copied and thus spread further. In other cases the prashasti describes how the patron approached the Bhattaraka in the temple, asking him to contact Raidhu with a request on his behalf.64 In three other instances the patrons directly go up to Raidhu and make their wish for a new composition known.65 (p.360) As for the patrons’ motivations, the prashastis make it clear that patronizing the composition of a work is as meritorious an act as having a sacred image installed. Kheu Sahu, a merchant from an Agraval (aïravala, ayaravala) family living in Gwalior, one day contemplates the wealth he has gathered and feels the need to serve the dharma (dhamma). He goes to the temple and asks Raidhu to compose a Megheshvara story for him, which ‘removes the dirt of the Kali [age] and brings prosperity’ (Mehesaracariu 1.7). In the long prashasti of the Sammattaguṇaṇihāṇakavva, the mahajana Kamalasiha of Gwalior approaches Raidhu in the temple.66 He, too, feels dissatisfied with having material wealth alone and desires karmic merit (punna, punya) in this age of Kali (1.6.19–1.8.17). In the past he already had an image of Rishabha consecrated under the auspices of king Dungar Singh Tomar himself and now he convinces Raidhu to compose a poem on the virtues of ‘correctness’ (sammatta, samyaktva) in Jain doctrine (1.14–16). Another work stages the conversation between the patron and the mediator of the work as follows: Khelha mentions to the Bhattaraka Yashahkirti how he once had an image of Jina Chandraprabha consecrated on Gopachala to honour his mother who had become a mendicant (jina-dikkha, Sammaïjiṇacariu 1.4.12–19). He now intends to fulfil his promise of sponsoring a biography of Mahavira and to donate her the religious benefit (punnaha phalu) accruing from this act in order to thank her for the gift of the auspicious human birth (Sammaïjiṇacariu 1.5).67

Page 12 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India Though these words suggest that the patrons considered their action first and foremost an act of religious merit, and that paying for (p.361) the composition of a Jain religious work generated good karma and enhanced the possibility of an auspicious future existence, Raidhu’s texts also hint at social prestige and fame as a secondary motivation of the patrons and their families. For instance, Nemidasa, the patron of the Puṇṇāsavakahākosu, tells Raidhu: ‘With good poetry, spread my name, so that my house will become steady and eternal!’ (1.6.16), to which Raidhu responds: ‘I will spread your fame here in the world’ (1.6.24). After suggesting Hemaraya as a suitable patron, Bhattaraka Kamalakirti tells Raidhu: ‘Make his name stand with good poetry!’ (Jasaharacariu 1.1.13). And in connection with another patron: ‘Again you must let the name of that prideless noble man rise in a poem, so that it [his name] spreads here within Bharata for a long time to come’ (Sammaïjiṇacariu 1.8.14). As we have seen, the long genealogies given in the long prashastis to these works reinforce the idea that the fame and prestige of the family were crucial motivations. As a professional poet, Raidhu was concerned with both maintaining his reputation and spreading his patrons’ fame.

Conclusions Why did Raidhu choose Apabhramsha, despite living in a time and place where the literature in the regional vernacular had begun to develop (see Chapter 13)? This chapter has sought to answer this question by drawing upon the information presented in his own texts, in particular the long prashastis of his sandhibandhas. The image of Gwalior’s Digambara families that emerges from Raidhu’s prashastis is that of a wealthy and confident transregional community that moved along trade routes and between cities like Delhi, Hisar, Gwalior, and Mandu that belonged to different sultanates and kingdoms. Some families in Gwalior had migrated from places hundreds of miles away in north India and kept up direct relations with them. In one work Raidhu describes the family of Teju Sahu from Kurukshetra, whose daughter had married Khelha, the son of Raidhu’s patron in Gwalior (Sammaïjiṇacariu 10.34.18–27). Bhattarakas also travelled long distances and crossed political borders (p.362) to serve their transregional lay constituency.68 Even Raidhu himself apparently travelled to Lahadapura where, as he states, he ‘composed this [Jasaharacariu] on the excellent premises of Shri Jodha Sahu’ (Jasaharacariu 4.18.21–22). The choice of Apabhramsha, a transregional language of literature, thus parallels the transregionalism of this Digambara community. Transregional, but locally rooted, active and well-connected, the migrating families of Raidhu’s prashastis were able to form close connections with local rulers, both Muslims and Hindus, and a number of them held official positions at the courts. They also invested in and moulded the local space, building temples, and consecrating images in their new home towns. Indeed, turning Gwalior into a Jain pilgrimage centre seems to have been desirable for both Raidhu and Gwalior’s ruler Dungar Singh Tomar. That the cultural (and economic) Page 13 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India transregionality of Jain merchants was a prominent and attractive local value is clear from king Dungar Singh’s statement when he recognized the merchant Kamalasiha first and foremost as a member of the wider community of Jain merchants who had brought prosperity to different regions in India long before the Tomar kingdom of Gwalior. In response to Kamalasiha’s petition to be allowed to erect a Jina image, Dungar Singh declares: Just like Saurashtra during the reign of Visaladeva, where [Jain] dharma could settle for a long time without any obstacles and where the excellent merchants Vastu- and Tejahpala constructed excellent tirthas, and just like Firoz Shah’s sincere graciousness towards [Jains] living in Yoginipura, where the famous Saramga Sahu zealously organised pilgrimages, in the same manner, o righteous and virtuous man, you may set up all these things here!… If something is missing, I shall provide it. Whatever you require, I shall provide. (Sammattaguṇaṇihāṇakavva 1.11.10–16) Dungar Singh here mentions two historical political regimes, one in Gujarat and one in Delhi, where Jain merchants were held in high regard and were allowed to perform their religious observances.69 Both in (p.363) their selfunderstanding and in the understanding of local rulers, then, Digambara merchants were seen as members of a powerful transregional community, whose sphere of activity crossed boundaries of time, space, and polity, and who needed to be encouraged to settle and invest locally. We have seen above that Raidhu himself preferred the term pāyaya or pāiya, that is, prakrita, Prakrit, despite the fact that we now recognize his works to have been in Apabhramsha and his preferred sandhibandha style to be an exclusive Apabhramsha genre. In one instance in the Jasaharacariu, he does mention Apabhramsha in a general statement about the three classical languages allowed for kavya: ‘Whichever language, Prakrit, Sanskrit, Apabhramsha, etc. there may be in an utterance, a handsome poem bewilders the mind of wise men.’70 Why then should he refer to the language of his own works as Prakrit instead of Apabhramsha? One explanation may be that by this time Prakrit had become more identified than Apabhramsha as a Jain language.71 In any case, whether Raidhu calls his language Apabhramsha or Prakrit, it is clear that he refers to himself as a poet composing in one of the traditional transregional languages of kavya. The confusion could also stem from the pragmatic fact that by Raidhu’s time, Prakrit and Apabhramsha were both cosmopolitan languages more or less exclusively linked to Jain literary production.72 A distinct advantage of the Apabhramsha sandhibandha style for Raidhu’s merchant patrons must have been that it left more room for prashasti information than most of the other classical forms. Even a cursory (p.364) comparison between the Apabhramsha prashastis in the second volume of Shastri’s Jain granth praśasti saṃgrah and the Sanskrit and Prakrit prashastis in Page 14 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India the first volume, confirms this. In Raidhu’s compositions in Prakrit and Hindi either no patron is mentioned at all or only his name appears. In order to enhance their social prestige and immortalize their names and those of their family members, as well as their contribution to the community, patrons then opted for Apabhramsha sandhibandhas narrating the stories of Puranic and nonPuranic heroes following the tradition started centuries earlier by Pushpadanta, who devoted several kadavakas of the first section of his Mahāpurāṇu to the description of his patron, Bharata. For Raidhu, positioning himself and Digambara religious activity in Gwalior within this hallowed tradition must have been crucial in his attempt to turn Gopachala into a tirtha, a sacred space, mainly through his sculptural, but also through his literary production.73 In order to realize this endeavour, it made sense for Raidhu and his patrons to use media that were both timeless and transregional. Although Raidhu appears to have been proficient with vernacular poetic forms, as we gather from his ‘Hindi’ compositions, and his patrons were perhaps no longer able to understand Apabhramsha compositions in full, cosmopolitan traditions which apparently offered little room for innovation still embodied significant value. Notes:

(1) The oldest traces of the recognition of Apabhramsha as one of the three ‘classical’ languages of kavya date from the seventh century, though unfortunately none of the early compositions have survived. See Bhamaha’s Kāvyālaṃkāra (I.16) and Dandin’s Kāvyādarśa (I.32–36); see Sheldon Pollock, The Language of the Gods in the World of Men, Berkeley: University of California Press, 2006, pp. 89–114. (2) For the use of Sanskrit among the Jains, see Paul Dundas, ‘Jain Attitudes towards the Sanskrit Language’, in Ideology and Status of Sanskrit, Jan E.M. Houben (ed.), Leiden: E.J. Brill, 1996, pp. 137–56. (3) See also the mention of Gwalior in Brac de la Perrière in this volume. (4) This overview is based mainly on Kochar’s outline of Apabhramsha literature and the prashastis of Apabhramsha compositions collected in P.J. Shastri (ed.), Jain granth praśasti saṃgrah, Delhi: Vidya Sewa Mandir, 1954–63, vol. 2. It deserves note that more than one third of the 121 works treated by Shastri are identified as dating to the fifteenth century. Many of those which are not precisely datable are also attributed to the fifteenth century. See H. Kochar, Apabhraṃśa sāhitya, Delhi: Bharatiya Sahitya Mandir, 1956. (5) K.C. Jain’s history of Jains in Malwa, for example, stops abruptly in 1305, implying that there was no significant Jain presence after the establishment of

Page 15 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India the Khalji sultanate; K.C. Jain, Malwa Through the Ages: From the Earliest Times to 1305 A.D., Delhi: Motilal Banarsidass, 1972. (6) Matra metres are quantitative metres with a fixed number of morae, whereas varna or syllabic metres, found predominately in Sanskrit, follow a fixed number and pattern of long and short syllables. Well over half of the impressive number of compositions discussed by Kochar (1956) are sandhibandhas. (7) A.K. Warder, Indian Kāvya Literature, Delhi: Motilal Banarsidass, 1972–2004, vol. 1, pp. 413ff. (8) The oldest known Jain Purana is Vimalasuri’s Paümacariyaṃ, which has been dated to either the first, the third, or the fifth century. After an apparent interval of several centuries (if we accept the earlier date for the Paümacariyaṃ), texts become more frequent from the seventh century onwards, with Ravisena’s Padmapurāṇa, Jinasena Punnata’s Harivaṃśapurāṇa, and the Mahāpurāṇa of Jinasena and Guṇabhadra. (9) J. Varni, Jainendra siddhānta kośa, New Delhi: Bharatiya Jnanpith, 1997–2000, vol. 4, pp. 8–26. A second list of 169 heroes also includes the mothers and fathers of the twenty-four Jinas, eleven Rudras, nine Naradas, twenty-four Kamadevas, and fourteen Kulakaras; see Varni (1997–2000), p. 10. (10) The sandhis are composed of smaller textual units of about ten verses called a kadavaka, which are further subdivided into three structural parts: (a) the kadavakas of a particular sandhi may begin with a verse of two to four rhyming padas; (b) a kadavaka always entails a middle part, and (c) a closing verse, called ghatta. (11) Marginal references to musical performance are, for instance, found in Sandhis 48, 49, and 81 of Svayambhudeva’s Paümacariu (ninth–tenth century); see H.C. Bhayani (ed.), Paumacariu of Kaviraja Svayambhudeva, Bombay: Singhi Jain Shastra Shikshapith, Bharatiya Vidya Bhavan (3 vols), 1953–60. (12) This is suggested by some of Raidhu’s reflections on the literary language he employed in his compositions, which are nowadays considered to be Apabhramsha. In the initial prashasti of the Santiṇāhacariu, for instance, Bhattaraka Jinacandra encourages the poet to: ‘start occupying yourself with a Prakrit poem’ (āraṃbhahi pāiya-kavva-vitti, 1.6.5). The colophons of the individual Sandhis of the Balahaddacariu explicitly state that the work is ‘bound in Prakrit’ (pāiya-baṃdheṇa), and in the Pāsaṇahacariu, while excusing himself for any errors he may have made in his work, the poet says: ‘I know nothing of Prakrit metres’ (pāiyachaṃdahã ṇaü muṇiu lesu, 7.6.1). (13) See V.P. Johrapurkar, Bhaṭṭārak sampradāy, Sholapur: Jain Sanskriti Sanrakshak Sangh, 1958, prastavana; V.A. Sangave, Jaina Community: A Social Page 16 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India Survey, Bombay: Popular Prakashan 1980, pp. 317–22; and Peter Fluegel, ‘Demographic Trends in Jaina Monasticism’, in Studies in Jaina History and Culture: Disputes and Dialogues, P. Fluegel (ed.), London: Routledge, 2006, pp. 339–60. (14) See Phyllis Granoff, ‘Mountains of Eternity: Raidhū and the Colossal Jinas of Gwalior’, Rivista di Studi Sudasiatici, 1, 2006: 31–50. (15) Johrapurkar (1958), p. 217 (lekhamka 557), 218 (lekhamka 560); Shastri (1954–63), vol. 2, prastavana p. 83n2. (16) H.C. Bhayani (1953–60), vol. 1, Introduction, pp. 18–20; Shastri (1954–63), vol. 2, prastavana p. 83n2; Johrapurkar (1958), p. 218 (lekhamka 559). (17) K.S. Lal, Twilight of the Sultanate, Bombay: Asia Publishing House, 1963, pp. 84, 112–13. (18) Note that this last statement is lacking in the manuscript used by Shastri (1954–63), vol. 2, p. 39. See Kochar (1956), p. 119; Shastri (1954–63), vol. 2, prastavana pp. 81–2, and pp. 38–41 (prashasti 21). (19) Imdaura indicates the hill fortress of Indur of the Mewati rulers near Tijara (Alwar dist., Rajasthan), see Lal (1963), pp. 104, 108–9. (20) See Kochar (1956), p. 122; Shastri (1954–63), vol. 2, prastavana p. 82, and pp. 41–4 (prashasti 22). Yashahkirti further composed a Jinarātrikathā, describing the night when Mahavira attained moksha, and a Ravivratakathā, neither of which appear to have been composed under explicit patronage; Kochar (1956), p. 359; Shastri (1954–63), vol. 2, prastavana p. 82, and pp. 44–5 (prashasti 23 and 24). (21) Johrapurkar (1958), pp. 219–21 (lekhamka 565–75). (22) These were fifteen manuscripts bundled together and kept in the Panchayati Mandir (Khajur Masjid) in Delhi, whose prashastis Shastri edited; Shastri (1954– 63), vol. 2, prastavana pp. 111–12, and pp. 103–5 (prashastis 50–65). (23) In most of his prashastis Shrutakirti describes himself as the successor of Tribhuvanakirti and describes Tribhuvanakirti as the successor of Devendrakirti; see Johrapurkar (1958), pp. 207, 209. (24) See Shastri (1954–63), vol. 2, prastavana p. 122, and pp. 111–12 (prashasti 85); and Kochar, who gives a wrong number of sandhis (1956), pp. 127–8. His other works include a Parameṣṭhīprakāśasāra (Parameṭṭhīpayāsasāra) of seven chapters in the sandhibandha style, which mainly gives instructions on the Jain teachings; a Yogasāra in two sandhis teaching correct behaviour for laymen; a Dhammaparikkhā of 179 kadavakas, and in his Yogasāra he mentions that he had Page 17 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India composed a Siddhacakkakahā; see Kochar (1956), pp. 373–4; Shastri (1954–63), vol. 2, prastavana pp. 122–3, and pp. 133–5 (prashastis 86 and 106). (25) Lal (1963), p. 64. (26) See Dasharath Sharma’s preface to Shastri (1954–63), vol. 2, pp. iii–iv; Lal (1963), p. 124. The patron was an Agraval from the Mittala gotra (mitana and mittana); Shastri (1954–63), prastavana p. 87, and pp. 50–4 (prashasti 28). Tejapala’s Varaṃgacariu, completed in 1450, tells the story of king Varanga who lived in the time of the twenty-second Tirthankara Nemi. No patron is mentioned for this work, though it is explicitly said to have been composed ‘through the grace of Muni Vipulakirti’; see Shastri (1954–63), vol. 2, prastavana p. 87, and pp. 54–5 (prashasti 29). In 1458 he completed a Pāsapurāṇa, a biography of Parshva, at the request of one Ghughali; Shastri (1954–63), vol. 2, prastavana, pp. 87–8, and pp. 124–5 (prashasti 99). Note that the Bhattarakas Tejapala names as his own gurus were active, according to Johrapurkar’s material, between 1524 and 1544, making it difficult to connect them to Tejapala’s Varaṃgacariu, dated 1450 (1507 VS); see Johrapurkar (1958), p. 115 (lekhamka 282). (27) Shastri (1954–63), vol. 2, prastavana pp. 127–8, and pp. 117–19 (prashasti 89); I was unable to identify this ruler. (28) Shastri identifies Asavala’s karahala, which the poet situates in the country of Kushartha (kusaññha), with modern Karhal, north of Etawah; see Shastri (1954–63), vol. 2, prastavana pp. 129–30, and pp. 128–9 (prashasti 101); Johrapurkar (1958), p. 96. (29) Granoff (2006). (30) Of all the Apabhramsha poets of this period, he is the only one to have received some, albeit limited, scholarly attention. Kochar describes the manuscripts of a few of Raidhu’s compositions, referring to earlier articles in Indian Jain scholarly journals Anekānt and Jain Siddhānta Bhāskar; Kochar (1956), pp. 116–8, 240–4. Unfortunately, I have not been able to consult these articles. The prashastis of most of Raidhu’s available compositions were first edited by Shastri in his anthology (1954–63), vol. 2. Rajaram Jain conducted the first thorough study of Raidhu’s work: he collected manuscripts of twenty-five individual compositions, edited and translated five of them into Hindi and studied many of the others in a separate volume, including a rough edition of some prashasti passages of sixteen of Raidhu’s texts; see R. Jain, Raïdhu sāhitya kā ālocanātmak-pariśīlan, Vaishali: Prakrit, Jainshastra aur Ahimsa ShodhSansthan, 1974; Raïdhu-granthāvalī (2 vols), Sholapur: Jain Sanskriti Sanrakshak Sangh, 1975–88; Bhadrabāhu-cāṇakya-candragupta-kathānaka evaṃ rājā kalkivarṇaṇa, Varanasi: Shri Ganesh Varni Di. Jain Sansthan, 1982; Puṇyāsravakathā(-kośa), Delhi: Shri Digambar Jain Sahitya-Sanskriti Sanrakshan Samiti, Page 18 of 24

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: OUP-Reference Gratis Access; date: 23 May 2020

Apabhramsha as a Literary Medium in Fifteenth-Century North India 2000. Aside from these, Nalini Balbir edited and translated one of his shorter poems; N. Balbir, ‘Souper de jour: quatrains’, Indologica Taurinensia, 14, 1987– 8: 47–77. Phyllis Granoff has recently discussed Raidhu’s activities as a consecrator of the Jina images in the walls of Gwalior fort in their contemporary political and religious context (2006). In the prashastis of his work, variants of his name are Raïdhu, Raïvaï (?Ratipati) and Raïdhura. Rajaram Jain believes the name Raidhu to be the equivalent for the Sanskrit name Ratidhava (‘husband of Rati’, as in Raïvaï). He also seems to have been known under the alias Simhasena; R. Jain (1974), pp. 35–9. (31) B. Jain, Bhārat ke digambar jain tīrth, Bombay: Bharatvarshiy Digambar Jain Tirthkshetr Kameti, 1974–6, vol. 3, p. 331. Contemporary websites mention Gopachala as an atishaya kshetra, attributing a miraculous event to the site; for example, http://www.jainteerth.com/teerth/gopachal.asp, http:// www.pilgrimagetourinindia.com/yatra-jain.htm, accessed on 1 December 2009. (32) This chapter is limited to Apabhramsha compositions whose manuscripts have been found and contents described. (33) So designated by R. Jain (1974), pp. 4 and 478 apparently basing himself on the information of a manuscript catalogue. The text, Bārahabhāvanā, itself does not appear to mention the language in which it was written; R. Jain (1974), pp. 447–52. (34) Raidhu apparent