Before the Divide: Hindi and Urdu Literary Culture [1 ed.] 8125038299, 9788125038290

Citation preview

Before the Divide

For our entire range of books please use search strings "Orient BlackSwan", "Universities Press India" and "Permanent Black" in store.

Before the Divide Hindi and Urdu Literary Culture

Edited by FRANCESCA ORSINI

BEFORE THE DIVIDE Orient Blackswan Private Limited Registered Office 3-6-752 Himayatnagar, Hyderabad 500 029 (A.P.), INDIA e-mail: [email protected] Other Offices Bengaluru, Bhopal, Chennai, Guwahati, Hyderabad, Jaipur, Kolkata, Lucknow, Mumbai, New Delhi, Noida, Patna, Vijayawada © Francesca Orsini, 2010 First published 2010 This paperback edition 2011 Reprinted 2016 eISBN 978-81-250-5339-2 e-edition: First Published 2018 ePUB Conversion: TEXTSOFT Solutions Pvt. Ltd. All rights reserved. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the publisher, except in the case of brief quotations embodied in critical reviews and certain other noncommercial uses permitted by copyright law. For permission requests write to the publisher.

Contents

Acknowledgements A Note on Transliteration 1. Introduction FRANCESCA ORSINI 2. Rekhta: Poetry in Mixed Language The Emergence of Khari Boli Literature in North India IMRE BANGHA 3. Riti and Register Lexical Variation in Courtly Braj Bhasha Texts ALLISON BUSCH 4. Dialogism in a Medieval Genre The Case of the Avadhi Epics THOMAS DE BRUIJN 5. Barahmasas in Hindi and Urdu FRANCESCA ORSINI 6. Sadarang, Adarang, Sabrang Multi-coloured poetry in Hindustani Music LALITA DU PERRON 7. Looking Beyond Gul-o-bulbul Observations on Marsiyas by Fazli and Sauda CHRISTINA OESTERHELD

8. Changing Literary Patterns in Eighteenth Century North India Quranic Translations and the Development of Urdu Prose MEHR AFSHAN FAROOQI 9. Networks, Patrons, and Genres for Late BrajBhasha Poets Ratnakar and Hariaudh VALERIE RITTER Contributors Bibliography Notes

Acknowledgements

T

he present volume is based on a workshop on ‘Intermediary Genres in Hindi and Urdu’ which took place at the Faculty of Oriental Studies, at a panel on ‘North Indian Literary Cultures Before the Divide’ organised by Professor Vasudha Dalmia of the University of California at Berkeley and myself at the 18 th European Conference on Modern South Asian Studies in Lund (Sweden) on 8 July 2005. I would like to thank the Smuts Fund (Cambridge) for funding the workshop, and Aditya Behl, Vasudha Dalmia and Samira sheikh for presenting papers which unfortunately could not be included in this volume. I would also like to thank Christopher Shackle, Shamsur Rahman Faruqi and Alok Rai for having encouraged my first forays into comparative Hindi-Urdu perspectives, and Vidya Rao for her particularly informed editing. The further I proceed along this path, the more I discover that R.S. McGregor had already been there, and this book is dedicated to his exemplary scholarship.

A Note on Transliteration

T

ransliteration broadly follows R.S. McGregor in the Oxford HindiEnglish Dictionary (Oxford University Press 1993) for Hindi. For Persian and Urdu words we have followed F. Steingass's A Comprehensive Persian-English Dictionary (various editions). We have used transliteration and diacritics only for the titles of works, quotations and unfamiliar words.

1 Introduction Francesca Orsini

This essay has benefited from prolonged discussions with the participants in the workshop I organised on ‘Intermediary Genres in Hindi and Urdu’ held in Cambridge (UK) on 15 September 2004 and the panel on ‘North Indian Literary Cultures Before the Divide’ set up by Vasudha Dalmia and myself at the 18th European Conference on Modern South Asian Studies in Lund (Sweden) on 8 July 2005. I am also indebted to Daud Ali, Sudipta Kaviraj, Imre Bangha, Lalita du Perron, Samira Sheikh, Katherine Brown and Jessica Bath, who took part in a monthly reading group that commented on the essays in Sheldon Pollock 2002.

P

recolonial India was a deeply multilingual society, with multiple traditions of knowledge and of literary production conducted in specific languages and a marked diglossia between ‘classical’ languages and the ‘vernacular’. Yet the first histories of north Indian literatures, written during the colonial and nationalist periods and deeply involved in crystallising communities around language and cultural identity, rewrote literary history in terms of separate, single-language traditions as the competitive and teleological histories of ('Hindu') Hindi and (‘Muslim’ or secular) Urdu. As a consequence, these literary histories have been marked by ‘appropriation, neglect and exclusion’ (Bangha in this volume). So far, the alternative to this fractious history of Hindi vs Urdu has been a narrative of ‘composite culture’, where selective syncretic traditions have been taken as definitive evidence that culture acted as a great cohesive force

in the mixed Indo-Muslim polity.1 Both narratives have had to exclude much of literary production to prove their point. The myth-making and exclusions that were involved in the construction of separate ‘Hindu-Hindi’ and ‘Muslim-Urdu’ literary traditions have been discussed at length by scholars in recent years, but an alternative to those flawed narratives is yet to emerge.2 This volume is the first attempt to rethink aspects of past Hindi and Urdu literary production outside the straitjacket of Hindi and Urdu literary histories. Some of the names and genres taken up by the individual authors will be familiar ones to students of Hindi and Urdu literary history—Keshavdas, Malik Muhammad Jayasi, Tulsi, Sauda, Upadhyay, ‘Hariaudh’, Braj Bhasha riti poetry, Avadhi epics and Urdu marsiyas—others, such as Vajid, Makhdoom Sarvarhalai or Maqsud, will be less familiar. But even the familiar inhabitants of the Hindi-Urdu literary terrain have been considered in an unfamiliar light. The authors move beyond the constraints of teleological narratives of Hindi and Urdu to explore the more spacious framework of ‘north Indian literature’. These reassessments of canonical texts and poets, and explorations of less familiar works and genres will prove to be crucial in reconstructing the multilingual and multi-dimensional literary world ‘before the divide’. CRITICAL ISSUES The first problem faced by nineteenth and early-twentieth century works on ‘Hindi’ and ‘Urdu’ linguistic and literary history, and in the debates that suffused the era of their composition, was that of language definition.3 As scholars have pointed out in recent years, the issue of language definition was first recognised as a ‘problem’ by colonial linguists, and the suggestions they put forward, often carrying the stamp of official authority, either became commonplace or provoked long-lasting debate and resentment.4 The problem centred on the obvious differences in script and vocabulary in what seemed to be the same language, and also on the uncertainty surrounding the correct name of the language(s). John Borthwick Gilchrist, the influential writer of language textbooks, professor of Hindustani at Fort William College in Calcutta and, as a consequence of that position, patron of Hindi and Urdu writers, thought of it as a

continuum, in which ‘Hindi’ was the rustic, unPersianised bottom register, Persianised Urdu the top register, and Hindustani the preferred middle level. Gilchrist was also perhaps the first to identify language with script and religion, suggesting that Urdu and Hindustani, in the Persian script, were largely the language of north Indian Muslims, while Hindi in the Devanagari script was that of Hindus. As such, Gilchrist has been reviled in Hindi and Urdu literary histories as a classic proponent of the colonial policy of divide and rule. With the benefit of hindsight, however, we can see that he was both reacting to the social and cultural hierarchy of north India at a time in which Persianised speech was prized, and to the need of identifying the one ‘common vernacular of Hindustan’, i.e. north India. His identification of language with script was no doubt problematic and opened a veritable can of worms for both colonial officials and Indian intellectuals. As far as language definition is concerned, we believe, with Imre Bangha (this volume), that Hindi and Urdu are names for ‘a multitude of north Indian vernacular dialects that from an outsider’s point of view were simply called Hindavi (language of India), or Bhakha, (simply, the spoken language), to distinguish them either from Persian and Arabic or from Sanskrit and Prakrit’. As new ideas about language met with a new sense of history and cultural identity centred around ‘community’, and as Indian intellectuals felt the need to rearrange literary traditions into literary histories, they also immediately confronted the problems of script and language definition. Was everything written in the Urdu (Persian) script automatically Urdu (or Hindustani), and everything written in Devanagari automatically Hindi? Or did an attribution of language depend on the religious identity of the author, or on the content of the work? What happened when the same texts had been copied in manuscripts in both scripts, not to speak of the other scripts current in north India (Gurumukhi, Kaithi)? Frances Pritchett and S.R. Faruqi’s work (2001) on Muhammad Husain Azad’s seminal Urdu literary history, Āb-e hayāt (1880), and King’s work (1974) on the literary-historical policy of the Nagari Pracharini Sabha (est. 1893), the most important Hindi literary institution of the time, suggest that several logics of exclusion and inclusion were at work. Script, the religious identity of the author, the content of the work and its quotient of literariness were the main standards.

Literariness, of course, meant the literary standards of nineteenth-century intellectuals, who operated in milieux that viewed ornate (riti) or devotional poetry and the ghazal as the distinctive features of, respectively, Hindi and Urdu literature. This meant that, to give just a few examples, the devotional Kabir and Jayasi became part of Hindi literary history ‘despite’ being Muslims, Raskhan was praised for his ‘love for Hindi’ because he wrote poetry on Krishna, and scores of Hindu poets of Urdu were marginalised or excluded from Azad’s influential account. As already suggested, these new literary histories created more gaps and problems than they solved. For Hindi literary historians it became a problem that hardly any Hindi manuscripts in the Devanagari script were found using Khari Boli, the form of modern standard Hindi, a lack which exposed Hindi to accusations of being a recent invention. Of course, had they looked for evidence of Khari Boli writing in the Urdu script and not limited their search to the categories of devotional and ornate poetry (the language of which was likely to be Braj), they would have found it!5 Urdu’s historical narrative had a curiously yawning gap in the middle, between Amir Khusrau’s early beginnings in the thirteenth century and the efflorescence of Urdu literary culture in the north around 1700. Why had no Urdu been written in the meantime? The later ‘discovery’ of the production of vernacular poetry in Dakkani and Gujri by Sufis and at Sultans’ courts seemed to fill this gap in Urdu’s historical narrative by way of a geographical detour, but the question about Urdu’s absence from the north Indian literary landscape still remained unanswered. Unlike other Indian vernaculars, which seemed to establish their regional literary presence either before or after the first millennium, Hindi and Urdu seemed to struggle on their own soil.6 How do Hindi and Urdu literary histories stand at the present moment? Two recent, and separate, surveys of pre-colonial Hindi and Urdu in Sheldon Pollock’s monumental Literary Cultures in History (2002) provide a useful point of reference.7 Drawing upon his unmatched knowledge of early neo-Indo Aryan and medieval Hindi texts, R.S. McGregor provides a truly erudite synthesis. His geographical map of the centres of Hindi literary production includes Gwalior, the Braj-Agra area, Orchha and Banaras, with additional centres in Avadh and Mithila.

Apart from stray evidence of the use of vernacular, the Gwalior court under the Tomar dynasty emerges in the fifteenth century as the ‘earliest identifiable centre of cultivation’ of Braj Bhasha poetry. Vernacular poetry had already started during the previous century in Avadhi with the sophisticated and ‘fully-bilingual’ Sufi romantic epics, while in Mithila, Vidyapati was experimenting with a form of Apabhramsha. Indeed, Apabhramsha seems unfortunately to fall between the cracks of McGregor’s essay and of Sheldon Pollock’s essay on Sanskrit (2002): the important role it played on a parallel with Sanskrit literary culture and as a precursor of neo Indo-Aryan literary genres, practised as it was in north India until the fifteenth century, deserved a separate treatment. By the fifteenth century, the confluence between western devotionalism and northern Nath yogi traditions produced its first north Indian Sant poets with Kabir and Raidas. In his earlier history McGregor (1984) had called the fifteenth and sixteenth centuries ‘the years of maturity’, and truly the sixteenth century witnessed the explosion of Krishna devotionalism and the consolidation of sampradayas in the Braj area, though ‘non-sectarian poets were also active and numerous’. This efflorescence is usually explained in terms of Akbar’s liberal policies, but in fact it was during the reign of the supposedly unsympathetic Sikandar Lodi that the Krishnaite sampradayas first came and ‘discovered’ Braj (see Entwistle 1987: 136). McGregor underlines the generative power of such phenomena for Braj Bhasha literature—from songs to hagiographic tales, to more erudite prabandha-kavya and theoretical texts of theology and poetics (first in Sanskrit and then also in Braj Bhasha). Parallel to this development, it was at the Bundela court of Orchha that the sophisticated style of Braj Bhasha poetry known as riti developed, a kind of poetry which took the whole of North India by storm and constituted one of the two elite forms of vernacular poetry for the next three centuries. McGregor sees Keshavdas, the accomplished initiator of this trend, as someone ‘conscious of the cultural role that is to be played by Braj Bhasha as the recipient and communicating agent of older tradition’ (2002 : 928); he mentions the ‘open access’ that Braj Bhasha afforded him to Birbal, Akbar’s minister at the Agra court, but in the context of a book on ‘Indian literary culture’ much more could have been said (as Allison Busch

has done in this volume) on the new and mixed audiences that Braj Bhasha riti poetry swiftly acquired. Ramaite poetry in the sixteenth century was clearly influenced by contemporary Krishna poetry and devotionalism, but Tulsidas’s great achievement was in producing an original synthesis which could ‘win over maximum assent for his larger view of religion and culture from the devotees of Krishna…’ (2002: 938). That he should choose Avadhi chaupai and doha as developed by Avadhi Sufis is ‘natural’ for McGregor, yet more could be said, as Thomas de Bruijn has done in this volume, about the subtle cross-influences between texts which, within Hindi literary history, have been presented as belonging to two separate streams (and two separate communities). This separateness is belied both by intertextual references and by the partly overlapping history of reception. If there is at all an agenda in McGregor’s essay, it seems to be to prove that ‘sanskritised Hindi’ was not an ‘invention’ of the nineteenth century, as both George Grierson and other critics of modern Hindi suggested, but a recurrent feature of the Hindi literary tradition. This seems indubitable, but there is a deeper level at which his essay does not question the basis of the Hindi master narrative, which sees Braj Bhasha and Avadhi (but not Urdu) as part of the cultural identity of the modern Hindi-Hindu community. As was suggested above, to take community and cultural identity as the unproblematic bases for the history of a set of traditions that are so fragmented and diversified in terms of language, region and taste made a lot of sense in the nationalist period, but a more dispassionate historical look now would view ‘textual communities’ and ‘cultural identities’ (in the plural) as changing over time. Were the many adaptations from Sanskrit into Braj Bhasha in the seventeenth century, for example, really due to ‘a wish for the reassurance of defining values of one’s culture in terms of achievements of the past during a time of social and political uncertainty’ (2002: 944–45) during the reign of Aurangzeb and after? Perhaps they were, but perhaps, also, they carried a different set of meanings. Shamsur Rahman Faruqi’s essay certainly employs a more revisionist approach, and he confronts several received assumptions in Urdu literary historiography head on. He is severe with Hindi historians for denying

Urdu’s claim to the ‘Hindi’ literary tradition. At the same time, he is also critical of traditional Urdu literary histories for having dismissed the popular, Indian aspects of Urdu, for having privileged Delhi as the centre and arbiter of Urdu literary culture and for having quietly silenced the contribution of other centres and of the many Hindu poets. Faruqi casts his net wider than older histories of Urdu, but it could have been wider still, as in Sayida Jafar and Gyanchand Jain’s five-volume history of Urdu literature before 1700 (1998). What Faruqi does admirably is to ask some fundamental questions: why is there such a gap between the first putative literary attempts at vernacular poetry by Ma’sud Sa’d Salman (1046–1121) and Amir Khusrau (1253–1325) and the first available Hindavi texts even in Gujarat and the Deccan: the Gujri songs by Sheikh Bajan (1388–1506) and Fakhr ud-din Nizami’s masnavi (1421/2) in the Deccan? And, more strikingly, why is there no Hindavi poetry and prose in north India at all before the seventeenth century, despite the fact that sources indicate that it was the common lingua franca of both Muslims and Hindus? He suggests that the efforts by elite poets such as Ma’sud Sa’d Salman and Khusrau should be considered casual and not in accordance with any established mode of writing. For this reason, they were not preserved in manuscripts. By contrast, since the Sufis addressed themselves to specific groups of followers and devotees, it was natural for their prose, poetry and speeches to be preserved in writing. Sufis in Avadh wrote in Avadhi as well as Persian, while apparently no Sufi, barring Sheikh ‘Abdul-Quddus Gangohi (1455–1538), made Hindi/Hindavi a vehicle of literary expression in the north. Faruqi suggests that the reason why Sufis did not adopt this language in the early centuries was the universal popularity and general understandability of Persian in the north, obviating the need to use Hindi/Hindavi for their popular discourse. Moreover, the popularity in the North of Rekhta, that is Hindavi and Persian mixed, seems to have retarded the growth of independent Hindi/Hindavi literature (Faruqi 2003: 837–38). In their history, Jafar and Jain have brought to light many stray vernacular verses by Sufis in north India as preserved in Persian malfuzat and maktubat from the fourteenth century onwards (Jafar and Jain 1988). What kind of Hindi they are in remains to be seen, and Faruqi’s argument that no Hindavi literary text was written (or preserved) still stands, but the

impression even from these occasional references is that Sufi saints who composed poetry were not ignorant of, or indifferent to, popular Hindi genres, an impression confirmed by Imre Bangha’s essay in this volume. Further, stray utterances and sayings in the vernacular suggest that their discourses may have been recorded in Persian, but that perhaps they were spoken in a language that at least contained vernacular expressions, much as Latin sermons did in contemporary Italy. While it is undeniable that a vernacular literary culture developed, on the basis of both popular and sophisticated Indian literary models,8 only at the regional Muslim courts of the Deccan, in Gujarat and in Jaunpur, there is nonetheless considerable evidence of vernacular literary activity also in Delhi and Avadh during the Sultanate and early Mughal periods. For example, the popularity of music and of song genres arguably paved the way for the later popularity of what came to be known as Braj Bhasha poetry among the Mughal and post-Mughal elites. Mir ‘Abd al-Vahid Bilgrami’s Haqā’iq-i Hindī (1566), for example, contained dhrupad songs and justified their use in Sufi sama’.9 Still, more pieces need to be put together. NEW QUESTIONS AND A NEW APPROACH How does this volume differ from the existing literary histories of Hindi and Urdu outlined so far? While the essays in the volume clearly do not amount to a new complete history, they do provide a new body of evidence and new categories that are needed to handle the task of reconceptualising the literary contours of pre-colonial north India. Instead of looking for precedents for modern Hindi and Urdu literary teleologies, the essays ask how we can differently envisage the literary landscape of north India before the divide. In the broadest sense, we need to ask what happened, between the fourteenth and the nineteenth centuries, as far as elite and popular conceptions of language and literature were concerned. At the more specific level of individual works, a new set of questions must be asked of texts before we fit them within existing categories. The first set of questions arises from an understanding of literary language (or dialect) as a flexible medium. Such an approach sharpens our

appreciation of nuances of register, accent, genre, alphabet and audience variation. In her study of internal variations of register within the same literary language, Braj Bhasha, Allison Busch cautions us against the danger of over-interpreting lexical choices by pre-modern poets. She observes that language practices such as Sanskritisation and Persianisation are clearly found in pre-nineteenth century Hindi texts; it is surely the meanings we assign to these practices that are candidates for re-evaluation, and not the practices themselves. (n. 93). Within the Urdu corpus, Christina Oesterheld suggests that marsiya poets used a register close to ‘Hindavi/Hindi/Braj’ because its sound qualities and connotations were deemed to be particularly suited to the expression of loss and pain. Thomas de Bruijn raises questions of dialogism and converging audiences when he brings together two works usually thought to belong to quite separate categories—Malik Muhammad Jayasi’s Sufi romance Padmāvat and Tulsidas’s Ramaite devotional epic Rāmcaritmānas. Both were written in Avadhi using the same metrical structure and were composed in the same region within a few years of each other: de Bruijn’s essay explores what else they have in common. A second set of questions arises from the re-evaluation of genres and works considered minor and marginal by both literary histories. It is true that indigenous taxonomies and quasi-histories were already developing in the eighteenth century, before the impact of colonial notions of literature and history, in some Braj riti manuals and Urdu tazkiras (poetic anthologies). But while recognising this natural process of hierarchisation within a literary culture, it is crucial that we also consider the other genres that were current in the literary culture more broadly conceived. Often, we find, such minor or ‘intermediary’ genres provide important clues about the circulation of literary tastes among different audiences. By combing through manuscripts in Persian and Urdu and catalogues of Braj Bhasha manuscripts, for example, Imre Bangha has been able to bring to light a continuous poetic engagement with Khari Boli (‘Rekhta’) in Sufi circles, courts and some devotional groups much before Urdu exploded on the north Indian scene in the eighteenth century. Francesca Orsini’s essay focuses on

another ‘intermediary’ poetic genre, the barahmasa, which is usually associated only with folk poetry. In fact it was practised by poets of the Avadhi epics, devotional and riti poets as well as more or less sophisticated Urdu poets, from the fourteenth to the nineteenth centuries. This allows her to bring together different kinds of poems in Hindi and Urdu on a common map, ‘trying to identify the levels which they occupied in the multilingual, stratified and evolving literary system of north India’ (p. 185).10 Finally, while the marsiya may not strictly be considered an ‘intermediary’ genre, Oesterheld argues that it still reveals a use of language, an audience and the creation of a particular aesthetic-emotional sensitivity on the part of Urdu poets that are all significantly different from that of the ghazal, which has dominated Urdu literary histories. Two more avenues of research, only suggested in this volume, will help in piecing together the body of material in Hindi and Urdu in order to form a new, composite mosaic. The first avenue is music history. As Nalini Delvoye has shown in the course of her numerous publications, music and singing were an area of early engagement of Sufis and local Muslim elites with Indian vernaculars, performers and music/song traditions.11 They were also an area of circulation between different geographical areas and social milieux, as the popular stories about Haridas and Tansen suggest. Lalita du Perron’s essay focuses on a later period and explores the extensive range of vocabulary of khyal songs found in some Persian manuscripts and in early recordings. Khyal songs partook of a shared repertorie, and her analysis illuminates the ways in which singers and composers were sensitive towards register and used it to convey particular moods. The second promising avenue of research is in the area of non-literary texts, like the Urdu translations of the Quran studied by Mehr Farooqi in her contribution. One of the problems that has plagued Hindi-Urdu debates on the ‘real’ language spoken by the people of Hindustan has been the tendency to use literary texts as if they were transparent evidence of language use, forgetting that, in north India as elsewhere, literary languages were more often than not, just that—literary dialects that were employed by people who would use other forms of language in the everyday. The Quran translations analysed by Mehr Farooqi, less self-consciously literary, offer much better evidence of language use among non-elite groups.

A final observation, north Indian society, we have stressed, was a multilingual society, and multilingualism remained a pervasive feature of the north Indian literary system up to the early twentieth century, as Valerie Ritter’s essay attests. One of the biggest shortcomings in the existing histories of Hindi and Urdu is that in each case the object of analysis has been taken to be a proto-national language, disregarding the multilingual context and interrelations. Multilingualism has been acknowledged only in terms of a one-to-one diglossia, with the classical language providing a pool of vocabulary and a literary model for the vernacular, respectively Sanskrit for Hindi and Persian for Urdu.12 While the essays in this volume focus on Hindi and Urdu texts and take only limited account of the other languages present in the literary system (Persian, Turki, Sanskrit, Arabic, Apabhramsha and so on), they all take multilingualism into account as a condition and a generative principle of the literary system. Thus, some of the macaronic poets analysed by Imre Bangha delight in displaying polyglot virtuosity, while for other Persian court poets, Rekhta was only an amusing sideline. Multilingualism is refracted in Allison Busch’s study of register variation within Braj Bhasha, which shows how poets put loanwords from Sanskrit or Persian to their particular use, sometimes exploiting their sound quality and at other times the cultural connotations they evoked. Within this multilingual and complex literary field, Thomas de Bruijn argues, genre was not primarily defined by a monologic and exclusive identity, but was rather a container that encompassed a paradigm of poetical aesthetics such as forms, metres and language, and a practice of reception and transmission. Moreover, the distribution of genres in the literary field referred not so much to completely distinct cultural forms but rather to different positions within a single but composite field. These positions represented different points in a complex matrix of differences that could be based on religious affiliation, clientship of particular patrons, relative distance to other traditions, caste or political grouping. This matrix was never fixed, as new positions could develop under new circumstances that called for the extension of existing genres or the definition of new boundaries or divisions. While pre-modern literary genre was thus, according to de Bruijn, dialogic, or ‘intermediary’, by nature,13 modern literary histories have tried to immobilise genres inside the divided field

(e.g. Urdu ghazal and Braj Bhasha kabitt) and have projected its categories back on to the medieval period, creating artificial divisions that are incongruous to the logic that once steered literary composition and appreciation. TOWARDS A NEW HISTORICAL NARRATIVE At present, Hindi and Urdu literary histories are trapped in competing historical narratives that do not allow a common history to emerge. The Hindi historical narrative is still largely grounded in Ramchandra Shukla’s pioneering study, first published in 1929. Shukla, as every student of Hindi will know only too well, divided Hindi literary history into four eras: the heroic era of Rajasthani epics, the devotional era that followed a move towards introversion after the ‘Muslim invasions’, the era of courtly poetry that reflected the cultural and moral waning of Hindi/Hindu energies and, finally, the modern period, when those energies were once again unleashed with the coming of British colonialism.14 Literature, in this historical narrative, was conceived as the soul of a people, which can be beaten and wounded but will ultimately resurge.15 Strongly steeped in moral and communitarian terms, this narrative clearly leaves little space for Urdu literature or for a common literary space, not to mention its deplorable imbrication in tropes about Muslim invaders and India’s supposed precolonial decline–a staple of colonial historiography. On the other side of the divide, histories of Urdu place it squarely in the context of Indo-Islamic cultural history. They follow mainstream historical treatments of ‘medieval’ north India in identifying culture with the court culture of the Mughals and their successors (Lucknow nawabs in particular), with Sufis and with the culture of the bazaar.16 Thus Sufi authors writing in Hindavi dialects other than Khari Boli, such as Avadhi, are sometimes included, but until quite recently a bhakti poet-saint like Kabir was not. And the presence of Hindu authors of ghazals and masnavis is taken as evidence of Urdu’s liberal outlook within the ‘confluence of culture’ that characterised Indo- Islamic history.

Clearly, if we want to re-imagine the history of Hindi-Urdu literary culture ‘before the divide’ we need to reject these narrow and exclusivist historical narratives of pre-colonial north India. We need a political history that is not concerned only with conquest, defeat or resistance but recognises that all polities are complex systems that involve accommodation, negotiation and competition. We need a social history that will include the whole spectrum of social groups and will not focus exclusively on courtly elites, religious figures and groups and undifferentiated ‘masses’. Above all, we need to re-conceptualise the (multiple) role of culture within this complex and multilingual polity, as a vehicle for expression, for patronage, for cultural experimentation, assimilation and/or cultural conflict. A strong awareness of the geopolitical terrain of north Indian history will also be necessary—of centres and outposts, of areas of influence and avenues of circulations—as other comparative historical enterprises of this kind teach us.17 This kind of history will have to be written jointly by historians and literary scholars. We hope that this volume will be considered an initial contribution to such a project, in particular to the methodological rethinking that needs to inform it. Since the essays in this volume take up select aspects of this history, let us now try to piece them together and tease out the pointers they offer towards envisoning a common historical narrative. The issue of beginnings is taken up with pithy clarity by Imre Bangha in his essay. He observes that claims to early beginnings have been made by Urdu literary histories with respect to Sufis like Baba Farid (c. 1175–1265) and the poet Amir Khusrau (c. 1253–1325), while Hindi literary histories trace them back to the shrouded figure of Gorakhnath or, later, Namdev. But ‘in the absence of early evidence such as manuscripts or dated references’, he argues, ‘the attribution and dating of all these poems are problematic and they may not reflect the linguistic situation of the times of their putative authors’ (p. 34). His suggestion that it is not useful to try and define the language of works that are lost (e.g. Amir Khusrau’s Hindi verses or Gorakhnath’s compositions) is providential. ‘Nevertheless’, he continues, ‘we need not be over-sceptical and should also take into consideration that “the language of Delhi” was already used and understood by literati during the Sultanate period’. Manuscripts attest the use of Persian and various forms of

Apabhramsha in the thirteenth and fourteenth centuries. But apart from the spectacularly early appearance of the Avadhi Sufi romance with Maulana Daud’s Candāyan (1379), for Khari Boli and Braj Bhasha we must limit ourselves to conjecture that poems, songs and tales in Khari Boli or Braj Bhasha templates (that is a base language or grammatical structure) circulated orally but are lost today.18 On the basis of evidence from malfuzat from the fourteenth century onwards, Imre Bangha demonstrates that Sufis and other musicians used Hindavi in their musical gatherings in north India, and this may have been a tradition going back to earlier times. It is only in the fifteenth century that we can speak of a ‘vernacular literary culture’ in north India, and it is a decidedly multilingual and multilocal one. The period between 1450 and 1550 appears particularly fertile for literature and performance, supported by a range of fiefdoms such as Lodis’ Delhi, Jaunpur, Malwa and Gwalior, which, despite pursuing different cultural and political objectives, corresponded and interacted with each other.19 The role of music in bringing local vernacular traditions to the attention and appreciation of Indian and Indo-Muslim elites cannot be overestimated, as Nalini Delvoye’s studies of dhrupad collections and IndoPersian musical texts have shown. At a more popular level, Mukund Lath has highlighted the role of itinerant singers (geyakāras) in spreading Sant songs all over north India,20 while Sufis’ contribution in the form of romances and songs is visible in Avadh, Gujarat and the Deccan. Imre Bangha’s essay draws attention to the early use of ‘mixed-language’ poetry (rekhta) by north lndian Sufis. By the early sixteenth century, he observes, ‘Dehlavi’ had become the vehicle of the Sufis of Gujarat and was cultivated in the courts of the Deccan and Gujarat as a literary language, as well as by the Jaunpur ‘Mahdi’ and by Chishti Sufis, while Khari Boli elements were also current in the mixed language of the Nirgun Sants as attested by the Guru Granth. It is from this period that the ‘rediscovery’ of Braj dates, and the establishment there of the devotional complexes that were so influential in the development of Braj Bhasha literature.21 Rather than confining our inquiry to major imperial capitals, then, we need to factor in several networks of cultural production and circulation, including khanqas (Sufi monasteries), regional courts, devotional circles, cities, qasbas and fairs.

While the advent of the Chaghatai-speaking Mughals is usually viewed as a period of flourishing of Persian culture,22 the evidence presented in this volume suggests a more cosmopolitan milieu, where experimentation with local vernaculars was not disdained and usually took the form of polyglot divertissement by the Mughal elite.23 In fact, Imre Bangha argues that evidence from Babur himself and from Rekhta verses attributed to some sixteenth-century Persian poets suggests that the earliest Rekhta writing may coincide with the beginning of Mughal times. The evidence from the time of Akbar and his successors also suggests a high degree of literary circulation and cross-influence. Bangha suggests, for example, that the relative popularity of Rekhta poems in the Nagari script among Dadu Dayal (1544–1603) and his followers may have been prompted by the similar practice of the Sufis and the increasing popularity of Rekhta at the Mughal court.24 But the literary traffic also flowed the other way round, from regional to imperial centres, as the case of Braj Bhasha courtly poetry shows. ‘From virtually the moment of its inception’, Allison Busch writes, ‘the Braj Bhasha courtly style attracted both Mughal patrons and poets—to the extent that the stunning transregional success of riti literary culture from the seventeenth century would be unthinkable without factoring in IndoMuslim communities. Whereas Sanskrit literature remained largely inaccessible except through sporadic Persian translations, riti literature was a cultural repertory in which Indo-Muslims could and did participate firsthand’, a style ‘situated at a kind of intersection, then, between Sanskrit and Persianate courtly traditions’ (p. 90). Braj Bhasha riti poetry, music and erotics were the three aspects of Indian culture that were translated for the Persian reader by Mirza Khan in his Tuhfat al-Hind (Gift from India, c. 1675). The exact volume and coordinates of this cultural and literary traffic still need to be assessed, both at a micro and macrolevel, but it is significant that a certain cosmopolitanism should be the marker both of secular and religious darbars (such as that of Dadu Dayal or of the Sikh gurus). The most significant literary phenomenon of eighteenth-century north India must be the stunning popularity of what we recognise today as ‘Urdu poetry’, traditionally associated with Vali Dakkani’s visit to Delhi in 1700.25 The growth of a distinctive and highly selfconscious poetic culture centred around the ghazal and selfdisciplined through the institution of the

ustad-shagird relationship has been well documented.26 Two significant consequences of the triumph of Urdu in eighteenth-century Mughal Delhi were the obscuring of all previous and current experimentations with Rekhta and Braj Bhasha on the part of the Mughal elites,27 and the obscuring of Urdu’s southern genealogy. Though the macaronic poetry studied by Imre Bangha continued and, in fact, increased in volume and range in the eighteenth century, it was Persianised Hindavi which became mainstream Rekhta. Yet, as Bangha importantly argues: the existence of this kind of poetry in North India before 1700 suggests that the fashion for Persianised Hindavi that Vali brought to Delhi did not ‘create’ Rekhta poetry in North India but rather displaced the pre-existing fashion for mixed language poetry (p. 87). Arguably it was this existing fashion which contributed to the quick acceptance of the new style introduced by Vali. Rekhta using Khari Boli or Braj Bhasha grammatical forms did continue, but was now associated with what were considered lesser genres, like marsiyas or barahmasas. The exact geography of the popularity of Urdu/Rekhta and Braj Bhasha among local elites, both urban and rural, still needs to be mapped: the continuing popularity of Indian song traditions, especially khyal, would suggest that Indo-Muslim elites remained familiar with Braj Bhasha poetry, at least in song form,28 while Imre Bangha notes that the success of Urdu may be the reason behind the new hybrid poetic form evolved by riti and Krishna bhakti poets. Valerie Ritter’s paper, examining evidence drawn from the literary practices of late nineteenth-century rural elites, suggests that poetic tastes were very eclectic, though often cultivated through distinct networks. Certainly, Mughal successor states like that of Lucknow show an increased patronage of local, popular forms, with courtesans embodying the whole range of literary and musical tastes that elites and aspiring elites needed to acquire. Also significant is the emergence of a popular urban culture—the culture of storytellers, street-singers and performers—that was to find expression in popular publishing and in the great success story of the nineteenth century theatre. As the evidence from nineteenth-century Hindi and Urdu barahmasas suggests, this was an eclectic literary taste that, to a much higher degree than Ritter’s Hariaudh, mixed tropes and registers.

Also, while Ritter shows that Hariaudh became increasingly illat-ease about his own eclectic poetic tastes and subscribed to the Hindi cultural project, in the domain of popular print and popular theatre, cultural nationalism could go hand in hand with linguistic and poetic hybridity.29 It is indeed ironic that the nineteenth century that produced the exclusivist discourses of Hindi and Urdu should also have produced their most eclectic mixture. If it is still true, as A.K. Ramanujan liked to argue, that every Indian is inescapably multilingual, the recognition of multilinguality as the defining feature of the pre-twentieth century cultural landscape requires an imaginative effort on the part of us literary scholars, now sadly monolingual in our research. The good news is that the first step to becoming multilingual is easy—we just need to learn another script.

2 Rekhta: Poetry in Mixed Language The Emergence of Khari Boli Literature in North India Imre Bangha

I am grateful to Govind Sharma, Dr Kishorilal, Hajnalka Kovács, Francesca Orsini, Allison Busch, Peter Diggle, John Gurney and V. Narayana Rao for their remarks on various aspects on this essay, and to Katherine Brown, Monika Boehm-Tettelbach and Azad Shamatov for drawing my attention to and providing me with some important source material. A grant from the Max Müller Fund made possible a study tour in India to examine and copy some Rekhta manuscripts. INTRODUCTION

I

n one of my classes a student was puzzled by a short poem inserted into an Urdu prose narrative. The poem had hardly any Perso-Arabic vocabulary but was written in the Urdu script as was the rest of the text. She complained that despite being a native speaker of Hindi who had learnt the Urdu script she could not tell the difference between Hindi and Urdu. This spontaneous eruption is in dramatic contrast with the political role the Hindi-Urdu divide played in twentieth-century India, manifesting itself in sentences such as Abdul Haq stating that ‘Pakistan was not created by Jinnah, nor was it created by Iqbal; it was Urdu that created Pakistan.’1

Although since the eighteenth century Hindi and Urdu have developed two distinct literary traditions, the borderlines between the two are far from being as clear as the political boundaries. The essays in this collection show that, apart from the script, the divide between Hindi and Urdu was blurred in certain intermediary literary genres even in the late nineteenth century.2 The example of the perplexed student shows that the uncertainty persists to the present day. It seems one might recognise their common linguistic and literary heritage in a plethora of north Indian vernacular dialects that from an outsider’s point of view were simply called Hindavi, (‘language of India’), or Bhakha, (‘language’), to distinguish it from Persian and Arabic on the one hand and from Sanskrit and Prakrit on the other. Instead, however, discourses on their early literature that evolved in the two languages from the eighteenth century onwards are marked by appropriation, neglect and exclusion. While histories of early Hindi literature tend to be integrative, often including the borderlands of Apabhramsha, Maithili or Dakkani, those of early Urdu either try to restrict themselves to the Khari Boli dialect and to Muslim authors3—making some allowance for Muslim authors writing in Hindavi dialects other than Khari Boli, such as those of the Avadhi masnavi tradition or, more catholically, for Hindu authors who show some input from Khari Boli. The latter approach is the one adopted from Muhammad Husain Azad’s Āb-e ḥayat (1880) to the most comprehensive recent history of early Urdu by Jafar and Jain (Tārīkh-e adab-e Urdū 1700 tak, 1998). Though this last work excludes the Avadhi masnavis, the authors are well aware of the vagueness of their approach. They give up the idea of restriction to Muslim authors on the basis that authorship is an element external to language4 and include poets central to the Hindi tradition such as Mirabai and Tulsidas because of the Khari Boli features of poems attributed to them. However, they also admit that calling this poetry Urdu would render the Hindi- Urdu distinction meaningless and therefore hail approaches, such as that of Sahil Bukhari, which examine the history of Khari Boli literature in the Perso-Arabic and in the Devanagari scripts together, showing the overlap of Urdu and Hindi traditions.5 It is indeed the most suitable approach to investigate the early development of this idiom, all the more because the use of Khari Boli is not closely linked to any

writing system. Apart from the Perso-Arabic and in the Devanagari script Khari Boli was written in Gurmukhi in the seventeenth century and later in the Kaithi script. The most influential recent study to deal with the origins of modern Hindi and Urdu is Shamsur Rahman Faruqi’s Early Urdu Literary Culture and History (2001),6 which is a close English reworking of his Urdū kā ibtidā’ī zamānā (1999). Faruqi’s view of Urdu literary history is also exclusivist, and early Urdu literary culture appears limited to Khari Boli literature by Muslim authors. In opposition to this is the general (and official) Hindi stand emphasising the composite aspect of Hindi, which encompasses a surfeit of dialects such as Avadhi, Bhojpuri, Braj Bhasha, Rajasthani, Khari Boli and others. This view is expressed in English in Amrit Rai’s A House Divided: The Origin and Development of Hindi-Urdu (1984), which in turn was based on Suniti Kumar Chatterji’s Indo-Aryan and Hindi (1942). Rai’s examples are taken from both Hindu and Muslim authors.7 The polemic is well illustrated by the search for the earliest poet. While Chatterji and Rai trace the development of early Hindi from Apabhramsha and consider Gorakhnath (eleventh century c.) and the Nathpanthis, followed by Amir Khusrau (1253–1325), to be the first Hindi authors, Faruqi suggests that Urdu literature began with Ma’sud Sa’d Salman (1046– 1121) of Lahore followed by Amir Khusrau. Both speculations are problematic, however. No Gorakhnath manuscript is available prior to the late seventeenth century,8 and thus we are not able to say with certainty what form of language Gorakhnath used. Similarly, the earliest quote from Khusrau’s Hindavi is in the Sabras of Vajhi of Golkonda written in 1636. Faruqi himself points to the fact that nothing of Ma‘sud’s and Khusrau’s Hindavi9 corpus is available today, and that ‘the first person whose Hindavi survives in substantial quantity, and with whom Urdu literature can seriously be said to begin’, is an author not from the north but from Gujarat, namely Shaikh Bahauddin Bajan (1388–1506) 10 of Ahmadabad. REKHTA AS MIXED POETRY

The Persian word rekhta (‘poured, interspersed, mixed’) had several technical meanings. Prior to the eighteenth century, it was part of musical terminology. It also referred to a mode of writing, namely to poetry written in a language that mixes lines, phrases and vocabulary from Hindi and Persian (the reference to Persian also includes the Arabic vocabulary imbibed by Persian), in which the Hindavi component is normally Khari Boli and sometimes Braj Bhasha or a mixture of the two.11 As a musical term, Rekhta appears in Alauddin Barnavi’s musicological treatise Cishtiya bihishtiya (1655). Barnavi defines Rekhta as a kind of text in which one sets the words of both languages to a raga and a tala.12 Although Mahmud Sherani’s suggestion that this definition of Rekhta originated with Amir Khusrau cannot be substantiated, this passage indicates an early link between Rekhta and Hindustani music.13 It is in the same year that another occurrence of Rekhta in the sense of ‘mixed language’ appeared in the colophon of a manuscript of Vajid I am to discuss below. So far no documentation has been found of the same technical use of the term prior to the mid-seventeenth century. In the eighteenth century, Rekhta appears also as the name of Khari Boli mixed with Perso-Arabic vocabulary—the language which is today called Urdu. The greatest Urdu poet of the century, Mir, referred to his language not as Urdu, but either as Hindi or as Rekhta.14 The meaning of the word, however, varied even within Mir’s usage: he used rekhta interchangeably with shi’r (verse).15 In his tazkira Nikāt ush-shu’arā, for example, he called rekhta ‘poetry which is in the style and manner of Persian poetry, but in the language of the exalted court of Delhi’.16 In this work, he distinguished between six kinds of rekhta, two based on style and four based on the linguistic mixing of Hindi with Persian, i.e. (1) Persian and Hindi lines alternating; (2) the same line half in Persian and half in Hindi; (3) the use of Persian verbs, prepositions and conjunctions within a Hindi line, and (4) the use of appropriate Persian phrases in Hindi.17 (The list in fact is not exhaustive, and Ali Jawad Zaidi in his History of Urdu Literature presents yet another linguistic type: (5) Persian couplets with Hindi refrains.18) In the eighteenth century Mir, as his poems illustrate, only approved of the fourth type and normally used the word Rekhta in a restricted sense to refer to the fourth type only. For the sake of convenience, I will refer the other

forms of rekhta as ‘macaronic’ poems, borrowing the term from Italian renaissance literature, where the word maccheronico referred to mixed Latin-Italian verse.19 The fact that in Gujarat or in the Deccan the mixed language was not called Rekhta, and Mir’s consciousness that he wrote in Rekhta suggests that in the eighteenth century Urdu poetry was perceived as the inheritor not only of Dakkani but also of earlier Rekhta experiments in north India. In this essay, if not indicated otherwise, I will call Rekhta any poetry in either the ‘extended’ Persian, Gurmukhi, Kaithi or Nagari script which consciously mixes the vernacular Hindavi (including Braj Bhasha) and the cosmopolitan Persian. This Rekhta is different from Sadhukkari, the spontaneously mixed literary language of the Sants, that blends elements from various north Indian dialects and languages. Although Sadhukkari may include Sanskrit and Arabo-Persian words, it is a spontaneous blend of several vernaculars. As we shall see, in north India, Rekhta was a literary idiom that was (a) first practised in certain Sufi circles from the early sixteenth or maybe late fifteenth centuries, (b) patronised in the Mughal court, (c) taken up occasionally by Nirgun sants, (d) by some Sikh authors of the Janamsakhis, (e) by Krishna bhaktas, (f) by syncretistic authors and (g) court poets in Rajasthan in the seventeenth and particularly in the eighteenth centuries. Early Rekhta poetry in the Persian script figures to a greater or lesser extent in most histories of Urdu literature,20 but no study exists of its counterpart in the Nagari and occasionally in the Kaithi or Gurmukhi scripts. Nagari Rekhta (i.e. Rekhta in these three scripts) was composed by mainly Hindu Vaishnava, Sant or Sikh authors since the late sixteenth century. It is based on Khari Boli (the dialect associated with Delhi in its origins) and written down in the Nagari, Kaithi or Gurmukhi script either at the moment of composition or later. This Rekhta is a literary language with a (usually loose) Khari Boli template (that is a base language or grammatical structure) and a relatively high Perso-Arabic vocabulary compared to Braj works. In this way, although linguistically not different from Urdu, which was also called Rekhta in the eighteenth century, Gurmukhi Rekhta is part of the Panjabi Sikh tradition, while its Nagari and Kaithi counterparts are included in Braj Bhasha or the Sant devotional

tradition. As such, Nagari Rekhta is entered in literary histories and manuscript catalogues along with Braj Bhasha texts since it tends to use the metres and themes of Braj and Nirgun Sant literature. Many of its authors wrote the majority of their works not in Rekhta but in Braj Bhasha or in Sadhukkari, the mixed language of the Sants. In spite of using Khari Boli and the Nagari script, this genre was not hailed as the precursor of modern Hindi literature, even though Rekhta was produced well into the nineteenth century and was, directly or indirectly, influential in the development and acceptance of modern Hindi. If the history of Nagari Rekhta is taken into consideration, then modern Hindi should not be considered as a language originating only from the artificial experiments of Fort William College but also as the continuation of a nowforgotten literary idiom. Yet Rekhta became neglected from the 1850s onwards, the time of Bhartendu Harishchandra. Instead of allying themselves with this literature, Bhartendu and his circle fought against ‘Urdu Begam’ and should probably be held responsible for denying the existence of literature in Nagari Rekhta as a possible meeting point between Hindi and Urdu. Today it is only a small group of Braj Bhasha scholars who know about the trend of writing Krishna poetry in Nagari Rekhta, current mainly in the eighteenth century. Although Nagari Rekhta is incomparably smaller in its output than the mainstream Braj Bhasha or Sant literature, and its poetry has not exercised such influence as the Avadhi narratives, there are beautiful pieces in it. Indeed, many of the best poets of the eighteenth century, such as Anandghan, Nagaridas and Brajnidhi, tried their hands at Rekhta, along with many lesser-known authors like Manohardas and Rasrashi. In this essay I will present a sketch of the history of Rekhta and Khari Boli poetry in north India. In the first part, I will talk about the unsubstantiated claims to early Rekhta and describe the development of Rekhta through its sixteenth-century extant versions manifest in the different varieties of macaronic poetry written chiefly in the Perso-Arabic script that were marginalised after the success of Vali’s Dīvān in Delhi. In the second part, I will show how in seventeenth century north India, Nagari-

script Rekhta coexisted with sporadic Urdu-Rekhta, and I will follow up its record in sectarian and court literature until the mid-nineteenth century, when it became neglected due to the exigency of defining clear linguistic and literary boundaries. A fundamental difficulty in writing the early history of Hindavi is the lack of philological background work to the texts studied. Even when we have critical editions based on manuscripts, we cannot be sure that the text in a later manuscript represents the same linguistic situation as at the time of its composition. One cannot state with certainty that the text of the critical edition of works such as the Bikaṭ kahānī, based on manuscripts dating from at least a hundred years after the death of the author, corresponds to the language of its birth in the early seventeenth century. As we will see, traditional attributions to early authors found in relatively late handwritten books are far from reliable since it was common in early modern South Asia to link poems to the prestige of established names. But the undependability of manuscript transmission is only one of the many problems. An immense part of early Hindavi literature still lies unpublished in manuscript collections, and the picture that we can get on the basis of published material is bound to be distorted. The published material is, more often than not, available in publications whose principles are far from that of a critical edition. The editors often standardise not only the orthography but also the language. Studies on early literature often give examples without specifying their sources and in this way the reliability of their quotes is uncertain. This paper aims to follow up the emergence of Khari Boli literature in north India by a search for works in early dated manuscripts. By using this material as a point of reference in language and style, poems with less reliable transmission can be examined comparatively. In this way I will present traditional attributions to sixteenth and seventeenth-century poets when the styles of the individual works are consistent with that of other works found in dated manuscripts. A philological approach is by definition restrictive since it cannot take into consideration the rich oral tradition that is almost impossible to document today. Already Amir Khusrau mentioned that he had composed poems in Hindavi, and there must be other Indo-Persian poets who also did so. This Hindavi poetry, however, did not initially enjoy much prestige and

was probably never committed to writing. We do not have many documents about the spread of the speech of Delhi, the ‘Dehlavi’, throughout the Delhi Sultanate as a lingua franca. It was from this lingua franca that the first documented literary languages based on Khari Boli, namely Dakkani and Gujri, emerged in areas south of the modern ‘Hindi belt’. Although poetry with Khari Boli features or macaronic stanzas may have existed in north India prior to the sixteenth century, due to the lack of reliable sources observation on the nature of such material can be more than conjecture. CLAIMS TO BEGINNINGS Khari Boli literature, like that of all modern languages, emerged at a certain point in history. It can be argued that the spoken language that, in all probability, had links with the literary Shauraseni Apabhramsha of northwestern India, developed into idioms of which Khari Boli was one of the literary versions. Some Hindu scholars argue for the continuity of linguistic forms in the literary languages Apabhramsha and Khari Boli. Their ideas are supported by lines such as the one from the Apabhraṃśaprakaraṇa of Hemachandra’s śabdānuśāsana. Bhalā huā ju māriā bahiṇi mahārā kantu (śabdānuśāsana 8, 4, 351)21. It is good that my husband was killed, my sister. This line indeed shows Khari Boli and Punjabi features indicating a period of development when the two idioms were not separated, a phenomenon attested also by the Punjabi elements of early Dakkani. Most of the Khari Boli features in this work and in other Apabhramsha compositions, however, are isolated instances and do not suggest any use of systematic Khari Boli as can be seen from the second line of the same couplet: Lajjejjaṃtu vayaṃsiahu jaī bhaggā gharu eṃtu I would have been put to shame among my friends, if he had run away (from battle) and come home.

It might therefore, be more fruitful to examine extant material that shows a systematic use of Khari Boli. Probably the best known example of macaronic Rekhta is a popular poem attributed to Amir Khusrau Dehlavi (c. 1253–1325) in which the first half of each line is in Persian and the second in Braj Bhasha (Perso-Arabic words are in bold type): Zi ḥāl-ī miskīn makun taghāful, durāye naina banāye batiyā̃ Ki tãb-i hijrãn na dāram ai jān, na lehu kāhe lagāye chatyā̃ … Ba-ḥaqq-i ān rūz-i faṣl-i maḥshar, ki dād mā-rā fareb Khusrau Samīpa man ke davāri rakhū̃ jo jān pāū̃ parāī rakhyā.22 Do not be negligent towards this poor one—You hide your eyes and invent excuses. Since I do not have the strength to bear the separation, o my love, why don’t you embrace me at once?… I swear by the Day of Gathering that she deceived me, Khusrau, I will keep a sentry near my heart if I find my beloved guarded by someone else.23 A similar composition is attributed to Khusrau’s contemporary Amir Hasan Sijzi Dehlavi (d. 13 3 7)24: Har laḥz̤ a āyad dar dilam dekhū̃ ūse ṭak jāy-kar; Gūyam ḥikāyati ḥijr-i khud bā ān ṣanam jīū lāya kar…. Bas ḥīla kardam, āy Ḥasan, bejān shudam az dam ba-dam; Kaise rahū̃ tujh jīū bin tum le gae jang lāya kar.25 Each moment I felt like going out to catch a glimpse of her, setting my heart on that beloved I myself tell the story of my separation…. I tried many ways, O Hasan, at every instant I lost my life. How can I remain without you, my life? You have taken it and brought war. There are similar claims for being the precursors of Khari Boli and of Rekhta literature by thirteenth- to fifteenth-century Sufi poets, such as Baba

Farid, Farid’s son-in-law. Ali Ahmad Sabir Kaleri (Kalyari) (d. 1265), Shaikh Hamiduddin Nagauri (d. 1273), Ali Qalandar Panipati (d. 1363), Shaikh Sharafuddin Maneri (d. 1381) and others.26 There are also candidates for early Rekhta and Khari Boli authorship with a Hindu background, such as the fourteenth-century poet-saint from Maharashtra, Namdev: Sāvadha sāvadha bhaja le re rājā; nahī̃ āvai aisī ghaḍī jū. Uttama naratanu pāyā re bhāī; gāphila kyõ huvā divāne jū. Jinne janma ḍārā hai tuja kūṃ bisara gayā unakā gyāna jū. Phira pastāyegā dagā pāyegā; nikala jāyegā āvasāna jū…. Aisī bāta sunake nāmā̃ sāvadha huvā; guru ke pāva miṭhī ḍārī. Mai ānātha dubale śarana saye tuja kū; āba jo merī lāja rākhī jū. (Pad 192)27 Worship the King with complete alertness; you will never have the opportunity again. O my Brother, you received the highest estate, a human body; why have you become so lazy, so negligent, so deranged! You abandoned the knowledge of the one who conferred life upon you. You will repent, will be cheated, and you will go away in the end… Hearing such words Nāmdev became alert and took the dust of the guru’s feet. I took refuge with you as a frail orphan; now, please protect my honour. In the absence of early evidence such as manuscripts or dated references, the attribution and the dating of all these poems are problematic and they may not reflect the linguistic situation of the times of their putative authors.28 To illustrate the pitfalls of traditional attributions let us have a closer look at the most famous of these early Rekhtas, namely that of Khusrau. As has been mentioned, no manuscript evidence for his Hindavi exists prior to the quotes in Vajhī’s Sabras (1636). The rekhta quoted above first emerged as Khusrau’s in the album of Partab Singh copied in 1719. After that, the poem appeared in tazkiras under the name of Khusrau. The same rekhta, however, is also present in an earlier album dated 1652/1656,

which was in possession of Mahmud Khan Sherani. Here, however, the takhallus, pen name, inserted into the last but one line is not of Khusrau but of a certain Ja’far, about whom nothing is known, Ba-mihr-i ān śokh carkh-i bad-mihr (ki) burd mā-rā śikeb Ja’far. Su pītā man mãh durāya rākhū̃ jo tūha pāõ parāna katiyā̃ .29 I swear by the love for that coquettish beauty that the unkind fate took away my patience, Jafar,30 I will hide my love in my heart, when I am killed at your feet. Moreover, Sherani demonstrated that the sixteen-morae form of its metre (∪ - ∪ / - - / ∪ - ∪ / - -) called mutaqārib fu’ūlu fi’lun shānzdah ruknī was not used before the mid-fifteenth century.31 Nevertheless, we need not be over-sceptical and should also take into consideration that ‘the language of Delhi’ was already used and understood by literati during the Sultanate period. It is also possible that there were poems in a Khari Boli template circulating orally that are lost today, since by the early sixteenth century ‘Dehlavi’ had become the vehicle of the Sufis of Gujarat and was cultivated in the courts of Deccan and Gujarat as a literary language. Khari Boli elements were also current in the mixed language of the Nirgun Sants, as attested for example in the vocabulary of the Guru Granth, where verbal and pronominal forms such as kiā, gaiā, mujha, tujha, mujhai, tujhai, tumhārā/tumhārī/tumhāre etc. figure in abundance. The search for mixed Hindavi-Persian, Rekhta, and for Khari Boli features shows that most early claims link Rekhta with Muslims rather than with Hindus, raising the expectation that the use of Khari Boli and of Rekhta was more closely linked to Muslims. But can anything at all be known about the literary Hindavi that Muslims used during the Sultanate period? After all, the dialect of the Hindavi romances was Avadhi, ever since Maulana Daud’s Candāyan (1379). References to the use of Hindavi as well as Hindavi phrases and sometimes even poems are embedded into Persian works, such as letters or the discourses (malfuzat) of leading Sufis delivered to a select gathering of

disciples and visitors.32 These discourses were embellished with didactic poetry, anecdotes and apophthegms. In the absence of early Hindavi manuscripts, it is in the works of the malfuzat genre that a systematic and critical search can reveal the earliest recorded occurrences of Hindavi poetry. Although works of this genre may date from centuries after the death of the pir, some of them are reliable sources of information about the times of the sultanate. Some malfuzat were discourses recorded soon after they were delivered by a spiritual master and some were collected by a descendant or disciple of a Sufi after his death. The most important source for early Hindavi, the Surūr aṣ-Ṣudūr33 belongs to the second category. It contains the sayings of Shaikh Hamiduddin Nagauri (d. 1273), the successor of Khwaja Mu'inuddin Ajmeri,34 as recorded by his grandson and successor Shaikh Fariduddin bin Abdul Aziz (d. 1334).35 What is attested in works from the fourteenth century onwards is that Sufis and other musicians used Hindavi in their musical gatherings in the North, and this may have been a tradition going back to earlier times. A spiritual discourse of Nizamuddin Auliya, dated 1316, tells us how the weaver Shaikh Ahmad Nahravani (fl. 1235)36 became the disciple of Faqih Madhaw, the imam of the Jami Masjid at Ajmer, who had been entranced by Nahravani’s Hindavi song and ‘said to him that it was unfortunate that he was just wasting his melodious voice in singing Hindavi songs, and advised him to memorise the Qur’ān.37 The use of Hindavi is also associated with the person whom Nizamuddin Auliya credited with the introduction of sama’ singing to Delhi, namely Shaikh Hamiduddin Nagauri (d. 1273).38 The Surūr aṣ-Ṣudūr quotes several Hindavi verses attributed to Hamiduddin: Jo bistārai to sabai sikata (jo) saṃkoya;39 Sau sau eka puruṣa ke nāṃva biralā jānai koya.40 Everything that expands, basks and shrinks: Hundreds of names for the one God—the outstanding man understands it. The Hindavi of this poem (with the exception of the -ā ending in the word biralā, which suggest Khari Boli usage) and that of the other ones

found in the same work41 is what later was called Braj Bhasha. This instance alerts us to the fact that the Hindavi favoured by north Indian Sufis in their gatherings was probably closer to Braj Bhasha than to Khari Boli or Rekhta.42 REKHTA AT THE MUGHAL COURT In the sixteenth century Rekhta seems to have been practiced both in Sufi circles and in the Mughal court. From Babur’s evidence below and the existence of Rekhta attributed to other sixteenth-century Persian poets such as Saqqa, Mu’aiyid and Mashhadi, one can argue that the earliest Rekhta writing may coincide with the beginning of Mughal times. One might even suspect that the poems attributed to Khusrau and to the other poets mentioned above date from this century. Prior to the early eighteenthcentury success of Vali’s Dīvān, however, no serious effort was made to record Rekhta poetry in the north. The lack of manuscripts is indicative of the neglect of poetry that had not found its way into a larger composition and also suggests that the use of Rekhta must not have been very widespread, or that it may have been an oral genre considered too frivolous or undignified for committing to writing.43 An important pre-Mughal religious lineage that used Rekhta is that of Miran Sayyid Muhammad from Jaunpur (d. 1505),44 who after a pilgrimage to Mecca travelled widely in India including Gujarat and Bidar in the Deccan and eventually died in Baluchistan. In 1497, in Ahmadabad, probably prompted by the approaching millennium of Islam in 1591/2, he declared himself the Mahdi, the leader who is expected to rise before Judgement Day. He is credited with the use of mixed language.45 Among the nine couplets attributed to him in various sources, one is in the Rekhta form in a Perso- Arabic metre: Agar faz̤ al kunī yak jau jīve jīve jīve; Agar ‘adal kunī yak jau muve muve.46 If you have mercy the size of a grain, then you live, you live, you live.

If you administer (mere) justice the size of a grain you die, you die, you die. His other examples, however, use Indian metres and thus his poetry is in line with the earlier tradition of Hindavi poems used in sama’ gatherings, and also with his contemporary, Abdul Quddus Gangohi, whose similar poetry under the pen name Alakhdas survives in considerable quantity.47 Although Sayyid Muhammad was banished from Gujarat, a large number of ulama and sufis accepted him as the Mahdi48 and his followers are known as the Mahdavis. His often-persecuted successors in Gujarat and Rajasthan also used Hindavi. The most important of them were Sayyid Khwandmir Matufi,49 who fell fighting in Patan in 1524, and Mian Mustafa (d. 1577),50 who at times was persecuted by local authorities and at times summoned and rewarded by Akbar. The Mahdavi poems suggest that Rekhta was initially cultivated in religious circles and appeared in north India as the result of the wanderings of religious personalities probably connected to the Deccan or to Gujarat. It is, however, the Mughals to whom the earliest firmly datable Rekhta can be attributed, and its author is none other than Emperor Babur. His Turkish Dīvān, preserved in a manuscript dated from 1529, includes a couplet partly in Khari Boli Hindi, partly in Persian and partly in Turki:51 Muj-kā na huā kuj havas-i mānak-o motī; Faqr ehliga bas bulgusidur pānī-o rotī. I had no desire for gems and pearls, For poor people, sufficient are water and bread. After Babur, there is not much dated early manuscript evidence for Rekhta in the north for more than a century. The most important manuscript is the album written by Jaimal Thal in 1652–56, which contained Ja’far’s Rekhta that was later attributed to Khusrau. Apart from Persian compositions, this album contains poetry in the mixed language by several poets, as can be seen from the pen names: Jamali, Faizi, Bairam, Jani, Sedan, Fatah Muhammad, Ja’far and an unknown author.52 With the exception of Fatah Muhammad, all these authors produced macaronic poetry with Persian

template. One poem later attributed to Khusrau is in Persian but two words can be read as a pun and be interpreted as Rekhta, Guftam gahī dar khāna-yi ma’mūn-i tū bāsham; Guftā ki darīn khāna balā’īst mamānī.53 I said that I would be for a while in your safe house; She said that calamity resides in this house—don’t stay! And reading it with the Hindavi meaning of the words māmūn and mumānī, I said that I would be for a while in the house of your uncle. She said that her aunt was a calamity in that house. There is a poem in the same album written by someone under the pen-name ‘Jamali’. ‘Jamali’ must be identical with Maulana Hamid bin Fazlullah known as Shaikh Jamali Kamboh (d. 1536). He travelled to Mecca and to other Muslim lands and in his later life lived in Delhi as a member of the Suhrawardi order. He maintained good contacts with Babur and Humayun whom he accompanied on his expedition to Gujarat. He was a Persian poet and author of Siyār ul-’ārifīn (Biographies of Holy Men, c. 1530–36), a tazkira on the lives of Chishti and Suhrawardi holy men dedicated to Humayun.54 Jamali’s poem in the album is in Persian with an abundance of Hindavi words:55 Ān parī rukhsār cūn shāna ba cotī mīkunad, jān darāz-i ‘āshīqān-rā ‘umr-e chotī mikunad. Cashm-rā qaṣṣāb sāzad khanjar az ghamza zanad; ‘ishqbāzān-rā judā botī zi botī mīkunad. Cūn zanad khanjar ba jānam khūn zi jānam mīcakad; hamcū murg-i nīm basmal lot-potī mīkunad. Bar darat āyam raqīb-at gūyadam dar khāna nīst; īn cunīn kambakht bā mā bāt khotī mikunad. Dar rah-i ‘iśhqat Jamālī gashta (cūn) zār-o-nazār; ‘āqibat az muflisī dar tah lãgotī mīkunad.

When that fairy-faced woman combs her hair, she makes the long life of the lovers short. She makes her eye into a butcher and turns her glances into a dagger. She cuts into pieces the enamoured ones. She thrusts her dagger into my heart, blood drips from it and it rolls about like a half-sacrificed fowl. I come to your door but your doorkeeper /my rival/ tells me that you are not at home; that wretched one speaks to me so falsely On the path of your love, Jamali has become miserable. Out of poverty, in the end, he ties a loincloth in several layers on his waist.56 Another, incomplete Rekhta of Jamali which, according to Sherani, is present in several tazkiras and albums, uses similar phrases: Khvār shudam zār shudam luṭ gayā; dar rah-i ‘iśq-i-tū kamar ṭūṭṭā hai. Garci badam guft raqīb-i kuṭṭan; uskā kahā mat karo yah jhūṭṭā hai. Gāh nagufta ki Jamālī tū baith; tham karo, kyā apnā karam phuṭṭā hai.57 I became wretched and weak; I am plundered. On the path of your love one breaks his back. Though my mischievous rival spoke ill of me, don’t do what he says, he’s false! Didn’t he say many times, ‘Jamali, sit down here, pull yourself together, is your fate broken?’ Though Ja‘far’s Rekhta in the album mentioned above used Persian halflines mixed with Braj Bhasha, the blend of Persian and Khari Boli of this last poem proved to be more popular among sixteenth-century authors. There are even more occasional Rekhta poems by Persian poets from Humayun’s and Akbar’s time, as the examples of Bahram Saqqa Bukhari and his contemporaries Mu’aiyid Beg Kur, ‘the Blind’, and Mashhadi show. The identical rhythm and the rhymes of their Rekhtas suggest that all may have been written for the same poetic gathering, and they are a one-off

experiment.58 The first quotation is by Saqqa, the second by Mu’aiyid and the third by Mashhadi, [1] Bāz hindū bacca-ī qaṣd-e dilam dhartī hai; kūch nahā̃ jāno az īn khasta (ki) kyā59 kartī hai. Cīn bar abrū zada barbasta kaṭāra60 ba-miyām; cal cal ay dil manigar, tūjh kane61 ū lartī hai.62 Hāt mahandī lāyhā63 dast farū burda ba khūn; kih base kushta zi dastān- i gham-ash martī hai…. Cup kar ay dil shuda Saqqā zi gham-e yār manāl; gar jafā raft ba jān tū miyān karti hai. An Indian girl wants to take my heart again. — You do not know what she is doing to this poor one! With brows knit and the dagger tucked at her waist, Go, get away, o my heart, do not watch, she is fighting with you. The henna on her hand is as if she had plunged it into blood, That many die at the hands of grief for her!… Be silent, o lost-in-love Saqqa, don’t lament the sorrow caused by the beloved, If torment penetrates your soul, you act as its sheath. [2] Hargah ān sāqī-yi-hindī ki t̤ arab kartī hai; kāsa-yi mai zi sharāb-i-lab-i khud bhartī64 hai. Khwāham aḥvāl-i dil-i khẉ īsh bigūyam bā tū; lek az nāzukī-yi t̤ ab ‘(yi) tū-am dartī hai. Gasht cūn qiṣṣa-o-afsāna ba har pīr-o-javān; ki Mu‘aiyid zi gham-i ‘ishq- i bate marte hai.65 Whenever the Hindi Saqi is making merriment she fills my cup with the wine of her lips.

I want to tell you the condition of my heart but I fear because your disposition is so fragile. The news, like stories and fairy tales, reached all, young and old, that Mu'aiyid dies out of his love for an idol. [3] Hindū-y cashm-i tū goftam ki ba-man laṛtī hai; raft dar khanda o goftā ki mughal ḍartī hai.66 I complained that her black eyes were ever at war with me. She burst into laughter, ‘Oh! the Mughal is afraid’. Common to the authors of these Rekhta poems is that they are Persian poets and their Rekhta is scant. While they use the Persian ghazal form, their rhymes and end-rhymes are in Hindavi and the rest of the lines are to a varying extent in Persian conforming to Mir’s third category, namely the use of Persian verbs, prepositions and conjunctions within a Hindi line. The feminine is not a fortunate solution in the rhymes. We might understand this use of the feminine in several ways. It could be due to Dakkani influence, where the protagonist can be feminine. Or perhaps it was used for the masculine in this from. Or else, the ī verb ending (as in dartī) could have been pronounced as -e for the masculine plural. Some critics have argued that this confusion is due to the authors’ limited aquaintance with Hindavi.67 Jaimal’s album also contains a poem with the takhallus ‘Bairam’. Bairam is a Turkish name meaning ‘the Festival of ‘Id’. The only literary figure known to us with this pen-name is Abdurrahim’s father, Bairam Khan (1525–1561), author of Persian and Turki Dīvāns.68 Even though the poem is not found in these Divans, and the attribution to Bairam Khan may be doubtful, we cannot exclude the possibility that the poem was composed by another person of this name. The Turkish pen-name suggests Mughal authorship. Dilā kun yād-i ān sā‘at darūn-i gor jab sove; ‘azāb-i sakht-tarīn69 bāshad ki lohū ānsuvān rove.

Na ānjā khẉish nai qurbat na sāthī bāp aur bhāī; na zan farzand ko belī dar-ān tārīk tanhāī. Biāyad jānsitān nāgah cū70 malak ul-maut darbārat; jo haigā jīv kar sancā kunad dar yak zamān gārat. Tihī raftand ān mardum jinhõ ke lākh the pāle; na bā-khud burd yak jītal ki rete hāth uṭh cāle…. Gumān dāram dar īn dunyā do gaz ghar bās arū mātī; pasārā dūr kar candīn cū luqmān bāndh rah tātī. (ki) Bairām naqd jo hove (to) ṣarf-i rāh-i ū kīje; are jo chāḍkar jānān harā’in khāe le līje.71 O heart, remember the hour when you’ll sleep inside a tomb. There will be such terrible torment that you will weep tears of blood. There is no family, no kinsman, no companion, no father or brother. No woman will protect her child in that dark desolateness. Suddenly the Angel of Death, Taker of Souls, arrives at your court. Whatever happens, concentrate in your heart, because it plunders you at once. Millions who were nourished72 were not even to take one jītal73 when they left with empty hands…. I am proud in this world to have two yards for home and soil. Running around, I have stretched as much as Luqman who fenced the road.74 Bairam, spend the money you have in God’s path; oh, if you abandon the Beloved, consider it defeat. There is another poem in the same album, with the pen-name Faizi. At least two known poets are candidates for its authorship. One is Sheikh Allah-dad Sirhindi, the author of the dictionary Madār ul- Afāẓil (Pivot of the Most Learned Ones, 1592) and of a contemporary history called Akbar-nāma (1601),75 the other one is Abu’l-Fazl’s brother and Akbar’s poet laureate Abu’l-Faiz Faizi (1547/8-1595). The poem in this album follows a Persian template with a strong input of Hindavī words:

Ay ān-ki hast la’lat cūn āb-i zindagānī; tā tishna lab namīram īnak pulāo pānī. Guftī fasāna gūyam jānān ba jān va lekin; tū sust man pareshān ky kar bane kahānī. Ay dil zi la‘l-i jānān kāmam nagasht ḥāṣil; zīrā-ki zar nadāram vo bastu hai birānī. Man dardmand-i ‘ishqam bar man kanūn vafā kun; ‘umram guzasht dar gam nis jāgte bahānī. Ba-shnū to faizī az man ba-gzār rū-yi jānān; tū ‘āshiq-ī va sāda vo zāt hai sayānī.76 O, you, whose ruby-lips are like the water of life; behold, here is rice and water, so that I don’t die of thirst. Oh beloved to my heart, you told me to tell my story, but you are lazy and me disturbed; how could the story build up? 0 my heart, my desire got no fulfilment from her ruby-lips, since I do not have gold and she is a treasure belonging to others. I am afflicted by love, fulfil your promise now. My life was spent in sorrow and I spend my nights awake. Listen to me, oh Faizi, leave the face of the beloved! You are a lover and are honest, and she is of a clever race. The above attributions to Bairam and to Faizi (whichever one he may be) remain somewhat uncertain. Nor can we exclude later appropriations of a famous author’s name as was the case with Khusrau. Nevertheless, the identification of Faizi and Bairam with Mughal noblemen would fit well with the syncretistic picture of the Mughal court. Other sources, too, seem to corroborate this attribution.77 As Babur’s example showed, it was not only Persian that was cast against Hindavi in Rekhta poems. The experimenting spirit at Akbar’s court is attested by the macaronic poems attributed to Abdurrahim Khankhanan ‘Rahim’ (1556–1627). He mixed Khari Boli with Sanskrit and used not the ghazal but the quatrain form in Sanskritic metres. His Madanāṣṭaka is in the malini metre, while the following poem in shardulavikridita:

Ekasmin divasāvasānasamaγe maĩ thā gayā bāga mē̃; Kācit tatra kuramgabālanayanā gula toratī thī kharī. Tāṃ dṛṣṭvā navayauvanām śaśimukhīm maĩ moha mẽ jā parā ; No jīvā mi vinā tvayā śṛnu priye tū yāra kaise mile78 One day at dusk I went to a garden Where a woman with eyes like a young gazelle stood plucking flowers. Glimpsing that woman in her prime, whose face shone like the moon, I fell in love. I do not live without you, listen my beloved, how can I meet you? It should be mentioned that multilingual compositions are not exceptional in Indian literature, or indeed in any literary culture which is either multilingual or is marked by diglossia between a classical language and vernacular(s).79 Sanskrit dramas already used different Prakrits according to the characters’ role and social status. Dialogue across linguistic boundaries in early modern times was also alive in south India both in historical writings80 and dramatic literature.81 Mixing idioms did not stop at drawing on two languages. The Maharastrian Jayarama Pindye’s Rādhāmādhavavilāsacampū used twelve, while in north India a quatrain written in Sanskrit, Braj, Gujarati, Marathi, Rajasthani, Khari Boli, Punjabi, Persian/Arabic and Telugu, a real virtuoso performance, is attributed to the above-mentioned Rahim: Bhartā prācīṃ gato me (Sanskrit) bahuri na bagade (Braj) ŝhū̃ karū̃ re have hū̃ (Gujarati) Mā̃ jhī karmāci goṣṭhi (Marathi) aba puna iśanasi (?) gāṃṭha dholo naī ṭhe (Rajasthani) Mhārī tīrā sunerā (Rajasthani) kharaca bahut hai (Kharī Bolī) īharā tābarā ro (?) Diṭṭhī taiṃḍī dilõ dī (Punjabi) iśaqa ila fidā (Persian/Arabic) oḍipo baccanādū82 (Telugu) My husband went east and is not coming back—what shall I do now? This is my fate.83 Please listen, I do not have a coin in my purse.

Listen to me, the expenses are high and there are many in the family. In order to see him I sacrifice my heart for love.—‘It is he who is coming!’84 All the Rekhtas surveyed in this section, apart from Babur’s couplet and possibly of Saqqa, Muaiyid and Mashhadi, are preserved in later manuscripts. Yet the relative abundance of macaronic Rekhta from the sixteenth century makes it difficult to question their authenticity on the same grounds as for Baba Farid, Amir Khusrau, Hasan Dehlavi or Namdev.85 In the following section I will consider the possible motives that induced poets to compose such macaronic poems. MOTIVES FOR LINGUISTIC HYBRIDITY Much research has been done in recent years on language choice in India’s multilingual society. Some scholars explain language choice through motives external to the language and its literary culture, while others search for internal forces. Some influential modern theories explain it in terms of its teleological contribution to some project such as proselytising or integration. According to an early idea of Richard Eaton, based on Annemarie Schimmel,86 Sufis in Bijapur adopted Dakkani Hindi as an instrument of proselytisation.87 In a similar vein, Muzaffar Alam has explained the Mughal choice of Persian on the basis of its non-sectarian aspect, which made Persian an effective tool for negotiating difference within Indian society and thus contributing to the consolidation of the empire.88 In his later research, however, Alam seems to have abandoned these ideas.89 Another theory based on external motives has been put forward by Sumit Guha, who has examined language choice in the earlymodern Maratha region in connection with the power of patronage. According to Guha, languages were marked by a tension between hybridisation and identity. In the case of administration the use of a vernacular invoked shared ethnic and territorial rootedness, while the higher Persianate register signalled cultural superiority as well as a wider subcontinental identity.90 He has argued that the same phenomenon can be seen in poetry and was sometimes used to

display poetic virtuosity as in the case of the above-mentioned Rādhāmādhavavilāsacampū. Moving away from political and religious explanatory paradigms, Allison Busch and Christina Oesterheld in this volume examine lexical hybridity within the field of literature as a genre- and context-sensitive issue. They both find that a more tadbhava register was used for a female voice in opposition to highly Sanskritised or highly Persianised registers.91 Making a similar point, Shantanu Phukan argued on the basis of works such as the Bikaṭ kahānī and a marsiya by Sauda that Hindi was perceived by Mughal elite male authors as ‘especially effective in moving emotions’ and was embedded in Persian or Persianate Urdu to invoke a domestic female tone as opposed to the male and the non-domestic female world of the ghazals.92 The use of a vernacular for female voice can be observed in a narrative poem from the sixteenth century by ‘Ishqi Khan (d. 1582),93 which describes how the Turkish, Tajik and Indian wives of a wealthy jagirdar talk in Turkish, Persian and Hindi respectively. This is, for example, how he is received by his Hindustani wife on his return home, Zan-i hindī zi yak t̤ araf gūyad; haũ tirī lauṇḍī tū mirā khvandgār; Tum jo mujh kõ piyār karte ho; haũ bhī kartī hū̃ tihārā pyār. Apne kothe pai maĩ bichāūṃ palang; ūs ūpar leṭ jīo pāõ pasār; Bīc tūṃ leṭ loṇḍiyā̃ cau-gird; ḥaramān ās pās tum backār.94 On one side the Indian wife said, I am your woman and you are my kind lord; The way you love me, I love you in the same way. Let me make up a bed in my room, come and lie down on it stretching your legs. Lie down surrounded by girls; there are women round here—be careful. And this is how an Indian wife receives a poor husband: Zan-e hindī zi yak t̤ araf gūyad; terī mā̃ golī terā bāp camār; Jhūṭh tujh thẽ bahut sunā mat bol; sac tirā haũ kahaũ mirā mat mār.

Tujh thẽ mujh ko na rotī o pānī; tujh thẽ mujh kõ nahīṃ savār95 o sĩgār; Ab na rāhūṃ tire khudā kī saũ; nikalūngī tihāre ghar thẽ bahār.96 On one side the Indian wife said, your mother was a cowherd, your father a camār; Don’t say a word, I have heard enough lies from you, if I tell you the truth, don’t beat me. I get neither bread nor water from you, nor carriage, delicious food nor ornament. I won’t stay with you, I swear, I will leave your house. Phukan briefly considered the possibility that the motive behind mixedlanguage could be irony, as was the case for Latin-Italian macaronic parodies of the Renaissance.97 Although he quickly dismissed the idea in favour of a more homely or pathetic effect, the humorous effect of the ‘Persian’ couplet attributed to Khusrau, Rahim’s macaronic poem or ‘Ishqi Khan’s mimetic lines seems inequivocable. Sometimes we have an indication from the users of the languages themselves on how they perceived the ‘ecology of Hindavi’ in their literary world. I have mentioned above Nizamuddin Auliya quoting—and apparently seconding—Faqih Madhaw’s opinion that Shaikh Ahmad Nahravani was wasting his time on Hindavi songs. We can see a more straightforward condemnation of Hindavi in the discourses of Shaikh Sharafuddin Maneri (d. 1381), who once forbade the singing of a Hindavi chakri saying that Chakri is found on the lips of women. It is a very free sort of thing. There were also some young men in the assembly. Can you tell me where one and all acquire the power to bear such things? Confusion would result, for ‘melodious songs are as enchanting as adultery’. For that reason it was forbidden. If, however, it takes place in privacy, and all present are ascetics, men of struggle with self and having much knowledge, as well as being capable of making lawful exceptions, then they can do so.98

At another musical gathering where after some Persian songs the minstrels had switched to Hindavi, Maneri said: Hindavi compositions are very fortright and frank in expression. In purely Persian verses, there is a judicious blend of allusion and what can be fittingly expressed, whereas Hindavi employs very frank expressions. There is no limit to what it explicitly reveals. It is very disturbing. It is extremely difficult for young men to bear such things. Without any delay they would be upset. This is why there are difficulties involved in allowing young men to listen to such things. The members of this group, however, experience only grief and pain.99 It is not difficult to imagine the growing fashion of Hindavi singing hinted at by these lines, and the perplexity of the older generation. We can also assume that since songs in Persian (and Arabic) and other languages alternated at Muslim musical gatherings, multilingual compositions must not have been out of place. In fact, Maneri’s condemnation seems to have been of no avail since Hindavi words and verses begin to appear in the Persian writings of his followers, especially in those of Muzaffar Shams Balkhi (d. 1400/01).100 Other authors similarly conversant with Persian and Hindavi such as Gesudaraz and, four hundred years later, Anandram ‘Mukhlis’ underlined the emotional capacities of Hindavi.101 In the earliest phase of Hindavi literature, Gesudaraz is credited to have emphasised the tenderness, clarity and musicality in this language102, while Anandram ‘Mukhlis’ spoke about the Hindavi romance Padmāvat as having ‘an eastern melody brimming over with pain’,103 evidence that Hindavi retained a similar emotional appeal in the eighteenth century. Sumit Guha suggests that embedding eastern Hindi dialects in Persian or Persianate Urdu texts was a choice that aristocratic men of letters made to evoke intimate domains of affection and loss especially connected to childhood, when they were surrounded by the rustic speech of the unlettered wet-nurses and attendants in the women’s quarters.104

Another possibility is that writing poetry in a mixed language meant imitating spoken usage. In pre-modern India, as today, informal speech very often mixed phrases and words of an Indian cosmopolitan language with those of a vernacular, as the example of the malfuzat showed. Mixing, however, also had its rules. Normally a vernacular was mixed with a cosmopolitan language, and it is rare that elements of two vernaculars were mixed consciously. The list of external and internal forces mentioned so far is far from exhaustive and further possible motives could have been at work behind linguistic choice and hybridity. For example, in the case of Sant poets such as Dadu Dayal, literary polyglossia was a powerful means of reaching out and impressing the audience. Noblemen of Turkish descent such as Bairam Khan, Rahim and ‘Ishqi may have felt encouraged by the innovative and hybridising spirit of the Mughal court to experiment with the several languages at their disposal, including those of the Mughals’ ancestors and of the people of the country. It was, however, not only Turki noblemen who tried Rekhta. The unprecedented nature of cross-cultural interaction at the Mughal court between intellectuals whose work belonged to Sanskrit and Persian traditions has already been noted by Sheldon Pollock.105 In Persian poetry a call for the new, and a dislike of imitation appeared as the preference for the tāza- gū’ī (freshness in composition)106, while in the imperial painting studios this spirit manifested itself as the ‘delight in originality’ of artists like Daswant107 and produced the unique Mughal style of miniatures uniting elements of Irani, Dakkani, Rajasthani and European painting. Motivation behind the use of mixed language composition cannot be explained with one factor or another but should rather be perceived as the working of multifarious rationale with different intensity at different places and at different stages. REKHTA IN THE NAGARI SCRIPT: EARLY STRAY POEMS (MUKTAKAS)

We have seen so far that Rekhta poetry in the Persion script was cultivated in the Deccan, in Sufi circles in north India and was patronised by the Mughal court already in the sixteenth century. In this section we will see that the same genre was taken up by Hindu religious poets in the second half of the sixteenth century. Since we have scarce material at our disposal, it is difficult to tell exactly under what circumstances Rekhta in the Nagari script emerge. The oldest corpus of Rekhta poems in the Nagari script can be found among Nirgun Sants, whose teachings often contested and blurred the Hindu-Muslim division and whose mixed language could sometimes be very close to Khari Boli. Indeed it is the linguistically most adventurous Sant poet, Dadu Dayal (1544–1603), who not only used elements from different languages or dialects with confidence but composed poems in Rajasthani, Gujarati, Braj, Punjabi, Persian and Sindhi. His use of Khari Boli may have been prompted by the similar practice of the Sufis and the increasing popularity of Rekhta in the Mughal court. His literature is attested in early manuscript material and his Khari Boli muktakas can be considered to be the earliest extant examples of Nagari Rekhta. The following song, rather Sufistic in content, is already present in a manuscript from 1636 and no substantial variant readings exist to it: 108 Alā terā jikar phikar karte haĩ; Ās̃ aka mustāka tere; tarasi tarasi marate haĩ. Salaka ṣesa digarā nesa; baiṭhai dina bharate haĩ. Dā̃ima darabārī tere; gaira mahala ḍarate haĩ. Tana sahīda mana sahīda; rāti divasa larate haĩ. Gyā̃ na terā dhyāna terā; isaka āgi jarate haĩ. Jā̃ na terā jyāda terā; pau sira dharate haĩ. Dādū dīvā̃na terā; jara ṣarīda ghara ke haĩ. (Dadu Pad 398)109 O God, I remember and reflect upon you. I am your passionate lover dying of intense longing. I have no other place [?] in the world; I spend my days sitting here. I am your permanent courtier—frightened outside your palace. My body is martyred, my soul is martyred; I fight day and night.

My knowledge is yours, my meditation is yours; I burn in the fire of love. My soul is yours, my life is yours; I bow my head to your feet. Dadu is your steward; I am of your house bought with your money. Dadu also experimented with the consciously mixed language of the Mughals by interspersing his Hindavi with long Persian phrases as in Pad 81.110 By the early seventeenth century the Sants developed their own vehicle of expression in a language that mixes various vernacular languages and dialects and what is by modern Hindi scholars called Sadhukkari (sādhukkarī bhāṣā). This must be the reason why, though many of Dadu’s disciples had a rich literary output in sadhukkari, most of them did not continue their guru’s experiments with Hindavi-Persian hybridity or with Khari Boli, with the exceptions of Sundardas and Vajid. Here is an example by Sundardas where the first line of the poem is almost entirely in Khari Boli (with one Persian word) and the second is in Braj Bhasha: Maĩ hī ati gāphila huī rahī seja para soi. Sundara piya jāgai sadā kyaũkari melā hoi.111 4 (Bandagī kau amga) 27 I was too negligent and remained asleep on the bed. My beautiful beloved is always awake.—In what way can we meet? Vajid112 (fl. 1600), today a relatively unknown author113, was, according to the Bhaktamāl of Raghavdas114, a Pathan Muslim. When he killed a pregnant gazelle, compassion arose in his heart. He broke his bow and arrows and without returning home set off in search of a guru, which he later found in the person of Dadu. From the mention of Khadgasen as his office-keeper and Khadgasen’s reference to him as ṭhākur we can surmise that he was a landlord. Raghavdas counts Vajid among the hundred disciples of Dadu Dayal and several of his padas and sākhīs are also collected in the Dadupanthi Sarvāṃgīs of Rajjab (1620?) and Gopaldas (1627) and in Jagannath’s Guṇ Gañjanāmo. His most celebrated works are his stray arillas that have been published four times in four different

books,115 and which inspired the modern guru Osho to deliver discourses on Vajid.116 Over hundred different works of his are mentioned in manuscript catalogues.117 Most of them fall into the Nirgun Sant tradition and many are about morals (nīti). He was, however, a prolific author whose literary output include entertaining religious works such as Andhā kūbṛā sagun ‘Omens of the blind and the hunchback’, Gun rājkṛt (or ‘The acts of the king’: The story of the previous birth of a king, a carpenter a merchant and a leper)’ or Guṇ mūrikh-nāmo ‘The book of the stupid’. His published works do not show any significant use of Khari Boli or Perso-Arabic vocabulary and the editor of his arillas in the Pañcāmṛt is astonished by the fact that ‘he used a very pure form of Hindi’.118 Nevertheless some titles suggest a greater influence of Persianate culture such as the Guṇ Sekh Saṃvād or the Guṇ Sūphī- nāmau while his works on separation such as Guṇ Virah-nāmau ‘The book of separation’, Virahvilās or Virah sumiraṇ hit upākhyān suggest a Krishna-ite context. Vajid’s Rekhta includes some technical musical terms such as Malhār and Ahīrī (names of ragas) or mandra (lower pitch), evidence that the author was at home in the world of music, and that it would be fruitful to examine further the earliest links between the use of Rekhta and music. Vajid is the author of a work called by its scribes Rekhta or śrī rekhtā thākur kā,119 ‘The Rekhta of the Lord’. This is a collection of 14 kavittas in Rekhta, of which I have found four manuscript copied in 1651, 1655, 1752 and one sometime after 1667. The grammar of this work is Kharī Bolī with very strong traces of Braj. One of the most salient features of the text is the high number of Perso-Arabic words: 104 different words of Perso-Arabic origin are used 155 times, which means a proportion of 19 per cent. In a few poems the Sufi idea of linking worldly love to transcendental love is dominant while the majority of the quatrains are concerned with the Vaishnava (and courtly) theme of the cowherdwomen’s separation from Krishna. The beloved is sometimes God described as a woman (kavittas 1, 4, 10) but more often it is Krishna. Sometimes the lovers speak directly to one another and sometimes we hear the words of a messenger as favoured

by the Indian tradition. A popular context is the cowherd women’s complaint to Krishna’s messenger, Uddhava, Gopī gāī gvālani tau behāla haĩ bihārī bina Hotā na māluṃma makasūda kyā tumhārā haĩ; Ināyāta rahai makāṃma kīne hai kamala naina Maiṃnamathi mārai mādhau cārā kyā hamārā hai; Jau tau takasīra kachu bhaī hai hamārī Hari Kījīye jū māpha tuma jīte hama hāryā hai; Tuma tau sakhā hau sāṃkhī sāṃcī kini kahau bali Udhau brajanātha braja kāhe taĩ bisārā hai. (7) Without Krishna the cowherd-women, the cows and the cowherds are despondent. ‘Nobody knows what you are up to. The lotus eyed one favoured us, taking up residence in us—We have been smitten by the ‘Soul-Churner’ Love, O Krishna, what is our way out? If ever we offended, Hari, please forgive us, you have won, we are defeated. You are his friend—whom else can we call a true witness?—Uddhava, why has the Lord of Braj abandoned us?’120 In the following poem the masculine verbal form in the expression khalak hī yāra huvā ‘the world has been made her lover’ suggests a male lover, Dila kī dilāsā sārī dunī kā tamāsā kuli Gama kā karāra dunī dekhai jīsa ūba hai; Vākī muhabati dekhaī̃ khātari maĩ yau kyau āvai Ālama kī sāhibi tau aisī jaisī dūbā hai; Rāga khūba raṃga khūba āṃkhaĩ khūba bhauṃhai khūba Hausa khūba hāmsī khūba, sabhā kaisī khūba hai; Umara kī khūbī para khalaka hī yāra huvā Kauṃna kauṃna khūbī kahaũ khūba mahabūba hai (1)

All consolation of the heart, the whole spectacle of the world and the permanence of sorrow—witnessing this one loses spirit. Seeing her love why should I show regard for this? The dominance of the world is only like a blade of grass. Her passion is splendid, her colour is splendid, her eyes are splendid, her eyebrows are splendid, her desire is splendid, her smile is splendid and how splendid is her purity. With the splendour of her prime youth all creation has been made her lover. Why pronounce on her splendour? The beloved is splendid. Other Sants outside Dadu’s lineage continued this tradition. For example, Malukdas (1574–1682?) from Kara (Allahabad) frequently used a Khari Boli template with Perso-Arabic words in his padas and especially in his kavittas, Bhīla kada karī thī bhalāī jiyā āpa jāna Phīla kada huā thā murīda kahu kisakā. Gīdha kada jñāna kī kitāba kā kinārā chuā Byādha aura badhika nisāpha kahu tisakā. Nāga kada mālā lai ke bandagī karī thī baiṭha Mujhako bhī lagā thā ajāmila kā hisakā. Ete badarāhõ kī badī karī thī māpha jana Malūka ajātī para etī karī risa kā.121 Has a tribal ever done any good intentionally? Has an elephant ever become a disciple of anyone? Has a vulture ever touched the edge of a book of knowledge? Has a fowler or a hunter done any justice to it? Has a snake ever welcomed anyone sitting with a garland? — I also had a rivalry with Ajamil. You have pardoned the sins of many wicked people, Why are you so angry with your worshipper, the casteless Maluk? Ten words out of 60 in this poem are of Perso-Arabic origin. Malukdas used an even more Persianised language (24 out of 83 words) abounding in Islamic technical terms, where a stronger Sufistic message was intended,

Terā maĩ dīdār divānā. Ghaṛī ghaṛī tujhe dekhā cāhū, suna sāheba rahamānā. Huā alamasta khabara nahĩ tana kī pīγā prema piyālā, Thāṛha hoū to giri giri paratā tere rãga matavālā. Kharā rahā̃ darabāra tumhāre jyõ ghara kā bandājādā, Nekī kī kulāha sira dīye gale pairahana sājā. Taujī aura nimāja na jānū̃ nā jānū̃ dhari rojā, Bāṃga jikira tabahī se bisarī, jaba se yaha dila khojā. Kahaĩ malūka aba kajā na karihaũ dila hī sõ dila lāyā, Makka hajja hiye me dekhā pūrā murasīda pāyā.122 I am crazy about seeing you. I want to see you every moment, hear me, o gracious lord! I became intoxicated, I do not know my body; I drank the cup of love. If I stand up, then I fall again and again—drunk with the colour of your love. Let me stand in your royal assembly as a slave born in your house. I donned the hat of virtue and wear its cloak on my shoulders. I know no arguments, no prayer, I do not know how to fast. Since I searched my heart I have abandoned the muezzin’s call and the remembrance of God. Malūk says, now I won’t make up for my missed prayers, I willingly fell in love. I have seen Mekka and the Pilgrimage in my heart, and received my perfect spiritual guide. Apart from the Perso-Arabic and in the Devanagari scripts Khari Boli was also written in Gurmukhi in early Sikh literature. Janamsakhis, such as Miharvan Sodhi’s Janama sākhī śrĩ guru Nānakdevjī or Hariji Sodhi’s Gosati guru Miharivānu,123 written in Sadhukkari, mixed Braj features with Khari Boli and Punjabi. Some works, such as the Ādi Rāmāin by Miharvanu Sodhi ‘Manohardas’ (1580–1640), the grandson of Guru Ramdas, are occasionally dominated by Khari Boli. In contrast with the Nirgun Sant works mentioned above, in the Sikh compositions Khari Boli is very

strongly mixed with other dialects and the high input of Perso-Arabic vocabulary is missing. Tab brahmādik kī bārī āī. Tab brahmādi ehi kahiā jī he srī dev jī mujh kaü ehī lamkā dehi. Tab srī Mahādev kahiā ji mai dīnī. Tab itne kahne sāth pārbatī karodhu dīā. Ji he adharmī tujh daü ehu bāt kiu kari kahinī āī hai? Mai aje iskaü dekhi bhī nahī nibadī. Paru jāhi je sati paramaisur hai. Taba je kou is lamkā ke bīe bādega so tatkāl hī bināsu hoi jāegā. Tab lamkā kaü pārbatī kā sarāpu bhaïā.124 Then it was the turn of Brahma’s son, Brahmadik. Brahmadik said ‘O great god, give me this very Lanka.’ Shiva said: ‘(It’s yours.) I have given it to you.’ When this much was said Parvati became angry: ‘O you unlawful one, how could you say this? I haven’t even seen it fully yet. But I swear by the highest God, if anyone enters this Lanka, he will die immediately. So Lanka was cursed by Parvati. Although the early practice of Nagari Rekhta can be found chiefly among Nirgun Sants, an early Krishna-poet also experimented with it. Here is an example of the use of Khari Boli with Perso-Arabic vocabulary which comes from the heartland of Braj, from Vrindaban. It is a song by Svami Haridas, the founder of a school of Krishna devotion who is also celebrated as the initiator of the dhrupad style of singing.125 In this case we lack manuscript evidence prior to the mid-eighteenth century, and there is a shade of uncertainty regarding its authorship and date, yet it is difficult to imagine that this odd poem would make its way to the limited sectarian Haridas corpus from outside and would withstand the ‘Brajifying’ tendencies of the scribes, Bande akhatiyāra bhalā; Cita na ḍulāva āva samādhi bhītar na hohu agalā; Na phiri dara dara pidara dara na hohu adhalā; Kahi haridāsa karatā kiyā su huvā sumera acala calā. (6 Asṣṭdash Siddhānta)126 O, worshipper, this choice is good. Waver not in mind, enter into profound meditation, be not an adversary;

Do not wander from door to door [or seek] your father’s door, do not be blind. Haridās says: what the creator causes, comes to pass—even immovable Meru moved. (Translated by L. Rosenstein)127 Although the poem is far from being pure Khari Bolī and has Braj Bhasha forms (like the repeated imperative hohu,‘be’) it is rather isolated in the Haridas corpus. It is possible that a poem with Khari Boli features and some Perso-Arabic vocabulary was smuggled into the Haridas corpus in order to justify the Rekhta attempts of eighteenth-century Haridasi poets such as Sahacharisharan and Sitaldas. If this quatrain is genuine then the reason behind the use of Khari Boli forms and Perso-Arabic vocabulary may be an early attempt to evoke the atmosphere of music patronised enthusiastically in Islamic courts. It is also interesting to observe that there is nothing specific about Krishna bhakti here and the poem is rather similar to one by Nirgun Sants. THE SEVENTEENTH CENTURY—REKHTA IN THE PERSIAN SCRIPT The seventeenth century saw an increased production of Rekhta in the Persian, the Gurmukhi and the Nagari scripts. According to Faruqi the earliest literary text in Hindi/Hindavi extant in north India is the Bikaṭ Kahānī (Dire Tale, 1625), a poem of 325 shi‘rs in the masnavi form describing a woman’s pangs of separation during the twelve months of the year, whose author, Muhammad Afzal, is also known by the half-Hindu half-Muslim name of Afzal Gopal.128 The problem with this poem is similar to that of the poems of Kabir and others. It was transmitted orally for a long period, and its text is preserved only in eighteenth-century or later manuscripts.129 It is possible that the language of the published text does not represent seventeenth-century features. This poem, Faruqi notes, was neglected by early Urdu tazkira writers and has only been reclaimed as part of Urdu literary history in the twentieth

century (Sherani, Zaidi, Faruqi, Jafar and Jain). What could be the reason behind this neglect? Faruqi argues that macaronic poetry has been rejected by eighteenth-century Urdu poets. It can be claimed that probably before the nineteenth century ‘Hindi’ and ‘Urdu’ represented a literary division within Hindavi manifest in metrical forms and genres (rather than in language or script). Padas, dohas or kabittas were not accepted as part of the high Urdu tradition no matter how Persianate their vocabulary. In much the same way no ghazal or rubai could be produced within the Hindi tradition even if it lacked Persianate vocabulary. However Bikaṭ Kahānī belongs at least as much to the popular Indian barahmasas, ‘twelve months’ poems rejected by Urdu high tradition, as to the Persian masnavi genre.130 Out of its 325 couplets 41 are directly in Persian, 20 have one line in Hindavi and one in Persian and another 20 lines are half in Persian and half in Hindavi.131 Since its Hindavi template is not pure Khari Boli but also shows Braj Bhasha features, Bikaṭ Kahānī is very close to the Nagari Rekhta poems. In the following lines, for example, jare and bādara are Braj Bhasha forms, Sakhī bhādõ nipaṭ taptī paṛe rī; tamām-e tan-badan merā jare rī. Siyāh bādar cahārõ or chāye; liyā mujh gher piu ajahū̃ na āye.132 My friend, the rainy season burns me severely; my entire body is aflame. Dark clouds have spread everywhere and surrounded me—my beloved has not come yet. Bikaṭ kahānī is thus truly in an ‘intermediary’ position—linked to Indian tradition through its genre and to Persian through its metre, which all Urdu Barahmasa writers adopted after Afzal.133 With works such as Vajid’s rekhta, the Bikaṭ Kahānī and the Prem Prakāsh by Shah Barkatuddin Marharvi/Bilgrami (1660–1729), a Sufi and Persian poet, who used the penname ‘Ishqī in his Persian compositions and Pemī in his Hindavi, we see the emergence of a hybrid linguistic and literary koine that combines and chooses between Persian language, poetic imagery and metres and Hindi (Khari Boli or Braj Bhasha) phrases, metres and poetic topoi and genres. Prem Prakāsh uses not Perso-Arabic but Braj Bhasha metres and shows an even more sustained engagement with Hindi poetic forms: it contains

202 dohas, 50 rekhtas in kabitt (quatrain) and pad (song) metres, 20 rekhtas in question-answer-form, 113 kabitts and pads, 1000 lines of rekhta besides irshad (guidance), ariza (humble petition134), barahmasa and sadrtu varnan.135 Pemi in his Rekhta mixes Persian phrases with Khari Boli inflected by Braj forms, Camke tere paṭ oṭa mẽ mukh rūp ujiyārā—jīvan sẽ136 badlī mẽ Bagzār ki dar rū-i tū bīnīm khudā-rā—ab sūnī galī mẽ.137 Under your veil your bright face shines like water in a cloud. Allow me to see God(‘s manifestation) on you(r face)—in an empty lane.138 Other seventeenth-century authors who composed poems with one line in idiomatic Hindi and one in Persian include Mullah Nuri Azampuri.139 Though this kind of macaronic poetry, Mir’s third category, continued in the eighteenth century,140 it was Persianised Hindavi, Mir’s fourth category, which became mainstream Rekhta. Several seventeenth-century Persian poets continued to use Khari Boli and the mixed language as in the previous century. The best documented example is that of Baba Fatah Muhammad (d. 1669), who was the son of Shaikh ‘Isa Jand Ullah, a friend of Abdurrahim Khankhanan.141 The following poem is in an Indian moraic metre (16+13 morae) and its language is Braj with some Khari Boli features. Isa kula bhĩtara mīta na koū, āpa svāratha saba dekhe; Tumha sāga janama akāratha bītā, jo bītā to kita lekhe…. Jhūṭha daghābāzī batamārī, ghāta bisāsī madhu-pītī; Aba kāhe pachatāvana lāgā, taba kyõ soca na mana kītī. Fataḥa muḥammada kyā samajhāve, nakha-sikha tū yõ ālūdā; Apanā āpa savāra divāne, aurana sõ kyā maqasūdā.142 No one is your friend in this family—everyone is after his own selfinterest. Your life has passed in vain. You lived, but for what reason?…

Falsity, deception, robbery, murder, treachery, debauchery—Why did you start to repent it only now? Why did not you think of it then? How much should Fatah Muhammad explain, you are defiled from head to toe. See to it yourself, madman, why do you expect it from others? The importance of this poem lies in the fact that it was included into a Persian and Rekhta album and that, despite its linguistic features, was considered to be part of the Persian/Rekhta tradition. The poem was called Rekhta by Sherani although it rather shows Braj Bhasha features with a minimum input of Khari Boli. The distance between the Rekhta of Bikaṭ kahāni and that of Fatah Muhammad shows two alternatives of mixed language. Unlike Afzal and Shah Barkatuddin, Fatah Muhammad used Indian metre and a Braj Bhasha template rather than Khari Boli. A poem attributed to Chandrabhan Brahman (1574–1662),143 who was Dara Shikoh’s mir munshi and later vazir, shows the blend of Hindavi and Persian that became common in the eighteenth century with Vali Aurangabadi. Although Jalibi claims to have found it in an old album, the first dated occurrence of the poem is in a tazkira called Khumkhāna-yi Jāved (1908), and some scholars question its authorship.144 Its imagery and language, however, suggest an early date of composition. Post-seventeenth century Rekhta is less likely to use archaic forms such as haman ko (us), lāe dālā (thrown) or the nasalised postpositions sẽ and kõ, Khudā ne kis shahar andar haman ko lāe ḍālā hai; Na dilbar hai, na sāqī hai na shīsa hai na pyālā hai. Piyā ke nā̃ va kī sumran kiyā cāhū̃ karu kissẽ; Na tasbī hai na sumran hai na kaṇṭhī hai na mālā hai. Khẉubān kī bāg mẽ raunaq hoe to kis taraḥ yārān; Na daunā hai na marvā hai na sosan hai na lālā hai. Piyā ke nāõ ‘āshiq kõ qatl karnā145 ‘ajab dīkhe; Na barchī hai na karchā146 hai na khanjar hai na bhālā hai. Barahman vāst̤ e ashnān ke phirtā hai bagiyyā sẽ; Na gangā hai na jamnā hai na naddī hai na nālā hai.147

Into what city has God thrown us? —There is neither beloved, nor cup-bearer; neither flask, nor cup. If I want to repeat the name of my beloved, by what means should I do it? — There is neither rosary, nor rudraksha, neither necklace, nor garland. If beauties were to illumine the garden with their presence, then how would they do it, o friends? — There is neither rosemary nor marjoram; neither iris nor tulip. To attribute the killing of the lover to the beloved—how strange it seems! — There is neither lance nor sword, neither dagger nor spear. The Brahman wanders in the garden in order to taking a bath. — But there is neither Ganges nor Jamuna; neither river, nor brook. Since neither Vali Ram nor Brahman composed a Hindavi dīvān, their Rekhta poetry has been neglected in the same way as that of many others.148 Modern Urdu scholars have been eager to rediscover early examples of Urdu Rekhta in north India, though they have not viewed it in the context of a continued engagement with Hindavi but as prehistory of Urdu poetry. There is also evidence of a widening of the domain of Rekhta in seventeenth-century north India. Some longer works produced in loose Khari Boli include Shaikh Maulana ‘Abdullah Ansari’s treatise on the religious jurisprudence of India entitled Fiqh-e hindī (1663),149 Shaikh Mahbub ‘Alam’s three long poems, the Mahshar-nāma (The Book of the Day of Judgement), the Masā‘il-e Hindī (The Precepts of the Prophet in Hindi) and the voluminous Dard-nāma on the life of Prophet Muhammad.150 The language of these works is influenced by Braj and Punjabi and comes thus very close to the mixed language of the Sants, as the following dohas from the Mahshar-nāma show, Rabbā merā eka tū̃ nāhī̃ koī dūjā; tujha sā sā‘ī̃ chāṛa kara kisa lāū̃ pūjā…. Sārī qudrat tū rakkhā cāhā so kīnī; ekõ kāyā chīna lī ekõ māyādīnī.

Eka rakkhe nita rovate rovẽ bahu bhātā; eka rakkhe nita sovate sovẽ dina rātā̃ .151 Only you are my lord, no one else. Abandoning a lord like you whom else should I worship?… You protect the whole creation and do whatever you want; You took away the body of one or put another in illusion; You keep one continuously weeping—weeping in many ways; You keep another continuously sleeping—sleeping day and night. In all probability Ansari’s and Mahbub ‘Alam’s works attempted to communicate the most important teachings of Islam to circles that were less familiar with Persian. The language sometimes relies more on Khari Boli, as in the opening lines of the Masā’il-e Hindī, Qayāmata ke aḥvāla mā̃ hindī kahī kitāba; maḥshara-nāma nāva hai jāno aiy aṣhāba. Maḥshara-nāma bīca hai suna va‘da aisa diyā; ina ‘ājiz darvesha ne būjho khola hiyā. Amra-nahī kī bāta mõ hindī bolī bola; shar‘ tarāzū dīna kī judī jo dū̃ gā tola. Bāra cauda barasa laga va’da lāgī ḍhīla; muḥammad jīvana yāra ne kahā āy beqīla. Va’da kõ ākhir karo amra-nahi kī bāta; likha diyo hindī bola kara bā̃ cū̃ maģ dina rāta. Talaba bahuta isa yāra kī dekhī sā̃ cī sūjha; likhī kitāba us vāst̤ e hindī bolī būjha. Masā‘il-i hindĩ nā̃ va aba isakā kaha de yāra; paṛho fātiha mujha upara je bakhshe karanāra.152 I have composed a book in Hindi on the final judgement; you know, it is called Maḥshar-nāma, o companions. In the middle of Maḥshar-nāma I, this lowly poor man, you know, openly made a vain promise that ‘I will tell the commands and prohibitions in Hindi. The Law is the scale of Religion even though I give different measurements.’ For some thirteen years I was lax with the promise, when my friend, Muhammad Jivan said: ‘O you speechless one, fulfil your promise of telling the commands

and the prohibitions, write them in Hindi and I will read them day and night.’ I saw that his eager solicitation was earnest; know it therefore, that I wrote this book in Hindi for his sake. Call it now by the name of Masā’il-e Hindī, my friend, and read the prayer for the soul above me so that God may forgive me. The number of the Rekhta works quoted so far testifies that by the seventeenth century a specific northern Rekhta koine had developed which was distinct from that used in the south. Northern Rekhta poets had a predilection for macaronic poetry and their Hindavi, though based on a Khari Boli template, showed strong Braj Bhasha influence and was free from Dakkani features. That this was not a fixed language in the seventeenth century is shown for example by Mahbub ‘Alam’s use of different registers. Indeed, writing on Muslim themes in the vernacular in the north was not restricted to Khari Boli or the mixed language. The tradition of writing popular works in Avadhi was kept alive by works such as Shaikh Faizullah’s translation of the Persian, Qiṣṣā-ye Jamjam.153 Moreover, Rekhta texts belonging to the Urdu-Persian masnavi tradition also began to be written in north India towards the final years of the century, after the genre had long become popular in Gujarat and the Deccan. Raushan ‘Ali wrote his Jang nãma (War chronicle), also called ’āshūr nāma (Tenth-day chronicle) in 1688/9, and in 1693/4 Isma’il Amrohvi wrote his Tavalludnāma-ye bībī fātima154 (Birth Chronicle of the Lady Fatima), a biography of Fatima. His other masnavi, Qiṣṣa-ye mōjiza-ye anār (The story of the miracle with the pomegranate) was written in 1709. Significantly, the language of these two masnavis shows strong Dakkani features, though their author proudly announces at the end of each that they were written near Delhi: Vat̤ an amroha merā hai shahar nām; isī jā par merā hai jā-(ye) qiyām.155 My homeland is the town called Amroha; in that place is my permanent home.

There is some indication that Rekhta activities in the North were strengthened also through an early interaction with Gujri.156 Faruqi observed that Raushan ‘Ali’s masnavi is modelled on Miskin’s (fl. 1681) Gujri Jang nāma-ye muḥammad ḥanīf (Battle chronicle of Muhammad Hanif) written in all probability in 1681.157 But while Miskin describes his language as Gujri,158 Raushan ‘Ali refers to his as 159 Hindī/Hindustānī/Hindavī on different occasions. Another example of traffic between Gujarat and north India and of linguistic and literary syncretism—in this case connected to religious syncretism—is Prannath (1618–94),160 a Gujarati kshatriya and a religious reformer with a wide outlook who even knew about Christianity. After travelling in Arabia in his youth, he eventually settled in Panna in Bundelkhand in 1683, where he became the spiritual mentor of King Chhatrasal (d. 1732). Prannath proclaimed himself both the Mahdi of Islam and Kalki of the Hindus and founded a sect that integrated Islamic and Hindu elements. His fourteen treatises in Hindi and Gujarati verse are collected under the name Quljum-sharīf (also called Kuljam-svarūp or Tārtam Vānī).161 In these treatises he made extensive use of Khari Boli and in some works, such as the Kiyāmat-nāmā (Book of the Day of Judgement), he shows a strong preference for Perso-Arabic vocabulary.162 His acquaintance with Gujri through his Gujarati background might account for this inclination. The Hindavi that Vali brought to Delhi did not ‘create’ Rekhta poetry in north India but rather displaced the pre-existing fashion for mixed language poetry. It can be argued that it was this existing fashion which contributed to the quick acceptance of the new style introduced by Vali. Rekhta using Khari Boli or Braj Bhasha templates did continue, but was now associated with what were considered lesser genres like marsiyas or barahmasas, as the neglect of tazkira writers suggests. NAGARI REKHTA IN THE SEVENTEENTH AND EIGHTEENTH CENTURIES

Rekhta poetry was also composed in the Nagari script in the seventeenth century although we know much less about it than about its versions in the Persian script. Nagari Rekhta is documented from the princely court of Jaipur as well as from the syncretistic movement of Prannath. One of the earliest courtly composers.163 Prannath Shrotriy in Jaipur who wrote his Bejārnāmā for Maharaja Ram Singh (r. 1667–89).164 Prannath’s disciple, Laldas composed his guru’s biography, the Bīṭak, in 1694 in Khari Boli,165 which at times is very close to modern usage166 and at times is influenced by Braj or Punjabi.167 The eighteenth century saw a particularly rich production of mixed poetry in the Nagari script, possibly influenced by the rising tide of Rekhta in the Persian script, and can be considered the heyday of Nagari Rekhta. Despite the overwhelming popularity of Braj Bhasha riti poetry, in the eighteenth century we find Nagari Rekhta all over the Hindi region, from Bihar to Rajasthan.168 We can identify three contexts for Nagari Rekhta activities: the circles of Nirgun Sants, a few Krishna Bhakti poets and courtly riti poets who mainly wrote in Braj Bhasha but also, significantly, experimented with Rekhta. While Nirgun Sant poets wrote in Khari Boli with a large admixture of Perso-Arabic, Panjabi and Avadhi words, riti and Krishna Bhakti poets evolved a new hybrid poetic form which ‘assimilated’ PersoArabic vocabulary, bending it to Braj Bhasha phonology and using a Khari Boli linguistic template. Titles like Ishk-caman, Ishk-latā, Ishkphandā etc. suggest that what is at work now is rather a poetic curiosity towards the new, and sensational, success of Urdu/Rekhta verse than Sufi influences, which had already been present for long. High Brajbhasha literature was more open towards literary influences than towards religious ones. Hindu poets like Nagaridas were cautious not to identify the beloved overtly with God.169 In his Ishk-caman, where he speaks about love with Islamic imagery and vocabulary, Nagaridas’s interpretation is different from that of the Sufis since he considers the lover, God and the beloved to be three different entities while in Sufism, Khudā, God, and maḥbūb, the beloved, are the same, Āsika pīra hamesa dila lagaĩ casma ke tīra, Kiyā khudā mahabūba kaũ sadā sakhata bepīra. (15)170

The lover’s heart is always tormented struck by the arrow of the glance; (But) God made the beloved to be continuously hard and unfeeling. When philosophical Sufi influences were accepted, this approach was rejected by strong circles within the Brajbhasha literary community as the case of Anandghan (or Ghan Anand, as he is also known) shows. The larger part of Anandghan’s quatrains can be read as relating to both his own secular and transcendental love, as was done by the scribe who tried to change the word for the beloved, sujān into expressions like ju syāma to make sure that it was not read as mundane. However, when the quatrain was too overtly mundane, then sujān was changed into su pyārī and the like to make sure that this ‘secular’ poem would not have any religious reference. It never happened in Hindi literature before Anandghan that the human beloved was identified with the Absolute as Anandghan’s double usage of the word sujān suggests. This twofold reading of the poems was peculiar to Persian and then to Urdu and was vehemently rejected by Anandghan's contemporaries.171 Let us now examine the various kinds of Nagari Rekhta separately. In the case of Nirgun Sants, quite often the choice for it is simply due to the fact that the spontaneously mixed language of the Sants tilted more towards Khari Boli. Dariya Sahab (c.1680–1723)172 of Bihar, who established a tradition indebted to the Kabir Panth, used a tinge of Khari Boli in his ātmakathā173 and also wrote some Rekhta in the mixed Sant language.174 Another poet of the same name from Mewar (1676–1758),175 a Muslim weaver, often composed poems with a strong Khari Boli input, Dariyā guru pūrā milā nāma dikhāyā nūra; Nisā sukha ūpajā kiyā nisānā dūra.176 Dariya found his definite guru—God’s name showed the light; I received the bliss of fulfilment and removed the sign. As we have seen in the case of court poets and Krishna-bhaktas, Rekhta poetry flourished due to literary rather than religious influences. Nirgun Sants and syncretists, however, were more open towards Sufi ideas, imagery and terminology. One example is Charandas (c.1703–82), who was

born in Kotwa in Rajasthan. He was brought up in Delhi and later he returned to his native place where he lived as a yogi and became the leader of a sect that combines the worship of the Bhāgavata’s Krishna with Sufism creating poetry that is close to that of the Nirgun Sants.177 His engagement with Sufism is reflected in some technical vocabulary (khudī, murśid) and the use of composite expressions (suhbat sādhõ kī), Do dina kā jaga mẽ jīnā hai, kartā hai kyõ gumān; ai besahūr gīdī ṭuk rām ko pichān. Dāvā khudī kā dūr kar apane tu dil setī; caltā hai akaṛ akaṛke jvānī kā jos ān. Mursid kā jñān samajh ke husiyār ho sitāb; gaflat ko choṛ suhbat sādhõ kī khūb jān. Daulat kā zauq aise jyõ āb kā dubāb; jātā rahegā china mẽ pachatāyagā nidān. Din rāt khovtā hai duniyā ke kārbār; ik pal bhi yād sā̃ i kī kartā nahī̃ ajān. Sukdev guru jñān carandās ko kahaī̃; bhaj rām nām sā̃ cā pad mukti kā nidhān.178 We live only for two days in this world. Why are you so self-important? O you drowsy dim-witted one, recognise God just a little! Remove the claim of egotism from your heart; the moment of vigour of youth goes away crookedly. Understand the guru’s knowledge and become aware at once! Give up negligence and recognise the speech of the true ones. Desire for wealth is like a bubble of water, it will go away in a moment and you will repent at the end. You waste your days and nights in wordly chores, and do not remember your lord even for a moment, you ignorant. Sukhdev gives this teaching to Charandas, worship God’s name, a true object, the treasury of liberation. Several other Nirgun Sants are listed in the catalogues as authors of Rekhta compositions,179 in particular a very high number of Niranjanis figure amongst them,180 as well as several followers of Bavri Sahab in Delhi and Avadh.181 Their Rekhta poetry shows that for them it was not just

poetry in mixed language but also in mixed metre. Some poems in Braj Bhasha without any considerable Perso-Arabic input, though sometimes with a tinge of Khari Boli are classified as Rekhta in manuscripts. More research is needed to investigate the basis of this attribution. Poems by Bulla Sahab (1693–1768) mix Braj Bhasha forms with Khari Boli ones so much that even the past tense of the verb ‘to be’ can appear in Khari Boli (huā) and in Braj Bhasha (bhaī) within the same poem.182 The Aliphnāmā, a poem by Bhikha Sahab (or Bhikhanand, d. 1791), a disciple of Gulal Sahab (who had been a disciple of Bulla Sahab) and followed him as the leader to the Bavri community,183 is an interesting composition in which each line starts with a consecutive letter of the Arabic alphabet (modelled on his Kakaharā, a composition on the Sanskrit alphabet). Though Bhikha is not bothered by the subtleties of the Arabic alphabet and does not distinguish between characters like jīm, zāl, ze, z̤ āl and z̤ o’e, the Aliphnāmā abounds in Perso-Arabic vocabulary. As in the case of Bulla Sahab, the use of Khari Boli varies in Bhikha’s poetry as well. The following poem in the jhūlnā metre (similarly to many other quatrains) is classified as Rekhta but is actually in Braj Bhasha, Pāpa au punna nara jhulata hī̃ḍolanā, ū̃ca aru nīca saba deha dhārī. Pā̃ ca aru tīni paccīsa ke basa paro, rāma ko nāma sahajai bisārī. Mahā kavalesa dukha vāra aru pāra nahĩ, māri jama dūta dẽ trāsa bhārī. Mana tohĩ dhirakāra dhirakāra hai tohĩ, dhṛga binā hari bhajana jīvana bhikhārī.184 Man swings on the swing of sin and virtue and takes high and low bodies. Controlled by the five substances the three qualities and the twenty-five elements he easily forgot God’s name. He suffers much anguish and grief and does not find the other shore. When he dies Death’s messengers heavily torture him. Shame on you, o soul, shame on you, shame, you wasted your life without worshipping God.

It was not only the Nirgun Sants and related poets that wrote in Rekhta, but there were also Krishna-bhaktas who continued the tradition of Svami Haridas, and we even find some mixed language compositions, this time with Punjabi elements, current under the signature of Surdas.185 One of the earliest documented authors is the Gaudiya Vaishnava Manohardas (fl. 1700) whose following quatrain resembles those of Vajid, Khosavakta dekhyā maũ pharajaṃda bābā naṃda jī kā Phaiṃṭā dhotī jarda baṃsī charī hātha līyā hai. Sā̃ valā nāma kīnā jara javāhira makaṃbŭla Jauvana kī maujẽ phaila jevāīsa dīyā hai. Jīva garakāva khŭbhsŭrata dariyāva bīca Saboroja mahajŭja mastahāla kīyā hai. Aba sahī jindagānĩ mahabŭba dilajānī1 Su huī dīdāvanī dāsa manohara jīyā hai. (Gaurangagunāvalī)186 In a happy moment I caught sight of Nanda’s son wearing a small turban and a yellow dhoti and holding a flute and a stick in his hands. The ‘Black One’ makes the names of ‘gold’ and ‘jewel’ shine. The surges and delights of prime youth spread …187 My soul drowned in the middle of the beautiful river. He made my days and nights happily intoxicated. Now my life is true, my love of the heart, I have a glimpse of you, and your servant, Manohar, lives. The Krishna devotees normally wrote Rekhta with a Khari Boli template, though it could also be Punjabi as in the case of Anandghan’s Ishk-latā, Ānãda ke ghana tuma binā mujanū̃ nahĩ bhāvai. Nayana asāḍe lāganai tujahī nū̃ dhāvai. Huna kyā kījai lāḍile vekhana nahĩ pāvaĩ. Julama karaĩ ye bāvare mujanū̃ tarasāvai. Taĩde mukha para tila abe ati khūna karãdā. Alakaĩ taiṃḍī yaũ chuṭī dvai nāgina lasãdā….67 (Ishk-latā 35–36)188

Cloud of bliss, without you I get no pleasure. My searching eyes run to you. What shall I do now? I cannot glimpse my beloved. These naive ones oppress and torture me. Alas, the black spot on your face slays me. Your loose locks of hair shine as two snakes…. We can conclude our survey of Nagari Rekhta devotional verse with two Haridasi ascetics belonging to the community of Tatiya Sthan in Vrindaban writing Rekhta in the early nineteenth century. In his Saras mañjāvalĩ Sahacharisharan, the abbot of the Tatiya Sthan between 1821 and 1837, continued the tradition of the māñjh (or mānj) metre with slight admixture of Punjabi. Sitaldas, a disciple of Thakurdas (1799–1811) who was allegedly infatuated with a boy called Lal Bihari (the name Lal Bihari occurs in his poems as a name for Krishna) used Khari Boli and a wide range of cultural referents in his Gulzār-caman, Ānand-caman and Bihārcaman, Majanū̃ Pharahāda Mādhavānala ye the maharama isa bastī ke; Lalai Shīrā̃ mẽ līna hue ura kāmakandalā kistī ke; Yaha ishka candrikā chāya rahī aba taka bāyasa isa mastī ke; Jānī ḍhuḍhe hĩ milate hai gāhaka is husnaparastĩ ke. (Ānand-caman 20)189 Majnun, Farhad and Mādhavānal were the relatives of this settlement, Their heart was absorbed in Laila and Shirin and in the boat of Kāmkandalā. This moon of love is still shining over the tree of this passion. My friend, costumers of this worship of beauty can only be found with difficulty. Finally, we find Nagari Rekhta literary activity at princely courts in Rajasthan already in the seventeenth century in all probability prompted by similar experiments at the Mughal court and by the popularity of the Nirgun

Sants’ poetry, while Jaipur (Amber) regularly patronised Rekhta writing, as the manuscript collection of the City Palace attests. A few decades after Prannath Shrotriy (fl. 1680), Kavikalanidhi Shrikrishna Bhatt composed Ishk-mahtāb, which used a wide range of vocabulary including Avadhi, Braj Bhasha, Apabhramsha, Persian depending on the sentiments evoked in the poem.190 In fact, by the eighteenth century the leading Braj Bhasha poets of the time wrote Nagari Rekhta, not just as occasional poems but even as longer works such as Nagaridas’s Ishk-caman, Anandghan’s Ishk-latā, Brajnidhi’s Rekhtā saṃgrah and Rās kā rekhtā and Rasrashi Ramnarayan’s Ishk-phanda,191 Ishk-latā,192 Ishk-paccīsī and Māhai mālik mukām kī or Ishk-daryāv.193 The high number of manuscripts of his Ishk-daryāv, written in the ‘Rekhta’ and ‘Ghazal’ metres during the reign of Maharaja Pratap Singh (r. 1778–1803), testifies to the popularity of this work.194 One of the latest Rekhta authors in Rajasthan was (Gangadas) Bakhtavar (1823–96) who served at the Udaipur court.195 Brajnidhi’s interest in Rekhta is also shown by the fact that he asked his court poet Rasrashi to compose similar verses. One such work composed for Brajnidhi is the Rasik-(or Rasrāshi-) Paccīsī.196 A somewhat intermediary position between Rekhta and Braj was occupied by poems which had a Braj template with a high input of PersoArabic vocabulary, such as in this passage from the Ishk-nāmā of Bodha (1748?–1803?),197 a court poet from Panna in Bundelkhand, Pahicāne prema rakāne je beparadā darada dariyāva hilai; Magarūra dikhāte ākhira yā dilasūra prema ko paṃtha pilai; Taki tabiyedāra udāra vāhi aru ganai na dhaka dai naina jhilai; Taba khūba iska bodhā āsika jaba mahirabāna mahabūba milai. (Ishk-nāmā 33)198 The one who openly drowns in the river of pain knows the ways of love. Even if the beloved appears proud, the lover after all trod the path of love. Deeming the beloved a generous companion, he does not count betrayals but his eyes are absorbed.

The love of the lover, Bodha, is great when he meets the compassionate beloved. This tradition continued well until the emergence of modern Khari Boli literature, and miscellaneous Rekhta under the name of several poets can be found in Nagari manuscripts dating from the nineteenth century. All these poets were primarily Brajbhasha poets who experimented with Rekhta. In their case this normally meant a Khari Boli template, though it could also be Punjabi as in the case of Anandghan’s Ishk-latā. CONCLUSION Hindavi was a literary language in India along with Persian since the fourteenth century, though there are strong indications that it has been used already some decades after the Muslim conquest of north India. Hindavi songs were sung along with Persian ones in Sufi sama‘ and maybe in other gatherings despite the initial opposition of leading masters such as Sharafuddin Maneri. However, Hindavi did not reach the same status as Persian, and the scarcity of Hindavi material also suggests that Hindavi works were considered inferior to Persian ones in prestige and remained limited in quantity. The few extant examples of Hindavi used at these gatherings are very close to Braj Bhasha both in language and poetic form, not to Khari Boli, which only later became more closely associated with Muslims. Indeed, the earliest use of Khari Boli and of the mixed language in north India can only be documented from the sixteenth-century Mughal court, although there are indications that it had been promoted somewhat earlier in Sufi circles. This suggests that, contrary to perceived notions, Khari Boli literature was cultivated in north India almost parallel to its counterparts in the South, namely Dakkani and Gujri. In north India under Mughal patronage it took the form of macaronic poetry written in Persian metres mixing phrases and half-lines of Hindavi and Persian. Although the normal templates of Rekhta were that of Persian or Khari Boli and the languages used are Hindavi and Persian, there was in fact a wide range of possibilities

within Rekhta, including Braj or Punjabi templates and the mixing of more than two languages or dialects. It was in the experimental and syncretistic environment of the Mughal court, which as Allison Bush also shows in this volume was receptive to Braj Bhasha riti poetry as well, that we find Mughal noblemen writing light-hearted poems with words and phrases from Hindavi. Often these words, phrases and images have to do with the feminine. ‘Ishqi Khan’s poem suggests a mimetic use mirroring the multilingual Mughal court. The example of Abdurrahim Khankhanan demonstrates, that composing in Hindi is a show of virtuosity for the multilingual poet. These are the compositions that can be most pertinently called ‘macaronic’, since they seem to have a mimetic/ ironic bend. While the earliest documented Rekhta poetry developed within Muslim circles, later in the sixteenth century the Nirgun Sant tradition that blended Muslim and Hindu ideas also developed its own Rekhta, written down not in the Persian script but in Nagari and using Indian metres rather than Persian ones. Their Rekhta, just like in the South and probably prompted by similar poems of wandering Sufis, mixed Persian words and phrases into a Hindavi grammar. The Persian-script version of this variety developed into the literary language today called Urdu. The earliest extant Nagari Rekhta texts originated in Rajasthan, namely with Dadu Dayal and his disciples. This use of Rekhta spread mostly among Nirgun Sants but we also find a stray example among the songs of a leading Vaishnava of Vrindaban in the sixteenth century, Svami Haridas. Due to the burgeoning world of Sadhukkari poetry Dadu’s disciples, with a few exceptions, did not continue with Nagari Rekhta and its occurrence remained scarce until the eighteenth century. The earliest extended composition in north India that can be considered Khari Boli is a work by Vajid, the ‘Rekhta of the Lord’, which survives in several early manuscripts and whose composition in all probability predates that of the Bikaṭ kahānī. The fairly complex structure of Vajid’s Rekhta suggests that by the late sixteenth or early seventeenth centuries Rekhta had become an independent genre and Nagari Khari Boli literature existed not simply as occasional rambles into the field but in highly developed

compositions. In Vajid’s case we see a move just opposite to Svami Haridas, a Krishna devotee who wrote a poem like the Nirgun Sants. Vajid was a Nirgun Sant who wrote Krishna poetry. Both authors seem to have been at home in the world of music, and it can be fruitful to examine further the earliest links between the use of Rekhta and music. Material from the seventeenth century suggests that Mughal macaronic poetry integrated the achievements of the Nirgun Sants’s experiments with a Khari Boli template. The best example of this integration is the Bikaṭ Kahānī. By the end of the century mixed poetry in Rekhta, and the domain of Hindavi, had spread quite widely in north India in popular Muslim and syncretic religious groups like Prannath’s sect. Even in Mughal aristocratic circles Persian poets dabbled with Rekhta. The existence of this kind of poetry in north India before 1700 suggests that the fashion for Persianised Hindavi that Vali brought to Delhi did not create Rekhta poetry in north India but rather displaced the pre-existing fashion for mixed language poetry. In the next century more and more examples of Nagari Rekhta are attested, both from courts and by Krishna devotees, showing that Rekhta pervaded all fields of Hindi literature. It was the court of Amber and Jaipur, perhaps the Hindu court most open to Mughal culture, that provided patronage to Rekhta activities in the seventeenth and eighteenth centuries, while among Krishna devotees the leaders of the Haridasi school produced some memorable Rekhtas. In the case of Vaishnavas Perso-Arabic vocabulary was normally assimilated into a familiar Braj Bhasha template. Rekhta production was less in quantity than that of poetry in Braj Bhasha or in Sadhukkari. Yet a careful sifting through manuscript catalogues reveals a significant number of Rekhta works, probably comparable to the heritage of other literary dialects such as Avadhi. We know, however, little about its readership, although several theories can be found attempting to explain the motives for its use. The motives lying behind the use of Nagari Rekhta vis-a-vis Braj or Sadhukkari Bhasha and other Hindi dialects are not entirely clear. Probably the growing success of its use by Muslims prompted some Hindu authors to experiment with it—perhaps to evoke a courtly atmosphere since the court par excellence was a Muslim one, that of

the Mughals. The use of Hindavi within the world of Persian is relatively better researched—Hindavi was used to reach out to people who may have been little acquainted with Persian and more importantly it was considered more effective than Persian in evoking emotions and was also perceived as closer to the world of women. Raja Rao said in his Foreword to Kanthapura that English is the language of our intellectual make-up-like Sanskrit or Persian were before-but not of our emotional make-up. Similarly, in early modern India, Persian may not have provided the same emotive strength as did Hindi and hence their was a continuous need of a mother-tongue for expressing emotions. Since the modern meaning of Urdu implies separation from and contrast to Hindi, it is unfortunate to refer to the language with this word before the late eighteenth or nineteenth century, when this contrast was not felt so markedly by its speakers and when the language was not even called Urdu. Instead of referring to the separate early histories of Hindi and Urdu, I propose to make a different distinction. The word Hindavi (or even Hindustani) can be used to refer generally to the varieties of Hindi and Urdu prior to the articulation of their separate identity and the linguistically neutral phase Khari Boli can refer to the idiom using the template of modern Hindi and Urdu. In all its various forms, Rekhta literature, though neglected by modern scholarship, is more than one of the most important meeting points between Hindi and Urdu; it is the shared early life of the two gradually separated languages.

3 Riti and Register Lexical Variation in Courtly Braj Bhasha Texts Allison Busch

An International and Area Studies Fellowship awarded by the ACLS/SSRC/NEH supported a research leave that enabled this and other projects. I am grateful to Francesca Orsini and Imre Bangha for thoughtful comments on earlier drafts. Thanks are due to F. Nalini Delvoye for sharing her rare collection of H.N. Dvivedi books; to Muzaffar Alam for help interpreting Rahim’s Persian couplets; to Richard Eaton for sharing his expertise on language practices at Dakkani courts; to Sadhana Chaturvedi and the staff at the Hindi Sahitya Sammelan, Allahabad, for allowing me to photocopy the Śṛṅgāramañjarī of Chintamani Tripathi; to the staff at the Alwar branch of the Rajasthan Oriental Research Institute for making available a manuscript of Chintamani’s Bhāṣā-piṅgal

W

ith its literary spaces shared across Hindu and Muslim courtly communities in Mughal-period India, the Braj Bhasha riti tradition is an instructive site for exploring the connections between language practices and other cultural, political, and religious affiliations. On the one hand, the riti corpus is literally defined by its adherence to Sanskrit literary norms, and the largely brahmanical episteme they represent. The word rīti means ‘method’, referring specifically to Sanskrit method, and one of the most prevalent genres of riti literature, the Rītigranth (Book of method), was designed at least in part to be a vehicle

for disseminating classical literary ideas in a vernacular medium. On the other hand, from virtually the moment of its inception the Braj Bhasha courtly style attracted both Mughal patrons and poets—to the extent that the stunning transregional success of riti literary culture from the seventeenth century would be unthinkable without factoring in Indo-Muslim communities. Whereas Sanskrit literature remained largely inaccessible except through sporadic Persian translations, riti literature was a cultural repertory in which Indo-Muslims could and did participate firsthand. Situated at a kind of intersection, then, between Sanskrit and Persianate courtly traditions, what might the writings of riti poets reveal to us moderns about the ‘Hindi’1 of its day—both as a linguistic phenomenon but also as an index of the larger social and conceptual worlds its users inhabited? Broadly speaking, by the seventeenth century the Hindi favoured at regional courts throughout north India, and as far away as Raigarh and Golconda (in modern Maharashtra and Andhra Pradesh, respectively) was Braj Bhasha. But the precise parameters of this ‘Braj Bhasha’ or ‘Braj’, as it is often affectionately called, are far from fixed. Indeed, considerable disagreement exists among scholars about such basic matters as the place and time of origin, and even the name, of early modern north India’s most popular literary language. For instance, is Braj Bhasha to be considered the same as the Gvaliyari (language of Gwalior) used as early as the fifteenth century by Vishnudas (fl. 1435) of the Tomar court?2 Or was Braj literary culture a much later development—a specific byproduct of the Vaishnava fervour that overtook the nearby Mathura region only in the sixteenth century?3 And who is to be the arbiter of competing modern narratives about early modern Hindi? Unfortunately, for the most part pre-colonial authors were not concerned with delineating precisely the language they used—often they just called it bhāṣā/bhākhā: language. Nor do language practices themselves exhibit the kind of homogeneity that might help anchor Braj as a fixed unit of analysis. It is a commonplace in north India that vernacular writers—in strong contrast to their Sanskrit counterparts— were generally indifferent to the delimiting mechanisms of prescriptive grammar until the colonial period. All this means that we find considerable internal variation within the loosely-defined larger rubric of Braj Bhasha.

Of concern to me in this article is one particular aspect of the fluidity of early modern Braj Bhasha: variation in lexical styles. Riti texts exhibit a phenomenon now widely associated with modern Hindi-Urdu, namely a spectrum of written registers ranging from the Sanskritised (tatsama) or semi-Sanskritised (ardhatatsama), to a more basic vernacular style (tadbhava), to a Persianised idiom. Some writers fell predominantly into one particular camp. Others tapped into more than one of these registers depending on context. Yet others used a hybrid style as a matter of course. As modern students of pre-modern north Indian literary cultures from before the Hindi- Urdu ‘divide’, it seems critical to probe the earlier significations and logic(s) of such divergent usages. In what follows I present case studies of different Braj Bhasha styles, examining texts by Keshavdas (fl.1600), Chintamani Tripathi (fl.1660), Bhushan Tripathi (fl.1673), Rahim (fl. c.1600), and Raslin (fl.1740). My aim is to see what larger conclusions can be drawn about the lexical tendencies of pre-modern Hindi authors, and how they may, or may not, differ from those of today. A few words about the unavoidable limitations of this endeavour are appropriate. The scholarly reach required to execute a thorough study of language usages across all riti-period genres—bhakti poetry, courtly poetry, martial ballads, scholarly treatises, and commentarial literature—is an expertise I do not pretend to possess. This article is intended to be suggestive rather than exhaustive. Moreover, some language usages simply do not lend themselves readily to scholarly analysis. Contemporary ethnographic studies abound that establish linkages between language and other kinds of social or political identities; interrogating past practices is not so easy. We cannot ask Keshavdas, or Bhushan, or Rahim, what informed their choice to write in a certain way in a particular poem. And normally these poets, content just to write their poetry, did not ease the scholarly burdens of posterity by deigning to comment on these issues. In the absence of many clear and direct pointers from the past, the modern interpreter of early modern language practices must tread carefully. In the very process of seeking out frameworks for reconceptualising prenineteenth century Hindi, the collective enterprise in which contributors to this volume are engaged, we confront the risk of reading too much of the present into the past. Some fully naturalised modern conceptual structures

—such as the notion that a singular language is a meaningful marker of a particular ethnic group or religious community—would perhaps not have made sense to early modern Indians. Or it may be that the Hindi-Urdu divide of the modern period is, improbable as the notion may initially appear, related to a range of pre-colonial phenomena. Scholarly arguments are invariably made about the colonial-period ruptures in Indian traditions, yet rarely do continuities make headlines. But tracing continuities must also be a part of the process of understanding the relationship between the Hindi-Urdu past and present. As we shall see in what follows, language practices such as Sanskritisation and Persianisation are found in prenineteenth century Hindi texts; it is surely the meanings we assign to these practices that are candidates for re-evaluation, and not the practices themselves. But what meanings are to be assigned? OVER-INTERPRETATION AND UNDER-INTERPRETATION The cultural semantics of riti-period lexical choice have not gone wholly unnoted, and a brief survey of common approaches to the issue is a useful backdrop for the case studies that will follow. In the Kāvyanirṇay (Critical perspective on literature, 1746), a rare premodern text that calls attention to Braj Bhasha’s lexical variance, Bhikharidas mentions both Sanskritised and Persianised language: Bhāṣā Bṛjabhāṣā rucira, kahaĩ sumati saba koi Milai saṃsakṛta pārasyau, pai ati pragaṭa ju hoi Every man of learning recognises Braj as a fascinating language. There are instances of Braj mixed with Sanskrit, and also Persian, but which still remain altogether clear.4 Bhikharidas treats these registers with a kind of neutral, pluralist attitude that is harder and harder to find in modern-day South Asia. It is also interesting to note that Sanskritised and Persianised registers were both considered possible without forfeiting comprehension (pai ati pragaṭa ju hoi).

Discussions of register also arise intermittently in more modern scholarship on riti literature, especially in response to texts marked by a prevalence of Perso-Arabic vocabulary. Some categories of analysis are not necessarily relevant to the pre-modern literary landscape, and we do well to tread cautiously in this conceptual minefield booby-trapped since the days of colonial- and nationalistperiod religious rivalries. One early twentiethcentury British scholar explained the phenomenon of Perso-Arabic style in the Satsaī of Biharilal along communal lines that were notably absent from the work of Bhikharidas: I have been struck with the comparatively large number of words of Persian and Arabic origin which appear with little or no change in this typical Hindi poem…. The extent to which foreign words are used in such a poem at such a date is a striking indication of the penetrative power of the language of the Islamic conquerors.5 According to this model Perso-Arabic lexical forms are indexes of Islamic rule as well as essentially ‘foreign’ rather than integrated into riti writing— an approach to lexical variance subsequently wellattested in Hindi scholarship.6 Other scholars (quite rightly, in my opinion) see the use of hybrid vocabulary in terms of its poetic enrichment of the language.7 Also found is a more functionalist theorization: rather than being viewed as some ‘penetration’ of the body of Braj by Muslim conquerors, using Perso-Arabic words aided in communicating with Indo-Muslim elites, who were patrons of riti literature.8 There is also a class of liberal-minded scholars who view riti-period multilinguality as a component in a larger cultural system of religious ecumenism.9 While this is a welcome departure from the ‘Islamic conquerors’ interpretation, it is an explanatory mechanism that may also call for some caution. Whether language practices are traced to Hindus and Muslims fighting or getting along, the binary logic is still grounded in present-day realities of Hindu-Muslim opposition. If we rely on heavyhanded communalist correspondences between language and religious identity are we obstructing access to other conceptual structures that may have been in place in the past?10 Whatever may be the case, some of these modern models of language use, grounded in a Hindi-versus-Urdu logic that is unreflectively ‘after the

divide’, seem about as delicate as a bull in a china shop when it comes to articulating basic features of riti style. We are surely guilty of over-interpreting if we do not make allowances for how the use of a particular Persian, Arabic, or Sanskrit word might be attributable to some absolutely straightforward cause. Or to a cause that is aesthetic rather than religious. For instance, the Braj corpus consists of far more poetry than prose, and versification had a demonstrable impact on lexical practices. The language choices for a poet working in the doha metre are not the same as those of someone writing a freeform prose passage. In the former case, vocabulary choices may be determined by rhyme scheme or syllable weight more than any other factor. Take the line by Biharilal: Rasa kī sī rukha sasimukhī, hãsi hãsi bolati baina The moon-faced girl of liquid beauty speaks her words laughingly.11 The Persian word ‘rukh’ (rukha in its Braj-ified form) is cleverly paired with the modified tatsama ‘sasimukhi’ (from the Sanskrit) to create a gentle rhyming effect. The Persianness of ‘rukha’ seems completely incidental— except perhaps in that the doubling of meaning across languages (rukh and mukh both mean face) creates an added layer of poetic charm. Similar examples of word choice being predominantly rooted in principles of poetic craftsmanship can easily be multiplied. Aesthetics more than any other principle is likely to have been at work in the occasional instances of Persian vocabulary found in King Jasvant Singh’s overwhelmingly popular Bhāṣābhūṣaṇa (Ornament to the Vernacular, c. 1600), a work that epitomises the courtly cleverness of riti poetry. Consider the mixed language of the following doha: Ati kārī bhārī ghaṭā pyārī bārī baisa Piya paradesa ãdesa yaha āvata nāhi sãdesa The dense clouds darken, a sweet girl in the bloom of youth. With her lover abroad, she is anxious—why has he not sent any message?12

In this ultra-concise rendition of a typical viraha theme, the two Sanskritderived words paradesa (foreign country) and sãdesa (message) are expertly linked with a Braj-ified form of the Persian word andesha (anxiety) to create beautiful alliterative effects and internal rhyme. That is all. There is no religious or political implication. Although Jasvant Singh knew Persian and had close ties to the Mughal court he was not trying to enhance communication with Muslims by employing Persian words. Nor were his Sanskrit words directed at Hindus. There simply is no larger point to be made about the matter. For a riti writer with broad exposure to a range of different social milieux, courtly style and elegance were possible in both Perso- Arabic and Sanskrit registers, and there even seems to have been a certain delight in mixing them. It is also undoubtedly the case that some Persian vocabulary was unmarked in Braj Bhasha usage. The word kāgad and its variant kāgar (from Persian kāgaz), for example, is commonly attested in riti literature, but this is hardly a matter for etymological deconstruction since the object in question was not available in pre-Islamic India. Thus it only makes sense that Braj writers would routinely employ a Persian loanword alongside the more Indic term pāti (leaf), which stems from an earlier technology of palm-leaf writing. Furthermore, many riti poets had close contact with branches of Mughal administration, and depending on the poet and the court and the time period, Persian words could be just as fully naturalised as Sanskrit ones. While recognising that the bulk of riti writers were brahmins and often well-versed in Sanskrit (who therefore could be expected to know whether a word was a Sanskrit derivative or not), who knows if they were always conscious, in the way scholars of Hindi-Urdu are today hyperconscious, of the roots of individual words they used? Are even highly educated modern speakers of English particularly aware of when they are using Latinate versus Germanic vocabulary? While alert to the problem of over-interpreting language practices in terms of crude modern ideologies that construct Persian lexemes as integral to Muslim language practice, enjoining Sanskritisation as the mark of ‘śuddh’ Hindi and Hindus, we also need simultaneously to resist an overlycautious tendency towards under-interpretation. It would be unwise to swing to the opposite extreme and begin arguing that they were used in a

free-form, value-free vacuum. Such an interpretation of early modern language practices is equally problematic, and demonstrably implausible. REGISTER IN CONTEXT In the lines by Biharilal and Jasvant Singh just quoted, no kind of consciousness with respect to choice of Persianised versus Sanskritised vocabulary seems to be in evidence, except perhaps for something that could be called poetic consciousness. But is this all there is to the story? Can other connotations to lexical choice be discovered, and theorised? One thing is certain: the use of register can vary considerably across the spectrum of riti poets, and even within the oeuvre of a single poet. Such variations do not appear to be random. In what follows I will track particular types of language use as they occur in their individual contexts, and reflect on the patterns that emerge.

1. The writings of Keshavdas The works of Keshavdas Mishra encourage an analysis of lexical register as a set of changeable rather than fixed language practices. Keshavdas was from a family of learned brahmins, well-versed in Sanskrit literary and intellectual traditions. But this traditionalist bent was mitigated by external forces: during the poet’s own lifetime his patrons’ kingdom, Orchha, became a tributary state of the Mughal empire. Thus, his writings afford an exceptional opportunity to study the language practices of a classicallytrained author who, through increasing contact with the Persianate world, was exposed to a new array of linguistic and cultural possibilities. True to Keshavdas’s reputation in the Hindi tradition as an ācāryakavi (scholar-poet), classicising tendencies are strongly evident, both stylistically and lexically, throughout much of his oeuvre. His major scholarly works (Rasikpriyā, 1591 and Kavipriyā, 1601) show a distinct predilection for Sanskritised language. But such a style was encouraged—if not necessitated —by his intellectual task: the exposition of technical aspects of Sanskritderived literary categories such as heroines and heroes (nāyikās/nāyakas), poetic moods (rasas) and rhetorical devices (alaṅkaras), which form the

basis of the Braj rītigranth genre. The following typical ‘definition’ (lakṣaṇa) from the Rasikpriyā illustrates the kind of linguistic imperatives at work: Sādhāraṇa-nāyaka-lakṣaṇa Abhimānī tyāgī taruna, koka-kalāni prabīna Bhabya chamī sundara dhanī, suci-ruci sadā kulīna Ye guna ‘Kesava’ jāsu mã soī nāyaka jāni Anukula dacha saṭha dhṛṣṭapuni caubidhi tāhi bakhāni General definition of a hero A hero is self-confident, willing to sacrifice, young, and skilled in the arts of love. He should be prosperous, forgiving, handsome, wealthy, wellgroomed and always from a good family. Says Keshavdas, these are the recognisable qualities of a hero. And the category of hero is held to be four-fold: faithful, expert, deceitful, and brash.13 The language employed here is almost pure Sanskrit with only the thinnest veneer of vernacularisation: except for typical Braj modifications of Sanskrit phonemes such as ‘cha’ for ‘kāa’ or ‘ba’ for ‘va’, all but line 3 consists mostly of words that are virtually indistinguishable from Sanskrit. Such a register, however, is chosen for certain contexts, and not others. Despite their scholarly focus even the ritigranth works contain other styles. It is typical of the genre to alternate between definition verses (lakṣaṇa) and poems that illustrate (udāharaṇa) the author’s theoretical propositions. The latter verses tend, significantly, towards tadbhava style. An illustration of the manifestly faithful hero (prakāśa anukūla nāyaka) taken from the very same page of the Rasikpriyā is far less Sanskritised than the definition: ‘Kesava’ sūdhe bilocana sūdhī bilokani kõ avalokẽ sadāī Sūdhiγai bāta sunẽ samujhẽ kahi āvati sūdhiyai bāta suhāī Sūdhī sī hā̃ sī sudhānidhi so mukha sodhi laī basudhā kī sudhāī Sūdhe subhai sabai, sajanī, basa kaisẽ kiye ati ṭeṛhe Kanhaī. Keshav says, Her eyes are straight, She always looks into your eyes with straightforward innocence

She is straightforward when she listens to you, And in the way she understands. Her charming replies are also straight-laced, Her laughter is straight, Her moon-face has absorbed the world’s straightness. Oh friend, her character is straight in every respect! How did she bring into her thrall this most-crooked Krishna?14 Here and elsewhere throughout the corpus of Braj poetry, tadbhava language is chosen for recounting the escapades of Krishna and the gopis. Although Keshavdas purports to be analysing the traits of a male character in this verse, the point of view is actually that of a woman, the girlfriend of the heroine, who comments on the power Radha has over Krishna. This is Keshavdas’s way of illustrating the nāyaka’s faithful or ‘anukūla’ qualities. Since the speaker is an uneducated, unsophisticated gopi, a de-Sanskritised register lends verisimilitude to the poet’s impersonation of a woman. A further factor in the less formal register of udāharaṇa verses is that, unlike the lakṣaṇas, they are not normally based on Sanskrit models. These less theoretical, and more poetic, portions of the ritigranths are the riti writers’ independent creations. De-Sanskritised lexical style is found elsewhere in the Keshavdas corpus with very different poetic effect, as in the poet’s first work, the Ratnabāvanī (Fifty-two verses about Ratna). Written in c. 1583, only a few years after Orchha capitulated to the Mughals, the Ratnabāvanī is a martial tale of resistance centring on the bold yet ill-fated efforts of the Orchha prince Ratnasena, who tried to protect his father’s kingdom from the Mughal onslaught. The subject of the work and its dominant mood of heroism (vira rasa) seem to demand a particular idiom. A combination of tadbhava words, Prakritising archaisms, heavy retroflex sounds, and a pronounced doubling of consonants create dramatic onomatopoeic effects, mimicking the cacophony of armies as they clash in battle: Tahã amāna paṭhṭhāna ṭhāna hiya bāna su uṭhṭhiva Jahã ‘Kesava’ kāsī-naresa dala-rosa bhariṭhṭhiva Jahã tahã para juri jora ora cahũ dundubhi bajjiya Tahã bikaṭa bhaṭa subhaṭa chuṭa ka ghoṭaka tana tajjiya

Jahã Ratanasena rana kahã caliva halliva mahi kampyo gagana Tahã hvai dayāla Gopāla taba biprabhesa bulliya bayana The battlefield was filled with countless Pathans shooting arrows, hearts intent on war. Keshav says, the Prince of Kashi (Ratnasena15) urged his warriors forward. The soldiers engaged their enemies with force, and the sounds of war drums rang loudly in all directions. Fearsome warriors went careening from their mounts, giving up their lives. Wherever Ratnasena led his soldiers in battle, the earth shook, and the skies trembled. Then suddenly the merciful Vishnu came to earth, disguised as a brahmin. He spoke to Ratnasena.16 Since Braj Bhasha poetry is otherwise characterised by an avoidance of conjunct consonants, the effect of the dense sound clusters here and elsewhere in the work is striking. Their choice is deliberate: the style is characterised by what Indian literary theorists call the ‘literary property of vigour’ (ojas-guṇa), and it hearkens back to old concepts that underlie both Sanskrit martial poetry and the Hindi rāso with its distinct ethos of opposition.17 Yet another aspect of Keshavdas’s style—testament to the poet’s immense versatility and range—can be traced in his experiments with Braj forms of Sanskrit courtly kāvya (refined poetry). His Rāmcandracandrikā (Moonlight of the deeds of Ram, 1601), Vīrsiṃhdevcarit (Deeds of Bir Singh, 1607) and Jahāngīrjascandrikā (Moonlight of the fame of Jahangir, 1612) favour verses written in an elevated style, the fashioning of which required the deployment of an entire arsenal of Sanskrit-derived rhetorical flourishes including classical topoi of kingly glory, a wide range of figures of speech (alaṅkāras), pronounced compounding, and complex metrical forms. In most respects the poet’s Sanskrit literary models seem to dictate

both style and lexical content. It is noteworthy that all three of these kāvya works are about kings: the ideal King Rama of epic lore, but also two of Keshavdas’s contemporaries: the poet’s patron Bir Singh Deo (r. 1605– 1627) as well as the Mughal emperor Jahangir (r. 1605–1627). Formal verses of praise (praśasti) and iconic descriptions of a ruler seem to demand a kāvya-idiom for which Sanskritised language is particularly well suited.18 The Vīrsiṃhdevcarit, for instance, may be seen in part as Keshavdas’s protracted literary argument about his patron’s fitness to rule, and the building blocks of this argument are elaborate, Sanskritised verses about Bir Singh’s kingly glory, the elegance of his court, and the wellbeing of his subjects. Here the poet also invoked Sanskrit literary models (especially Bana’s Kādambarī and the Vāsavadattā of Subandhu19), and in the cantos leading up to the coronation, where establishing the moral authority of Bir Singh is of paramount importance, he even goes so far as to weave purely Sanskrit verses into his tale.20 Keshavdas’s Jahāngīrjascandrikā, a panegyric to the Mughal emperor written towards the end of the poet’s life, may well be more Sanskritised than any of his other works, its Indo-Muslim hero notwithstanding. Several verses are almost completely tatsama in their construction, such as the following one comparing Jahangir to Indra, king of the gods: Kavi, senāpati, kusala kalānidhi, gunī gīrapati Sūra, ganesa, mahesa, śeṣa, bahu bibudha mahāmati Caturānana, sobhānivāsa, śrī dhara, vidyādhara Bidyādharī aneka, mañju ghoṣādi cittahara Dṛṣṭi anugraha-nigrahani juta kahī̃ ‘Kesava,’ saba bhā̃ ti chama Imi Jahā̃ gīra suratāna aba dekhahu adbhuta indra sama. See how the emperor Jahangir is as astonishing as the god Indra In his court are poets and generals, skilled artists and discerning scholars, Warriors, officers, stable masters, shaikhs, masterminds The clever, the glamorous, the lustrous, A range of entertainers and their companions. There are beautiful songs, haunting to the soul. Keshavdas says, Jahangir is a capable ruler in every respect— He is kind to the deserving, and harsh towards those who break the law.21

This is almost pure Sanskrit! Sanskritised language for an Indo- Muslim ruler? Such a style defies our expectations. But these are modern expectations. The Jahāngīrjascandrika is a telling example of how language practices ‘before the divide’ need to be approached in ways that tease out their unfamiliar rationales. Sanskrit is not a ‘Hindu’ language in this text: it is a lexical code chosen precisely because it speaks to moral perfection and kingly authority like no other. Still, there is a twist to this verse, which can be read as an extended double entendre (śleṣa), in which Jahangir and the Hindu god Indra are simultaneously glorified.22 Peeping out from behind the hyper-Sanskritised style, it turns out, is a multi-lingual pun that hinges on two possible pronunciations of the word ‘śeṣa’. Read as a Sanskrit word in relation to Indra’s court, it means Sheshanaga, the serpent companion ofVishnu. But the same word, when pronounced in the Braj manner, sounds like ‘Shaikh’, allowing it to double as the Arabic word for spiritual master. This brings us to a final factor that must be considered in any analysis of lexical register in Keshavdas’s poetry: a new tendency towards PersoArabic vocabulary in select parts of his last two works. Persianised vocabulary is virtually nil in Keshavdas’s writing until the Vīrsiṃhdevcarit, which was commissioned in 1607—at a time when Bir Singh Deo’s political ascent was being backed by Jahangir. Here we begin to see an occasional Persian word entering Keshavdas’s diction, particularly for Mughal contexts.23 And this tendency becomes more of a habit by the time of the Jahāngīrjascandrikā written five years later, as when Bir Singh Deo Bundela is accorded a new Persianised title ‘bakhata bilanda’ (high-fated, i.e. fortunate) in the following mixed-register doha: Nakhata somaṭata nakhata so, bakhata bilanda bisekhi Bhāga, birājata kauna yaha, kahijai nakha-sikha dekhi. He is especially fortunate, Like a star crossing the edge of the moon. Oh Fate, who is this illustrious man? Describe him from head to toe.24

In this short verse tatsamas, tadbhavas, and Braj-ified Persian forms interplay to produce rich alliterative resonances. These are clearly the careful choices of a masterly poet-craftsman. Nonetheless, unlike in the poems of Bihari or Jasvant Singh cited above, here hybridity in language seems to carry more than a purely poetical value. The mixed language of the verse captures perfectly the stately cosmopolitanism of Keshavdas’s patron Bir Singh Deo Bundela sitting in attendance at the Mughal court, and it speaks powerfully to the new alliance between his patron and the emperor Jahangir. Similarly, given the lexical profiles of Keshavdas’s earlier compositions it is difficult not to be struck by something new when Jahangir is addressed as ‘Ālama panāha kulli ālama ke ādamī ’ (Shelter of the world, man of the whole world, vv. 167–168) or his son Khusrao is praised as the recipient of a ‘khalaka kī khūbī ko khajāno’ (treasure house of all earthly good qualities, v. 55). These and other Persianised phrases probably did not just flow naturally from Keshavdas’s pen. In the context of his corpus they are anomalous and seem carefully studied. The skilled manipulation of Perso-Arabic vocabulary can be seen as partly an aesthetic touch that was intended to produce a ‘Mughalising’ effect. Smidgeons of Persian evoke the Mughal courtly environment, but they also seem to be indicative of a sense of cultural rapprochement with the Mughals, which evolved in the course of Keshavdas’s oeuvre no less than in the political climate of the Orchha state. It will be clear from even this brief sketch that studying the works of the single poet Keshavdas yields a tremendous range of register profiles. And these cannot be classified neatly along lines of religious affiliation. Although Persianised vocabulary is likely to be found in a Mughal scene and not elsewhere, Sanskritised language is found in various contexts, and these cannot be construed as Hindu— particularly when such a register is considered suitable for Jahangir. Tadbhava language also has its identifiable literary spaces such as feminine speech or, when configured slightly differently, martial scenes. Are these observations confirmed, nuanced, or countered by the register profiles of other riti writers?

2. The writings of Chintamani Tripathi

Variability of lexical register is also a feature of the work of Chintamani Tripathi, who was alongside Keshavdas, one of the leading poets of the early riti tradition. Although a fully accurate evaluation of Chintamani’s oeuvre awaits the publication of his complete corpus, his available works provide enough clues to allow some exploration of trends in lexical choices. In many respects the logic is similar to what we find in the writings of Keshavdas. Like Keshavdas, Chintamani was a brahmin well versed in Sanskrit traditions; he too was keenly interested in crafting vernacular renditions of the principles of Sanskrit literary theory, and most of his known works are ritigranths. His Kavikulakalpataru (Wish-fulfilling tree for the brotherhood of poets, c. 1670) and Śṛṅgāmmañjarī (Bouquet of Passion, c.1666) are frequently Sanskritised in their scholarly style, especially in the lakṣaṇa verses.25 Both draw heavily on Sanskrit sources; indeed, the Śṛṅgāramañjarī is even a fairly direct translation of Akbar Shah’s Sanskrit text of the same name.26 A close connection to Sanskrit source material and the nature of the technical subject matter explain the tendency towards tatsama style. The Śṛṅgāramañjarī is of particular interest to any would-be theorist of pre-modern language practices because it contains extensive prose passages. Prose is relatively rare in Braj—Keshavdas, for instance, does not use it in any of his eight works. Significantly, Chintamani’s prose style, unconstrained by the exigencies of rhyme or metre, is strongly inclined to Sanskritised vocabulary. The work opens with what must surely be the longest compound in all of Braj Bhasha literary history, which, in its nearly one hundred word abundance and tatsama lexicon, hearkens back to the most complex of Sanskrit ‘gadya’ styles.27 Whereas current linguistic patterns of Hindi language nationalism in India suggest that Sanskritised Khari Boli is a modern practice, I think R.S. McGregor is correct to draw attention to its Braj antecedents. His study of the Braj commentaries of Keshavdas’s patron Indrajit (c.1600) finds ‘clear evidence that a Sanskritised style of speech of high prestige existed and was well recognised’ in early modern India.28 Chintamani’s Ŝṛṅgāramañjarī, like the prose writings of Indrajit, is a useful reminder that a Sanskritbased Hindi prose was neither a colonial invention, nor an exclusionary by-product of

modern Hindu-Muslim rivalries. Certain scholarly contexts seem to have encouraged or even necessitated its use hundreds of years ago. What about Persianisation in the works of Chintamani—are there patterns to be detected in this register? After all, his translation of the Ŝṛṅgāramañjarī was commissioned in an Indo-Muslim cultural setting. The exact conditions surrounding the commission are unknown, but Chintamani’s work was probably produced at the Golconda court since Akbar Shah, the text’s purported author, was son of the preceptor to the Qutb Shahi ruler Abul Hasan ‘Tanashah’ (r.1672–1687).29 It turns out that one looks in vain for any strong tendency towards Perso-Arabic vocabulary in the work—even in the lengthy introduction containing verses in honour of several Indo- Muslim notables (including praśasti verses to Akbar Shah himself). In fact, except for two Persianised lines, the entire prelude to the poem, ninety-six lines in length, is, if anything strongly, Sanskritised.30 Asking whether Perso-Arabic vocabulary should be used for Muslims is simply the wrong question to pose in the case of this poet. As we have seen, Persianisation was something that crept into Keshavdas’s writings over time and does in his case seem truly to be a marker of contact with the IndoMuslim world, as his patron Bir Singh Deo Bundela forged ties with Emperor Jahangir. But Chintamani’s case was different. He was writing in a later epoch; he was a cosmopolitan poet patronised by a range of courts: Hindu and Muslim, Northern and Deccani. His usages are harder to classify. There is no little irony in the fact that a verse in honor of the Hindu king Shahaji Bhonsle from the opening to Chintamani’s unpublished Bhāṣāpiṅgal31 (Treatise on vernacular prosody, c. 1662) is far more Persianised (in lexicon if not in imagery) than any of his praśastis to IndoMuslims: Kavina ko rājai-bhoja, voja ko saroja-bandhu dīnani ko dayāsindhu, lāja-sīla ko jahāja” Koṭi kāma sundara hai, sāhibī purandara hai, Mandaru hai vairī-bala vāridhi-mathana kāja Janga mai jālima, avalamba kuli ālama ko, bālama dharā ko, saba sūrana ko siratāja, Vikrama apāra, sakra sujasa ko pārāvāra

bhārī bhāratha mana samattha sāhi mahārāja Maharaja Shahaji is a King Bhoja to his poets, When it comes to lustre he is the sun. To the poor he is an ocean of compassion, He is a large ship when it comes to the extent of his good character. He is attractive like a crore of Kamadevas, In grandeur he is Indra himself, He is like mount Mandara, poised to churn the ocean of enemy powers. He is ferocious in battle, and a stronghold for the whole world. The earth’s darling, the crown of warriors, His prowess is endless—he has attained the boundless fame of Indra. His courage withstands even the heaviest battle.32 This is a startlingly mixed verse with both Sanskritic compounding (bandhu-dīnani ko dayāsindhu…) and strong Persianisation (Janga mai jālima … kuli ālama); in which the poet takes the liberty of juxtaposing the Arabic word ṣāḥibi with the tatsama Purandara (Indra). Rhyme is obviously a major consideration in the vocabulary choices, but the kind of Persianisation in evidence was probably not something remarkable in its day. We should not forget that Persian was part of the cultural repertory of a certain class of Hindu court poets from the seventeenth to nineteenth centuries. Perhaps we can also ascribe Chintamani’s language style to local conditions in the mid seventeenth-century Deccan, where Persianised style and hybridity were both regular occurrences.33

3. The writings of Bhushan Tripathi We also find the traits of hybridity as well as a dense distribution of Persian vocabulary in the writings of Bhushan Tripathi, who is widely thought to be Chintamani’s younger brother. Bhushan is famous in the annals of Hindi literary history as the court poet of Shivaji (Shahaji Bhonsle’s son), for whom he wrote his magnum opus the Sivarājabhūṣaṇ (Ornament to King Shivaji) in 1673. The work is ostensibly a ritigranth, where one might expect the use of more Sanskrit than Persian, but this work is filled with

surprises. First of all, for the most part, Bhushan proves himself to be manifestly uninterested in any scholarly pursuit of poetic theory. Instead of composing his own lakṣaṇa verses he was content to copy those of Matiram Tripathi (like Chintamani, in all likelihood, one of Bhushan’s brothers).34 His udāharaṇa verses also depart dramatically from the bucolic love scenes of Krishna and Radha—the usual fare of Braj poetry—focusing instead on the clamour of Shivaji’s battles and the laments of his beleaguered enemies. It is in Shivaji that Bhushan proves to be most interested, perhaps because the work was in all probability commissioned as a praśasti for the occasion of his patron’s coronation.35 The multi-layered resonance of the word bhūṣaṇ (ornament) in the title constitutes a brilliant feat of semantic acrobatics, and speaks to the complexity of the author’s project. In offering up his poem as an ornament to his patron Bhushan invites us to think of it as a panegyric, but bhuṣan is also a synonym for alaṅkāra or rhetorical device, the literary topic under discussion in the work. Bhūṣaṇ is additionally, of course, the signature (chāp) and title of the poet himself.36 The linguistic profile of the Śivarājabhūṣaṇ is unusual, in keeping with the work’s atypical combination of objectives. It is written in a mixed style, with extreme feats of both Sanskritisation and Persianisation in evidence. The praśasti aspects of the text seem to invite the dense levels of figuration characteristic of Sanskritised kavya style and in this respect, the work bears comparison with the Vīrsiṃhdevcarit and Jahāngīrjascandrikā. Like Keshavdas, Bhushan— a fellow brahmin, after all—was perfectly capable of using recherché Sanskritic words, compounding techniques, and all manner of classical poetic devices. The Sanskritised register invokes an age-old moral vocabulary of rectitude and valour to present an idealised vision of his patron. But Sanskritisation is only one side of Bhushan’s language-coin. The flip side is the Persianised register to which we have already alluded. Much of the Persianised language consists of ordinary workaday words and is not particularly associated with Indo-Muslim characters. A heavy degree of Persianisation makes a certain amount of sense in the text because the Śivarājabhūṣaṇ is profoundly about seventeenth-century politics, and Persian was the language of power. Perso-Arabic and even Islamic epithets for Shivaji, such as ‘gāzī’ (‘victor over the unbelievers’),37 however, give

rise to cognitive dissonance in the unprepared reader, in whom modern Indian cultural memory has enshrined Shivaji as the ultimate ‘Hindu’ rebel fighting his ‘Muslim’ enemies. A well-informed reader may also be aware that Marathi took a turn toward the Sanskritised during Shivaji’s reign—but this was only in the final years of his life. Bhushan’s work, written seven years before the death of Shivaji, shows no evidence of such dePersianising measures.38 His writings, like those of his brother Chintamani, serve as a powerful reminder that Persianised language was not the distinct marker of a particular religious or cultural community in the seventeenthcentury Deccan. By drawing on both Sanskritised and Persianised language Bhushan had all the bases covered: he invoked an old Hindu authority bolstered by hundreds of years of traditional kingly representation; he also spoke the language of the court politics of the here and now. It has been suggested that by drawing on Persian, particularly the words that were common in the heavily Persianised Marathi of the day, Bhushan, a northerner, could make his work more intelligible to an audience that lacked fluency in Hindi.39 There may be at least some truth to this assessment, although such a functionalist explanation is not wholly adequate. Interestingly, it was Khari Boli Hindi rather than Persian that apparently had Indo-Muslim associations for Bhushan, who occasionally seems to go out of his way to use Khari Boli verb endings instead of Braj ones for recorded Muslim speech.40 All of this underscores the multi-valence and flexibility of Braj Bhasha. During the seventeenth century it became a language that travelled vast distances, and along the journey it encountered a range of courtly contexts and regional linguistic practices, to which poets adapted. The writings of Matiram Tripathi, who worked for small scale Hindu patrons in northern India rather than Deccani or Mughal rulers, are far less Persianised in style, and this variability of language practices among brothers underscores the point that a poet’s literary language is not a given of birth or caste or community, but one of choice. There are some additional features of Bhushan’s own language choices, features we have not yet encountered in this study. While partly a celebration of the military feats of Bhushan’s famous patron, the Śivarājabhūṣaṇ is also a strong statement of Shivaji’s disillusionment with the Mughal political establishment—and even, in places, a denunciation of

major figures such as emperor Aurangzeb. It can be a strongly, sometimes bitterly, satirical text, and the satirical effects in some cases stem precisely from Bhushan’s deft manipulation of language in ways that could not be more different in spirit from what we observed in the poetry of Keshavdas. If Keshavdas’s experiments with mixed language style are by all indications gestures of cultural inclusiveness, Bhushan uses that same style on occasion to create a mocking, hostile mood. Consider his etymologically corrupt but thematically brilliant handling of Aurangzeb’s name. In Persian the word Aurangzeb is a flattering title, meaning ‘adorning the throne’. In Bhushan’s hands the word ‘Aurang’ is Braj-ified into ‘Avaranga’.41 According to Braj phonetics this is a plausible enough pronunciation of the emperor’s name, but it also invokes the combination of the Sanskrit lexemes ‘ava’ and ‘raṅga’, which together mean something like ‘sickly pale’—a point that could hardly have been lost on a brahmin like Bhushan. This deliberate Sanskritisation of the emperor’s Persian name suggests Aurangzeb’s overwhelming trepidation in the face of Shivaji, transforming his noble title into a source of derision. Examples of derisive word play in the Śivarājabhūṣaṇ could easily be multiplied. Some stem from precisely this peculiar feature of Braj Bhasha: the ability for particular words to be read simultaneously in both Sanskrit and Persian registers. Like Keshavdas, Bhushan also employs punning techniques from Sanskrit (in this case yamaka, the repetition of a single word that invokes more than one meaning), but to dramatically different effect. Note the play on the word pīra in these lines: Sāhitanai sivarāja kī dhākani, chūṭi gaī dhṛti dhīranhū kī Mīrana ke ura pīra baṛhī yau, ju bhūli gaī sudhī pīranhū kī. Shivaji, son of Shahaji, struck such terror in the hearts of Muslim nobles that even the bravest lost their nerve, Their affliction grew such that they forgot the teachings of the Sufis. The first usage of the word pīra invokes the Sanskrit meaning ‘pīḍā’ (affliction). But turning to the Persian lexicon the same word as it is

typically written in Braj can also mean a Sufi Pir. A similar bilingual mocking of the enemy is evident in: Dīnadayālu na to so dunī aru mleccha ke dīnahĩ māri miṭāvai There is no one in the world as merciful to the oppressed as you And you wipe out the faith of the Mlecchas.42 Here dīna first occurs as part of a Sanskritised compound meaning ‘merciful to the poor’, an appropriate kingly epithet for Shivaji. In the second half of the line, however, the same word is used in the Arabic sense of religious faith, which Shivaji is said to be wiping out. This last line seems to be a deliberate inversion of the more typical image of Muslim rulers razing Hindu temples and religious artifacts. One thing is clear: Bhushan uses both Sanskritised and Persianised vocabulary to striking effect in his work, and these practices require a far more complex analysis than a simple division along the lines of Hindu versus Muslim would allow. In considering the cases of Keshavdas, Chintamani Tripathi, and Bhushan Tripathi we have noted the multiple rationales that seem to underlie differential patterns of register. Sanskrit is the language of technical literary jargon for a ritigranth, but also the language of kingly perfection appropriate to praśasti-oriented genres. Persian is a workaday language for some courts; it is the language of politics; it is sometimes but not invariably employed for Indo-Muslim contexts; multi-lingual puns are also employed with radically different intentions. A consideration of riti writings by select Indo-Muslim authors reveals additional patterns.

4. The writings of Rahim The poetry of the Mughal administrator Abdurrahim Khan-i Khanan (1556– 1627) is a particularly promising site for an investigation of Hindi register. He was voraciously multi-lingual, and this trait seems to have had a tremendous impact on his Hindi literary style. Rahim naturally knew Persian, the major imperial language of the Mughals, and was a famously generous patron of Persian poets in his day.43 His generation still had a

connection to Turki, the native language of the earliest Mughal rulers, as evinced from his Vāqi’āt-i Bābarī, a translation from Turki into Persian of Emperor Babur’s memoirs. Rahim was also conversant with a range of Indian regional languages. He is even said to have learned both Sanskrit and Portuguese. As far as composing poetry is concerned, he is credited with some verses in Sanskrit and Persian, but the bulk of his literary output seems to have been in Hindi.44 Half a dozen collections of his Hindi poems have come down to us, which, if authentic, would be compelling testimony to his multilingual poetic skill. Unfortunately, however, none of the texts has been dated, and a thorough review of available Rahim manuscripts remains a desideratum. Still, one can at least venture some preliminary findings about his writing on the basis of the existing published works.45 Rahim’s literary talents in Hindi ranged across many dialects (Avadhi, Braj and Khari Boli), and within these, various lexical registers ranging from Sanskritised to tadbhava to pure Persian are all attested. In analysing Rahim’s Hindi style(s) the first observation to make is that the variety within the texts embodies a set of cultural practices in the outside world: the poetry through its mixed language enacts a kind of Mughal cosmopolitanism. And this seems to be precisely the point—or at least one of the points—of Rahim’s poetic experiments with Hindi. Without wishing to belabour stereotypes about early Mughal ecumenism, there was something about Rahim’s particular historical moment that brimmed with cultural newness and exploration. Mughal power was expanding and, as one of the empire’s key purveyors and protectors, Rahim travelled extensively throughout the subcontinent. European outposts dotted the coasts, their ambassadors visited the Mughal court, trading in a range of cultural currencies—from Flemish painting to Christian religious precepts. It really should not surprise us, then, if the poetry of this leading Mughal notable deeply reflects its multicultural surroundings. Rahim’s register of Mughal cosmopolitanism is evident throughout his oeuvre, but perhaps nowhere more dramatically than in his Madanāṣṭaka (Eight verses of love).46 The title, the metre (mālinī) and the aṣṭaka genre all proclaim the Madanāṣṭaka’s partial location in a Sanskrit literary field. But the base language as determined by verb structures and postpositions is, incongruously, Khari Boli. Contributing to the text’s pronounced upending

of customary literary practices are the stark juxtapositions of both Sanskritised and Persianised vocabulary in configurations that vary from verse to verse. The first line begins with pure Sanskrit: Śarada-niśi niśīthe… (at midnight on an autumn night…) And then our poetic expectations are suddenly thwarted when the poet completes the line with the plodding long vowels typical of Khari Boli: …cā̃ d kī rośanāī (the light of the moon). The verse turns out to be about Krishna leading the gopis away from their respectable family lives into the forest for love-play on a moonlit night—a typical theme of its period. But there is nothing typical about the language. A line about Krishna reads: Zarad basan-vala gul caman dekhtā thā (The one with the yellow garment looked at the rose garden, v. 5). The phrase zarad basan is a mixed register calque on pitambara, a common Sanskrit/Braj epithet for Krishna. And mixture is the name of Rahim’s poetry-game. Both in lexicon and topoi the Madanāṣṭaka transports us back and forth from the kuñj of Vrindavan to the poetic world of the Persian ghazal, surprising the reader at every turn. Krishna plays his flute on a moonlit night, enchanting the gopis in a manner familiar from centuries of Indic poetic representation; but he is also enraptured by a gopi’s hair (expressed in the language of ghazals: zulfẽ), and sips from the proverbial cup (pyāla) of the lovelorn, getting drunk in a style reminiscent of images from Persian poetry. Rahim unites Indic and Persian language, as well as motifs, in dramatic fashion in this text. Rahim’s other collections of Hindi poetry may not be as boldly macaronic as the Madanāṣṭaka, but they are still generally mixed in lexical profile.47 It is not an easy task to pinpoint why, but Rahim’s default register is atypical of others in the Hindi literary tradition. Rahim and Keshavdas, for instance, were exact contemporaries— and they almost certainly knew one another48—but their language practices are very different, particularly in the extent and choice of Perso-Arabic vocabulary. How do we theorise

this difference in degree and style of Persianisation? Along the lines of Hindu versus Muslim? Courtly context? Cultural orientation? Or can we theorise it at all? The opposition of Hindu versus Muslim seems least likely to yield any useful analysis since there is almost nothing Islamic about Rahim’s Hindi texts.49 Courtly context and cultural orientation are more promising, if not fully satisfactory. As a Mughal courtier Rahim travelled in a world that prized refined Persian speech and poetry. And when he was not speaking Persian his default vernacular probably tended towards a relatively Persianised Khari Boli. But how do we account for Rahim’s use of Sanskritised style and even, on occasion, pure Sanskrit?50 Partly it is his cosmopolitanism at work, but Rahim’s hyper-variegated lexical practices may also be seen as a kind of revelling in the poetic power of Braj Bhasha. Rahim’s work serves as a powerful reminder that there is nothing natural about writing poetry. It is a highly conscious act, necessitating the careful selection of words for particular effects. Perhaps here more than anywhere we must be careful about over-interpretation. For it is precisely its quirky mixes and hybridization that give Rahim’s Hindi poetry so much of its charm, and an overly-reductionist deconstruction of what it all means not only risks spoiling the beauty of the enterprise, it isn’t even adequate to the task. More than any of the poets discussed so far, Rahim seems to mix vocabulary as a gesture of poetic playfulness. The playful effects are intensified by his modifying words in a highly idiosyncratic manner. Take, for instance, his strange Avadhi-fication of both Persian and Sanskrit words in his Barvai nāyikābheda.51 The addition of the suffix ‘va’, sometimes accompanied by a shortening of the preceding vowel, produces a diminutive effect in Eastern Hindi dialects like Avadhi. But Rahim plays with this ‘va’ suffix obsessively—even ludicrously— throughout the work. Particularly incongruous is its repeated application to Sanskritic compounds: Madhya vipralabdhā nāyikā Dekhi na keli-bhavanavā, nandakumāra Lai lai ū̃ca usasavā, bhai bikarāra.

The middling type of frustrated heroine She did not see Nanda’s son in the pleasure-house Sighing long and hard, she became restless.52 Forming a diminutive from a tatsama word like ‘kelibhavana’ (pleasurehouse) was just not done—not in the real world, at any rate. But this is the world of poetry. Nor are Persian words spared, as in the treatment of ‘gumān’ (pride/haughtiness) in this verse: Adhama nāyikā Berihi bera gumanavā, jani karu nāri Mānika au gajamukutā, jau lagi bāri. The lowest kind of woman Oh woman, don’t get in a huff time after time, Otherwise I’ll have to buy rubies and pearls.53 These idiosyncratic modifications of words to generate the impression of eastern language are poetic effects, stemming perhaps partly from a concern to generate the right metrical weight in each part of the tightly-controlled and ultra-concise barvai line; they are also, surely, a delightful exploration of new literary possibilities in a vernacular language that was eminently suited to experimentation. The Hindi of this period could be manipulated in ways that were possible in neither Sanskrit nor Persian. The grammar was not fixed, so words could be bent and shaped creatively. The literary manoeuvres of Rahim, although in a manner very different from those of Keshavdas, also seem to speak of a cultural rapprochement between the Mughals and their local Hindu subjects—but this time the flow moves in the other direction. As Keshavdas occasionally used Persian words in his later works for Mughal courtly scenes, Rahim embraced not only Indic lexical styles but also themes, and if anything, Rahim’s gestures are far more striking than those of his Hindu contemporary. There is no evidence in his entire corpus that Keshavdas knew anything meaningful about the Indo-Muslim world, its religion or larger cultural and intellectual practices.54 Rahim, however, seems to have been conversant with many aspects of Indian culture: a whole range of languages, Vaishnava bhakti,

Indian mythology, as well as diverse technical details about Sanskrit and Hindi literary systems. Rahim’s other collection of barvai is partly an experiment with the Indic barahmasa form; it also shows mastery of Krishnaite poetic conventions from the Braj tradition.55 These are poignant poems spoken in the voice of a gopi, who expresses her chagrin that Krishna has not returned in time for the monsoon. Rahim’s manipulation of register here shows both great sensitivity and skill. The text’s predominantly tadbhava style could not be more appropriate to the expression of feminine pain and longing.56 Sanskritised vocabulary is used sparingly, only for the opening invocations to Hindu deities.57 Persianised vocabulary, when it does occur in a handful of verses, is understated and seems largely instrumental to the task of creating end-rhyme.58 There are, however, four verses (not in the gopi’s voice) composed entirely in Persian, which express the absent Krishna’s love-sickness but in a more formal, masculine, and urbane register.59 These are a message delivered from Uddhava—didactic bore and perennial killjoy of Braj lore—and the Persian register seems perfectly calculated to heighten the poignancy of Krishna’s new preoccupation with city life in Mathura and his increasing distance from the lovelorn gopis.60 But it is overall the tadbhava simplicity that dominates in the poems, conjuring up a delicate blend of rusticity and pathos that bear testimony to Rahim’s sensitive handling of bhakti literary sensibilities. Rahim stretched himself culturally more than most, and this is evident everywhere in the poet’s multiregister virtuosity.

5. The writings of Raslin The language profile of the writings of Sayyad Ghulam Nabi ‘Raslin’ Bilgrami (1699–1750) is less varied than that of Rahim, but his facility with the linguistic and literary heritage of non-Islamicate India is no less striking. As suggested by his full name, Raslin hailed from Bilgram in what is now Uttar Pradesh, a famous centre of Indo- Muslim intellectual life.61 Like Rahim, Raslin was active in the Mughal army, but he is today mostly remembered for how he wielded his pen rather than his sword. Raslin wrote

only in Braj Bhasha. Given the educational setting of Bilgram it seems certain that this poet was trained in Arabic and Persian, so becoming a Braj Bhasha poet seems to have been a conscious choice. His principal works are the short Nakh-śikh Aṅgdarpan (Mirror of the body, 1737) and a substantial rītigranth entitled Rasprabodh (Understanding of sentiment, 1742); several dozen miscellaneous (mutafarriq or phuṭkal) verses are also attributed to him. Many aspects of Raslin’s poetry suggest that he carefully cultivated an Indianised aesthetic. His preferred takhallus ‘Raslīn’ (‘absorbed in sentiment’) as well as much of his imagery and style declare his orientation toward riti subjects.62 Even the distinctly Islamic opening to the Rasprabodh, with its verses in praise of Allah and Muhammad, is infused with Indic rather than Persianised terminology, as when Allah is hailed as ‘alakha anādi ananta nita pāvana prabhu karatāra’ (invisible, without beginning or end, eternal, purifying, lord and creator); or Muhammad is said to have bound mankind with a ‘satya dharma kī ḍorī’ (cord of the true moral code); or when the prophet’s goodness is said to be inexpressible by even the 1000 tongues of Sheshanaga.63 Raslin’s writings are the expressions of a pious Muslim, but one who was completely conversant with Indian literary motifs. In harmony with the themes of Raslin’s compositions is a distinct lexical style. In the two riti works the poet chose a simple tadbhava register, with only the occasional foray into tatsamas for either invocations to god (maṅgalācaraṇ) or technical vocabulary from Sanskrit literary theory. It is precisely this quality of purity in Raslin’s writing—it is about as close to unmarked either through Sanskritisation or Persianisation as one could get —that is so arresting. And it was a Muslim—not a Hindu—who wrote in this manner. Of course, if we have learned anything in our discussion of register thus far, purity is decidedly not a characteristic of Braj, which often appears to be congenitally impure, that is to say, hybrid and multiregistered. When compared to all the riti authors thus far discussed, Raslin’s vocabulary is by far the least Persianised, except perhaps for that of Keshavdas early in his career. For an eighteenth-century Mughal soldier with Raslin’s background, this complete lack of Persianisation must have been deliberate. It is as though his work is not so much un-Persianised as

de-Persianised, that is, actively avoiding Persian-derived forms. It is not clear what factors would have prompted Raslin to write his particular style of Braj Bhasha. Braj was not the Hindi dialect spoken in his region of Bilgram, so it was definitely a learned language for him—a language, which, alongside its literary tradition, he obviously took great care to master. Although he wrote in Braj rather than Avadhi, Raslin’s chosen style bears comparison with that of premākhyān authors like Manjhan and Jayasi (fl. 1540s)—similarly de-Persianised in lexicon.64 Perhaps he was inspired by the practices of these earlier Indo-Muslim authors, who had Indianised their Sufi materials in both lexical and thematic presentation.65 That Raslin did not avoid Persianised language in all his poems underscores the deliberation behind Rasprabodh and Aṅgdarpan. His mutafarriq verses show that he did mix his Braj with Perso-Arabic words on some occasions, as in the following hymn of praise to the twelve Imams: Ādi dai Alī puni Hasana kõ jasa suni, Jāhira Husaina guni jāne khās o āma ke, Puna Jain ābadīna Bākara mahāprabīna, Jāfara se haĩ amīna Kājima kalama ke, Alī Rajā ke samāna Takī Alī Nakī jāna, Akasarī tẽ bakhāna Mẽhadĩ tamāma ke Dūra kai sakala kāma dhyāna dhari āṭhõ jāma, Japata haũ sadā nāma dvādasa imāma ke First place is given to Ali, then hear of the fame of Hasan, And all people humble and noble know Hussein is clearly to be counted. Then there are Zainul Abedin, and Baqir—the greatly clever. The words of Jafar are trustworthy like those of Kazim, And know Taqi Ali and Naqi to be the equal of Ali Raza. They say that (the son of) Askari is the last: Mehdi. Putting aside all other matters and meditating day and night I constantly repeat the names of the twelve imams.66 The key point is that like Rahim, Raslin had the competence to write in both Persianised and non-Persianised registers, and when he wrote in the latter it was his choice to do so. It should also be stressed that this poet’s keen

interest in the subtleties of Braj Bhasha poetry was part of a larger literary trend. From the sixteenth century well into the eighteenth Braj Bhasha was a popular literary language that was cultivated by a range of cultural groups both Muslim and Hindu: from brahmin pandits to kings and courtiers (whether Mughal, Rajput or Dakkani). Only a couple generations before Raslin, the Mughal court intellectual Mirza Khan had written his Tuḥfatu’l Hind (Gift from India, c. 1675), a Braj grammar and glossary in which the language was ardently praised, and its literary principles expounded for precisely the type of Indo-Muslim poet and connoisseur embodied in the later figure of Raslin.67 Writers like Raslin and the corpus of riti literature more broadly are emblematic of an age when the cultural field of Hindi was far more fluid than it has become today. For whatever the factors to which we attribute the Hindi-Urdu divide, whether it was the Persianised style popularised in Delhi by the Dakkani poet Vali from the early eighteenth century, or later trends at Fort William College in Calcutta, or an evolving colonial and nationalist discourse about language and religious identities, or all of the above, in the case of even a relatively late riti poet like Raslin this divide was not on the horizon yet. CONCLUSION No monolithic understanding of language practices—particularly not one based on language as a marker of religious identity—can account for the rich and varied semantic terrain we find in a broad cross-section of riti textuality. The five case studies presented here provide a basis for identifying and theorising a range of lexical practices from a world not yet burdened with strict communitybased divisions along the lines of modern Hindi versus Urdu. The use of Sanskrit and/or Perso-Arabic words in Braj Bhasha seems to have conjured up various context-sensitive meanings. To be sure, not all practices can be explained with any strict coherence of logic. But this is probably for the best. Modern language ideologies, the product of a very specific world that has been deeply penetrated by colonialism and the cultural politics of nationalism, suffer from being too coherent, and perhaps we would do well to be suspicious of altogether clear-cut

explanatory models. These case studies prompt observations of a different order. We do not see strict correspondences between language styles and religious communities, but a close study of the texts yields suggestions, if not always bold directives, of how else we might interpret apparent trends. Some practices are familiar from the modern period, others not. Probing the less familiar ones is particularly necessary because it is outside our conceptual comfort zones—beyond our naturalised ways of thinking—that we stand the greatest chance of apprehending critical features of language practices in Indian pre-modernity. Sanskritised language was one major register available to riti poets, but it is not particularly ‘Hindu’ in its orientation. In the riti world a Sanskritised register was often chosen for scholarly writing, where it added a necessary complexity of expression that it would not have been easy to obtain using simpler tadbhava style. Highlighting the existence—and even prominence when it comes to scholarly genres—of Sanskritised language in riti texts serves as a useful corrective to the commonly held notion that Sanskritisation originates in the nineteenth century and is driven by divisive imperatives. Another place where preference for Sanskritised over Persianised language is seen is in the panegyric form, which tended to be written in a high kāvya style. What is interesting here, given modern language dichotomies, is that Indo-Muslim rulers and notables (Jahangir, Akbar Shah) could be portrayed according to a Sanskritising aesthetic in precisely the same manner as Hindu ones (Bir Singh Deo Bundela, Shivaji). Even as late as the seventeenth century, Sanskrit kāvya style maintained a hold over certain discourses of moral and political authority, regardless of the religion or cultural orientation of the ruler. Incorporating Perso-Arabic vocabulary into Braj poetry also needs to be seen in terms of a range of interpretive possibilities. At times there may not have been any special meaning to such usages, which is to say that Persianised language was chosen either for aesthetic or largely functional reasons. Regarding the former, the choice to use Perso-Arabic alongside Sanskrit and tadbhava registers was an attempt to fashion the most beautiful poem possible with the best ingredients from any language available. Riti

poets had an extensive lexical palette to choose from and a poet—perhaps like a painter selecting his colours—could range between languages and dialects—according to what best suited the context, or produced the most interesting literary effects. One of these literary effects was rhyme, which was largely unknown in the riti poets’ (predominantly) Sanskrit models. Mixed language is yet another profile, with many different permutations. Poets may switch from one register to another as they move between scenes or genres. Some patterns of mixing register and language may be seen as part of an aesthetics of rapprochement; but the same technique may also engender an aesthetics of reproach. Keshavdas’s use of Persianised language in his later works suggests a new spirit of cooperation between the Mughals and his regional kingdom of Orchha. Similarly, the hybridity and macaronic style we find in his contemporary Rahim illustrate a particular moment of cultural openness and experimentation. Hindi with its flexibility in registers and dialect forms was particularly suited to such experimentation. But hybridity may have harsher overtones, too: Bhushan’s hostility towards the Indo-Muslim political establishment finds expression in a trenchant multi-lingual style. Raslin’s is the least hybrid of the registers examined here. His tadbhava style is, ironically, somehow cultivated in its simplicity and the lack of Persianised language is an intentional silence in this writer’s voice. For Raslin’s register is not just tadbhava: in his riti works he actively eschews Persian-derived forms—a reminder that conscious experiments with dePersianisation and de-Sanskritisation long predate the modern period. In some sense Raslin’s style makes him—a Muslim—the ultimate Hindi poet. In sum, there is every indication that language register was manipulated with great sensitivity and in a range of contexts by the Hindi poets of early modern India. It may now be difficult for modern readers to retrieve the multiple nuances of such a diversity of language practices, but it is instructive to try to do so. The lexical orientations of particular authors were not concomitant to being a member of a given community, but a matter of careful choice. The choices were not the same as today’s choices, but they are choices that we would do well to pay attention to in any reconstruction of the Hindi linguistic and literary past.

4 Dialogism in a Medieval Genre The Case of the Avadhi Epics Thomas de Bruijn

This article develops some points that I raised in my paper at the EASAS Conference in Lund, Sweden, in July 2004. Some examples of the comparison between Jayasi’s Padmāvat and Tulsidas’s Rāmcaritmānas have been published earlier in a different context: the exchange of religious symbols between traditions. See de Bruijn 2005. DIVISIONS AND DIALOGUES

T

he notion of a profound and unbridgeable division between [indus and Muslims in modern Indian culture that surfaced at the beginning of the twentieth century imposes an antagonism between cultural traditions that may not have been there in the same form in earlier periods. This imposition calls for a careful reexamination of pre-modern cultural forms and the divisions that created them but have been obscured by projecting modern political divisions on the medieval situation. The present essay will investigate some general characteristics of premodern cultural categories and divisions before focusing on the composite genre of medieval Avadhi epics. This literary genre presents us with an example of a cultural form that was developed by Indian Sufi poets, but in which a Bhakti poet like Tulsidas (ca. 1532–1623) found himself very much at home when writing his Rāmcaritmānas (ca. 1574). The presence of the

image of Rama in Sufi Avadhi poetry before Tulsidas, especially in the Padmāvat (947H/ 1540–41) of Muhammad Jayasi (d. 949H/1542?), raises the question whether this can be seen as a possible intertext for Tulsidas’s epic. If so, this invites further examination of the role of the Avadhi epics as a ‘dialogic’ genre that encompassed different religious contexts, mainly through its role in the competition of both Sufi and Bhakti institutions for patronage and political prestige. Pre-modern religious poetry in the Indian vernaculars is divided in various corpuses made up of texts which have some common stylistics, literary form or religious orientation. A prominent division seems to be between the shorter, strophic poems and the long narrative poems. There is also a division on the basis of language that is not directly related to the local origins of the poets and is, to some extent, independent of the religious orientation but seems surprisingly consistent with the formal divisions: shorter strophic poems seem primarily the domain of Braj Bhasha and a set of other vernaculars, excluding languages such as Avadhi. There is also the Vaishnava-orientation of the Braj Bhasha poems, separating them from the short poems of the Sants and other radical Nirgun movements of popular religion. These do not have a preferred linguistic domain—Sant poetry covers a wide range of languages, as the spread of the works of Kabir show —and seamlessly spill over in the songs of the Bengali Bauls or other regional traditions of singing and wandering mystics. The longer poem has a domain of its own with the epic tradition of Rajput and Dingal bards, the epics in Avadhi and a range of folktales in almost every language of the subcontinent. It has a pedigree in the earlier strata of the archaeology of Indian writing such as Apabhramsha. The two domains are not completely separated: the longer poem can include shorter forms such as the doha, gitikachand or soratha that also have a separate life as moralistic short strophic poems. This simplified overview gets more complicated as one zooms in on specific traditions, such as the Avadhi epics by the Sufi poets that came about around the fourteenth century. Not only are these epics part of a composite culture, where Indian Sufi poets converted local stories into epics of mystical love, they also cross borders in the way they spread in the

literary field, with manuscripts in Persian Nastaliq, Devanagari and Kaithi scripts, in large illustrated copies and plain textbooks. Retracing the trajectory of a pre-modern text through the collections of Indian manuscripts often involves crossing the borders between various titles and headings under which the same material is catalogued, dividing for example ‘Sanskrit’ manuscripts from ‘Hindustani’ and ‘Persian’ sources. Scholarship on medieval poetry has also laid bare the relative value of classifications of texts and poets into neat categories of religious background or cultural outlook. The painstaking work on the Sūrsāgar by Hawley and Bryant1 has been groundbreaking in correcting the sectarian view on Surdas and his works. This revision has also unleashed a large number of problems of attribution, as the early collections contain various poets besides Surdas and blur the distinction between saguna and nirguna poetry. Strict text-based divisions on the basis of genre, language and form certainly do not exhaust the number of possible categories within premodern traditions: social structures add divisions such as caste, gender, local versus universal, that seem to be very productive in placing poets, texts, patrons and readers into specific positions in the field of cultural production. One can even speculate on divisions that have a much longer durée than one particular period and are able to connect seemingly unrelated cultural formations in South Asia. Oppositions such as itinerant versus settled, grāma versus āraṇya or transcendent versus worldly have a powerful hold on the structures of cultural production in various communities. For the present essay, the division between premodernity and early modernity seems a relevant dividing line, separating a dialogic process of cultural formation from the monologic fixation of religious and cultural canons and the political exploitation of identity. One of the most intriguing aspects of pre-modern Indian writing is that, despite the many hybridities, there is no unifying force that joins the different discourses into one single form of expression. It is even a point of stark contrast with the literature after the divide: the modern Hindi novel has stylistically more in common with the modern Urdu novel than Mirabai has with Tulsidas. Despite the construction of categories and other

trajectories of unification in the field there is a point at which division provided better chances of survival than unification, as is attested by the divisions we see in pre-modern writing. One possible reason for this is that all creative artists were cognisant of, or actively involved in, various discourses.2 It may be more productive, then, to look for a logic of the aesthetic and practical paradigms in the medieval literary field that allows it, to cross divisions of genre and language in contrast with what the textbook histories of Indian writing want us to believe. Pre-modern poetic discourse in Indian vernaculars developed a vocabulary and an idiom that was surprisingly homogenous and flexible enough to accommodate different doctrinal or religious paradigms. This idiom could be taken to various positions in the field and used to express meanings that were specifically relevant to a particular position or identity.3 In this structure, genre was not primarily defined by a monologic and exclusive creative identity, but was rather a container that encompassed a paradigm of poetical aesthetics such as forms, metres and language, and a practice of reception and transmission which ensured the spiritual authority of the poet and his support by patrons. The latter could take the form of an association with a religious institution. Not a closed and monolithic entity in this structure, genre allowed for, and was often constructed as, a dialogic exchange with other genres, to which it referred in its use of images or which it completely encorporated, as was the case with the gnomic doha that became part of a narrative genre (the premakhyans). Moreover, the distribution of genres in the literary field referred not so much to completely distinct cultural forms but rather to different positions within a single but composite field. These positions represent different points in a complex matrix of differences that could be based on religious affiliation, clientship of particular patrons, relative distance to other traditions, caste or political grouping. This matrix was never fixed as new positions could develop under new circumstances that called for the extension of existing genres or the definition of new boundaries or divisions.

The pre-modern literary genre distinguishes itself from the modern counterpart by its dynamic, dialogic exchange between various positions and by the constant re-definition of difference and convergence between these points. This genre is dialogic, or ‘intermediary’ by nature,4 unlike the monologic fixation of genres in the modern, divided north Indian literary field (e.g. Urdu ghazal and Braj Bhasha kabitt) which has led to a highly polarised development of literary difference. In the construction of a literary history, this divided field has projected its categories back on to the medieval period, creating artificial divisions that are incongruous to the logic that steered distinction and differentiation in the premodern genres. THE AVADHI EPICS This theoretical model can be put into practice by taking the Avadhi epics as an example and examining the details of the cultural dialogue that characterised this literary form. The tradition of long narrative poems in the Avadhi dialect was mainly an extension of earlier forms of narrative poems in Apabhramsha by Jain poets. They combined a narrative couplet (in the paddhariya metre) with a short gnomic poem (in the dhatta metre) to intersperse narration with a specific doctrinal explanations of the story. The Avadhi texts use the chaupai metre for the narrative part and conclude each couplet of 7 or more lines in this metre with a didactic verse in the form of a doha or soratha. These concluding verses show a remarkable resemblance to the short strophic poetry of the Nath yogis and the Sant poets.5 The main impetus for the development of the format in Avadhi came from Sufi poets, who used it to frame Indian folktales or heroic stories into mystical ‘romances’ that glorify the sacrifice for mystical love. The earliest example of this genre is Maulana Daud’s Candāyan of 1379 AD. Later, sixteenth century poets such as Shah Manjhan and Muhammad Jayasi brought this literary genre format to perfection and expanded its semantic possibilities. Aditya Behl has described the formation of this genre as an instance of the ‘translation between Persian and Hindavi’ by Indian Sufi poets and scholars.6 A correction to this characterisation should be made with regard

to the various voices in the Avadhi/Hindavi genre. Manjhan’s work is in many more ways a ‘translation from Persian’ than Jayasi’s poems, which show evidence of being more deeply inserted in Indian cultural forms and practices. As Behl rightly remarks, there is no evidence of a similar use of the Avadhi/Hindavi genre by ‘Hindu’ authors before Tulsidas composed his Rāmcaritmānas in this format. It is not often that the great ‘Hindu’ epic by Tulsidas, which has become such a dominating presence in modern Hinduism, is seen as belonging to a literary format that had a much more composite nature than the modern paradigm of cultural and religious division perceives. From the perspective of a medieval poet, however, the distinction between a ‘Hindu’ and a ‘Muslim’ epic in Avadhi was probably not as evident as it is from a modern point of view, as a comparative analysis of Jayasi’s Padmāvat and Tulsidas’s Rāmcaritmānas shows. On the one hand, as we shall see, the exemplary kingship of Rama was a well-developed theme in the Sufi poem, and on the other hand Tulsidas’s epic is much more deeply rooted in the semantic potential of the genre as was created by the Sufi poets and in its position in the world of patronage than a reading from a ‘divided’ perspective is likely to bring out. Muhammad Jayasi’s Padmāvat is based on the historical tale of the sack of the Rajput stronghold of Chitor by Sultan Alauddin Khilji in the fourteenth century—not the most obvious example of a ‘Muslim’ poem. The poet leaves no doubt as to the nature of his beliefs in the introductory verses to the poem, praising Allah, the Prophet and his spiritual teachers who are Sufi Pirs connected to a local branch of the Chishti lineage. Yet, in the story, the cultural roles are reversed: the Rajput king Ratansen is the hero and the Muslim Sultan, who lays siege to the Rajput fortress in order to obtain the beautiful princess Padmavati, is the villain and fails to lay hands on her as she commits sati on her husband’s funeral pyre. The conflict between Ratansen and Alauddin is not one between Hindu and Muslim but rather between the righteous ruler, who has achieved spiritual wisdom by means of his yoga of love for Padmavati, and the brutal worldly ruler who is guided by lust and who, by demanding the wife of his Rajput vassal, breaks the code of conduct of the feudal overlord. The Sultan’s worldliness is set against the ‘immortality’ Ratansen has acquired through his sacrifice and his union with the divine

beauty of Padmavati. The first part of the epic is therefore a mirror-image of the battle with the Sultan as it shows the Rajput’s quest for Padmavati who is the daughter of the king of Simhala, and which involves a difficult voyage and Ratansen’s complete sacrifice for love. Jayasi gives his tale of divine love a worldly and wider semantic dimension by connecting it, through numerous intertextual allusions, to other love-stories of the Persian and Indian tradition. His references are not so much the courtly Persian tradition of the masnavi but rather to the dastans of Sikandar (Alexander the Great) and Amir Hamza and other stories from popular traditions.7 This insertion of the poem in a local, Indian context is what distinguishes Jayasi’s works from earlier premakhyanas such as Manjhan’s Madhumālatī which remain much more in the domain of a Sufi allegory in an Indian language. Jayasi’s Chishti pedigree seems to have placed him in closer contact with the sensibilities of Indian devotional poetry. This means that his poems seem to be intended to make sense in the composite setting of sixteenth-century north India, combining an appeal to the devotees, both Hindus and Muslims, of the Sufi shrines as well as the ‘inner circle’ of murids.8 One of the elements that give his Padmāvat a firm Indian grounding is the structural parallel between Ratansen’s voyage to Simhala and Ram’s voyage to Lanka. The intertextual connection between the tale of Padmavati and the Rama story is far from unintentional: it frames the Sufi poem in a cultural position that is recognisable for an Indian audience and immediately gives the story a thematic coloration. The intertextual link with the Rama story works both ways: Jayasi compares scenes and events from the Rama story with elements of his Padmāvat but also imposes his own ideology on these elements. In the same way, the tale of Ratansen and Padmavati ‘borrows’ from the semantics of Rama’s tale and expands its relevance in its cultural context. This intertextual style characterises Jayasi’s works and has a very significant parallel in his retelling of the Krishna material from the Bhāgavata Purāṇa in the Kanhāvat, a work that he wrote in the same year as the Padmāvat. The link with the Rama story makes very obvious sense if one regards the position of the Padmāvat and other premakhyans in the exchange

between the Sufi dargahs and local landowners or nobles as patrons of the religious institutions. The poem presents an alternative to authority based on brute force and warfare and proposes a legitimation for worldly rule based on a paradigm of mystical enlightenment. This ideal of kingship, which was already present in the Persian tradition in the image of Sikandar/Alexander, Jayasi now brings to the Indian context. The theme had a definite relevance in the context of north Indian politics in a period in which the traditional system of shifting alliances and internal conflict was being replaced by the imperial hierarchy that was fully deployed by Akbar. The relation of a Pir and his pupils as a metaphor for that between the emperor and his vassals was in fact later reified in Akbar’s political concept of the dīn-i ilāhī.9 By reframing Ram in this context, Jayasi brought him much closer to the frame of reference that was prevalent in the Indian political scene of the period. The references to the Rama-story in the Padmāvat take different shapes. The simplest form is the use of śleṣa (double entendre) on Avadhi words such as ‘rāma’ and ‘rāvaṇa’, which can double for ‘rāmā (wife), ‘rāu’ (king), and ‘ramana’ (lover); ‘laṅka’ (waist or Simhala); ‘lakhana’ (Lakshman, or jewellry, lakṣaṇa). As the following examples show, Jayasi uses the semantic possibilities this figure offers by joining the erotic, the mystical and the heroic aspect of Ratansen’s quest for Padmavati.10 In the first example, Padmavati returns from the temple of Shiva, where she has performed the puja of spring, and meets Ratansen who faints at the sight of her. She goes to sleep and has a dream in which she sees the meeting of sun and moon, predicting her future marriage with Ratansen. On waking up, she tells her sakhis of her vision and asks them to explain its meaning: Dina au rāti jānu bhae ekā, Rāma āi Rāvana ga£h chemkā Tasa kichu kahā na jāi nikhedā, Arajuna bāna Rāhu gā bedhā Doha: Janahũ laṅkā saba lūsī hanūṃ bidhāṃsī bāri. Jāgi-uthiũ asa dekhata sakhi so kahahu bicāri It was as if day and night had merged, Rama (or the woman) had come and laid siege to the fortress of Ravana (or the lover).

There is no way to describe their duel, it was as when Arjuna pierced Rahu with his arrow. Doha: It was as if the whole of Lanka (or the waist) was destroyed; Just when I saw this I woke up; please friend, explain to me what you think of it. (Padmāvat 197.6-doha) When describing the wedding of Padmavati and Ratansen, the poet remarks: Hulasi laṅka ki Rāvana rājū, Rāma Lakhana dara sājahī̃ sājū. Lanka (the waist) was thrilled with Ravana’s rule (a good lover); Rama and Lakshman equipped their armies (a group of beautiful women prepared her attire). (Padmāvat 280.5) After the marriage Padmavati withdraws to be dressed up for the first night with her husband. Ratansen is overcome by the pain of separation and wakes up when he hears about the coming of his guru ‘Gorakha’: Gorakha sabada suddha bhā rājā, Rāmā suni Rāvana hoi gājā. The king regained his consciousness when he heard the word ‘Gorakha’; He roared like Ravana (the lover) when he heard about Rama (the woman). (Padmāvat 304.1) After the consummation of the marriage, the poet assesses the damage: Kahaũ jujhi jasa Rāvana Rāmā, seja bidhaṃsi biraha saṅgrāmā; Līnha laṅka kaṅkana gaṛha ṭūṭā, kīnha singāra ahā saba lūṭā. I will tell you, it was like the battle between Ravana (the lover) and Rama (the woman); the bed was destroyed in conquering separation.

He took Lanka (her waist) and destroyed the golden fortress; he robbed all the ornaments that adorned it. (Padmāvat 318, 1–2) After the marriage, Ratansen and Padmavati remain in Simhala for a year, enjoying all aspects of their love. Padmavati seduces Ratansen into staying and making love to her, instead of going back to his first wife, Nagamati, whom he left behind in Chitor. Kālhi na hoi rahe saha Rāmā, āju karau Rāvana saṅgrāmā. Now it is not like yesterday, when you were with your wife (when you were Rama); fight the battle of love as my lover today (fight with Ravana). (Padmāvat 333.5) Another level of intertextual reference is to be found in the structural overlap between the scenes in the Padmāvat that have a parallel in the Rama story such as, for example, the crossing of the oceans, the fight with the demon-king (the Sultan), and the return to Ayodhya (Chitor). A very direct and interesting use of the intertext of the Rama story is the appearance of characters from the latter in the Padmāvat. This happens when Ratansen is in Simhala and threatens to immolate himself in the fire of love. Hanuman is also present, watching from a nearby mountain, and warns Mahesha (Shiva) and Parvati to rescue the world from this allencompassing fire. The divine couple appear and Parvati proceeds to test Ratansen’s devotion to Padmavati by showing herself in the form of a beautiful apsara. But even this seductive display cannot divert the king from his sacrifice for Padmavati. In order to prevent it, Mahesha points him to a secret passage into the fortress of Lanka by which he may reach his beloved (Padmāvat 207–216). The motif of the test by Parvati (Sati) corresponds to a test of Rama’s divinity which is not part of Valmiki’s Rāmāyaṇa but figures in Tulsidas’s Rāmcaritmānas (I, 52–53): Sati wants to perceive Rama’s divine status and appears to him in the form of Sita; Rama immediately sees through her disguise, confirming his divine nature.11

Intertextual links with the Rama-story pervade the Padmāvat and prove that the theme of the deeds of Rama was already present to the writers of Avadhi epics before the composition of Tulsidas’s Rāmcaritmānas. The theme of righteous kingship and the value of abiding by moral codes of conduct in order to uphold the order of the world are primary elements in Jayasi’s message to his patrons. This message places the religious charisma of the Sufi Pir in a central position as a source of legitimacy for worldly rule. The poet mediates this exchange of symbolic capital that can also be seen as a very real practice of patronage for the Muslim and Hindu religious institutions. The link between this political context and the symbolism of the Padmāvat is expressed in a verse that places the theme of pride as a danger for righteous kingship in the context of Rama’s tale. Rāvana garaba birodhā Rāmū, au ohiṃ garaba bhaeu saṅgrāmū. Tehi Rāvana asa ko varibaṃḍā, jehi dasa sīsa bīsa bhuaḍaṃḍā. Sūraja jehi kai tapai rasoī, baisandar niti dhotī dhoī. Sūka soṃṭiyā sasi masiārā, pavana karai niti bāra buhārā. Mīcu lāi kai pāṭī bāṃdhā, rahā na dosara ohi sau kāṃdhā. Jo asa bajara ṭarai nahiṃ ṭārā, sau muā tapasī kara mārā. Nāti pūta koṭi dasa ahā, rovana hāra na ekau rahā. Doha: Ocha jāni kai kāhūṃ jani koī garaba karei Oce pārai daiya hai jīt patra jo dei. Ravana opposed Rama because of his pride, and from this pride the fighting started. Was there ever one as strong as that Ravana, who had ten heads and twenty arms? The sun cooked his dinner, fire used to wash his loincloth. Shuka was his staff-bearer, the moon his torch, the winds used to sweep his gate clean. Death he made into his bedpost, there was no one who was up to his stature.

But even he, who was so powerful that he could not be shaken, was killed by ascetics. He had a hundred million sons and grandsons, but not one of them remained to mourn him. Doha: Let no one look down upon others out of pride, As it is God who protects the weak and decides who will prevail. (Padmāvat 266) TULSIDAS’S RĀMCARITMĀNAS AND THE AVADHI EPICS Although Tulsidas’s text is usually discussed in relation to its ‘roots’ in the tradition of retellings of the Rama-story in sectarian Sanskrit texts and the classical epic of the Rāmāyana, the Rāmacaritmānas uses the language and literary form of the Avadhi epic. Both Charlotte Vaudeville and Philip Lutgendorf note this fact and mention the Padmāvat by Muhammad Jayasi as a prime example of this genre, but neither of them elaborate on possible intertextual exchanges between these texts which, after all, belong to the same literary genre. Their focus is more on a diachronic archaeology of Tulsidas’s concept of Rama bhakti in Indian sources or the context of performance.12 Lutgendorf does mention how manuscript traditions show that the epic of Tulsidas followed a similar pattern of transmission to that of the Padmāvat: manuscripts are plentiful in both Devanagari and Persian scripts and testify to the diffusion of both texts in circles of merchants and local nobility. While it is important for a good understanding of Tulsidas’s text to place him in a tradition of sectarian retellings of the story of Rama in Sanskrit that expressed a particular theological doctrine through that famous tale, the choice of the literary format of the Avadhi epic is nonetheless significant. It adds a synchronic, horizontal dimension to his retelling of the Rama story that incorporates the connotations that prevailed in this genre. Other works by Tulsidas such as the Kavitāvalī testify to the fact that his poetical activities were wider in scope than the Rāmcaritmānas alone.13 He also

wrote in Braj Bhasha and seems to have been a ‘professional’ religious poet who shared a lot of the cultural habitus14 that went with that status with others in the same field, including the Sufi poets. This habitus was shaped by the patronage extended by religious institutions, which gave a new, contextualised meaning to descriptions or glorifications of kingship. This could go hand in hand with a representation of the Rama story with all the theological implications that Tulsidas’s text had, professing a unified devotion to the deeds of Rama as a means to cross the differences between sectarian, saguna and nirguna visions of Rama. The genre of the Avadhi epics had become imbued with this double role, giving the religious poet the role of intermediary, telling the specific meaning of his story, as do the many narrators in the Rāmcaritmānas. In this genre, a special role had been taken by the mystical idiom of the Nath yogis, which the Sufi poets exploited to describe the path to religious liberation and from which Tulsidas took the concept of the name of Rama as the main object of devotion. These elements bring the Rāmcaritmānas closer to its contemporary and contextual background through the form of the genre of the Avadhi epics. This genre provided only a matrix of semantic paradigms and a position in the local literary field that could accommodate messages of various religious connotations. The genre itself and the matrix it provided were not bound to a religious identity. If one accepts the model of the Avadhi epic as a dialogical matrix, earlier interpretations based on a strict identity of the genre seem less convincing. I am thinking here of Paul Hacker’s concept of ‘Inklusivismus’, which he attributed to Tulsidas’s text as a stratagem to persuade followers of other doctrines.15 Or of the view of the Rāmcaritmānas as a final rallying call to the forces of Hinduism against the rise of Muslim power, cited by Charlotte Vaudeville and at the heart of present-day Hindu nationalist reconstructions of the literary context of the medieval work. Both interpretations seem to ignore the dialogical qualities of the genre.16 In order to see the semantic matrix at work it is worth looking closely at some examples from the Rāmcaritmānas which show how this text was related to its contemporary literary background. There are more than a few

instances in the Rāmcaritmānas showing that Tulsidas was at least aware of the Sufi paradigm. The most well known locus is the verse in which he includes the words garībnewāj and sāhab in his introduction. He restores what is lost, he is a refuge for the poor; Raghuraja is a righteous and powerful lord. Wise men recognise this and praise his glory; thus they make their own speech fruitful and blessed. (Rāmcaritmānas I, 13.4) The word sāhab reoccurs several times in the text to indicate the divine. It reflects an almost colloquial manner of speaking and is not developed in a theological sense by Tulsidas. The use of the term garībnewāj (gharībnawāz) is more significant: it is the sobriquet of the founder of the Chishti saint Muinuddin Chishti, whose tomb in Ajmer is a major site of pilgrimage and worship. It is also its position in the text that attracts our attention. In the Avadhi Sufi epics the introduction is a place where the poet establishes the thematic and doctrinal colour of the work. It is here that Sufi poets praised their spiritual guides, their patrons, Allah and the Prophet and the king of the day. As mentioned before, the concept of the ‘name of Rama’ is a central tenet in Tulsidas’s transformation of the story and of its religious meaning. The poetic idiom that describes this nirguna concept was developed in medieval poetry in the corpus of Sant poetry such as the poems ascribed to Kabir, Raidas, Nanak and others. The basis for this idiom was the tantric imagery of the Nath yogis. The Sufi poets who used Indian vernaculars always made important use of this material in their works. They used this idiom to rephrase the Persian and Arabic terminology of Sufi mysticism into the local languages. In the Padmāvat, for example, Nath imagery is used to describe the stages the yogi of love (Ratansen) must cross on his way to Simhala, i.e. the gradual destruction of the body and detachment from worldly attachments in the yogi’s sacrifice for love. The introductions to the Sufi works also allude to the supreme power and status of the name by associating it with the devotion to Allah and his Prophet.

Tulsidas used the same Nath yogi concept and the poetic idiom that went with it to frame his notion of the name of Rama. His introduction to the Rāmcaritmānas expresses this most explicitly in verses that insist on paradoxical nature of Rama: his name escapes description or praise, but its presence and power pervade everything. Compare these images with the Sufi’s praises of Allah and Muhammad in the opening parts of their premakhyans. The form and nature of the name is an ineffable story; it gives happiness to those who understand it, but it cannot be told. His Name speaks for both the attribute-less (God) and for the one (God) who has attributes; it is a clever interpreter explaining the nature of both. (I, 21.4) That Creator is without marks, without outer form, without colour; (yet) he is engaged with everyone, everyone is engaged with him. He pervades the universe, whether He is hidden or obvious; the righteous one perceives him, the sinner does not recognise him. (Padmāvat 7.1–2) God made him (the Prophet) his interpreter in the world; the two worlds were saved when they praised his name. For him who does not praise his name, He made a place in hell. (Padmāvat 11.6–7) The introduction to the Rāmcaritmānas is a complex and significant part of the text that is vital to an understanding of the theological tour de force its author performs by unifying a wide array of sectarian visions on Rama. One of the central pieces of his introduction is a description of the Manasa lake, the place where the devotee of Rama comes to find the highest religious experience. The ghāṭs of the lake are the dialogues through which the most part of the story is told and represent the various doctrines that merge into the tale of the deeds of Rama. Compare it to the image of the Manasa lake in Jayasi’s Padmāvat that is present in more than one scene. Here the image has a strong thematic function as it refers to the yogic paradigm of purification and mystical achievement, where the beauty of the

lake and its nectar-like water (amṛta) symbolise the attainment of the highest spiritual enlightenment. In the Padmāvat, the descriptions of the lake of Chitor and of Simhala are shot through with this symbolism. How wonderful the Manasa lake looks; it is boundless like the ocean and just as deep. Its water is spotless like pearls; it has the colour of amṛta and the sweet smell of camphor. From the island of Lanka slabs were brought and these were bound together as its ghāṭs. On every side steps are made; on all four sides people go up and down. It is red with flowering lotuses; a shelter has been made for thousands of birds. Oysters are lying open, the pearls have come out; geese pick these to adorn themselves. Most beautiful silver birds swim there; it is as if they are statues made out of gold. Doha: On the dikes on all four sides are trees which all have fruits of nectar; If one observes the beauty of this lake, all thirst and hunger disappear. (Padmāvat 31) Jayasi uses the same image again when, after an arduous journey and the dangerous crossing of the seven seas which divide Simhala from the mainland, Ratansen and his followers see the Manasa sea: They came to the seventh sea, the Manasara; he who had been courageous in his truthfulness (sata) obtained a thousand siddhis by reaching it. Seeing the pleasing beauty of the Manasara, the excitement in their hearts unfolded like a lotus. The darkness had gone away and the ink of the night had disappeared; the dawn broke and the rays of the sun broke through.

All the companions (of the king) called out: ‘Astu, astu’ (It is he, it is he!), God has opened the eyes of us who were blind! As the lotus flowers blossomed there, their bodies also began to smile; as bees, their eyes started to enjoy its flavour. The hamsa birds smiled and enjoyed themselves; they picked up the gems, pearls, and diamonds. When one comes here, having performed such penance and yoga (as Ratansen did), one’s hope is fulfilled and one can enjoy the pleasure of rasa. Doha: The bee that has his mind fixed on the Manasara comes to it and enjoys the rasa of the lotus; The woodworm that does not have this courage eats dry wood. (Padmāvat 158) The images that are used in this description mix poetical conventions with the thematic connotations of the mystical poem. This expansion of the conventional image of the Manasa lake is also present in Tulsidas’s image of the lake of the deeds of Rama. Here is an example of his description in the translation by F.S. Growse: Doha: This pure and lovely lake has four beautiful ghāts, viz. the four charming dialogues contrived by divine wisdom. Chaupai 37 The seven books are its beautiful flights of steps, which the eyes of wisdom delight to look upon: the unqualified and unsullied greatness of Raghupati may be described as its clear and deep expanse; the glory of Rama and Sita as its ambrosial flood; the similes are its pretty wavelets; the chaupais as its beautiful lotus leaves thickclustering; the elegance of expression as lovely mother-ofpearl; chhands, sorathas, and dohas as many-coloured lotus flowers: the incomparable sense, sentiment, and language as the pollen, filaments and fragrance of the lotus; the exalted action as beautiful swarms of

bees; the sage moral reflections as swans; the rhythm, involutions, and other poetical devices as diverse graceful kinds of fish; the precepts regarding the four ends of life, the wise sayings, the thoughtful judgments, the nine sentiments (or rasas), the prayers, penance, abstraction and asceticism, of which examples are given, are, all beautiful living creatures in the lake; eulogies on the faithful, the saints and the holy name are like flocks of water-birds; the religious audience are like circling mango groves, and their faith like the spring season; the expositions of all the phases of devotion and of tenderness and generosity are like the trees and canopying creepers; self-denial, morality and holy vows are their flowers, and wisdom their fruit; the love for Hari’s feet as the sound of the Vedas: and all other stories and episodes as the parrots and cuckoos and many kinds of birds. Doha 37 The hearer’s emotion is some grove, garden or parterre, where sportive birds symbolise his delight and Piety the gardener pours a stream of devotion from (the water-pot of) his beauteous eyes.17 The description continues in verse 38 (my translation): Those who recite these deeds with care, they are the guardians of this lake; those men and women who hear it constantly with reverence are the high gods who care for this lake. Those who are vile and lustful, they are like the poor cranes and the crows who do not come near the lake; because there are no snails, frogs or scum here in the form of rich and juicy stories. That is the reason why lustful people do not have the heart to come here, like the poor crows and cranes; it is very difficult to reach this lake and one can only get here with the grace of Rama. Bad companions are a hard, difficult and crooked road; their words are like lions, tigers and snakes (on the road); the entanglements of domestic chores are like large and insurmountable mountains.

Pride, infatuation and arrogance are like the dense woods; prejudices are like so many frightening rivers. Doha: For those who are without the provision of faith and do not have the company of good people, the Manasa lake is out of reach as they are not in the favour of Raghunatha (Rama). The image of the lake as a source of amṛta reappears in a remarkable way in Rāmcaritmānas: when Rama has finally beaten Ravana, the demon sees the divine identity of his killer and converts into a bhakta of Rama. It is therefore a truly significant moment in the epic. On this occasion Tulsidas narrates how Rama was told by Vibhishana that there was a source of nectar in the navel of Ravana which made him invincible. As Rama prepares for the final battle suddenly all sorts of omens appear: darkness falls and both sun and moon are invisible (Rāmcaritmānas V, 103). At this moment Rama hits Ravana’s navel and is able to defeat him. Similarly, in the Padmāvat, the goal of Ratansen’s journey is the amṛta-pool of Simhala, while the eclipse of sun and moon is a yogic image of spiritual enlightenment that Jayasi refers to many times. CONCLUSION The elements from the Padmāvat and the Rāmcaritmānas presented here for comparison are but a small sample of the instances that invite further research into the contextual position of the latter work. These examples indicate that, by choosing the format of the Avadhi epics, Tulsidas inserted his text in an intertextual framework (i.e. a genre) that provided a model for narrating a religious epic and presenting it in a form that was respected and patronised in a local context. The examples also warrant a reading of the genre of the premakhyan as a much more open and dialogical format than one could infer by looking only at the Sufi texts without comparing them with the Rāmcaritmānas. Our investigation of the image of Rama in the Padmāvat has highlighted how the Sufi poets connected the typical theme of the premakhyans—spiritual authority as a supreme source of power over brute force—to the classical Indian hero. This connection made the genre

an ideal format for presenting the universalised figure of Rama that Tulsidas wanted to put forward. There is little to be gained by linking the Padmāvat and the Rāmcaritmānas through some kind of religious ‘syncretism’. Jayasi used the image of Rama to bring his concept of enlightened kingship closer to the Indian audience, and the genre allowed him to include a very wide array of images and even characters from Indian traditions without stepping outside the literary format and thus blurring his rhetorical position. It is the dialogical nature of the genre that allows him to create this polyphony. The same goes for Tulsidas, who moved his version of the Rama story from the universal Sanskrit tradition to a localised discourse and used the specific aesthetic paradigms of the genre and its position in the social and political arena, where it functioned in the dialogue between worldly patrons and religious institutions. Tulsidas’s choice meant affiliating his text to the genre historically developed by the Sufi poets. This process of affiliation can be observed throughout the Rāmcaritmānas, not in the form of direct word-for-word concordances or extended borrowings but in the use of a common dialogical semantics. There is no doubt that the Rāmcaritmānas was directed at a different audience from that of the Avadhi Sufi epics and made a theological case that is incomparable to Jayasi’s vantage point. This intertextual interaction creates a fine balance between extending the meaning of a work by incorporating elements from other traditions, and delimiting and focusing the meaning and scope of a text for a specific position in the field. It allows texts of different religious outlooks to coexist in a dialogic manner without the need of a syncretic doctrine. Genre offers a matrix of positions that combine convergence and difference, in line with a social environment. The most tangible evidence for this mode of operation is provided by the manuscripts of the Sufi texts and of the Rāmcaritmānas, which followed the same trajectories of reproduction and collection. The nexus between genre, social structures and content created dialogical divisions that made sense in pre-modern Indian literature, but are at odds with the monologic divide that modernity placed between Hindu and Muslim culture.

5 Barahmasas in Hindi and Urdu Francesca Orsini

T

he aim of this collection of essays is to move away from the received historical narratives of Hindi and Urdu, each with its own teleology, and look afresh at the textual material available in order to attempt a picture of north Indian literary culture that is at the same time more complex and layered and also more attuned to the nuances of register, accent, language choice, genre and audiences. As a start, we thought that it would be useful to concentrate on ‘intermediary genres’, i.e. those genres that fall in between the high literary tradition of Hindi, riti poetry in Braj Bhasha, and the high literary tradition of Urdu, centred mainly on the ghazal. As is the case with minor poets or works, the value and usefulness of these intermediary genres do not lie necessarily in themselves, but in how they force us to consider a broader range of texts and tastes than those envisaged by the canon and thereby reconsider the possible options and choices north Indian poets had before them. Why did a Sant write in Nagari Rekhta? Why did a Persian poet or an Avadhi Sufi mix Hindavi and Persian? Why did a sophisticated Rekhta poet like Uzlat write a barahmasa? Whatever their specific motivations, all these cases speak of an awareness of multiple literary models and a keenness to experiment with other literary or oral traditions that go against the purist intentions of modern literary historians. In this essay I will consider one particular ‘intermediary genre’, the barahmasa or poem/song of the twelve months. In this kind of poem an

abandoned woman (virahinī or birahinī) pines for her absent lover or husband and describes her pitiful state, month after month, against the backdrop of seasonal changes and ritual events. Today, barahmasas are perceived only as a kind of folk song, but a significant literary tradition attests to the attractiveness of this template for poets in all the literary languages of north India for centuries. In fact barahmasas are found at the beginning of literary writing in several neo-Indo Aryan languages and the genre is characterised by remarkable ubiquity and flexibility.1 As such, barahmasas represent the kind of ‘open’ or ‘dialogic’ genre that Thomas de Bruijn theorises in his essay in this volume. The female voice, her pining for and loyalty to, her absent lover and the calendrical cycle are the core elements of a matrix that poets from a wide variety of cultural backgrounds took up and modulated according to their specific taste and inclination. In the first part of this essay I will consider a range of these early vernacular poems before focusing on barahmasas in Urdu in the second part. Finally, I will consider briefly their transition into print in the 1860s, when the same barahmasas were printed in Devanagari and the Urdu script, and offer an example of popular literary taste that survived the Hindi-Urdu divide. A FLEXIBLE GENRE Within the multilingual and multiform literary field of medieval north India, Thomas de Bruijn argues in his contribution to this volume, genre ‘was not primarily defined by a monologic and exclusive creative identity, but was rather a container that encompassed a paradigm of poetical aesthetics such as forms, metres and language, and a practice of reception and transmission’. Genre ‘allowed for, and was often constructed as, a dialogic exchange with other genres, to which it referred in its use of images or which it completely incorporated’. This definition of genre as ‘open’, ‘dialogic’ and flexible applies particularly well to barahmasas, as their various manifestations will show. For a start, the three chief characteristics of barahmasas—the woman’s voice, the pain of separation from the beloved (viraha) and the catalogue of images pertaining to the seasons—are not exclusive to this genre, they are part of the Prakrit and Sanskrit poetic tradition from its early stages. The

theme of the ‘six seasons’ (ṣaḍṛtu) was a great favourite and indeed a setpiece for the aspiring Sanskrit and Prakrit poet. Underlying both poetry and medicine—for poetry about the seasons is also found in Sanskrit medical texts—was a common perception of the qualities of each season and the activities, food, dress and behaviour appropriate to each of them. Thus medical texts grouped the six seasons in two sets of three. The first set— śiśir, vasanta and gṛīṣma (late winter, spring and summer)—was qualitatively hot and dry. Varṣā, śarad and hemanta (the rainy season, autumn and early winter), the second set, was characterised as cold and wet. The ‘hot and dry’ months were deemed debilitating for the human body, while the ‘cold and wet’ months were said to be invigorating.2 The poetic ‘description of the seasons’ (ṛituvarŇana), rich in metaphoric connections and expressing time as a cycle, ‘constantly on the fulcrum between memory and expectation’, expressed a ‘generic way of thinking about time’ that went much beyond the realm of poets.3 However much they shared in terms of the stock of images and of the underlying moods and imagery connected to the seasons, barahmasas seem to have been in origin a purely popular genre, distinct from classical court poetry.4 That ‘six seasons’ and ‘twelve months’ were perceived as separate set-pieces expressing different moods is well brought out by the sixteenthcentury Sufi poet Malik Muhammad Jayasi in his Padmāvat (see de Bruijn in this volume).5 According to the common pattern of these epics, the hero leaves his first wife in his quest for the heroine. The preparations (śṛṇgāra) of the heroine, Padmavati, and the pleasures she enjoys over the course of one year with the hero, Ratansen, are described in terms of a ‘six seasons’ format and draw on the classical tradition in terms of tropes and diction. This ṣaḍṛtu is immediately followed by the sufferings of Ratansen’s first wife, whom he abandoned at home in his pursue of Padmavati. Her torments are instead expressed in the form of a viraha barahmasa. All the natural and climatic elements that made the lovers’ dalliance all the more delightful in the first piece are reversed in sign: the same clouds, rains and bird’s cries that heightened pleasure there now heighten the pain, and even the special insects that come out in the rainy season can be symbols of the restless birahinī. Here is Padmavati’s rainy season in the ṣaḍṛtu: Ritu pāvas birasai piyu pāvā; sāvan bhādaũ adhik sohāvā.

Kokila bain pā̃ ti baga chūtī; dhani nisarī jeũ bīr bahũṭī Camakai bijju baris jaga sonā; dādara mora sabada suṭhi lonā. Raṃga rāti piya saṃga nisi jāgai; garajai camaki cauṃki kanṭha lāgai. Sītala bunda ū̃ca caubārā; hariyara saba dekhiya sansārā Malai samīra bāsa sukha bāsī; beli phūla seja sukha ḍāsī Hariyara bhummi kusumbhī colā; au piya sañga racā hinḍolā Doha: Pauna jharakke hiya harakha lagai siyari batāsa Dhani jānai yah paunu hai paunu so apnī āsa. [337] In the rainy season, with one’s lover near, Sāvan and Bhādõ are so much more pleasant. The cuckoo speaks, rows of herons fly, women flock out like scarlet raininsects. Lightning flashes, the earth glows in the rain, lovely sound the frog and the peacock. Awake all night in love-play with her lover, she clutches him startled by thunder. Cool raindrops, high-raised platform, the whole world looks verdant. Scented is the malay breeze, and happy, she spread a bed of tendril flowers. Against the green earth, wearing a saffron bodice, she tied a swing with her beloved. Doha: The gusts of wind gladdened their hearts, cool felt the breeze, She thought, this wind brings contentment. But here is Nagmati’s Savan in the barahmasa: Sāvan barisa meha atipānī; bharani bharai haũ biraha jhurānī. Lāgu punarbasu pīu na dekhā; bhai bāuri kahã kanta sarekhā. Rakata ka āṃsu pare bhuī ṭūṭī; rengi calī janu bīr bahūṭī Sakhinha racā piu saṃga hinḍolā; hariyara bhuī kusambhi tana colā.

Hiya hinḍola jasa ḍolai morā; biraha jhulāvai dei jhāṃkorā. Bāṭa asūjha athāha gāṃbhīrā; Jiu bāura bā bhavai bāṃbhīrā. Jaga jala būḍi jahā̃ lagi tākī; Mora nāva khevaka binu thākī. Doha: Parabata samunda agama bica bana behaṛa ghana ḍhankha. Kimi kari bheṭaũ kanta tohi nā̃ mohi pāṇv na pankha. [345] Savan clouds are pouring down; fields are flooding but I burn with viraha. Punarvasu’s constellation has appeared, my husband has not. I’m going mad, where can he be? My tears of blood fall on the earth and scatter like scarlet rain-insects. My friends have hung swings with their lovers, green is the land and saffron their clothes. My heart swings, too, to and fro, tossed by viraha with violent blows. Deep and unfathomable is the road; crazed I wander like a bhambhira fly. As far as you can see water covers the earth, [but] the boat of my life is without my oarsman. Doha: Inaccessible mountains, oceans, forests and jungles stand between us, How can I come and meet you, my spouse, without feet or wings!6 This incorporation of a barahmasa into an epic or a romance is found both in Rajasthani rasaus like the Bīsaldev Rāsau (oldest manuscript, 1576) and in Avadhi Sufi romances right from the first extant example, Mulla Daud’s Candãyan (1379).7 In these works, the barahmasa works as a stock element, barely woven into the narrative of events. Both in the frame and in the body of the barahmasa, the narrative element usually remains negligible.8 In her study of barahmasa folk songs, Susan Wadley has emphasised the close correspondence between the natural description of the months, the mental and physical state of the heroine and ritual events marking the calendrical cycle. The physical and psychological distress of the deserted

heroine, she argues, is compounded by her inability to perform the critical ritual activities of women and to enjoy her lover’s presence at the times of the year (the rainy season, the cold month of Pus) when sexual activity is encouraged, if not mandatory.9 However, although farmers’ barahmasas do include the critical events of the agricultural calendar, the same is not true for most barahmasa songs. Nor are women’s ritual events that central, either. Thus, we could say instead, that this genre implies the climatic, ritual and affective characteristics of each month and season but is flexible enough to allow many possible emphases. Though originally folk songs, barahmasas were taken up by vernacular poets from the earliest stages of neo-Indo Aryan literatures, possibly in their search for available models. As this section will show, this appropriation of an oral form opened the genre to a host of possible changes according to the level of sophistication of the poet, the poem’s function and its prospective audience. The barahmasa was flexible at the formal level itself: it could be incorporated into longer poems (e.g. Avadhi premakhyans and Rajasthani rasaus), provide a poetic ‘base’ for palace wall-paintings and approximate Perso-Urdu masnavis.10 In this sense, the barahmasa can be understood as a ‘matrix’ of semantic paradigms and of positions in the literary field, in the sense used in this volume by Thomas de Bruijn. This flexibility is evident even in what may appear as a stable element of the genre, i.e. the adoption of the female voice by male poets.11 Although this element played a significant part in the appeal the barahmasa genre had for poets and in their poetic strategies, it should by no means considered in uniform terms. Ann Gold and Gloria Raheja have alerted us to the importance of context and subtext in order to understand women’s folk songs.12 In considering the following corpus of barahmasas authored by male poets, I would similarly like to keep in mind the fact that the suffering woman’s voice became a trope which could be used to flatter the courtly patron as the ideal lover, to express spiritual thirst and the strain of devotion, to provide visions of a pastoral world or simply to explore a whole new set of possible metaphors that the male subjectivity of the ghazal could not create (see also Oesterheld in this volume).

A rough typology of Hindi and Urdu barahmasas is enough to show the multiple positions they came to occupy in the literary field: a. religious poems (Jain barahmasas, Mirabai, Guru Nanak, Guru Arjan); b. part of longer narratives (Padmavāt, Bīsaldev rāsau); c. Braj Bhasha riti poems (Keshavdas, Lakhansen), sometimes to accompany miniature albums and palace wall paintings; d. Urdu Rekhta poets both before and after 1700 (Uzlat, Afzal). These are often longer poems (300–400 vv.) with greater focus on the heroine’s mental and physical distress. Metre and imagery are partly drawn from Perso-Urdu poetry, but the language remains very mixed. And, printed in the nineteenth century: e. poems by popular urban Urdu poets (Maqsud, the anonymous author of Sundarkalī and Ranj). f. songs printed singly and in collections in both Nagari and Urdu script (Khairashah, Harnam, etc.) g. short religious songs (Benimadho, etc.) So far, this essay has examined barahmasas in Avadhi romances. The following sections take up examples from the other groups. RELIGIOUS BARAHMASAS One typical change in function took place at the hand of religious poets, who either interpreted the heroine’s viraha in spiritual and/ or devotional terms—as the pain of the soul’s separation from God— or else grafted the template onto a story of spiritual conversion or tagged a didactic message onto every month.13 The earliest known literary example, a short poem in late Apabhramsha by the Jain monk Dharam Suri, combined prose and verse on the months of the year with verses of self-praise and fragments of songs on viraha.14 Charlotte Vaudeville has documented the significant production in old Gujarati of Jain barahmasas on the subject of Neminath’s

desertion of his wife Rajmati on their wedding day to follow a life of renunciation.15 Barahmasas were taken up for a religious purpose also by Nath Yogis, Sufis, Sants and Bhaktas. Krishna, the absent god-lover par excellence, fitted into the role very well, as the following barahmasa attributed to Mirabai shows, Piyā mohiṃ darsaņ dījai ho. Ber ber maiṃ ṭerahuṃ ahe kripā kījai ho. [Refrain] Jeṭh mahīne jal viṇā panchī dukh hoi ho. Mor āsārḥaṃ kuralahe ghan chātrag soī ho. My love, give me the vision of Yourself: Again and again I call You, have pity on me, ho! [Refrain] In the month of Jeth, for want of water the bird is in pain, ho. In Asarh the peacock throws its cry, the chātaka bird calls the cloud, ho.16 BRAJ BHASHA BARAHMASAS Roughly contemporary to Mirabai, Keshavdas (fl. 1600) perfected a sophisticated poetic idiom in Braj Bhasha that found immediate and wide success in courtly circles all over north India (see Allison Busch in this volume). His barahmasa seems to have achieved regional fame and was the one most often reproduced in palace paintings. According to the conventions of riti poetry, the verses in his poem are highly alliterative and dense with poetic figures. The metres are savaiya and doha and the protagonist are the ideal nāyak and nāyikā, identified as Krishna and Radha. The mood is, intriguingly, not one of viraha. Rather, the nāyikā describes the natural and ritual characteristics of each month and urges the nāyak not to go away—in a way, the classical treatment of the ‘six seasons’ adapted to a twelve-months format. The aim is to convey the mood of each month in its various aspects, the ‘generic way of thinking about time’ in Selby’s words, rather than to express mental and physical suffering. The paintings often show the couple in the foreground, while descriptive elements from

the verses offer clues to painters and establish a basic vocabulary for the depiction of every month. One painting of the month of Bhadon of the Bikaner school, for example, shows an elephant, a lion and a tiger, all mentioned in the verses: Ghorata ghana cahũ ora; ghosa nirghosini maṇḍahi, Dhārādhara dhara dharani musala dhārana jala chaṇḍahi. Jhillī gana jhankāra, pavana jhuki jhuki jhakajhorata, Bāgha-siṃha guñjarata, puñja kuñjara taru torata. Nisi dina viśeṣa nihi seṣa miti, jāta suolī oṛiai. Desahĩ piyūṣa paradesa viṣa, bhādaũ bhauna na choṛiai. Dark clouds have gathered all around; they thunder and thunder. Rain pours down in heavy torrents. Cicadas chirp, strong winds blow fiercely. Tiger and lion roar, elephants dishevel trees. Night and day, there is no difference, Home is nectar, travelling abroad is poison, one should not leave home in the month of Bhādõ.17 This is clearly a sophisticated composition, a courtly piece that, combined with the pictorial representation, formed an appealing ‘set’ on śṛñgāra for the princely patron, who could recognise himself in the nāyak and his sexual attractiveness, and could appreciate the richly decorated settings.18 Sets based on Keshavdas’s poem were painted at Malwa, Jaipur, Bikaner, Kangra, Datarpur, and Alwar up to the nineteenth century.19 In the hands of other Braj Bhasha poets, the result could be a kind of simplified śṛñgāra with seasonal details.20 Here is an example, in a manuscript dated 1776, by a poet called Lakhansen: Caṛhe asāṛh sukhahi ke māsā; jehi vidhi dehi so karai vilāsā. Dukh sukh ghar apne auva; ge pardes abhāgā chāyo. Chāīni mandir aur caupārī; raci raci chāyeni bārahduārī. Pākyau ām subhaga amrāī; amrita phala yek cār suṭhāī. Pākyau kaṭahar baḍaha kerā; pākyau pāna tamolina kern….

Sendūr cīr mai dharāvo hariyā; sudina devasa piyā dīkh ughārī. Dekhata nā̃ h leva ura lāā̃ ; ham ranga [sic anga] mori rahava alsāī. Doha: Āṭhau anga rasa bhīnī khāv adhara mukh pāna. Taba sukh hoi parāpati jab ghar āvahi kānha.21 Asarh has come, the month of happiness; those to whom God has granted it, enjoy themselves. Joys and troubles at home; he went away and left me wretched. [Others] thatch houses and roofs and make twelve-pillared pavilions. Mangoes ripen in the lovely groves, four or five lovely amritphal, Jackfruit, badahal and banana ripen. The betel-leaf tree of the betelseller too. Quick, put on sendur and clothes, to show yourself on the happy day your beloved appears. Seeing her lord she took him to her heart: ‘All my body feels languid’. Doha: Her whole body is drenched in rasa, her lips are red with betel, Only then she will find happiness when her spouse comes home. In Hindi critical vocabulary, this poem of 120 verses would be called rītimukta, i.e. devoid of literary tropes. The language is Bagheli, i.e. a mixture of Avadhi and Bundelkhandi (as seen from forms like ‘ge’ for ‘gaye’). The mood is also mixed, sensual rather than pathetic, with a bountiful nature (jackfruit, mangoes, areca nut and bananas are ripe, and a sensual heroine whose betel-red lips invite desire). As with Keshavdas, this shift in emphasis can be understood in terms of an audience of poetry practitioners and connoisseurs who would see the barahmasa as another form of erotic poetry. Although the Braj Bhasa poetry of Keshavdas and other riti poets did get printed in the early poetic anthologies (see Bangha 2001) and also in individual volumes, often sumptuously produced for princely patrons,

barahmasas do not seem to have figured in such publications, nor were they reprinted in the collections we will survey later. URDU BARAHMASAS A rather different treatment of the genre is the one we see at the hands of Muhammad Afzal ‘Afzal’, whose Bikaṭ Kahānī has attracted critical attention as the first significant poem in Urdu/Rekhta composed in north India (see Bangha and Oestherheld in this volume). Bikaṭ Kahānī was only the first in a considerable production of Urdu barahmasas documented by Dr Tanvir Ahmad Alvi (1988), which started in the seventeenth century and continued throughout the nineteenth century and well into the world of print. Alvi’s edited collection of twelve Urdu barahmasas shows that although the genre did not become part of the high Urdu literary tradition that developed in north India in the eighteenth century around the ghazal, it was nevertheless widely practiced, all the way from Rajasthan to Bihar. Three kinds of poets seem to have written Urdu barahmasas in the eighteenth and early nineteenth centuries: Sufis like Hazrat Jauhri of Bihar, Maulvi Hafizullah Qadiri of Badaun and, in the nineteenth century, Shaikh Ibadullah Badal ‘Ranj’ and Shaikh Abdullah Ansari Kanpuri; educated provincial officers/ servicemen, like Ilahi Bakhsh; and ‘bazaar poets’ like Maqsud of Lucknow. They were all part of the wider world of Urdu poetry but probably outside the aristocratic poetic circles of ustads, shagirds and musha’iras.22 Generally speaking, in the hands of these poets barahmasa developed either on the lines of a romantic masnavi, sometimes with spiritual but usually with secular overtones, or else closer to the perceived realm of the pastoral which drew upon secular and/or Krishna love matter, as can be seen especially for the theme of Holi. Only a few more sophisticated poets like Uzlat and Vahshat (see below) brought to bear the treasure trove of Perso-Urdu poetic themes and style and crafted metaphors that developed the subject of love-suffering. Instead, most Urdu barahmasa writers, even when drawing upon Perso-Urdu poetic vocabulary, used only first-level metaphors and simple tropes. Some followed Afzal’s example and continued to use ‘mixed’ (Rekhta) language, alternating Persian and Hindi lines. Most went for a colloquial register full of Hindi tadbhava

words. While Christina Oesterheld’s essay on marsiyas in this collection suggests that such a register was used by marsiya poets to evoke the domestic world of women and their grief, in the case of barahmasas we will see that the colloquial register rich in idiomatic expressions and tadbhava words evokes a mundane world of natural details, seasonal tasks and festivities.23 Nature itself is a mixture of poetic stereotypes (the peacock and the cuckoo, the black monsoon clouds) and of more minute details, like the bīr bahoṭī (Buccella carniola or ‘rain insect’), the particular red insect that comes out during the rains, or the names given to the rain in each specific fortnight of the monsoon (Maqsud). At a time when Urdu poetry in north India was developing along sophisticated Persian models in the wake of Vali's sensational ghazals, barahmasas offered an appealing folksy idiom and a not-too-implicit sensuality, linked to the feminine voice and the cycle of seasons. Though clearly not ignorant of the Perso-Urdu poetic tradition, Urdu barahmasa poets strove to use a different idiom, and mixed their Khari Boli with Braj Bhasha and local words, proverbs and Persian words as ‘mispronounced’ by uneducated people to achieve a ‘popular’ (for want to better term) effect. Afzal’s Bikaṭ Kahānī is a lyrical (not a narrative) poem in which Afzal returns again and again to the sufferings of the birahini. It is a typical viraha barahmasa right from the first verse: Suno sakhyo bikaṭ merī kahānī; bhaī haū̃ ‘ishq ke gham sõ divānī Na mujh kõ bhūk din nā nīnd rātā; birah ke dard sõ sīnā pirātā. Tamāmi lok mujhi baun kahi n; khirad gum karda majnū̃ ho rahi n (Afzal 1991: vv 1–3) Listen, friends, to my sad story. I have become mad with the pain of love. I have no hunger in the day, no sleep at night. Separation hurts my chest. The whole world calls me crazy. I’ve lost my reason, I’ve become mad. Afzal really set the ground for the development of the genre in Urdu, and his example was acknowledged either explicitly (Ilahi Bakhsh) or implicitly (Vahshat, Najib, Ranj) by later poets. For one thing, he used a metre common to Persian and Urdu masnavis (ḥazaj-i musaddas-i maḥzūf) which

all Urdu barahmasa writers adopted after him. Secondly, his poem is considerably longer than folk barahmasas (vv 321), and most Urdu barahmasas after his varied between two and four hundred verses. According to Tanvir Ahmad Alvi it was transmitted mostly orally, and the oldest manuscript available dates from almost two hundred years after its composition. While the relatively large number of manuscripts from all over north India and the Deccan testify to its enduring popularity, it is hardly mentioned in tazkiras of Urdu poets.24 The poetic conventions that Afzal introduced would have alerted his readers and listeners that this was a poem like a masnavi: thus the first 57 verses consist of a general statement on love.25 Love (‘ishq) is an illness which no medicine, doctor or exorcist can cure (v. 4). It is a fire in which the whole world burns, and only those who experience it know the pain it brings. Love is everywhere, in the mosque, the Ka’ba and the temple, and everything we see is the work of love. Next, the birahini speaks of the madness and selfforgetfulness that love has induced in her: like Majnun, it has set her apart from her family and led her to a metaphorical desert, so that now she has lost all sense and, wandering like a beggar, she is made the object of derision. Madness has planted its flag in the kingdom of life and has played havoc in the house of reason. It has stolen her patience and burnt her sense of comfort; it has made her drink the cup of madness and forget herself. Finally, love is described as magic, which can turn a wise man into a madman and vice versa. An initial moment of happiness with her beloved aroused the envy of heaven, which then separated them again.26 This temporary union (vaṣl) followed by a long and painful separation is not an uncommon theme in Perso-Urdu poetry, but only Afzal, and a century and a half later, Vahshat used it in their barahmasas.27 The barahmasa proper consists of months of varied length (between vv 9 and 30), beginning with Savan. Phagun, with the Holi festival, gets the most attention (vv 40).28 At the end, when the husband/ lover has come back in Asarh after the birahini had a premonitory dream (vv 293–300), a few lines warn against making love, for it is not easy and deprives you of all comfort (vv 308–20).29 The hypothesis that Afzal might have taken the cue for his barahmasa from folk songs is corroborated by the formulaic filler-verses (‘Sajan

pardes, maī̃ nit dukh bharū̃ rī’; ‘[Agahan] dukh de gayā ab [Pūs] āyā’; ‘Piyā ke dard se bis khā marū̃ rī’, etc.)30 that are practically absent from the later, more polished Urdu barahmasas. But to the template of the calendrical cycle, with its web of climatic, ritual and emotional connections, and with its motifs of the pandit and the mullah, the bird-messenger, the dream and leaving home as a jogin— to all this, Afzal added the symbolism and language of love of Perso-Urdu poetry. For example, if Afzal’s depiction of the onset of the rains in the month of Savan and its effect on the birahini is a typical example of the theme: Caṛhā sāvan bajā mār31 au naqqāra; sajan bin kaun hai sāthī hamārā Ghaṭā kārī cahārõ or chāī; birah kī fauj ne kīnī caṛhāī. Savan has come with drums and kettledrums; without my love I have no company. Black clouds overcast; the army of birah has attacked me. (Afzal 1991: 28, 49–50) In the next section, the raincloud (ghaṭā) is equated with the raincloud of sorrow that is about to burst (into tears) in her breast, and the drumming noise of the clouds (like naqqāra) is ‘sounding the drum of death for mec. Perhaps because of direct borrowing from women’s folk songs or because it was an intriguing voice to adopt, there are several verses in which the birahini reproaches her absent husband in the strongest terms—a kind of pleasure in colourful and down-to-earth ‘woman’s language’ that we find again in other Urdu barahmasas and in the mid-nineteenth century masnavis by Mirza Shauq Lakhnavi.31 Here, for example, the birahini asks the crow-messenger to go and tell the pardesī piyā: ‘Ahad kar gae ajhū̃ na āe; are kin saut ne ṭone calāe. Daghābāzī musāfir sõ na kīje; itā dukhṛā ghatībõ ko na dīje. […] Gayī so jāne de, ab āo ghar re; are z̤ ālim khudā kā khauf kar re. You went with a promise, but never came back; what stepwife has ensnared you? Do not cheat the poor traveller, do not afflict the helpless. […]

Let bygones be bygones, come home now. Oh cruel one! Fear God at least!32 The suspicion that this kind of reproach is a pose that the male poet (and audience) much enjoyed is supported by some other verses in which this kind of anger is immediately followed by protestations of love, indeed of subjection: Are z̤ ālim tere paiyā̃ paṛū̃ ri; dil o jān tujh upar qurbā̃ karū̃ rī. Terī bā̃ dī kī bā̃ dī ho rahū̃gī; jo kuch mujh ko kahegā so karū̃gī. Kahegā so karū̃gī āo re hāy; mukh apnā ṭuk mujhe dekhāo re hāy. You cruel one, I fall at your feet. I’ll give my life for you. I’ll serve you as your servant’s slave. I’ll do everything you say. I’ll do what you say. Oh please do come, just show me your face.33 As Imre Bangha has already amply shown in his essay in this volume, the language of Afzal’s poem is a Rekhta that alternates between Khari Boli and Persian (see also Allison Busch in this volume). This remained customary for several Urdu barahmasas up to the nineteenth century, even after Mir’s kind of Rekhta had become dominant—as if to show the lasting impact that Afzal’s poem had on Urdu readers and poets despite its noncanonical status.34 A hundred and fifty years later, Mufti Ilahi Bakhsh, a jamādār (police officer) in the service of the Raja of Kishangarh wrote quite a sophisticated barahmasa under the same title and clearly modelled on Afzal’s, whose popularity he explicitly acknowledged:35 Sunī hogī bikaṭ tumne kahānī; sab uske khāme kī ātish fishānī. Valekin maī jo dekhā usko sārā; t̤ abī’at ne merī ik josh mārā. Ye āyā dil mẽ, kahie ek kahānī; bayā̃ ho jis mẽ soz-e dil nihānī. You must have heard the Bikaṭ kahānī; everyone praises its pen to the sky. But after I had seen it all, I was taken by a sudden zeal I thought, why don’t I tell a story which describes the inner pain of the heart?

Those who wrote barahmasas were not, it is worth restating, the famous and cosmopolitan Urdu poets. The only exception is Mir Abdul-Vali Uzlat (1750ca.) from Surat, a friend of Mir’s, whom he helped by providing details about poets in the Deccan for his tazkira. A great traveller, he is known to have been interested in painting and to have composed poetry in other ‘Hindi’ genres, including a ragamala, and several kinds of riddles (do arthi, shish arthi), as well as kabitt, chand, dohra, soratha, jhulna and other popular genres.36 His choice of a barahmasa thus seems in line with his other interests.37 His is a much shorter and compact poem (79 vv.), also in the ḥazaj metre and with an introduction on love (14 vv.) followed by regular sections of five verses each, starting with Asarh. The language is Khari Boli. The interest of this poem for us lies in the way it shows an accomplished poet in Persian and Urdu bringing metaphors and wordplay into the template in order to craft verses dense with literary echoes. Thus, the onset of the rains, already described by Afzal as the attack of an ‘army of clouds’, is expressed by Uzlat in the following terms: Asāṛh āyā hai bādal kı faujẽ; ki teghẽ barf kı man ha> maujẽ. Girī̃ haī̃ ashk sang uṭh uṭh malve; hamāre lakht-e dil ke pal pal ole. Asarh came, and armies of clouds; gale after gale blowing swords of ice. Tears of stone fell, ruins galore; hail crushed my heart to pieces.38 ‘Armies of clouds’ strike with ‘swords of ice’; teardrops, implicitly equated with raindrops, are so hard and heavy that they reduce everything to ruin, while hailstorms break her heart into pieces. The other women’s pleasure in swinging on swings in the rainy season is juxtaposed as follows to the predicament of the birahini, whose life swings dangerously on the thin thread of her breath: Jinhõ ke pī haĩ ghar so caṛh hinḍole; khushī se pẽgtĩ hai kalole. Jhūlātī hū̃ mai; jhũlā sā̃ s kā hāy; jo pī āvẽ to dil kā t̤ ifl sukh pāy. The women whose husbands are at home climb on swings, surging and swinging. I swing on the thread of my breath, let my beloved come, the ‘child of my heart’ will rejoice.39

The birahini ‘burns’ in the fire (tap, tapnā) of separation even as everybody else is shivering in the cold, and she feels all the hotter in the hot season. In the hot month of Baisakh, wrote Jayasi, even a sandal-scented bodice ‘scorches’ Nagmati.40 Often this burning has explicit sexual connotation, but in this literary Urdu barahmasas heat and fire are producive of other metaphors as well. This is how Uzlat writes about Baisakh: Jab āyā āg sā ye mās baisākh; umīdẽ dil mẽ sab jal kar hui rākh. Gudāz-e yās se jyū̃ dil pighal jāe; dil ānsū ho palak par āke ḍhal jāe. When Baisakh came like a fire; it burnt all hope in my heart to ashes. As it melts out of despair, my heart dissolves into a tear on the eyelashes.41 Before we turn to more folksy barahmasas, we should mention the Bikaṭ Kahānī by Ilahi Bakhsh, composed probably a few decades later in the early nineteenth century. This poem is not only evidence of the enduring popularity of Afzal’s model among amateur Urdu poets, but it also gives a good sense of what became of a barahmasa in the hands of a provincial officer with a Persian education and a literary sense. His poem, like Vahshat’s, is rather longer than Afzal’s (410 vv.) though in the same metre.42 Linguistically, it is in Khari Boli with some Braj Bhasha orthography and alternative grammatical forms (e.g. karat hai as well as kare hai). Also in terms of vocabulary, the range is strikingly broad. Persian and even Arabic words are used freely, according to both the original and the Indian pronunciation. In the verse ‘Yakāyak ān kar jab ra’d garjā/dhamak se ūs kī merā jī larjā’ (v 125), for example, the Arabic word ra’d is used for ‘cloud’ instead of the usual megh and bādal, but the Persian verb larzīdan is spelt as it is pronounced in Hindi, i.e. larajnā.43 Is this a conscious imitation of local pronunciation? Verses that are typical of folk barahmasas like: Gaī yek ‘umr, pī ab tak na āe; kaho kyõ ab tak pardes chāe. A whole life has passed, yet my love is not come yet; tell me why has he gone abroad (v. 271)

are offset by Urduising others, like this description of spring: ‘Ajab dilkhvāh mausam suhāvnā hai Sakhī gulshan mū̃ har ek gul khilā hai. Wondrously pleasing is the weather. My friend, in the garden every flower is in bloom. (v 327) The mixed lexicon and imagery is best shown by this example from the month of Baisakh, where typical Indian birds (like the kokilā and the parevā) are mixed with those of Perso-Urdu poetry (the bulbul, the qamrī, the fākhtā): Karẽ cuhalẽ caman mẽ jāke bulbul, macāvẽ shor aur ḍālẽ bahut ghul. Phire qamri khushī āvāz kartī, sanobar sāt milkar nāz kartī. Pukāre fākhtā kūkū udhar se, lagāve āg mujh dukhiyā ke bar se. Are jab kokilā ne kūk mārī, lagī dil mẽ mere birahā kaṭārī. Yakāek jab kahī̃ bole parevā, ajī mujh birahinī kā jān levā. Ma> sun āvāz ko ḥairā̃ ho hū̃, piyā kī yād kar gham se jarū̃ hū̃. Nightingales make merry in the garden chirping loudly and noisily. The happy partridge struts and coos, flirting happily with the fir tree. While the ring-dove cooing aloud sets my poor breast on fire. Oh anywhere I hear the cuckoo’s cry; it’s a dagger piercing my poor heart. When the turtle dove raises its sudden cry, it takes away this poor woman’s heart. These calls and cries have made me distracted; the memory of my beloved is burning me. (Alvi 1998: 273, vv 350–355) This hybrid language well corresponds to a hybrid poetic idiom and imagination. For example, although the birahini conforms by and large to the type of the ‘gopi pining for Krishna’—for example, she goes and searches the forest and rails at his treacherous dalliances with other women —there are other touches, too, that bring her closer to the flirtatious and argumentative heroines of Shauq’s masnavis, for example her threat to die

and bring rusvā-e ‘ālam on the cruel beloved. Typically, Persianising verses describe the lover’s pitiful state, while the vocabulary turns to Khari BoliBraj Bhasha when rituals and festivals are mentioned. Here, for example, the merriment of other women during Holi is described in greater detail and with greater enthusiasm than usual: Sajan par apne nakatoṛā karẽ rī, khushī se rang mẽ pī ko bharẽ rī. Liye hātho mẽ pickārī phirat haĩ. ‘Abīr apne piyā ke mukh malat haī̃, Gulāl aur rang sab kuch khelte haī̃, gulāb aur mashq us me malte haī̃. Are ‘ālam mẽ horī saj rahī hai, daf aur mirdang kī gat baj rahī hai. Harek mahalõ mẽ bas bājẽ haī̃ jhānjan. Suhāgin nār ke pāõ ke bichvan. Rangīlī cū̃rī par rang ḍārẽ, bhaī horī mẽ baurī sārī nārẽ. They sneer at their husbands, those happy ones, and merrily splash them with colour. They strut holding their pichkaris, rubbing their faces with colour. They play with colours and coloured water, smeared with rose and musk. Holī is warming up all around, you can hear the tambourine and the drum. Anklets and toe-rings jingle from every house, those happy married women. Throwing colour on their colourful chunris, all the women flow crazed in the stream of Holi (270–71, vv 315–22). What we see at work is a move within the genre away from Afzal’s exclusive focus on love as suffering to the combined appeal of a pining and/or argumentative heroine and the worldly pleasures that the other women, the suhagins, actually enjoy. It is precisely this kind of combination that we shall find in the more popular barahmasas printed time and time again in the nineteenth century. The barahmasa by Maqsud, a Lucknow watercarrier (bhishtī) and a contemporary of Mushafi who acquired some fame and following as a ‘bazaar’ poet and is mentioned begrudgingly in Mushafi’s Tażkira-yi hindī (completed 1794–95), is a strikingly confident poem in standard Khari Boli, of medium length (208 vv) and in the usual hazaj metre. In size and in

language, Maqsud’s barahmasa stands somewhat in the middle between the folk songs and the more literary Urdu narrative poems.44 Uniquely, the occupation that keeps the birahini’s husband away is actually mentioned, and several verses are devoted to fantasies about that pardes: he is a soldier at the service of a hakim, an officer, and the heroine wishes him to be transferred (taghrīr) so that her husband will come home (vv 126–28). Stylistically, the birahini’s voice is very similar to what was projected as ‘woman’s speech’ in nineteenth-century masnavis—a language full of expletives, proverbs and insults. In fact, Maqsud appears equally at ease with both sets of vocabulary, that of Perso-Urdu poetry and that of local, everyday speech, and is able to mix them in an easy flow: Birah kī fauj ne ā mujhko gherā. Nahĩ̤ mujhme ab hosh merā. Nikal ākhõ se lakht-e dil kī boṭī. Zamī̃par ho gaī bīrā bahoṭī. Piyā jis roz ghar se thā sidhārā. Qamr-e ‘aqrab me āyā sitārā. Sitārā naḥus thā aur bad ghaṛī thī. Shab-e furqat merẽ sir par khaṛī thī. The army of birah came and surrounded me, leaving me out of my wits. My heart broke in pieces out of my eyes; it fell on the ground like scarlet rain-insects. The day my love left home; the star came into the moon of Scorpio. It was an ominous star, an ill-fated moment; the ’Night of Separation’ came upon me. (Alvi 1988: 229, vv 52–55) At first Maqsud’s poem focuses on ritual events,45 but in the second part his attention wanders away from the calendar to other mundane details: thus during Kartik ‘people start making quilts and mattresses’ (v 102), other girls poke fun at the birahini by calling her crazy and saying that she is under the influence of an evil spirit (sāya, vv 121–24), and she turns to various religious specialists promising generous rewards for their services once her husband comes back: Shagun tũ khol pothī dekh pānḍe. Piyā āvẽge mukh se bol pānḍe. Piyā kā jis ghaṛī darsan karū̃gī. Terā ghī-khānḍ se maĩ mukh bharū̃gī. Pande, open the book and look at the signs, tell me my love will come.

The moment I see him come back, I’ll fill your mouth with ghee, sweets and sugar. (vv 89–90). She calls him again a few months later to avert the evil astral influence that keeps her husband away, and gives him offerings (dān-ṭīkā, 172), but she also goes to the dargah to pray on the Pir’s tomb and promises to donate sweets and a cover if her husband comes back (vv 176–80, 203). In short, Maqsud’s barahmasa provides an example of a successful blend of a folk template, Khari Boli poetic diction and some Perso-Urdu imagery which we find replicated in the printed barahmasa songs and collections (see below). In fact, his poem must have achieved some popularity, since we find it printed in Devanagari in the 1870s, as well as copied down in Devanagari by individual readers. That Maqsud’s peculiar stylistic blend was not unique is confirmed by a similar text that was also printed several times in Lucknow in the 1870s as a single poem (12 pp), i.e. the anonymous Bārahmāsi Sundarkalī. This is even more of a hybrid work, in that it brings in narrative elements (it is framed as the story of Prince Ratan Kunvar and Princess Sundarkali) and takes greater liberties with the description of the months. Thus, the poem begins in Asarh but with no mention of the rain (which only comes in Savan), and the cold nights fall not in Pus but in Magh. Other seasonal and ritual events are absent, apart from Holi, which takes up the whole of the month of Phagun. The ‘fire of biraha’ that consumes the heroine from the very beginning is here presented in decidedly physical terms. We are told that they were married as children and he immediately left ‘for the East’: Kite din sõ mujhe huī javānī. Javānī ne kiyā mujhko divānī. Javānī ne kiyā mere upar josh. Kaṭe bāl apne mẽ ab huā hosh. Rangīle sej par ārām kīje. Khushī o rang-ras kā kām kīje. I reached youth a long time ago, my youth has made me wild. Youth has made me burn I realise my childhood is over. Do rest on this delightful bed, make merry and be happy. (214, vv 12–14)

The cold season and warm clothing are described through significant objects, which signify the pleasures that warm beds offer to those who can purchase them. Mahīnā māgh kā paṛtā hai pālā. Sakhī sab ghar-baghar oṛhe dushālā. Bichaunā toshak aur oṛhẽ nihālī. Sabõ ko ek taraf apnī taiyārī. Kisī ko chīnṭ gujrātī o rṅumī. Kisī ko bandarī paṭne kī mṅṅṅumī Mujhe har vaqt ye jāṛā satāve. Nihālī aur razāī ko oṛhāve. Bichaune ke upar phulõ kī kaliyā̃ . Piyāre pīu kahā̃ vo rangreliyā̃ . The month of Magh, frost has set in, all my friends are home wrapped in their shawls. Tucked into their quilts and covers, they are all well prepared. One wears chintz from Gujarat, one from Rum, Portbandar or the waxed one of Patna. I am, instead, plagued by the cold, forced to burrow under mattress and quilt. These flower buds embroidered on the covers [remind me] of the merry times with my love! Piyā kāran merī phaṛtī hai ā̃ khiyā̃ . Jo āve pīu lagāū̃ chātiyā̃ . Palang ũpar do tan lipṭāe rahte. Khushī ho rang-ras kā bhog karte. Ki jāṛe mẽ mazā hotā anṅuṭhā. Jo guṛ-ghī se bhī lāge hai mīṭhā. Agar ānā hai to jald se āo. Mazā jāṛe kā hai ākar cakhāo. Because of him my eyelids flicker; if he comes I’ll take him in my arms. We’d lie together on our bed, and take our pleasure to the full. Unique is the pleasure of the cold season, sweeter than guṛ and ghī. If you are coming, come quickly; give me a taste of the pleasures of this season. (Alvi 1988: 217–18, vv 85–90, and 219, vv 94–97) Finally, for Chait the flowers in bloom and the gardener taking his pick in the garden offer the possibility of extending the metaphor to the heroine herself with further erotic double-meanings, in a language that recalls that of thumris and of contemporary theatre songs:

Hamāre bāgh me phūle hai vo phūl. Vo kis ke bāgh me huā hai mashghūl. Camelī sī dehī sakhī hamārī; rasīlī ā̃ kh yū̃ nargis he pyārī. Nayan ke pās campe kī kalī hai. Balak ye nāk campe se bhalī hai. Merā mukh hai ki jyū̃ gulāb kā phūl. Na jānū̃ bāghbān kyā ho gayā bhūl. Vo joban hai ki jyū̃ anār kā phal; bharī miṭhās se misrī kī jyū̃ dal. Bahār āyā, dukhanhār na āyā; anār āyā, cakhanhār na āyā. That flower has bloomed in my garden, but he has lost himself in another. My body is like a jasmine, my friend, my eyes, liquid like narcissi. Between the eyes is a champa bud. No, my nose is even prettier than that! My mouth is like a rose bud; God knows what the gardener forgot! My breasts, two pomegranates: full of sweetness like sugar. Spring has come, my saviour has not; the pomegranate is ripe, but there’s no one to taste it. (Alvi 1988: 221, vv 128–32, 136) This overview of Urdu barahmasas complements the argument made by Imre Bangha and Christina Oesterheld in this collection, i.e. that parallel to the sophisticated Urdu literary culture centred on the ghazal and on PersoUrdu models, another poetic culture existed that was not ignorant or oblivious of the former but drew upon Hindi poetic models and elements from folk culture and from contiguous modes like the pastoral-devotional and produced a range of poems which evoked a kind of alternative aesthetic to that of the ghazal.46 By tracing the evolution of the genre we have been able to see the ways in which courtly, qasbah and bazaar poets variedly exploited the folk template to create quite different poetic compositions. Afzal’s poem, for example, declined the figure of the suffering lover in the feminine gender, Keshavdas presented the ruler and his courtiers with ideal versions of their own, flattered selves, while Ilahi Bakhsh, Maqsud and others played on the juxtaposition between the suffering heroine and the happy, sensuous suhagins. In a shift recurring in several Urdu poems, the twelve-month cycle remained but the birahini did not necessarily link her suffering to the specific climatic, affective and ritual mood of each month. Also, while all

the barahmasas used the birahini’s feminine voice, some poems worked the pathetic emotional register, others the flirtatious and argumentative, and yet others the sensuous image of ‘wasted youth’—or a mixture of all these. In the end the barahmasa in Urdu became a narrative poem in ‘woman’s speech’, whose attractiveness lay in the expressions used by the birahini, in her latent sexuality, but also in the quaint details about her eclectic religious practices.47 This emphasis on worldly pleasures and on women’s eclectic religious practices is to be found again in the barahmasas that flooded the print-market from the 1860s onwards. It is to these that we now turn. BARAHMASAS IN PRINT Barahmasas were perhaps the first substantial genre in the boom in commercial publishing in north India of the 1860s.48 Unlike qissas, they first appeared in Devanagari and then gradually in Urdu as well. Crucially, the cheapness of lithography allowed the same printers to produce books in both scripts at low cost, without having to purchase a full set of type fonts: it was enough to copy a manuscript and print it. Lithography also allowed variants of writing and spelling and was therefore suitable for multilingual books and books in local dialects. Visually, lithographed books reproduced the experience of manuscripts, albeit with different formats and pagination. Barahmasas in Devanagari were printed in small, 16mo lithographed booklets of 8 or 16 pages, mostly and with a woodblock illustration on the cover, while in Urdu they were printed in the standard 12mo size of Urdu lithographed books, either singly or in collections of six or eight. The first edition I have found so far is a Bārahmāsā Khairāśāh printed in Agra by the Matba Ilahi in 1862.49 The small format and the fact that it was brought out in Devanagari by a printer which otherwise mostly published Urdu books suggest to me that it was perhaps a sideline initially, aimed at the audience which might have bought books occasionally at fairs and read in Devanagari and not Urdu—i.e. women. After five further editions in Devanagari, the Bārahmāsā Khairāśāh was printed in Urdu, too, in Delhi in 1870; by 1876, it was a staple part of Urdu barahmasa collections.50 Broadly speaking, two kinds of barahmasas dominated the market. First, short religious songs. Second, especially after 1870, longer songs like

Bārahmāsā Khairāśāh which combined a folk-song template with popular Urdu verse in the manner of Maqsud or the Bārahmāsā Sundarkalī also came into the market. Devotional barahmasas were poems of only a few, rather elementary verses, often attributed to Surdas or Tulsidas, and appear to be of a kind with other compositions meant for group singing:… Kṛṣṇa hī Kṛṣṇa raṭat narnārī (men and women repeat the name of Krishna) is the refrain of one of them, while another has on the cover a woodblock image that depicts a Sant (Surdas) sitting in front of a group of women, all singing a bhajan, with a temple in the background.51 The template of twelve months was used to evoke in the two, short lines of each stanza incidents or līlās connected to the deity, while the refrain brought the mood back to devotional fervour or viraha. Thus in the barahmasa attributed to Surdas, also published as Bārahmāsā Benīmādho, each stanza contained one single detail from Krishna’s exploits, while the name of the month at the beginning sets the beat along the calendrical cycle; Rāmcandra kā bārahmāsā evokes moments of Ram’s quest, and Dropadījī quickly retells the events related to the game of chess in the Mahabharata. George Grierson’s 1880s collection of Bihari folk songs quotes a number of them which have the gopis and Udhav, Kaushalya and Lakshman as protagonists; some of these songs collected from ‘ladies of Shahabad’ were, he noticed, in ‘nearly pure Bhojpuri’, others had probably been originally in Braj Bhasha and were transmitted orally by the Bhojpuri-speaking women.52 The picture suggested is that of the reproduction and transmission through print of songs that were sung among groups of women. This is an example from the Bārahmāsa Benīmādho or Sūrdās; the language is clearly Braj Bhasha, perhaps slightly affected by the transition into Urdu script: Kātik kalol karẽ sab sakhiyā̃ , Rādhā bicār kar man mẽ rī, Mādho piyā ko ān milāo, nāhĩ prān bace chan mẽ rī. Refrain: Hamkau chā̃ ṛ cale Benimādho, Rādhā sog kare man mẽ rī, Agahan gend banāy sā̃ vare, jāy khele ṭaṭ Jamnā ke rī, Khelat gend girī Jamnā mẽ, kālī nāg nāthṅu chan mẽ rī. (nd: 3)

Katik, all girl friends make merry, but Radha is deep in thought: Make me meet with Madho my love, or my life will leave me any moment now. Refrain: Benimadho has left me and gone, Radha grieves in her heart. Agahan, the Dark One has made a ball and plays on the banks of the Jamuna. As he plays, the ball falls into the river, where the black snake is. By far the greatest circulation, though, was achieved by worldly barahmasas by poets like Khairashah, Harnam, Allah Bakhsh and Shambhuray. Khairashah’s barahmasa was printed twenty times in Devanagari and Urdu in less then twenty years, Allah Bakhsh’s at least fifteen times in a decade, and Harnam’s at least six times between 1869 and 1878.53 These poems elaborated on folk songs and roughly followed Indic metres with stanzas formed by a doha and usually eight verses, for a total of around a hundred verses. The numerous internal rhymes, the short and regular stanzas and the presence of a refrain all suggest that they were indeed meant for singing, while the presence of the intercalary month of laund, absent from more literary barahmasas, suggests that they were closer to folk songs.54 Unlike folk songs, though, they were not in a local dialect but in a demotic language, a mixture of Braj Bhasha and Khari Boli; the vocabulary consisted almost exclusively of Hindi tadbhava words, mixed occasionally with simple Perso-Urdu vocabulary. In formal terms, therefore, they are to be distinguished from folk and devotional songs as well as from the poems in local Hindi dialects, from the courtly compositions in Braj Bhasha and from the Urdu tradition of Afzal’s Bikaṭ Kahānī. In terms of emphasis, they continue the shift away from the lyrical expression of the birahini’s suffering and towards a representation of a worldly aestheticethic that harnessed the popular aesthetic and structure of feeling connected to the seasons and the calendrical cycle in order to produce a more diffuse aesthetic of worldly pleasure expressed through the trope of the woman. The similarities between the poems—they vary mainly in the arrangement of tropes and expressions—suggests that the pleasure they

offered was one based on seriality and recognition, and on the variation in the selective mention of ritual and other details.55 After the initial focus on the birahini in the month of Asarh, with the customary parallel between the rain and her tears, the fear induced in her by lightning and her suffering enhanced by the birds’ cries, these songs move to the festivals and pleasures of the ritual calendar that she is missing out on. The major festivals of Dussehra in Kvar and Divali and Naurata in Kartik are mentioned, but the focus seems once more to be more on the worldly pleasures associated with the festivals, melas and clothes that the heroine is missing out on, than on ritual acts. This is Allah Bakhsh: Sakhī cauthā mahīnā kvār, ā jā dildār, kahtī yū̃ nār dvār par khariyā̃ . Ham kis par pahrẽ sakhī rī harī cũṛiyā̃ . Sab sakhī Nauratā khelatī, dekhtī mele piyā sang, leke gunṭhtī chaṛiyā̃ . Phūlõ kī barkhā lagā rahī jhaṛiyā̃ . Maĩ kis par dharū̃ Dasahra, ghar nahī̃ sājan merā. (vv 27–31) My friend, it’s the fourth month of Kvār, come my beloved, say the women standing at the door, For whom should we wear our green chunris? All our friends are ‘playing’ Nauratas, see the fair with their beloveds and weave sticks. Flowers are showering down, Who should I deck Dussehra for, if my love is not at home? Apart from the aesthetic effect created by the oil lamps, for Divali Harnam mentions the customary gambling playing: Ghar ghar divālī nauratẽ, sakhī pūjtī khelẽ juā. Lāl rang causar bichā hai, ek dā̃ v mẽ badkar dharā. Sakhiyõ ko pau-bārah, mujhe shish-panj do pāsā paṛā.56 Kar ās Gangā maĩ calī, raste kī sudh na bāṭ kī. Pyāre piyā Harnām bin, ghar kī rahī na ghāţ kī. In every house my friends worship Divali and Naurata and gamble. A red chausar mat is spread out; we betted our throw of dice.

My friends got full score; I got only ‘five and six’. I went to bathe in the Ganges; lost to myself and lost the way. Without my dear love, Harnam, I have nowhere to go.57 None of the barahmasas I have seen mentions Karva Chauth, Rakhi, Nagpanchami or Janmashtami or any of the other rituals deemed important for women. Instead, women’s religion in the barahmasas consists of celebrating festivals and parading in nice clothes for and with their husbands/lovers. All without fail mention Holi, where the frolicking and flirting of the other women contrasts with the heroine’s ‘empty youth’. The language of these barahmasas is significant for its hybridity. If the history of the nineteenth century is the history of the creation of modern standard forms and separate linguistic identities around Hindi and Urdu and of the reform of literary tastes, a process accelerating in second part of the century, then these texts point to the continued existence and currency of a highly mixed language and poetic idiom. The language is standard Khari Boli with some Braj Bhasha forms and occasional Perso-Urdu inflections, very far from that of the Urdu ghazal and of sophisticated Braj Bhasha poetry. Tadbhava forms abound, as they would have in the current speech of the time. This was a language that was clearly accessible to both Hindi and Urdu audiences, since the barahmasas were printed in both scripts with minimal variation. Print did tend to fix the language one way or another, mostly in line with existing written models: thus editions in Urdu script (which of course do not differentiate between the verb ending ‘ai’ ‘y’, ‘e’ and ‘ī’, e.g. karai, kare) tended to bring the language closer to Khari Boli by spelling Braj Bhasha verb forms as Khari boli ones (‘paṛe’ for ‘parai’, ‘choṛ’ for ‘chā̃ ṛ’) and occasionally using words more in line with Khari Boli Urdu (‘divānī’ for ‘dukhārī’, ‘thālī’ for ‘thārī’, etc.). The popularity in print of this kind of barahmasas shows that a genre which had known until then a multiple and layered history in women’s songs, courtly poetic exercises, experiments by Perso-Urdu literati and urban popular poets, acquired a new function of entertainment for the Hindi-Urdu reading public, to which it offered an attractive image of a woman who was flatteringly dependant on men for her fulfillment, unlike the defiant beloved of the Perso- Urdu ghazal. Whether it was the birahini

drawing attention to her wasted but youthful body or the beautiful suhāgins who enjoy sexual pleasures in harmony with the seasons, the message was a simple one: a happy woman is not just a married woman (for the birahini is also married after all) but, as Hafizullah Qadiri’s Bārahmāsa neh had already put it succintly: ‘A woman who is wanted by her husband is a suhāgin’ (‘Jise cāhe sajan vo hai suhāgan’. See ‘Alvi 1998: 208). After months of complaints and lamentations addressed to the husband who deserted her and sent no news of himself nor asked any of her, the heroine is usually overjoyed to welcome him again at the end of the poem, her position as a suhāgin and her sensual appearance immediately reinstated— with no admonitory lesson for the male audience. CONCLUSION This essay has brought together different kinds of barahmasa poems in Hindi and Urdu on a common map, trying to figure the levels which they occupied in the multilingual, stratified and evolving literary system of north India between the fourteenth and the nineteenth centuries. The advantage of focusing on a ‘minor’ genre of folk origin lies in the evident ease with which poets from different cultural milieux manipulated it and refitted it according to their own intentions, without any worry about its pedigree. Barahmasas represent the kind of open, dialogic genre that Thomas de Bruijn (in this volume) considers characteristic of the pre-modern north Indian literary system. Thus Sufis, from Mulla Daud to Jayasi, encapsulated barahmasas in their premākhyāns, the birahini’s suffering echoing a spiritual longing, but in pure Avadhi terms. At the turn of the seveteenth century, a Braj Bhasha riti poet like Keshavdas loaded the twelve-month template with poetic ornaments in a way that downplayed the birahini’s suffering and provided the perfect inspiration for palace paintings and miniatures. Only a few decades later, another poet, Afzal, this time coming from an Indo-Persian background, mixed the folk template with Persian phrases and expressions and produced a barahmasa that, together with its title Bikaṭ Kahānī, remained a blueprint for Indo-Persian poets for the next two centuries.

Imre Bangha (in this volume) has rightly called Afzal’s Bikaṭ Kahānī an example of ‘Mughal macaronic poetry’ and has argued that The existence of this kind of poetry in north India before 1700 suggests that the fashion for Persianised Hindavi that Vali brought to Delhi did not ‘create’ Rekhta poetry in north India but rather displaced the preexisting fashion for mixed language poetry. (87) The eleven other Urdu barahmasas from the eighteenth and ninteenth centuries collected by Tanvir Ahmad Alvi testify to the fact that poetry in Persianised Hindavi (the Rekhta that Mir preferred over all others), while hegemonic, did not completely displace the taste for this kind of poetry, whether in mixed language or in a colloquial idiom that combined a few, common Perso-Arabic words and tropes with tadbhava words and images. The results were nonetheless formally varied. While poems like Maqsud’s or the Bārahmāsā Sundarkalī were clearly narrative poems formally modelled, like Afzal’s, on the masnavi, the barahmasas that came into vogue in print in the 1860s and 70s were strophic song-poems. Linguistically and aesthetically, however, they point to a common fund of words, expressions and images which remained the same whether the template was pure Khari Boli or mixed Khari Boli-Braj Bhasha. Typically, the same song-poem would contain more Braj Bhasha verb forms and spellings when printed in Devanagari, and fewer when printed in Urdu. Rhyme would pair together words originally belonging from different poetic words (e.g. dildār and nār as in Allah Bakhsh above), creating a completely hybrid poetic idiom. The fact that the barahmasa was in the feminine voice of a woman pining for her absent beloved was clearly an element of attraction for the poets and allowed them to experiment with what Bakhtin called ‘speech genres’, in most cases in the direction of a colloquial, idiomatic register rich in tadbhava words to signify mundane, seasonal activities and natural details. It is true to say, however, that the ‘feminine voice’ was not a stable signified. It could be a pathetic effect of great emotional intensity, when the voice returned relentlessly to the heroine’s mental and bodily suffering. It could be sensuous, when the lament drew attention to the woman’s ‘wasted

youth’ or when it became an excuse to evoke the pleasures that the seasonal festivals offered to the women whose husbands were not away. Finally, it could evoke a homely world of seasonal tasks, consultations about omens with religious specialists and the nice things to be had and enjoyed at different times of the year. Taken together with the feminine voice of Urdu marsiyas that Christina Oesterheld writes about in this collection, or the flirtatious voice of Sa‘adat Yar Khan Rangin’s (1756–1834/5) Rekhti and Mirza Nawab Shauq Lakhnavi’s masnavis, we can see that there were several alternatives to the male voice of the ghazal.

6 Sadarang, Adarang, Sabrang Multi-coloured poetry in Hindustani Music Lalita du Perron

A draft of this paper was presented at the 18th European Conference on Modern South Asian Studies, Lund, Sweden, 8 July 2004.

I

n all genres of north Indian art (Hindustani) music, the language of the compositions is Hindi—though not as we know it today. Although true to the nature of an oral tradition we cannot always reconstruct the original texts of the songs, there is enough evidence (sometimes in manuscripts) to indicate that a wide variety of dialects of Hindi have always been represented in Hindustani music. The word ‘dialect’ should be used carefully, as the main literary form of Hindi, Braj Bhasha, which forms the foundation of many north Indian songs (though often modified in performance), long preceded modern Hindi as a literary language. However, as Stuart McGregor points out, Braj Bhasha may ‘properly be regarded as falling within a composite literary tradition of Hindi’.1 The question of a ‘composite literary tradition’ is significant, especially when discussing the genre khyal which is the focus of this paper. Song texts are rarely included in discussions of the Hindi literary tradition2, a hiatus my work aims to address. Part of the bias may be that Hindustani music was rarely contained within the boundaries of ‘north India’ (a somewhat problematic concept in itself) or indeed of the Hindi-speaking geographical area. The songs of the north Indian art music genres have therefore been

exposed to a variety of cultural, historical and linguistic influences. This broad base is, especially in the case of khyal, reflected in the wide diversity of topics and linguistic conceits represented in the songs. In this paper I look at the language of the songs of khyal on the strength of manuscript evidence as well as songs composed in the twentieth century. The aim is to arrive at an impression of how khyal fits into the Hindi lyrical tradition, and to conjecture how this sense of belonging has been achieved. It is no doubt salient that in spite of being aimed at aristocratic and sophisticated audiences who are likely to have been well-versed in Persian poetics, there is no evidence of this in khyal. This is why I speak of the genre being selfconsciously located within the Hindi lyrical tradition, to which its creators clearly wanted it to belong. Yet khyal occupies a very particular place which it shares with no other genre. It will be useful to contextualise khyal alongside its companion in Hindustani music, thumri. Thumri, which used to be the courtesans’ song form par excellence, can be divided into two major styles. Bol banav thumri, the form most current today, is characterised by its emotional expression using short phrases. The older form of the genre, bandish thumri, tends to be faster, usually set to the 16-beat tental. Its lyrics are more narrative in scope than its modern counterpart, and may have less emotional scope. Many bandish thumris have found their way into the khyal repertoire, where they are sung as fast compositions in the second part of performance. When we look at fast, or drut, khyals therefore, we may actually be analysing texts that have moved around the genres of Hindustani music. There are in fact songs that even today are known as either thumri or khyal depending on whom one asks. It is very likely that even in the nineteenth century certain compositions were known as both thumris and khyal, depending on the context of performance. A song performed by a well-known male vocalist may have been called a khyal; that same song sung by a courtesan would have been considered a thumri. Thumri occupies a small world: its inhabitants are few and their experiences are limited. We hear of village girls, mothers-in-law, Krishna’s flute, the joy of union, the pain of separation. Especially the latter is essential to many thumri compositions. The khyal universe is infinitely larger, as I will show in detail later on in this paper. A text describing a

girl’s fine attire sits alongside a composition dedicated to Allah. A yogi, with his magical power acquired through meditation and abstinence, can be described in one text with deference and devotion, and in another composition with fear and intrigue: what might happen should he decide to unleash his hidden vigours? Advice may be given to young lovers, sitting on a swing, enjoying love’s early flusters. The seasons are described, birds and other animals, awaiting the arrival of the monsoon. And so it goes on. What, then, makes a good khyal? Though in both thumri and khyal the basic linguistic matrix is certainly Braj Bhahsa, this presupposition needs to be both clarified and modified. As the lyrics are inevitably short it is difficult to reach any definitive conclusions as to their linguistic provenance. In most dialects of Hindi the clearest markers of identity are found in the verb forms, and these are often used in an extremely simplified form in both these musical genres. For example, auxiliary verbs and gendered endings of participles are commonly absent, giving karata rather than kartī hai or karati āhi. Furthermore, as lyrics invariably change in performance, Khari Boli forms are making inroads, though there is also an awareness that too much influence of the modern language is not conducive to creating the right mood. A good khyal, therefore, is one in which there is enough space for elaboration but also plenty of scope for mystique. A text with a neatly delineated narrative and an abundance of modern Hindi would be too narrow in scope, and much too prosaic. Khyal texts need to be adaptable and creative, containing short phrases which can be used for melodic elaboration in performance. Most of all, khyal texts need to evoke a world which is known to the audience, yet cannot be described within the constraints of the short text on which the musical performance is based. Khyal songs are known as bandishes (bandiś) because they are to some extent ‘fixed’ (which is what bandish means). They are the one aspect of the musical tradition which is handed down relatively intact. The ‘bandishness’ of bandishes is located in the lack of space to improvise and the fact that the words recur on a more or less predictable basis. It does not follow, however, that a certain song will always be performed in exactly the same manner. Far from it— the enormous fluidity in khyal songs, from performer to performer as well as from performance to performance, is one of the

salient features of texts that belong to a musical tradition, along with the fact that there is virtually no traceable evidence of historical changes and that the texts as they are performed consist of small units rather than poems in their entirety. In the present discussion we may pay special attention to the continuity in language use and subject matter between lyrics that are likely to have been composed at different times in different places, drawing on a variety of traditions. Khyal has been the prevalent genre of vocal music in north India for some two hundred years. Khyal songs are not defined by written representations, but are transmitted orally, committed to memory, and recreated through performance. The origins of khyal have recently been subject to a major revision by the historian Katherine Brown, but it will take some time before her findings (locating khyal much earlier than has commonly been accepted) will filter through into popular awareness. The received view is that khyal was developed (even ‘invented’) by the eighteenth century bīn player and dhrupad singer Niyamat Khan who composed under the penname Sadarang. Alongside him, the singer Firoz Khan, who wrote under the pseudonym Adarang, is also often mentioned as a major influence on the genre. (Sometimes Firoz Khan is referred to as Niyamat Khan’s nephew.) Both were court musicians at the Delhi court of the Mughal emperor Mohammad Shah (r. 1719—1748), who himself wrote compositions under the guise of ‘Rangile’. Like the later nawab of Avadh, Wajid Ali Shah, who was instrumental in promoting the arts and especially the vocal genre thumri in Lucknow, Mohammad Shah was a great supporter of the life of music at his court. Many of the compositions ascribed to Sadarang praise or at least mention his royal patron. A genre named ‘khyal’ is mentioned in much earlier accounts than those of Mohammad Shah’s reign. It is often linked—as are so many genres and forms—with Amir Khusrau Dehlavi, and Faqirullah’s Rāga Darpan of 1666 mentions a list of musicians at the court of the Mughal emperor Shah Jahan which includes two khyal performers. In this work many patrons of khyal are also mentioned, including Faqirullah himself. So although Mohammad Shah of Delhi was not the first major patron of the genre at all, he remains firmly associated with the promotion of khyal in popular

imagination, and it is believed that during his reign khyal began to usurp the place of court music traditionally held by dhrupad. Patronage is an essential aspect of the history of khyal (and one much too complex to be part of this paper). After Muhammad Shah the courts of Lucknow and Gwalior were the main employers of khyaliyas and their accompanists. The Gwalior gharana (‘school’) remains to this day one of the most prominent gharanas of khyal. Other styles of khyal singing associated with princely states are Rampur-Sahaswan, Jaipur and Patiala/Punjab. Although the boundaries of gharana are breaking down in the modern age due to the free movement of knowledge as a result of, for example, initially and most importantly, the advent of railways, then broadcasting and later the wide availability of cassettes, most performers still ally themselves with a particular style of singing belonging to a particular tradition. The actual compositions, however, seem to travel well across stylistic and geographical boundaries. Bandishes appear to be especially shared amongst the two major gharanas of Agra and Jaipur. There are many stories of compositions being given as dowries between musicians from the main schools. In addition there are of course the inevitable legends of thefts and death-threats made over the possession and use of compositions. These bandishes clearly constitute a major part of a gharana’s intellectual property. Although Sadarang and Adarang are considered to have composed many khyal bandishes (some of which have survived), there has been a tendency in the twentieth century (and probably earlier as well but we have no reliable record of this) for musicians to create their own songs. In recent history this may have largely been motivated by a political and educational impetus. In the early twentieth century the Gwalior gharana, for example, began to use poetry from the world of bhakti, including Tulsi and Mira, in their repertoire, in order to both ‘Hinduise’ and sanitise the existing tradition. It should be noted that at this time the Muslim ancestors of the gharana had almost entirely been replaced with Hindu musicians. Although it would be undesirable to establish a strict boundary between Hindu/Muslim musicians and the languages they may have used in their

compositions, the political milieu of the music world at the end of the nineteenth and the beginning of the twentieth centuries makes it virtually impossible not to be drawn in that direction. With some noteable exceptions, the Hindu/Muslim distinction follows along the same lines as middle-class/hereditary, educated/illiterate, ‘progressive’/‘backward’, and there was a concerted effort by Hindu reformers to retrieve and even ‘rescue’ the great art of music (which very much focused on khyal at that stage) from the hands of its traditional custodians who were considered lacking in moral credibility. It is therefore from the perspective of mapping out a ‘pre-divide’ landscape that the modern ‘Hindu equals Hindi vs. Muslim equals Urdu’ dichotomy carries no weight at all. Although the nature of my research, along with my belief in the relevance of aural evidence, has made me focus on recordings and live performances of khyal, for the sake of an historical overview I have been studying a number of Persian manuscripts held at the British Library in London. Although this can never be substantiated, I am working on the assumption that at least some of the compositions in the manuscripts were performed at some stage of their lives, and so they provide a significant comparison with some of the compositions performed today. I have looked at three manuscripts, all of which are unhelpfully not dated but are unlikely to be much more recent than 1820, possibly 1800. Only one manuscript is named, being the Kanz al-Mūsiqī subtitled Gulshan-i khyāl, dedicated to someone called Mo’azzaz Khan.3 The latter manuscript is headed as being in ‘bhākhā4 in the label on the front (presumably classified as such by an early cataloguer), and also describes the manuscript as being in ‘Hindavi with a Persian preface’.5 This manuscript is inscribed as belonging to Sir Charles Wilkins, who was a founder-member of the Asiatic Society in Calcutta (now Kolkata), and who died in 1836. The other two manuscripts describe khyal songs as being dar zabān- i Braj, that is, in the language of Braj.6 Working my way through the compositons in these manuscripts (and they are by no means all khyals, but also include rekhta, tappa, tarana and other genres), I was primarily struck by the continuity of language used in the songs then and in those performed today. In one of the unnamed

manuscripts,7 we come across Sadarang as well as Muhammad Shah and his penname Rangile. There are also compositions with the name of Sadarangile, a composite of the chāp of the main court musician and that of his royal patron. All these names also appear in songs that are still current today, but I have unfortunately not yet come across compositions in the manuscripts that are performed on the modern stage. As stated, Sadarang and Adarang are said to have composed many khyal bandishes, though it seems the latter’s have not survived in the same measure. We do not of course know whether compositions which include their names were actually composed by those poets or not.8 Sadarang’s bandishes have either been better preserved and continued, or he has inspired more compositions to be dedicated to his name. The latter is not an unreasonable assumption, as he is traditionally regarded as the ‘father of khyal’. Some musicians contend that khyals which were really composed by Sadarang can be recognised as such in terms of their musical value, and so in theory a ‘critical edition’ could be made. This view is unlikely to be substantiable, and to date as far as I am aware no attempts at sifting out the ‘real thing’ from any pretenders have been made. Research carried out so far suggests that roughly half of khyal compositions include a penname or chāp. In probably another fifty per cent of those compositions, it is impossible to trace who these supposed poetcomposers actually were. In any case, it is from a scholarly perspective not immediately relevant, as in a fluid and oral tradition changes invariably occur, even to such seemingly intrinsic aspects of a poem as its author’s name. Nevertheless, verifiable authenticity and historicity notwithstanding, most musicians and audiences find the stated provenance of compositions meaningful as it locates a song in a particular context or lineage. There are opposing views as to the extent to which performers interact with the compositions they are singing. I include in this paper some examples of songs ascribed to the singer on that particular recording, assuming a larger degree of engagement with the text when the performer is also the poet. I begin my textual discussion with some compositions by Sadarang and Adarang as they are thought of as the ‘founders’ of the genre. Before we begin our discussion of texts I start with some observations on perceptions of language in khyal.

Some authors on khyal make haphazard references to the literacy or otherwise of the composers and performers of the genre, especially with reference to their ability to negotiate Hindi. Bonnie Wade, for instance, writes that ‘when singers learn a cīz9 with a text in a language not their own —or even when they do know the language—there does not seem to be much effort exerted towards the text. The result can be incorrect grammar, the mixing of dialects, or words which do not seem to mean anything.’10 Such rather reductive observations on language are common among musicologists, whose primary focus and area of expertise is, after all, music. From a linguistic perspective, however, we must question the validity of pronouncing grammar ‘incorrect’, wonder what the boundaries of the dialects are so that we can term them ‘mixed’, and investigate if words ‘that do not seem to mean anything’ may not be—perhaps in their very meaninglessness itself—in fact highly salient. In other words, there may be more to the poetry of songs than meets the eye and ear. However, we must also acknowledge that some performers do indeed have a somewhat cavalier attitude towards the words of the songs they are singing, resulting in, if not anything immediately deemable ‘incorrect’, at least a considerable challenge for the transcriber. Vamanrao Deshpande, in Indian Musical Traditions, goes one step further than Wade and suggests: A cheej is said to contain an element of poetry. But it is never actually described as poetry or a poem. It is called only a cheej…. A cheej has musical value. Poetry possesses literary value. The immediate proof of this fact is that cheej-s have been written or composed by the singers themselves and not by poets as such…. To compose a song is not a poet’s business.11 Deshpande also sets out the various arguments for and against composing and performing texts in Marathi as the modern vocalists in Mumbai and surroundings apparently—according to him—have trouble with the idiosyncrasies of the language of khyal songs. Discussions with musicians and music scholars do indeed suggest that, perhaps seemingly paradoxically, the more educated vocalists in the urban

centres of India can find the use of non-standard Hindi (Wade’s ‘mixed dialects’) problematic. A prime example of linguistic confusion is the text in raga Todi: Laṅgara kāńkariyā̃ ji na māro, ‘Rascal, do not throw pebbles’. The word jina means ‘do not’, used in conjunction with an imperative verb (in this example māro, ‘throw’). Although this adverb is common in the dialects, it does not exist in modern standard Hindi. A well-educated vocalist and music teacher in Mumbai assured me the text read ‘Langara kankariyā̃ jina māro’ in which the adverb jina had become ji (‘Sir’) and na (‘not’). When I suggested the text should include the word jina, I was disdainfully told: ‘There is no such word’. As a foreign researcher I felt it would be eminently rude to point out to this well-respected and established music scholar that there are dozens of dialects of Hindi in which jina is a perfectly normal component of the language. By contrast, traditional musicians without scholarly inclinations have much less trouble with the dialects of the texts as they are already comfortable with Hindi in its nonstandard forms. In the twentieth century many singers wrote their own khyal compositions, often inserting their pennames. These bandishes are relatively easy to trace as we know who many of the chaps refer to, and so when a bandish is sung by that particular singer we can, at least in theory, assume a high degree of faithfulness to the text as intended. Nevertheless, we can never completely rely on the assumption that the person whose chāp is in the text actually wrote the entire composition. He may have added one or more lines and hence included his name in that portion of the text. He may have wanted to claim the bandish as his in the course of a particularly inspired performance. The ironic situation thus arises that a modern poet may ascribe his own composition to Sadarang to bestow a certain amount of ‘authenticity’, or conversely, a modern performer may insert his own penname in a bandish that already existed. Clearly the attribution of particular compositions to particular poets must be viewed with caution. Although bandishes by historical figures like Sadarang and Adarang are nowadays sung in most lineages, the more recent compositions can still be seen as the property of the descendants of that musician. Let us now turn to some phrases taken from the compositions in the Persian manuscript 2019 here transcribed in Roman and Devanagari:

8v Bhairav: मैं तोरी नगरिया चल बसूँ Maĩ torī nagariyā cala basū̃ I will settle down in your town Bibhas: पायलिया मोरी बाजी मोरी बाजू किस कर ठ† नूँ प्यारो तुमरे पासू॥ मोहॠ‡ ं मकर परोस दूजी लोग आन हँसू मोहीं कल न परत है ता तें पलक नहीं लागी ।। Pāyaliyā morī bāje Morĩ bājū kisa kara ānū̃ pyāro tumare pāsū || Mohẽ makara parosa dūjī loga ān hãsū Mohī̃ kala na parata hai Tā tẽ palaka nahī̃ lāgī || My anklets jingle and jangle. How can I make my way to you, love? The neighbours and other people come and laugh …(?) I don’t get any peace And no sleep either. 9v Bibhas (?): मोरी सास नन्दिया द्यौरनठ¿à¤¯à¤¾ जेठनिया़। पैंयाँ लागूँ मैं तोरी सैठ‚या मुहममद शाह ।। Morī sāsa nandiyā dyauraniyā jeṭhaniyāl Paiyā̃ lāgū̃ maī torī saiyā Mohammad śāh

My mother-in-law and sisters-in-law I fall at your feet, my lord Mohammad Shah. 22v Kedārā: अली निस दिन नींद न आवे न à¤à¤¾à¤µà¥ ‡ मैंको पिया बिन ना सुहावे सेज।। कौन जन्म की बैरी हो मेरी ॠ¤à¥¤ Alĩ nisa dina nī̃da na āve Na bhāve maī̃ko piyā bina nā suhāve seja || Kauna janma kī bairī ho merī Friend, sleep doesn’t come in the night, nor day I don’t like it without my lover, the bed doesn’t please me Who is my life’s enemy? 25r Bihag: बन बन बाजी बँसी हर की या मुरली मॠ‡ रो मन बस कर लॠ€à¤¨à¥‹ सदा तन मन धन सब तुमपर वारूँ ।। Bana bana bājī bãsī hara kī Yā murlī mero mana basa kara līno sadā Tana mana dhana saba tuma para vārū̃ In every forest Krishna’s flute resounds That flute has taken control of my heart, forever (forever = sadā; is this a reference to Sadarang?) Body, heart, wealth, I devote it all to you. 25r Jaijaivanti: हम जान लियो री प्यारी तठ¿à¤¹à¤¾à¤°à¥€ बात।

Hama Jāna liyo rī pyārī tihārī bāta. Beloved, I have come to know all about your ways. The compositions in this manuscript rely heavily on the formulae or formulaic building blocks that are an essential component of both thumri and khyal. Apart from including in its subject matter the usual suspects of the female in-laws, the neighbours, chiming anklets and Krishna’s flute, the texts are also full of stereotypical phrases such as kala na parata hai (I get no peace), paiyā̃ lāgū̃ torī (I fall at your feet), nī̃nda na āve (sleep doesn’t come), janama kī bairī (life’s enemy—usually a reference to the co-wife, or the mother-in-law), muralī mana basa kara līno (the flute has taken control of my heart), piyā bina na suhāve seja (without my lover the bed doesn’t please me), tana mana dhana saba tuma para vārū̃ (body, heart, wealth, I devote it all to you), etc. We find all of those, and similar phrases and conceits in the bandishes that remain popular on the modern stage. The reference to the ‘pleasing bed’, however, would not be as frequent as the other phrases as in the modern middle-class environment of the Hindustani music world such a phrase would be considered too risqué. Thus some singers might substitute this reference to the bedroom with other phrases. If we take a composite of the phrases above, and put them in a slightly more formal, metrically conventional framework, we could imagine something like the following stanza from Raskhan, a Vaishnav Braj Bhasha poet who probably lived in the mid-sixteenth century:12 बॠ‡ णु बजावत गोधन गावत ग ्वालन के सँग गो मधि आयो ।ॠ¤ बाँसुरी में उन मेरोई नाम स ग्वाल कॠ‡ मिस टॠ‡ रि सुनायो ।। ए सजनी सुनि सास के त्रासनि नंद के पास उसास न आयो। कैसी करों रसखानि नहीं हित चैन नहीं चित चोर चुरायो ।।

Beṇu bajāvata godhana gāvata gvālana ke saṅg go madhi āyo. Bā̃ surī mē̃ una meroī nāma sa gvāla ke misa teri sunāyo. E sajanī suni sāsa ke trāsani Nanda ke pāsa usāsa na āyo. Kaisi karaũ Rasakhāni nahī̃ hita caina nahī̃ cita cora curāyo. Playing his flute and singing a herding-song he comes with his cowherds amidst the cows. Through his flute, he sounded my own name on a pretext of [calling] the cowherd boys. O friend! Hearing it, for fear of my mother-in-law, even with my sister-in-law near by I could not catch my breath; How should I act Raskhan, I who have no well-being, No peace have I, for the stealer of hearts has stolen it. In the following bandish too, many of the above mentioned formulae make an appearance. Here is a bandish in raga Nat Bihag, as performed by Faiyaz Khan (Record number HSB527H-355; 1936 approx): Text 1 झन झन पायल बाजे मा

Jhan13 jhan pāyala bāje mā

जागे मोरी सास ननदिया

Jāge morī sāsa nanadiyā

और दौरनिया

Aura dauraniyā

हाँ रे जिठनिया मा।

Hã̄ re jiṭhaniγā mā

अगल सुने मेरो बठ—ल सुने जो

Agala sune mero bagala sune jo

जो सुन पावे सदारठ‚गीले ।।

Jo suna pāve sadāraṅgīle

My anklets jingle-jangle, friend My mother-in-law and sisters-in-law will wake up My neighbours will hear, They will hear, Sadarangile. The presence of the chāp Sadarangile in this text suggests this composition may be contemporary to the manuscript quoted from above. The motif of the jangling anklets is a common one. The noise of the small bells prevents the heroine from meeting her beloved—be he her husband or an illicit lover. Even married women do not have free access to their husbands in the early stages of marriage as too much intimacy between the newly-weds is frowned upon. The bride has to prove her allegiance to her new family first (best achieved by producing a son) lest she should coax her husband away from his blood-relatives, thus breaking up the joint family. This text, apart from providing us with a glimpse of reality, also shows some effective means of achieving poetic impact. The repetition of jhan in the first line may seem facile, but works well in performance. It also provides a nice continuity with following ‘j’ words jāge, jiṭhaniyā, and jo. The cluster of diminutive words in the references to the sisters-in-law is also an efficient way of establishing continuity: nanadiyā dauraniyā jiṭhaniyā. The use of rhyme and repetition within short phrases is an important aspect of khyal: agala sune mero bagala sune jo. The motif of the ankle-bells links this bandish with the dance of the courtesan, and so it connects the realities of the life of the dancer neatly with the glimpses of ‘real life’ that such compositions aspire to transmit. The Rajasthani poet-saint Mirabai frequently mentioned ankle-bells in her poetry, usually in the motif of herself as one who dances under the influence of ‘divine madness’. Mirabai’s text पग घुँघरू बांध मीरा नाची रॠ‡ (Paga ghū̃gharū bāndha Mīrā nācī re—Tying her anklebells on her feet Mira danced14) was in fact recorded as a khyal by the singer Onkar Nath Thakur.15 The next text also includes the name of Sadarangile but is written in a very different idiom. Like the previous text describing an aspect of the

reality of a married woman’s existence, this song tells us of the happy occasion of a visit by the son-in-law or daughter-in-law’s mother. The reference to the sautana in the penultimate line is unusual in its mood: here the co-wives seem to co-exist peacefully, as they are milake, together. In most texts the co-wife or rival is the enemy, who is the sole reason for the heroine’s lonely state. Text 2: Miyan ki Malhar, performed by D.V. Paluskar: (HMV-N-35289; 1947) आई समधन मोर रे घठ°à¤µà¤¾ मा ।

Āī samadhana mora re gharavā mā

सरबस मन बराई।

sarabasa mana barāī.

संदल गुलाब सानके

Sandala gulāba sānake

कर à¤à¤° गारन लगाई

kara bhara gāran lagāī

सब सौतन मिलके सदारंगीले

saba sautana milake sadāraṅgīle

को पियरवा बनाए ॠ¤à¥¤

ko piyaravā banāe.

My relative16 has come to my house Everyone’s heart is pleased. Mixing sandalwood and rose With hands full we embrace All the co-wives together Make Sadārangīle their beloved. I tried to discuss this text with Bala Sahab Poochwale, an elderly vocalist who is one of the main exponents of the Gwalior gharana. (The performer D. V. Paluskar is of the same lineage). After some deliberation he dismissed

the text as being ‘of the Chhattisgarhi type’. This reference to the area of Chhattisgarh17 explains—in Poochwale’s view—both the somewhat ‘folksy’ images presented in the text (it is certainly not a courtly subject matter, nor does it belong to the formulaic viraha-inspired repertoire) as well as the distinct use of language. In actual fact, there is nothing particularly ‘Chhattisgarhi’ about this text. The suffix -vā used in lines 1 (gharavā) and 6 (piyaravā) is indeed a feature of eastern Hindi, but more of the slightly more northern dialects Avadhi and Bhojpuri. For someone living in Gwalior (just over the border with UP in Madhya Pradesh) Chhattisgarh is simply an area to the east, and so his appraisal of the language of the text is not at all unreasonable. His location of the language of the text in ‘foreign parts’ indicates that musicians are not necessarily concerned with gleaning the exact meaning from the texts. This does not however have to equate to being disinterested. Bala Saheb admitted to having sung this composition, and inspite of an overall ignorance of the fine detail of the poetry, having liked the ‘natural sweetness’ of the lyrics. This suggests that an analysis of ‘mood’ is as relevant as one of semantics.18 In spite of the conciseness of the lyrics, the poetry of this song is more complex than it may seem. Many words have clusters of short a’s, providing a sense of internal rhyme even within a word itself: samadhana, sarabasa, sandala. This sense of rhyme is enhanced by the ubiquitous presence of sibilants. In addition to the ones just-mentioned, we find sānake, saba, sautana, sadāraṅgīle. The third and fourth lines show the same scansion in their respective opening statements: sandala gulāba sānake/kara bhara gāran lagāī.19 Even though we cannot analyse the text in terms of conventional meters, there certainly is metric support within the lyrics. As mentioned before, compositions ascribed to Firoz Khan ‘Adarang’ are not very common. Pandit Amarnath, in Living Idioms in Hindustani Music, suggests that ‘all the compositions of Adarang available today reflect Sufi thinking’ (1989:19). It is not entirely clear how this ‘Sufi thinking’ is represented but it may be that Amarnath is alluding to the spiritual flavour of the texts, such as the following famous bandish. Two aspects of this text impress most: one is the absence of visual references, the other the relative harshness of the sounds. The former may contribute to the ‘Sufi’ flavouring

of the text, in that it is more philosophical than narrative. The relevance of the sounds of a text reminds us of the major argument that arose around the end of the nineteenth century, when Khari Boli (Modern Standard) Hindi began to be used for poetry after a centuries-long domination of the poetic field by Braj Bhasha. Most musical texts conform to some extent to the belief that Khari Boli Hindi is not suitable for poetry because of the harshness of its sounds, and when we do see a relatively modern text it is indeed striking. Text 3: Raga Desi, performed by D.V. Paluskar: साँचि कहत है अ दारंग यह

Sã̄cī kahata hai adārańga yah

नदी नाव संजोग।

Nadī nāva sañjoga.

कौन किसी के आए जाए

Kauna kisī ke āe jāe

दाना पानी किसमत लाए

Dānā pānī kisamata lāe

यही कहत सब लोग।।

Yahī kahata saba loga.

Adarang says this truthfully The union of river and boat. Who knows what is ahead Fate will provide Everyone says this. This text reads somewhat as a limerick, with its ABCCB rhyme. The couplet inserted as lines 3 and 4 (CC) feels out of place, with its harsh sounds and tight rhyme scheme. The use of the word qismat in line 4 may be a further trigger for assuming Sufi influences. This Urdu word is not commonly used in khyal texts and makes much more frequent appearances in the poetic genre ghazal. Khyal lyrics do not often draw on Perso-Arabic

vocabulary. Considering that up to recent times most performers and composers of the genre were Muslim, the manner in which khyal bandishes fit into the mould of the Hindi lyrical tradition is remarkable though certainly not coincidental. The poet-composers of khyal (and other courtly genres such as thumri) seemed to have found their inspiration in the Hindi devotional and folk traditions, where the natural sweetness of the language and the portrayal of the lovelorn woman were ideally suited to being incorporated in the sensuality of the courtly performance. A further composition ascribed to Adarang rather invalidates Amarnath’s suggestion that all such bandishes show Sufi influences. Text 4: Raga Malkauns, performed by Amir Khan [HMV-EALP 1258; 1961] आज मोरे घर आइला बालमा

Āja more ghara āilā bālamā

करोगे अदारंग सों रंग–रलियाँ

Karoge Adāraṅga sõ raṅga-raliyā̃ .

अतर अरगजा सुगठ‚ध बसन पहरूँ

Atara aragajā sugandha basana paharū̃

फुलनव सेज बिछावॠ‚ँ चुन चुन कलियाँ ||

Phulavana seja bichāvū̃ cuna cuna kaliyā̃

My lover has come to my house today You will make merry with Adarang!20 I will wear perfume, essence and sweet-smelling attire I will scatter flowers on the bed, having picked the buds. Even in a tradition where the erotic is quite easily transformed into the spiritual it is difficult to put a religious spin on such an overtly physical text, with the blatant eroticism of the image of ‘picking buds’ leaving the audience in not much doubt as to what is going on.21

Although as stated earlier Sadarang’s and Adarang’s compositions remain popular, many twentieth-century singers created their own khyal poetry, or at least claimed such poetry as their own by inserting the penname by which they were known. The main exponent of the Patiala gharana Bade Ghulam Ali Khan (d. 1968) used the name Sabrang in many of the texts he sang. It is striking how well Bade Ghulam Ali Khan understood the requirements of the khyal idiom, as the following examples demonstrate: Text 5: Raga Megh, performed by Bade Ghulam Ali Khan (Hindustan Records H-910, undated) मोरे मदिरिया को आवो साजन

More mandiriyā ko āvo sājana

पैयाँ परत मैं तो बिनती करत आ जी।।

Paiyā̃ parata maĩ to binatī karata ā

हम संग करत मुख की बतिया

Hama saṅga karata mukha kī batiyā

जागत जागत बिन पॠ€ रतिया

Jāgata jāgata bina pī ratiyā

सबरंग पिया तोरी बिनती करत ।।

Sabaraṅga piyā torī binatī karata.

Lover, come visit my house I fall at your feet, implore you. He speaks with his mouth [but not with his heart] Every night without my lover I am awake Sabrang Piya, I plead with you. This text has considerable shortcomings as a written poem: the first line doesn’t rhyme with anything, and although there is an ABCCB rhyme scheme, the second and fifth lines end with the same phrase, a sure sign of

pedestrian poetry. As a khyal text, however, this song works wonders. We start with a nice alliterative phrase, both words of which firmly belong to the diction of non-standard Hindi: more and mandir, the latter word used in its pre-modern sense of ‘house’.22 Furthermore, mandir appears in its diminutive form mandiriyā, adding to the ‘folksy’ flavour of the text. The sam, the beat which receives most and continued emphasis in performance, falls on ā of āvo—‘come’—which happens to be the most crucial syllable in the sentence. The second line once again opens with alliteration, followed by a phrase which rhymes with the opening statement: paiyā̃ parata/ maĩ to binatī karata. We then move on to the second part of the song: hama and saṅga provide a measure of rhyme, in the next line the repetition of jāgata gives a sense of continuity as well as emotional power (‘I am awake and awake, this endless night’), and the attractive phrase bina pī ratiyā (the night without my lover) rhymes beautifully with mukha kī batiyā. The latter is an interesting phrase in itself, as one would expect a bāta (used here diminutively as batiyā), ‘speech’, to always come from the mouth (mukha). The implication therefore is that this speech is from the mouth only, i.e. not from the heart. The juxtaposition of speech of the heart and the mouth is not just assumed. Bade Ghulam Ali Khan himself employed the corresponding phrase mana kī bāta, as we can see in the next composition. Text 6: Raga Shuddh Sarang [HMV-EALP-1364; 1971] बेग दरसवा देहूँ सजनवा

Bega darasavā dehā̃ sajanavā

तुम बिन मैंका कल न परत ।

Tuma bina maĩkā kala na parata.

तुम ही जानत हो हमरे मन की

Tuma hī jānata ho hamare mana kī

बात सबरंग

bāta Sabaraṅga

कैसे कटे बरखा रुà¤

Kaise kaṭe barakhā

¤ के दिनवा।।

ruta ke dinavā?

Quickly show me a glimpse, lover Without you I get no peace. Only you know what’s in my heart, Sabrang How will the days of the rainy season be bearable? Occurring also in text 2, the use of the suffix -vā is commonly employed in khyal, though it is not used as formulaically as it can even if the texts were created in recent times. This usage is more evidence of khyal poets choosing their language consideredly. be in thumri. Once again, Bade Ghulam Ali Khan ‘Sabrang’ shows his awareness of the requirements of khyal in the two phrases of the opening line, which rhyme and which are metrically parallel. The use of the word maĩkā in the second line is noteworthy. This word, ‘to me’, is an older variant of the modern standard Hindi mujhko. Yet it does not appear in any dictionary I have come across, nor does it appear to be used in poetry other than thumri and khyal where it is relatively common. The first two words of the last line come with startlingly plosive noise: kaise kaṭe. This plosion is reflected in the meaning of the phrase, literally ‘how will they be cut’, i.e. ‘how will they be spent, how will they be bearable’—the subject being ‘the days of the rainy season’. The last word of the text once again employs the suffix -vā, taking us full circle. In addition to their poetic merit in terms of the khyal idiom, Bade Ghulam Ali Khan was clearly happy to embrace songs describing the standard motifs of the heroine pining for her wayward lover and the heroine being embarrassed by the antics of the male protagonist. The language of both bandishes is characterised by a sweetness more readily associated with thumri, a genre of which Bade Ghulam Ali was a famous and skilled performer. However, these lyrics also sit comfortably alongside the Sadarangile compositions we came across earlier. These texts, though written in an idiom that conforms to expectations of the songs of the genre, are not exclusively representative. The usual expectation might be that the poetic world inhabited by khyal is small, but the khyal universe is in fact much wider than these lyrics suggest. This

demonstrates among other things the wide base of its textual inspiration. This point is well-illustrated by the following list, which divides themes and motifs in khyal into four categories:

The khyal universe Protagonists who speak in the first-person voice the woman pining for her absent lover the woman complaining that her lover is spending time with her rival or co-wife the woman expressing adoration of her lover the devotee expressing adoration of his/her chosen deity the woman expressing happiness that her lover has come home the woman complaining that Krishna, or an unnamed male protagonist, is harassing her in the street the woman who wants to meet with her husband, but who is afraid of her mother-in-law disapproving of their intimacy; her chiming toe rings or anklets will give her away. the woman who wants to meet with her illicit lover, but who is afraid of her mother-in-law hearing her leave the house due to her chiming toe rings or anklets. the Muslim who sings of his religion, or love for Allah, or of God’s wisdom. the yogi who talks of mystical matters Mirabai, Krishna’s devotee, who is visiting from the world of bhajan Protagonists described by third-person observers Krishna, as divine lover, or as a young child

people from Krishnaite mythology, like the poor man Sudama, who comes home one day to find his hut transformed into a palace as a result of his kindness to Krishna other gods, like Shiva and Rama the yogi, who is not always approved of as he subverts the rules of societal propriety the heroine, all dressed up the hero, all dressed up the young couple, sitting on a swing, or being told how to conduct a love-affair birds and animals, primarily in songs that describe the seasons Situations and circumstances moral pronouncements: how to attain spiritual or musical excellence the seasons, which can bring happiness as well as sadness family life and festivals People hovering in the background the poet-composer, infrequently mentioned by name the protagonist’s friend, who is often addressed but never gets a starring role the patron, whose presence can be felt in the shifting mode of address Mohammad Shah ‘Rangila’, at whose court khyal was promoted in the eighteenth century. Sometimes his name appears in the composition Sadarang and Adarang, popularly received as the ‘inventors’ of khyal. Their names appear in a number of compositions, suggesting that they composed the song. (This is often doubtful and in any case hard to prove, or indeed, disprove.)

The wide variety of topics and conceits of this imagined universe shows that khyal has indeed become a confluence of a multitude of traditions and ideas. Any of these ideas can be referred to in the text, but the lyrics cannot be so powerful as to interfere with the musical performance—the text is subservient to the musical requirements of khyal. The following text is an illustration of the import of texts from the dhrupad tradition into khyal. Text 7: Raga Bahaduri Todi performed by Mallikarjun Mansoor, [HMVECLP-2384; 1969] हे महादेव देवन पठ¤à¥‡ पारवती पते

He Mahādeva devana pate

ईरवरेरा

Pārvatī pate Iśvareśa,

नील कंठपंचानन दुख हरन।

Nīlakaṇṭha pañcānana dukha harana.

वाम देव महादेव ठ—ंगा शिखर

Vāmadeva Mahādeva gaṇgā śikhara,

जटा जूट डमरू डम डम बाजे

Jaṭā jūṭa ḍamaru ḍama ḍam

सुख करन ।।

bāje sukha karana.

Oh lord of the gods, lord of Parvati, God of gods, Blue-throated and five-faced, remover of suffering. God of the left-handed path, great god whose matted locks are the peaks through which the Ganga flows, Who plays sweet rhythms on his damaru, and brings happiness. This text reads much like a Sanskrit stuti. It primarily consists of epithets of the god Shiva, and has no storyline or narrative scope. Although this song

was almost certainly not composed as a khyal, it is not unwelcome in the repertoire. Made up of small segments with much alliteration (deva devana, pate pārvatī pate, jaṭā jūṭa) and a simple rhyme scheme (dukha harana— sukha karana), it does in fact conform to the basic requirements of the genre. The same can be said for the following bandish, whose provenance is unclear but is sung as a khyal: Text 8: Raga Bhankar, performed by Faiyaz Khan [HMV-EALP-1365; 1971]. हे करतार ।

He karatāra

जागे पुकार।

Jāge pukāra.

तू करीम रहीम

Tū karīma rahīma

पाक परवरदिगार ॠ¤à¥¤

Pāka paravaradigāra.

O creator This call in me is bursting out. You are merciful and compassionate The pure provider. This composition relies for its cohesion on the rhyming of small segments: karatāra-pukāra, karīma-rahīma and alliteration in pākaparavaradigāra. The last word in turn rhymes with both line-final words of the sthāyī. Although in tune with its Islamic overtones this text draws on Perso-Arabic vocabulary (in the same way that text 7 relied on Sanskrit), its aesthetics nevertheless suit the needs of the khyal repertoire. Most of the bandishes discussed in this paper show a certain fluidity in language and even register. The song in Raga Desi (text 3) is compact with a solid rhyme scheme, and reads like a convincing piece of poetry on the page. Text 1 by contrast is much more typical of the oral tradition, as well as Faiyaz Khan’s idiosyncratic performance style. Text 2 draws on eastern Hindi for inspiration, consolidating the ‘homely’ atmosphere of the text. A

similar conceit occurs in text 4, where the eastern āilā in the first line contrasts with the Khari Boli karoge of the second. Bade Ghulam Ali Khan was a skilled song-composer whose poetry (texts 5 and 6) shows great awareness of how to use the language of khyal effectively and efficiently. In keeping with the nature of khyal performance, much of the rhyme depends on short phrases. He similarly relies on forms from eastern Hindi as well as on diminuitive words to create a suitable atmosphere. Songs that have made their way into khyal from the dhrupad repertoire tend to rely more heavily on a Sanskritised vocabulary. Text 7 is essentially a catalogue of epithets of the Hindu god Shiva. By contrast, text 8, an exhortation to prayer, uses epithets from the Islamic tradition. It is my contention that a dichotomy of Hindu vs. Muslim texts is either a result of khyal becoming a melting-pot of a multitude of musical and poetic traditions, or a modern construct whereby there has been an Islamic ‘backlash’ to the overt Hinduisation of khyal texts at the beginning of the twentieth century.23 None of the nineteenth-century manuscripts or pamphlets that I have come across suggest that khyal compositions could be divided into such polarised categories. On the contrary, it seems evident that that the poet-musicians of khyal were self-consciously locating themselves within the boundaries of the Hindi lyrical tradition. A modern model of a Hindi-Urdu division along religious and cultural lines was probably not applicable to the bandishes of khyal as they were originally conceived. Yet it is likely that as the musical milieu of the twentieth century did begin to follow the now almost stereotypical divide, a division between ‘Hindu’ and ‘Muslim’ bandishes was introduced. The penultimate text of this paper starts off like a fairly run-of the-mill love song, possibly addressing Krishna, the heart-stealer: ‘I am devoted to you’. However, by the time we get to the second part of the song, we are on more solidly religious terrain: ‘Fulfil the wishes of the wretched.’ Text 9: Brindabani Sarang, performed by Bhimsen Joshi [HMV-EASD1515; 1974]. Jāū̃ maĩ topai balihārī

जाऊँ मैं तोपै बलिहारी तुमही मेरो मन हर लीनो।

Tumahī mero mana hara līno.

गरीब की इच्छा पुरन कर दो

Garība kī icchā purana kara do

तुम तो गरीब नवाज़।।

Tuma to garība navāza.

I am devoted to you You have stolen my heart. Fulfill the wishes of the wretched You are the saviour of the wretched. It is of course debatable if the first and second part of this song were actually written together as one composition, but the fact that they were performed together as one item (and recorded on a disc) shows that playing with register is an acceptable, perhaps even desirable, aspect of khyal. There is a long-standing tradition of ‘floating antarās’, whereby the second part of one song can be used in another song as well. This does not always lead to textual consistency, but does provide performers with the freedom to adapt their performances as they see fit. In this particular instance, a text which starts off as a somewhat light-hearted lovesong concludes with the serious, and Perso-Arabic, phrase garība navāza. We can compare this couplet by the eighteenth-century sant poet Charandas24 with the opening lines of the khyal Text 10 below: तुम साहब करतार हो हम बन्दे तेरे रोम रोम गुनहगार है बकसो हरि मेरे Tuma sāhaba kartāra ho hama bande tere Rom rom gunahagāra hai, bakso Hari mere.

You are the Master, you are the Creator, we are your servants I am a sinner in every pore of my body. Spare me, my Lord. Text 10 तुम रब तुम साहिब करतार घट घट तू रहे जलथल घर बार। तुम ही रहीम तुम ही करीम गावत गुनि गंधर्व सुर नर सुरतार (सुरताल) ॥ Tuma raba tuma sāhiba kartāra ghaṭa ghaṭa tū rahe Jalathala ghara bāra. Tuma hī rahīma tuma hī karīma Gāvata guni gandharva sura nara suratāra. You are the Lord, you are the Master Creator, you are in all things, On water and earth, inside and outside. Only you are merciful, only you are compassionate The wise celestial beings, gods and men sing the tone and beat. Khyal then does not hesitate to draw on motifs and images outside of the world of Krishna-bhakti that is so commonly associated with the Hindi lyrical tradition. But its musical allegiance is all too obvious. In text 10, the praise of the almighty is turned into a musical pun in the very last line, in the double meaning of the word sura, meaning ‘god’ as well as ‘note’ (Sanskit svara), and the reference to the beat sūl tāl (suratāra) commonly used in dhrupad. The audience is humourously drawn back into the musical realm, confirming that khyal is comfortable alongside other poetry from the Hindi tradition, yet its point of reference remains firmly musical.

7 Looking Beyond Gul-o-bulbul Observations on Marsiyas by Fazli and Sauda Christina Oesterheld

U

rdu literature and Urdu culture are still seen predominantly from the perspective of the Urdu ghazal.1 In popular perception Urdu is often identified with ghazal poetry. Urdu cultural institutions all over the world regularly organise international mushairas which draw large audiences. For many practitioners and critics, the ghazal is the most prestigious genre. Open any literary journal in Urdu, and you are bound to find scores of ghazals. The unbroken popularity and prestige of the ghazal is not a problem in itself—even if it may arouse feelings of jealousy in prose writers and nonghazal poets. However, it turns into a serious problem when in retrospect the whole tradition of written and spoken Urdu (as far as we are aware of it) is reduced to ghazal poetry, and that too of a limited period and a specific style. As a result, the tendency to draw on Persian motives, Persian imagery and idioms in the eighteenth century come to be taken as representative of Urdu culture as a whole. Verdicts on the language divide caused by Urdu poets through their estrangement from the Indian environment are based on this selective perspective.2 This view of the literary tradition does not take into account that even in ghazals there has always been a variety of styles and, moreover, that this largely ignores other poetic genres. Although the range of ghazal themes is wide enough, still quite a number of topics, modes and voices remain outside its orbit. The ghazal perspective by and

large is that of the male lover/ mystic/philosopher etc; its space is the public sphere. There is hardly any room for female voices and for the familial. The few exceptions that do exist came into focus only in the second half of the twentieth century when serious attention began to be devoted to the female voice in Urdu poetry. The female voice in pre-classical and classical Urdu poetry has been studied from different perspectives and angles. Of late, the works by Carla Petievich, who has focussed primarily on ghazal poetry, come to mind. Of special importance in our context is her essay on Dakkani.3 In this work she discusses the female voice/female narrator as lover and bhakta—in line with the Indic tradition— exemplified by poems of the famous Dakkani poets and rulers, Quli Qutb Shah (r. 1580–1612) and Ali Adil Shah II (r. 1656–72). She clearly shows how the verses quoted by her draw upon the Gītagovinda, the poems of Vidyapati and of Mira Bai. Many of the familiar themes from both bhakti poetry and shringar rasa of non-devotional poetry are recognised here (the physical torment, the sense of betrayal mixed nonetheless with attestations of eternal devotion); and it is difficult to conclude otherwise than that they are deliberate transplants by sophisticated connoisseur-artists, artists who were open to experimentation across the ‘lines’ of a range of cultural/ aesthetic systems (Petievich 2000: 121). She continues: such ‘manipulations’ were successful precisely because they were based upon genuine compatibilities within the aesthetics of Indic love poetry and that of the Perso-Islamicate ghazal (p. 123). Petievich goes on to contrast the attempts of those Dakkani rulers to visibly adapt themselves to their Hindu environment, with the conscious process of differentiation between late-Mughal elites and their subject environments (p. 123). We cannot go into any detail here. Suffice it to point to another example of early Urdu poetry in the Indic tradition, this time from north India: the famous barahmasa Bikaṭ kahānī by Afzal Panipati (d. 1625).4 The genre of barahmasa remained very popular throughout the nineteenth century and was among the mass products of the printing presses

in the late nineteenth and early twentieth centuries. Though the echo of bhakti poetry remains present in this type of poetry, its overall character is secular, worldly and sensual rather than devotional. A somewhat different case is that of the frivolous, voyeuristic rekhti poetry, a playful sub-variety of the classical Urdu ghazal. This phenomenon has been studied, among others, by C. M. Naim and, again, Carla Petievich.5 Rekhti poetry, written almost exclusively by men, employs a largely imaginary, exaggerated ‘feminine’ voice to entertain and amuse a male audience. Here we are as far removed from devotional poetry as could be. But we also encounter an equally imagined and stylised female voice at the opposite end of the scale: in the deeply religious genre of the marsiya— a fact that has already been discussed at some length by Shantanu Phukan6. In the present paper this aspect of marsiya texts will be further elaborated and illustrated.7 Marsiya8 means threnody on the death of a relative, a patron or friend. In Urdu it has come to refer specifically to elegies on the martyrdom of Imam Husain and his followers at Karbala in 680 AD which are recited before a Shiite audience, though it is still used also in the broader sense. C.M. Naim defines the context of a marsiya recital: It is to be declaimed in a somewhat dramatic fashion at a majlis-e aza, i.e. a gathering of devout people seeking to obtain religious virtue by listening to the story of Imam Husain and his companions and by shedding tears over their tragic fate. Thus, it should be kept in mind that a marsiya in Urdu is primarily not meant for private perusal in moments of leisure, that it has a particular public-religious context, and that it also has a somewhat edifying goal beside the usual literary purposes that any good poetry has.9 There is no scope here to outline the historical background of the battle of Karbala. The following events form the basis for marsiya narratives: Together with a small group of followers and relatives, Imam Husain, the grandson of the Prophet, headed for Kufa on the tenth Muharram 61 AH At Karbala, about 25 miles north-west of Kufa, he was surrounded by the troops of his adversary Yazid which were about 4000 in number. They cut

him off from the water of the nearby Euphrates. When after some negotiations Husain refused to surrender, he and his two hundred (approximately) followers were killed. Husain’s head was sent to Damascus. His martyrdom is commemorated every year by Shiite Muslims. Poems in different forms are sung or chanted during the first ten days of the month of Muharram. Originally marsiyas were written in different verse forms. In this field, too, Dakkani works predate those of north India. Only in the nineteenth century, the stanza form of musaddas (six-line stanza) with the rhyme scheme aa-aa-bb, introduced for marsiya writing by Sauda, was finally accepted as the standard form of marsiya poetry. Sadiqur Rahman Kidwai comments as follows on the suitability of this form: The last two lines, as it were, heighten the effect of the emotion, impression or event which has been build up in the first four lines of the musaddas.10 With its rhyme scheme and the resulting sound quality this form also lends itself particularly well to public recitation. The two last lines of the first stanza may be repeated in every stanza (cc-cc-bb, dd-dd-bb etc.), thus creating coherence, an almost hypnotic and sometimes also dramatic effect, depending on the wording of this refrain. The technical term for this variety is tarji’ band. A marsiya may narrate the whole sequence of events or highlight a single episode to illustrate the heroism, the devotion and the suffering of Imam Husain and his followers. The descriptions of places, armour, dress etc are usually taken from an Indian milieu, even in later periods when the ghazal drew more heavily on Persian sources. Some important topoi/episodes are: the arrival of Husain and his group at the battlefield; the death of Imam Husain’s nephew Qasim (the eldest son of Hasan); the death of Ali Akbar, Imam Husain’s eldest son; the death of the Imam’s baby child Ali Asghar whose neck was pierced by an arrow when Husain took him into his arms; the martyrdom of Imam Husain; the transportation of the women from Husain’s camp to Syria, where Yazid awaited them. Each death is followed

by lamentations. It is especially here that women enter the field, mourning the dead. However, female voices in marsiyas are not limited to this function, and lamentations not necessarily form the bulk of marsiya texts. Thus, in a marsiya which Naim has analysed in his paper, only about one fourth of the stanzas were of a griefinducing (mubkī) nature.11 But obviously such mubkī verses are essential for the pathos and the cathartic function of the narrative. And it is here that the sound qualities and connotations of ‘Hindavi/ Hindi/Braj’, which are deemed to be particularly suited to the expression of loss and pain,12 come in. Let us therefore look at some examples. The first text I would like to turn to is a prose adaptation (with numerous verse translations inserted) of the Persian Rauẕatushshuhadā by Kashfi (809 AH) which was very popular in Muharram sessions. The Urdu version by Fazl Ali Fazli under the title Karbal kathā, also known as Deh majlis, was composed in Delhi in 1732–33 and revised in 1748–49.13 Listen first to the words of Qasim’s mother at the time of his departure for the battlefield: Gayā mujh āge se maĩ āh bajān hū̃ tujh bin Tū kahā̃ hai ki na jānū̃ maĩ kahā̃ hū̃ tujh bin14 You have left me, I am full of anxiety without you Where are you? I don’t know where I am without you. And Qasim’s bride: Kyõkar mujh bālī̃ ab jīvnā dubhar paṛi Bābā calā, dūlhā calā aur ko’ī na dhartī hā’e hā’e (p. 159) How hard it is to live now for a girl like me, Father has gone, my bridegroom has gone, no one is left to me. When Qasim’s dead body is brought to the camp, Imam Husain’s wife Shahar Bano mourns his death in the following words: Ab ranḍāpā mile hai dūlhan kū̃. Noctī hai vah apne abran kū̃ Dekh Qāsim tū̃ us kī tapan kū̃. Yād kartī har ek sukhan terā (p. 161) Now the bride has turned into a widow. She tears her jewellery apart,

Qasim, look at her torment. She remembers your every word. This is followed by the long lament/dirge of Qasim’s mother from which only a few lines can be given here: Hā’e kyā kyā maī tere bain karū̃. Terĩ s̱ūrat kī yād mẽ maĩ marū̃ Terĩ ā̃ khõ bhā̃ võ kū̃ yād karū̃. Yā yah ghunca sā mukh dahan terā Hā’e Qāsim ammā̃ ke jīv jigar. Ab calā mā̃ dukhiyā kū̃ tū̃ taj kar Qāsim ab terī bālī dūlhan par. Jam ho lāgegā yah maran terā (p. 162) Oh, how can I mourn you. I die for the memory of your face. Shall I think of your eyes and brows, or of your mouth delicate like a rose bud? Oh Qasim, your mother’s heart and soul! You are gone, leaving her in deep sorrow. Qasim, now that you are dead, your young bride, too, will die. The death of the young men and children is mourned not only by the women but also by Imam Husain himself. He, too, sheds tears and the tone and style of his laments resembles that of the women, but they are much shorter, in most cases restricted to two or four lines. Thus his suffering is shown to be deepened by the pain of losing the persons dearest to him, but his restraint and subservience to the will of Allah are equally testified. One of the most touching scenes in any marsiya is the death of Ali Asghar. His mother’s milk has dried up because Husain’s camp has been cut off from the water of the Euphrates. The Imam takes his son in his arms toward the river and pleads with his enemies to give water to his son who is dying of thirst. While he is thus pleading for mercy for his son, the baby is killed by an arrow from the enemy’s camp. Handing over the dead body to his wife, Husain says: Lo is t̤ ifli-i śahīd kū̃ ki ẖauẕi-i kaus̤ ar se serāb huā (p. 189) Take this martyred infant who has had his fill from the spring of Paradise. He weeps and utters only one more line. The mother’s lament for her dead son, however, fills four pages. It is obvious that women, and especially mothers, are allowed to indulge in their personal grief and to express it

openly to an extent which is not granted to men. Men, too, lament the death of the martyrs, as you will see later, but on a more impersonal plane, as a loss to the community. Fazli remarks in his preface to Karbal kathā that he created this (in his words) ‘commonly understandable Hindi’ version at the request of his patron and the ladies of his patron’s house. It was felt that most of the women were not able to follow the original Persian text. Therefore Fazli was ordered to provide the pious ladies and gentlemen of the house with a text that they could understand.15 What he felt to be easily understandable at his time very often needs comments for the modern Urdu reader. The edition under review contains numerous footnotes explaining unfamiliar words and expressions, usually of ‘Hindi’ or Sanskrit origin. In the passages quoted above, such unfamiliar words are: āb(a)ran (jewellery, as derived from Sanskrit ābharana), tapan (pain, torment), jam (death, derived from Sanskrit yama). Fazli’s work is today recognised primarily for its historical value as an early specimen of Urdu prose in north India. His prose is not without its own merits, capable of dramatic as well as emotional intensity. His verses, however, generally lack fluency and elegance when compared with contemporary Urdu poets, leave alone later masters such as Sauda and Mir. But apart from these reservations, it is interesting to analyse the type of language he deemed to be ‘a Hindi close to commonly understandable language for pious Muslim women and men’ (Hindī qarību-l-fahm-i-‘āma-i mūminīn-o- mūmināt). This language consists of a free mix of vocabulary from different sources. The dominance of one register over the other varies according to subject matter. Utterances of women contain the highest share of Indic vocabulary which seems to be close to the language used in actual practice of mourning/lamenting a death in the family, singing about the absence of a son or lover, the pain of living in a strange land, etc. We can hear the echo of folk songs sung by women, and of expressions used by women while describing how they bring up their sons, how they nurse them, feed them their milk, how they rock them in their arms or in the cradle, how they wean a baby, what dreams they have for their son’s wedding, etc. These passages take on a very intimate air which forms an effective contrast to the religious and martial/heroic pathos of the battle

scenes and the preparation for battle. In marsiyas these familial themes most often take on the character of dreams and aspirations linked with the son that are shattered by his death. It must be noted, however, that these passages, too, are the result of a long literary tradition transmitted in genres such as the barahmasa and in romantic epics in different north Indian languages. They contain a large number of stereotypical and stylised expressions which yet by no means diminish their effect. It is impossible to study the interaction between real women’s speech and literary adaptations of the female voice by male poets in a historical perspective because there are no records of women’s voices. Women’s speech as reported in Insha’s Daryā-i lat̤ āfat16 is based not on his own experience but on the Rekhti dīvān of Rangin.17 Moreover it is limited to individual words and idioms. It does, however, reveal a remarkable share of Indic vocabulary, and Insha does not find fault with that. He remarks: Students of language have to be informed that the language of the ladies of Shahjahanabad—as distinct as that of men—is more polished than that of all other ladies of Hindustan. They have their own language and style. Any word that has been adopted by them, be it from Arabic, Persian, Syrian, Turkish, Punjabi or Purbi, Maṛwari or Dakkani, Bundelkhandi or whatsoever, is Urdu.18 Thus his attitude toward the ‘purity’ of language is very simple: The languague usage of Shahjahanabad (old Delhi) sets the standard, and whatever is used in Shahjahanabad is chaste, proper, eloquent Urdu. There is a very pronounced pro-Shahjahanabad bias, but no bias as far as different source languages are concerned. Let us now turn to some excerpts from marsiyas by Mirza Muhammad Rafi Sauda (1713–80), one of the great Urdu poets who was famous for his mastery of language. He wrote a large number of marsiyas in different forms.19 I have selected one marsiya in the musaddas tarjiś band form mentioned above. The scene again is the mother’s lament after the death of Ali Asghar: Mā̃ As̱ghar kī din aur rain. Ro ro kartī haĩ yah bain. Hā’e re As̱ghar ibn-i H̲usain. Kis ko dekhẽ ge yah nain.

Tujh bin mere nūru’l- ‘ain Kyõkar ho is dil ko cain. Kartī thī maĩ tujh ko pyār. Hotī thī yah jān nisār. Tū merā thā gal kā hār. Kin ne ḍālā tujh ko mār. Tujh bin mere nūru’l-‘ain Kyõkar ho is dil ko cain. Kyõ tū paṛā us jā pah niḍhāl? Kyõkar lāgī tīr kī bhāl? Akhẽ khol e mere lāl. Mujh dukhiyā kā dekh aẖvāl. Tujh bin mere nūru’l-‘ain Kyõkar ho is dil ko cain. Rotā maĩ kis ko bahlā’ū̃. Dūdha thapak kar kis ko pilā’ū̃. Chātī āge kis ko sulā’ū̃. Jhūle mẽ ab kis ko jhulā’ū̃. Tujh bin mere nūru’l-‘ain Kyõkar ho is dil ko cain. Bacca re dil mẽ hai yah cā’o Mā̃ ko s̱ūrat ẓarā dikhā’o. Maĩ tujhko kahtī hū̃ ā’o. Yā to apne pās bulā’o. Tujh bin mere nūru’l- ’ain Kyõkar ho is dil ko cain. Cā’o thā dil mẽ terā biyāh, dil ke karū̃gī khā̤tir khvāh Dī na ajal ne furs̱at, āh. Kaise ho merā nibāh. Tujh bin mere nūru’l-‘ain Kyõkar ho is dil ko cain (pp. 226–28). Day and night, Asghar’s mother weeps and calls him. Oh, Asghar, son of Husain, whom shall these eyes now see. Without you, oh light of my eyes, How will my heart find peace. How I loved you, son, devoted my whole life to you. You were the garland round my neck. Who came and killed you? Without you, oh light of my eyes, How will my heart find peace. Why are you lying there so still? How did that arrow pierce your neck? Open your eyes, my darling child, and see what a state I’m in. Without you, oh light of my eyes,

How will my heart find peace. Whose tears can I now wipe? Whom can I feed my milk? Whom put to sleep in my arms? And whom rock in the cradle? Without you, oh light of my eyes, How will my heart find peace. Son, I’m full of longing, Just let me see your face. I tell you, either come, or call me to your side. Without you, oh light of my eyes, How will my heart find peace. My dreams of celebrating your marriage, fulfilling my heart’s desire, Were all shattered by fate. How can I now live on? Without you, oh light of my eyes, How will my heart find peace. Look at the perfect blending of words from different source languages! Can there be a more beautiful expression of a mother’s grief? I think that these verses are among the most touching and at the same time aesthetically most accomplished verses in the Urdu language.20 Here, again, the generic ties with the Indian poetic and epic tradition are obvious. The quoted passages also prove that Mir Babar Ali Anis (1802–1874) was not the first marsiya writer who depicted women in their different roles in the family, as mothers, sisters and wives, as Kidwai claims: ‘Before Anis’s marsiya we do not find such truthful images of women’s lives in Urdu poetry which tell us about their actual place in family and society. Anis shows women embedded in their familial and social relations that shape the way in which they experience happiness and sorrow and manage crises. The gender-based role and character of women in family and society has been depicted in so much detail and so skilfully only by Anis. The woman and family we come across in Anis’s marsiya was devised on the model of the life of the ashrāf (genteel, noble) Muslims of Avadh.21

Anis without doubt is the most popular and famous marsiya writer, but he is in no way the first to have dealt with women in relation to the members of their family on the backdrop of the events of Karbala. The theme occurs in all earlier marsiyas and was elaborated with exemplary skill by Sauda. Referring to another marsiya by Sauda, Phukan remarks: The loss which Sauda explores here is specifically a domestic and familial loss. It is, in other words, a loss unimaginable in the topography of the ghazal which, though also a poetry of loss, does not accommodate sorrow within the confines of the home. Thus, the marsiya locates itself in the specificity of gender, family relationships, and domestic settings. The emotions it explores are often quite unambiguously women’s emotions. And the speech which corresponds to an outpouring of such emotions is pointedly un-Persianised.22 Sauda’s treatment of the theme is unsurpassed, as the passages quoted above demonstrate, but changes in the language and in literary taste perhaps made his work sound obsolete to later audiences. Another marsiya by Sauda mourns the death of Imam Husain, the ‘king’ (śāh). See the following lines: Suno mujh bāt kahū̃ maĩ tum se, ro’o Gham hai śāh kā āj khuśī ko dil se dho’o Jisko jag mẽ log kahẽ the dīn kā sult̤ ān Sīs kaṭā ab us kā tan hai khūn se ghalt̤ ān Kahtā hū̃ maĩ ab tumhẽ mukh par khāk lagā’e Dīn-o-dunī ke tāj kā sar neze par jā’e Karte haĩ sab bain yahī hāth apne mal mal Ham bhī marẽ jahā̃ hai śāh kā maqtal Un bin jag mẽ suno hamārā kyā hai jīnā Jin bin sūnā āj huā hai śahar Madīnā Ek nahī̃ jītā rahā jin kā levẽ nām Vāris̤ to sab mar cuke ab jīnā kis kām Bakhtõ se jo kāú udhar se jīte āvẽ Phir Madīne mẽ jā kar mũh kis ko dikhlāvẽ

Pūche jo ko’ī āj kidhar tum khoyā sarvar. Kah kyā devẽ javāb use e dīn ke rahbar Lākh t̤ araẖ kā hamẽ ātā hai afsos Man ko apne rah ga’e yõ hī masos masos (p. 231). Listen to me, I tell you, cry! Today we mourn the King, forget all joy. He whom the world called King of Faith Rolls beheaded, spilling his blood. I tell you, smear ashes on your faces. The king of the world and faith’s head—on a spear. We all mourn him, wringing our hands. Let us die where the King was murdered. How can we live without him in this world When today Medina lies deserted. Nobody is left whom we could turn to With all heirs gone what shall we live for? If by chance anybody happened to survive How can he face the people at Medina? What shall he say, believers, when they ask, Where did you lose the Master? We are overcome with grief, Holding fast to our throbbing hearts. These verses are not pronounced by a female persona but by the poet who extols the community or the congregation to mourn the Imam’s death. Thus, it is a male voice we are hearing. But the language of these lines, too, is very simple, clear and straightforward. It is a functional language designed with the intended effect, to appeal to an audience which might not be so well-educated and comprises women and children. Perhaps this, compared with Sauda’s panegyrics, and somewhat more homely idiom which was also understood to be better suited to the occasion as it might have deepened the sense of community. Naim in an earlier work formulated this point as follows:

The marsiyas are written to be read before an audience in a majlis, and to make the listeners cry. To succeed in its chief goal a marsiya has to be firmly rooted in the intimate and the local. Though in purely linguistic terms the later marsiyas are as different from the Dakkani ones as the later ghazal is from the earlier, in their operative aspect, i.e., in the mubkī (tear-jerking) aspect, they are not that different. They too utilise indigenous value systems, emotions, and material culture to bring the tragedy closer to home to the devout audience.23 It is, however, difficult, or rather impossible, to ascertain how far an oral tradition of marsiya recitation contributed to the written form and whether north Indian Urdu poets at that time were aware of the Dakkani marsiyas. Following Naim’s line of argument that the death of the martyrs marks their transition into eternal life, that they lose the battle but win the war, one may argue that then there is no need to lament their deaths. Indeed, both men and women happily accept their fate and sacrifice their own or their children’s life for the just cause, but nonetheless they are also shown to mourn the martyrs by lamenting, weeping, beating their breast, etc. Outside the marsiya universe they may be criticised for their excessive grief, but within the marsiya context their laments are fully justified and crucial to the intended impact of the text. These intense emotional sequences are the main location of an informal style, of household expressions and intimate addresses and thus contain a high percentage of Indic vocabulary. Until the late nineteenth century, nobody seems to have bothered about the many tadbhava words and the expressions which were attributed to not only women’s, but also men’s speech in the marsiya genre. A gradual shift toward a more Persianised style seems to have occurred in the course of the nineteenth century, which is quite obvious when one looks at the language of the famous marsiya poet Anis. It seems that a general modernising and streamlining of language reduced the extent of ‘Hindi’ words and ‘women’s speech’ in contemporary marsiyas. My conclusion is based on very few examples. Thus I cannot claim general validity for it. Naim’s remark, however, that Anis and Dabir succumbed to the dictum of the purity of the language, corroborates this impression.24

CONCLUSION The perception of Urdu has for centuries been focused on ghazals, so much so that Urdu came to be identified with ghazal poetry. In our times, ghazal no doubt is the most popular form of Urdu literature, most easily performed and consumed, most likely to bring the author public acclaim and publicity. In the classical period, ghazal enjoyed the highest prestige and was the central subject of the literary discourse of connoisseurs. This fixation on the ghazal led to the eclipse of other genres and forms of Urdu poetry, and of the colloquial register of spoken Urdu, too. In narrative poetic forms and genres such as masnavi and marsiya, we encounter a much larger variety of linguistic registers than in the highly conventional, abstracted ghazal world. Despite the conventions and stylisations which are present in those forms as well, they more often draw on concrete details of everyday life and use appropriate language. (Just think of Mir’s description of his house in the rainy season!) Thus, we here find representations of different registers of language which are more or less absent from ghazal poetry and which belong to the common ground of a spoken Hindi/Urdu. Irrespective of the literary value we may attach to these forms, they all are a part of the vast body of Urdu literature, and in the case of marsiyas a very vibrant, functional part of it. What I have said about the ghazal mostly refers to the classical model. In contemporary ghazal poetry a remarkable shift away from the classical style toward a simpler, more colloquial language is taking place (not always to the better, though, since this often reduces the ambivalence and complexity of meaning considerably). Keeping in mind the widespread and highly publicised notions of the Persianising of Urdu in the late eighteenth century, I conclude as follows: We must not confuse prescriptive utterances in prefaces, treatises and critical remarks on poets with the actual practice of writing. The language used by individual poets has always been more varied than prescriptive texts would make us believe. The limited focus on the ghazal tradition, to the exclusion of other genres, has led to a narrow and limited view of Urdu literature and its language. Urdu writers of all periods have chosen those language

registers and styles which were appropriate for their respective genres and subject matter. Hence, the poetic diction may vary widely even within one author, leave alone different times and authors. When other genres are taken into account, it becomes impossible to uphold stereotypical notions about the language of ‘Mir’ (Taqi Mir) and ‘Mirza’ (Sauda). As far as marsiyas are concerned, Sauda’s style is much more simple, colloquial and ‘Indian’ than Mir’s. Language choices were not based on the etymological origin of words. There was no religious bias in the choice of words. Words of Indic origin could be freely used in religious texts for Muslims. If such words were put on the index, it was because they were denounced as uncivilised, plebeian or coarse, but not because they were identified with Hindu culture. Probably the background for disqualifying such words was the attempt to distinguish Urdu from the (as yet more powerful?) Braj tradition. The fixation on the ghazal has tended to exclude the world of the spoken language as it existed outside the literary discourse. Though influential writers and scholars such as Sirajuddin Ali Khan Arzu (1687/88–1755/56), Shah Hatim (1699–1783) and Mir Insha’allah Khan Insha referred to the language spoken by a certain social group in Delhi as the only reliable source for the standard of Urdu, very few specimens of that language have been preserved in writing. However, it is unlikely that the spoken language was much influenced by the prescriptive pronouncements mentioned above, rather the literary language was attuned to the current usage. At the same time, the liberal choice of words by any author does not necessarily preclude prejudices as far as language communities are concerned. Insha’s statements might have been intended as satire,25 but then they seem to have satirised attitudes which were current at his time (and continue to be so). The literary discourse of the ghazal excluded the sphere of the household, the family and especially of respectable women. Only occasionally do we get a glimpse of the language of that sphere in the

literary representation of women in marsiyas and masnavis. We need to be aware, however, that these specimens of women’s speech, too, are highly stylised and conventional. Opening up the canon of Urdu literature to genres and forms apart from the ghazal does provide a wealth of material for linguistic research and allows us to revise the image of Urdu literature as alienated from its Indian roots.

8 Changing Literary Patterns in Eighteenth Century North India: Quranic Translations and the Development of Urdu Prose

Mehr Afshan Farooqi

I would like to acknowledge the University of Virginia for providing me with funds to travel to India in the summer of 2004 when I collected material for this paper. I am grateful to Shamsur Rahman Faruqi for directing my attention to some of the textual materials related to this research. I am also indebted to my colleagues in the Department of Asian and Middle Eastern Languages and Cultures at the University of Virginia, especially Robert Hueckstedt and Griffith Chauseé, for commenting on drafts of this paper. I want to thank Richard Cohen for his thoughtful comments and invaluable help. Though I was unable to attend the conference which has served as the catalyst for the collection in which this article appears, I have benefited from the suggestions of Francesca Orsini, the editor of this volume. For the sake of consistency, Persian words have been transliterated as they are pronounced in Urdu.The iẓāfat is indicated with –e. Indic words have been treated as though they were written phonetically in the Urdu script.

I

n his history of Delhi College Malik Ram summarily dismisses the Urdu prose texts that were written in the eighteenth century as ‘elementary’ and/or ‘religious’. He argues that these prose texts were authored by scholars of Persian and Arabic and written with a view to improving morals and educating the readers about Islam. He further adds that these works are not good examples of prose, especially Urdu prose, especially Urdu prose, because they are laden with Persian and Arabic lexis and encumbered with Persian syntactical devices. Their prose is so dominated by metaphors and similes drawn from Persian poetry that it almost reads like poetry.1 According to Malik Ram and numerous other scholars (far too many to name here), modern Urdu prose is a ‘gift’ from the British (dar aṣl angrezõ kī den hai)2 and began with the establishment of Fort William College around 1801. The second important objection to early Urdu prose comes from Ram Babu Saksena, who in his history of Urdu literature trivialised the importance of early Urdu prose, especially of the Deccan, simply because, according to him, the rhythms of those texts do not sound like that of ‘modern’ Urdu prose i.e. ‘Urdu prose ki taraqqī yāfta shakl, jo āj kī nas̤ r kahī jā saktī hai.’3 Thus, modern Urdu scholars have rejected pre-Fort William Urdu prose for the following reasons: It is religious or moralising. It is laden with Perso-Arabic lexis. It is encumbered with Persian syntactical devices. Its rhythm is not modern. This outright rejection of early religious or moralising prose texts as not belonging in the history of any language, this essay will argue, is incorrect, misleading, even disingenuous. The example of early English prose springs to mind. Many scholars have pointed out that English prose begins with the translation of the Bible in the fourteenth century. As for the argument that the rhythm of those texts is not modern, one may ask: What exactly is meant by the rhythm of modern Urdu prose? No literary style springs fully developed like Athena from the head of Zeus; all styles and manners of language are parts of a continuum. Sometimes a new style emerges from the old and both strive side by side until one of them—generally the newer one — edges out the other. Or, it will also happen that the old style flows into a new style with no apparent sense of change. Here again one can cite

examples from English prose, like the prose of Shakespearean plays or that of the King James Bible. In addition to rejecting the early prose texts as ‘religious’, Malik Ram’s clubbing together the ‘few’ north Indian prose works like Fazl Ali Fazli’s Karbal kathā (‘The Story of Karbala’, circa 1732), Isvi Khan Bahadur’s Qiṣṣa-e Mehr Afroz-o Dilbar (‘The Tale of Mehr Afroz and Dilbar’, circa 1731–55) and Husain Ata Khan Tahseen’s Nau t̤ arz-e muraṣṣ’a (‘The New and Well-Decorated Style’, circa 1775)4 and declaring their prose as inferior (‘ham unhẽ ūrdū nas̤ r kā koi ā’lā namūna nahīn kah sakte’)5 is also difficult to accept, because each of these works has a different prose style. While Isvi Khan Bahadur’s prose has less overlay of Persian, Tahseen’s is highly Persianised. Karbal kathā is a mixture of the colloquial and the learned and is clearly intended to be delivered orally or read as an oral narration of an event which has a special religious importance for a large section of Muslims (see also Christina Oesterheld in this volume). In fact, what we have here is a panoply of styles and registers. True, none of them can be considered the source for ‘modern Urdu prose’. This has rather to be sought in the simpler style of early religious prose texts. This essay will therefore first briefly examine the three narrative prose texts mentioned above before turning to some eighteenth-century Quranic translations in order to assess the literary and historical value of these neglected examples of early Urdu prose. EARLY URDU NARRATIVE PROSE We know practically nothing about Fazli’s life except that he was from the Delhi area and was perhaps the son of Nawab Sharaf ‘Ali Khan, who was related to the Sayyids of Baraha. Some fourteen years later after composing Karbal kathā, he revised it (in 1748) before allowing it a wider currency. It appears that he was a ẕākir or a ‘speaker’ at Shia gatherings during Muharram, and expounded from the Rauẓat ush-Shuhadā (see fn. 4, also Christina Oesterheld in this volume). In his Introduction to Karbal kathā he mentions that the idiomatic Persian of the Rauzat ush-Shuhadā (Mausoleum of the Martyrs) went over the heads of the women in the audience and did not move them enough to bring tears, so he ventured to produce a Hindi

version peppered with suitable metaphors and colourful idioms that would be easier to understand and more effectively move the audience. While the Introduction utilises Persian vocabulary, the main text is rendered in a style closer to the spoken language. Here is a paragraph from the Introduction: Bandah-e ḥaqīr, pur taqṣīr hasbul irshād ūs qiblah gāh ke khāṣ Rauẓat ush-shuhadā kā sūnātā thā lekin m‘ani uske nisa’ aur ‘aurāt ke samajh mẽ na āte the aur fiqrāt-e pursoz-o gudāz ūs kitāb-e maẕkūrah ke basabab-e lughāt-e fārsī ūn kõ na rulāte the[…] Mujh aḥqar afqar kī khāt̤ ir mẽ guzrā ke tarjuma is kitāb kā ba-rangin ī ‘bārat-o ẖusn-e ist’ārāt-e hindī qarībul fahm-e ‘āma’-e mominīn-o momināt kīje to bamūjib is kalām-e bānīz̤ ām ke […] This humble, inadequate servant as per request from the mentor used to recite the Rauẓat ush-Shuhadā, but the ladies and the women in the assembly could not understand the meaning. And the emotionally stirring and moving parts from that remarkable book could not bring tears to their eyes because of the difficulty of the Persian vocabulary. It occurred to this humble one that if the above-mentioned book were translated in a colourful idiom with Hindi metaphors, it would be easily understood for the general public, both men and women.6 From the main text: Tab Hur āge ā, kahā, ay ‘Umar-o Sa’d! Husain sāth laṛegā? Kahā, hā̃ . Hur kahā, is laṛāī mẽ bahut tan be sar hoẽ ge. Phir Hur ghoṛā pherā maidān mẽ ā, apne bhāi kũ kahā, ay bhāt, mẽ ne bihisht ikhtiyār kiyā, aur ghoṛā uṭhā Haẓrat pās ā. […]7 Then Hur came in front and said, ‘O Umar and Sa’d! Will you fight Husain?’ [He] said ‘yes’. Hur said, ‘Many bodies will lose their heads in this battle.’ Then Hur turned around his horse, came to the battlefield and said to his brother, ‘O Brother, I have chosen Heaven.’ Then leading his horse, came near Hazrat Imam Husain […] As the examples show, Fazli’s Introduction is composed in a style and language very similar to that of Persian ornate prose; it was not meant for

oral delivery but simply for the edification of the ẕākir (narrator) or the general reader. By contrast the main text is clearly intended for oral delivery and contains fast action and high drama meant to elicit emotion in the listener(s). The manuscript of the Qiṣṣa-e Mehr Afroz-o Dilbar lay in oblivion for some two hundred years until it was discovered at Hazrat Ji’s dargāh at Gwalior in 1929 by Agha Haider Hasan. A scholarly edition was produced by Masud Husain Khan in 1966.8 According to Ali Jawad Zaidi, Isvi Khan was well versed in Sanskrit and Braj Bhasha and is well known for a commentary entitled Rascandrikā (1752) on Bihari’s Satsaī. This dastan is in two parts, the second being a Naṣīẖat nāmah or ‘Document of Advice.’ Isvi Khan’s prose vocabulary favours Braj Bhasha over Persian and Arabic. Aur ā̃ nkē̃n uskī kaun nargis kī munāsibat dīje to nargis to chashm-e ḥairān rakhtā hai aur iskī ānkhen to rasīlī hain. Aur khanjan mẽ kī jo munāsibat dīje cancalāpan kī, to is mẽ cancalāpan nahīn hai.9 And if you were to compare her eyes with a narcissus, its eyes are wide open [ḥairān] while hers are dreamy [rasīlī]. And, if you were to compare them for sauciness to a wagtail, they don’t have that kind of sauciness. The language is imaginative and poetic. It draws upon the well-known Persian-Urdu metaphor of the nargis but favours the Indic aesthetic sensibility of rasīlāpan over ḥairānī; it speaks of chanchalāpan and brings in the wagtail, not the bulbul. Isvi Khan’s language thus represents a style distinct from Fazli’s Karbal kathā. By contrast, Tehseen’s retelling of Qiṣṣa-e chahār darvesh in a heavy ornamental style produced many admirers because it was perceived as ‘literary’ and displaying the writer’s erudition. The prose of Nau t̤ arz-e muraṣṣ’a is also so highly Persianised that it reads like Persian sprinkled with Hindi verbs and particles. Here is an example: Jab t̤ a’er-e zarrīn bāl aftāb ke ne rukh bīch āshiyāna-e maghrib ke kiyā, aur baiẓa-e sīmīn māḥtāb kā batn, murgh-e mushkin, shab-e tār ke se namūdār hūā, mustaḥfiz̤ ān-o muḥārisān-e shahr ke ne ke bamujib ḥukm-e

valī is dayār ke māmūr the, darvāzah shahr panah kā masdūd karke rāh-e āmad-o raft, ṣādir-o vārid kī band kī.10 When the golden-haired bird, the sun, turned its face towards its home in the west, and the silver egg of the moon emerged from the scented black hen of night, the protectors and defenders of the city, in accordance with the order of the governor and the practice in those parts, secured the gateway of Shahr-e Panah and suspended the passage of newcomers and residents. Tahseen’s ornate prose is artfully constructed and relies heavily upon Persian idioms and Arabic vocabulary to make it acceptable as ‘high’ literature, while Isvi Khan’s qissa is a delightful mixture of the colloquial and the learned. Yet it was Tahseen’s qissa which became the exemplar for the dastan texts. Its popularity is partly responsible for the modern critical belief that all pre-Fort William Urdu prose was essentially in this ornamental style. I have quoted from the three most well-known eighteenth century Urdu prose works. There are a few more examples of fictional prose works in north India,11 but they all largely conform to the three examples examined so far. None of these prose styles was really suited for the explication of abstract, conceptual ideas and themes, and none of them seems to have been the model for works written for Fort William College. In other words, the question raised at the beginning of this essay still remains: if modern Urdu prose really began with Fort William, what was the model for the Fort William prose? Was the matter-of-fact, unembellished yet limpid prose style of the Fort William oeuvre accomplished by copying western models, or simply by weaning out Persian-Arabic lexis? Allegedly, Lalluji Lal, the putative originator of ‘modern Hindi’ prose, asked the same question; along with Sadal Mishra, Beni Narain and others, he decided to prepare a text in which tatsama and/or Braj Bhasha words replaced the Arabic and Persian vocabulary.12 Earlier, Gilchrist had tried to create ‘Hindostanee’ readers, which were hybrid texts resembling the Rekhta or ‘zabān-e Hindi’. Replacing Arabic and Persian with Sanskrit words could not have worked for Urdu. Because of the nature of the Urdu script, which effectively and

intimately impacts on the nature of Urdu phonology, and vice versa, Urdu could not, and even now does not, accommodate tatsama vocabulary. Also, replacing Perso-Arabic vocabulary with Sanskrit would have rendered Urdu almost unintelligible to all but very few speakers.13 As we have seen, by the eighteenth century Urdu had developed, mainly with the aid of Perso-Arabic and Hindi vocabulary, a sophisticated literary style in poetry and, as this essay will show, a workable style in prose that was used for expressing concepts, abstract ideas and fine distinctions or nuances.14 It may be worth recalling that Hindavi/Urdu had developed its own ways of creating new vocabulary through the free use of Arabic, Persian, and tadbhava words, resulting in new hybrids, derivative units which cannot be defined strictly as Arabic or Persian or Prakritic. This borrowing process was one of the special features of Urdu, which defined it and made it different from modern Hindi and modern Urdu, which coin neologisms purely from Sanskrit or from Arabic roots. For example, some of the most common words for good (khush, khūb, nek) and bad (bad), are all Persian. Urdu has combined them with Arabic words to generate numerous combinations which are neither Persian nor Arabic: khūbṣūrat, khūbsīrat, khush akhlāq, khush khalq, khush tab’a, khush mizāj, khush khabar, bad mizāj, bad kirdār, badm‘āsh, badtamīz. While some of these may be found in Persian, it is possible that they may have travelled from Urdu to Persian and not the other way around. Certainly many of them are indigenous to Urdu even if a Persian speaker could understand their general drift with some effort. Then there are words derived from Arabic or constructed on a notional Arabic model such as nazākat, faẓīhat, ẕahniyat—words which have both components of Arabic but are not Arabic, such as lā khairā, and other hybrids created by joining Persian and Arabic words according to Persian syntax (such as gulbadan, johardār, etc.). Urdu also combines Persian with Hindi suffixes to form verbs such as garmānā, sharmānā, āzmānā, farmānā; in the same way, Persian suffixes are attached to Hindi words to create new adjectives and nouns (e.g. samajhdār, thānedār, chaukīdār) according to the Persian syntax. EARLY TRANSLATIONS FROM THE QURAN

The early Urdu translations from the Quran were written in the eighteenth century, almost at the same time as the narrative texts examined above were being composed. The first, albeit partial, Urdu translation of the Quran is Shah Muradullah Ansari Sambhali’s Khudāī Ne’mat or ‘Godsend’ (completed in 1770).15 It is a tafsīr in what he calls ‘Hindi’ of the Quran’s thirtieth para (section) and the surah Fātiḥa. Two decades later Shah Rafiuddin and Shah Abdul Qadir each independently completed the task of translating the entire Quran. Shortly afterwards, Shah Muhammad Haqqani Marehara (1792) finished his own Urdu translation. In all, twelve translations, some full, some partial, were completed by different scholars in a period of a little over twenty years.16 The second half of the eighteenth century was a memorable period in many ways—the literary milieu of Delhi was full of extraordinary personalities. Shah Alam II (1759–1806), the ruler in Delhi, was himself a poet and writer and the author of the extended Urdu prose dastan, the Ajāibul-Qiṣṣas (1792). Under the nurturing attention of the poet-linguist-scholar Sirajuddin Ali Khan Arzu (1689–1756) and the Sufi poet Mirza Mazhar Jan-e Janan (1699–1781), there appeared two remarkable generations of Urdu poets, including such masters as Hatim, Abru, Yaqin, Dard, Qaim, Sauda and Mir. In taṣawwuf, or mysticism, Shah Waliullah (1702–1762) and his family were the leading personalities of time. His three sons Shah Abdul Aziz Muhaddis Dehlavi (1745–1846), Shah Rafiuddin (1750–1818) and Shah Abdul Qadir (1754–1826) were extremely influential in the fields of fiqh (Islamic jurisprudence), hadīth and Quranic studies. Shah Rafiuddin and Shah Abdul Qadir were the earliest Urdu translators of the complete Quran. The translation and diffusion of the Quran and other religious literature in the vernacular is usually set by scholars in the context of nineteenth-century reform movements and the debate around the use of print. How do we explain these Urdu translations almost a century earlier? Were they connected to a decline in Arabic and Persian learning—was Arabic being replaced by Urdu as a preferred language in the madrasas? Second, was there a significant change in literacy levels, so much so that a wider readership had been created and this new readership was more comfortable with Urdu? And, finally, what role if any can be assigned to the reformist

spirit that was clearly abroad among the Muslims of north India from the middle of the eighteenth century under the influence of Shah Waliullah and his school? The answer to the first question—whether these translations were connected to a decline in Arabic and Persian education and learning— must be answered in the negative. The classical education that was offered at the madrasas and maktabs continued to provide instruction in Arabic and Persian. Urdu was not among the subjects taught in these institutions during the time of the early translations, i.e. the last quarter of the eighteenth century, though, as one translator put it, ‘In this country, Hindustan, teachers who teach from books of Arabic and Persian, if they can explain the meaning in Hindi (Urdu) then the students can comprehend, otherwise not a word registers with them and they understand nothing’ (see Haqqani below). While Urdu was gaining status as a language of poetry and literary prose even among those persons who would normally use Persian as the language of refined social communication and learned discourse,17 there is no indication that it was gaining ground over Persian or even running parallel with Persian as a language of intellectual discourse. In his contribution to this volume Imre Bangha has collected evidence of Urdu/Rekhta verse by Sufis and by Persian poets at court from the time of Akbar, but it is safe to say that only in the eighteenth century did Urdu move from the domain of the oral and the occasional to that of the written and recorded, as had already been the case in the Deccan and Gujarat. Urdu’s literary base was developing, and problems of script, such as how to write the retroflex sounds, were being ironed out.18 It was slowly evolving as a written language. The answer to the second question—whether there was a significant rise in vernacular literacy in this period to warrant the hypothesis of a new readership, is more difficult. We have no concrete data on literacy levels in the eighteenth century. Internal evidence from the translations suggests that the intended audience was composed either by those who had some basic training in a madrasa but could not comprehend Arabic or Persian without a ‘Hindi’ translation (Haqqani), or by those men and women who had not had a formal education, i.e. knew no Persian or Arabic, spoke ‘Hindi’/Urdu and had perhaps learned to read it a little on the side, and were keen to read or

listen to expositions of the Quran in the vernacular (Shah Muradullah). While, as we shall see, at least some of these translations reproduced spoken language and were transcriptions of discourses, there is also a recognition that ‘these days the fashion is for Rekhta’.19 The third question holds perhaps the key to the motivation behind these early translations-translations which did not displace Arabic and Persian as the languages of Islamic learning and piety but expanded them to those Muslims who were not formally educated. A growing interest in providing access to the Quran gave rise to translations and commentaries explaining the sacred text through the medium of the vernacular. Besides, and this is important in the context of modern Urdu prose, these translations provide alternative instances to the kind of literary prose that was coming into use parallel to the language of poetry. Closer to the spoken language of the time, of all the styles current at this time, the prose of the Quranic translations appears to be an ‘intermediary’ stage of the language before it was divided into ‘Hindi’ and ‘Urdu’ by educational and political agendas in the nineteenth century. In fact, there are more tafsīrs (translation with explication) of the Quran in Urdu than in any language.20 Although there is some disagreement as to the exact number of complete translations in Urdu, there is no doubt that there have been at least 150, a number unrivalled in any other language, not even in Persian. While it has remained a tradition in all Muslim societies to learn how to read and recite the Quran in the original Arabic, the practice of translating, explaining and commenting on the Quran in local languages is a widespread phenomenon historically and geographically. The very first translation—that of surah Fātiḥa, the single most important surah of the Quran—was prepared in Persian during the time of the Prophet himself.21 The earliest complete Persian translation of the Quran recorded was in the third century hijri.22 In India, partial translations are reported to have been begun in zabān-e hindī in the sixteenth century, when a translator from Sindh was sent to Kashmir for this purpose.23 In 1604 CE, Hazrat Makhdoom Nuh Sarvarhalai Sindhi completed a Persian translation of the Quran. This translation precedes Shah Waliullah’s (1738) more famous translation by more than a century.

The Makhdoom’s translation is unique in many ways, most remarkably in the different interpretations of the Quranic verse Bismillāh ar rahmān ar rahīm that occur at the head of every Quranic chapter.24 According to Mustafa Qasimi, who has edited the translation, the verse was interpreted in context with the contents of the surah that it preceded. A few examples of Makhdoom’s Persian renderings are worth mentioning here: Banām-e khudā-e rozī dihanda va raḥmat kunanda In the name of God, giver of sustenance and the dispenser of mercy. Banām-e khudā-e ‘aṭā bakhsh khatā posh raḥmān raḥīm In the name of God, the forgiving, the concealer of faults, the munificent, and the merciful. Banām-e khudā-e raḥmat kunanda ‘ām-o khāṣ In the name of God the dispenser of mercy to the commoner and noble man. Makhdoom Sarvarhalai’s interpretive translation of Bismillāh employs the Persian word Khudā for Allāh. In Persian Khudā is the generic term for indicating Divinity and its main sense is that of Master; but it is used in many secular situations also, like deh khudā (master of the land) kad khudā (master of the house) nākhudā (master of the boat) and so on. The Arabic word Allāh precludes all such images. In fact the Arabic Allāh is a contraction of al ilah and signifies ’uniqueness'.25 Not only is the Persian word Khudā bereft of these nuances, its secular sense is very nearly ever present in the language; that is why its plurals like khudāyān or khudāigān are common, whereas there is no plural for Allāh. For this reason most translators, even in non-Quranic contexts, prefer to leave the word Allāh untranslated. The fact that Makhdoom Sarvarhalai freely uses the word Khudā for Allāh indicates that he would not have had any apprehension of having disapproval directed towards him for doing so. It also shows that the Persianate culture had so permeated the learned Muslim environment of that time and place, in this case Sindh, as to make the word Khudā synonymous with Allāh. This suggests a certain degree of liberalism in the environment and also in the tradition of Quranic scholarship, especially translation.

Translations were not frowned upon as individualistic approaches and interpretations, as long as they are not seen as infringing on basic rules of understanding and interpretation. An interesting and perhaps even more remarkable example of such an individualistic approach to Quranic translation is found in the fragments of the Quran translated into Avadhi (Purbi or ‘eastern Hindi’ according to the editor) by the nineteenth-century Sufi Maulana Shah Fazle Rahman Ganj Moradabadi (d.1895). He called his translation Manmohan kī bātẽ (Words of the Heart-Enticer) and freely refers to Allah as ‘Manmohan’.26 Such variety in translation even of a sacred text like the Quran underscores the possibilities of interpretation that are inseparable from the act of translation itself. Sarvarhalai’s Persian translation is relatively unknown today, especially if compared to the popularity of Shah Waliullah’s translation. Probably its individualistic and unconventional approach did not appeal to a populace that was shifting from liberalism to reformism and sectarianism as a form of cultural defence. (When Urdu began replacing Arabic and Persian as the language of religious as well as literary and elite discourse, there was a reluctance to encourage individualistic scholarship because there was safety in conservatism.) Unsurprisingly, the Urdu translation by Shah Waliullah’s son, Shah Abdul Qadir (ca. 1790), is so closely modelled on Shah Waliullah’s that it has often been described as an Urdu rendering of Shah Waliullah’s Persian. But even for later Urdu translators of the Quran, Shah Waliullah’s remained the model. Let us now look more closely at the authors, text and language of these early translations. Shah Muradullah belonged to Sambhal, a qasbah in the Moradabad district, a hundred miles to the east of Delhi, in what is now western Uttar Pradesh. He was a disciple of the eminent Sufi and poet, Mirza Mazhar Jane Janan of the Naqshbandi silsilāh and was himself a celebrated scholar with a love for poetry. In the dībācha or ‘Introduction’ to his tafsīr, Shah Muradullah explains the need for a Hindi (Urdu) translation. Lākhõ kaṛorõ musalmān jo hindī zabān bolte hain ‘arabī fārsī zabān sẽ kuch vāqif nahīn hain. Jin logõ ne mat̤ n-e qurān paṛhā thā ūn ko qurān kī āyatõ kī tafsīr hindī zabān mẽ ma’nī sunāonā thā. Sunne vāle mard

bībīyān bahut ikhlās se, shauq se sunte the. Is ḥāl mẽ ba’ẓe ikhlāṣ mandõ ne kahā, jo ham ko bhī qurān kī āyatõ kī yeh tafsīr ma’lūm rahtī, sūratõ ke ma’nī yād rahte to kyā khūb bāt hotī. Ŭs vaqt Allāh ta’ālā ne ūn ke sāce shauq aur ikhlās kī barkat sẽ is ‘ājiz bande khāksār ke dil mẽ yeh bāt ḍāl dī jo is hindī taqrīr ko kaghaẓ ke upar qalamband kar, likh kar unko paṛhā dīje to dīn ke’ulūm kī bātẽūn ke upar khūb t̤ araḥ se ma’lūm ho jāvẽ, yād rahẽ, kām āvẽ, acce ‘āmāl karne kā shauq baṛh jāve. Lakhs and crores of Muslims who speak Hindi don’t know any Arabic or Persian… For those who had read the Quran in Arabic (but did not understand Arabic) I used to explain the meaning (tafsīr) in Hindi. The listeners, both men and women were very attentive and eager… Several among the audience of sincere listeners said how wonderful it would be if they could understand the Quran and know the meaning of what they read… Thus God granted their prayer: their earnestness and honest desire to understand what they read put this idea in the head of this humble one to pen my discourse in Hindi so that it could be read…27 Shah Muradullah further adds that there was a corpus of Quranic studies in Persian but no scholarly work was undertaken in Hindi: Those whose native language was Persian followed the lead from Arabic and wrote thousands of books on theology, hadīth, tafsīr, sharah for the sake of fellow Persian speakers and for those who knew Persian, and thus brought all the knowledge into Persian. Not one of the buzurgs or ‘ālims [learned persons] wrote a religious treatise in Hindi.28 ****** The thirtieth para of the Quran comprises short surahs that deal with a variety of subjects. This allows a lot of variation in the prose of the tafsīr. On occasions the tone is one of narration, at other times it is of explication. The first paragraph quoted below is an example of narrative while the second is explicatory: Sūrah qadr makkī hai. Is me pānch āyaten, tīs kalima, ek sau bārah ḥarf hain. Tafsīrõ mẽ likhā hai ke sabab nāzil hone kā is ṣūrah ke yeh hai-jo Banī Israīl kī qaum me ek shakhs thā, Sham’ūn un kā nām thā. Ba’ẓe

kahte hain Maimūn nām thā. Hameshā din ko rozah rakhtā thā aur rāt ko namāz paṛhtā thā.29 Surah Qadr was revealed in Mecca. It has five ayats, thirty kalimas and one hundred and twelve words. According to the tafsirs the reason for its revelation was the following: In the tribe of Bani Israel, there was a man by the name of Shamun. Some say his name was Maimun. He always fasted during the day and prayed through the night. Kaho yā Muẖammad un logon ke javāb mẽ jo Khudā ta’ālā kī ṣifāt pūchte hain. voh Khudā ek hai; apnī ẕāt mẽ, ṣifāt mẽ, aur koī dūsrā uskā sājhī nahīn. koī uskā sharīk nahīn. Na ẕāt kisī kī Khudā kī ẕāt hai, na ṣifāt kisī kī Khudā kī sī ṣifāt hai.30 Speak, O Muhammad, in response to those who ask about the attributes of God. There is only one God, unique in being, in qualities and there is none his equal. God is the Eternal, the Absolute. Such limpid and lucid prose was being written in Urdu for the first time. This prose is not encumbered with Perso-Arabic syntactic devices and vocabulary, yet it is eloquent and free in its choice of synonyms (sājhī, sharīk). Even though it favours a mode of oral narration, the text is elegant enough to be styled as literary. Shah Muradullah’s language certainly has the capability to explain polysemic Arabic lexis. Khudāī Ne’mat was first published in Bengal in 1831 and went through several editions in the nineteenth century. It undoubtedly prefigures the style that was developed and disseminated from Fort William some thirty years later and found keen practitioners in Sir Syed Ahmad Khan and others who followed his lead. But despite its elegant prose this tafsīr finds no place in Urdu’s canonical literary history and hardly figures in accounts of the beginning of Urdu prose. Urdu literary historians have dismissed it on account of it being a religious text in their self-conscious attempt to project a secular image of Urdu. Shah Rafiuddin, the second son of Shah Waliullah, was born and brought up in Delhi. He was a Sufi scholar like his father and older brother Shah Abdul Aziz, and the author of thirty-seven books on various aspects of

Islam. His translation of the Quran is the first full Urdu translation and was completed around 1788, though it wasn’t published until 1840. An interesting aspect of this translation is that Shah Rafiuddin did not pen it himself—it was prepared by a disciple, Syed Najaf Ali Khan, from the notes the latter took while reading the Quran with the Sufi master. Shah Rafiuddin read through the draft and corrected it where needed. A similar practice was followed for compiling Tafsīr-e Rafi’ī, a commentary cum translation of surah Baqar.31 This lesser known tafsīr employs colloquial language and has an oral style. The translation is free flowing and not word for word; explication follows the translation. Shah Rafiuddin’s younger brother Shah Abdul Qadir also translated the Quran into Urdu. His tafsīr, which is presumed to have been finished shortly after Shah Rafiuddin’s, around 1790, was published under the title Mūẓih alQurān, in 1829. The significance of the title is that it contains a tārīkh or chronogram commemorating the date of its completion. One wonders why a second translation was undertaken soon after the first, if it represents a different approach and whether Shah Abdul Qadir acknowledged and/or benefitted from his older brother’s work. Qadir’s translation is known to be an Urdu replica of his father’s Persian one, but it is significantly different from Shah Rafiuddin in that it flows very well. It pays attention to Urdu syntax, instead of reproducing the original word for word. In the dībācha, Shah Abdul Qadir discussed his technique and thoughts on translation: It is not mandatory to provide a word for word translation because the Hindi syntax is quite different from the Arabic. If one follows the Arabic word order it will be difficult to understand the meaning from the translation. Instead of Rekhta more Hindi synonyms have been employed to make it easy for comprehension of the general reader.32 For example, Arabic lexis has been replaced with words from a common parlance: chamak instead of ziyā’ (luminescence), ujālā for nūr (radiance/light), gorī for ḥūr (houri) and so on. Common Persian words have also been substituted with local colloquial ones, such as pūch instead of pursish (reckoning); pīche in place of bad (afterwards), changā for ṣeḥatyāb (healthy), etc. A comparison of the translation of surah Laḥab (surah 111, from the thirtieth para) will illustrate this point:

Shah Rafiuddin: Halāk hojiyo hāth Abu Laḥab ke, aur halāk ho voh; na kifāyat kiyā usko māl uske ne, aur jo kuch kamāyā thā; shitāb dākhil hogā āg sholay vālī mẽ, aur jorū uskī uṭhāne vālī lakṛiyon kī, bīch gardan uskī ke rassī hai post-e khajūr kī se.33 Shah Abdul Qadir: Abu Laḥab ke hāth ṭūṭ jāyen aur voh barbād ho jāye; na uskā māl uske kām āyā, na uskī kamāyī ab baiṭhegā deg mārtī āg mẽ; aur uskī jorū sar par liye phirtī īndhan; uskī gardan mẽ rassī hai mūnj kī.34 Summarising: (Both hands of Abu Lahab will be broken, and he will perish. His wealth and what he earns will not avail him. He shall soon burn in the flaming fire; his wife the bearer of fuel. Upon her neck a halter of twisted mūnj rope.) While Shah Rafiuddin seeks to maintain the rhythm of Quranic recitation, Shah Abdul Qadir’s clear Hindi/Urdu reads like modern Urdu prose. Both use common tadbhava words and words of common parlance, but Shah Abdul Qadir even more so than his brother: he prefers barbād over halāk, kām ānā over kifāyat, āg over sholay, mūnj kī rassi over post-e khajūr and so on. Shah Muhammad Haqqani’s tafsīr and translation, ‘Ināyat Rasūl ki (Gift of the Prophet), was completed in 1791–92 and remains unpublished. This tafsīr too begins with a helpful introduction.35 Among other details pertaining to the translation, Shah Haqqani, like the previous translators, also gives an explanation for preferring Urdu over Persian as his choice of language: Not to mention Arabic, now readers of Persian are hard to find. In this country, Hindustan, teachers who teach from books of Arabic and Persian, if they can explain the meaning in Hindi (Urdu) then the students can comprehend, otherwise not a word registers with them and they understand nothing. This humble, poor and unassuming one realising that now the fashion is for zabān-e Hindī, it is the spoken language and it is called Rekhta employed it for interpreting/translating the Quran…36

Each one of the dībāchas emphasises the decline in Arabic and Persian learning and the need to produce texts for a readership comfortable with the spoken vernacular, zabān-e hindī. It appears that almost all the translators were keen to write in what we can call the ‘spoken language’, because the objective was to maximise comprehension. Zabān-e hindī is the elusive ‘Hindostanee’, the metonym of Gilchrist et. al., the colloquial Urdu that continues to be spoken but that could not survive as a single common literary language for various reasons that are beyond the scope of the present paper.37 Closer to the spoken language of the time, the prose of the Quranic translations appears to be, among all the styles current at this time, an ‘intermediary’ stage of the language before it was divided into ‘Hindi’ and ‘Urdu’ by educational and political agendas in the nineteenth century. Less tied to formal literary models and moved by the urge to communicate abstract religious ideas in a simple language, these translations also provide a clue to the development of modern Urdu prose that has been neglected by modern historians of the language. A comparative analysis of the Shah Muradullah, Shah Rafiuddin, Shah Abdul Qadir and Shah Haqqani translations of the same single surah will further illustrate their use of language and rhythm, their ability to convey meaning and also the uniqueness of each translator’s style.38 In order to exemplify the notion of what ‘the sound or rhythm of modern Urdu prose’ should be I am including an excerpt from a contemporary translation, by Maulana Fateh Muhammad Jallandhari, that has received both popular and scholarly acceptance. I am providing additional language samples by the way of an appendix, with a view to afford those scholars interested in linguistics material for analysis. Shah Muradullah:39 Qasam hai āftāb kī aur uskī roshnī kī z̤ uhā ke vaqt kī. Aur qasam hai chānd kī jab sūraj ke pīche nikaltā hai. Chaudhvīn rāt kā chānd. Aur saugand hai chānd ki jo nikaltā hai rāt ko sūraj ke pīche. Aur qasam hai din kī jab roshan kiyā khudā ta’ālā ne āftāb ko din mẽ. Aur roshan kiyā din ne rāt ke andhiyāre ko.

Tārīkī jātī rahī aur duniyā kī zamīn sab roshan ho gayī. Aur saugand hai rāt kī jab chipā letī hai, dhānp letī hai apne andhiyāre se sūraj ko, zamīn ko, duniyā ko.40 Aur saugand hai āsmān kī aur jisne banāyā āsmān ko Aur qasam hai zamīn kī aur jisne use banāyā. Aur qasam hai har ek nafs kī, har tan kī, har jān kī, aur uskī jisne har ek tan ko, jān ko rāst kiyā. Shah Rafiuddin:41 Qasam hai sūraj kī aur uskī dhūp chaṛhne kī Aur chānd kī jab āve uske pīche [yā’nī sūraj ḍūbne ke sāth nikale] Aur din kī jab usko roshan kare Aur rāt kī jab usko dhānk leve Aur āsmān kī aur jaisā usko banāyā Aur zamīn kī aur jaisā usko phailāyā Aur jī kī, aur jaisā usko ṭhīk banāyā Phir samajh dī usko dhitāī kī aur bach chalne kī. Shah Abdul Qadir:42 Qasam hai sūraj kī aur uske dhūp chaṛhne kī. Aur qasam hai chānd kī jab ke sūraj ke pīche lagā āve. Aur qasam hai din kī jab roshan kar dikhāve jahān ko. Aur qasam hai zamīn kī aur jisne use bichāyā hai. Aur qasam hai badan-e ādam kī aur jisne usko durust kiyā aur pūrā banāyā. Shah Haqqani: Aur qasam hai sūraj kī aur uskī roshnī kī. Aur qasam chānd kī jis vaqt nīche jāve sūraj ke. Aur qasam din kī jis vaqt roshnī kare. Aur qasam rāt kī jis vaqt chupāve roshnī ko din kī.

Aur qasam āsmān kī aur uske banāne vāle kī. Aur qasam ẕāt-e ādam kī aur uskī jisne durust kiyā aur banāyā us ādam ko. Maulana Fateh Muhammad Jallandhari: Sūraj kī qasam aur uskī roshnī kī Aur chānd kī jab uske pīche nikale Aur din kī jab use chamkā de Aur rāt kī jab use chipā le Aur āsmān kī aur us ẕāt kī jisne use banāyā Aur zamīn kī aur uskī jisne use phailāyā Aur insān kī aur uskī jisne uske a’ẓā ko barābar kiyā Phir usko badkārī se bachne aur parhezgārī karne kī samajh dī Aur jisne apnī nafs ya’nī rūḥ ko pāk rakhā voh murād ko pahũchā Aur jisne use khāk mẽ milāyā voh khasāre mẽ rahā.43 Shah Muradullah’s version is elaborate and tends to be interpretative while Shah Rafiuddin’s, even though it is some twenty years later, is slightly archaic; it is also compact and prosaic. A slightly different form of verb is used: chupāve instead of chipā le, jāve, āve, etc. Shah Muradullah’s language is graceful and flowing. He shows imaginative flight and prefers idioms which have a poetic content. His lexical repertoire has more variety and he chooses to offer multiple glosses for singular Arabic terms such as nafs, for a fuller and comprehensive explication of meaning. Nafs (not to be confused with the Persian nafas) in Arabic means ‘being, inner being, soul, existence or person.’ Shah Muradullah’s interpretation, Aur qasam hai har ek nafs kī, har tan kī, har jān kī, (And by each and every soul, each and every body, each and every being) brings out all aspects of the meaning of nafs. Shah Rafiuddin’s choice of jī (aur jī kī) as the gloss for nafs in a certain sense can be said to subsume all these meanings.44 The problem with the phrase aur jī kī and other similar phrases is its awkwardness in terms of the Urdu syntax—clearly Shah Rafiuddin is ignoring the Urdu and staying close to the Arabic original. Shah Abdul Qadir employs badan-e ādam (body of man) and limits the meaning of the verse to the concept of

human being as a perfect machine. Shah Haqqani’s gloss of ẕāt for nafs (aur qasam hai ẕāt-e ādam kī) is more inclusive than badan except that ẕāt is generally not used in Urdu.45 Maulana Fateh Muhammad Jallandhari’s widely read, more contemporary translation glosses nafs as insān, i.e. human being (aur insān kī), giving a more clear-cut meaning but one devoid of mystical nuances. CONCLUSION Since the first stirrings of the nationalist movement in South Asia, both scholars and politicians have taken various rhetorical positions regarding the origins and the respective individuality of the ‘two’ languages, Hindi and Urdu. Many have placed rather too much emphasis on the role played by the texts written at Fort William College, and this has resulted not only in historical misrepresentations but has also diverted scholarly attention from a closer investigation of the period before Fort William College. It is an oversimplification to argue that what happened at Fort William College should be given the sole credit for the origin and establishment of modern Urdu prose, shorn of rhetorical flourishes and flamboyant vocabulary, that is, an Urdu prose which was capable of conveying model ideas and subjects of a more literary nature. The translations-cum-commentaries that I have presented here have been read and heard by countless numbers of people and must have played a formative and normative role in popularising Urdu prose. The prose of each one of these tafsīrs tends to minimise borrowing from Arabic and Persian, indicating that to write such prose was possible, at least by the 1770s. They even share the style of naming or giving a title to the tafsīr in an informal tone that sounds exactly like the ‘modern Urdu’ with which we are familiar. Scholars have focused on the language programmes of Fort William College because chronologically they appear at a critical time in our understanding of the foundations of British hegemony, cultural as well as political. To a certain extent we fulfill an expectation which is founded on the assumption that in order to secure control and expand their influence the British facilitated the invention of what they call ‘Hindostanee’ prose and further politicised the situation by promoting ‘Hindi’ and ‘Urdu’ prose only

a few years later. While it may not have been a ‘diabolical plan’ inspired by guile, the Fort William College programme should be understood to have existed to serve emphatically the interests of the East India Company, which apparently recognised the need to standardise a vernacular for administrative purposes. This understanding was quickly transformed by the British reading of Indian society, which tended to emphasise differences between social hierarchies and religious communities, for the sake of serving moral and legal arguments and policy-making. It is a fact that under the direction of Governor-General Wellesley (1798–1805), Gilchrist and his hand-picked munshis actively promoted the production of vernacular prose texts and their use in education. To this extent, scholars are correct to identify the early period of Fort William College (ca. 1800) as the ‘smoking gun’ in the promotion of Hindustani (read: zabān-e Hindī) prose. My argument is that our historiographical research on this subject requires that we look past this episode and that we can and must assume there were antecedents to what developed full-blown in the nineteenth century into ‘modern Urdu prose’. The same can be said for ‘modern Hindi prose’. The problem is even deeper in that scholars of Urdu literature have narrowed their own perspective on the history of Urdu, excluding the prose of the eighteenth-century Quranic tafsīrs as not inherently literary. But the ornate and heavily Persianised language of eighteenthcentury narrative texts such as Nau t̤ arz muraṣṣ’a cannot be said to be the precursor of modern Urdu prose. By contrasts, the Quranic translations linked the spoken and literary language and led the way towards the ‘Fort William’ style of Urdu prose. My point is that prose language, when deployed in the context of translating and commenting on important religious texts, in particular a universal text such as the Quran, is a form of literary prose. If it happens to precede other forms of literary prose, then it should be recognised as participating in the development of those later literary prose forms. The complicated history of both Hindi and Urdu in the nineteenth and twentieth centuries obscures the rather politically uncomplicated situation of Shah Abdul Qadir’s zabān-e hindī of the last quarter of the eighteenth century. It is important to realise that the terms Rekhta, Hindi and zabān- e Hindī were used synonymously as well as in distinction to one another by such eighteenth century authors as Shah Muradullah, Abdul Qadir and Shah

Muhammad Haqqani. This suggests that a measure of fluidity existed at that time in the way people thought of and referred to the spoken language which was being adopted for the purposes of writing prose and poetry. It is also possible that there were differences in the level of sophistication and scholarship among these authors, which may account for their equating or not equating these terms. As noted earlier, the vernacular was not taught in preparatory schools such as maktabs and madrasas. Similarly, the vernacular was not formally studied in Sanskrit tols and paṭhśālās.46 Thus, the literati were not trained to form or hold a particularised viewpoint with respect to the name and identity of the non-classical or informal language. The eighteenth century was a self-assertive century for literature and culture in its own right. It deserves to be seen through the eyes of the selfaware, remarkable generations of poets, Sufis, scholars and linguists who participated in the emergence of Urdu as the preferred literary language of Delhi and the Gangetic plain, and who fostered the literary culture that took pride in nurturing the cultural ambience and context in which Urdu prose was accepted, finally to become an ideal literary vehicle. We can not ignore and we must understand the forces at work after the 1850s which set the stage for separate ‘Hindi’ and ‘Urdu’ literary cultures, utterly transforming the meaning of the terms themselves. That such paradoxical developments occurred should neither divert our attention from nor confuse our understanding of the situation in the eighteenth century and earlier, a period in which the identity of the vernacular as deployed in various literary contexts was not so self conscious.

9 Networks, Patrons, and Genres for Late Braj Bhasha Poets Ratnakar and Hariaudh Valerie Ritter

I

n the late nineteenth and early twentieth centuries, Braj Bhasha poetry was positioned at the brink of the Hindi-Urdu divide. While its supporters held this literary dialect to be the ‘Hindi’ poetic language of the past and present, Hindi/Nagari supporters objected that Braj Bhasha could not be the poetic language for the future and gradually made the choice to embrace poetry in ‘language near to the language of men’,1 Khari Boli. This language was however Sanskritised—so as not to resemble Urdu poetry—and morphologically standardised—so as to resemble the lingua franca Hindustani. Braj poetry in this period of the divide offers us further complexities for our assessment of the cultural changes wrought by the Hindi movement. This essay focuses on two poets who shared the distinction of persisting to write in Braj Bhasha in a period when Braj Bhasha was receding, apparently forever, as an option for modern poetry. Jagannath Das ‘Ratnakar’ (1866–1932) was known as the ‘last great Braj Bhasha poet’; Ayodhyasingh Upadhyay ‘Hariaudh’ (1865–1947) was a poet in both Braj Bhasha and Khari Boli.2 Sustained by patronage structures in flux, in a world of transforming literary values and politics, the works of these

authors highlight the renegotiation of the power of Braj Bhasha poetry in terms of canon. In many ways, Ratnakar and Hariaudh stand out as archetypal poets of the late Harishchandra and Dvivedi eras. Their demographic profiles were typical of supporters of Hindi: small zamindars, teachers, qanungos (registrars of landed property in a subdivision of a district), brahmins and Agrawal merchants.3 Both were members of the Nagari Pracharini Sabha (hereafter ‘NPS’), the association that promoted the use of (Deva)Nagari and, later, of sanskritised Khari Boli.4 This same association instigated the first major collection and editing of manuscripts in Nagari, most of them in Braj Bhasha, thereby establishing the Hindi literary historical tradition as we know it. As Imre Bangha points out in his contribution to this volume, this history largely disclaimed Urdu and connected with Braj Bhasha and other old Hindi dialects as the forebears of modern Hindi literature in Nagari. Both Hariaudh and Ratnakar attended events at the NPS headquarters in Varanasi and both published under its auspices. Both attended poetic gatherings of this and other literary associations, on occasion as sabhapatis (conveners/keynote speakers) and well-known elder literary figures. Both were educated in Persian, appreciated Urdu poetry and, to a degree, wrote on similar subjects in the same poetic genres. Because of these commonalities, evidence about their social networks and their shifting avenues of patronage can inform and complicate our view of the dividing north Indian literary landscape. This landscape was, I find, dividing along several lines at once, with overlapping avenues of patronage, so that Braj Bhasha poetry came to inhabit a mixed oral and print culture with metamorphosing definitions of the literary public. Literary people met within both the ‘traditional’ spheres of court and temple, and newer institutions of education and civic associations, both private and official. The new mixture of patronage and association seems to have produced new sorts of ‘cultural legislation’, through which modern Braj Bhasha poetry, in linguistic and formal features, ultimately waned. Frederic Pincott described the linguistic and literary views of Hindi supporters of this period succinctly, in an unusual London publication of 1889, ‘Khaṛī Bolī kā padya: The Poetical Reader of Khari Boli, compiled

by Ayodhya Prasad Khatri, of Muzuffarpore’ and edited with a preface by Pincott.5 In one of the first modern literary Hindi books printed in England, Pincott noted that the sixty or seventy million people of north India speaking various dialects of the Hindi language, have been accustomed for generations to express their poetic conceptions in the dialect of Braj…. The sacred character of the [Braj region] caused the dialect of Braj to be employed in the literature of the North-West Provinces; and nearly the whole of the literature formerly consisted of religious poems. ‘Urdu or Hindustani’, in contrast, is a ‘cultivated’ idiom to the extent that ‘the natives of the North-West, of which it is supposed to be the vernacular, are compelled to learn it in schools, like a foreign tongue’.6 This volume purported to display the ‘non-Islamitic form of Hindi’, Khari Boli, which had languished without government patronage, having been esteemed by Indians themselves to be an ‘uncourtly idiom of the vulgar’. Pincott then cited Ayodhya Prasad Khatri’s object to induce his countrymen to abandon the use of the archaic Braj dialect in their poetic effusions, and to persuade those who favour Urdu to use Nagari. This programme, Pincott opined, would remove the greatest obstacle to the intellectual development of northern India. The absurdity of talking and writing prose in one language, and poetry in what is virtually another language, is beginning to make itself felt’ (ibid: vi) Braj Bhasha as a medium of poetry formed an ‘anomaly’ that is ‘inconvenient’, and therefore Babu Ayodhya Prasad was endeavouring to confer a substantial boon on his countrymen, by inducing them to clothe all their ideas in one common form of speech, written in one common character (ibid: v–vi). Besides, ‘the tone of all these compositions is unexceptionable’, and therefore the work would help to ‘raise the character of Indian literature’

(ibid: viii). Thus, in the following decades Braj Bhasha continued to be considered a barrier to ‘progress’ and to the consolidation of Hindi vs Hindustani/Urdu. Furthermore, the erotic content of Braj Bhasha poetry was now considered problematic in a Victorian elite context that both condemned and fetishised this feature of Indian culture.7 Of course, the matter of Braj Bhasha was even more complicated for its authors, who, like their forebears, loved its poetry. While Braj Bhasha publishing did function somewhat to reclaim a nonUrdu literary heritage for Hindi, and therefore buttress a political divide between Hindi and Urdu, evidence put forth below suggests that poets of Braj Bhasha considered it a unifying poetic mode across other inexorably widening divisions. The patronage structures that produced and reproduced Braj Bhasha poetry were complex and similarly motivated by a certain nostalgia for tradition, but the very sense of the past that Braj Bhasha invoked may have excluded it from the future of canonised modern Hindi poetry.8 In this paper, Hariaudh and Ratanakar, friends who together straddled various social structures supporting the new Hindi literary scene, will exemplify the erstwhile relevance of modern Braj Bhasha to ‘the Hindi public sphere’, as Francesca Orsini has put it.9 ORIGINS Hariaudh and Ratnakar were both connected with the Varanasi literary world, and both were among the last well-known Braj Bhasha poets. However, they came from rather different backgrounds. Hariaudh was born in 1865 into a brahmin family in Nizamabad, near the district town of Azamagarh in the eponymous district somewhat north of Varanasi.10 His family, along with others in Nizamabad, had traditionally maintained ties to what would now be called Sikh practices, such that Hariaudh often maintained a ‘Sikh’ appearance, albeit one less strictly prescribed than in the present day. Next to Hariaudh's home lay a gurdwara commemorating Guru Amar Das, and its anteroom was the site of poetic gatherings, including the recitation of all of the kinds of poetry of north India: Urdu shers, chaupais, dohas, kavittas, savaiyyas and samasyapurtis in the old literary dialects of Hindi, and of course the poetry of the Guru Granth

Sahib. Hariaudh’s grandson recounts in his biography of his grandfather that, while still a boy, Hariaudh proved himself in this company with his exegesis of a verse of Sant Kabir. Hariaudh attended the local school, passed the Hindi intermediate test in 1879 at the age of fourteen and received a scholarship the same year to study English at Queen’s College in Varanasi. However, he returned home after only six months and continued studying with his uncle, a pandit who performed brahminical duties in Nizamabad. With this uncle, he read Sanskrit and Punjabi and studied prosody and astrology. He also studied Persian extensively with a local maulvi, to the point that his ability in Persian equalled his ability in Sanskrit. He began his career as a schoolteacher in the local district school at Azamgarh in 1884 and passed the normal exam (i.e. the teacher- training exam) in 1887. Appointed as a sub-district qanungo in 1891, his writing career began to flourish; before his retirement in 1923 he would publish over twenty works, mostly poetry and criticism. Azamgarh being a district town and centre of handicraft production (black pottery and zari saris, to name just two, both associated with the sizable Muslim community), he did not lead an insular life, although his location was clearly somewhat remote from the cultural hubs of Varanasi and Patna. Hariaudh associated somewhat with the well-known local Urdu scholar Shibli Numani and presumably other Urdu litterateurs of Azamgarh. An English sahib supervised Hariaudh at the District Collector's Office of Azamgarh, the Nizamabad gurdwara had probably some pilgrimage and administrative traffic from Punjab, and a local postmaster tutored him in Bengali. The traffic through Azamgarh was not insubstantial and, like the sarais that dotted the district roads, this town formed a real, though underestimated site of exchange for intellectual goods. Ratnakar, by contrast, emerged from the centre of the elite literary life of Varanasi. He was the son of a wealthy Agrawal merchant of Varanasi. His great-grandfather was a court-connected gentleman of Lucknow with ancestral links to the court of the Delhi Mughals. Ratnakar’s father, Purushottam Das, was a friend of Bharatendu Harishchandra, so Ratnakar literally grew up within Harishchandra's mitra-mandal, as biographers call the wealthy and influential upper caste merchants and intellectuals who comprised Harishchandra’s circle of friends. Purushottam Das had expertise

in Persian, and had memorised the Quran, attesting to this mixed cultural/intellectual milieu.11 Ratnakar studied English (and presumably Bengali) at the Bangali Tola High School, and then, later, English at Queen’s College. As would be expected, he also studied Persian and Urdu. As a young man, he wrote some Urdu poetry with the takhallus ‘Zaka’. He was intimately connected to the burgeoning print culture of the era, as a friend of Devakinandan Khattri, whose serialised novels of the early 1890s, in the ‘Hindustani’ register, would subsequently fall out of favour for ‘Hindi literature’ in coming years. His father and the Harishchandra-mandal encouraged his taste for Braj Bhasha poetry. This also is not surprising, considering their predilection for courtly arts and for Krishna devotion. Ratnakar thus emerged from a cultural milieu of a merchant class of growing economic and cultural power, which embraced high cultural forms of many varieties. These included not only the company of poets and circulation of texts in Braj Bhasha, Urdu and Persian, but British-style public organisations for educational or scholarly pursuits, and a burgeoning press.12 Ratnakar passed his BA in English, and began an MA in Persian and an LLB degree, but was unable to complete either. POETIC NETWORKS AND THE CASE OF SUMERSINGH SAHABZADE Poetic networks formed along predictable channels: personal, associational, political. However, in the late nineteenth century, they had a notably distinct texture. Along with the informal communities, such as the group at the gurdwara in Azamgarh, the group at the Gopal temple in Varanasi and the ‘salon’ of Harishchandra, other venues for interpersonal connections involved literary associations, which attracted colonial administrators and somewhat reflected the structures of British associational models. Additionally, publishing houses themselves served as social spaces for literary activity, as was the case with Hariaudh’s publisher, the Khadgavilas Press of Bankipur in Patna (a publisher of Harishchandra’s works, and the major press in Bihar excepting the government press at Patna, and henceforward KVP) and probably others in north India. The KVP’s lending library in Patna was a centre for interaction between a variety of people:

Rajput princes who invested in the company, authors from Varanasi and the surrounding regions, and the same Britons who joined Hindi literary associations and published on Hindi literary and linguistic matters in the British press, e.g. Sir George Grierson13 and Frederic Pincott. The Press hosted kavi-goṣṭhīs (gatherings of poets), as did the associations to which its authors belonged.14 The relationship of the Press with its authors was not one of simple monetary exchange. For example, in place of monetary royalties from the sale of books, authors were paid in kind with luxury goods and formal exercises of respect and honour when they visited the Press offices. While certainly the small market for books in Hindi played a part in this mode of payment, this (perhaps courtly?) mode of payment through gifting persisted well into the twentieth century. A certain personage connected with the press, who traversed various sites of literary performance throughout North India formed a common link between Hariaudh and Ratnakar. The older Baba Sumersingh Sahabzade (1847–1903)15 was a native of Hariaudh’s Nizamabad, which had a longstanding Udasin Nanakapanthi community.16 A friend of Bharatendu Harishchandra of Varanasi and the mahant of the Harimandir gurdwara in Patna from 1885, Sumersingh took on Hariaudh—then known as Ayodhyasingh—as his protege, even introducing the boy to Harishchandra himself.17 Thus also, when in Nizamabad, Hariaudh attended the meetings at the gurdwara. Sumersingh shared his library with the young Hariaudh and eventually got him a publisher, Ramdin Singh of the Khadgavilas Press. Sumersingh’s poetic name was ‘Sumer Hari’, from which the young Ayodhyasingh is reputed to have derived his own nom-de-plume, ‘Hariaudh’.18 Because Hariaudh resided largely in Azamgarh and Nizamabad, Sumersingh figures as an important influence from the relative metropolises of Patna and Varanasi. Sumersingh eventually founded the Patna-Kavi-Samaj at the urging of Patna University students, and this group met at the Press library for samasyapurti.19 Besides this informal connection with the University, Sumersingh’s letter of congratulation in the introduction of Max A. Macauliffe’s Sikh Religion might indicate that he held some stature on the pan-Indian Sikh intellectual scene.20

Many of Sumersingh’s publications were Sikh in subject matter, though other subjects and genres were also represented. One, for instance, consisted of kundaliyas21 based upon the more Krishnaite and riti-oriented Bihari Satsai.22 He wrote other pieces on poetic ornament (alamkara), couplets (dohas) on Sikh topics, and a commentary on the Japujī.23 Shivanandan Sahay remembered him as a bhakta of Tulsi’s Avadhi Rāmcharitmanas as well, which he edited at the Press along with a commentary on the Manas by yet another Sikh author.24 Hariaudh described Sumersingh’s works in 1934 as ‘Hindu through and through’, and he noted in the same paragraph that Sumersingh’s works had become unavailable in the intervening years because ‘some Sikh sampradaya members do not consider themselves Hindu, and they were opposed to [these works]’.25 If anyone did in fact attempt to erase Sumersingh’s poetic works, they achieved some measure of success. Little record remains about him in the secondary literature,26 although two Punjabi authors have recently rediscovered and reprinted two of his Braj Bhasha texts on Sikh themes, with the Braj Bhasha texts in Gurumukhi script. His poetic commentary on the Japuji, Japu siddhānta suryoday, completed in Ferozepur in 1897, can be found under the name Sumersingh Singh Bhalla. The editor of his Sri guru pad prem prakāś (ca. 1881) mentions his family biography within the text and reviews a few facts gleaned from other sources, i.e. that his family had settled in Nizamabad in 1761, and that he travelled to Amritsar and Lahore.27 On one of his trips to the Punjab, circa 1888–89, Sumersingh went to Patiala to visit Gaurishankar Vajpeyi, the son of well-known Braj Bhasha poet Chandrashekhar Vajpeyi, who had been poet at the court of arts patron Raja Narendrasingh and the author of works on Vaishnava, Sikh-oriented and classical rasa themes in Braj Bhasha. Sumersingh took with him on this trip the younger, soon-to-be English graduate Ratnakar, who had recently met Sumersingh through his wealthy vaishya in-laws in Patna.28 In Patiala they collected some of Chandrashekhar's manuscripts in what seems to have been a literary-historical project similar to the ones the NPS was supporting at the time. The Raja of Varanasi had already requested Ratnakar to edit an

old text in a quite similar project. By the turn of the century, Ratnakar had edited and published ten or so texts in Braj Bhasha, most of them riti in orientation, some of them as books and others in journals such as the one he edited with Devkinandan Khattri, the Sāhitya sudhānidhi. Hence, manuscript collection and editing was not simply a modern project of history-making, organised by public associations with the language politics of today. Rather, rajas and connoisseurs themselves undertook such projects, usually with mixed modes of text scholarship, ranging from that of manuscript copying (although into print) to the European styles of critical editing. While the previous integration of the present-day categories of ‘Sikh’ and ‘Hindu’ is no new scholarly discovery,29 the extensive network that Sumersingh moved in for religious, literary and, presumably, business purposes, deserves note. Mahant at Patna, litterateur in Varanasi, frequent visitor to the Punjab, and local intellectual in Nizamabad, Sumersingh moved fluidly between and within regions, and in varying social circles, ranging from the wealthy city merchant class of Varanasi, Brahmins in the districts, and English sahibs. Interestingly, what we do know shows Sumersingh’s literary world as a nexus of religious spaces (such as gurdwaras in Nizamabad and, presumably, Patna, and the Gopal temple in Varanasi) and religious texts. These spaces were ‘Sikh’ or ‘Vaishnava’ or both, interfacing with courtly Braj Bhasha genres and with print-culture projects of canon-formation, all with the cultural/intellectual projects of colonialism in the background. Further research on the interconnectedness of Sikh texts with Braj Bhasha poetic genres may yield a more complex picture of the cultural transformations evident in the ‘modernising’ Hindi poetry of the early twentieth century. For instance, some of Hariaudh's progressive ideals bear some similarity to Nanakpanthi ideas such as anti-casteism, nirgun bhakti and the veneration of gurus as human moral models.30 These Sikh/nirgun ‘progressive’ ideas dovetailed with both elite Vedantic movements and liberal ideological agendas inherent in British colonialism. Tat Khalsa rigidification, already begun in Sumersingh’s lifetime, would obscure these kinds of connections. But Hariaudh wrote in 1934:

…in the subject of his poetry [Sumersingh] had great hopes. He wanted his poetry to be spread among the public, and he said that the destruction of differences between Sikhs and Hindus might be accomplished by this means.31 Poetry thus functioned for him as a connecting agent in a political context that was dividing along religious lines—Hindu, Muslim and Sikh. PATRONAGE, PATRONS AND LANGUAGE PROBLEMS At the same time, the generic context divided—albeit unevenly and fitfully —along categories of ‘modern’ literature and ‘medieval’ literature, so that Braj Bhasha as a medium, packaged with its metrical forms and its largely Krishna—or śṛṇgāra—oriented content, underwent widespread disassociation from ideas of the modern, at least among many of those whose aspirations for literature in the Devanagari script went beyond the ‘merely’ devotional, or ‘decadent’ courtly pleasures. The most well-known target of criticism was the genre of nāyikā-bhed, the taxonomy of heroines, and its associated nakh-śikh (toe to head) descriptions, which M.P. Dvivedi lambasted in a famous 1901 article.32 By 1926, Sumitranandan Pant had invoked ‘the three-foot-tall world’ of nakh-śikh descriptions metonymically for Braj Bhasha poetry in its entirety.33 By 1928, criticism of Braj Bhasha was strong enough for Makhanlal Chaturvedi to write defensively (and incorrectly) in his review of Ratnakar’s Gaṅgāvataraṇ that ‘the time of disrespect for Braj is over, the time for its renaissance is coming’.34 However, by the 1930s the defence of women from Braj Bhasha erotica and the campaign to write poetry in the language of prose had merged. Hence, the late history of Braj Bhasha is an ambivalent one of strong feelings of attachment and strong feelings of disdain, much similar to the allegations of courtly decadence in the content of Urdu poetry and the ‘foreignness’ of its linguistic idiom. How did the problem of Braj Bhasha’s pre-modernity interact with, on the one hand, the modern quest for a vernacular literary history in Devanagari-script Hindi-Urdu and, on the other, with the continued courtly patronage systems in nineteenth and twentieth century north India, which

had commissioned and circulated many of these early manuscripts in the first place? How did courts, especially the smaller, more local ones, approach poetry in the age of print? How did the apparatus of the Department of Public Instruction and civic associations guide the production of poetry along different paths? Looking at the circumstances of the production of Hariaudh and Ratnakar’s Braj Bhasha poetry may be useful when approaching these questions. The ongoing communalisation of the literary landscape directed Hariaudh and Ratnakar away from Urdu and towards Braj Bhasha, a cosmopolitan literary dialect that effectively represented ‘Hindi’.35 In a very real sense, Khari Boli Hindi still seemed an impossible medium for good poetry. The newer and less circumscribed domain of prose, however, was more easily amenable to Khari Boli. Hariaudh had his first brush with ‘Hindi’ as a language different from ‘Urdu’ when he was teaching school in Azamgarh; around 1889 the district Deputy Inspector of Schools, Shyam Manohar Das, requested Hariaudh to translate a published ‘Urdu’ translation of Shakespeare’s Merchant of Venice into ‘Hindi’. Hariaudh reported his response to this request as being ‘Urdu is itself a version of Hindi, what [would be] a translation of it!’36 Hariaudh was quick to figure it out, and he refined his prose Hindi register prolifically and variously in the subsequent decades. Thus, it appears that Hariaudh’s introduction to the Hindi movement, and likely his ‘conversion’ to this movement, occurred with this prose translation, presumably meant for use in schools. Yet for poetry, at this point, Braj Bhasha stood in for ‘Hindi’ by dint of its script and its subject matter, and with less concern for lexical origins. The problem of its distance from Khari Boli and other newly objectionable qualities of metre and content were still somewhat intractable. In 1900, Hariaudh published a trilogy of poetic volumes through Khemraj Krishnadas’s Shri Venkateshwar Press of Bombay, publisher of Hindi and Sanskrit works that were distributed throughout India and even worldwide:37 Premāmbu pravāh, Premāmbu prasravaṇa, and Premāmbu vāridhi (The current of the water of love, The flowing of the water of love. The ocean of the water of love). These small books were distinctly Braj

Bhasha in their lexical and grammatical forms, but inflected somewhat according to the grammar of Khari Boli. All three are rather conventional expressions of Krishna bhakti in the various metrical forms of Braj Bhasha verse, displaying Hariaudh's thorough knowledge of this category of literature in subject and formal execution. In yet a fourth title on the topic of love in the year 1900, Hariaudh published Premprapañca arthāt premsambandhī kavitāvalī (Manifestations of love: a collection of poems on love). This text demonstrates the poetic status of Braj Bhasha as the functional ‘Hindi’ alternative to Urdu. The work comprised transcreations of verses by two different Urdu authors into Braj Bhasha: Surur’s verses from the 1824 narrative Fasāna-e’ajāib (Tale of Wonders), along with a marsiya by Anis (1802–1874), mixed with verses of his own composition.38 While certainly this piece exemplifies a work of linguistic ‘othering’, its content suggests that authors writing within the divide also looked for ways of retaining connection with ‘the other’s’ poetry, demonstrated by Hariaudh's invocation of a certain universalism.39 Although the ‘love’ texts do not speak to patronage of a literary association or a court per se, they indicate the extent to which the Hindi movement embraced Braj Bhasha as a medium of division from Urdu. However, it was not long before Braj Bhasha was considered unviable, linguistically and metrically, as a medium for modern, national poetry. Those using Braj Bhasha were fast becoming objects of public criticism. As cited by Christopher King, in 1914 Maithilisharan Gupta ‘called advocates of Braj Bhasha enemies of India’s national language, Khari Boli Hindi’.40 Although earlier, in his 1909 Kāvyopavan, Hariaudh had attributed the ascent of Khari Boli to Indian political power, stating that Braj Bhasha ‘is incompetent in promoting our own… political power’, he also reflected with some nostalgia and poignancy on the waning currency of Braj Bhasha: In ten years there has been a strange change in the taste of the people of this region…[Today] articles on ways to remove Braj Bhasha are being written…. [Now] there is an obstruction in the unimpeded stream of śṛñgāra rasa that had been flowing unbroken’. In its stead, he continued,

people now are enthused especially for description of natural scenes (prākṛtik dṛśya), portrayals of character (svabhāva citra), the promotion of love of one's country, ways to uplift (unnati-sādhan) jāti, country and society.41 In a later Braj Bhasha work from 1917, Ṛtu mukur arthāt ṣaṭ ṛtuō ki uttamottam kavitāyẽ (Mirror of the seasons or most excellent poems on the six seasons), he practically apologised for using Braj Bhasha in the first place, and his subsequent Braj Bhasha output in book form was relatively sparse until his 1931 work Rasakalaśa: Rasasambandhī anūṭhā kāvyagranth (A jug of ras: a unique poetic work on rasa). Half of its contents consisted of an essay interpreting rasa in light of ‘modern’ concerns with ‘usefulness’, science and national character; the other half featured Braj Bhasha-like kavittas and savaiyas on topics of rasa and nāyikā-bhed but with a new slant, such as deś premikā, jāti-premikā, etc. On his retirement from government service in 1923, Hariaudh turned down an offer to work at the court of Chattarpur42 to join Banaras Hindu University at the invitation of Madan Mohan Malaviya. Through the 1920s and 30s he experimented with a variety of lexical ‘styles’ and metres, including some from Urdu, and edited a collection of Kabir padas for the NPS, Kabīr vacanāvalī (1934). During these same years, he revised his famous Priyapravās of 1914 with an eye toward excising what were now considered Braj Bhasha lexical forms.43 Ratnakar, the more urbane of our pair, had at first styled himself an Urdu poet as a young man, using the nom de plume ‘Zaki’. But according to biographers, he ‘ripped up and threw away’ his youthful Urdu attempts.44 Little is known about Ratnakar’s ‘conversion’ away from Urdu but, like Hariaudh, he certainly received ideas promoting ‘Hindi’ from his immediate elders. Around the time he went off to work briefly for the Raja of Avagarh (ca. 1890), he visited Mathura and began to write in Braj Bhasha, and Braj Bhasha alone. Ratnakar’s poetic life involved new social forms like the public association and also literary-historical projects. Ratnakar (and Hariaudh also) had joined in the Kashi Kavi Samaj, a group connected with the Vallabha sect Gopal temple in Varanasi. The indexes of their samasyapurti

publications indicate the participation of individuals with distinctly Brahmin names, and honorific titles suggesting royalty or courtly status (including one courtesan), or perhaps high status within the Vallabha sect.45 In the same period, Ratnakar worked on several editing projects of Braj Bhasha manuscripts, including the manuscripts obtained from Chandrashekhar’s son at the court of Patiala, works by poets Sundar, Kriparam and Ghananand. He published a chapbook embellished with the swinging Radha and Krishna, Hiṇḍolā (1894) on sambhoga-śṛṅgāra, reputedly inspired by his connection to the Gopal temple, well known as a centre of Harishchandra's religious and literary circles. Ratnakar thus worked in a milieu in which Vaishnava sentiments and Hindu-inflected courtly culture mixed with the new kinds of associations and print technologies of the north Indian fin de siècle. In 1903 Ratnakar became personal secretary to the Raja of Ayodhya, a man who himself had literary interests. As Orsini has shown via the memoirs of Padumalal P. Bakhshi, the cultural domain of the raja in a small town extended well beyond palace walls, as the court was the [economic] centre of town life… Since the Raja was a patron of music and of Hindi literature, the local cultural life reflected his tastes. Temples would host poets and poetic performances of courtly works, and the raja would usually publish a work or two of his own.46 Maharaja Pratap Narayan Singh of Ayodhya had published a large and fancy illustrated tome, Rasakusumākar or a Novel Book on Literature: Rasakusumākār arthāt sāhitya kā ek anūṭhā granth, through the Indian Press in 1894. This work itself deserves attention for what it might indicate about the world of courtly text production and Braj Bhasha in the late nineteenth century. On the one hand, this text is a normal Braj Bhasha-medium exposition on Sanskrit poetics, although with prints of paintings from various sources as illustrations.47 The book’s high production value—gilt, prints, and more than the usual amount of floral Victorian print art between chapters and at page margins— reminds one of the ‘ornamented object’ of the hybrid Victorian Indian court explored in Joanne Waghorne’s work on the court of Pudukkottai.48 At base, the Maharaja’s book is a print-culture take on the

courtly illustrated manuscripts of previous centuries, repositories of art as much as text.49 However, its contents hold our attention here: a visually formatted outline precedes the arrangement of verses, indicating the author’s purpose to resystematize his subjects of rasa-nirūpaṇa and nāyikābheda. In the testimonial epilogue, Fredric Pincott lauded the Maharaja’s analysis as ‘strictly scientific’, and possibly ‘a stepping-stone to correct appreciation of the practical affairs by which a nation lives and thrives’ and ‘something to discipline the inclinations of his compatriots’. While this indication of the work’s amenability to a Briton’s taste suggests something about the changing nature of courtly texts, another feature suggests that the text was not meant for the cognoscenti of Braj Bhasha poetry in the first place: in addition to glosses throughout, a glossary rephrases poetics terms like upamā as tashbīh, suggesting that the audience for this text was undergoing a transformation, and perhaps that the Maharaja was consciously reintroducing Braj Bhasha specifically instead of current Urdu poetic practices.50 The visual outlines from the Rasakusumākāra were later anthologised in the first Hindi literature textbook, at Calcutta University, compiled by Ayodhya's own Lala Sitaram,51 prolific translator and no doubt an acquaintance of the Raja. It was for such a rasika that Ratnakar worked as private secretary (the title as such, in English), and then later for his widow the Maharani, who commissioned his 1921 Gaṅgāvataraṇa. She awarded him 1000 rupees for his effort, and he in turn endowed the money to the Nagari Pracharini Sabha for an annual prize for Braj Bhasha poetry, for ‘the advancement [unnati] of Braj’. Based on the story from the Rāmāyaṇa, later critics valued Gaṅgāvataraṇa primarily for its ‘nature description’. Hence, with Ratnakar Braj Bhasha achieved a measure of ‘modernity’ on the interregional Hindi stage. Ratnakar’s later Uddhava-śataka (1929) consisted of a set of kavittas reputedly culled from earlier poetry performances and organised into a narrative around the bhramara gīt theme, and this work remains part of the Hindi M.A. curriculum to the present day. Notably, it has been reprinted twice after Independence in a fancy illustrated edition by the Rasik-Mandal of Allahabad.

Ratnakar’s persistence in the modern canon, such that he is commonly considered the ‘last great Braj Bhasha poet’, is more interesting for the fact that he clearly was little troubled by concerns with a ‘lack of modernity’, and seemed rather less concerned to ‘purify’ his Hindi of Persian and Arabic than some of his peers.52 Ratnakar clung tenaciously to the traditional content of kavitta and savaiyya verses, but as years went on, it seems he wrote fewer poems on sambhoga śṛṅgāra, as in his first Hiṇḍolā. Although some said he occasionally overstepped the bounds of propriety,53 his late Uddhava- śataka does seem to have less ‘bodily content’ than it might have had.54 Still, thoroughly Braj Bhasha in both idiom and sentiment, the kavittas of this work resemble countless others of the bhramara gīt genre of centuries past, replete with the sobbing and sorrowful gopis whom the younger authors sought to exile from modern Hindi poetry. Much of Ratnakar’s poetic output was published only in periodicals during his lifetime, and in his later works we find a number of kavittas on topical themes of nationalism, filtered through the lens of vīr rasa. With a special dispensation from the Maharani, then after his retirement in 1928 and until shortly before his death in 1932, Ratnakar worked in Varanasi on the massive Sūr Sāgar critical edition for the NPS, a project funded through hybrid sources, both governmental and royal. In 1925, at the request of Shyamsundar Das of the NPS, he published his most well-known editing project, the Biharī-Ratnakar ṭīka, still in common use today, for which he had used his connections through the rajas of Varanasi and Darbhanga to gain access to the famous manuscript of the Bihārī-satsaī at the Jaipur royal library.55 Notably, Ratnakar was not alone in his persistent dedication to Braj Bhasha. Interest in editing, analysing, and canonising riti poetry continued unabated through the twenties, despite the many obscenity concerns voiced in periodicals of the day. From his original publications, and undoubtedly for his editorial work, Ratnakar emerged as a ‘standardiser’ of Braj Bhasha, in the way that M.P. Dvivedi is known as the standardiser of Hindi. The critic and Braj Bhasha poet Ramashankar Shukl ‘Rasal’ would write that Ratnakar gave Braj Bhasha its ultimate ‘ekrūptā’ (unity, homogeneity).56

Ratnakar defined his persona through the courtly status symbols ofhis day, which themselves reflected the colonial cultural presence. He liked Tennyson, bought automobiles and published in the Journal of the Royal Asiatic Society.57 But, interestingly, he also took on the accoutrements of earlier courtly culture—he ‘dressed like a talluqedar’58 in silks, used classical gestures while reading his poetry and generally lived the life of a rasika bon vivant, including pastimes like writing verses on grains of rice and discussing the attractiveness of women (nārī-lāva ṇya).59 Nanddulare Vajpeyi wrote that ‘he lived in a medieval atmosphere, and despite having studied English he did not have any special interest in modernity…to the extent that he had the gestures of those medieval times’.60 Shyamsundar Das remarked that ‘looking at him, one would barely think that he had a B.A. in English ’.61 His conservative apparel aside, Ratnakar felt that Braj Bhasha could provide a vehicle for the poetic present. In a speech as sabhapati to the first Akhil Bhartiya Kavi Sammelan in 1925, he argued that Braj Bhasha poetry should change with the times.62 Hariaudh, on the other hand, developed a more sant-like persona, typically wearing a turban, a beard and a white cotton shervani and keeping habits that were more ascetic.63 The two were friends during Hariaudh’s years in Varanasi, and Ratnakar’s death greatly distressed the already fragile Hariaudh. He replaced Ratnakar on the editorial board of the Sūr Sāgar project, and until his death in 1947 he kept up the habit of writing one Braj Bhasha couplet every day. These were published that same year as Hariaudh Satasaī. The similarities between the oeuvre of these two poets are striking: topically, both struggled to reframe śṛṅgāra in terms of ‘nature poetry’; both attempted to use Braj Bhasha forms, either grammatically, metrically, or both, for socially-concerned poetry. However, Ratnakar clearly had a more limited range of production, less interest in experimentation with new or innovative metrical genres even compared to the metrically conservative Hariaudh. Viewing these differences in light of their respective overlapping milieux, their sources of patronage may help explain them. The patronage sources and venues for Braj Bhasha overlapped to a very large degree: royal houses funded Braj Bhasha poetic production and

manuscript collation by public associations, and members of these courts populated less formal poetic associations as well as institutions such as the NPS. Religious centres and public associations also formed interlocking networks of patronage, fostering editions of older manuscripts as well as original Braj Bhasha poetry. That being the case, poets outside the milieu of courts and their culture of text production had probably more encouragement to innovate, although still with a view towards canon. Thus, Ratnakar's court context might be seen as an explanation for his poetic predilection of Braj Bhasha and Braj Bhasha alone. However, he clearly understood and valued the newer literary ‘polities’ of his era: in 1930 he spoke as the sabhapati of the Hindi Sahitya Sammelan meeting in Calcutta and praised the organisation for its examination programme in Hindi literature, which received recognition from municipalities, zila boards, the stri-samaj, Muslim students, and princely states. In this speech from the Sammelan platform, the court appeared as but one of many worthy structures for literary patronage in a social world of burgeoning governmental bureaucracies and public associations.64 Thus, linguistic register, metre and content collapsed together: while to Hariaudh, Urdu verse appeared ‘coquettish’, as he claimed in his Premprapañca, and courtly Braj Bhasha less so, for others Braj Bhasha itself signified descriptions from a sexualised gaze and wordplay at the expense of deep meaning. Braj Bhasha had become an over-determined linguistic mode that was detatched from modernity by its metric, topical and tropical genre associations. The homology between language morphology and content forced the abandonment of Braj for the future Hindi canon. In the words of Orsini, ‘the mere use of Braj Bhasha evoked the nāyak-nāyikā ethos’, just as a couplet of Urdu invoked another ethos.65 GENRES AND PUBLIC PATRONAGE In order to tease out a still more nuanced difference between courtly vs noncourtly Braj Bhasha authors, we can turn to attitudes towards text and how a text might function. Both the small, early-twentieth century courts and the networks of Hindi literati went to some effort to preserve, produce, and culturally revive Braj Bhasha texts—different conceptions of the ‘work’ of

texts may distinguish these two realms further. Ratnakar dearly felt the pleasure of gaming that courtly poetry held in particular. Anuprāsa (alliteration and assonance) remained his stock-in-trade, for linguistically clever verses reminiscent of Keshavdas (see Allison Busch in this volume): Dīna daśā dekhi. . . ..... Sūkhe se srame se sakabake se sake se thake, bhūle se bhrame se bhabhare se bhakuvāne se, haule se hale se hūla-hūle se hie maĩ hāya, hāre se hare se rahe herata hirāne se. (Uddhava-śataka, 1967 [1929]: 33) His language glittered with verbal tricks, if his sense was less elaborate: Seeing their wretched state… … [Uddhava was] despondent, exhausted, overwhelmed, alarmed, worn out, lost, confused, anxious, fretting, Shaken with fright, alas, stabbing prods in his heart, Defeated, taken, searching as if lost. In the neo-classical vein of Alexander Pope, whose Essay on Criticism he had translated for the NPS magazine, Ratnakar upheld the poetic pleasures of ‘the ancients’at the same time as he upheld a certain degree of modern ‘realism’. Ultimately, though, he envisioned his riti-style poetry as part of a beneficent political future, courtly or otherwise. After arguing in his own added verses for the continued value of poetic anuprāsa in the face of poetic change, he exhorted in conclusion, ‘Let the poet and the pleasuregiving art of poetry be true again,/May an abundance of progress [unnati] remain forever in India’.66 Hariaudh, by contrast, tempered his neo-classical commitment to Braj Bhasha forms with the social awareness that he felt was necessary for the times. In his Rasakalaśa of 1931 he self-consciously merged the taxonomies of heroes and heroines with current propriety and with nationalism. His new nāyikās for his era, however, never gained much

attention or critical praise, and perhaps deservedly so. The following is the kavitta ‘Deś-premikā’, the lover of her country: Gauravita satata atīta gauravõ te hoti Gurujana-gurutā hai kahatī kabūlatī Mudita banati avanītala maĩ phaili phaili Kīrati kī kalita latā ko dekhi phūlatī ‘Hariaudh’ prakrti alaukikatā avaloki Prema ke hiṇḍore pai hai pulakita jhūlatī Bhārata kī bhāratī vibhūti te prabhāvita hvai Bhāmini bhalī hai, bhāratīyatā na bhūlatī She is glorious forever, since ancient glories She proclaims her allegiance to the wisdom of her elders. Pleased, she pervades the earth Blooms as she sees the beautiful creeper of praise [for God]. Hariaudh says, gazing at the sublimity of nature She swings, thrilling, on the swing of love. Affected by the splendour of the Sarasvati [or ‘voice’] of India The woman is good; she doesn’t forget her Indian-ness. (Hariaudh 1951 [1931]: 99] The diction is almost Khari Boli, but still sprinkled with Braj-isms. The content is on one hand a prosaic conservative stance, and on the other hand evocative of the love of Radha and Krishna. Hariaudh thus attempted to modernise Braj Bhasha forms, but with less success than his friend Ratnakar, whose language and content were more consistent with the genre historically. Ratnakar thereby, ironically, achieved more success as ‘modern’ poet by adhering to the canonical Braj Bhasha poetic content. CONCLUSION It appears that the dictum of Wordsworth, the ‘language of men’ for poetry did prevail, in the long view, in the newer patronage structures of education and in the poetry market in general. Princely support had already been shifting to the large associations, which princes had joined from their

inception. In a classic example, Sir Virasimhju Dev, the Raja of Orchha, once the court of renowned rīti poet Keshavdas, began to fund a book series through the publishing arm of the Hindi Sahitya Sammelan in 1937 which would be dedicated to ‘the modern poets’ of Hindi.67 In such a way, the former patrons of individual poets shifted their cultural concerns to a public sphere governed by civic associations and canon-driven publications. It is unclear what sort of aesthetic effect the associations, princes and education committees imagined the waning of Braj Bhasha would create: the language of Romantic revolution was surely in their consciousness, but what ‘breaking free’ of convention would actually mean—linguistically, metrically, and topically—had yet to be theorised completely in this context. In summary, the late nineteenth century literary landscape bore more divisions than simply that between Hindi and Urdu, but these divisions developed more slowly on the ground than canonical histories would suggest. Baba Sumersingh Sahabzade is a fascinating example of a transregional intellectual involved in a range of venues of poetic production —courtly, religious and associational. While Hariaudh’s and Ratnakar’s poetic careers demonstrate how Braj Bhasha as a medium of courtly and devotional poetry evolved briefly as a self-consciously ‘modern’ idiom, and then largely withered in the modern Hindi canon. Supported only by the remains of a precolonial courtly milieu and continuing devotional use, Braj Bhasha in a sense became a ‘period piece’, much like Ratnakar’s style of dress. In rethinking the role of Braj Bhasha in the cultural life of north India, we must consider the range of personal networks between poets that remain understudied and under-imagined. The various patrons of poetry also deserve more scrutiny as partial legislators of poetic change. Ultimately, the shift away from Braj Bhasha genres was not merely social nor governed by patronage policies, but also by personal journeys. Ratnakar, Hariaudh and their various patrons can teach us something about the complications and vagaries, personal and political, of defining the pleasure in the text. Finally, the question of genre remains, and of why a genre might fade away with its linguistic register. Perhaps it was the strange inextricability of

Braj Bhasha’s taxonomically-described women in the face of the strong movements for cultural reform. Perhaps it was the inherent nostalgia of Braj Bhasha genres, of a rustic-yet- cosmopolitan idiom and the (ironically) affected country speech of unaffected gopis, which somehow doomed Braj never to be ‘modern enough’, nor cosmopolitan in the right sort of way for the twentieth century.

Contributors

IMRE BANGHA is Lecturer in Hindi at the Oriental Institute, University of Oxford. He is the author of several books and articles in English and Hindi on Braj Bhasha poetry and on textual transmission and Hindi manuscript culture. He is currently in preparing critical editions of several Braj Bhasha texts. ALLISON BUSCH is Assistant Professor of Hindi-Urdu Language and Literature at the University of Columbia. She has published several articles on Braj Bhasha riti poetry and the literary and intellectual life of seventeenth-century imperial and sub-imperial courts. Her book on Mughalperiod Hindi literary culture is forthcoming. THOMAS DE BRUIJN is an independent scholar based in Leiden University, The Netherlands. After his PhD in 1996 on the poetry of the sixteenth century Indian Sufi poet Muhammad Jayasi, he worked as a researcher at the IIAS (Leiden) on a project on the literary field of post-Independence Hindi literature. He has been guest-lecturer at INALCO (Paris), where he completed his habilitation. His research interest in recent years has been in the aesthetics of both pre-modern and contemporary Hindi writing. LALITA DU PERRON is Associate Director of the Center for South Asia at the University of Wisconsin at Madison. She is the author of Hindi Poetry in a Musical Genre: Thumri Lyrics (2007), and of a forthcoming book on the lyrics of Khayal, with Nicholas Magriel. MEHR AFSHAN FAROOQI is Assistant Professor of South Asian Literature at the University of Virginia. She has recently published the two-volume Oxford India Anthology of Modern Urdu Literature (2008) and is the author of several articles on the shaping of modern genres in Urdu. CHRISTINA OESTERHELD teaches Urdu at the Department of Modern South Asian Studies (Languages and Literatures) at the South Asia Institute,

University of Heidelberg. Her main research interest is Urdu fiction from the nineteenth century to the present, including various forms of popular literature. She has translated Urdu short stories and modern poetry into German. FRANCESCA ORSINI is Reader in the Literatures of North India at the School of Oriental and African Studies, University of London. She is the author of The Hindi Public Sphere (2002), Print and Pleasure (forthcoming) and has edited a collection of essays, Love in South Asia. Her new research project seeks to reconsider the literary history of north India from a comparative, multilingual perspective. VALERIE RITTER is Assistant Professor of South Asian Languages and Civilizations at the University of Chicago. She is the author of several articles on late Braj Bhasha poetry, and her book Kama's Flowers: Nature in Hindi Poetry and Criticism, 1885–1925, will be published by SUNY Press. Her new research explores the activities of literati working within Hindu courts in colonial north india.

Bibliography

Abbasi, Z. A. 2003. Qulliyāt-e Mīr. Delhi: Qaumí Kaunsil barā-ye Furūg-e Urdū Zabān (2nd edn). Abdul Qadir, Shah Muhaddis Dehlvi. nd. Al Qurān al-Hakīm ma‘ Tarjuma o Tafsīr Muẓīh al Qurān. Delhi: New Crescent Publishing Company. Adib, Masud Hasan Rizvi. 1984. Shumālī hind kī qadīmtarīn urdū naẓmẽ. Lucknow: Adabsitan. Agraval, Vasudevsharan. 1988. Padmāvat. Malik Muhammad Jāyasī kṛt mahākāvya. Chirgaon: Sahitya Sadan. Alam, Muzaffar. 1996. ‘Assimilation from a Distance. Sufi Accommodation in Awadh Society’. In R. Champakalakshmi and S. Gopal (eds): Tradition, Dissent and Ideology: Essays in honour of Romila Thapar. Delhi: Oxford University Press. ––––– 1998. ‘The Pursuit of Persian: Language in Mughal Politics’ Modern Asian Studies 32:2. ––––– 2002. ‘Persian in Pre-Colonial Hindustan’ In Sheldon Pollock (ed.): Literary Cultures in History: Reconstructions from South Asia. Berkeley: University of California Press: 131–98. ––––– 2004. The Languages of Political Islam in India c. 1200–1800. New Delhi: Permanent Black. Albelisharan and Sarma, Govind (eds): 1999. Saras Caman. Vrindaban: Sri Lakshmimohan Seva Trust. Ali, Abdullah Yusuf. 2002. The Quran. English translation. Istanbul: Asir Media. ‘Ali’, Kishorisharan. 1997. Rasbhakti dhārā aur uskā vāṇī sāhitya. Vrindaban: Ras Bharati Sansthan.

Alvi, Tanvir Ahmad. 1988. Urdū mẽ bārahmāse kī rivāyyat. Muṭāla’ aur matn. Delhi: Urdu Academy. Amin, Syed Muhammad. 2001. Shāh Haqqānī kā Urdu tarjuma-o tafsīr-e Qurān: Ek tanqīdī aur taḥqīqī jā’iza New Delhi: Kutub Khana-e Amjadiya. ’Arun’, Atmaram Sharma. 2000. Ratnākar dvārā upeksit kavi maņdal: Navīnatam khoj evam śodh par ādhārit. Delhi: Bhagyavanti Prakashan. Bahura, Gopal Narayan. 1976. Literary Heritage of the Rulers of Amber and Jaipur. With an index to the register of manuscripts in the Pothikhana of Jaipur. Jaipur: Maharaja Sawai Man Singh II Museum. Bailey, T. Grahame. 1930. ‘Early Urdu Conversation’. Bulletin of the School of Oriental Studies VI (Part I): 205–208. Bangha, Imre. 2001. ‘Lover and Saint: The Early Development ofAnandghan’s Reputation’ Journal of the Royal Asiatic Society 11/2 (July): 175–90. ––––– 2004a. ‘Dynamics of Textual Transmission in Premodern India: The Kavitavali of Tulsidas’ Comparative Studies of South Asia, Africa and the Middle East 24 (2): 33–44. ––––– 2004b. ‘Perzsa muhabbat és szanszkrit rāga: Szerelem kevert nyelven: A Modern Hindi elveszett korai irodalma’ In László Horváth and László Takács (eds): Genesia: Tanulmányok Bollók János emlékére. Budapest: Typotex: 516–36. ––––– 2005. ‘Nāgarī lipi mẽ kharī bolī sāhitya kā ārambh’. Vāgārth (October). ––––– 2007. ‘A Poet of Court and Ashram in the Eighteenth Century: Ānandghan’s contacts with Kishangarh-Rûpnagar and with Salemabad’ In Csaba Dezső (ed): Pāṭhāntarāṇi: Studies by Hungarian Indologists in honour of Prof. Csaba Töttössy. New Delhi: Manohar. Bansal, Nareshchandra. 1988. Rasik Karṇābharaṇ. Vrindaban: Vrindaban Research Institute.

Bārahmāsā Allāh Bakhś. 1867. Agra: Matba Ijad Kishan. British Library: 279.49.a.4. Bārahmāsā Khairāśā. 1862. Agra: Matba Ilahi. British Library: 14158.c.61 Bārahmāsā Harnām. 1869. Delhi: Matba Ahmadi. British Library: VT1202. Bārahmāsa Sūrdās. n.d. Agra: Matba Rada. British Library: 14109.a.10. Bard, Amy. 2002. ‘Desolate Victory: Shi’i Women and the Marsiya Texts of Lucknow’ Unpublished PhD, Columbia University. Barthwal, Pitambardatt. 1978. Traditions of Indian Mysticism Based upon the Nirguna School of Hindi Poetry. New Delhi: Heritage [reprint]. Bayly, C.A. 1988. Rulers, Townsmen, and Bazaars: North Indian Society in the Age of British Expansion, 1770–1870. New Delhi: Oxford University Press. Beach, Milo Cleveland. 1992. Mughal and Rajput Miniature. Cambridge: Cambridge University Press. Beg, Mirza Sangin. 1982. Sairul Manāzil. Edited and translated from Persian into Urdu by Sharif Husain Qasimi. New Delhi: Ghalib Institute. Behl, Aditya. 2002. ‘Premodern Negotiations: Translating between Persian and Hindavi.’ In Rukmini Bhaya Nair (ed): Translation, Text and Theory: The Paradigm of India. New Delhi, London: Sage Publications: 89–100. Bhatnagar, Renu. 1991. Keīav-kāvya par saṃskṛt sāhitya kā prabhāv. Delhi: Parimala Publications. Bhīkhā sāhab kī bānī aur jīvan caritra. 1909. Allahabad: Belvedere Steam Printing Works. Bihari Lal. 1979 [1925]. Bihārī-Ratnākar: Bihārī-satasaī par Ratnākar kī ṭīkā. Varanasi: Tara Publications. Blumhardt, James Fuller. 1900. Catalogue of the Library of the India Office. Hindustani Books, vol. II, part II. London: Eyre and

Spottiswoode. ––––– 1902. Catalogue of the Library of the India Office. Hindi, Panjabi, Pushtu and Sindhi Books, vol. II, part III. London: Eyre and Spottiswoode. Bora, Rajmal. 1987. Bhūṣaṇ aur unkā sāhitya. Kanpur: Sahitya Ratnalay. Bourdieu, Pierre. 1980. Le Sens Pratique. Paris: Editions de Minuit. Brown, Katherine. 2003. ‘Hindustani Music in the Time of Aurangzeb’ Unpublished PhD dissertation, School of Oriental and African Studies, University of London. ––––– 2006. ‘If Music be the Food of Love: Music, Masculinity and Eroticism in the Mughal Mehfil’. In F. Orsini (ed): Love in South Asia: A Cultural History. Cambridge: Cambridge University Press: 61–83. Bryant, Kenneth E. 1978. Poems to the Child-God: Structures and Strategies in the Poetry of Surdas. Berkeley: University of California Press. Bullā sāheb kā śabdasār aur jīvan caritra. 1910. Allahabad: Belvedere Steam Printing Works. Busch, Allison. 2003. ‘The Courtly Vernacular: The Transformation of Brajbhāṣā Literary Culture (1590–1675)’ Unpublished PhD dissertation, University of Chicago. ––––– 2005. ‘Literary Responses to the Mughal Imperium: The Historical poems of Keśavdās’. South Asia Research 25 (1): 31–54. Callewaert, Winand. 1992. The Life and Works of Raidās. New Delhi: Manohar. ––––– 2000. The Millennium Kabīr Vāṇī. New Delhi: Manohar. ––––– (ed). 1993. The Sarvāṃgī of Gopāldās: A Seventeenth Century Anthology of Bhakti Literature. New Delhi: Manohar. Callewaert, W. and B. Beck, (eds). 1991. Nirgun bhakti sāgar. New Delhi: Manohar, 2 volumes.

Callewaert, Winand and Lath, Mukund (eds). 1989. The Hindi Songs of Namdev. Leuven: Department Orientalistiek and New Delhi: Motilal Banarsidass. Caraṇdāsjī kī bānī, 2 vols. 1908. Allahabad: Belvedere Steam Printing Works. Ceserani, Remo and Lidia De Federicis. 1986. Il Materiale e l’immaginario. Torino: Loescher. 10 volumes. Chatterji, Suniti Kumar. 1942. Indo-Aryan and Hindi: Eight Lectures on the History of the Aryan Speech in India and on the Development of Hindi (Hindusthani) delivered before the Research & Post-graduate Department of the Gujarat Vernacular Society in 1940. Ahmedabad: Gujarat Vernacular Society. Chaturvedi, Jawaharlal. 1965. Braj bhāsā rīti śāstra grantha koś, Allahabad: Hindi Sahitya Sammelan. Chaturvedi, Parashuram (ed). 1957. Hindi sāhitya kā bṛhat itihās (IV): Nirgun bhakti. Varanasi: Nagari Pracharini Sabha. Chaudhuri, Jatindra Bimal. 1954. Khān Khānān Abdur Rahim (1557 1630 AD) and Contemporary Sanskrit Learning (1551–1650 Kolkata: Prachyavani Mandir.

AD– AD).

Cohn, Bernard. 1985. ‘The Command of Language and the Language of Command’. In Ranajit Guha (ed): Subaltern Studies IV. New Delhi: Oxford University Press: 276–329. Dadu, Sukhdayal. 1976. Mahaṛṣi VāzidjīMahārāj ke aril. Delhi: Sri Dadudayal Trust, VS 2033. Dale, Stephen F. 2004. The Garden of Eight Paradises. Babur and the Culture of Empire in Central Asia, Afghanistan and India (1483–1530). Leiden-Boston: Brill. Dalmia, Vasudha. 1992. ‘Harishcandra of Banaras and the reassessment of Vaishnava Bhakti in the Late Nineteenth Century’ In R.S. McGregor (ed): Devotional Literature in South Asia: Current Research, 1985–88. Cambridge: Cambridge University Press.

––––– 1996. The Nationalisation of Hindu Traditions. Harishchandra of Banaras. New Delhi: Oxford University Press. Dariyā sāhab bihār vāle ke cune hue pad aur sākhī. 1913. Allahabad: Belvedere Steam Printing Works. Dariyā sāheb mārvāṛ ke prasiddha mahātmā kī bānī aur jīvan caritra. 1909 (2nd edn) Allahabad, Belvedere. Steam Printing Works, Das, Brajratna. 1948. Bhāratendu Hari§candra. Allahabad: Hindustani Academy. Das, Narayan (ed). 1997. Khānkhānāviracitam Kheṭa-kautukam. Varanasi: Caukhamba Vidyabhavan. Daudi, Khalilur Rahman (ed). 1962. Qawā’id-e Zabān-e Urdū mashūr ba Risāla-ye Gilchrist. Lahore. de Bruijn, Thomas. 1996. The Ruby Hidden in the Dust: A Study of the Poetics of Malik Muḥammad Jāyasī’s Padmāvat, (327pp.). PhD dissertation, Leiden University. ––––– 1999. ‘Images of the Last Day: Jāyasī’s Ākhirī Kalām and Indian Islam’. In Alan W. Entwistle et al (eds): Studies in Early Modern IndoAryan Languages, Literature and Culture. New Delhi: Manohar: 71–85. ––––– 2000. ‘Visions of the Unseen, Rhetorical Strategies in Sufi and Bhakti Poetry’. In Mariola Offredi (ed.): The Banyan Tree: Essays on Early Literature in New Indo-Aryan Languages, Volume 1. New Delhi: Manohar: 69–82. ––––– 2005. ‘How Many Roads to Lanka? The Intercultural Semantics of Rama’s Quest.’ In Jacqueline Suthren Hirst and John Zavos, Teaching across South Asian Religious Traditions, Vol. 14.1, 2005 ISSN print 09584935, ISSN online 1469–364X. Delvoye, Françoise Nalini. 1994. ‘The Thematic Rage of Dhrupad Songs Attributed to Tansen, Foremost Court-Musician of the Mughal Emperor Akbar’. In Alan W. Entwistle and F. Mallison (eds): Studies in South Asian Devotional Literature. Delhi: Manohar: 407–29. ––––– 1995. ‘Indo-Persian Literature and Art-Music: Some Historical and Technical Aspects’. In F.N. Delvoye (ed): Confluence of Culture:

French contributions to Indo-Persian studies. Delhi: Manohar: 93–130. Desai, Vishakha N. 1985. Life at Court: Art for India’s Rulers, Sixteenth to Nineteenth Centuries. Boston: Museum of Fine Arts. Deshpande, Vamanrao. 1973. Indian Musical Traditions: An Aesthetic Study of the Gharanas of Hindustani Music. Bombay: Popular Prakashan. Dewhurst, R.P. 1915. ‘Persian and Arabic Words in the Satsai of Bihari Lal’. Journal of the Royal Asiatic Society: 122–27. Dikshit, Triloki Narayan. 1963. Hindī sant sāhitya. New Delhi: Rajkamal Prakashan. Dionisotti, Carlo. 1967. ‘Geografia e storia della letteratura italiana’. In Geografia e storia della letteratura italiana. Torino: Einaudi. Divakar, Krishna. 1969. Bhonslā rājdarbār ke hindī kavi. Varanasi: Nagari Pracharini Sabha. Dvivedi, Ganesh and Caturvedi, Parashuram. 1974. Hindī sant kāvya saṃgrah. Allahabad: Hindustani Academy. Dropadījī kī bārahmāsī. n.d. Delhi: Matba Jvalaprakash. British Library: 14158.c.3. Dvivedi, Harihar Nivas. 1955. Madhyadeśīya bhāṣā. Gwalior: Vidyamandir Prakashan. ––––– 1976. Gwāliyar ke Tomar. Gwalior. Dvivedi, Mahavirprasad. 1995. Mahavīraprasād Dvivedī racanāvalī. Volume 2. Edited by B. Yayavar. New Delhi: Kitab Ghar. Dvivedi, V.P. 1980. Bārahmāsā. The Song of Seasons in Literature and Art. Delhi: Agam Kala Parkashan. Eaton, Richard. 1978. The Sufis of Bijapur. Princeton: Princeton University Press. Entwistle, Allan W. 1987. Braj: Centre of Krishna Pilgrimage. Groningen: Egbert Forsten.

Ethé, Hermann. 1980. Catalogue of Persian Manuscripts in the India Office Library. London India Office Library and Records. Faruqi, Nisar Ahmad. 1989. Naqd-e Malfūẓāt. Delhi: Maktabah Jamia. Faruqi, Shamsur Rahman. 1999. Urdū kā ibtidā’ī zamāna: Adabī tahzīb va tārīkh ke pahlū. Karachi: Aj ki kitaben. ––– 2001. Early Urdu Literary Culture. New Delhi: Oxford University Press. ––––– 2002. ‘A Long History of Urdu Literary Culture. Part 1. Naming and Placing a Literary Culture’. In Sheldon Pollock (ed): Literary Cultures in History: Reconstructions from South Asia. Berkeley: University of California Press: 805–63. Faruqi, Ziya-ul-Hasan (ed). 1996. Fawā‘id al-Fu’ād: Spiritual and Literary Discourses of Shaikh Niẕāmuddīn Awliyā Originally Compiled by Amīr Ḥasan ‘Alā’ Sijzī Dehlawī. New Delhi: D. K. Printworld. Fatehpuri, Farman. 1977. Urdū imlā aur rasmul khaṭ. Lahore: Sang-e Meel Publications. Fazle Rahman, Shah Ganj Moradabadi. 1990. Manmohan kī bātẽ. Introduced by Nazar Ali Khan. Patna: Khudabakhsh Oriental Library. Fazli, Fazl Ali. 1965. Karbal Kathā. Edited by Malik Ram and Mukhtaruddin Ahmad. Patna. Gallego, María Angeles. 2003. ‘The Languages of Medieval Iberia and their Religious Dimension’. Medieval Encounters 9 (1): 107–39. Gangohi, Abdul Quddus. 1971. Alakh Bani or Rushd Nama. Tr. S.A.A. Rizvi and S. Zaidi. Aligarh: Bharat Prakashan Mandir. Garcin de Tassy, Joseph-Héliodore-Sagesse-Virtue. 1850. ‘Analyse d’un monologue dramatique indien’. Journal Asiatique, October: 310–28. Gilchrist, John A. 1796. A Grammar of the Hindostanee Language or Part Three of Volume First, of a System of Hindostanee Philology. Kolkata: Chronicle Press.

Giovanardi, Claudio. 1994. ‘Il bilinguismo italiano-latino del medioevo e del Rinascimento’. In Storia della lingua italiana, Vol. 2, Lo Scritto e il Parlato. Torino: Einaudi: 435–67. Girish, Girijadatt Shukla. 1959. Mahākavi Hariaudh: Caritātmak samīkṣā. Allahabad: Ramnarayan Lal. Gordon, Stewart. 1993. The Marathas: 1600–1818. New Delhi: Cambridge University Press. Grierson, George A. 1884. ‘Some Bihárí Folk-Songs’. Journal of the Royal Asiatic Society of Great Britain and Ireland, Vol. 16: 196–246. ––––– 1889. The Modern Vernacular Literature of Hindustán. Kolkata: Asiatic Society. Growse, F.S. 1879. ‘The Sect of the Prán-náthís’, Journal of the Asiatic Society of Bengal vol. XLVIII: 171–80. (With text and translation of Prannath’s Kiyāmat-nāmā.) ––––– 1987 [1877–1880]. The Rāmāyaṇa of Tulasidasa translated from the original. by F.S. Edited and revised by Growse; R.C. Prasad. Delhi: Motilal Banarsidass. Guha, Sumit. 2004. ‘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): 23–31. Gupta, Charu. 2001. Sexuality, Obscenity, Community: Women, Muslims, and the Hindu Public in Colonial India. New Delhi: Permanent Black. Gupta, Jagdish. 1961. Rītikāvya Saṃgrah. Allahabad: Sahitya Bhavan. Gupta, Kishorilal (ed.). 1965. Nāgarīdās Granthāvalī I-II. Banaras: Nagari Pracharini Sabha. ––––– 1966. ‘Hariaudh jī ke kāvya-guru: Bābā Sumer Siṃh Sāhabajade’, in Hariaudh śati smārak granth Edited by V.L. Shaida and K.L. Gupta. Azamgarh: Hariaudh Shodh Samsthan. Gupta, Mataprasad. 2002 [1942]. Tulsīdās: ek samālocanātmak adhyayan. Allahabad: Lokbharati (6th edition).

Gupta, Mataprasad (ed). 1961. Mañjhankṛt Madhumālatī. Allahabad: Mitra Prakashan. Guru, Kamataprasad. 1920. Hindī-vyākaraṇ. Varanasi: Nagari Pracharini Sabha. Hacker, Paul. 1983. ‘Inklusivismus’, in: Inklusivismus: Eine Indische Denkform, herausgegeben von Gerhard Oberhammer. Wien-Delhi: Brill-Motilal Banarsidass. Hadi, Nabi. 1995. Dictionary of Indo-Persian Literature. New Delhi: Indira Gandhi National Centre for the Arts. Hansen, Kathryn. 1992. Grounds for Play. The Nautanki Theatre of North India. Berkeley: University of California Press. ––––– 2003. ‘Languages on Stage: Linguistic Pluralism and Community Formation in the Nineteenth-Century Parsi Theatre’ Modern Asian Studies 37 (2): 381–405. Haq, Qazi Fazl. 1931. ‘Urdū mughalõ ke darbār mẽ.’ Oriental College Magazine (Aug.): 98–102. Haqq, ‘Abdul. 1951. ‘Vafāt nāma-ye Bībī Fātima’. Risāla-ye Urdū (April). ––––– 1961. Qadīm Urdū. Karachi: Anjuman Taraqqi Urdu. ––––– 1977. Urdū kī ibtidā’ī nashonumā mẽ ṣūfiyā’e karām kā kām. Karachi: Anjuman-e Taraqqi-e Urdu. Haqqani, Shah Muhammad,‘Ίnāyat Rasūl kī. Manuscript at the Khanqah-e Barkatiya, Marehara. ‘Hariaudh’, Ayodhyasingh Upadhyay. 1909. Kāvyopavan nānārasamayī kavitāvalī. Patna: Khadgavilas Press.

arthāt

––––– 1928. Venis kā bā̃ kā. Varanasi: Pathak and Son Bhasha Bhandar Pustakalay. ––––– 1934. Hindī bhāṣā aur uske sāhitya kā vikās arthāt Bābū Rāmadīn Simh Readership ke sambandh se Paṭnā University mẽ diye gaye Paṇḍit Ayodhyāsiṃh Upādhyāy (Hariaudh), Professor, Hindu University, Banaras, ke vyākhyānõ kā saṃgrah. Patna: Patna University.

––––– 1951 [1931]. Rasakalaśa. Varanasi: Hindi Sahitya-Kutira. ––––– 1996 [1914, revised 1941]. Priyapravās: Khaṛī-Bolī kā sarvaśreṣṭh mahākāvya. Varanasi: Hindi Sahitya Kutir. Hasan, Mushirul. 2004. From Pluralism to Separatism. Qasbas in Colonial Awadh. New Delhi: Oxford University Press. Hashimi, Nurul Hasan and Khan, Ma’sud Husain (eds). 1979. Muḥammad Afẓal: Bikaṭ kahānī: Bārah māsah. Lucknow: Uttar Pradesh Urdu Academy. Hawley, John Stratton. 1983. Krishna, The Butter Thief. Princeton: Princeton University Press. ––––– 1984. SurDas: Poet, Singer, Saint. Seattle: University ofWashington Press. Husain, Muhammad Basheer. 1968. Fihrist-e makhtūtāt-e Sherānī, vols. I-II (Annotated catalogue of a collection of manuscripts donated by Mahmud Khan Sherani to the Punjab University Library, Lahore). Lahore. Husaini, S. Bilqis F. 1988. A Critical Study of Indo-Persian Literature During Sayyid and Lodi Period 1414–1526 ad. Delhi: M.S. Publication. ‘Insha’, Insha Allah Khan, and ‘Qatil’, Mirza Muhammad Ahsan. 1962. Daryā-yi Lat̤ āfat. Edited by Maulvi Abdul Haq. Karachi: Aftab Academy. Islam, Khurshidul and Russell, Ralph. 1969. Three Mughal Poets. London: Allen and Unwin. Jackson, Paul SJ. 1987. The Way of a Sufi: Sharafuddin Maneri. Delhi: Idarah- i Adabiyat-i Dilli. Jafar, Saiyida and Jain, Gyanchand. 1998. Tārīkh-e adab-e urdū, 1700 tak. New Delhi: Qaumi Kaunsil baraye Rurugh-e Urdu zaban, 5 volumes. Jaiswal, Usha. 1956. Ratnākar aur unkā sāhitya.Varanasi: Hindi Pracharak Pustakalay.

Jalibi, Jamil. 1975. Tārīkh-e adab-e urdū. Lahore: Majlis-e Taraqqi-e Adab, 3 volumes. ––––– 1977. Tārīkh-e adab-e urdū. Vol. I. Delhi: Educational Publishing House. Jawán, Mirza Kazim ’Ali. 1812. The Bārahmāsā. A Poetical Description of the Year in Hindoostan. Calcutta: Hindoostanee Press. Jones, Kenneth. 1973. "Ham Hindu Nahīṃ: Āryā-Sikh Relations, 1877– 1905" Journal of Asian Studies 32 (3):457–75. Karim, Irtiza. 1987. ‘Ajā’ibul Qiṣṣas kā tanqīdī mut̤ ala’. Delhi. Kesavan, B. S. and A. Gupta. 1997. History of Printing and Publishing in India: A Story of Cultural Re-awakening. New Delhi: National Book Trust. Keshavdas. 1993. Jahāngīrjascandrikā. Kishori Lal (ed). Allahabad: Sahitya Bhavan. Khan, Masud Husain (ed). 1965. Qadīm urdū. Vol. 1. Hyderabad. Khan, Masud Husain. 1966. Muqaddama-ye tārīkh-e zabān-e urdū; Urdū ke ’ahad ba ’ahad irtiqā kī tārīkh aur us kī ibtidā se muta’alliq lisāniyātī naẓariyõ kā jā’izah. Lahore: Urdu Markaz. Khan, Mirza. 1935. A Grammar of the Braj Bhakha: The Persian text critically edited from original MSS., with an introduction, translation and notes, together with the contents of the Tuḥfatu-l-Hind, by M. Ziauddin. Calcutta, Vishva-Bharati Book-Shop. Khan, Muhammad Habibulrahman (ed.). n.d. Nikātush-Shu’arā. Badayun: Nizami Press. Khan Bahadur, Isvi. 1966. Qiṣṣa-e Mehr Afroz-o Dilbar. Edited by Masud Husain Khan. Hyderabad. Khatri, Ayodhya Prasad (comp.). 1889. Khaṛī bolī kā padya: The Poetical Reader of Khari Boli. London: W.H. Allen. Kidwai, Sadiqur Rahman. 1972. Gilchrist and the Language of Hindoostan, New Delhi: Rachna Prakashan.

––––– 2004 ‘Anis kī marsiya go’ī urdu naẓm kī tārīkh mẽ’ śabkhūn No. 282 (July). King, Christopher R. 1974. ‘The Nagari Pracharini Sabha of Benares, 1893–1914: A Study of the Social and Political History of the Hindi Language’ Unpublished PhD dissertation, University of Wisconsin. ––––– 1994. One Language, Two Scripts. The Hindi Movement in Nineteenth Century North India. New Delhi: Oxford University Press. Kishorilal. 1971. Rīti kaviyõ kī maulik den. Allahabad: Sahitya Bhavan. Kochar, Harvamsha. n.d. Apabhraṃśa sāhitya. Delhi: Sahitya Mandir. Lefèvre-Agrati, Corinne. 2006. The Court of ’Abd-ur-Raḥīm Khān-i Khānān as a Bridge Between Iranian and Indian Cultural Traditions’ Unpublished paper delivered at the Nineteenth European Conference on Modern South Asian Studies, Leiden. Lelyveld, David. 1993. ‘The Fate of Hindustani: Colonial Knowledge and the Project of a National Language’. In C.A. Breckenridge and P. van de Veer (eds): Orientalism and the Postcolonial Predicament. Philadelphia: University of Pennsylvania Press: 189–214. ––––– 1994. ‘Zuban-e Urdu-e Mu’alla and the Idol of Linguistic Origins.’ Annual of Urdu Studies 9: 57–67. Lutgendorf, Philip. 1991. The Life of a Text: Performing the Ramcaritamānas of Tulsidas. Berkeley: University of California Press. Macauliffe, Max Arthur. 1963 [1909]. The Sikh Religion: Its Gurus, Sacred Writings, and Authors. Delhi: S. Chand. MacLean, Derryl N. 2003. ‘The Sociology of Political Engagement: The Mahdaviyah and the State’. In Richard Eaton (ed). India’s Islamic Traditions: 711–1750. New Delhi: Oxford University Press: 150–66. Madani, Zahir ud-Din. 1990. Gujrī mas̤ naviyā̃ . Gandhinagar: Gujarat Urdu Academy. Maheshvari, Hiralal. 1980. History of Rajasthani Literature. New Delhi: Sahitya Academy.

Majmu’a bārahmāsa. 1876. Delhi: Matba Brahm Press. British Library: 14119.d.5. Majmu ’adilpasand. 1876. Kangra: Matba Fauq Kashi. British Library: 14119.d.6. Malik Ram. 1975. Qadīm Dillī College. New Delhi: Maktaba-e Jamia. Malūkdāsjī kī bānī: Jīvan caritra sahit. 1912. Allahabad, Belvedere Steam Printing Works. Mangaldas, Svami (ed). 1948. Pañcāmṛt. Jaipur: Shri Svami Lakshmiram Trust. Mathur, Ramesh. 1974. Padmāvata: An Etymological Study. Kolkata: Intercultural Research Institute. McGregor, Ronald Stuart. 1968. The Language of Indrajit of Orcha: A Study of Early Braj Bhasa Prose. Cambridge University Press. ––––– 1984. Hindi Literature from its Beginnings to the Nineteenth Century. Vol. 8, pt. 6 of A History of Indian Literature. Edited by Jan Gonda. Wiesbaden: Harrassowitz. ––––– 2000. ’A Narrative Poet’s View of his Material: ViṣŇudās’s Introduction to his Brajbhasa Pāṇḍav-carit (ad 1435)’ in Mariola Offredi (ed): The Banyan Tree, Vol. 2. Delhi: Manohar. ––––– 2000. ‘The Formation of Modern Hindi as Demonstrated in Early "Hindi" Dictionaries’. Gonda Lecture published by the Royal Netherlands Academy of Arts and Sciences, Amsterdam. ––––– 2002. ‘Progress of Hindi: Part 1’ in Sheldon Pollock (ed): Literary Cultures in History. Berkeley: University of California Press 942–944. Meisami. Julie Scott. 1987. Medieval Persian Court Poetry. Princeton: Princeton University Press. Menariya, Motilal. 1942. Rājasthān mẽ hindī ke hastalikhit granthõ kī khoj, Vols. I-IV. Udaipur. ––– 1978. Rājasthānī bhāṣā aur sāhitya. Allahabad: Hindi Sahitya Sammelan.

Miner, Allyn. 1993. Sitar and Sarod in the Eighteenth and Nineteenth Centuries. Delhi: Motilal Banarsidass. Mishra, Bhagirath (ed). 1956. Akbarsāhkṛtsṛṇgārmañjarī. Lucknow: Lucknow University. ––––– 1972. Hindi sāhitya kā bṛhat itihās (VII): Rītikāl: Rītimukt. Varanasi: Nagari Pracharini Sabha. Mishra, Vidyanivas and Govind Rajnish, (eds). 1985. Rahīm granthāvalī. New Delhi: Vani Prakashan. Mishra, Vishvanath Prasad. 1952. Ghanānand granthāvalī. Banaras: Vani Vitan. ––––– 1954–1956. Keśavgranthāvalī. Allahabad: Hindustani Academy, 3 Vols. ––––– 1957. Bhikhārīdās granthāvalī. Varanasi: Nagari Pracharini Sabha ––––– 1965. Bihārī. Varanasi: Vani-Vitan Prakashan. ––––– 1972. Bhāsābhūṣan. In Jasvant Siṃha granthāvalī. Varanasi: Nagari Pracharini Sabha. –––––(ed). 1974. Bodhā granthāvalī. Varanasi: Nagari Pracharini Sabha. ––––– 1994 [1953]. Bhūṣaṇ granthāvalī. New Delhi: Vani Prakashan. Mishrabandhu [Ganeshvihari, Shyamvihari and Shukdevvihari Misra]. 1972 [1913]. Misrabandhu vinod. Hyderabad: Ganga-Granthagar. Munis, Parkash. 1978. Urdū adab par hindī kā as̤ ar. Delhi: Maktaba Jamia. Muradullah, Shah Ansari Sambhali. 1867. Khudaī ne‘mat al-ma’ruf tafsīr-e muraddiya. Bombay. Nagendra (ed). 1973. Hindī sāhitya kā bṛhat itihās. Vol. 6. Varanasi: Nagari Pracharini Sabha. Nahta, Agarchand (ed). 1965. Rāghavdās kṛt Bhaktamāl: Caturdās kṛt ṭīkā sahit. Jodhpur: Rajasthan Oriental Research Institute.

Naik, Chhotubhai Ranchhodji. 1966. Abdu’r-Raḥīm Khān-i-Khānān and His Literary Circle. Ahmedabad: Gujarat University. Naim, C.M. 1983. ‘The Art of the Urdu Marsiya’. In Milton Israel, and N. K. Wagle (eds): Islamic Society and Culture. Essays in Honour of Professor Aziz Ahmad. New Delhi: Manohar: 101–16. ––––– 1999. Zikr-i Mir: An Eighteenth Century Autobiography. New Delhi: Oxford University Press. ––––– 2001. ‘Transvestic Words? The Rekhti in Urdu’. Annual of Urdu Studies 16 (1): 3–26. ––––– 2004. Urdu Texts and Contexts. The Selected Essays of C.M. Naim. New Delhi: Permanent Black. Oberoi, Harjot. 1994. The Construction of Religious Boundaries: Culture, Identity, and Diversity in the Sikh Tradition. Delhi: Oxford University Press. Orsini, Francesca. 2002. The Hindi Public Sphere, 1920–1940: Language and Literature in the Age of Nationalism. New Delhi: Oxford University Press. ––––– 2007. ‘Traces of a Multilingual World: The Uses of Hindavi in Persian texts’. Paper delivered at the conference ‘After Timur Came: Multiple Spaces of Cultural Production and Circulation in Fifteenthcentury North India’ SOAS, London, 28–30 May. ––––– [forthcoming]. ‘Hindi and Urdu Barahmasas in Print’ Print and Pleasure: Popular Literature and Entertaining Fictions in Colonial North India. New Delhi: Permanent Black. Osho. 1995. Kahai Vājid pukār: Ośo dvārā Vājīd vāṇī par die gae das amṛt pravacanõ ka apratim saṃskāraṇ. Poona: The Rebel Publishing House. Padmavati. 1973. Mīrā: vyaktitva aur kṛtitva. Varanasi: Hindi Pracharak Sansthan. Panchal, Paramanand. 2001. Amīr Khusro: vyaktitva aur kṛtitva. New Delhi: Hindi Book Centre.

Pandey, Sudhakar (ed). 1987 [1969]. Raslīn granthāvalī. Varanasi: Nagari Pracharini Sabha. Petievich, Carla. 1990. ‘Boy-Love in the Urdu Ghazal’. Annual of Urdu Studies 7: 1–20. ––––– 1993. ‘The Feminine and Cultural Syncretism in Early Dakani Poetry’, in Annual of Urdu Studies 8: 120–30. ––––– 2000. ‘Dakani’s Radha-Krishna Imagery and Canon Formation in Urdu’. In Mariola Offredi (ed): The Banyan Tree: Essays on Early Literature in New Indo-Aryan Languages. New Delhi: Manohar, Vol.1: 113–30. ––––– 2002. ‘Doganas and Zanakhis: The Invention and Subsequent Erasure of Urdu Poetry’s ‘Lesbian’ Voice.‘ In Ruth Vanita (ed): Queering India: Same-Sex Love and Eroticism in Indian Culture and Society. New York: Routledge: 47–60. Phukan, Shantanu. 2000. ‘Through a Persian Prism: Hindi and Padmāvat in the Mughal Imagination’ Unpublished PhD dissertation, University of Chicago. ––––– 2001. ‘Through Throats Where Many Rivers Meet’: The Ecology of Hindi in the World of Persian’. Indian Economic and Social History Review XXXVIII (1), January-March: 33–58. Pickthall, Marmaduke. n.d. Holy Quran. English and Urdu Translation by Maulana Fateh Muhammad Jallandhari. Delhi: Kutubkhana Ishayatul Islam. Pinch, William R. 1996. Peasants and Monks in British India. Berkeley: University of California Press. Pindye, Jayarama. 1989. Rādhāmādhavavilāsacampū. Pune: Varada Books. Platts, John T. 1960. A Dictionary of Urdū, Classical Hindī and English (reprint). Oxford University Press. Pollock, Sheldon. 1998. ‘India in the Vernacular Millennium: Literary Culture and Polity, 1000–1500’. Daedalus, 127 (3): 41–75.

––––– 2001. ‘New Intellectuals in Seventeenth-century India.’ Indian Economic and Social History Review 38 (1): 3–31. –––––(ed). 2002. Literary Culture in History: Reconstructions from South Asia. Berkeley: University of California Press. Prakash, Om. 1956. Hindī-alaṃkāra-sāhitya. Delhi: Bharati Sahitya Mandir. Prasad, Pushpa. 1990. Sanskrit Inscriptions of Delhi Sultanate 1191–1526. New Delhi: Oxford University Press. Pritchett, Frances. 1985. Marvelous Encounters: Folk Romances in Hindi and Urdu. Delhi: Manohar. ––––– 1994. Nets of Awareness. Urdu Poetry and Its Critics. Berkeley: University of California Press. ––––– and Shamsur Rahman Faruqi: 2001. Āb-e ḥayāt: Shaping the Canon of Urdu Poetry. New Delhi: Oxford University Press. Punzo Waghorne, Joanne 1994. The Raja’s Magic Clothes: Re-visioning Kingship and Divinity in England’s India. University Park: Pennsylvania State University Press. Quli, Dargah Khan. 1982. Muraqqa’-e Dehlī. Edited and translated from Persian into Urdu by Nurul Hasan Ansari. Delhi: Delhi University. Rafiuddin, Shah. n.d. Qurān Sharīf Mutarjim ba do tarjuma. Delhi: Kutub Khana-e Rashidiya. Raghavan, V. (ed). 1951. śṛṅgārmañjarīof AkbarShah. Hyderabad: Hyderabad Archaeological Department. Raheja, Gloria Godwin and Ann Grodzins Gold. 1994. Listen to the Heron’s Words. Berkeley: University of California Press. Rai, Amrit. 1991. A House Divided: The Origin and Development of Hindi/ Hindavi. New Delhi: Oxford University Press. Rai, Alok. 2000. Hindi Nationalism. New Delhi: Orient Longman. Rajguru, Govindnath (ed). 1974. Hariji Sodhi kṛt gosati guru miharivanu. Chandigarh: Punjab University Publication Bureau.

Rajnis, Govind. 2004. Pañcāmṛt aur pañcraṅg: Madhyakālīn sant-kaviyõ kā prāmāṇik pāṭh-sampādan. Jaipur: Maya Prakashan. Ram, Shivmangal. 1996. Svāmī Lāladāsa kṛta mahāmati Prāṇanātha bīṭaka kā madhyakālīna bhāratīya itihāsa ko yogadāna. New Delhi: Hindi Book Centre. Randhawa, R.S. 1994 (1962). Kangra Paintings on Love. Delhi: Ministry of Information and Broadcasting. Rao, V. N., David Shulman, and Sanjay Subrahmanyam. 1992. Symbols of Substance: Court and State in Nāyaka Period Tamilnadu. Delhi: Oxford University Press. Ratnakar, Jagannath Das. 1897. ‘Samālocanādarśa’. Nāgarī Pracāriṇī Patrikā, 1 (1). ––––– 1950[1933]. Ratnākar arthāt golokavāsī śrī Jagannāthadās Ratnākar ke sampūrṇ kāvyõ kā saṃgraha. Vol. 1, edited by S.S. Das. Varanasi: Nagari Pracharini Sabha. ––––– 1967 [1929]. Uddhav śataka. Allahabad: Rasik Mandal. Ray, Lallan. 1958. Ratnākar aur unkā ‘Uddhavaśataka’: Kavivar Ratnākar ke kāvya kī viśad ālocanā aur ‘Uddhavaśatak’ kī vistṛt vyākhyā. Varanasi: Nandakishor and Brothers. Ritter, Valerie. 2004. ‘The Language of Hariaudh’s Priyapravās: Notes towards an Archaeology of Modern Standard Hindi.’ Journal of the American Oriental Society 124 (3). Rizvi, Sayyid Athar Abbas (ed). 1957. Haqāyaq-i Hindī—Mīr Abdul Vāhid Bilgrāmī. Varanasi: Nagari Pracharini Sabha. Rizvi, Sayyid Athar Abbas. 1965. Muslim Revivalist Movements in Northern India in the Sixteenth and Seventeenth Centuries. Agra: Agra University. ––––– 1978. A History of Sufism in India. Delhi: Munshiram Manoharlal, 2 Vols. Rosenstein, Lucy. 1997. The Devotional Poetry of Svami Haridas: A Study of Early Braj Bhāṣā Verse. Groningen: Egbert Forsten.

Ross, E. Denison. 1910. The Persian and Turki Dīvāns of Bayram Khān, Khān- khānān. Calcutta, Asiatic Society of Bengal. Saksena, Ram Babu. 1927. A History of Urdu Literature. Allahabad: Ram Narain Lal. Sarvarhalai, Hazrat Makhdoom Nuh Sindi. 1981. Qurān al-Ḥakīm (Persian translation of the Quran). Edited by Abu Said Mustafa Qasimi. Hyderabad: Sindi Adabi Board. Sauda, Mirza Muhammad Rafi. 1984. Kulliyāt-e Saudā. Edited by Muhammad Hasan. New Delhi: Taraqqi Urdu Bureau. Schimmel, Annemarie. 1975. Classical Urdu Literature from the Beginning to Iqbāl. Wiesbaden: Otto Harrassowitz. ––––– 1992. ‘A Dervish in the Guise of a Prince: Khan i Khanan Abdur Rahim as a Patron.’ In Barbara Stoler Miller (ed): The Powers of Art: Patronage in Indian Culture. Delhi: Oxford University Press. Schomer, Karine. 1983. Mahadevi Varma and the Chhayavad Age of Modern Hindi Poetry. Berkeley: University of California Press. Schomer, Karine and H. McLeod (eds). 1987: The Sants: Studies in a Devotional Tradition of India. Delhi: Motilal Banarsidass. Schulze, Benjamin. 1977. Urdu Grammar. Tr. Abul Lais Siddiqi. Lahore: Majlis Taraqqi-e Abdab. Selby, Martha Ann. 2000. Grow Long, Blessed Night. Love Poems from Classical India. New York: Oxford University Press. Sengar, Shiva Singh. 1970 [1878]. śivasiṇh saroj: Hindī sāhitya ke itihās kā ādigranth pramāṅik evam saṭippaṇsamskāraṅ . Edited by K.L. Gupta. Allahabad: Hindi Sahitya Sammelan. Shah Alam Sani. 1965. ‘Ajā’ibul Qiṣṣas’. Edited by Rahat Afza Bukhari. Lahore. Sharma, Mukunddev. 1956. Hariaudh aur unkā sāhitya: Hariaudh jī ke jīvan aur unke sampūrṇ sāhitya kā ālocanātmak vivecan. Varanasi: Hindi Sahitya Kutir.

Sharma, Purohit Harinarayn. 1931. Brajnidhi Granthāvalī. Varanasi: Nagari Pracharaini Sabha. Shastri, Premsagar (ed.) 1989. Adi Rāmāiṇā. Patiala: Bhasha Vibhag. Shauq, Mirza ‘Lakhnavi’ 1982. Zahr-e ’ishq, or Poison of love: A Love Narrative from Awadh. Translated by Shah Abdus Salam. Delhi: D.K. Publications. Sherani, Hafiz Muhammad. 1926. ‘Rekhta’ In Maqālāt-e Sherānī Vol. I: 19. (Originally published in Oriental College Magazine, May 1926) ––––– 1930a Panjab mē̃ urdū. Lahore: Nur Hasan. ––––– 1930b. ‘Urdu ke fiqre aur dohre āţhvī̃ aur navī̃ ṣadī kī fārṣī tasnīfāt se’ In Maqālāt-e Sherānī Vol. I. Lahore: Majlis-e Taraqqi-e Adab: 132– 58. (Originally published in Oriental College Magazine, August 1930). ––––– 1931. ‘Dasvī̃ ṣadī hijrī ke ba’z jadīd daryāft-shude rekhte.’ In Maqālāt-e Sherānī Vol. II. Lahore: Majlis-e Taraqqi-e Adab: 82–94. (Originally published in Oriental College Magazine, May 1931). ––––– 1940–41. ’Dā’ira ke mahdaviyõ kā urdu adab kī ta’mīr mẽ ḥiṣṣa.’ In Maqālāt-e Sherānī Vol. II. Lahore: Majlis-e Taraqqi-e Adab: 146-242. (Originally published in Oriental College Magazine, Nov. 1940–Feb. 1941). ––––– 1966. ‘Urdu kī shākh haryānvī zabān mẽ tālīfāt’ In Maqālāt-e Sherānī Vol. II. Lahore: Majlis-e Taraqqi-e Adab: 351–418. Sherani, Mazhar Mahmud (ed). 1966. Maqālāt-e Sherānī, 2 Vols. Lahore: Majlis-e Taraqqi-e Adab. Shukla, Ramchandra. 1994 [1929]. Hindī sāhitya kā itihās. Varanasi: Nagari Pracharini Sabha, 29th edition. Siddiqui, Mahmud Husain. 1979. The Memoirs of Sufis Written in India: Reference to Kashaf-ul-Mahjub, Siyar-ul-Auliya, and Siyar-ul-Arifin. Baroda: M. S. University. Simha, Dhirendranath. 1986. Ādhunik Hindī ke vikās me Khaḍgavilās Press kī bhūmikā. Satna: Bihar Rashtrabhasha Parishad.

Singh, Bacchan. 1989. Ācārya Śukla kā itihāsa paṛhte hue. Delhi: National Publishing House. Singh, Kamal. 1983. Gorakhnāth aur unkā hindī sāhitya. Muzaffarnagar: Sudha Kamal Prakashan. Singh, Kirpal (ed). 1962. Janama sakhi sri guru Nānaka Devaji/Meharbān Ji Sodhi. Amritsar: Sikh History Research Department, Khalsa College. Singh, Maharaja Pratap Narayan. 1894. Rasakusumākār, or a Book on Rhetoric: Rasakusumākār arthāt sāhitya kā ek anūṭhā granth. Allahabad: Indian Press. Singh, Samar Bahadur (ed). 1961. Abdurrahīm Khān-Khānā. Jhansi: Sahitya Sadana. Singh, Shivprasad. 1958. Sūrpūrva bhāṣā aur uskā sāhitya. Varanasi: Hindi Pracharak Pustakalay. Sitaram, Lala, Rai Bahadur. 1926. Selections from Hindi Literature, Ars Poetica (book V), and Other Poets (book VI, part II). Calcutta: University of Calcutta. Snell, Rupert. 1991. The Hindi Classical Tradition. A Braj Bhāṣā Reader. London: School of Oriental and African Studies. ––––– 1994. ’"Barvai" Metre in Tulsīdās and Rahīm’ in Alan Entwistle and Françoise Mallison (eds): Studies in South Asian Devotional Literature. Delhi: Manohar: 373–405. Sprenger, Aloys. 1854. A Catalogue of the Arabic, Persian and Hindûstāny Manuscripts of the Libraries of the King of Oudh, Vol. 1, Containing Persian and Hindu’sta’ny Poetry. Calcutta: J. Thomas. Srivastava, Savitri. 1981 [1976]. śrī Jagannāthadāsa ‘Ratnākara’ kṛt Gangāvataran kā kāvya-śilpa. Varanasi: Ratna Publications. Statement of Particulars Regarding Books, Maps etc. Published in the North Western Provinces and Oudh. 1894–1940. Also Quarterly List, variously titled. British Library: SV 412/38. Streusand, Douglas. E. 1989: The Formation of the Mughal Empire. New Delhi: Oxford University Press.

Sumersingh Sahabzade. 2000 [ca. 1881]. Srigur-pad prem prakāsh, krit Bābā Sumer Simh. Edited by A. S. Kahlom. Patiala: Punjabi University. ––––– 1987 [1897]. Atha Japu sidhānt (Atha japu sidhānt suryoday arathāt Srī Japujī dā bhāwa aratha, jiu sūraja caṛiyā: sammat 1954 [1897 ce] dī krit likhat-Bābā Sumer Singh Bhallā Jī walõ kāwi ṭīkā). New Delhi: Ishar Singh Nara. Tara Chand. 1939. Hindustānī kyā hai. New Delhi: Makatbah Jamiah. ––––– 1944. The Problem of Hindustani. Allahabad: Indian Periodicals Limited. Tehseen, Mir Muhammad Husain Ata Khan. 1958. Nau t̤ arz-e Muraṣṣa’. Edited by Nurul Hasan Hashmi. Allahabad: Hindustani Academy. Twining, Thomas. 1893. Travels in India A Hundred Years Ago, with A Visit to the United States. Being Notes and Reminiscences by Thomas Twining. Preserved by his Son Thomas Twining of Twickenham and Edited by The Rev. William H. G. Twining. London: James R. Osgood, McIlvaine and Co. Vajpeyi, Jagdish. 1964. Ādhunik Brajabhāṣā-kāvya: Hindī sāhitya ke ādhunik- kāl mẽ Braj-kāvya kā vikās. Muzaffaranagar: Ajanta Prakashan. Valdẽs, Mario J. and Djelal Kadir (eds). 2004. Literary Cultures of Latin America: A Comparative History. New York: Oxford University Press, 3 Volumes. Var, Dan Bahadur Pathak. 1965. Ratnākar kī sāhitya-sādhanā. Agra: Vinod Pustak Mandir. Vaudeville, Charlotte. 1955: Étude sur les sources et la composition du Ramayana de Tulsi-Das. Paris: Adrien-Maisonneuve. ––––– 1986. Bārahmāsā in Indian Literatures. Delhi: Motilal Banarsidass. Visharad, Ayodhyasingh. 1932. Vājīnd kī arelāṃ. n.p. Vyas, Kantiram Baldevram (ed.). 1981. Apabhraṃśa of Hemacandra. Ahmedabad: Prakrit Text Society.

Wade, Bonnie. 1997. Khyal: Creativity within North India’s Classical Traditions. Delhi: Munshiram Manoharlal. Wadley, Susan S. 1983. ‘The Rains of Estrangement: Understanding the Hindu Yearly Cycle’. Contributions to Indian Sociology, Vol. 17 (1): 51–85. Wordsworth, William. 1800. Preface to Lyrical Ballads, With Other Poems (Second edition). London: Longman and Rees. Yucel, Bilal (ed). 1995. Bābür dīvāni. Ankara: Atatürk Kültür Merkezi. Yusufi, Khurshid Ahmad Khan. 1990. Khumkhāna-yi Jāved. Edited by Kaifì Dihlavi and Brijmohan Dattatreyah, Vol. 1. Islamabad: Mauqtadirah-yi Qaumi Zaban. Zaidi, Ali Jawad. 1993. A History of Urdu Literature. New Delhi: Sahitya Akademi. Zvabitel, Dusan. 1961. ‘The Development of the Baromasi in the Bengal Literature’, Archiv Orientalni 29.

Notes

C

1

1 For a recent example of a history grounded in this belief in ‘composite culture’, and in culture’s unifying role, see Mushirul Hasan 2004. 2 See Christopher R. King 1994, Vasudha Dalmia 1996, especially ch. 10, Shamsur Rahman Faruqi 2001, and Alok Rai 2000. 3 I will not rehearse here the terms of the debate, for King, Dalmia, Faruqi and Rai have already defined them admirably. In this volume we have chosen to retain the terms Hindi and Urdu since they help to orient ourselves within the two broad literary repertoires that developed out of the same language. It would make little sense to start calling ghazals and marsiyas ‘Hindi’, or kabitt and savaiyas ‘Bhakha’. At the same time, as the volume will make amply clear, we believe that Hindi and Urdu literature existed within a common space, with significant overlaps and exchanges; that they are more to be understood in terms of register and affiliation to a literary repertoire than in terms of alphabet (since the script in which manuscripts are found depends on the transmission and subsequent life of a text); and that they should by no means be identified with religious communities. 4 For Orientalist understandings of the ‘vulgar speech’ of north India, see Bernard Cohn 1985, and Dalmia 1992, ch. 4, and David Lelyveld 1993. For the Orientalist invention of the term ‘Urdu’, J.B. Gilchrist’s role in promoting the term ‘Hindustani’ and a comparison with indigenous categories of Indian intellectuals such as Insha, see S. R. Faruqi 2001, ch. 1. For inconsistencies among colonial views of north Indian languages, see King 1994, ch. 10. 5 As Imre Bangha has done, showing also evidence of Khari Boli poetry in Nagari: Dadu 1637, Vajid 1651; see Bangha in this volume. 6 See Sheldon Pollock (1998), for a comparative discussion of ‘vernacularisation’ in India. 7 They are Shamsur Rahman Faruqi, A Long History of Urdu Literary Culture Part I: Naming and Placing a Literary Culture (2002: 805–63), and R.S. McGregor, The Progress of Hindi Part I: The Development of a Transregional Idiom (2002: 912–57).

8 Poets like Sheikh Ahmad Gujarati claimed knowledge of Telugu and Sanskrit. 9 S.A.A. Rizvi 1978: 350; also Muzaffar Alam 1996: 174. 10 ‘As is the case with minor poets or works’ Orsini writes about barahmasas, ‘the value of these intermediary genres does not lie necessarily in themselves, but in how they force us to consider a broader range of texts and tastes than those envisaged by the canon and thereby reconsider the possible options and choices north Indian poets had in front of them’ (pp. 150–51). 11 See e.g. Françoise Nalini Delvoye 1994: 407–29 and 1995: 93–130. See also Katherine Brown 2003 and 2006. 12 By contrast, Shantanu Phukan has rightly suggested that, at least until 1700, in north India we need to think in terms of a diglossic relation between Hindi and Persian. See S. Phukan 2001: 33–58. 13 See all the examples in Imre Bangha’s essay in this volume. 14 See Ramchandra Shukla 1994. For a critical study, see Bacchan Singh, Ācārya Śukla kā itihās paṛhte hue, 1989. 15 For Ramchandra Shukla’s idea of literature, see my The Hindi Public Sphere, 2002. 16 For further evaluation of Urdu literary histories, with particular reference to the problem of origins and beginnings, see Imre Bangha in this volume. 17 I am thinking here of R. Ceserani and L. De Federicis’s 5-volume history of Italian literature (1986), which developed insights first enunciated by Carlo Dionisotti in his essay ‘Geografiae storia della letteratura italiana’ (1967). And of Valdès and Kadir, A Comparative History of South American Literature (2004), which rejects national and linguistic boundaries and speaks instead of geographic factors, the formation of cultural centres, linguistic diversity and the production of literary cultures in the plural. 18 For the literary use of Persian during the early Sultanate, see Muzaffar Alam (1996). For Sanskrit inscriptions of this period see Pushpa Prasad 1990. For Apabhramsha texts of this period see Harivamsha Kochar (nd). Shivprasad Singh (1958) argued in his Sūr pūrva bhāṣā aur uskā sāhitya (Language and literature before Surdas), that Braj Bhasha literature developed out of a form of Shauraseni in the twelfth to fourteenth centuries. While spoken words (ukti) are preserved sporadically in Sanskrit linguistic texts, the earliest extant manuscripts of literary works date from the late fourteenth century: the earliest known Braj Bhasha work is a Pradyumnacarit, written in Vikram Era 1411 (CE 1354). 19 For Persian works written during this period, see S. Bilqis F. Husaini 1988.For Tomar Gwalior and its patronage of literature and music, see H.H. Dvivedi 1976, and F.N. Devoye 1992. 20 See his ‘Musicians and Scribes’ in W. Callewaert and M. Lath (eds) The Hindi Padavali of Namdev (1989). 21 See A.W. Entwistle 1987. 22 See Muzaffar Alam 1996.

23 Alternatively, as was later the case with Urdu barahmasas, marsiyas and rekhta, the vernacular was used for the female voice. 24 See also the evidence in Muzaffar Alam 1996. 25 For a spirited discussion of this ‘event’, see S.R. Faruqi 2001. 26 See Frances Pritchett 1994, and S.R. Faruqi 2001. 27 And the allegiance of Persian-educated Hindus to Urdu as well as Persian literature to the point of ‘forgetting’ other local vernacular literary traditions, as attested by Shantanu Phukan (2001). The neglect of poems like Afzal’s Bikaṭ kahānī can be attributed to the same phenomenon; see Imre Bangha in this volume. 28 See Lalita du Perron in this volume; for a historical outline see Allyn Miner 1993. 29 See for example, Kathryn Hansen (1992, 2003) on the language of Nautanki and of Parsi theatre.

C

2

1 Abdul Haq 1961: 22. Quoted in Alok Rai 1991: 264. 2 See also Orsini (forthcoming) on nineteenth-century Barahmasa printed books. 3 The latest and most coherent example of such efforts is Faruqi 2001. 4 Jafar and Jain 1998 vol. I: 371. 5 Sahil Bukhari, ‘Urdū kā qadīmtarīn adab’. Nuqūsh 105 (May 1945) and Urdū kī kahānī, Lahore, 1975: 160–98. Quoted in Jafar and Jain 1998, vol. I: 451. 6 A shorter version of this essay appeared in Pollock 2002. 7 This Hindi expansionist tendency is refuted by Faruqi, who examined the origins of Urdu as independent of the Hindi and indeed Hindu tradition and labelled Rai’s work as ‘full of tendentious speculation’ and dismissed it in a footnote. Faruqi 2001: 1, n1. 8 Kamal Singh 1983: 56. 9 The word Hindavi in this article refers generally to the varieties of Hindi and Urdu prior to the articulation of their separate identity. 10 Faruqi 2001: 71. 11 This is what was theorised by Mir in the eighteenth-century; see p. 27 below.

12 ‘Every farsi—farsi being consonant with Hindavi khyal—in which one sets the words of both languages [Persian and Hindavi] to a raga and a tala is called rekhta. This rekhta is composed in every parda [musical mode], [thus] increasing pleasure and enjoyment’. Tr. based on that by Katherine Butler Brown 2003: 240. The original reads: ‘ Har fārsī ki ba maẓmūn-i khyāl-i hindavī mutābiq bāshad va alfāz̤ -i har do zabān-rā dar yak tāl va yak rāg bar bast namūda bāshand… ān-rā rekhtā-rā dar har pardā mibandand va zauq va lazzati afzūn mīdahanď. Shahab Sarmadee 1996: xxxii–xxxiii. Farsi at this stage was a musical composition sung in the ‘Khusravi’ style of qawwals in Delhi (i.e. that of qaul and tarana later known as ‘qawwali’: see Brown 2003: 239). 13 Sherani 1926: 3. 14 E.g. Guftagū rekhte mẽ hamse na kar, yah hamārī zabān hai, pyāre. [Kulliyāt- e Mīr I. 548] ‘It is my own tongue, my dear, don't contend to me in rekhta.’ (Tr. Faruqi 2001: 23). 15 Sar sabz-e hind hĩ mẽ nahĩ kuch ye rekhte; hai ghūm mere shi’r kī sāre dakhan ke bīc [Kulliyāt-e Mīr II. 790]. ‘My poetry grows green not only in the northern planes. In all the Deccan too the praises of my verse resound.’ (Tr. Islam and Russell 1969: 215) 16 ‘Rekhta ki shi’r ast ba-ṭ’aur-i shi’r-i fārsī ba-zabān-i urdū-i mu ‘allā-i shāhjahānābād- i dehlavī’.; Khan, n.d: 1. 17 Khan n.d: 186–87. 18 Zaidi 1983: 30; this type can also be conceived as a variety within Mir’s first category. 19 Ugo Enrico Paoli, Il latino maccheronico, Firenze 1959. Quoted in Phukan 2001: 42. 20 E.g. Jalibi 1977: 19–83; Zaidi 1983: 19–35; Jafar and Jain 1998, vol. I: 371–462, vol. V: 9–97. 21 Vyas 1981: 8. 22 This text is quoted on the basis of Partab Singh’s album in Sherani 1931: 88. Panchal (2001: 75) gives a variant version without referring to an original source. The final couplet, for example, is as follows: ‘Ba-ḥaq ām mah ki roz-e maḥshar, badād mā-rā fareb Khusrau. So pīt man kī durāya rākho, jo jāya pāū̃ piyā kī khatyā. 23 The meaning of the last line is unclear. 24 See Jafar and Jain 1998 vol. 1 419–22. 25 Zaidi 1993: 28. Jalibi (1977: 35) and Jafar and Jain (1998, vol. I: 420) give a slightly different text based on ‘Abdul Qadil Sarvari: ‘Aligarh tārīkh-e-adab-e- Urdū, pp. 148–49. Jalibi says that the poem was found in an old album at the Anjuman-e Taraqqi-e Urdu (Pakistan). Neither Zaidi nor Sarvari has given any indication of source. 26 See Jafar and Jain 1998, vol. I: 383; 422–28; Zaidi 1993: 19 and Rizvi 1978, vol. I: 327–28. 27 Callewaert and Lath 1989: 360.

28 Baba Farid’s earliest documented poems as quoted by Bajan (1388–1506) and in the Guru Granth (1603–1604) do not show the systematic use of Khari Boli features with which some later authors credit him (see Jalibi 1977: 27, 35). As far as Maneri is concerned, two independent collections of his spiritual discourses agree that the mystic was inimical to Hindavi singing because of the frankness of its expression (see below). Although Namdev’s songs are present in manuscripts from the second half of the sixteenth century, namely in the Goindwal pothis and in the Fatehpur manuscript, the padas with substantial Khari Boli elements and Perso-Arabic vocabulary are not there. The song quoted above, for example, was found only in an undated copy of a manuscript from Pandharpur. The poem found in the undated Pandharpur manuscript is taken into the critical edition by Bhagirath Mishra and Raj Narayan Maurya (eds): Sant Nāmdev kī Hindī padāvalī, Pune 1964. Similar claims to Kabir’s Khari Boli poems (e.g. Rahanā nahi desa, birānā hai or Sumirana bina gotā khāoge, see Rai 1991: 157–59) cannot be confirmed due to the lack of early written evidence, since none of these poems is present in the earliest available manuscripts dating from between 1570 and 1681 (see Callewaert 2000). In his History of Rajasthani Literature, Hiralal Maheshwari (1980: 9) claims that occasional Khari Boli usages (sometimes mixed with Punjabi) occur in Rajasthani bardic literature such as in Badar Dhadhi’s (fl. c. 1450) Vīr māyan written in the nisani metre and in Sandu Mala’s (1573–1679) jhulnas (Mahārājā Rāysinghjī rā, Akbar Pātsahjī rā, Dīvān Pratāpsinghjī rā etc.) as well as in the arillas (also called chandrāyanas) of Kesaudas Godara and Vajind and in the songs attributed to Qazi Mahmud (fifteenth century). Most of these works are still inaccessible. There are, however, indications that Maheshwari’s statement should be treated with caution. The Vīr māyan, for example, dates not from the fifteenth but from the nineteenth century (Menariya 1978: 222). In spite of the fact that Vajind or rather Vajid was the author of a Rekhta composition (not mentioned by Maheshwari), his arillas do not show Khari Boli features. 29 Sherani 1931: 88. 30 The line can also be interpreted as ‘I swear by the love of that coquettish, unkind Fate, who took away my patience.’ 31 Sherani 1931: 91–94. 32 Bailey 1930: 205–208 based on works by Muhammad Sherani, Shamsullah Qadri and Abdul Hay Nadvi. See also Jalibi 1977: 26–29; Orsini 2007. 33 The manuscript of this work is found in the Habibganj Collection, Aligarh Muslim University. See also Nisar Ahmad Faruqi, 1989: 26–27a. 34 Ziya-ul-Hasan Faruqi 1996: 426. 35 Literary histories also refer to the malfuzat containing the sayings of Baba Farid (e.g. Zaidi 1993: 19). It is recorded, for example, that Baba Farid, who is supposed to have used Hindavi in his conversations when living in Ajmer and Delhi, used to call a friend ‘bhaiya’ (Bailey 1930: 205). The malfuzat containing his sayings, Ali Asghar’s Javāhir-i Farīdī, dates from the Mughal era, written three hundred years after the time of the Shaikh with the aim of glorifying the spiritual achievements of Baba Farid and his descendants (Rizvi 1978: 13). 36 Nizamuddin Auliya stated that Nahravani had been present at the sama’ gathering in 1235 where a Persian verse produced such powerful extasy in Shaikh Qutbuddin Bakhtiyar Kaki that he died a few days later. Shaikh Bahauddin Zakariya (1182–1262) also talked with respect about him. See Faruqi 1996: 325–26.

37 Faruqi 1996: 326. 38 Faruqi 1996: 426. 39 The first line of this couplet seems to be corrupt. The metre is close to that of a dohā. 40 Surūr aṣ-Ṣudūr p. 69, quoted in Rizvi 1978, vol. I: 327. 41 On p. 74: Birale cīna jo rogina gaī jaugina karī guna gaī ko dosa, ayana rasāyanasaṃcarai raṃga jo mārai osa. On p. 302: Ausadhi bhejana dhani gaī ou bhaī birahīna; ausadhi dosa na jānaī nāri na cetai tīna. Both are quoted in Rizvi 1978, vol. I: 327. 42 The same phenomenon is recorded in the sixteenth century in Mir Abdul Wahid Bilgrami’s Haqā’iq-i Hindī (1566). In this work Bilgrami interpreted the Radha-Krishna imagery in a Sufi light and the examples of Hindavi quoted are in Braj Bhasha. 43 I owe this last suggestion to Allison Busch. 44 A study of their contribution to Urdu can be found in Sherani: ‘Dā’ira ke mahdaviyon kā urdū adab kī ta’mīr meñ ḥiṣṣa’ In Maqālāt-e Sherānī vol. II: 146–242. On the beginnings of Mahdism and its connection to Jaunpur see Rizvi 1965: 74–75, Maclean 2003. 45 Jafar and Jain 1998, vol. I: 428–33. 46 The poem is found on the margin of Miyan Valiji’s Inṣāf-nāma (p. 217), quoted in Jafar and Jain 1998 vol. I: 429. The metre is mujtass, musamman makhbūn mahzūf (u – u /uu ––/u – u /––). 47 Abdul Quddus Gangohi, Rushd-nāma (Alakh-bānī), Aligarh, 1971. 48 Rizvi 1978, vol. I: 284. 49 Rizvi 1965: 107–10. A couplet attributed to him was found by Sherani in a manuscript of a work called Zād al-fuqarā in possession of Sayyid Khayruddin Sahib Wakil Thakanjat, Jaipur. See Sherani 1940–41: 199. Jafar and Jain 1998, vol. I: 433–34. Jalibi (1977: 134) found the same couplet in an old album in possession of Afsar Siddiqi. 50 Rizvi 1965: 114–18. 51 Yücel 1995: 500 and f. 88b. This is a critical edition based on ms Nr. 3743 at the University of Istanbul Library and dated 1265AH/1848/9 CE. The poem is missing from the earliest manuscript of Babur’s Dīvān preserved in Paris and dated from 1515 CE. The edition mentions that this couplet is on folio 17b in the 1529 Rampur manuscript. 52 The identification of the poets with known personalities is not without problems. The names of Sedan and Ja’far do not figure anywhere else in literary histories. (There is however, a Braj Bhasha work on conjuring called Adbhūt vilās written in 1638 and attributed to a certain Mira Sedan Guhar who may be identical with our author. A manuscript of this work is mentioned in Menariya 1942, vol. III: 228–29.) Three other names present a different challenge since there were several poets with the names Faizi, Bairam and Jani.

53 Sherani 1931: 76. 54 On Jamali and his Siyār ul-‘ārifīn see Siddiqui 1979: 82–98. 55 Jalibi 1977: 52 says that the poem is found in a manuscript album (nr. 3/633) at the Anjuman-e Taraqqi-e Urdu, Karachi. This might well be the same album since it does not figure in the catalogue of Sherani’s manuscripts donated to Punjab University. See Husain 1968. 56 There is a work play here. If a loincloth can be tied in several layers, it means that its wearer is extremely emaciated. The idiom ‘to tie a loincloth’ also means to renounce the world. 57 Sherani 1930: 16; Jalibi 1977: 52. 58 The poems appear in an album which was possibly written between 1556 and 1572 and was found by Nazir Ahmad at the Library of Habibganj. See Nazir Ahmad: ‘Salāt̤ īn-i mug̱aliya kā nayā kalām.’ Fiqr-o naẕar-e alīgarh, Jan 1963, quoted in Jafar and Jain 1998, vol. I: 444. Sherani (1931: 78–79) quotes f. 183 of the manuscript of Saqqa’s Dīvān as the source of the poem but does not give the details of the manuscript. Jafar and Jain (1998, vol. I: 444) mention that copies of this Dīvān are found in the Khuda Bakhsh Library, Patna and at the Bhandarkar Oriental Research Institute, Poona. 59 Emendation from kī. 60 Emendation from katāra. 61 Emendation from tū ci kunī. 62 Zaidi (1993: 29) gives a metrically correct variant reading of the first two couplets, but his source is not indicated. 63 This word is problematic. The metre would require two short syllables and a long one. 64 The original reads bartī; the emendation is by Jafar and Jain. 65 Jafar and Jain 1998, vol. I: 444 and Zaidi 1993: 29. 66 Ibid. 67 Mas’ud Hasan Rizvi, ‘Mirāṣī-yi rekhta’, in Taḥrīr-i Dillī. April-June 1971: 8. Quoted in Jafar and Jain 1998, vol. I: 445. 68 Ross 1910. 69 Emendation from the sakht-tar. 70 This word is hypermetrical here. 71 Sherani 1931: 84. 72 An alternative meaning is: ‘People who protected millions are left empty-handed.’

73 I was not able to interpret this word. 74 This must be a reference to the legendary, pre-Islamic sage, Luqman. Luqman’s name is linked to many proverbs and fables. I was not able to find a reference to this story of Luqman’s fencing of the road. 75 Hadi 1995: 146. 76 Sherani 1931: 83. 77 Sherani, for example, mentioned that he had seen other Rekhtas attributed to Abu’l Fazl and to Faizi (1931: 83). 78 Mishra and Rajnish, 1985: 173 and Chauduri 1954: 22. Neither of them mention their source. 79 See e.g. María Angeles Gallego, ‘The languages of medieval Iberia and their religious dimension’, Medieval Encounters 9, 1, 2003: 107–39 and Claudio Giovanardi, ‘Il bilinguismo italiano-latino del medioevo e del Rinascimento’, Storia della lingua italiana, vol. 2, Lo Scritto e il Parlato, Torino: Einaudi 1994: 435–67. I owe this information to Francesca Orsini. 80 Subrahmanyam 1999: 280–321. 81 The Telugu play Annadāmahānātakamu, for example, uses colloquial Tamil. Rao 1992: 334. 82 The Telugu part seems to be a corruption of āḍĕ-po vaccinḍu. (Oḍipo vaccinḍu means ‘he came to be defeated’.) 83 Literally ‘This is the accumulation of my karma’. 84 Translated on the basis of the Khari Boli transliteration in Mishra and Rajnish 1985: 174. 85 Surveys of sixteenth century Rekhta in north India can be found in Jalibi 1977: 51–61, Zaidi 1993: 28–35 and Jafar and Jain 1998, vol. I: 428–50. 86 Annemarie Schimmel, ‘The Influence of Sufism on Indo-Muslim Poetry,’ in Strelka, Joseph P. (ed.), Anagogic Qualities of Literature. University Park: Pennsylvania State University Press, 1971: 196. Quoted in Eaton 1978: 143. 87 The same idea occurs in Zaidi 1993: 20. 88 Alam 1998: 349. 89 On Eaton, see Phukan 2001: 38, note 11. Alam did not emphasise the secular aspect of Persian in Alam 2002. 90 Guha 2004: 21–22. 91 See Allison Busch and Christina Oesterheld in this volume. 92 See Phukan 2001.

93 ‘Ishqi Khan, a descendant of the Turkish spiritual guide Isma’il Tash, was a mir munshi during Akbar’s reign and authored a Persian Dīvān. In his Persian qasida, Sard-o-garm-i zamāna, he used some Hindi and Turki stanzas. Haq 1931: 101. 94 Haq 1931: 101. An alternative interpretation of the last expression is, baci kār ‘what are you doing?’ 95 Emendation from savār. 96 Haq 1931: 102. 97 Phukan 2001: 33–58. Giovanardi 1994, however, distinguishes between ‘pedantic’ texts of a serious nature, which combined the morphology of the vernacular with vocabulary from classical Latin, medieval Latin and the vernacular, and macaronic poems whose phonology and morphology looked like Latin for parody. I owe the latter reference to Francesca Orsini. 98 Bahr ul-ma‘ānī (Ocean of Meanings, unnumbered manuscript in the Fatuha (sic!) collection in the Khuda Bakhsh Library, Patna, 759 AH/1358), translated in Jackson 1987: 111. 99 Mukhkh ul-ma‘ānī (The Core of Spiritual Realities), Agra: Mufid-i Am Press 1903: 154. Translated in Jackson 1987: 134. 100 Jackson 1987: 135. 101 While discussing the early use of Hindavi in sama’ gatherings, Athar Abbas Rizvi, the author of the monumental A History of Sufism in India, voices a similar opinion claiming that Hindavi songs ‘were not composed for propaganda purposes but were a natural evolution from the deep and personal involvement of … mystics with their environment. Hindavi was a more convenient language in which to utter the feelings of a heart filled with divine love’. Rizvi 1978, vol. 1: 327. 102 Jāmi al-kalām: (Compendium of works) Khwājā Gesūdarāz Sayyid Muhammad Akbar Husainī kī vānī. Intizāmī Press, Usmānganj, 1937/8: 172–73. Trans. into English in Rai 1991: 121. 103 Ibid. Also Phukan 2001: 34–35. 104 Guha 2004: 20. 105 Pollock 2001: 20. 106 Alam 2002: 172–74. 107 Beach 1992: 41. 108 Dādū Mahāvidyālay No. 12. It is also present in mss from VS 1715, 1733, 1765 and 1770. On the solid manuscript tradition of Dadu see Callewaert and Beeck 1991, vol. I: 14. 109 Callewaert and Beeck 1991, vol. I: 263. Further examples include padas 75, 81, 84, 271 etc. 110 Callewaert and Beeck 1991, vol. I: 213.

111 Callewaert and Beeck 1991, vol. I: 517. 112 On Vajid and his Rekhta see Bangha 2005. 113 Apart from two padas and some 180 sākhīs (dohās and arillas) in the Sarvāṃgī of Gopaldas the only published poems of Vajid are his arillas in modern collections such as the Pañcāmṛt (Mangaldas 1948). Indeed it is his arillas that, even today, are current as popular sayings in Rajasthan. (Maheshvari 1980: 126). 114 Raghavdas: Bhaktamāl 428; Nahta 1965: 201–2. 115 Ayodhyasingh Visharad, 1932 Vājīnd kī arelāṃ. Mangaldas, 1948, Pañcāmrt. Dadu 1976 (VS 2033); Maharṣi Vājidjī Mahārāj ke arill; Rajnis, 2004, Pañcāmṛt aur Pañcrang: Madhyakālīn santkaviyoṃ kā prāmāṇik pāth-sampādan: 43–72. 116 Osho, 1995, Kahai Vājid pukār: Ośo dvārā Vājīd vāṇī par die gae das amṛt pravacanõ kā apratim saṃskāraṇ. 117 More than sixty of his compositions can be found in the City Palace collection in Jaipur. We have considerable manuscript material from his lifetime or from right after his death. His earliest manuscript, containing his eight works is dated from 1600 and we have a manuscript of his Guṇ Gañjanāmo from 1613, of his padas and Guṇ ajāib-nāmo from 1636. A critical edition of his works is being prepared by Daksha Mistry and myself. 118 Mangaldas 1948: p. ka. 119 Pothikhana 2422, 3404, Rajasthan Oriental Research Institute, Jodhpur no. 10902(1) and Hindi Sahitya Sammelan, Allahabad no. 2145(3): 1354. 120 Some phrases are addressed to Krishna and the last line is directed to Uddhava. It is possible that the cowherd-women unconsciously address Uddhava as Krishna. Such double vocatives are also found in Surdas. 121 Malukdāsjī kī vānī 1912: 30. The transmission history of Malukdas’s poems has not yet been studied critically. There is a possibility that some of his songs belong to later poets of the same name. 122 Ibid: 6. 123 Singh 1962, Rajguru 1974. 124 Shastri 1989: 4. 125 For Svami Haridas, see Rosenstein 1997. 126 Rosenstein 1997: 108. 127 Rosenstein 1997: 189. 128 ‘His actual poem may be earlier, but the only datum we have about the poet is 1625, the year of his death, though it is generally assumed that he completed the mas̤ navī not long before his death.’

Faruqi 2001: 113. 129 Hashimi and Khan 1979: 18–22. 130 After Mas’ud Sa’d Salman the genre was neglected in Indo-Persian tradition. See Orsini in this volume. 131 Faruqi 2001, p. 117. 132 Khan 1965, pp. 418–9 quoted in Rai 1984, p. 123–4. 133 See Orsini’s essay in this volume. 134 Probably the Urdu form of the Braj vinaya. 135 Zaidi 1993: 21; Jafar and Jain 1998, vol. V: 93–94 and Khan 1966: 130. 136 Sẽ is the reading of Jafar and Jain 1998, vol. V: 93. Khan 1966 has the uninterpretable saṣ here. 137 Khan 1966: 130. 138 The interpretation of the tags at the end of both lines is problematic. 139 E.g. Har kas ki khiyānat kunad albat̤ t̤ a bitarsad; Bīcāra-yi Nūrī na kare hai, na ḍare hai. (Everyone who does treachery is certainly afraid; this poor Nuri does not do it, and he is not afraid.) Jalibi 1977: 59, on the basis of Qa’im Chandpuri’s tazkira Makhzan-i nikāt (Storehouse of subtle points), Anjuman-e Taraqqi-e Urdu, Aurangabad 1929: 3. Also quoted in Sherani 1931: 87. This poet is not identical with Shaikh Muhammad Nur, the author of a prayer in Hindavi published in Sherani 1930: 244–46. 140 As attested by a marsiya of Sauda (Kulliyāt II, Allahabad 1971: 376). See Phukan 2001: 45 note 22. Ghazals attributed to Munshi Vali Ram Vali, a Persian poet at Shahjahan’s court, follow a similar blend but avoid Braj Bhasha forms. See Khazīnā al-’ulūm (Treasury of the sciences) by Durgaparshad Nadir (1879), quoted in Sherani 1930: 235. Also quoted in Jalibi 1977: 71, Zaidi 1993: 64–66, and Jafar and Jain 1998, vol. V: 35–36. 141 Fatah Muhammad was the author of a book on religious and Sufi beliefs, the Fatḥ ul-‘aqā’id (The Triumph of Religious Tenets) and a tract on prayers, the Miftāḥ uṣ-Ṣalāt (Key to Blessing). See Hadi 1995: 117. 142 Sherani 1931: 86–87. The poem was found by Sherani in Jaimal’s album dated 1652–56. 143 On Brahman see Rizvi 1978, vol. II: 414, Jafar and Jain 1998, vol. V: 36–38. 144 Yusufi 1990, See also Jalibi 1977: 72, Zaidi 1993: 65 and Jafar and Jain 1998, vol. V: 37, who also discuss Hashimi and Khan’s views. 145 Jalibi’s text has a lacuna here and Jafar and Jain have bā. The conjecture karnā is by C. M. Naim.

146 Jalibi’s text has karchī here. Jafar and Jain suggest karcī. The emendation karchā is by C. M. Naim. 147 Quoted in Jalibi 1977: 72, from an old handwritten album preserved at the Anjuman-e Taraqqī-e Urdū, Karachi. No further reference is given. 148 See the Hindavi verses attributed to Sufis like Shaikh Junaid (fl. 1618), Shaikh ‘Usman Jalandhari (fl. 1625), Shaikh Faziluddin Batalvi (d. 1738), Sayyid Atal Narnauli, Shaikh ‘Ata Banka, as well as the stray poems by Mir Ja’far Zatalli (1659?–1713), Mirza Mu’izuddn Muhammad Musavi Khan Fitrat (1640–89), Naushah Ganjbakhsh; Sherani 1930: 240–41, Jalibi 1977: 81, Jafar and Jain 1998, vol. V: 86–97. See also the marsiyas in the albums described by Ma’sud Hasan Rizvi Adib (Shumālī hind kī qadīmtarīn urdū naẓmẽ , Lucknow, Adabistan, 1984; also Faruqi 2001: 114–5) or those attributed to the notoriously difficult Indian-Persian poet Shaikh Nasir ‘Ali Sirhindi (d. 1697); Sherani 1930: 232–33 and 242–43, Jafar and Jain 1998, vol. V: 34–35, 52–79. Even the most famous Indo-Persian poet of his times, Mirza Bedil, is credited with some couplets in Mir’s tazkira Nikāt ush-shu’arā, but their language is too modern to be Bedil’s (Jafar and Jain 1998, vol. V: 88–89). 149 Jalibi 1977: 79 and Sherani 1930: 236–39. See also Jafar and Jain 1998, vol. V: 38–45. Sprenger (1854: 617), in his description of an undated manuscript in the Royal Collection of Avadh (nr. 644) mistakenly identifies the Fiqh-e hindī with ‘Ālam’s Maḥshar-nāma. See Jafar and Jain 1998, vol. V: 48. 150 Jafar and Jain 1998, vol. V: 45–51. 151 Jafar and Jain 1998, vol. V: 49. The earliest manuscript of the Maḥshar- nāma dates from 1745 (1158AH). 152 Jafar and Jain 1998, vol. V: 50. 153 Ibid: 52–57. 154 Faruqi 2001: 113; Jafar and Jain 1998, vol. V: 79–86. This masnavi was published by ‘Abdul Haqq under the title Vafāt nāma-ye bībī fātima (Death Chronicle of the Lady Fatima) in Risāla-ye Urdū Karachi, April 1951. 155 Jafar and Jain 1998, vol. V: 83. They suggest that in spite of this statement Isma’il Amrohvi must have lived in the south, where his literary language imbibed its Dakkani characteristics. 156 The dictionary of the Capuchin Friar François Marie prepared in Surat in 1703, the Thesaurus Linguae Indianae, is a dictionary of Gujri with a significant component of Perso-Arabic vocabulary but the Hindavi words are represented in the Nagari script thus hinting at further Nagari-Rekhta connections in Gujarat. McGregor 2003: 947–48. A manuscript copy of the dictionary is at the Bibliothèque Nationale in Paris. 157 Jafar and Jain 1998, vol. IV: 293–95. 158 Zahiruddin Madani, Gujrī mas̤ naviyā̃ , 1990: 25–26, quoted in Faruqi 2001: 113. 159 Faruqi 2001: 113–14.

160 McGregor 1984: 141–42. Barthwal 1978: 260–62. 161 A manuscript in the Persian script is found at the Asafuddaula Public Library, Lucknow and a portion of it at the headquarters of his sect in Panna. 162 Ram 1996: 124. 163 A Rekhta in a manuscript of 21 folios is attributed to Maharaja Jaswant Singh of Jodhpur in the catalogue of the Hindi Sahitya Sammelan, Allahabad (ms no. 3–193/2145–3). 164 He also wrote stray Rekhta as well as works with titles that suggest a high Khari Boli input such as Vasant kī khabri hai and Vaid Phajīhatī. 165 McGregor 1984: 205. 166 E.g., Taba gujarāta se āye dīva mẽ bhāī sāthī jayarāma ke ghara uṭha mile ānanda so baṛo sukha pāyo dekhkar. (Then he arrived from Gujarat in Dīv to the house of his brotherly companion, Jayram. He came to receive him with joy and was overwhelmed with delight to see him.) Bīṭak 20,3 quoted in Ram 1996: 180. 167 E.g., Bhīmasyāma bhaṭṭa sunī karane āye dīdāra Caracā ita baṛī bhaī uno kiyā baṛā pyāra. (Having heard this Bhimshyam Bhatt came to see him. They had a long discussion and he showed him great affection.) Bīṭak 21,25 quoted in Ram 1996: 181. 168 Most of this Rekhta is still unpublished and the following survey is based on the various manuscript catalogues printed by the Rajasthan Oriental Research Institute, Nagari Pracharini Sabha (Varanasi), Hindi Sahitya Sammelan (Allahabad), Bihar Rashtrabhasha Parishad (Patna) and other smaller centres. 169 On Nagaridas and the literary atmosphere at Kishangarh-Rupnagar see Bangha 2007. 170 Gupta 1965, vol. II: 49. 171 Bangha 2001: 175–90. 172 For the date of birth, see McGregor 1984: 145. Barthwal 1978: 265–66. For the date of his death see Mishra 1972: 160. 173 Mishra 1972: 164. 174 Dariyā Sāheb 1913: 9–19. 175 McGregor 1984: 145. 176 Dariyā Sāheb 1909: 3. The first half of the second line is hypometrical. 177 McGregor 1984: 146, Menariya 1978: 295–96. Barthwal 1978: 266–67. 178 ‘Cetāvanī’ shabda 23 (Caraṇdās 1908 pp. I. 113–14). ‘Cetāvanī’ sabdas 22, 25 (Caraṇdās 1908 pp. I. 112–15) are similar Rekhta poems. In his work about the development of Hindi language and

literature Hariaudh, usually attentive to traces of modern Hindi, did not notice that Charandas used Khari Boli. Hariaudh 1934: 455. 179 E.g. there are several poems attributed to Ramdas of Marwar (1726–1798), who after trying to follow the teachings of eleven various gurus found his spiritual guide in Hariramdas of the Ramsnehi school. Rasik Govind was the author of Kalijug rāsau (1807) a longer composition classified as nītikāvya, poetry on morals. He is not identical with the Sikh Guru Ramdas (b. 1534, see McGregor 1984: 55). See Bhagirath Mishra 1972: 499. 180 Such as e.g. (Jan) Haridas (d. c. 1645), Tursidas, Mohandas, Hariram (maybe identical with Hariramdas and/or with Kisna), Gangadas, Kisandas (maybe identical with Kisna) and Sevadas (1640?–1741?), the successor of Tursidas. For (Jan) Haridas see Menariya 1978: 307, for Sevadas see Chaturvedi 1957: 221–23. 181 Some Rama worshippers also composed Rekhta, e.g. Tulsi Sahab (1763–1843) of Hathras, who considered himself a reincarnation of Tulsidas, and (Svami) Ramcharan or Ramcharandas (b. c.1760), who produced many works such as a commentary on the Rāmcaritmānas. See Gupta 1942, p. 74 and McGregor 1984: 170. The manuscript catalogues mention Rekhta by various authors about whom nothing can be known, such as Ratanlal, Kalramji Maharaj, Vallabh Gosvami, Daulesh and Jan Dhiram. 182 Rekhta 2 in Bullā sāheb 1910: 23. 183 McGregor 1984: 144. His autobiography is claimed to be Rekhta by a cataloguer of the Nagari Pracharini Sabha, Saṃkṣipt vivaraṇ II: 332. The selection as published in the Hindī sāhitya kā bṛhat itihās, however, does not show any input of Perso-Arabic vocabulary but rather a mixture of Braj Bhasha, Khari Boli, Avadhi and Punjabi thus nearing Sadhukkari. Examples are found in Chaturvedi 1968: 239–40. 184 Bhīkhā sāhab 1909: 61. 185 Rāmjī kā bārahmāsā ms no. 4241/1 at the Hindi Sahitya Sammelan, Allahabad. 186 Bansal 1988: 20–21—quoted from an undated manuscript (no. 4487- GI at the Vrindaban Research Institute). 187 The text is unclear at this point. 188 Mishra 1952: 180. 189 Published in Albelisharan and Sharma 1999: 113. 190 In battle descriptions for example he imitated Pṛthvīrāj Rāsau both in dialect and verse form. Chaturvedi 1968: 5 and 389. His other known work, the Durgā-bhakti-taraṃginī (1712) is based on the Sanskrit Durgā Śaptasatī. 191 Ms no. 27787 (9) at Jodhpur RORI ff. 106–08 (copied in VS1857). 192 Ms no. 9720 (5) at Jodhpur RORI ff. 14–16.

193 Ms no. 30075 (1) at Jodhpur RORI ff. 1–5 (copied in VS1916). 194 Four manuscripts are found in the Khasmohar Collection in the City Palace of Jaipur (Nr. 2367, 2373, 7799(2), 1546(1) and one in the Wellcome Library, London (Ms. Hindi 365.03)–erronously given under the name of Ghananand in the catalogue). His Kavitta ratnamālikā saṃgrah comprising 801 poems by earlier authors and 108 by Rasrashi was compiled for Singi Jivraj, Pratap Singh’s divan. The Mishrabandhu Vinod (p. 839, no. 950) puts his floruit around 1770 (VS1827). 195 RORI, Jodhpur Ms. no. 10847(2) 33–4. 196 This work on the Bhramargīt theme is preserved in manuscripts at some branches of the Rajasthan Oriental Research Institute: Jodhpur 9720(1) ff 1–5, Udaipur 31, ff 1–9, 465 ff 1–12 and 3029 ff 1–3. His other preserved works are his Prem Patrikā and Chandāṣṭak. Jodhpur RORI 9720 (3–4). 197 Also called Virahī-subhān-dampati-vilās. See Mishra 1974: 2. 198 Mishra 1974: 6.

C

3

1 Putting quotation words around the word ‘Hindi’ at every usage, or awkwardly appending it to ‘Urdu’ in a hyphenated compound, quickly becomes tedious. Acknowledging here the problematic nature of the word Hindi—particularly as a designation for the language(s) of pre-modern north Indian literary cultures—I will henceforth dispense with extra punctuation. 2 R.S. McGregor’s studies of Vishnudas make it clear that he considers the literary language of the Tomar court at Gwalior an early form of Braj Bhasha. See, for instance his ‘A Narrative Poet’s View of his Material: Viṣṇudās’s Introduction to his Brajbhāṣā Pāṇḍav-carit (AD 1435)’, in Mariola Offredi (ed), The Banyan Tree, 2000, vol. 2: 335–42. 3 If this is so, then it is anachronistic to speak of a Braj language or literature before the sixteenth century. In this vein, Hariharnivas Dvivedi rejects the designation Braj Bhasha and makes a case for ‘Gvaliyari’ or ‘Madhyadeshiya’ as more authoritative terms for early Hindi. See his Madhyadeśīya Bhāṣā, Gwalior 1955. Shivprasad Singh, for his part, proposes different stages in the development of Braj Bhasha. He posits a ‘transitional Braj Bhasha’ from the mid-twelfth to mid-fourteenth centuries before Braj develops into a fully-fledged language from c. 1350–1550. Shivprasad Singh, Sūr-pūrv brajbhāṣā aur uskā sāhitya, 1958: 71–237. 4 Bhikharidas, Kāvyanirnay, in Vishvanath Prasad Mishra, (ed), Bhikhārīdās granthāvalī, 1957, vol. 2: v: 14. 5 R.P. Dewhurst, ‘Persian and Arabic Words in the Satsai of Bihari Lal’, Journal of the Royal Asiatic Society, Part 1, 1915: 122–23.

6 Compare the attitudes of Ramchandra Shukla (1994: 132–33) who generally looks unfavourably upon Persianising poets of the Braj tradition (although he is somewhat lenient when it comes to Bihari). 7 A good example of this perspective is Kishorilal 1971: 480–83. 8 See Krishna Divakar 1969: 437. 9 Jagdish Gupta 1961: 119–25. 10 Shantanu Phukan has usefully cautioned against religiously overdetermined approaches to old Hindi Sufi texts. See Phukan 2000: 22–26. 11 Biharilal, Satsai, in Bihārī, 5th ed., Vishvanath Prasad Mishra 1965: 230, v. 561. 12 Jasvant Singh, Bhāṣābhūṣaṇ, in Jasvant Siṃha granthāvalī, Vishvanath Prasad Mishra (ed), 1972, v. 204. 13 Keshavdas, Rasikapriyā, Keśavgranthāvalī, vol. 1, Vishvanath Prasad Mishra (ed) 1954: vv. 2.1– 2. 14 Rasikpriyā: v. 2.5. 15 The Orchha kings traced their lineage back to a branch of the Gahadavala dynasty with a connection to Kashi. See Kavipriyā, Keśavgranthāvalī, vol. 1: vv. 7–8. 16 Keshavdas, Ratnabāvanī in Keśavgranthāvalī, vol. 3 Vishvanath Prasad Mishra (ed),1956: v. 8. 17 Some aspects mentioned here of the Ratnabāvanī, and of other historical works by Keshavdas (discussed below), draw on my earlier article, Busch 2005: 31–54. 18 But even Keshavdas’s highly ‘kavyaesque’ works can exhibit a degree of internal variation. Scenes more geared toward action or reporting may be more tadbhava in form, and fashioned in less ornate metres. 19 See discussion of Vīrsiṃhdevcarit in Renu Bhatnagar 1991: 239–63. 20 See Keshavdas, Vīrsiṃhdevcarit, Keśavgranthāvalī, vol. 3, cantos 28–32. 21 This interpretation is based on the Hindi translation of the verse in Keshavdas, Jahāngīrjascandrikā, Allahabad 1993: v. 114. 22 A second translation of the verse from the Indra perspective: See how the emperor Jahangir is as astonishing as the god Indra. In his court all kinds of wise deities are present: Venus and Kartikeya, the clever moon, learned Jupiter, The sun, Ganesha, Shiva, Sheshanaga, Brahma, Kamadeva, Vishnu, the Vidyadharas and their lovers,

And the apsaras like Manjughosha to captivate a man’s heart And Indra is a capable ruler in every respect— He is kind to the deserving, and harsh towards those who break the law. For uses of śleṣa by Jayasi in his Padmāvat, see de Bruijn in this volume. 23 See Vīrsiṃhdevcarit, cantos 5–6. 24 Jahāngīrjascandrikā:v. 78. 25 The udāharaṇa verses, as in the ritigranths of Keshavdas, tend to be simpler in style. 26 To complicate this profile of linguistic interactions further, the Sanskrit Śṛṅgāramañiarī is itself a translation (chāyā) of a Telugu work—as proclaimed in the text itself. See Śṛṅgāramañiarī of Akbar Shah, V. Raghavan (ed), Hyderabad: 1951, v. 15. 27 The lengthy compound is found in Akbarsāhkṛt Śṛṅgāramañiarī, Bhagirath Mishra (ed), Lucknow: 1956: 5. It is a translation of a similar compound found in the Sanskrit source. 28 See The Language of Indrajit of Orcha: A Study of Early Braj Bhasa Prose, Cambridge 1968: 5. In his work on the thesauruses (kośas) of Nanddas, McGregor adduces further evidence in support of this point. See ‘The Formation of Modern Hindi as Demonstrated in Early “Hindi” Dictionaries’, Gonda Lecture, published by the Royal Netherlands Academy of Arts and Sciences, Amsterdam, 2000: 7–9. Christopher King, reporting on the perceptions of John Gilchrist at Fort William College in Calcutta, also suggests that classicising trends (whether Sanskrit, Arabic, or Persian) could be found among early munshis at Fort William College. See King 1994: 26–27. 29 That Akbar Shah rather than a scholar at court is the actual (as opposed to attributed) author is doubted by the editors of both the Sanskrit and Braj texts. See, respectively, Raghavan 1951: 7, and Mishra 1956: 11–13. 30 The Persianised phrases are ‘dastagīra pīra pātsāha siratāja ke’ (‘protector and spiritual mentor of the crown of kings’, v. 5) and ‘pīra dastagīra e jāhira azamati eka’ (‘protector and spiritual mentor of this single manifest glory’, v. 13). 31 Bhāṣāpiīgal is the most common name of the work. This particular Ms, no. 4805 of the Rajasthan Oriental Research Institute, Alwar, Rajasthan, is labeled ‘Chandalatā’. In the margin of the text the abbreviation ‘pin. la.’ appears, probably short for Piṅgal-latā—yet another name from within the same semantic field. 32 Perso-Arabic vocabulary is given in bold-face. Verse from Ms no. 4805, Alwar, p. 1. A brief survey of the mansucript suggests that Persianisation is not otherwise prominent in the lakṣaṇa or udāharaṇa verses. 33 In a recent article Sumit Guha has drawn attention to the phenomenon of polyglot literary competition at Shahaji’s court. He also notes that commoners were often familiar with at least the bureaucratic registers of Persian during this period. See Guha 2004: 23–31. 34 Compelling examples of plagiarism are exposed in Rajmal Bora, Bhuṣaṇ aur unkā sāhitya, Kanpur 1987: 233. According to Om Prakash, Bhushan borrowed more than one quarter of his

lakṣaṇas from Matiram’s Lalitlalām. See Hindī-alaṅkāra-sāhitya, Delhi 1956:101–102 (quoted in Nagendra (ed) 1973: 343). 35 Shivaji’s coronation had originally been planned for 1673, but ended up being postponed until the following year. For a discussion of the circumstances see Stewart Gordon, The Marathas: 1600– 1818, 1993: 87–89. 36 The title ‘bhūṣaṇ’, which was so thoroughly to eclipse the poet’s given name that the Hindi tradition came to know him only by this sobriquet, was bestowed by patron, Rudra Shah Solanki of Chitrakut. The poet mentions the incident in Śivarājabhūṣaṇ, in Bhūṣaṇ Granthāvalī [reprint of 1953 edition], Vishvanath Prasad Mishra (ed) 1994: v. 28. 37 Śivarājabhūṣaṇ, vv. 60, 144, 172, 186, 194. 38 Shivaji commissioned the Rājavyavahārakośa (Dictionary of administrative terminology) to formulate Sanskritic equivalents to the Persianised vocabulary that had permeated the language of the region. See Guha 2004: 27, 29. 39 See, V.P. Mishra 1994: 40. 40 Ibid: 39. 41 See Śivarājabhūṣaṇ, vv. 58, 74, 113, among others. 42 Śivarājabhūṣaṇ, vv. 110, 167. 43 On Rahim’s role as a major patron of Persian poets see Chhotubhai Ranchhhodji Naik 1966: 280– 462, as well as Annemarie Schimmel 1992: 202–23. 44 An overview of Rahim’s roles as both patron and poet in Persian and Hindi—and the discrepancies between his strong patronage of Persian and weak patronage of Hindi (as well as his relatively scanty Persian oeuvre when compared to its Hindi counterpart)—is found in Corinne Lefèvre-Agrati 2006. 45 The text used here is Rahīm granthāvalī, Vidyanivas Mishra and Govind Rajnish (eds) 1985. The lack of basic text-critical infrastructure in Hindi continues to hamper the field. 46 This work exists in several recensions—testament to the kinds of manuscript problems that plague a would-be scholar of Rahim. Although there are major differences in the order of lines and verses, the overall flavour and thematic content of the poetry are shared across the recensions. And all versions exhibit the same polyglot profile. The version used here is the Nagari Pracharini Sabha recension published in Rahīm granthāvalī. The Hindi Sahitya Sammelan and Asani recensions are published in Abdurrahīm Khānkhānā, Samar Bahadur Singh (ed) 1961. 47 The Kheṭakautukam, a technical work on astrology attributed to Rahim is, however, macaronic in style. It features a Sanskrit grammatical and metrical infrastructure overlaid with significant amounts of Perso-Arabic vocabulary. A recent edition is Khānkhānaviracitam Kheṭa-kautukam, Narayan Das (ed.) 1997. For further discussion of Rahim's Rekhta and macaronic style, see Bangha in this volume.

48 Keshavdas’s Jāhāngīrjascandrika opens with panegyric verses in honour of Rahim’s father Bairam Khan, Rahim himself, and his son Iraj Khan. The prominent placement of these verses at the beginning of his work suggests that the family held considerable importance for him. See vv. 3–8. 49 McGregor 1984: 121. 50 For the Sanskrit verses attributed to Rahim see Mishra and Rajnish (eds) 1985: 169–74. 51 An informative discussion of the barvai form is Rupert Snell 1994: 373–405. 52 Barvai Nāyikābheda: v. 63. Compare ‘kopabhavanavā’ (anger-house) in v. 49. The eastern/diminutive forms of tatsamas have also been remarked by McGregor 1984: 122, and Snell 1994: 382. 53 Barvai Nāyikābheda v. 5. The nayaka has presumably been with another woman. To appease her he will now have to lavish jewellery upon her. A similar instance of word play is the treatment of ‘gulabavā’ (rose) in v. 18. 54 Keshavdas’s seeming ignorance of basic details about Islamic heritage is discussed in Allison Busch 2003: 232–34. 55 For the tradition of Urdu barahmasas see Orsini in this volume. 56 On Hindi as a vehicle for feminine expression in the Persianate literary imaginary see Christina Oesterheld in this volume, and Phukan 2000: 100–39. 57 Rahim, Barvai (Bhaktiparak), in Rahīm granthāvalī: vv. 1–5. 58 See for instance Barvai (Bhaktiparak): vv. 42, 68. 59 Barvai (Bhaktiparak):vv. 86, 94–96. 60 On Persian as a masculine domain (contrasted with feminised vernacular literary registers) see Phukan 2000: 56–64. 61 The vibrantly multilingual educational and literary practices of Bilgram are outlined in Raslīn granthāvalī, Sudhakar Pandey (ed) Varanasi 1987: 49–60. See also Mushirul Hasan 2004. 62 Raslin also uses, though rarely, parts of his given name as a takhallus: either ‘Nabi’ (prophet) or ‘Gulam Nabi’ (slave of the prophet). Raslin has also been credited with a now-lost nāyikābheda work in Rekhta. Pandey, Raslīn granthāvalī, 1987, editor’s introduction, p. 6. 63 Rasprabodh, in Raslīn granthāvalī v. 2, v. 9, v. 10. 64 A glossary to the Madhumālatī prepared by Mataprasad Gupta contains only 7 words of PersoArabic derivation. See Mañjhankṛt Madhumālatī, 1961: 489–504. The much lengthier Padmāvat, for its part, is said to contain only about 130 words of Perso-Arabic origin. Ramesh Mathur 1974: xxi. 65 Perhaps the Urdu poet Insha’s Rānī Ketakī kī kahānī, a later undertaking in de-Persianised style— this time in Khari Boli—needs to be seen not as a colonial-period innovation but as part of a longer

tradition of Indo-Muslims experimenting with a tadbhava register. 66 Mutafarriq v. 11 in Raslīn granthāvalī (Perso-Arabic vocabulary highlighted in bold type). 67 A recent discussion of this intriguing work, probably written at the court of Azam Shah (son of Aurangzeb), is R.S. McGregor 2002: 924–44.

C

4

1 See John Stratton Hawley 1983, 1984; Kenneth E. Bryant 1978. 2 Evidence of the multiple roles of poets in the field is there in the differences within the corpus of the poetry of Tulsidas, Kabir, Surdas and other medieval authors. Sectarian hagiographies often attribute the differences in theme and religious attitude to a moment of conversion. Such has been the case with Surdas whose vinaya poems are of a completely different theological outlook than his Krishna poetry. Tulsidas wrote also in Braj and his Kavitāvalī expresses a different point of view than his Rāmcaritmānas in Avadhi (see Bangha 2004). Also in the works of Jayasi, we find a substantial difference in the outlook of his premākhyāns, such as the Padmāvat, and more didactical works such as Akhirī Kalām and Akhṛāvat. See de Bruijn 1996, 1999, 2000. 3 It is striking, for example, how the idiom of the Nath yogi treatises on mystical asceticism pervaded nearly all traditions in the field of medieval religious poetry, from Sant poetry to Tulsidas and the Sufi poets. 4 See all the examples in Imre Bangha’s essay in this volume. 5 Karine Schomer and A. McLeod (eds) 1987: 74. 6 Aditya Behl 2002: 89–100. 7 See Julie Scott Meisami 1987. For popular traditions in Persian and Urdu see William Hanaway 1972; and Frances Pritchett 1985. 8 On the poetics of Padmāvat see de Bruijn 1996. 9 See Streusand 1989: 148ff; and Alam 2004. 10 See Padmāvat ed. by V.S. Agrawal: Mālik Muhammad Jāyasī kṛt Padmāvat: Mūla aur sañjīvam vyākhya. Ciragamva: Sahitya Sadana: 52.8; 102.4; 198.5 304.1, 318.1, 323.6; 324.1; 333.5; 384.5; 402.7; 459.7; 475.2. All references to the Padmavāt in this essay are to this source. 11 The parallel is reiterated later on in the Rama-story, when Hanuman is given entrance to the city by the demon Lankini (RCM V, 4–8). 12 See Vaudeville 1955: xi, and Lutgendorf 1991.

13 See Imre Bangha 2004. 14 For the concept of habitus see Pierre Bourdieu 1980: 88–89. 15 Paul Hacker 1983: 11–28. 16 Charlotte Vaudeville 1955: 327. 17 Growse 1987: 34–35

C

5

1 For a full study of early barahmasas, see Vaudeville 1986. 2 Selby 2000: 6. 3 Ibid: 9. 4 Vaudeville examines the possible origin of the form in chaumasa, or rain song of viraha (1986: 27– 32); also see Zvabitel (1961) for Bengali baromasi. 5 In this Jayasi followed other Avadhi Sufi poets before him who had already used barahmasas in their romances, starting from Maina’s lament in Mulla Daud’s Candāyan (1379) and Rupmini’s in Qutban’s Mirgāvatī, see V.P. Dvivedi, Barahmasa. The Song of the Seasons in Literature and Art, 1980: 41–42. 6 Text and tr. (slightly altered) in Vaudeville 1986: 65, 75. 7 See Vaudeville 1986: 56–63. 8 As Vaudeville remarks, ’…the fact that a barahmasa is included in a narrative work does not imply that the barahmasa itself is narrative in character’. Vaudeville 1986: 9. 9 Susan Wadley 1983: 62, 68. 10 It is worth noting that we find illustrated manuscripts of Daud’s poem. The earliest known version —which could well be the first pictorial depiction of a barahmasa extant—is a manuscript dated 1550; see Dvivedi 1980: 83. 11 Mirabai is a notable exception; see below. 12 Gloria Raheja and Ann Grodzin Gold (1994) argue that among themselves women do not feel that they have to adopt the postures of submission and modesty they have to in the presence of men, and their song genres contain much sexual banter and innuendo.

13 The barahmasa attributed to Guru Arjan in the Guru Granth is purely didactic, whereas the one attributed to Guru Nanak is of a devotional nature. Even Nagmati’s lament quoted above has tinges of a spiritual experience (e.g. the ‘unfathomable path’). 14 Found in a late twelfth-century manuscript at Patan; see Vaudeville 1986: 18. 15 According to Vaudeville, the viraha barahmasa of Rajimati pining for Nemi for twelve months before herself embracing the ascetic life ‘shows that Jain munis were quick to use viraha-gītas of the chaumasa or barahmasa type to disseminate their ascetic teaching, and the change from virahā to vairāgya, i.e. from a lament for solitude to an attitude of contempt for the world and its fleeting joys, occurs as a natural development’. Vaudeville 1986: 27. 16 See Vaudeville 1986: 53, 56. In several of the printed, manuscript and oral barahmasas of the nineteenth century attributed to Surdas or Tulsidas, the template was used to recollect the adventures of Krishna or Rama or the marriage between Krishna and Rukmini. The names Krishna, Murari, Shyam and also Raghubir also crop up regularly in secular barahmasas, evidence of the blurred boundary between secular and religious love that poetry in the public domain also display, and of the many and layered resonances that had gathered around the Krishna-gopi motif. See e.g. Bārahmāsā Benīmādho (below), Rāmcandrajī kī bārahmāsī by Ganeshdas (Fatehgarh 1868); on the marriage between Krishna and Rukmini, Bārahmāsā Shambhū Rāy (Delhi 1875). 17 Another painting of the Bikaner school depicting the summer months shows the hero and heroine seated before a pavillion with a maid in attendance, while two wayfarers rest under the shade of trees in the foreground, and a tiger and cobra are seen together in the lower right corner; see V.P. Dvivedi 1980: 105. Keshavdas’s verses run as follow: [The sun is so bright and scorching that] The five elements—air, water, sky, earth and fire have become one—like fire. Deserted are the roads, the tanks parched dry [seeing which the elephants do not go out]. Cobras and lions sleep inside. My love, even strong creatures become weak, the whole world is ill at ease. Keshavdas says, the elders say one should not go out in this month. 18 In most cases, sets of twelve months were painted along with Devanagari captions, either on a panel at the top or on the back of the painting, thus establishing beyond doubt that the paintings were based on the literary descriptions; V.P. Dvivedi 1980: 81. 19 Other poets whose barahmasas were illustrated were Govind, Datta, Anandram, Kashiram and Ramkishan. Govind’s barahmasa is one of viraha and the text is more similar to that of the folk songs, though the language is literary Braj Bhasha. The nāyikā is depicted without her lover; in the painting she is (still fully dressed) surrounded by a maid and two female musicians in a garden pavillion with many flowers, while the moon shines in the sky. See V.P. Dvivedi 1980: 141–52. 20 Chaturvedi’s rough list of Devanagari manuscripts in the hands of individuals in UP and copied in the eighteenth or nineteenth centuries includes thirty barahmasas. Five of them are ones that were also printed in Devanagari and Urdu, from where they may have made their way into personal manuscript collections. Gopal Narayan Bahura’s list of manuscripts held in Amber and Jaipur lists another two pages of Barahmasas by local court poets. Jawaharlal Chaturvedi 1965: 163–70; Bahura 1976: 175–76. 21 A copy of the manuscript text was generously given to me by Dr Kishorilal.

22 For brief biographical information, when available, see Tanvir Ahmad Alvi 1988. 23 E.g. Hafizullah Qadiri’s Bārahmāsa neh for example explicitly quotes a proverb about the first cold spell of Agahan that rekindles desire for hot food: ‘Agahan kā teṛi honā kyā kaṭhin hai/ kahāvat hai agahan hanḍiyā adhan hai’. (Times passes with difficulty in Agahan; as the saying goes ‘the frying pan is sizzing’); Alvi 1988: 246, v. 119. Maulvi Hafizullah Qadiri was the son of a Qadiriyya Shaikh from Badaun, and after a traditional education in Rampur he was himself initiated in the silsilā and became the Khalifa. He settled in a village in Bilaspur district and lived there all his life, writing spiritual works in Persian and dispensing guidance and advice. According to the editor of his barahmasa, he was also a poet in Persian, Urdu and Avadhi, but his barahmasa circulated orally until recent times; see Alvi 1988: 80. 24 See Alvi 1988: 22. 25 Later Urdu barahmasa writers also often began with an introduction on love, and some—e.g. Mufti Ilahi Bakhsh, Maqsud, Rabbani—prefaced it with formal praise (ḥamd) of God and the Prophet. 26 Text in Hashmi and Khan (eds) 1979: vv 4–54. 27 Alvi has suggested that the fleeting meeting, which only serves to heighten the intensity of the lover’s suffering, may indicate the influence of the Sufi theory of the stages (marhala) that the disciple must undergo in his spiritual training (personal communication). 28 Ibid: 40 vv. Apart from Afzal’s poem, which starts with Savan, the second month of the rainy season, and Vahshat’s, which starts in spring with the month of Chait, all follow the folk-song template by starting with the first month of the rainy season, Asarh. 29 Perhaps also under the influence of other narrative poems (masnavis), some poets like Maqsud (1790s?), the author of Sundarkalī, Najib (1830s) and Ranj (1872) framed the barahmasa through a narrator, who either hears the story from the woman in question or who intervenes in the narration, especially towards the end when the meeting between her and her returning lover-husband takes place. 30 ‘My love is abroad and I’m suffering constantly’, ‘[The month of Agahan] has left after giving me much grief, now [Pus] has come’, ‘The pain for my beloved makes me feel like taking poison’. 31 Ibid: 29, 30, vv. 76, 82. 32 Hashmi and Khan (eds) 1973: 37, vv 181–82, 186. 33 Ibid: 40, vv 227–29. 34 I have not come across any edition of Afzal’s poem printed in the nineteenth century, but Alvi (1988) states that a large number of manuscript copies are extant in north India and the Deccan. 35 He was born and educated in the 1780s according to Alvi 1988: 94; the verse is quoted on p. 100. 36 Alvi 1988; A.J. Zaidi 1993.

37 Interestingly, the only other Urdu courtly poet to compose a barahmasa, Mir Bahadur Ali ‘Vahshat’ from Lucknow, quickly lost his way and ended up writing a Saqinama instead A soldier/courtier and a contemporary of Mir Hasan, he is mentioned in several tazkiras as the author of Urdu verse in the manner of his day (i.e. iham) though he did not write much, but only the tazkiras nearest to him in time ever mention his barahmasa, which is never quoted. A long and rambling poem (515 vv.), it is the furthest from the barahmasa template: it starts unusually in Chait, soon forgets about describing seasonal and ritual events and instead inserts verses to the cupbearer to link each month to the next, and what more the lover-protagonist is a man. In language, too, it is pure Perso-Urdu Khari Boli, and introduces metaphors that are unsual in barahmasa poetry. See Alvi 1988: 186–213. 38 Alvi 1988: 162, vv 15.17 39 Ibid: 165, vv 18–19. 40 Agraval 1988: 353. 41 Alvi 1988: 165, vv. 65, 69. 42 Like other barahmasas by Muslim poets, it begins with the praise of God (27 vv.) followed by an introduction on love (17 vv.), in which the poet states in clear terms that this is a poem about worldly love. The months vary in length (between 20 and 40 verses); the longest month is the first one, Asarh, which does not begin with the rains but with the heroine looking for her beloved in every house; see Alvi 1988: 255–77. 43 Persian participles (like nālān, sūzān) and Arabic words like ibtidā also do not normally feature in barahmasas. 44 Maqsud’s is one of the few barahmasas to claim to be a ‘story’ told by a narrator about the birahini: after the praise of God (vv 1–27), it introduces the birahini ('Sunī hai hamne birahan kī zabānī/kahā kartī thī ro ro yah kahānī/29) and the customary reflection on the power of love (28–45), which introduces the combination of love lament and seasonal calendar. 45 He mentions Gobardhan puja in Asarh, Saluno during Savan, Kanagat in Kvar, Divali and bathing in the Ganges in Kartik, and of course Holi during Phagun. 46 Witness the informed opinion of Garcin De Tassy, who was aware of Afzal’s poem and even translated Hazrat Jauhari’s Barahmasa kanval daī into French under the title of ‘Un monologue dramatique’, in Journal Asiatique, October 1850: 310–28. 47 Therefore elements other than the description of the season and the mood of the month could thus take centre-stage, as with the character of the parrot in Barahmāsā Sundarkalī, whose sub-plot takes up two whole months. 48 For a fuller account, see the chapter on barahmasas in my Print and Pleasure: Popular Literature and Entertaining Fictions in Colonial North India, forthcoming. 49 Bārahmāsā Khairāśāh, Agra 1862, IOL 14158.c.61(1). The same is true of the earliest editions of all the most popular titles, like the Barahmasas by Allah Bakhsh (Agra: Matba Ijad Kishan, 1867; IOL: 279.49.a.4) and Harnam Delhi Matba Ahmadi, 1869 IOL: VT1202.

50 E.g. Majmu’a bārahmāsa 1876 IOL: 14119.d.5; Majmu’a dilpasand 1876 IOL: 14119.d.6. 51 The first is Dropadījī kī bārahmāsī, printed [in Delhi] at Matba Dilkusha, 16mo, 16pp. (IOL 14158.c.6); another edition was printed in Delhi at the Matba Jvalaprakash in the 1870s (14158.c.3); the second is Bārahmāsā Sūrdās, printed in Urdu characters in Agra by the Matba Rada (14109.a.10). 52 Grierson 1884: 204, 208. 53 See Blumhardt 1900 and 1902. 54 See Wadley 1983. 55 If we compare the poems by Allah Bakhsh and Harnam we notice that for Savan, Harnam mentions celebrating Saluno, while Allah Bakhsh mentions Tij and the singing of Malhar. Both mention the joyous swinging on swings by the other, happy and bedecked women. 56 Harnam here plays a pun on the heroine’s state (pau-bārah = ace and 12 in dice; śiś-panj = lit. ‘six and five’, to be in a state of confusion or anxiety. 57 Bāramāsa Harnām in Majmu’-e dilpasand, Delhi 1876: 7

C

6

1 R.S. McGregor 2002: 912. 2 Clearly many of the stalwarts of the Hindi tradition wrote poetry that was sung, and could therefore technically be classified as a ‘song text’. I am not suggesting that Surdas, Mirabai and Swami Haridas did not write ‘song’, or indeed that they are overlooked in literary analyses. My point is that songs that exist solely in an orally transmitted culture and have no life as poetry outside of a musical context have rarely, if ever, been included. 3 Kanz al-Mūsiqī, Ethe 2021. Unnamed mss, Ethe 2019 and 2020. See Ethe 1980. 4 Bhākhā was an unspecified term used for all types of Hindi in the seventeenth to nineteenth centuries, though usually only when written in the Devanagari script. 5 Hindavi is a Persian term essentially meaning ‘Indian’. The term has been used since at least the thirteenth century to refer to the mixed speech register of Delhi and surrounding areas at first, though it came to include most of north India. Although earlier usages of the term could refer to either the Persian or Devanagari scripts, by 1800 the manuscript cataloguer is likely to have only used Hindavi for the former. 6 This description refers to Braj Bhasha,‘the language of Braj’, though whether that is based on substantial linguistic evidence or the mere impression that most song texts must be in the poetic register associated with Braj Bhasha is unclear.

7 Ethe 2019 in the Persian catalogue of the British Library. 8 The problem of author and authority has been well-documented, most notably by Hawley 1984. 9 The word cīz, literally ‘thing’, is frequently used to describe a bandish. 10 Wade 1997: 23. 11 Deshpande 1973: 103. 12 Text and translation taken from R. Snell, The Hindi Classical Tradition, 1991: 118–19. 13 Although the inherent ‘a’ of Hindi is traditionally transcribed in poetry to reflect the fact that it tends to be pronounced in performance, I have omitted it here as it interferes with the onomatopaeic jhanjhan sound of the anklets. 14 As quoted in Padmavati, Mīrā: vyaktitva aur kṛtitva, Varanasi 1973: 83. There are many other but similar versions of this poem. 15 It is unusual for classical poetry to be used in the khyal repertoire. Usually a Mirabai text would be sung as a bhajan. However, the inclusion of this song shows the adaptability of khyal to the existing Hindi ‘canon’. 16 Samadhana: a daughter- or son-in-law’s mother. 17 This region in eastern Madhya Pradesh has recently become a state in its own right, but has existed as a linguistic and cultural entity for centuries. 18 Interview with Bala Saheb Poochwale, Gwalior, January 2003. I am fortunate to have met him as he sadly passed away in 2004. 19 San in sandala is read as short in this analysis, i.e. sā. 20 A first-person verb ‘I will make merry with Adarang’ (karūṅgi) would make more sense here and there are indeed instances of that. Whether the more ‘correct-sounding’ version is indeed the older or more authentic one is as always a moot point as it is virtually impossible to substantiate, and in any case, this is the version popularised by Amir Khan. 21 The sexual connotation of the word kalī (‘bud’) is shown by the phrase band kalī (‘closed bud’) denoting a virgin. 22 Those concerned with reading religious meaning in song lyrics are often pleased at the use of this word, which means ‘temple’ in modern Hindi. It is clear, however, that its original sense of ‘house’ is kept in khyal and thumri, even if the texts were created in recent times. This usage is more evidence of khyal poets choosing their language consideredly. 23 These two possible developments are not mutually exclusive of course and indeed, a combination of the two is the most likely scenario. 24 Quoted in Triloki Narayan Dikshit 1963: 271.

C

7

1 I use ‘ghazal’ in the narrow sense here, as a genre, not as a poetical form. In this sense the ghazal is used to convey specific sentiments and a specific atmosphere, mostly romantic or mystic longing, metaphysical speculation, a general discontent with society and the times, brooding over one‘s suffering, or romantic idealism. Ghazals in this sense are also distinguished by their specific sound quality, by a harmony of expressions, a preference for smooth sounds and avoidance of consonant clusters. In the broader sense, that is as a poem of the rhyme form aa-ba-ca-da etc., a ghazal may be used for any subject matter. 2 A prominent example is Amrit Rai’s A House Divided: The Origins and Development of HindiHindavi, 1991. 3 See her ‘Dakani Radha-Krishna Imagery and Canon-formation in Urdu.’ Petievich 2000. 4 For a detailed discussion of the female voice in Bikaṭ kahānī see Shantanu Phukan 2001, and Orsini in this volume. 5 C.M. Naim 2001; C. Petievich 1990, and 2002. 6 Phukan 2001: 44–48. 7 For the purpose of this paper, only marsiyas written by male poets will be dealt with. There is a considerable amount of marsiyas written by women which are discussed by Amy Bard in her PhD dissertation ‘Desolate Victory: Shi’i Women and the Marsiya Texts of Lucknow’, 2002. 8 For a detailed account of the history of the genre in Urdu, see C. M. Naim, ‘The Art of the Urdu Marsiya’, in Milton Israel and N. K. Wagle (eds), Islamic Society and Culture: Essays in Honour of Professor Aziz Ahmad, 1983. 9 Ibid. 10 Sadiqur Rahman Kidwai,‘Anīs kī mars̤ iya go’ī urdū naẕm kī tārīḵh mẽ’, July 2004: 17. All translations from the Urdu are mine. 11 Naim 2001:108. 12 Imre Bhangha in this volume. 13 I have used the edition by Malik Ram and Mukhtaruddin Ahmad. See Fazli 1965. 14 No attempt has been made at a poetic rendering in English. The translations serve only to provide an idea of the content of the verses. 15 See Karbal kathā, Fazli 1965: 14–15. 16 Insha’allah Khan Insha’s (1753–1817) Daryā-i lat̤ āfat was written in Persian and published in the beginning of the nineteenth century. It contains numerous samples of the speech of different

localities, ethnic and vocational groups as well as of women from specific quarters of Delhi. 17 Sa’adat Yar Khan Rangin (1756–1834) is considered to be the inventor of Rekhti. 18 Daryā-i lat̤ āfat (Urdu translation by Pandit Brajmohan Dattatreya Kaifi, ed. by Maulavi Abdul Haqq) 1988: 136. 19 I have used Kulliyāt-e Saudā, vol. 2, ed. by Muhammad Hasan 1984. In this edition, marsiyas fill 206 pages. 20 These lines, to my mind, are one more proof of Allison Busch’s statement that language choices, more often than not, were made for aesthetic reasons. See her paper in the present volume. 21 Kidwai 2004: 21. 22 Phukan 2001: 47, 48. 23 C. M. Naim 1983: 9. 24 C.M. Naim 1983:9. 25 See David Lelyveld 1994: 61.

C

8

1 Malik Ram, Qadīm Dillī College, 1975: 8. Malik Ram’s opinion is important because he is an editor of the first scholarly edition of Fazl Ali Fazli’s Karbal kathā (c. 1732), perhaps the earliest Urdu prose work in north India. See Fazl Ali Fazli, Karbal kathā, Malik Ram and Mukhtaruddin Ahmad (eds) 1965. 2 Ibid: 8. 3 Ram Babu Saksena 1927: 8; 239–41. 4 Karbal kathā, is a religious narrative, an abridged or creative translation of Vaez Kashfi’s (d.1504) Persian work, Rauẓat-ush Shuhadā; the other two are romances or dastans. Isvi Khan’s qissa is the earliest extant work of the dastan genre in north India. Tehseen’s story is based on the famous Persian Qiṣṣa-e chahār darvesh (Story of the Four Darveshes). The same story is retold in simpler language by Mir Amman’s Fort William classic, Bāgh-o bahār. For more details regarding these texts see Jamil Jalibi 1975, Vol. 2, Part II, Chs. 3–5. 5 Malik Ram 1975: 9. 6 Karbal kathā: 37–38. Punctuation in the English translation has been added. All English translations appearing in this essay are mine.

7 Karbal kathā: 144. 8 Isvi Khan Bahadur, Qiṣṣa-e Mehr Afroz-o Dilbar, Masud Husain Khan (ed) 1966. The manuscript did not mention either the author’s name or title. A different hand (other than the scribe) had penned in the title of the work and the writer’s name. Masud Husain Khan could not identify Isvi Khan Bahadur. Much later, Parkash Munis identified him as the author of the commentary Rascandrikā. See his Urdū adab par Hindī kā as̤ ar, 1978. 9 Ibid: 35. 10 Husain Ata Khan Tehseen, Nau t̤ arz-e muraṣṣ’a. Nurul Hasan Hashmi (ed) 1958: 86. 11 E.g. Mehr Chand Khattri’s Nau Ā’in-e Hindī (or Qiṣṣ;a-e Maḥmud-o Getī Afroz, 1788). Like Tehseen, Khattri presents a new style, or nau ā’in-e Hindī, albeit a different, and simpler one. Then there are Emperor Shah Alam II’s (1759–1806) voluminous dastan, the Ajāib ul-Qiṣṣas or ‘Wonderous Tales’ (1792), which he describes as a romance in Hindi, Rustam Ali Bijnori’s Qiṣṣa-e Aḥvāl-e Ruhela (1776), and Syed Husain Shah Haqiqat’s Jaẕb-e ‘Ishq, (The Allure of Love, 1796). See Jalibi 1975, ch. 5 for a more detailed description of these early fictional prose works. 12 See Tarachand’s presentation in the All India Radio broadcast series entitled, Hindustānī kyā hai (1939). The talks presented in the series were published soon after by Makatbah Jami’ah, New Delhi. ‘They found a way out by adopting the language of Mir Amman, Afsos and others, but they excised Arabic/Persian words from it, replacing them with those of Sanskrit …’. p. 11. 13 For the impact of the wholesale introduction of Sanskrit tatsama vocabulary into the Akashvani news broadcasts beginning in the 1950s, see David Lelyveld 1994. 14 For example, the two Arabic words, vāḥid and aḥad which have generally the same meaning, namely ‘one’. However, in terms of strict categorisation, vāḥid means ‘one’ but could be, in theory ‘the sum total of many parts which add up to one’. But aḥad means ‘one name indivisible, not admitting of any parts, and not consisting of fragments which total up to one’. Now it is clear that these two cognate terms are extremely useful in many kinds of philosophical and theological disputation and Urdu would be at a great disadvantage if it were asked to shed these two words in favour of Sanskrit lexemes. 15 Note the use of Khuda in the title instead of Allah. The title yields a tārīkh or chronogram 1185 hijri, approximately 1771–72. In many printed versions it is erroneously inscribed as Khudā kī Ne‘mat. The tafsīr has also been published as Tafsīr-e Muraddiya, in editions from Calcutta, Bombay, Kanpur and Lahore. For this paper I have used a copy of the Bombay edition of 1284 Hijri (1867 CE). Hafiz Mahmud Sherani’s collection at the Punjab University, Lahore, has a manuscript of the tafsīr that includes a dībācha or Introduction which is not included in the printed version(s). 16 For details see Syed Muhammad Amin (2001). 17 This is supported by the fact that by the middle of the eighteenth century Hindus were entering the realm of Urdu poetry in ever larger numbers than before. Even a leading Persianist and lexicographer, Tek Chand Bahar, occasionally composed Urdu verses. Similarly, from the middle of the eighteenth century professional and so-called lower classes in Delhi started producing Urdu poetry in ever larger numbers, as is attested by Qudratullah Qasim’s tazkira, or compendium, Majmū‘a’e Naghz.

18 Benjamin Schulze completed his grammar of Hindustani in 1741. See Benjamin Schulze, Urdu Grammar, translated by Abul Lais Siddiqi 1977. According to Jamil Jalibi, Hadley, a British army officer posted in Bengal prepared a grammar of Hindustani for the British soldiers in 1765 that was published from London in 1770 and went through several editions. Ferguson’s Hindustani Dictionary was published in 1773. From 1787 onwards, Gilchrist began his work on problems of script and transliteration. Insha’ Allah Khan Insha’ and Mirza Muhammad Ahsan Qatil’s Daryā-i Lat̤ āfat, was completed in 1801. This was an interesting work, albeit in Persian, on ideas of expanding the Urdu alphabet, different kinds of speech, and faṣāhat or ‘good’ prose. 19 Shah Muhammad Haqqani, ‘Ināyat Rasūl kī. Ms, Khanqah-e-Barkatiya, Marehara. 20 Tafsīr: ‘exegesis’. The closest English term sounds very formal and also doesn’t cover the various aspects of tafsīr; nor does the gloss ‘commentary’, which is commonly used, convey the exact meaning. Tafsīr always includes translation along with the commentary. It is never commentary alone. It is used for the Quran and Hadith. Other technical terms associated with Quranic translation are: ta‘bīr: ‘interpretation’, sharah: ‘explication’, tarjuma: ‘translation’. As far I could ascertain there are only 52 translations of the Quran in Persian. 21 According to Allama Sarkhesi the people of Persia requested Hazrat Salman to translate the surah Fātiḥa and the Hazrat complied. The Iranians recited the Persian translation in the namāz until they became comfortable with reciting the namāz in Arabic. Cited from the Introduction to the Quran al Hakim, Persian translation of the Quran by Hazrat Makhdoom Nuh Sarvahalai Sindi, edited by Abu Said Mustafa Qasimi, and published by the Sindi Adabi Board, Hyderabad, Sind, 1401 hijri (1981 CE). 22 This translation was carried out during the reign of Nuh Samani (265–350 hijri; 879–961 CE) and is currently in print, published from Tehran in 1970. 23 See the Introduction to Sarvarhalai 1981. 24 Tarjuma-e Makhdūm Sarvar bisyār khūbīhā dārad. Dar aks̤ ar jā tarjuma-e bismillah rā hasb-e ta‘lluq ḥarf-e jār mukhtalif bayān mī kunad va īn jiddat ast ke dar hech dīgar tarjuma-e fārsī na dīdah shudah. Quoted in the Introduction to Sarvarhalai. The pages of the Introduction are not numbered. 25 Al is the definite article in Arabic; followed by ilah, meaning ‘god’, the word Allah gives the meaning of ‘the One and the Only God’. 26 Shah Fazle Rahman Ganj Moradabadi, Manmohan kī bātẽ; introduced by Nazar Ali Khan, published by Khudabakhsh Oriental Library, Patna, 1990. Shah Fazle Rahman translates Bismillāh ar raḥmān ar raḥīm as ‘Pehle hī pehl nām letā hūñ manmohan kā jo baṛī miyah moh kā, mehrvālā hai’. It was common practice in certain Sufi works, especially those in Avadhi, to Indianise Islamic vocabulary. For example, the Creator is referred to as sirjanhār in Maulana Daud’s Candāyan and as kartār in Jayasi’s Padmāvat. According to Jamil Jalibi, colloquial words such as bisrām, bhasm, manmohan, sangat, acraj, jīv, jag, nit, etc. were not unique to Karbal kathā but also used by poets of the times like Abru and Naji, and even by the likes of Mir and Sauda. Perhaps the choice of the word manmohan, usually referring to Krishna in Braj Bhasha, stems from the popularity of its usage in eighteenth and nineteenth century poetry. See Jalibi 1975: 1039.

27 Tafsīr-e Muraddiya (dībācha): Manuscript in the Sherani Collection, Punjab University, Lahore. Cited from Jamil Jalibi 1975: 1044. Translation mine. 28 Ibid: 1045 29 Khudā kī Ne‘mat. Muradullah 1867: 306. These are the opening sentences of Shah Muradullah’s introduction to Surah Al-Qadr (surah 97). 30 Khudā kī Ne‘mat 376. These are the opening sentences of the Tafsīr of surah Ikhlāṣ, (surah no.112) of the thirtieth para. 31 See Shah Rafiuddin, Tafsīr-e Rafi’ī, Matba’ Naqshbandi, 1272 hijri. Cited from Jamil Jalibi 1975: 1050. Surah Al-Baqar is the second surah of the Quran. 32 Dībācha Mūẓih al-Qurān, ed. Shaikh Muhammad Panipati, Nuqūsh, no. 102, Lahore, May 1965. Cited in Jamil Jalibi 1975: 1053. 33 Shah Rafiuddin, Qurān Sharīf Mutarjim ba do Tarjumah, the first translation (being a true copy of the original first published translation of the Quran) by Shah Rafiuddin, the second by Maulana Ashraf Ali Thanvi, Kutub Khana-e Rashidiya, nd: 841. 34 Shah Abdul Qadir, Al Qurān al Hakīm ma’ Tarjumah-o Tafsīr Mūẓih al- Qurān, Shah Abdul Qadir, New Crescent Publishing Company, nd: 731. 35 This tafsīr should not be confused with the Tafsīr-e Ḥaqqānī also known as the Fath al-Mannan of Abu Muhammad Abdul Haq Haqqani. The latter is in eight volumes and was first published from Delhi in 1880. 36 Shah Muhammad Haqqani, ‘Ināyat Rasūl kī, manuscript at the Khanqah-e Barkatiya, Marhara, p. 53. Cited in Syed Muhammad Amin, 2001: 218. 37 The spoken variety of the language was almost always referred to as Hindi. For a detailed discussion of this issue see Shamsur Rahman Faruqi 2001: ch 1. 38 I have chosen ten verses of the surah Ash Shams from the Thirtieth para. The surah Ash Shams is one of the early Meccan revelations. It begins with a fine nature passage and proceeds to how we need to be aware of our soul and inspire it to do the right and avoid corrupting it through wrong doing. 39 Shah Maradullah gives the following rendering of Bismillāh ar raḥmān ar raḥīm: Shurū ’ kartā hūn Allāh ke nām se jo bahut mehrbān, nihāyat raḥm vālā hai. 40 At this point Shah Muradullah shifts from a translation mode into a more expansive explanatory mode. I am providing part of the deviation from the excerpt above: Allāh ta’ ālā un chizõ kī qasam kartā hai jo har ek chīz parvardigār kī qudrat kī baṛī nishānī hai. sūraj kī bahut baṛī ‘alāmat hai. Allah ta’ālā kī qudrat kī bahut bahut khalq sūraj se mutā’liq hain. tamām jahān āftāb ke nikalne se roshan ujālā hotā hai, kyõkī anāj dhūp hī mẽ pakte hain. jis chīz kā sukhānā darkār hotā hai dhūp hī mẽ sukhāte hain. jāṛā sūraj hī kī garmī se dafa’ hotā hai, aur jo kuch fā’de hain […] aur chānd bhī dūsrī nishānī hai pāk parvardigār kī qudrat kī. Bahut, bahutfā’de

chānd mẽ rakhe hain. ek baṛā fā’da yeh hai jo chānd kī ghaṭ-baṛh se shumār barsõ mahĩnõ kā, dinõ, kā jānā jātā hai. See Muradullah 1867: 246–47. 41 Shah Rafiuddin gives the following rendering of Bismillāh ar raḥmān ar raḥīm: Shurū ‘Allāh ke nām se jo baṛā mehrbān, nihāyat raḥm vālā hai. 42 Shah Abdul Qadir gives the following rendering of Bismillāh ar raḥmān ar rahĩm: Shurū‘ kartā hūn main sāth nām Allāh bakhshish karne vāle mehrbān ke. 43 Holy Quran; English translation by Marmaduke Pickthall and Urdu by Maulana Fateh Muhammad Jallandhari, Kutubkhana Ishayatul Islam, nd: 753. 44 Platts gives the following meanings: jī, Prakrit jīvo, Sanskrit jīva; life, soul, self, spirit, mind, heart, courage, disposition; any living thing. 45 Platts gives the following meanings: essence, substance, nature, radical constituent, soul, body, person, self. 46 This is not surprising. Our notion of ‘language’, the name(s) by which it is known and the formal knowledge of grammar only developed in the subjects we now comprehend after 1825. English was not offered as a subject in English schools until the mid-nineteenth century.

C

9

1 This trend was not as much in keeping with Wordsworth’s desire ‘to keep my Reader in the company of flesh and blood’, but rather more in order to ‘[abstain] from the use of many expressions… which have been foolishly repeated by bad poets’. William Wordsworth 1800: xxi, xxii. 2 There were in fact many poets in Braj in the late nineteenth and early twentieth centuries. Indeed, some remain today. In the period under discussion, roughly 1870–1935, the definition of ‘Braj Bhasha’ itself was in flux, and cemented in the latter years around particular morphological and lexical features that were deemed non-standard by arbiters of Hindi prose (Mahavirprasad Dvivedi most famous among them). Ratnakar and Hariaudh are merely two of the more well-known Braj Bhasha poets among at least a score or two of poets, including Bharatendu Harishchandra, Shridhar Pathak, the young Jayshankar Prasad, Satyanarayan ‘Kaviratna’, and Viyogi Hari. See Jagdish Vajpeyi 1964 and Orsini 2002: 40–41, n. 51. Generally speaking, Braj Bhasha has remained virtually absent from the secular pan-regional literary stage since the 1940s. 3 Francesca Orsini (2002) neatly outlines the social spheres of early Hindi supporters and readers in her Introduction to The Hindi Public Sphere, 1920–1940: Language and Literature in the Age of Nationalism. 4 Details about this organisation can be found in Christopher R. King 1994. See also, for post-1920 history, the references in Orsini 2002.

5 Khatri, 1889. Frederic Pincott (1836–96) was a London publisher and promoter of Hindi. He established an Urdu business journal in London, Ain-esaudagari, and promoted the inclusion of Hindi for the civil service exam in England, and as a government language in India from the 1860s. He was a friend of several Hindi litterateurs, and wrote in Sanskritised Hindi himself, including a children‘s textbook (1893), and a paean to Queen Victoria (1895). See Dhirendranath Simha 1986: 237–44. He advised Braj and Khari Boli poet Sridhar Pathak on his translation of Goldsmith and was a regular contributor to The Indian Magazine, published in London under the auspices of the National Indian Association in Aid of Social Progress and Education in India and supported by the Dufferins and many Indian royals, major and minor. 6 Pincott, Introduction to Khatri 1889: iv. Page numbers from this text. 7 See Charu Gupta 2001. 8 I intend to focus my work on poetry within and on the margins of the grand narrative of Hindi literary history, bracketing popular tracts, modern bhajans, and folk literature. See Statement of Particulars regarding Books and Periodicals Published in the United Provinces for a more complete picture of Hindi poetic production. 9 Here I refer to the central concept of Orsini’s The Hindi Public sphere, 2002. 10 The following biographical information was culled from two biographies, Mukunddev Sharma, 1956, and Girijadatt Shukl ‘Girish’, 1959. These two authors were Hariaudh’s grandson and protege, respectively. 11 Lallan Ray 1958: 3; and Dan Bahadur Pathak ‘Var’ 1965: 2. 12 Vasudha Dalmia (1992: 281–93) most succinctly describes the milieu I evoke here. C.A. Bayly’s Rulers, Townsmen, and Bazaars (1988) is also of note in elucidating nineteenth century merchant culture in Varanasi. 13 Grierson would facilitate the adoption of one of Hariaudh’s ‘ṭheṭh’ (authentic) Hindi novels for the Indian Civil Service Exam, and commission another ‘ṭheṭh’ novel, both from the KVP. Sir George A. Grierson (1851–1941) was an Irish civil servant and scholar who conducted the Linguistic Survey of India in 1898–1928. He had held various appointments as Magistrate, Collector, and Opium Agent, in various locales in present-day Bihar and Bengal, notably Patna and Gaya. Grierson was deeply involved in the Hindi publishing world and literary circles, and promoted the continued development of vernacular literature of north India. He also undertook literary historical projects concerning Hindi/Hindustani literature, e.g. The Modern Vernacular Literature of Hindustan 1889. See B. S. Kesavan and Asha Gupta 1997: Appendix I, 321–33. 14 Dhirendranath Simha, 1986. 15 Of a family descending from Guru Amar Das, hence the title ‘Sahabzade’. Also known as Sumersingh Bhalla. 16 The origin story had to do with a wandering Udasi sadhu granting the boon of sons. Defining this ‘udasi’ classification is difficult. Suffice it to say that the historical roots of this Nanakpanthi group were in Bihar, where they have maintained a controversial separate identity during the solidification

of the now hegemonic brand of Sikhism in the Punjab. Udasis are often put in the category of ‘sahajadhari Sikhs’. See Oberoi 1994: 76–80 and 127–31, and William R. Pinch 1996: 36, n54. 17 Brajratna Das, in Bharatendu Harishchandra (1948), wrote that Sumersingh and Harishchandra were close friends, that they would discuss religion and poetry, and Sumersingh received all of his famous friend‘s publications (174). 18 ‘Hari’ being a synonym for ‘siṃha’ (lion), Sumersiṃh could also read as Sumer Hari. The young Ayodhyasingh took ‘Hari’ for his ‘siṃha’ as well, formed the contraction ‘audha’ from ‘Ayodhya’, and reversed the two elements, resulting in his nom de plume ‘Hariaudh’, sometimes spelled Harioudh in English transliteration. 19 D. Simha 1986: 134. ‘Samasyapurti’, means completion of a verse, i.e., provision of a first part of a verse for a ‘samasya’ or problem which is the latter part of that verse. 20 Max Arthur Macauliffe 1909, 1963: xii. 21 Kundaliya is a verse form in which ‘the stanza resembles a coiled serpent’ (kundali), and which showcases the poet‘s virtuosity. Its form consists of a doha (couplet) and a rola (quatrain), in which the last foot of the couplet is repeated at the beginning of the following quatrain, and the end of the quatrain repeats the first word of the beginning couplet. See Rupert Snell 1991: 22. 22 Kishorilal Gupta 1966: 59. 23 As cited in D. Singh 1986, his titles were as follows: Guru-pada prema prakāś, Nitya-kīrttan, Gurukīrti-kavitāvalī, Vedī va ṃśottama sahasranāma, Sorho sahasranama, Khalsa pañcāśikā, Siṃha suryodaya, Gurucarit-darpan, Bihari-Sumer, Bedīvar dohāvalī, Dāridradukha-khaṇḍan dohāvalī, Sumer-bhūṣaṇ, Śravaṇa-maraṇa, Premprabhākar, Premsudhākar, Sandesā, Ulāhanā, Gurukul praśnottarī, Śabdāṅk, Jagat naikarī, Durvāsā-māhātmya, Purāṇopakramaṇika, Gurubhaktamāla, Rahatdarpaṇa, Vivek-varidhi, and Khālsā-śataka. 24 D. Simha 1986: 133. 25 Hariaudh 1934: 513. 26 Aside from Gupta’s article (1966), brief mention of him is made in the literary compendium Misrabandhu vinod and passing mention in the Śivasimh saroj. Most of the information I have found was published in Simha’s book on the Khadgavilas Press (1986), which published some of his work. 27 Foreword, Śhri guru-pad prem prakāś, kṛt Bābā Sumer Siṃh, A. S. Kahlon (ed), Patiala: Punjabi University, 2000: ix. This editor, and that of the Japu sidhant (Atha japu sidhānta suryodaya arthat Śri Japujī da bhāva arth, jiu sūraj chaṛiya: sammat 1954 [1897 CE] di krit likhat Bābā Sumer Singh Bhallā jī valõ kāvi ṭīkā (1987), were not aware of, or not interested in, Sumersingh’s connection to the authors of the modern Hindi canon. 28 ‘Arun’ [Atmaram Sharma], 2000: i-iii. Arun cites ‘Bihari satsai sambandhi sahitya’, Nāgarī Pracāriṇi Patrikā, Shravan 1928: 348–49. Interestingly, this author believes that Ratnakar (at least partly) plagiarised his Braj poetry from the old manuscripts he collected in Patiala and other sources. I have yet to find compelling direct evidence in Arun’s several publications, though further analysis is required. The following extract exemplifies Arun’s stance (2000: iv): Ratnakar did not publish all

of the acquired texts just as he found them… In relation to this, my belief is that Ratnakar did some alteration of… the poetic texts he acquired… Perforce of this very altering, he became the Ratnakar of Braj Bhasha… with considerable effort to keep all the many facts hidden, making many learned poets ignored in the world of literature, he has disestablished them forever. 29 See Harjot Oberoi 1994, and Kenneth Jones 1973: 457–75. 30 Hariaudh voiced these ideals through his Radha in Priyapravās (1914), who reinterprets nine-fold bhakti as particular types of social service. 31 Hariaudh 1934: 514–15. 32 Sarasvatī, June 1901; Reprinted in Bharat Yayavar (ed) Mahāvīra-prasāda Dvivedī racanāvalī, Vol. 2. New Delhi: Kitab Ghar 1995: 55–58. 33 See Karine Schomer 1983. 34 Written under the pseudonym ‘Ek Bhāratīy ātmā’ in Viśāl Bhārat, July 1928. Cited in Ray 1958: 16. 35 In contrast, it seems that Avadhi, the language of courtly, Sufi and Bhakti poets like Jayasi and Tulsidas (see de Bruijn in this collection), did not figure as an option for either Hariaudh or Ratnakar, despite their location in Avadh. This attitude indicates the status of Braj in the intervening centuries as the more widespread courtly literary dialect. 36 Hariaudh, Introduction to the second edition of Venis kā bāṅkā, 1928: 3. 37 This publisher is still in existence, and was a major distributor of Hindi publications in India and to Indians abroad, etc. See http://www.khe-shri.com/ khemraj.htm. 38 Although Hariaudh engaged with these works in order to disengage Braj Bhasha-cum-Hindi from Urdu, we might also surmise that Hariaudh chose these works with aesthetic concerns in mind. The Fasana was reputed to be an indigenous precursor to the European style novel; Anis was considered to have had a ‘realistic’ bent and also the quality of ‘karuṇa’. 39 Through the transcreation of Anis’s couplets into quatrains (chaupai) on baby Krishna, playing off Anis‘s original use of ‘lālā’. An article is forthcoming on this topic, co-authored by myself and Amy Bard. 40 Christopher King 1994: 36. 41 Hariaudh, Introduction to Kāvyopavan arthāt nānārasamayī kavitāvali, 1909: 2–3. 42 A small princely state in Bundelkhand and the place of employment for author Gulabray (1887– 1963) and later the ‘political’ Braj poet, Viyogi Hari (1896–1988). See biographies in the Appendix of Orsini 2002. Lala Sitaram also mentions procuring a seventeenth-century text from a munshi in Chattarpur in his Selections from Hindi Literature Book VI, part II, Other Poets, 1926: 1, testifying to Chattarpur’s status as cultural ‘capital’, Braj-wise. 43 For details, see Valerie Ritter 2004.

44 Ray 1958: 4. The only remnant of his Urdu poetry I have found thus far (in a secondary source) is an incomplete couplet about ‘guru-bhakti’, addressed to his ustad, Mirza Muhammad Hasan, and it is not ascribed to any publication, Var 1965: 3. 45 See Orsini 2002: 1.3.3, for description of the samasyāpūrti event, which took place in courts and in private parties, often with prizes, and which found new life in the later Hindi kavi-sammelans. 46 Orsini 2002: 39–40. 47 A quite realistic painting of a wooded scene illustrates ‘Nature’ for the volume, and a print of a Caucasian child, clearly culled from Victorian commercial art, illustrated hāsya-rasa. 48 Joanne Punzo Waghorne, 1994. 49 Illustrated Braj Bhasha and Sanskrit texts, among others, are surveyed in Vishakha N. Desai 1985, especially with essays by B.N. Goswamy and A.T. Embree. 50 Maharaja Pratap Narayan Singh 1894: 2–4. 51 See biography in Appendix of Orsini 2002. Singh’s chart is compiled in Selections from Hindi Literature (Book V, Ars Poetica [Calcutta: University of Calcutta, 1924]), pp. 1–4. 52 A case in point is the following kavitta, in which the gopis tell Uddhav to convey their pain of viraha to Krishna, here designated with the title sirtāj, a term found in Surdas but perhaps less common now:

Ausara mile au saratāja kachū pūchahī tau, Kahiyau kachū na dasā dekhā so dikhāiyau. āha kai karāhi naina nāra avagāhi kachū, kahibe kaũ cāhi hicakā lai rahi jāiyau. Ratnakar 1967: 103 If it happens that our lord [saratāja] asks about us, Don’t tell him anything, show him our state you’ve seen: Sighing and moaning, drown yourself in tears, Want to say something—but keep on sobbing. 53 Usha Jaiswal 1956: p. 119. 54 Ray 1958: 28. 55 Introduction to Bihari Lal, Bihārī-Ratnākar: Bihārī-satsaī par Ratnākar kī ṭīkā, Jagannathdas Ratnakar (ed) 1979 (1925): vii. 56 Introduction to Ratnakar, Uddhava-śataka, 1967 (1929): 51. This said, Ratnakar did not standardise Braj in a formal sense. He was reputed to have developed an unpublished Braj Bhasha grammar. Dhirendra Varma decided to undertake such a project in 1922, and it was only in 1937, with his Brajbhāṣā Vyākaraṇ, that literary Braj Bhasha obtained a grammar in Hindi. Ratnakar’s

hand in editing the early Hindi grammar of Hindi by Kamtaprasad Guru (Hindī-vyākaraṇ 1920) likely solidified his usages of ‘Braj’ versus ‘Hindi’; See Ray 1958: 5. 57 Jaiswal 1956: 15–18. 58 Ray 1958: 8. 59 Jaiswal 1956: 18. 60 Cited in Savitri Srivastava 1981 (1976): 2. 61 Introduction to Ratnakar in S.S. Das (ed) 1950 (1933): Vol. 1: 4. 62 Cited in Jaiswal 1956: 11. 63 Girish noted that ‘he did not wear khadi but he didn’t use foreign cloth either. Having a poet’s heart, he liked to wear elegant svadeshi clothing. In this matter he followed the great poet Rabindranath Tagore,’ Girish 1959: 3. 64 Jaiswal 1956: 102. 65 Orsini 2002: 50. 66 ‘Those in whose hearts are harsh restrictions on anuprāsa, / Why don’t they give up these poetic restrictions and write prose? / Anuprāsa never diminishes the power of a good poet, / But in truth gives rise to new perceptions in the heart’. Verse 157 of ‘Samālocanādarśa’, Nāgarī Pracāriṇī Patrikā, 1:1, 1897. The last six verses (154–59) comprised his addendum to his translation of Alexander Pope’s ‘Essay on Criticism’, which was later published separately by the NPS in 1919. 67 This is the ‘Ādhunik kavi’ series with which every Hindi literature student is familiar. At the front of each of these books is a narrative of the Raja’s donation, framed in terms of Orchha’s patronage history.