Introduction

Francesca Orsini and Katherine Butler Schofield

Khwāndan: To read, to recite [...] to sing.1

What would an auditory history of South Asia sound like? As we walk down a city street or a neighbourhood lane in contemporary India, loudspeakers hanging from electricity poles spread the tune of a choral bhajan or amateur singing at a ritual wake, while few neighbourhoods are out of aural reach of an azan, the call to prayer. Public spaces are routinely occupied by religious processions with drums, marriage processions with band baja, or political demonstrations with loud slogans and public speeches. Every town has public spaces—a maidan, lila ground, karbala, or park—where religious performances and “programmes”, fairs (melas), and political rallies regularly attract visitors and broadcast their activities through their lively noises. Amidst the cacophony of traffic sounds—extra-loud car horns, shrill cycle-rickshaw bells, the deeper grumble of buses and trucks—people’s mobile ringtones advertise their musical taste: Punjabi beats, melodious ghazals, or the latest Bombay dance number. Several times a day, when your own mobile rings and you pick up, a jingle or a verse addresses you for no apparent reason. This soundscape is not static and unchanging. On the Delhi metro, the bilingual warnings “metro paridhan ko ganda karna ek dandaniy apradh hai” and “mind the gap” have replaced the scratchy audio cassettes on buses as accompaniments on one’s daily journey through the city, while fewer autorickshaws seem to ride with their radio on full-blast.

South Asia’s visual culture has been the object of much study in recent years, from calendar art to photography, from truck art to political statuary. Yet it would be hard to deny that making sound and hearing or listening to music, songs, speeches, sermons, and stories have been equally constitutive of South Asian social and cultural history until the present day.2 But how has the mosaic of sounds, voices, and tellings changed over time? More fundamentally, how can we even write the history of sound at all, given that its nature is ephemeral: over in a moment, gone forever, and never fully captured in words on a page?

This volume explores the interconnected histories of singing, storytelling, and oral performance in early modern and contemporary North India (and Pakistan), in an attempt to restore the auditory realm to the literary and cultural history of South Asia. It does not aim at comprehensive coverage—there is no essay that deals with the rich performance traditions of Punjab, for example3—but presents strategically identified case studies that show different uses of texts in performance, give an idea of the wide range of performance practices, and highlight the significant circulation of aesthetic concepts and ideas about the beneficial effects of music, singing, and storytelling.

In the past two decades, an interest in what has been labelled acoustic or auditory history—the history not just of “music” but of historical soundworlds in their broadest possible sense—has begun to emerge in the study of Western music. This new move has received considerable stimulus from parallel work in ethnomusicology on contemporary soundscapes,4 research that has been foundational in the new interdiscipline of sound studies or “acoustemology”, which takes in fields as diverse as geography, anthropology, environmental science, and music.5

While most scholarship on acoustic history and auditory cultures has been restricted to the period of recorded sound (the late nineteenth through the twenty-first centuries), Emma Dillon’s magisterial 2012 monograph The Sense of Sound definitively turns musicology’s attention to the European medieval past and to the difficult task of disinterring auditory history, and the history of music as it was sounded and embodied, from its textual traces.6 Although such a task is made easier by the existence of detailed musical notation for European music that allows some sense of sounded reality to echo in the present, Dillon’s methodology offers much of relevance to our quest to understand the acoustic and somatic lives of North Indian texts. Historians, too, have recently turned to the auditory, with Mark Smith’s pioneering volume Hearing History opening our ears to the sounds of early modern Europe and America.7 Typically, though, the pre-twentieth-century acoustic histories of other literate cultures so far remain largely untold. The recent works of Gary Tomlinson and Barbara Andaya, on the soundscapes of the pre-colonial Aztec and Malay worlds respectively, thus provide stimulating foils for us, with their insistence that sound was heard and understood differently in earlier cultural contexts and that sonic power and song were of vital significance and signification to Southeast Asian and American polities.8

In this volume, therefore, we undertake a task that has not, to our knowledge, been attempted before for the literary cultures of North India.9 We seek to theorise the deep interdependencies of written text, sound, performer, audience, and meaning that the case studies in this volume make audible, in a situation where in most cases we no longer possess nor can ever recover the soundedness of the texts with which we work (see Schofield in this volume). By and large, texts that were once recited, sung, danced, and enacted have been territory ceded to scholars of literature and religion, who have generally acknowledged their aural and performative dimensions but gone little further.10 The aural domain, in turn, has been relegated to ethnomusicologists studying the highly exclusive soundworld of North Indian art music.11 And the very few ethnomusicologists who have brought tools of sonic analysis to bear on the textual archive have restricted their studies to instrumental genres and raga examples, whose mnemonic notation systems have been easier to translate, if only partially, into sound.12 Perhaps concerned that their literary understanding might be found wanting, or perhaps because of the sheer inaudibility of most historical texts, music historians have steered clear of explaining how poems subtitled with raga and tala names might have sounded, or what other sonic hints might tell us about the lives of North Indian texts as they circulated as sounded and embodied entities in a resonant world.

In any case, the overwhelming majority of ethnomusicologists study the present, not the past: music, singing, and storytelling in North India have primarily been the domain of anthropologists of music and scholars of orality, who rarely view the living traditions they study as part of a cultural and literary field that can be mapped historically. This is the connection that this book aims to make: the contemporary ethnographies by Amy Bard, Christian Novetzke, Richard Widdess, and Richard Wolf show what uses—sometimes startling and counter-intuitive ones—words, texts, and books are put to in contemporary formal and informal performance practices, and the subtle nuances of tone, emphasis, and effect that each performance brings. They show the textual scholars what they are missing and suggest exciting possibilities. Conversely, the essays by textual scholars and music historians reveal that even texts, when examined in this light, turn out to provide a surprising amount of clues about what Stefano Pellò calls the pre-textual and con-textual life of poetry, sermons, and stories: the anecdotes, gossip, and discussions that accompanied and explained how certain texts came into being (Cort, Pellò, Sharma); the correct knowledge that was required in order to understand and appreciate sophisticated oral performances (Hawley, Khan, d’Hubert); the performance needed to bring the texts we have to life (Horstmann, Novetzke, Busch, Bangha, Orsini). Through a number of recurring key words like raga (musical mode), rasa (juice, sentiment), rasika (connoisseur), and bhava (emotion), these essays also show how aesthetic and/or spiritual cultivation and understanding were crucial to the listening of music and tales in the early modern period. Such words were part of a code that straddled the multilingual poetic, musical, and visual arts—though declined with subtle differences in the courtly, sufi, and bhakti (devotion) contexts where songs and tales circulated (see particularly Miner, d’Hubert, Schofield, and Busch).13

Oral-literate, Multilingual, and Intermedial

General accounts of orality in South Asia typically begin with the oral-mnemonic tradition of Vedic and post-Vedic knowledge (the Upanishads, Panini’s grammar, etc.). But as Sheldon Pollock has pointed out, this much touted orality, while undoubtedly and bedazzingly true, has too often been taken as emblematic of a general Indian “indifference” to writing. The cultural premium on memorised knowledge (kanthastha or “held in the throat”, as Pollock reminds us) “left indelible traces in secular written culture”. And “from the moment writing was invented literary culture, the culture of kavya, became indissolubly connected to writing, so much so that the history of the one becomes unintelligible without taking into account the history of the other”.14 Similarly, Richard Widdess has noted that although there has been perhaps an even stronger emphasis placed on the superiority of oral-aural modes of transmission in Indian musical discourse, “many systems of ‘oral notation’ exist, and have existed since ancient times. These systems use solmization or other mnemonic syllables, and are primarily recited or sung, although they can also be written down”—and have been used to notate musical examples in written treatises since the Gupta period.15 Like Pollock, Stuart Blackburn notes of South India that even the early Sangam corpus of Tamil poetry (third century CE, but edited and anthologised only in the eighth) valorises both orality and writing: “Many of the poems are presented as if spoken or sung by bards, while, on the other hand, many give prominence to the role of the poet-scholar (pulavar)”.16 Both Blackburn and Pollock note the endurance of practices of orality in South Asia “as both fact and ideal” well into the modern period, and their persistence into the present day is particularly obvious in their continuing predominance in music pedagogy.17 Christian Novetzke notes that pre-modern sants and performers in Maharashtra lived in a milieu where literacy was a fairly ordinary and widespread skill. Yet the public culture of bhakti and the logic of performance meant that, though literate, kirtankars would still privilege orality. In fact, as he puts it, the kirtankar “might be considered an intermediary between text and orality” (p. 180).

Velcheru Narayana Rao coined the very useful term “oral-literate” to describe pandits, poets, and storytellers who operate within a culture that is both orally transmitted and literate at the same time—and this is a term that applies in different ways to almost all the people, cases, and genres presented in this volume: from preachers to kirtankars, from poets to musicians, from musical treatises to song-poems, from poetry manuals to tazkiras that record the gossip around poets, from early modern tales to the contemporary niyaz kahani pamphlets of Amy Bard’s essay. Perhaps the most surprising group that qualify as oral-literate are musicians; the consensus modern view that Hindustani ustads were illiterate is belied by a series of treatises and song collections written by hereditary musicians and their disciples from Mughal times down to the present day, demonstrating literacy in Sanskrit, Persian, Brajbhasha, and Urdu.18 Indeed most of the essays prefer to use the term performative rather than oral, in order to stress that sight, gesture, and sound were all involved.19 About qissa storytelling, for example, Pasha M. Khan notes that the seventeenth-century manual

Ṭirāz al-akhbār makes it clear to us that the term “qissa-khwan” does not convey the full range and force of the storyteller’s activities. Impressive as it seems that storytellers like Fakhr al-Zamani recited and improvised the interminable Dāstān-i Amīr Ḥamza from memory, they did not simply read them, but performed them. In his description of the presentation of the qissa, Fakhr al-Zamani prescribes not only modulations of the voice, but gestures and postures for the storyteller. (Khan, p. 198)

The same is true of many of the performance practices covered in this book: singing, poetic recitation, and storytelling. Conversely, the mere physical presence of a book during a performance—even if it is not consulted—may work symbolically as an authorising gesture, as Widdess shows in his essay.

Methodologically, however, although scholars of literature and religion concerned with the past are aware of the all-important oral dimension of performance, and ethnomusicologists and scholars of orality recognise that the living traditions they study are part of cultural and literary fields that have much longer histories, by and large it is difficult for one person to have the technical training to do both. The documents we have from the past—written texts, manuscripts, visual images, and written descriptions—often bear only scant traces of their oral-performative contexts, or else describe them in terms that are minimal or opaque, as in the musical notation of ragas or attempts to describe aesthetic experience. Conversely, current performance traditions that have been orally transmitted, especially the further we move from institutional centres, often bear only oblique traces of their history.20 We will come back to the relationship between texts/books and orality below.

Trying to reconstruct the oral-performative history of early modern North India, as this book tries to do, presents additional challenges. While certain aspects have been well studied—bhakti sayings, songs, and performances; the circulation of songs and singers in devotional circles and across North Indian courts (especially the Mughal imperial court); sufi romances in Hindavi21—others remain unclear. Texts of the time often contain lively religious discussions, goshtis, and repartees, but what relationship do those iconic representations bear to real events and/or practices? Numerous musical treatises contain notated raga examples, but what do these actually tell us about how the ragas sounded in performance? Many texts that suggest concurrent oral-performative practices such as Puranas or sufi malfuzat, or that were explicitly offered to patrons, were written or copied in the high languages of Persian and Sanskrit—does this mean they reflect speech practices, or rather protocols of writing? Much path-breaking work is currently being done on Sanskrit-Persian interactions in the Mughal period,22 and questions like “did the Mughals (or sufis) really know Sanskrit?”, or “did they really speak in Persian?”, regularly arise. While the answers necessarily vary according to the educational capital and background of patrons and audiences, we must recognise that the multiple diglossia of the time means that texts written in the high languages existed in an oral context that was vernacular and multilingual.

To state it more clearly, we begin from the premise that the linguistic economy of North India can be described as one of “multiple diglossias”,23 with several high languages—Arabic, Persian, Sanskrit—and a general spoken vernacular (what we call here Hindavi) that was variously written in the Persian, Kaithi, or Devanagari scripts. We intentionally use the term Hindavi, which was the term (together with Hindi and Hindui) employed for North Indian vernaculars in Persian sources, in order to avoid the split history of Hindi and Urdu that has dominated modern scholarship and language consciousness.24 A concurrent premise is that language and script were a function of written transmission and the competence of patrons and copyists: script was not intrinsic to a language, and the script and language of writing did not necessarily reproduce the language of oral performance or exchange. The protocols of high language meant that discussions in Hindavi between a sufi pir and his disciples would be written down in Persian, or that the Hindavi song-poems composed by Persian literati were referred to but not included in Persian-language histories and anthologies.25

Once we are aware of these premises, we begin to see that texts that appear to exist in separate domains sealed by boundaries of script and literacy could and did circulate thanks to oral transmission, translation, exposition, and memorisation. We also see that at times the texts are themselves translations of oral vernacular tellings/performances. Thus Jack Hawley’s essay on Surdas’s reworking of a passage from the Sanskrit Bhāgavata-purāṇa shows us that the Brajbhasha poet-singer Surdas, traditionally memorialised as being blind, knew the Sanskrit canonical text well enough to riff on it and could expect his audience to understand his game.26 His quasi-contemporary, the poet Alam, declared that “since few listen to this tale in Sanskrit, I have bound [this tale] together in chaupais in bhakha”.27 Does he mean that he knew the Sanskrit versions of the tale? The case of ‘Abd al-Rahman Chishti’s Mir’āt al-makhlūqāt, analysed in Muzaffar Alam’s essay, is even more intriguing: the Persian text purports to be a translation of a Sanskrit Purana and its author declares and cites evidence of his knowledge of Sanskrit and of the text. The aim is to render public something that the keepers of the original text—the Brahmins—have tried to keep secret, i.e. that Shiva himself predicted the advent of Adam and the Prophet Muhammad and the takeover of India by their descendants, who intermarried with the local population. The text faithfully reproduces in Persian the generic conventions of the Purana—the chain of narrators, the explanation of past and future events, etc.—but what is the relationship of this text to orality? Were ‘Abd al-Rahman or his readers familiar with the Sanskrit text and/or oral expositions of the Bhaviṣya-purāṇa; did he sermonise on it; and if so in which language?

Several essays in this volume seek to tease out the oral-performative dimension of written texts and genres, particularly those in the high languages or whose accessibility is uncertain. Taken together, the contributions of Busch, Schofield, Sharma, and Khan show that texts that have come down to us as part of separate and sophisticated traditions—Brajbhasha riti poetry, Hindustani music theory and song lyrics, and Persian poetry and storytelling—were all consumed by the same people at the Mughal court. Thus, while seeking to understand the logic internal to the formation and transmission of each archive (courtly and madrasa Persian, Jain, courtly bhakha, bhakti, sufi), it has been crucial for us to also question their limits and exclusions, and to place them within the larger framework of orature and oral transmission.28 An approach sensitive to oral circulation and performance highlights the dynamics through which these connections took place. The result is a map of a richer and more densely interconnected cultural and social world.

Of course there is a way in which music in particular has long been viewed as constitutive of premodern India’s “composite culture”, the aesthetic glue that held the otherwise fractious/centripetal polity together. While parallel religious nationalisms have positioned Hindu (and Jain) and Muslim communities as inevitably hostile and barely reconciled, the discourse of “composite culture” has upheld music and painting as evidence that Hindus and Muslims had been friends and had cultivated similar tastes.29 In this discourse, music or bhakti and sufi religions are held up selectively as special cases of “synthesis”, representing a “bridge” that connected what are still perceived as separate communities, each with their own traditions. In the process, music, bhakti, and sufism get extracted from their social histories and charged with a mysterious agent-less intentionality that obscures the messy and much larger-scale social processes of conscious mixing and intentional borrowing by thoughtful and knowledgeable men and women that historically must have occurred to produce any kind of “composite”.30 In contradistinction, while our evidence shows much circulation and translation of music and song genres, singers and performers, stories, and even aesthetic categories, we see these as normal products of a culturally diverse and multilingual polity—a regular multilingualism—with multilayered, distinct, yet interlocking contexts: courtly, urban, ritual/devotional, rural. The evidence also leads us away from the idea of the “composite” to thinking in terms of individuals actively appropriating across cultural and linguistic thresholds and between media (from poetry to musical sound to painting, etc.) to produce a widely shared early modern aesthetic of borrowing and reuse that revelled in virtuosity, brilliance, and multilayered depth and richness.31

Spaces of Performance and Performers

A major advantage of a multilingual and intermedial approach to orality and performance traditions is that it allows us to explore literary culture beyond the court, to understand the links between forms and performers outside and within the court, and to examine the dynamics of classicisation and popularisation. It also allows us to attend to the oral-performative aspects of poetic culture and wit, so obviously valued as cultural assets (see Pellò in this volume), and to consider the performers who enacted/produced these verbal forms, their social position, their self-presentation, and their own mobility.

Ever since Françoise “Nalini” Delvoye’s pioneering studies of dhrupad texts in Persian sources, of the circulation of songs and of song-poets (vaggeyakar) from the court of Gwalior to that of the Sultan of Gujarat and thence to Akbar’s court and sub-imperial centres, and of the relationships between Tansen, Swami Haridas in Vrindaban, and Muhammad Ghaus Gwaliori, we have been alerted to the intense circulation of songs, musicians, and musical knowledge between courtly and devotional/ritual domains.32 The striking flexibility in song themes and “retooling” of song texts as well as poems, so that a ruler’s name could be substituted by another, or by the name of Krishna, were a direct consequence of this circulation, as Busch reminds us in her essay.

The essays in this volume cover a wide range of performance spaces and domains. Sharma, Busch, Khan, and Schofield explore the culture of poetic, musical, and storytelling performances at the Mughal court from Akbar’s reign to Muhammad Shah’s. Sharma’s essay, for instance, details the kinds of Persian poetic and prose texts that were recited and discussed at the Mughal court, and notes that “in the Mughal context storytelling, poetic recitation, and discussion also functioned as a form of re-enacting and validating the canon in the face of new literary developments and challenges, especially when it came to poetry” (p. 288). He notes that Emperor Akbar preferred literary gatherings that involved storytelling to poetry recitations (musha‘iras), whereas his son and successor Jahangir was fond of listening to and discussing Persian poetry, particularly the ghazal, during long night gatherings. He also observes that Emperor Shah Jahan was particularly interested in literary works concerned with contemporary history, while Aurangzeb ‘Alamgir was especially fond of Jalaluddin Rumi’s Mas̱navī. Sharma points out that during Shah Jahan’s period (r.1628-1656), many Persian poets at court wrote short topical poems in masnavi form to be recited at court, suggesting that “these poems gradually replaced the ceremonial qasida as the traditional poetic form to mark formal occasions” (p. 292). As far as the performance of the ghazal is concerned, anecdotes show a certain overlap between courtly and sufi practices by the sixteenth century, and khayal, now known only as a genre of classical music, enjoyed a dual existence in both the dargah and the darbar from its sixteenth-century appearance until at least the time of Muhammad Shah (r.1719-1748).33

Allyn Miner’s essay shows how much of that musical knowledge was already cultivated at pre-Mughal courts such as the Sharqi court in Jaunpur, and how the musical knowledge codified there and in other apparently peripheral locations such as Rewa34 circulated among three different kinds of constituencies: one made of connoisseurs and music specialists who wanted to learn the specific qualities of each raga; another constituted by those who only wanted the basic vocabulary and took pleasure in the imaginative aspects of raga visualisation; and a third one made of practitioners and religious devotees who employed a more restricted range of ragas for ritual singing.

Several essays tackle performance texts and traditions at regional courts (Miner, Busch, d’Hubert, Orsini). Orsini’s traces the emergence of kathas or tales for local courts such as the Baghela of “Amarpur” in the wake of epic and Puranic retellings. Allison Busch’s pathbreaking work has shown how riti poets like Keshavdas, working in the small court in Orchha, created a new literary culture in the early sixteenth century by carefully studying Sanskrit models and reproducing them in the vernacular.35 She has also shown how, in the wake of political alliances between local rulers like the Bundelas of Orchha and Mughal princes, this literary culture spread into the heart of the Mughal imperial court and found ready patronage not only in the imperial entourage, but also among its ministers (like Todar Mal, whom Alam also mentioned in admiring terms in his 1582 Mādhavānal Kāmakandalā) and Rajput mansabdars, and who in their home territories developed their own sub-imperial courtly cultures and employed their own array of poets, genealogists, and storytellers.36 In her essay for this volume Busch focuses on the oral and performative dimensions of this literary culture in the form of memorising verses and rules as a necessary preparation for extempore poetic performances, of the retooling of verses by itinerant poets for successive patrons, and on the functions of poetry at these courts, including the performance of martial poetry on the battlefield itself, with the expressed aim of enthusing the warriors. Nor should we forget that these local rulers were also major patrons of ritual and devotional performances and sponsored a whole range of temples, monasteries (maths), and festivals.37 A particularly interesting case is that of the sophisticated seventeenth-century poet Alaol, the subject of Thibaut d’Hubert’s contribution, whose Bengali narrative poems/romances were informed by Persian, Sanskrit, and Hindavi poetics and literary models. The case of Alaol points to a kind of cosmopolitanism that was directly produced by the confluence of multilingual literary traditions, translated into the local literary language for a small but heterogeneous court in Arakan (now Myanmar). Widdess’s essay also touches on the circulation of North Indian musical and literary culture at the Malla court in Nepal. Not only did sangitashastra texts circulate there, but it is also here that some of the earliest manuscripts are to be found. He further notes that some of the Malla kings’ dapha compositions are present in contemporary dapha songbooks—a tantalising “tenuous textual continuity can thus be demonstrated between the early seventeenth-century palace context and twenty-first-century farmers’ music” (p. 234)—while the wonderful detail in his Figure 8.3 shows how non-courtly genres and performers were visualised in a hierarchical spatial fashion.

Several other essays focus on urban spaces and activities among a range of merchant, service, professional, and artisanal groups. Stefano Pellò shows how tazkiras of poets written in Persian from the eighteenth and nineteenth centuries focus on the activities and the professionalisation of contemporary poets and their disciples outside the court, in the urban spaces of Delhi and Lucknow, and in doing so eulogise and canonise them. But these texts also display keen interest in humbler and more marginal figures in the urban landscape such as madmen, opium-addicts, jesters, and street performers. John Cort’s essay focuses on the parallel world of Jain laymen in Mughal Agra, who met regularly in a temple courtyard and constituted themselves into debating circles that produced newly-authoritative doctrinal and liturgical texts and virtually emptied the figure of the bhattarak of authority.38 This was a process of vernacularisation of knowledge that happened independently of royal initiatives or patronage, the fruits of which are still part of Jain ritual practices today. Novetzke’s essay on kirtan performances and Widdess’s on dapha both straddle premodern texts and contemporary performances and include urban ritual performances among artisanal/subaltern groups, while Richard Wolf’s takes us to the diasporic urban streets of Karachi for rhythmically sophisticated perfomances of Nizami drumming.

With the essays by Horstmann, Novetzke, Hawley, Bangha, and Alam we are taken to spaces that range from small-town (qasba) sufi establishments and villages, maths, and festivals (mahotsav) where Dadupanthi acharyas would deliver their sermons.39 The great circulation of Surdas’s songs and Bajid’s arilla verses, and of manuscripts with Bajid’s short, humorous tales, points to a popular realm of religious, entertaining, and instructive performance that would definitely include Novetzke’s kirtans, which nowadays usually comprise “a story from the life of a sant that goes along with the song, and usually other songs or texts are brought in that can range from Sanskritic philosophy to sufi mysticism to the wisdom of political leaders and popular adages and sayings of unknown provenance in any language, including English” (p. 171). That Bajid’s manuscripts are also held in royal libraries like the Pothikhana in Jaipur alerts us to the fact that, just as there is a sophisticated oral knowledge that does not depend on literacy (Hawley, Wolf), so elite tastes could and did include popular genres. Conversely, the career of the early-seventeenth-century Iranian émigré storyteller Fakhr al-Zamani shows the remarkable mobility from urban to courtly spaces of both performer and genre. “If Fakhr al-Zamani’s progress is any indication, qissas that began at the ‘popular’ level could, given a chance and perhaps with some stylistic alterations, eventually be performed in the courts of nobles and preserved as manuscripts in their libraries”, notes Khan (p. 192). And while his gestural style of storytelling recalls the figure of the naqqal, a professional actor who conveyed “a story with words and actions, attempting to embody the narrative and its characters”, Khan reminds us that naqqals were lower on the scale of professions than qissa-khwan,40 and Fakhr al-Zamani never used the word naqqali for what he did (Khan, p. 198).

Contemporary Ethnography and History

Because of the challenges outlined above, we considered it to be vital to bring ethnomusicologists and scholars of contemporary performance practices into conversation with the more historically-oriented among us. We do not wish to suggest naively that contemporary performances reflect past practices. But what their study reveals holds great imaginative power for historical work. For example, contemporary ethnographies show us a consistent combination of recitation of oral/written text with extempore exposition (arthav or, in raga performance, vistar) right across the genres and contexts we are looking at.41 For the earliest tales, the text is indeed all we have; but were any of these tales—particularly those rich in ritual, technical, or esoteric meanings—accompanied by exposition? It seems likely, particularly in cases where the length of the stanzas and the narrative “density” in extant copies vary.42 There are, for instance, obvious markers of ritual beginning in most of our texts—but how much richer is Philip Lutgendorf’s description of kathavachaks (called Vyasas in this instance) taking their seat after worshipping the seat and garlanding the book, etc., in katha performances of the Rāmcaritmānas!43 Christian Novetzke makes the useful distinction between “didactic kirtan” (which includes a wide range of modes, as we shall see) and “ecstatic kirtan” on the basis of the protocols of sitting and standing. In the ecstatic “Varkari kirtan” everyone stands and dances, whereas in the other forms of kirtan only the performer (kirtankar) stands while the audience sits, though the audience still participates in many ways, “singing along with the songs, finishing well known verses along with the kirtankar, sometimes interacting with the kirtankar, and so on” (p. 172).

In many cases, all we have for past musical performances are musical treatises or manuscripts with song texts. Mukund Lath and Winand Callewaert have argued that the form in which the song-texts are collected and ordered in a manuscript can tell us whether it was a singer’s own workmanlike collection or a systematisation, e.g. for ritual purposes, though Miner in this volume suggests that the very presence of a raga in bhakti texts indicates, at the very least, “that the original compiler or composers moved in or were connected with court or temple circles” (p. 399). Christian Novetzke has made a crucial distinction between formal pothis and informal badas (more on which below). In the case of the book used in dapha performances, we could muse over the meaning and interpretation of the song texts, were Richard Widdess not to tell us that in dapha performance sound and key words and effort are much more important than the text.

We can imagine Bajid’s short and entertaining tales retold in intimate familial contexts, and similar qissas were printed in chapbook forms not dissimilar from the niyaz kahanis that Amy Bard writes about. But while Bajid’s mock subversion of the moral ending puzzles us, Bard gives us a rich account of the various ways in which tellers and listeners interpret the kahanis and relate them to their own life experiences. Both Bajid’s irreverent stories and Amy Bard’s contemporary formal and informal miracle tales show an informal relationship between written texts and orality. One intriguing notion she puts forward concerns the different quality of listening in formal and informal niyaz kahanis—reading them is perceived as hardly efficacious. Although both types of mo‘jizat emphasise listening, formal kahanis efface narrators, are less “personal“, less interactive, and more listening-oriented. In casual mo‘jizat, which speak to local, familial needs with clear geographical anchoring, listeners are likely to “talk back“. (Though why Osho was so keen on Bajid and what use he made of his verses and tales remains an intriguing question!)

Books and Performance

The conference from which this volume draws was originally entitled “Tellings, Not Texts”, but one participant pointed out that “Tellings and Texts” was a more appropriate title, since texts—in their material form as books—were often present in the performances we talked about. What roles do texts play in performance, we asked, and what is the relationship between them? Which way does the directionality go—from text to performance or from performance/oral exposition to text? Here, too, early modern North India shows a great range of possibilities and choices.

At one end of the spectrum, we see a great deal of interest in books as material objects. Already at North Indian Sultanate courts illustrated manuscripts (and illuminated Qur’ans) were valued and copied in Persianate and Indic styles and provide a tantalising glimpse of the circulation of shared tastes among elites that impacted, for example, upon Jain book-copying and book-dedicating practices.44 It is surely not by chance that the earliest illustrated manuscripts of the Sanskrit Bhāgavata-purāṇa also appeared in this period. The dazzling and profuse production of the imperial Mughal workshop (karkhana) has tended to absorb most scholarly attention, and we still await a comprehensive picture of illustrated book production in this period that devotes parallel and equal consideration to non-imperial manuscripts and book circulation.

Among the religious groups of the period, too, we find a striking investment in books. The most obvious example is that of the pothis (compilations) of the early Sikh gurus. Guru Nanak himself, who “believed that he had been assigned by God the vocation of singing his praises […] and that his hymns were the result of direct communication from God”, nonetheless urged writing God’s name as a devotional act. This has been taken as an implicit hint towards the fact that already in his lifetime disciples wrote down his hymns in a pothi.45 As G.S. Mann points out, Guru Nanak referred explicitly to the role of the Qur’an in Muslim devotional life and must have been familiar with the practice common in sufi khanqahs of placing the Qur’an in the open to allow for full access.46 Having a pothi became crucial to the authority claims of his descendants and disciples at their various seats (gaddis). Textual history shows that the early pothis were subsequently added to, though much emphasis has traditionally been placed within the Sikh tradition on the singularity and unbroken continuity of the Guru Granth—whose status is very much that of a sacred text, to be read, recited, and sung to raga but also worshipped in private and public rituals.

The Sikhs were not the only sant group who invested in writing, compiling, and copying books. Dadu Dayal’s disciples also compiled his utterances (vani) together with those of other sants (panch-vani), as well as enormous and literally comprehensive compilations (lit. sarvangi). They also wrote the first biographies/hagiographies (parchais and bhaktamals) in the vernacular in North India, and the manuals for sermons that Monika Horstmann writes about.47 She notes that, especially from the eighteenth century, “the wealth of manuscripts often of great length and calligraphic quality indicate that the patrons of these were men—and occasionally women—of considerable means” (p. 45). These developments form a striking parallel to the already existing but growing production in sufi circles of compilations of sayings (malfuzat) and biographical dictionaries (tazkiras) devoted to one’s master—a sure way of placing him on the map. Another related phenomenon concerns the considerable growth in the eighteenth and early nineteenth centuries of writing about and by elite lineages of musicians, especially kalawants and those with lineal connections to Delhi. Far from confirming the “illiterate ustad” stereotype we have inherited from nationalist musicologists,48 the later Mughal period saw major professional musicians and socially prominent amateurs writing their own names into history in the form of two kinds of books: tazkiras and song collections49 in Persian and Hindavi.50 This oral-literate field of musical knowledge transmission probably developed in part as a way of ensuring the longevity of lineages and the preservation of lineal musical property threatened in the eighteenth century by increased economic migration away from traditional centres of familial oral transmission. The evidence of lineages of hereditary performers who, alongside oral transmission, simultaneously contributed to the written lineage of Indo-Persianate musical knowledge furthermore underlines the central importance of community and silsila, or what Indrani Chatterjee calls “monastic governmentality”, in sustaining written as well as oral knowledge systems in early modern India.51

In this period we thus see a marked expansion of textuality—not just in Persian and Sanskrit, but also in the vernaculars—to new groups and new genres, often as the writing down of oral genres, if not whole performances. As Bangha points out, until the sixteenth century only

Puranic, epic and historical narratives and Sufi romances—composed normally in the doha-chaupai metre, conveying important religious or political messages, and usually of a performative nature—[…] had been deemed worthy of being committed to writing in the vernacular. Towards the end of the sixteeenth century, instead, books began to appear in Brajbhasha that were composed in order to be read and studied and not primarily to be performed (see Busch in this volume). This is also the time when we can spot the beginning of an ever-increasing activity to commit to writing Hindi songs that have so far been transmitted in oral performance. (p. 359)

About Bengali mangalkabyas, d’Hubert notes that the poet is typically represented but as the conduit of the Goddess’s wish to tell the story. The ritual act by which he grabs a flywhisk marks not only the beginning of the performance but also the moment in which he remembers the text: “in terms of representation of the literary activity, it is not only that performance is the main way to share the content of a written text; rather it means that no text is ever able to come to existence without a setting of ritual performance” (p. 428).

Courtly literary culture, Busch reminds us, was very much a written literary culture—she quotes a description of the court of Bir Singh Deo Bundela by Keshavdas in which “there sat countless writers writing, hundreds, and thousands of them” (p. 254). While her essay painstakingly teases out the performative elements and qualities of the poetic, historical, martial texts, and manuals written by Brajbhasha riti poets, she is also keen to stress that this “riti corpus was underwritten first and foremost by a textual engagement with the Sanskrit past”. Sunil Sharma points out that while history records that Akbar had Persian classics like the Shāhnāma and the Gulistān read out in court and these works were meant to be read aloud in sections, “their orality was accompanied by an equal value placed on these works as books, and it was usually through the copying and use of manuscripts of these texts that they were transmitted with the seals of the members of the royal family and nobility” (p. 288).

Inside and outside the court, the Puranas (which both Muzaffar Alam and Jack Hawley touch upon) present a tantalising case. While Puranas remained written in Sanskrit in North India and the Bhāgavata-purāṇa attracted important Sanskrit commentaries by the main Vaishnava acharyas, Hawley reminds us that the Bhāgavata-purāṇa was performed in a number of ways both in Sanskrit and in the vernacular: through recitation; reading and exposition involving sermons and songs; or in rituals (yajnas). Indeed, the Bhāgavata māhātmya manuscripts that began to appear at the turn of the eighteenth century expressed the need to regulate these performances. In their different ways, both ‘Abd al-Rahman Chishti’s Persian Purana and Surdas’s “song commentary” are evidence of this culture of Puranas circulating between writing and performance.52

By comparison, John Cort’s essay explores the process by which Jain hymns in Sanskrit (stotras) were skilfully translated by Banarsidas and his circle in Agra in the early seventeenth century. (As he points out, they chose to translate not Jain scriptures, but ritual texts.) The written translation was itself the result of oral debates, and in turn it became the subject of further debates. Yet “no one wanted to read a kirtan or an abhang, after all—they wanted to see it, hear it, and experience it displayed before them”, as Novetzke forcefully argues. Perhaps the most unexpected arguments about the directionality between performance and text and the role of texts in performance come from his essay and Widdess’s. By contrasting the bada (notebook) of kirtan performers—“loosely organised and often hastily constructed with lots of margin corrections, lines crossed out, and other emendations”—with the orderly pothi (book) of theologians and institutional figures, Novetzke argues that badas represent the logic of kirtan performance. His argument is worth quoting in full:

[The bada] is a format that privileges performance. Text here is submitted to the demands of performance. The text is a tool, a means and not an end. Furthermore, its role is not preservation—it may function as an archive, but its composition is not intended to ossify a text and convey it into the future as a fixed form. Instead, the text is dynamic, meant to trigger and prompt a performance, existing as a kind of outline of a kirtan, but containing no narrative, typological, or historical logic independent of performance. In other words, it cannot stand alone, as can a pothi or a fixed, complete composition. (p. 179)53

As the archetypal kirtankar, Namdev expresses suspicion of books and scriptures and expresses a theory of practice in which devotion “is on display, through the body and voice, and never on paper”. “Despite the rather impressive rate of literacy among the Marathi sants”, Novetzke argues, “the logic of practice of Varkari religious expression has been oral, or rather performative, and not written”. Kirtan is the performance of devotion and is central to bhakti practice because it is public and open to all.

Monika Horstmann likewise makes the case that the thematic and musical arrangement of sant vanis, and the fact that all these texts are used as manuals by preachers, means that “there can be no doubt that the Dādūvāṇī originated from the redaction of sayings by, or used by, Dadu in live homiletic contexts” (p. 59). During Dadupanthi festive occasions today, preachers do not consult a prepared script but “may hold the open sacred scripture in front of them”, though their sermons will draw upon material taken from a range of books. And if they are preaching about a verse in the scriptures they might read the verses out. “The live performance of a sermon is therefore oral in its fullest sense”, she notes, “although it draws on memorised or printed material and perhaps on notes the preacher has taken while preparing for the task” (p. 36). The manuals that she and Khan write about can be viewed in a similar light—not as free-standing texts but as aids to performance. Indeed, she points out that notwithstanding the respect accorded to Mahant Pokhardas’s spiritual stature, his sermons were acknowledged to be hard to enjoy since he read out the manual!

While Fakhr al-Zamani’s seventeenth-century Persian manual on storytelling, the Ṭirāz al-akhbār, resembles the Dadupanthi sermon manuals in containing useful lists of terms, lexical aids, assorted sayings, and nutshell versions of longer stories, it seems closer to the ritigranth manuals (Busch) in operating as a proof, almost a certificate, of its author’s own skills as a storyteller, and in elevating storytelling to an art that requires a manual. Khan notes that the text not only tells us much about the process of qissa performance but also shows us “one manner in which the qissa was defined or ‘encoded’” (Khan, p. 186). In the case of the qissa, the manual’s approach to genre as an assemblage of elements (Khan uses the term bayaz or scrapbook) that could and should be combined and recombined in new and attractive ways “undermines the very idea of monolithic genres”, because if we look at the range of intertextual and extra-literary material that each qissa-khwan was supposed to have “in his throat”, the range is stupendous and in fact crosses the line between mimetic and veracious genres that Fakhr al-Zamani otherwise subscribed to (p. 186). Each qissa performance was therefore something of an “intertextual tapestry” that “can only be comprehended as a complex of multi-generic intertexts that fall under the order of the master genre code of the qissa” (p. 205).

By contrast, in the Newar devotional singing (dapha) studied by Richard Widdess, the book containing the song-texts is important—but only as a ritual object. It is never consulted, and the boys who are taught the songs from it are not taught to read them. For singers and audience the texts only provide clues to meaning: “a word, name or phrase evokes a network of meanings and associations, which may in turn determine when and where the song is sung”. What is important, he notes, is the performance, “which has its own meanings and values independently of the text: as religious exertion, as an expression of social identity, and as a component in urban ritual” (Widdess, p. 244). When he asked one group why they go to the trouble of singing the entire Sanskrit poem Gītagovinda in one all-night performance when they do not understand the words, “they explained that the objective is to sing all night. Therefore they choose to sing the Gītagovinda, because it takes that length of time” (p. 240).

Beyond Text:
Musical Sound and the Inexpressible

This, finally, leads us to the question of the limits of the verbal—what does music, what do singing or vocally heightened forms of recitation, add or do to a text that supersedes what a text can do alone? In considering this issue in particular, the insights of our contemporary ethnographic case studies have proven especially helpful.

Rhythmic patterns are sometimes inserted verbally in performance scenes within tales or poems (Orsini, Busch), and the notated non-lexical syllables of courtly taranas that were vocalised and embodied by drummers and dancers in performance are interlaced with Persian couplets in eighteenth-century song collections.54 But in Richard Wolf’s essay, drum syllables become a code unto themselves, speaking of the performers’ mastery and enjoyment of skills recognised by each other and by those few in the audience—like Madhavanal with Kamakandala (Orsini)—who can understand their abstract relationships with verbal texts. And as Widdess concludes, cases like dapha, khayal, or the marai kirtan in Bengal suggest that “some South Asian musical forms escape such conventional categories as orality and text, but are better understood as performance: a process in which text may be present, in written and/or oral form, but is subsumed by musical elaboration and the enactment of religious and social meanings” (p. 245). The presence of cosmic sound within the body is key for bhakti poets like Kabir, and for their contemporary performers like Prahlad Singh Tipania, who evoke the continuous inner sound, the subtle voice (nada, jhini avaz).55

For the period before recorded sound, the question of what music does that goes beyond text is more difficult to answer because we no longer have access to the object under discussion. Since Plato,56 music in many cultures, past and present, has been used to deal with excess; to go beyond words, to express the inexpressible; to soothe, magnify, and charm the emotions through the senses rather than the intellect. Because music’s mode of communication is abstract, music’s semantic and affective associations are more ambiguous than language, capable of expressing a greater depth and number of nuances, often simultaneously.57 This is not to say that music’s significations are endless: music always exists within a culturally specific net of meaning that determines its limits. What it does mean is that the very choice to recite or to sing, and in many of our cases the choice of raga in which to set a text, can both enrich and multiply the possible meanings of text—or conversely render a lyric less ambiguous by intimating one meaning among many.

What is more, the raga itself exists both as a specifically musical formula (a set of notes that need to be sung in particular orders and ways), nadamaya, and as an icon, devatamaya, that conveys a distinct mood and set of associations accompanied by specific instructions as to the correct time or season to stimulate those associations.58 In the Mughal court, a clear “scientific” correlation was drawn between each of the seven notes of the Hindustani scale, the seven celestial bodies, and the four elements presiding over each, embedding raga firmly in the fields of Islamicate cosmology and Unani medicine.59 But this was in large part an attempt to interpret in new terms why the Indic system associated each raga with particular iconographies, deities and nayak-nayikas, timings, and rasa. At the heart of both knowledge systems was an attempt to account for each raga’s effect on the listener, its ta‘sir. This effect was not merely subjective, it was specific; and it was not merely affective, able to transport listeners into a state of meditation or restlessness or tearfulness, it was supernatural—it rearranged the natural world, whether that be a heart sickened by melancholy or a gathering rain cloud refusing to burst. Although the rational, “scientific” underpinnings of the explanation of ta‘sir had attenuated by the nineteenth century, the Ma‘dan al-mūsīqī (1869) nonetheless describes in magical terms how the singer’s command over ta’sir should work. The singer was in effect a necromancer (an ‘amil), and the melodic form of the raga was a magical formula (‘amal; du‘a ba tarkib) that, when voiced at the right cosmological time, would conjure up a spirit (a muwakkal) that would do his bidding. That spirit was the ta‘sir of the raga.60

In other words, sound in and of itself was powerful—effective and affecting—and the silencing of this dimension to the texts with which we work is definitively a loss. We are not ‘amils, and cannot conjure up the inexpressible experience of sounds that have long passed into silence. But all our work in this volume is underpinned by the conviction that we must be constantly mindful of the fact that texts in North India were (and are) frequently experienced as live auditory entities with a fuller range of experiential possibilities than may be apparent on the surface of the page.