Indian Economic & Social History Review http://ier.sagepub.com/

How to do multilingual literary history? Lessons from fifteenth- and sixteenth-century north India Francesca Orsini and (SOAS) Indian Economic Social History Review 2012 49: 225 DOI: 10.1177/001946461204900203 The online version of this article can be found at: http://ier.sagepub.com/content/49/2/225

Published by: http://www.sagepublications.com

Additional services and information for Indian Economic & Social History Review can be found at: Email Alerts: http://ier.sagepub.com/cgi/alerts Subscriptions: http://ier.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://ier.sagepub.com/content/49/2/225.refs.html

>> Version of Record - Jun 18, 2012 What is This?

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? Lessons from fifteenth- and sixteenth-century north India Francesca Orsini (SOAS) How can we conceptualise multilingual literary culture, and how can we research it? Using the turbulent ‘long fifteenth century’ in north India as a site, this article questions research models based on single languages (Hindi, Urdu) and engages critically with early modern taxonomies and archives. The article focuses on the materiality of the archive—the language, script and format in which texts were written down and copied—on the spaces and locations in which literature was produced and performed, and on the oral-performative practices and agents that made texts circulate to audiences in ways not bound by the script in which the texts appear to us. Not only are the models of composite culture and language-specificity questioned as a result, but the sites of literary production move from the court to a series of intersections, and areas that were peripheral move into view and connect with others. Keywords: multilingualism, literary culture, north India, Hindi/Hindavi

The Problem The first histories of north Indian literatures, written in the colonial and nationalist periods, were involved in crystallising communities around language and cultural identity. While each in its own way had to negotiate the origin and growth of its particular object, Urdu or Hindi, within the multilingual environment, they also quickly established boundaries, a canon, selective affiliations and, of course, significant exclusions. Acknowledgements: This article is the result of a three-year AHRC-funded research project on ‘North Indian Literary History from a Multilingual Perspective: 1450–1650’. I am grateful to the AHRC and to the many participants of the project, whose work I refer to in the essay. I would also like to thank Bodhisattva Kar for his critical reading of the essay, and the anonymous reviewer of IESHR for the valuable suggestions.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46 SAGE Los Angeles/London/New Delhi/Singapore/Washington DC DOI: 10.1177/001946461204900203

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

226 / FRANCESCA ORSINI

In šb-e ƒay€t, for example, Muhammad Husain Azad argued that Urdu had its origins in Braj Bhasha—perhaps a gesture to early specimens of bilingual rekhta— but then quickly established Urdu as a ‘north Indian’ vernacular whose literary affiliations were exclusively with Persian genres and tropes. In doing this, he downplayed the significant history of Dakkani, of Gujri, of non-Persianate vernacular poetry in the North (even that written and transmitted in the Persian script), as well as of Hindu poets writing in Urdu.1 George Grierson’s Modern Vernacular Literature of Hindustan embraced the catholic view that Hindi literary history included all its medieval dialects, a view shared by Hindi literary institutions of the time and subsequent historiography.2 But while Avadhi Sufi poets, whose earliest manuscripts were in Persian script, were included within Hindi literature, though their religious identity remained a ‘problem’ that required justification, rekhta (that is, Urdu) poets in Khari Boli were left out. The Hindi literary tradition appeared as a consequence to be formed exclusively by devotional and courtly poetry, and even the oral epics, tales and songs that Grierson the folklorist–linguist eagerly collected were not included in this early canon of Hindi literature. As Imre Bangha has noted, the criteria for inclusion were inconsistent, based on script, genre, topic or poetic language, and they seem to follow a preordained cultural logic: thus ghazals and masnavis were automatically classified as Urdu, while devotional or Braj Bhasha courtly (riti) poetry, even by Muslim poets, came under Hindi.3 Responding to their nineteenth-century politics, such literary histories were teleologically motivated. They encapsulated, and further promoted, a historical view in which Muslims (and Urdu and Persian) were regarded as ‘foreign’ elements in Indian culture. In this view, Urdu (and Indian Muslims) had little or no affiliation to local cultural forms but belonged to the Arabo-Persian ecumene. Undoubtedly new ideas of language and literature as the expression and property of the ‘people’ (jati or qaum) deeply influenced the way in which nineteenth century scholars related to the literary past.4 And while certain types of diglossic relations—Persian–Urdu, Sanskrit–Braj Bhasha—were valued and foregrounded, others that had been equally important to the literary culture, like Persian and Hindavi and Persian and Braj Bhasha, were sidelined.5 To what Azad, šb-e ƒay€t, p. 6. King, ‘The Nagari Pracharini Sabha of Benares’. 3 Bangha, ‘Rekhta: Poetry in Mixed Language’. 4 Thus Hafiz Mahmud Sherani viewed the early instances of Hindavi in the fourteenth and fifteenth centuries (8/9c H) as evidence of the creation of Urdu as a Muslim language: ‘These words and expressions, in my opinion, are enough evidence for the antiquity of the Urdu language, and in truth it can be said that this language was commonly spoken among Muslims in this period... we see that Muslim peoples (aqw€m) created a special language for themselves in India and as they spread thanks to their conquests and victories, this language spread eastward, westward, to the North and to the South as well, together with them’; ‘Urdu fiqre aur dohre’, in Maq€l€t-e ž€fiz Maƒm™d Sher€n…, ¯ p. 132, emphasis added. 5 Phukan, ‘Through throats where many rivers meet’. 1 2

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 227

extent these nineteenth-century taxonomies followed early modern ones is an intriguing question. As we shall see, earlier archives had their own strategies of exclusion. The myth-making and exclusions involved in the competitive and teleological Hindi and Urdu literary histories have been critiqued by several scholars in the past decade or so (Faruqi, King, Dalmia, Bangha, Busch). Recent Urdu literary histories by S.R. Faruqi, S. Jafar and G.C. Jain, and J. Jalibi have contested existing generic boundaries and widened the scope of Urdu literature to a great extent, yet these histories still assume a single-language object/tradition, Urdu. Given the institutional and ideological investment in ‘Hindi’ and ‘Urdu’ and the historical baggage of suspicion, it is of course unlikely that this state of affairs will end. Hindi and Urdu departments will obviously continue to exist in universities. This article focuses on literary culture in north India in the fifteenth and sixteenth centuries, the period of efflorescence of the north Indian Sultanates and the beginning of the Mughal empire. Rather than viewing the fifteenth and earlysixteenth centuries as the ‘twilight of the Delhi Sultanate’ (Lal), it considers the ‘long fifteenth century’ as a period of considerable regional political, cultural and religious dynamism—much like the eighteenth century—and as the beginning of widespread vernacular literary production in north India.6 After all, this is the time when the powerful voices of Kabir, Nanak, Surdas and other early saint– poets emerged, when the sophisticated Hindavi Sufi romances of Qutban, Manjhan and Jayasi were written, when Vishnudas of Gwalior retold the epics in the vernacular, and when singers and songs circulated intensely among courtly and religious milieux. What this article suggests is that in order to develop an alternative historical vision to the distorted one of exclusive, single-language histories it is necessary to take the multilingual reality of north Indian literary culture seriously. This requires a comparative perspective that takes in both cosmopolitan and vernacular languages, both written archives and oral performances, and texts and genres that ‘circulated’ in the same place and at the same time although they were ‘transmitted’ in separate traditions. Only then can we pull together the different parts of the same cultural and social world in order to draw out areas of convergence, silences and exclusions within its constituent parts. It is necessary to both critically take on board and question early modern taxonomies, since they often ignore and exclude aspects of the multilingual linguistic and literary world that were undeniably present in the fifteenth and sixteenth centuries. A focus on the materiality of the archive—paying attention to the language, script and format in which texts were written down and copied, and to combinations of texts and genres that were copied together—on the spaces/locations in which literature was produced and performed, and on the oral–performative practices and agents 6 See the forthcoming volume edited by Samira Sheikh and myself, provisionally titled After Timur Left.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

228 / FRANCESCA ORSINI

that made texts circulate to audiences in ways not bound by the script in which the texts appear to us, can help us greatly in trying to imagine the multilingual world of early modern north Indian literary culture. In other words, it is important to approach multilinguality as a set of historically located practices tied to material conditions of speech and writing, rather than as a kind of natural heterogeneity. Contexts, Categories and Definitions Before we consider the three elements mentioned above—the materiality of the archive, the spaces/locations of production and circulation, and oral–performative practices and agents—a few words about contexts, categories and definitions. Local taxonomies and ‘emic’ nomenclature are clearly important when trying to reconstruct the cultural world of fifteenth and sixteenth century north India. Was what to us appears as a single linguistic region, the ‘Hindi belt’ stretching from western Rajasthan to Bihar and from the tarai to Malwa and encompassing the states of Uttar Pradesh, Bihar, Uttaranchal and Madhya Pradesh, perceived as a ‘region’ then? What were emic understandings of literature and literary genres? Were they identical in the different languages and socio-literary communities, and did they coincide with our understandings today? How were languages called and differentiated? Political regions were, as ever, crucial, and in the period under review north India was divided into the Sultanates of Delhi, Jaunpur and Malwa, with a significant ‘Rajput’ presence in the countryside and in a few forts and fortified towns (for example, Tomar’s Gwalior). Geographical nomenclature referred to these entities or to individual towns and villages. Thus in fifteenth-century sources Awadh designates the city of Awadh/Ayodhya rather than the region, though it became an administrative unit (suba) under the Mughals. Nonetheless, going by Sufi sources, circulation and trade across north India between Bengal, Jaunpur, Awadh and Delhi and between this region and that of Gwalior and Malwa was easy and intense, despite the occasional destruction brought by the battles and raids between Sharqi Jaunpur, Lodi Delhi and Khilji Malwa, and despite Timur’s invasion of 1399 and occasional rebellions by local elites.7 In terms of language, the modern regional linguistic definitions of Braj Bhasha, Avadhi, Bhojpuri and Khari Boli are not reflected in the sources, which instead speak of a generic bhakha (bhasha) or Hindavi/Hindui/Hindi8 (in Persian texts) as the vernacular of north India. The perception one gets from the great circulation of tales and songs is of a general intelligibility of bhakha/Hindavi across the region in this period—Daud’s ‘Avadhi’ romance Chand€yan was recited from the 7 For seasonal circulation of peasant armies in this period, see Kolff, Naukar, Rajput, Sepoy. For rebellions by local ‘Rajput’ elite in Awadh, see Sinha, Subah of Allahabad under the Great Mughals. 8 To avoid confusing this early definition of the north Indian vernacular with modern Hindi (Khari Boli), I will use Hindavi in this essay.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 229

pulpit of a mosque in Delhi without any comment on the ‘eastern’ flavour of the language.9 At least until the late sixteenth century the general terms bhakha and Hindavi seem to denote not just a lack of grammatical and taxonomic interest towards the vernacular but also a continuum, with locally produced songs and tales that could travel and be understood over the whole of north India. As we shall see later, oral performers and performance contexts were crucial in this respect. Performers were able to modify inflections and replace words that were too local while keeping to the metrical scheme. These changes would then be written down by scribes and resurface as the great linguistic variation visible between, for example, the ‘eastern’ and ‘western’ recensions of Kabir’s songs.10 Not surprisingly, the elevation of Braj Bhasha to the status of courtly poetry with an alankarashastra pedigree in the late sixteenth century led to its recognition as a named vernacular. Braj Bhasha became a ‘cosmopolitan vernacular’ (Pollock) with a standard poetic language that needed to be learnt properly through manuals (riti-granthas) and with teachers (if not grammars per se, until the Persian Tuhfat al-Hind, 1675).11 As is well known, the growing status and popularity of Braj Bhasha poetry induced poets like Tulsidas to use it for his songs and verses.12 Thus, while north India was not a homogenous region in political terms, it seems to have been a fairly well-connected cultural and linguistic region. Its linguistic economy can be described as one of ‘multiple diglossias’.13 This means that there were several written/High languages—Arabic, Persian, Sanskrit14— and a general spoken vernacular (what I call here Hindavi) written in either Persian, Kaithi, or Devanagari scripts. A simplified form of Persian also seems to have been a spoken lingua franca, while individuals and groups also maintained their own spoken languages (for example, ‘Turki’ or Pashtu) for generations. Material traces of this Persian–Turki–Hindavi multilingualism are scant yet unmistakeable: Persian dictionaries compiled in India in this period are particularly multilingual and include Turki and Hindavi synonyms, and in the case of Turki also many lemmas. The poor command over Persian of some Pashtu-speaking 9 Cf. instead Anandram Mukhlis’s comment on the ‘sweetness of the purabi tongue’ when he heard it recited by his servant; Phukan, ‘Through throats where many rivers meet’, p. 35. 10 See Vaudeville, Kabir. For word substitution by singers, see Callewaert and Lath, ‘Musicians and Scribes.’ 11 See Busch, Poetry of Kings. 12 The songs collected in his KŠa G…t€val… (ca. 1590), G…t€val… and the poems in the Vinaya Patrik€, Kavit€val…, and many of the verses in the Doh€val…; see R.S. McGregor, Hindi Literature, pp. 114–57. 13 See María Angeles Gallego, ‘The languages of medieval Iberia.’ 14 Prakrit had a very limited but symbolically important status for Jains, who also continued to write in Apabhramsha until the sixteenth century; one of the major Apabhramsha poets, Raydhu, lived and wrote in Gwalior in the fifteenth century; see Eva de Clerq, ‘Apabhramsha as a literary language’, and Granoff, ‘Mountains of Eternity.’

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

230 / FRANCESCA ORSINI

Afghan amirs is occasionally commented upon. And a few compositions of the Sikh gurus show that a simplified form of Persian was current as a spoken language in Punjab in the late-fifteenth and sixteenth centuries.15 Persian words had of course entered everyday language use—particularly in the domains of administration, power and warfare—and can be found in a number of texts in spellings inflected by local phonology. Even a text famously known for its low percentage of Perso-Arabic lexicon such as Jayasi’s romance Padm€vat (1540) employs a number of such words when describing the Sultan’s [Alauddin Khalji] army, suggesting at the same time a heteroglot body of soldiers: Dhani sult€n jehika sam . s€r™, Sabai turuk sirt€j bakh€ne, L€khanha m…r bah€dur jang…, Jeb€, kholi, r€ga sõ mahe, Barana barana au p€ntihi p€nt…,

Uhai ka˜aka asa jorai p€r™. Tabala b€ja au b€ndhe b€ne. Jantra kam€naine t…r khadang…. Lejim gh€li ir€kinha cahe.[...] Cal… so sen€ bh€ntihi bh€nt….

Behara behara sab kai bol…,

Bidhi yah kh€ni kah€n saun khol…. [499, 1–4, 6–7]

Happy is the Sultan who owns the world, and who can assemble such an army! All sing praises of the Turkish chiefs, with drums and war-attire. Thousands of Mirs and brave warriors, with mechanic bows and khadangi arrows. Ready with armour, cannons, and leg-covers, with iron-stringed bows they mounted Iraqi [horses] Kind after kind, row after row, the varied army went. All differed in speech—where did God open such a treasure?!16 There are comparatively few traces of this multilingualism in the texts and in the internal organisation of the archives. For example, Persian biographical dictionaries may mention that certain individuals composed or listened to poetry and songs in Hindavi as well as Persian, but they never quote the poems and songs.17 These silences and exclusions of the early modern archive tell us a lot about cultural hierarchies of the time, but it is equally important to question them. 15 For Indo-Persian dictionaries, see Baevskii, Early Persian Lexicography; for references to Afghans’ poor command of Persian, see Siddiqui, Waqi ‘at-e-Mushtaqui, p. 9; for the ‘spoken Persian’ hymns of the early Sikh Gurus, see Shackle, ‘Approaches to Persian Loanwords’. 16 Agraval, Padm€vat, p. 527. See also the vernacular tale composed in 1493 for Baghela patrons who were, as Simon Digby puts it, familiar with ‘the governmental framework of Muslim power in the Gangetic plain’; ‘Two Captains’, p. 165. The tale shows a comparatively high number of PersoArabic words related to the military: jin/zin (reins), samser (sword), tirandaz (archer), taslim . (subordination), asrar/israr; Bhima Kavi, angvai Kath€. 17 For example, Badauni, Muntakhabu-t-Tawarikh.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 231

For this reason, though diglossia is the term most often used to describe the relationship between vernacular and classical languages, because of the hierarchical connotations of the term and of the dynamics it describes (High vs low language) it is too restrictive for the range of phenomena and relationships that we are trying to grasp. Diglossia undoubtedly fits in some cases, for example in the case of Braj Bhasha riti granth in relation to Sanskrit texts of poetics.18 But what about Persian and Hindavi? Did Sufi and courtly poets versed in Persian literary genres consider Hindavi verses and stories equally literary, or literary at all? Did the authors of the sophisticated Hindavi Sufi romances consider their works inferior to Persian poetry? The fifteenth-century Awadhi Sufi Abd al-Quddus Gangohi, who wrote in Persian and knew Persian poetry but also composed Hindavi verses with the pen-name ‘Alakhdas’ (the literal translation of Abd al-Quddus), posited a definite parallelism between Persian and Hindavi poetry.19 Poets like Manjhan, Jayasi and Qutban recognizably borrowed elements from Persian masnavis for their Hindavi romances, but expressed clear pride at the beauty and depth of meaning of their works.20 And what about the intensive multilingualism of an author like Vidyapati? For his courtly patrons and audience in Tirhut, in Mithila, he composed several works in Sanskrit on ritual, devotional practices, and the law, as well as a letter-writing manual and a kind of Mirror for Princes, the Purua Par…k€.21 His Sanskrit play Gorakavijaya contains songs in Maithili, and he became best known for his vernacular songs devoted to Krishna, which however were not collected until much later. He himself championed the local version of Apabhramsha, which he called avaha˜˜a, for literary composition in his K…rtilat€, a work narrating the journey of king Kirtisingh to the city of Jaunpur to enlist the military help of Ibrahim Shah Sharqi.22 Though such a high level of sophisticated literary writing in Sanskrit, Apabhramsha, and the vernacular (Maithili) by the same author is rare, if not unique, it tells us about the growing bi- or tri-lingualism of literary culture in this period. As writers on diglossia themselves are eager to point out, a low language, that is, a vernacular, is not always used for ‘minor’ functions and texts, or always in an inferior relationship to a High language.23 For this reason, I have used diglossia only when the relationship between two languages was clearly a hierarchical one, that is, when it was perceived as such. In the other cases I have preferred the terms bilingualism and multilingualism (or heteroglossia). Busch, Poetry of Kings. See S.A.A. Rizvi, Alakh B€n…. 20 They often followed Persian poets in using precious stones as metaphors for their precious words, i.e., ‘threading pearls’ (Plukker, The Mirig€vat… of Kutubana, p. 5) or ‘a ruby hidden in the dust’ (Agraval, Padm€vat, p. 22, also the title of de Bruijn’s monograph). 21 Vidyapati, K…rtilat€, ‘Introduction’, pp. 12–13. 22 McGregor, Hindi Literature, pp. 29–30. 23 Giovanardi, ‘Il bilinguismo italiano-latino.’ 18 19

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

232 / FRANCESCA ORSINI

Finally, while there is enough evidence of vernacularisation, this does not mean that a monolingual vernacular literary culture replaced the multiple diglossias. On the contrary, given the continuing and in fact expanding role and status of Persian as a necessary requirement for administrative jobs and elite culture, the symbolic role of Sanskrit, and the persistent heteroglossia that rendered ‘rough and ready bilingualism’ (S. Guha) a must for anyone engaged in trade, the army, religion, and performance, literary culture in north India witnessed a parallel growth in Persian and the vernacular from the fifteenth to the eighteenth centuries. This was a continuous and enriching process of in-, out- and cross-translation between classical languages (usually through the vernacular), or between classical languages and the vernacular and vice versa. Scripts, Writing and Copying From the perspective of modern scholars looking for the earliest specimens of Hindi and Urdu, it was an easy step to either take script as proof of language identity, or else to brush aside the evidence of script when it went against one’s conception. Thus Hafiz Mahmud Sherani, whose pioneering work on vernacular traces in Persian texts in the 1930s is still unsurpassed, considered those words and phrases as evidence of Urdu just because they were written, by Muslims, in the Persian script (see fn 5). By contrast, Mataprasad Gupta disregarded the evidence of script and insisted on positing a putative Devanagari/Kaithi ‘original’ for Jayasi’s Padm€vat because to him this was a Hindi text.24 Because of the ravages of war and climate there are unfortunately hardly any extant manuscripts from the fifteenth and sixteenth centuries, and even fewer documents that would help trace non-literary language and script use. Yet a few examples still remain that can be discussed. Moreover, even a few general pointers can help us avoid misapprehensions regarding a putative identity of script, language, and community such as those just mentioned. The first point to be made is precisely one regarding the relationship between script and language. Script was first and foremost a skill linked to education and professional use—only for religious texts, and even then not in all cases, can we posit an association between language and script at this time (for example, Islamic and Sikh scriptures). Though little material evidence has survived, historical sources tell us that the chancellery of the Lodi Sultans and of Sher Shah Suri was bi-lingual and bi-scriptual and included both Persian and Hindavi scribes.25 Gupta, J€yas…-granth€val…. A land-grant from a village near the qasba of Sandila (now district Hardoi) to a local Sheikh from Sultan Ibrahim II, son of Sikandar Lodi, [dated 927H (1520)] records the deed in Persian taliq and in Hindi Kaithi script below, and there is a similar (though illegible in print) grant from Sher Shah’s time; Shafi, ‘Three Old Documents’, pp. 281–85. This shows that Abbas Sherwani’s statement that Sher Shah had appointed two karkuns in every pargana, one to write Hindi and the other to write 24 25

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 233

Sumit Guha, who has reflected on this ‘biscriptual situation’ and the specialised scribal groups behind it, argues that in the Sultanates ‘a working arrangement soon crystallised, by which high prestige and public functions were controlled by a Persianate elite’, which formed a ‘caste-like’ body endowed with special skills and practices and recruited by patronage and cooption, and which used Persian for government business and Hindavi with outsiders. Less lucrative state functions might have been contracted out to local functionaries who enter the historical records only occasionally.26 Some local functionaries may have known Persian,27 but even local karkuns only literate in their own specialized script (Kaithi rather than Nagari in north India) heard the stereotyped Persian expressions of official documents frequently enough and incorporated them in their own vernacular documents.28 As is well known, Todar Mall’s decision ca. 1582 to make Persian the language of administration even at the lower levels of district administration gave a tremendous impetus to Persian literacy and education (including literary education) among Hindu scribal groups, and Persian-knowing Hindu scribes came to dominate the scribal profession at all levels. But there is evidence of continued biscriptualism in local documents such as the ‘Vrindaban documents’ (early 17c)—and of course in the chancellery of sub-imperial states such as Amber/Jaipur.29 The two languages and two scripts in these documents speak of different audiences and socio-cultural communities, but in the case of literary texts we do well Persian, had its roots in Lodi’s times; cf. Momin, The Chancellery and Persian Epistolography under the Mughals, p. 28. 26 Guha, ‘M€rg…, De… and Y€van…’, p. 132. 27 Although there are references to the fact that already in the time of Sikandar Lodi Hindus ‘learned to read and write the Persian script, which had not been common among whom until then’ (‘k€fir€n bakhw€ndan o–neveshtan-i khat-i f€rs… ki t€ €n zam€n dar miy€n-i …sh€n ma‘m™l nabud, ¨ pard€khtand’); Tabaqat-i Akhbarshahi, quoted in ‘Sikandar Lodhi aur uske ‘ahad ke ba‘z farsi musannifin’, p. 29. 28 At the same time, vernacular words and expressions crept into Persian letter-writing, especially in the ‘functional’ parts of grants and correspondence; Guha, ‘M€rg…, De… and Y€van…’, p. 134. The Lodi grant mentioned above speaks of ‘zam…n-i banjar’ (uncultivated land, p. 283), while a private letter (dated 955H/1548) written by a courtier of Islam Shah warns the addressee against using force (‘dhakka nakunand’) against a local Sheikh with good contacts at court. Perhaps because written in a clearly agitated state of mind, the letter is written in very simple Persian and lets the odd Hindi word in; Shafi, ‘Three Old Documents’, p. 287. The writer also uses an odd causative, ‘farm€n ... €dar m…kan€n…dand’ (‘they had an irrevocable order issued’), that looks like a calque of the Hindi causative karv€n€. 29 The letter known as Krishnadasa’s document (1637) from Vrindaban is somewhat different from the Lodi example, with Braj Bhasha and Persian side by side, left and right respectively: here the Braj Bhasha side is surprisingly free of Persian expressions and the Persian side is instead full of untranslated Hindi words (anucar, asthal, kun~j)—perhaps because written by a Hindu scribe? See Mukherjee and Habib, ‘Land Rights in the Reign of Akbar’. For examples from the Jaipur chancellery, see Horstmann, In Favour of Govinddevji. I am grateful to Monika Horstmann and Najaf Haider for their training sessions held at SOAS in 2007 and 2008 on how to read these documents.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

234 / FRANCESCA ORSINI

to consider the agency of the copyist and the literacy of the person commissioning the copy in order to understand the relationship between text and script, rather than immediately assume an identity between text, script, and socio-textual community. For example, we find the same narrative and song genres and texts— dhrupad, bishnupad, katha as well as doha—written in both Persian and Kaithi script, depending upon the literacy of the patron or copyist.30 Thus script is important evidence of their circulation across different and/or mixed audiences rather than of any perceived identity of the text. Jayasi’s Padm€vat for example, notes Thomas de Bruijn, was copied in Nasta‘liq, Devanagari and Kaithi scripts from the earliest stage of its textual tradition. Moreover, the difference in size and appearance among the manuscripts in the same script also suggests significant differences in the nature and purpose of the transmission of the text.31 The only copy of Malik Muhammad Jayasi’s tale of Krishna, Kanh€vata, that has a date (1067H/1657) and a colophon tells us that it was copied by a certain Sayyid Abdulrahim Husain, the son of a drug seller and a resident of Masauli near Kannauj, for a local kayastha, Rajaram Saksena of Qasimpur, also in Kannauj district: this suggests both that Rajaram Saksena was literate in Persian and that this ‘Sufi’ text circulated beyond Jais and beyond Sufi circles. Further clues are provided by the actual material form of the book, its size and quality and whether it was illustrated or not. As far as I know, books copied in Sufi circles were not illustrated, so the presence of several illustrated codexes of Hindavi Sufi romances, though lacking colophons and thus precise information about patronage, suggests that these tales were copied for elite patrons, possibly local amirs. But what to say of a Mirig€vat… codex copied in Kaithi ca. 1525?32 Sheikh Qutban had written the Mirig€vat… in 1503 and dedicated it to the Sultan of Jaunpur. Was this particular copy produced and illustrated for a local Hindu chieftain? Or, less likely, for an amir who preferred Kaithi? Illustrated books offer powerful evidence of the trend towards vernacularisation of Sultanate literary culture from Gujarat to Malwa. The same format of codex with Persian-script text and often full-page illustrations in a similar range of styles was used for Shahn€mas, Hamzan€mas and other Persian texts as well as for the Chand€yan and Mirig€vat….33 This was not the case with Mughal patronage of 30 For evidence of bishnupad and dhrupad in Persian script see Bilgrami’s Haq€’iq-i Hind… and the Sahasraras (ed. Sharma, see also Delvoye, ‘Collections of lyrics in Hindustani music’). In both cases only the song texts are in Hindavi/Braj Bhasha, while the commentary and introduction are in Persian. 31 de Bruijn, The Ruby in the Dust, ch. 1. 32 Held in the Bharat Kala Bhavan, Banaras (250 out of 253 folios); see Khandalavala and Chandra New Documents of Indian Painting, p. 107 (and Pl. 26 on p. 106). I am grateful to Preeti Khosla for this information and the reference. 33 See Khandalawala and Chandra, New Documents of Indian Painting; Goswamy, A Jainesque Sultanate Shahnama, and Brac de la Perrière, L’Art du livre dans l’Inde des sultanats, pp. 66–67.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 235

illustrated books, which seems to have been limited exclusively to Persian books— even when they were translations of Sanskrit or Hindavi texts. At the same time, the proliferation of illustrated books on paper (both in palm-leaf/manuscript and codex format) in the fifteenth and sixteenth centuries suggests the emergence of a common elite taste for the book as a precious object among Jain, Muslim and Hindu elites. And yet, in a context where so many texts (poems, songs, tales) were routinely read out and recited, the script in which a text or genre was transcribed or copied cannot be taken as an indicator of the limits of its audience, especially in the case of vernacular texts. Allison Busch has carefully examined the issue of the Mughal emperors’ taste for Braj Bhasha songs and verses, which was not dependent on their ability to read the Devanagari or Kaithi script: their experience was an auditory one.34 Persian poets who wrote complete texts in and on Braj Bhasha poetry always did so in Persian script, and with Persian commentaries, while other Muslim poets—especially in Rajasthan—composed only in Braj Bhasha and in the Nagari script.35 Should we think of bilingual literary elites in the first case, and of a more mono-lingual context in the second? A further intriguing point about script and writing in this bilingual environment is that while we may assume that at least some Persian-educated scribes had knowledge of the Kaithi (and Devanagari?) script, it appears that most scribes using Persian did not have a standardised orthography for Hindavi in mind when they transcribed Hindavi words and texts, and none was developed until ... I do not really know when! The five different copies of Abdul Quddus Gangohi’s fifteenthcentury Rushdn€ma, for example, are perfectly consistent in their Persian orthography but quite inconsistent when it comes to the Hindavi verses.36 Only in very few cases were Hindavi words written in the Persian script carefully vocalised in order to ensure correct reading.37 In most cases, Persian orthography of Hindavi did not even care to separate words. While we must assume that scribes knew what they were writing and were confident that those commissioning the copy would also be able to read the text (or find someone able to read it out for them), it is ironic that Persian was written down correctly though probably often spoken according to local phonology, while Hindavi was copied in Persian without any care for consistency.38 Busch, ‘Hidden in Plain View’ and Poetry of Kings. For Persian poets who also wrote in Braj Bhasha see Bilgrami, Ma’€thir al-Kir€m; for Muslim poets whose work is extant in the Nagari script, like Raskhan, Raslin, or Vajid, see McGregor, Hindi Literature. 36 Weightman’s ‘The text of Alakh B€n…’ compares the manuscript versions. 37 For example, Aqil Khan Razi’s transcription and translation of Jayasi’s Padm€vat, as shown by Shantanu Phukan, ‘Through a Persian Prism’. 38 We get clues about the local phonology of Persian from the way in which Persian words are spelt in Hindavi texts, and with exceptional clarity in the Persian hymns of Guru Nanak and Guru Arjan in the Tilanga section of the Adi Granth; see Callewaert, ®r… Guru Granth S€hib; Shackle, ‘Approaches to Persian loanwords’, and Orsini and Pellò, ‘Bhakti in Persian’. 34 35

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

236 / FRANCESCA ORSINI

Finally, and this leads us to the following section, although several of the manuscripts mentioned in this section were copied for individual patrons, writing and copying were activities most intensely connected with institutional locations and agents, or with an impetus towards institutionalisation and codification.39 The places—sacred and temporal—that were the main sites of literary production and circulation in the long fifteenth century included courts and the towns around them (Jaunpur, Delhi/Yoginipura, Gwalior, Mandu, etc.), but also centres of Sanskrit learning like Kashi and Tirhut, and the booming pilgrimage centres of Brindaban and Mathura in Braj, patronised by kings and courtiers as well as by ordinary pilgrims.40 While the imperial Mughal court as a centre for the production of Persian literary culture (and illustrated books) has been assiduously scrutinised, more recent scholarship has begun to trace the other, sub-imperial literary and artistic cultures that were enabled by the structure of, and mobility within, the empire, and the patronage given to religious groups in the shade of the empire.41 Among the most active copyists and collectors of vernacular songs and verses in this period we find religious groups—panths and sampradays like the Nanak and Dadu panths and the Vallabha sampraday, as well as Sufi lineages and khanqahs, and of course Jain lineages and families. Panths were also the first to write down and codify life-histories of vernacular poets in collections such as bhaktamals and vartas—thus harnessing the popularity of the poets and the power of their poetry to their own names. Though this kind of research has yet to be undertaken systematically for north Indian vernacular texts, in the context of Marathi Christian Novetzke has made a very useful distinction between two kinds of books, the more canonical pothis and the badas. While the former are ‘codifying’ collections usually written and copied by institutional agents and centres, the latter are notebooks copied by individual performers of the kirtan genre.42 In the context of north India, we may note a similar distinction between the fairly stable pothi tradition of Kabir in the šdi Granth and the Dadu-panthi Pancha-vani collections as against the more ‘unstable’ Kabir (and Surdas, and Mira and other 39 For the remarkably stable and systematic copying of manuscripts of Chaitanya’s biography, the Cait€nya-Carit€mta, thanks to the joint effort of the early disciples and the sponsorship of the Malla king of Bishnupur, see Stewart, The Final Word, chapters 1 and 7. 40 For Brindaban, the pioneering work is still Entwistle, Braj: Centre of Krishna Pilgrimage, but more recently see Pauwels, ‘The Saint, the warlord, and the emperor’; for a recent re-evaluation of the role of Kashi, which the poet Vidyapati from Mithila also referred to a centre of learning in the early fifteenth century, see O’Hanlon, ‘Speaking from Siva’s temple’. 41 Busch, Poetry of Kings; see Chatterjee, ‘Cultural Flows and Cosmopolitanism’ for the cultural impact of Raja Man Singh in Bengal; and Saha, ‘Between the Temple and Court’ for the patronage to the Vallabha sampraday. 42 Novetzke, Religion and Public Memory, chapter 3.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 237

popular poets) of singers’ ‘notebooks’ such as the 1582 Fatehpur manuscript Pada S™rad€saj… k€.43 Location and Circulation The most influential and compelling argument about the relationship between language, literature and politics, and between a High Language (Sanskrit) and the vernaculars has been put forward by Sheldon Pollock. Marshalling an impressive range of genres and examples, he has argued that (a) vernacular languages were first liberalised (that is, written down, usually for documentary purposes) and then, often after a considerable gap, literalised, that is, used for literary, imaginative, performative and expressive purposes (what he calls ‘workly’); (b) any vernacular innovation is linked to a reconfiguration of the culture-power order, when in place of cosmopolitan imperial polities more regional, vernacular polities emerge; (c) vernaculars then themselves became cosmopolitan and were used simultaneously, as had been already the case of Sanskrit, both for political and literary discourse, ‘with the court functioning as engine for the stimulation of literary production of a textualized sort’;44 (d) in order to become literarised, the vernacular [had to] emulate the superimposed models of literature of the cosmopolitan language, for ‘there is no parthenogenesis in culture’ (id.: 318). As he puts it, ‘vernacular poets achieved expressivity by appropriating and domesticating models of literary-language use from superposed cultural formations’ (id.: 298). Pollock is very clear that what he means by literature is what his authors meant by literature, that is, k€vya and s€hitya,45 a set of genres and discourses highly regulated by the ‘science of literature’, s€hitya€stra. ‘By contrast, the world of the “uncultured”, that is, of the uncourtly and noncosmopolitan languages of Place [deabh€€], was subliterary: a domain of the sung, the unwritten, the oral.’46 And since his Sanskrit theorists disdained orality and literature in non-cosmopolitan languages and consigned songs to a different order of discourse (g…ta), for Pollock, too, literature that does not follow the courtly practice of k€vya is simply not literature at all.47 43 See Bahura, Pada S™rad€saj… k€; also Callewaert and Lath, ‘Musicians and scribes’, in Hindi songs of Namdev, pp. 55–117. 44 Pollock, The Language of the Gods in the World of Men, p. 337. 45 And prashasti, that is, workly political discourse in inscriptions. 46 Ibid., p. 299. 47 At the end of a very articulate and convincing plea for the importance and radical innovation of writing for literature, Pollock comes to some rather startling conclusions: ‘Only authors of written work are included in the canons included in ethnohistorical accounts of literature; the oral poets stands entirely outside of history’ [Kabir, Surdas, Mira Bai?]; ‘...such oral culture is not only unknowable in its historicity, it is excluded from the literary history made by committing texts to writing ... It is no redundancy to say that a literary work does not exist until it becomes literate’; The Language of the Gods, pp. 317, 318. Elsewhere he acknowledges that ‘the ongoing interaction between

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

238 / FRANCESCA ORSINI

This model of the emergence of vernacular literary culture—with the making of vernacular polities and the literarisation of the vernacular through emulation of the superimposed model of Sanskrit by (mostly Brahmin) poets at royal courts who knew Sanskrit very well and shared the conceptual framework of Sanskrit poetics but wanted to innovate—does not work well for north India for a number of reasons. First and more obvious is the substantial presence of the dominant High Language of literary and political discourse, Persian, which continued to spread over north India through the Sultanate administration, madrasa education, and the culture of the Sufis—and as we have seen had significant spoken currency as well in a simplified form. Second, while Sanskrit-educated Brahmin poets did produce the kind of superimposed vernacular literary culture described by Pollock later on—notably with Keshavdas at the Bundela principality of Orchha—they were by no means the only or principal agents of literature at this time.48 Most of the authors of poems and narratives in fifteenth-century north India were either Muslims, kayasthas, Jain panditas, low-caste or of unknown or mixed background—even if later tradition may have strived to ascribe hidden Brahmin pedigrees to them. It is then unsurprising that the literary genres they preferred and the literary models they followed were less Sanskritic than in Kannada and other similar regional literary cultures. As we have seen, fifteenth-century Hindavi literature consists mainly of songs, doha couplets and kathas, narratives. Some were indeed produced at regional or even smaller courts, but others in the open ‘Bhakti public sphere’ (Agrawal) of towns and villages. Songs (and singer-composers, vaggeyakaras) were highly prized and at the centre of courtly performances, as well as of devotional practices and temples. Both kathas, songs and dohas were genres practiced by a range of different poets—Naths, Sants, Sufis, Jains, bhakha and sometimes also Persian court poets—and the high degree of intertextuality in terms of titles, names, tropes and images shows that they circulated among all these domains, evidence of a general intelligibility of genres and aesthetics.49 Thus, instead of Pollock’s model of court-centred vernacularisation sketched above, it seems better to understand literary culture in fifteenth-century north India as a multilingual and multi-locational literary culture—with a trend towards the oral and the literate constitutes one of the most remarkable and unique features of Indian literary culture. If oral compositions could be literised, literised compositions could also return to oral circulation, and the interplay between oral and literate composition and transcription could become dizzyingly complex’; ibid., p. 316. 48 This is the kind of literary culture analysed by Allison Busch in The Poetry of Kings; but it seems unwarranted to say that it was ‘singularly influential form of culture that occupied the entire conceptual domain of aestheticised language use’; Pollock, The Language of the Gods, p. 322. 49 The late Aditya Behl argued convincingly the redeployment of key concepts of Sanskrit aesthetics such as rasa, bhava, and rupa by Avadhi Sufis; see his series of lectures ‘Love’s Subtle Magic: An Indian Islamic Literary Tradition, 1379–1545’, delivered at SOAS, London, November 2008.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 239

Persian–Hindavi bilinguality in the domains of politics and literature of the various regional Sultans and in Sufi religious and literary practice, of Apabhramsha– Hindavi/bhakha bilinguality in Jain circles, of vernacular literary production with significant gestures towards Sanskrit in [some] Rajput polities, and with the emergence of strong vernacular voices in the ‘public sphere of Bhakti’. This more diverse picture mirrors the balance of social forces that were active and vocal in the polities of the regional Sultans and local Rajput chiefdoms and in the ‘religious marketplace’ (Sheikh) of the time: rulers and chieftains, merchants and artisans, religious leaders and groups of various kinds. One of the challenges in linking these different locations is that literary histories have come down to us with their own geographies. Spread on the north Indian terrain, these look like a series of overlapping maps based on language and content that bear little or no relation to each other. Thus there is the Hindi map of early Rajasthani (rasau) and Avadhi poems, followed by the Krishna Bhakti poetry of the Braj area and the Ram Bhakti poetry of Ayodhya and Banaras, and the riti poetry of Rajput courts. There is the Urdu map with early, phantom traces in Lahore and Delhi with Mas‘ud Sa‘d Salman and Amir Khusrau followed by two centuries of silence, the flourishing of Dakkani Sufi and courtly poetry, and Vali’s journey north in 1700, after which the map of Urdu poetry centres on Delhi and Lucknow with a host of ‘satellite’ cities and qasbas. The Persian map is further divided between darbar and khanqah, courtly and Sufi, with Persian courtly poetry making a mark with Amir Khusrau and Mir Hasan Sijzi, then ‘disappearing’ in the long fifteenth century, and bursting again on the scene with the large immigration of Iranian poets at the courts of Akbar, Jahangir and Shahjahan and the local talents of Faizi, Chandrabhan Brahman, etc.—all taking place at the imperial court. The Persian Sufi map is more dispersed and ‘regional’, but how exactly it relates to the Persian courtly one is only intuitively understood. While such maps maybe useful within the context of a silsilah or sampraday, or to understand the circulation and transmission of a specific taste (for example, Persian poetry) or the affiliation of a poet to a particular tradition, they seem inadequate when it comes to understand the workings of poets within this multilingual literary culture. Even if tazkiras and bhaktamals and poetic vamshavalis are themselves selective, and even if authors in a particular language ignored those in another language in their own town or area, surely this ignorance or lack of interest is to be studied as a function of the genre or of the field rather than reproduced. Moreover, these maps leave too many gaps, too many areas and periods where nothing seems to be happening. A geographical sensibility within a comparative approach can go a long way in highlighting and then filling these gaps. Often works that signal the oral communication and circulation of a narrative or a song genre from one language or one tradition or one performance style into The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

240 / FRANCESCA ORSINI

another are the most suggestive in this respect. Thus Nalini Delvoye’s work tracing the trajectory of dhrupad from the Tomar court in Gwalior to the courts of Gujarat, Rewa, and the imperial court at Agra and the sub-imperial one of Jodhpur, and the links and affiliations between Tansen, Swami Haridas and the Shattari Sufi Muhammad Ghaus Gwaliyari, has done much to show precisely how these different locations—both courtly and religious—were connected, and how singers, songs and musical aesthetics circulated between Sufi sama’ sessions, courtly mahfils and sabhas, temple functions and devotional gatherings.50 Other times it is works that are not central to the self-definition of a tradition, works that appear off-site, that force us to look around for a possible context until we can place them into a new literary geography. Thus for example Jayasi tells us that he composed his version of the Krishna tale (Harikatha), the alreadymentioned Kanh€vata (1540), in Jais after he ‘heard Ahirs singing and dancing at Divali, clapping the beat.’51 This forces us to look around for traces of Krishna bhakti and performance in Awadh, which is generally left out altogether of Krishna bhakti maps, usually centred on Braj, Puri and Bengal.52 Then we start finding references to singers and tellers of the Harikatha, such as Parmanand, originally from Kannauj but settled in Prayag, who attracted a regular audience and disciples: ‘a group of listeners always gathered around Param€nanda Sv€m… at about eight o’clock at night,’ and on special occasions he would sing all night long.53 It was also in Kannauj that the Sufi shaikh Abd al-Wahid Bilgrami (1510–1608), from the nearby qasba of Bilgram, spent part of his life later in the century. He must have heard a lot of Krishna songs (bishnupad), for he expounded on the Sufi mystical interpretation of the terms found in dhrupad and bishnupad songs in a Persian text he called The Truths of India (Haq€’iq-i Hind…, 1566). This included a systematic treatment of terms related to the story of Krishna, and from the tenor of his explanation it is clear that while the songs appealed to him aesthetically (and emotionally), he was not interested in the theology of Krishna bhakti.54 In the case of Krishna songs (bishnupad and kirtan), then, we can imagine them circulating in Awadh through groups of singers, sometimes taught and managed 50 See, for example, Delvoye, ‘Indo-Persian Accounts on Music Patronage in the Sultanate of Gujarat’, and ‘Collections of lyrics in Hindustani music: the case of Dhrupad’, but more generally all her research; also Allyn Miner’s ‘Ragas and raginis’. 51 Gupta, Kanh€vata, p. 140. 52 For a fuller treatment of this issue, see my ‘Krishna Bhakti and Sufis in Awadh’. 53 Richard Barz, ‘The V€rt€ of Param€nandad€sa’, p. 144. 54 Muzaffar Alam has considered the text within a paradigm of ‘assimilation from a distance’ which the Sufis of Awadh worked with, both out of their belief in the (inclusive) doctrine of waƒdat al-wuj™d, but also out of political necessity in the face of continued threats and attacks from local Rajput zamindars. In Haq€’iq-e Hind…, according to Alam (‘Assimilation from a distance’, p. 174), Bilgrami ‘sought to reconcile Vaisnav symbols as well as the terms and ideas used in Hindu devotional songs with orthodox Muslim beliefs’ within the ‘syncretistic religious milieu’ of Awadh qasbas.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 241

by wandering bairagis, resident swamis and even Sufi Shaikhs, and performing in urban centres, at fairs, at Sufi sama’ gatherings, as well as for private seva and worship. As for the Harikatha, the life of Krishna based on the tenth book (Dasam Skandha) of the Bh€gavata Pur€Ša, we find that the first (incomplete) Hindavi vernacular version, in chaupai-doha, was in fact composed in Awadh. This is the Haricarit by Lalac Kavi, a kayastha (or possibly a halvai) from ‘Hastigram’ (present day Hathgaon) near Raebareilly, concluded in 1587Vi/1530.55 This is only 10 years before Jayasi heard the story (and ‘read it’), and only 40 miles away from Jais. A geographical sensibility allows us to see these texts in Persian, Awadhi and Braj Bhasha as somehow connected. While in sectarian Krishna-bhakti accounts Awadh is just a stop-over for Vallabhacharya, a site where he initiated a few more devotees and singer-poets into the sampraday, we may read the accounts against the grain and note evidence of independent activity and circulation of Krishna songs and the Harikatha. Lalac Kavi’s Haricarit deserves only a brief mention in Hindi literary history because it is not a sophisticated literary achievement: yet both in its location (Kannauj, urban), the background of the author—kayastha, halwai? In any case not a patronised court-poet—and the wide geographical spread of its manuscript copies it signals its significance as a vehicle of circulation of the Harikatha. Jayasi’s elaborate and sophisticated Sufi tale tells us of another agency, that of the Ahirs as seasonal performers of some kind of raslila, and of a reworking that seeks to place itself in the open arena. Finally, the almost contemporaneous Haq€’iq-i Hind… signals the popularity of bishnupad (Krishna-bhakti songs) in the region, certainly not among Sufis alone. As a result, thanks to this multiligual map, Awadh appears traversed with Krishna bhakti performers and performances, a veritable palimpsest of religious entertainment that needs to be excavated carefully to appreciate the temporal and spatial connections of its constituent parts. If we take Jaisi’s Kanh€vata again as an example, the basic outline and features of the Krishna story remained the same but there was ample space for rearrangement, selection and addition. The historical palimpsest of north Indian literary (and religious) culture in this period meant that songs or tales (Vaishnava, Sant and other) circulated among different interpretive communities, and the same

Partly translated into French by Pavie, Krichna et sa doctrine (p. 3 for the date); according to R.S. McGregor (Hindi Literature, p. 96n), it was well known in the 18c. Lalacdas is also credited ~ with a ViŠupur€Ša (Vi. 1585/1528), see Suras, Avadh… KŠna k€vya, p. 40n, quoting ‘Avadh… me KŠa k€vya ke praŠet€: Kavi L€lacd€s’, Hind… Anu…lan’, XIV, 3, p. 18. According to Udaya Shanker Dubey, scores of manuscript copies of the Haricarit can be found in eastern UP and Bihar (all in Kaithi script) and as far afield as Malwa and Gujarat; personal communication, Allahabad, August 2009. 55

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

242 / FRANCESCA ORSINI

images and symbols became by necessity multivocal, that is, there was no simple interpretation that was valid for all, and participants knew this and accepted it as such. Conclusions The alternative to selective single-language literary histories, then, is not a narrative of ‘composite culture’, where equally selective syncretic traditions are taken as definitive evidence that culture (selectively: music, Sufism, Sant Bhakti) acted as a great cohesive force in the Indo-Muslim polity. Both the single-language and the ‘composite culture’ narratives exclude large swathes of literary production, arbitrarily set language boundaries, construct chronologies that do not match, and answer questions of language and literary choice spuriously along an unproblematic continuum of script–language–religious identity and community. What this essay suggests, instead, is that in addition to the usual literary–historical narratives based on genre, patronage and the rise of new trends and fashions, a multilingual literary culture also requires specific tools as well. This article has taken fifteenth-century north India as its case study, but the tools it proposes could be fruitfully applied to other regions and periods of literary history. First, an awareness of the material and of historically located practices of writing, where literature is but one domain of language use, to be relativised and compared to, for example, the domains of administration, trade and the army. This helps us reflect on the important phenomenon of biscriptualism and its consequences for the circulation of literary texts, the formation of archives and, in the final analysis, for literary history. Second, local taxonomies and written archives are precious in the affiliations and fault-lines they set up, but they need to be set against wider categories of language, literature and socio-textual communities, where oral performance and the circulation of genres and performers and a kind of regular multilingualism gave shape to a multilingual and multi-local literary culture. Thus, we noted that even scant bilingual traces within texts help question and relativise otherwise strict protocols such as those of Persian texts. Or, as Imre Bangha and Allison Busch have done, we can read register as a subtle form of bilingualism, a tool through which poets modulated their responses to other (poetic) repertoires in the multilingual world they lived in. These traces lead us to the all-important aspect of orality: oral performance, oral circulation, oral enjoyment. In turn, these help us explain the ubiquity, intelligibility and flexibility of songs, the striking intertextuality of certain texts (like Sufi Hindavi romances), the circulation of tales, titles and motifs, and the development of a broad-based aesthetic sensibility that can be summed up in terms like bhava and that was embodied in the person of what Katherine Schofield has called the ‘Mughal rasika’.56 56

Schofield, ‘The Mughal Rasika.’

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 243

Finally, within this multilingual and comparative outlook, attentive to the multiple locations of literary cultural and the oral aspects of performance, circulation and enjoyment, a geographical sensibility together with acute historicity can bring to light unexpected links. Thus, rather than a model of literary culture centred around either religious sites or around royal courts, we can see north Indian literary culture as constituted precisely by the interrelated efforts of singers, poets, patrons and audiences at courtly darbars and sabhas, in the open spaces of chaupals in towns and villages, in temples and khanqahs. To study these voices and texts, and to study them in connection with each other and within a wider comparative framework, means attempting to write a different, at the same time thicker and more comprehensive, history than that usually available in textbooks. Textbooks generally start from ‘hard evidence’ and documents such as coins, inscriptions, and historical chronicles typically compiled in centres of political power, whereas literature and the arts are added as supplementary ornaments to the hard core of power, usually under the rubrics of ‘patronage’ and, in the case of vernacular devotional literature, of an undefined ‘popular culture’. This is an exaggeration, of course, but not too far off the mark. Instead, especially for a richly fluid century like the fifteenth century in north India, literary texts are often the only way we have to write social history, to write individuals and groups, their self-representation and worldview into the picture, which is otherwise a hopelessly empty and schematic one of court and people, rulers and dynasties, Muslims and Hindus, of course men and hardly any women at all. Multilingual literary history, as we have seen, requires a perspective open to elements and agents not immediately present in the texts, an awareness that each text and author exists in a context that is more complex and varied than the one he gives us to believe. Most of all, in a region where vernacularisation seems to have been generally much slower and more partial than, say, in Karnataka or the Deccan, we need to remember that even texts in High languages were written by people who were still part of the vernacular world. References Agraval, Vasudevsharan. Padm€vat: Malik Muhammad J€yas… kt mah€k€vya, Chirgaon, 1988. Alam, Muzaffar. ‘Assimilation from a Distance: Confrontation and Sufi Accommodation in Awadh Society’, in R. Champakalakshmi and S. Gopal, eds, Tradition, Dissent and Ideology: Essays in Honour of Romila Thapar, Delhi, 1996. Anon. ‘Sikandar Lodhi aur uske ‘ahad ke ba‘z farsi musannifin’, Oriental College Magazine, Vol. 32, 1932. Azad, Muhammad Husain šb-e ƒay€t, Lucknow, 1880, 1993 reprint. Badauni, Abd al-Qadir. Muntakhabu-t-tawarikh, translated from the original Persian and edited by W.H. Lowe, vol. 2, Calcutta, 1898–1925. Baevskii, Solomon I. Early Persian Lexicography: Farhangs of the Eleventh to the Fifteenth Centuries, translated, revised and updated by John R. Perry, Folkestone, 2007 [1989].

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

244 / FRANCESCA ORSINI Bahura, Gopal Narayan. Pada S™rad€saj… k€, facsimile edition, with an essay by K. Bryant, Jaipur, 1984. Bangha, Imre. ‘Rekhta: Poetry in Mixed Language’, in F. Orsini, ed., Before the Divide: Hindi and Urdu Literary Culture, New Delhi, 2010, pp. 21–83. Barz, Richard. The Bhakti Sect of Vallabhacharya, Faridabad, 1976. Bhima Kavi. angvai Kath€, Shivgopal Misra, ed., Allahabad, 1966. Bilgrami, Mir Abd al-Wahid. Haq€’iq-i Hind…, translated into Hindi by S.A.A. Rizvi as Hak€yake Hind…, Kashi, 1957. Bilgrami, Ghulam ‘Ali Azad. Ma‘€thir al-Kir€m, Lahore, 1971. Brac de la Perrière, Eloïse. L’Art du livre dans l’Inde des sultanats, Paris, 2008. Busch, Allison. ‘Hidden in Plain View: Brajbhasha Poets at the Mughal Court’, Modern Asian Studies, Vol. 44(2), 2010, pp. 267–309. ———. Poetry of Kings: The Classical Hindi Literature of Mughal India, New York, 2011. Callewaert, Winand M. and Lath, Mukund. Hindi songs of Namdev, Leuven, 1989. Chatterjee, Kumkum. ‘Cultural Flows and Cosmopolitanism in Mughal India: the Bishnupur Kingdom’, Indian Economic and Social History Review, Vol. 46(2), 2009, pp. 147–82. De Bruijn, Thomas. The Ruby in the Dust: The Poetics of Muƒammad J€yas…’s Padm€vat, Leiden, 2011. De Clerq, Eva. ‘Apabhramsha as a literary medium in fifteenth century North India’, in F. Orsini and S. Sheikh, eds, After Timur Left, forthcoming. Delvoye, Françoise. ‘Nalini’. ‘Indo-Persian Accounts on Music Patronage in the Sultanate of Gujarat’, in M. Alam, F.N. Delvoye and M. Gaborieau, eds, The Making of Indo-Persian Culture: Indian and French Studies, New Delhi, 2000, pp. 253–80. ———. ‘Collections of lyrics in Hindustani music: The case of Dhrupad’, in J. Bor, F.N. Delvoye, J. Harvey and E. Te Nijenhuis, eds, Hindustani Music: Thirteenth to Twentieth Centuries, New Delhi, 2010, pp. 141–58. Digby, Simon. ‘Two Captains of the Jawnpur Sultanate’, in J. Gommans and O. Prakash, eds, Circumambulations in South Asian History: Essays in Honour of Dirk H.A. Kolff, Leiden, 2003, pp. 159–78. Entwstle, Alan W. Braj: Centre of Krishna pilgrimage, Groningen, 1987. Gallego, Maria Angeles. ‘The languages of Medieval Iberia and Their Religious Dimension’, Medieval Encounters, Vol. 9(1), 2003, pp. 107–39. Giovanardi, Claudio. ‘Il bilinguismo italiano-latino del medioevo e del Rinascimento’, in Storia della lingua italiana, vol. 2 ‘Lo Scritto e il Parlato’, Torino, 1994, pp. 435–67. Goswamy, B.N. A Jainesque Sultanate Shahnama and the Context of Pre-Mughal Painting in India, Zürich, 1988. Granoff, Phyllis. ‘Mountains of Eternity: Raidh™ and the Colossal Jinas of Gwalior’, Rivista di Studi Sudasiatici, Vol. 1, 2006, pp. 31–50. Guha, Sumit. ‘M€rg…, De… and Y€van…: High Language and Ethnic Speech in Maharashtra’, in M. Naito, I. Shima and H. Kotani, eds, Mar¯ga: Ways of Liberation, Empowerment, and Social Change in Maharashtra, New Delhi, 2008, pp. 129–46. Gupta, Mataprasad. J€yas…-granth€val…: Padam€vata, Akhar€va˜a, škhir… kal€ma, aura Mahar… B€…s…, Allahabad, 1952. Gupta, Parmeshvarilal, ed. Kanh€vata, by Malika Muhammda Jayasi, Banaras, 1981. Horstmann, Monika. In Favour of Govinddevji: Historical Documents Relating to a Deity of Vrindaban and Eastern Rajasthan, in collaboration with Heike Bill, New Delhi: Indira Gandhi National Centre for the Arts in association with Manohar, 1999. Khandalavala, Karl and Chandra, Moti. New Documents of Indian Painting: A Reappraisal, Bombay, 1969. King, Christopher. ‘The Nagari Pracharini Sabha of Benares, 1893–1914: A Study in the Social and Political History of the Hindi Language’, unpublished Ph.D. dissertation, University of Wisconsin, 1974.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

How to do multilingual literary history? / 245 Kolff, D.H.A. Naukar, Rajput and Sepoy: the Ethnohistory of the Military Labour Market in Hindustan, 1450–1850, Cambridge, 1990. McGregor, R.S. Hindi Literature: From the Beginnings to the Nineteenth Century, Wiesbaden, 1984. Miner, Allyn. ‘Ragas and Raginis, Sufis and Sants: Music in north India in the Early Sixteenth Century’, in F. Orsini and K. Schofield, eds, Tellings and Texts: Singing, Story-Telling and Performance in South Asia, forthcoming. Momin, Mohiuddin. The Chancellery and Persian Epistolography Under the Mughals, from Bábur to Sháh Jahán, 1526–1658: A Study on Inshá, Dár al-Inshá, and Munsh…s Based on Original Documents, Calcutta, 1971. Mukherjee, Tarapada and Habib, Irfan. ‘Land Rights in the Reign of Akbar: The Evidence of the Sale-deeds of Vrindaban and Aritha’, Proceedings: Indian History Congress (1989–90), Vol. 50, 1990, pp. 236–55. Novetzke, Christian L. Religion and Public Memory: A Cultural History of Saint Namdev in India, New York, 2008. O’Hanlon, Rosalind. ‘Speaking from Siva’s Temple: Banaras Scholar Households and the Brahman “Ecumene” of Mughal India’, in R. O’Hanlon and D. Washbrook, eds, Religious Cultures in Early Modern India: New Perspectives, London, 2011, first published as Special Issue of South Asian History and Culture, Vol. 2( 2), April 2011, pp. 253–77. Orsini, Francesca. ‘Krishna Bhakti and Sufis in Awadh’, in V. Dalmia and M. Faruqui. eds, Religious Interactions in Mughal India, New Delhi, forthcoming. Orsini, Francesca and Stefano Pellò. ‘Bhakti in Persian’, unpublished paper, presented at the panel on ‘Rethinking Bhakti in North India’, 21st EASMAS conference, Bonn, 2010. Orsini, Francesca and Samira Sheikh (eds.). After Timur Left: Cultural Production and Circulation in the Long Fifteenth Century, New Delhi, forthcoming. Pauwels, Heidi. ‘The Saint, the Warlord, and the Emperor: Discourses of Braj Bhakti and Bundel€ Loyalty’, Journal of the Economic & Social History of the Orient, Vol. 52(2), 2009, pp. 187–228. Pavie, Theodore. Krichna et sa Doctrine, Paris, 1852. Phukan, Shantanu. ‘Through a Persian Prism: Hindi and Padmavat in the Mughal Imagination’, unpublished Ph.D., University of Chicago, 2000. ———. ‘“Through Throats Where Many Rivers Meet”: The Ecology of Hindi in the Persian Imagination’, Indian Economic Social History Review, Vol. 38(1), 2001, pp. 33–58. Plukker, D.F. The Mirig€vat… of Kutubana: Avadh… text with critical notes. Amsterdam, published Ph.D., 1981. Pollock, Sheldon. The Language of the Gods in the World of Men, Berkeley, 2006. Rizvi, S.A.A. Alakh B€n…, Aligarh, 1971. Saha, Shandip. ‘Between Temple and Court: The Pushti Marga in Mewar’, paper presented at the 20th European Conference on Modern South Asian Studies (ECMSAS), Manchester, UK, 8–11 July 2008. Schofield, Katherine. ‘The Mughal Rasika’, paper presented at the conference ‘Tellings, Not Texts: Singing, Story-Telling and Performance’, School of Oriental and African Studies, 8–9 June 2009. Shackle, C.S. ‘Approaches to the Persian loans in the šdi Granth’, Bulletin of SOAS, Vol. 41(1), 1978, pp. 73–96. Shafi, M. Mohammad. ‘Three Old Documents’, Proceedings of the Idara-i Maarif-i Islamia, Lahore, 1936, pp. 281–85. ———. ‘Urdu fiqre aur dohre €˜hv…n aur nav…n ad… hijr… k… f€rs… tanif€t se’, Oriental College Magazine (August), 1930, reprinted in Maq€l€t-e ž€fiz Maƒm™d Sher€n…, Lahore, 1966. Sherani, Hafiz Mahmud. ‘Makhd™m Shaikh Bah€’ud-d…n Barn€w…’, Oriental College Magazine, 1927, part 1 (August), pp. 41–58, part 2 (November), pp. 9–26. Siddiqui, Iqtidar Husain. Waqi‘at-e-Mushtaqui of Shaikh Rizq Ullah Mushtaqui, New Delhi, 1993.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012

246 / FRANCESCA ORSINI Sinha, Surendranath. Subah of Allahabad under the Great Mughals (1580–1707), New Delhi, 1974. ‘Suras’, Murarilal Sharma. Avadh… KŠa k€vya aur unke kavi, Agra, 1967. Stewart, Tony. The Final Word: the Caitanya Carit€mrta and the Grammar of Religious Tradition, ° New York, 2010. Vaudeville, Charlotte. Kabir, Oxford, 1974. Vidypati, Thakkur. K…rtilat€, B. Saksena, ed., Banaras, 1967. Weightman, S.C.R. ‘The text of Alakh B€n…’, in R.S. McGregor, ed., Devotional Literature in South Asia. Current Research, 1985-1988, Cambridge, 1992, pp. 171–78.

The Indian Economic and Social History Review, 49, 2 (2012): 225–46

Downloaded from ier.sagepub.com at SOAS London on June 28, 2012