2. Making it Vernacular in Agra:
The Practice of Translation by Seventeenth-Century Jains1

John E. Cort

Languages and Translation

In one sense, everyone “translates” all the time. Translation is simply the basic interpretive process of rendering external data into terms that “make sense” to a person. Even between two speakers of the same language, the process of trying to understand each other involves a mode of translation. Less ubiquitous, but still almost universal, is the process of translating between and among different languages. As long as there have been languages, people have had to cross the linguistic borders between them. Everyday functioning in polylingual social settings requires translation, the finding of rough and ready equivalencies and similarities of meaning across languages so that one can engage in both necessary and casual communication with others. For some people, this process is elevated into a more consciously pursued practice. A person works carefully and painstakingly to bring a text in one language—a legal contract, a government document, a bill of lading, a poem—into a second language. Issues of accuracy and precision become more important. Greater control over both languages, the source and the target, becomes a concern. Training and experience in translation may be desirable skills on the part of the practitioner.

In some settings this practice of translation becomes a performance. The tour guide, the pilgrim priest, the simultaneous translator at a diplomatic conference, the person signing a speech or performance for hearing-impaired members of the audience—each is performing translation. The concept of performance can also be extended to the playful and painful rendering of a powerful hymn from a classical language understood by few into a vernacular tongue, so that everyone who hears or reads it can appreciate and experience its beauty and influence. If the concept of “practice” in translation emphasises the need for discipline and application, the concept of “performance” reminds us that translation can be a virtuoso act aimed at receiving the appreciation of an audience.

I advance these general observations about translation as everyday experience, as practice, and as performance, in order to provide a setting for the specific discussion that follows, in which I examine the practice and performance of translation by a handful of Digambar Jains in seventeenth-century North India. In addition to being poets and translators, the men in question were businessmen, government servants, and professional intellectuals. They engaged in translation as part of their day-to-day life in the polylingual metropoles of Mughal India. Due to their occupations and their interests, they also engaged in the more explicit and formal practice of translation. Because the particular texts that they translated were intended to reach and appeal to a target audience of their fellow Jains, they furthermore engaged in the performance of translation. To gain a fuller understanding of seventeenth-century North Indian literary and intellectual culture, therefore, it behooves us to pay attention to the widespread practice and performance of translation in this culture, a practice and performance that have hitherto been largely ignored by scholarship.

It is an obvious truism to say that India has always been multilingual. In this, India has been no different than any other culture, as the monolingual culture is an invention of the modern nation-state, aided by print-capitalism.2 Languages and dialects (the difference between the two also largely an invention of nationalism) changed within short distances, so people inevitably needed to be able to shift language to speak to others whom they met in a regional market or on pilgrimage, when men left home for work, or in marriage negotiations. The languages of governance were often different from the local vernaculars. Religious languages—which we often term as “classical”—differed, often sharply, from the vernaculars. Literary languages also differed from the spoken vernaculars; the idea that literature should represent, if not be identical with, the spoken tongue is another creation of modernity (and literary high modernism). Commerce, military service, and religious mendicancy all generated their distinctive forms of linguistic communication.

As I said, there is nothing uniquely Indian about this. However, when one compares the cultural and intellectual attitudes to this multilingual condition in India to those found in the European and Mediterranean cultures that we loosely term as “Western”, there is a striking difference. From the times of classical antiquity, and the first productions of Christian scripture in Greek and Latin as opposed to the earlier Hebrew and Aramaic, philosophers, theologians, and writers have thought long and hard about what it means “to translate” from one language into another.3 One would think that a culture in which language itself (vac) has a sacred presence, and in which grammar rather than mathematics formed the foundation for logic, would have generated a tradition of inquiry into the meaning and possibility of translation. For most of Indian history, however, this has not been the case. This is evident in the Hindī sāhitya koś (Dictionary of Hindi Literature), edited by Dhirendra Varma (1963). In the first volume of this massive work, totaling 997 pages, which is a glossary of literary terms (paribhashik shabdavali), not a single entry is devoted to a discussion of translation per se.4 This has continued into English-language scholarship.To cite just two influential recent examples, in neither of the massive volumes by Sheldon Pollock on South Asian literary cultures, totalling some 1,600 pages, is translation a topic of analysis.5

In part this is a reflection of a broader cultural and academic devaluation of translation and translators, which has led to what Lawrence Venuti has called “the translator’s invisibility”.6 Publishers often either omit the name of a translator, or bury it in fine print on the copyright page. Academic tenure and promotion committees place little if any weight on translation as evidence of “real” scholarly activity. This neglect extends into scholarship, as it is common to find detailed studies of important authors in which scant attention is paid to those author’s translations. The major biography of Banarsidas, for example, devotes only about one page out of 332 to his two iconic translations.7 Translation is not considered to be “original”, and so it is often dismissed as unimportant, uncreative, and ultimately derivative and mechanical work.

There were some exceptions, of course, especially in the context of religious communities such as the Jains and Buddhists. They developed their own sacred languages, which acquired sacred status within the specific communities; but they then had to come to terms with the privileged status of Sanskrit in classical and medieval India as the pre-eminent language of all intellectual discourse. Many Buddhist texts such as the Dhammapada were translated from Pali into Sanskrit, and the famous Buddhist translation workshops in China indicate that the Buddhists had no ideological objection to translation. Further, the number of Sanskrit and Pali texts that were translated into Chinese multiple times indicates some level of concern for accuracy and faithfulness in translation.

We can see possible tensions over translation expressed in the story of the famous fourth- or fifth-century CE Shvetambar Jain monk Siddhasena Divakara.8 He was a convert, from being a Brahmin to being a Jain monk, and from Sanskrit to Prakrit, but was still enthralled by Sanskrit intellectual culture. He arrogantly offered to render the Jain scriptures, composed in Ardha-Magadhi Prakrit, into Sanskrit.9 The other Jains were angered at the proposal. They said that Siddhasena was implying that the enlightened and omniscient Jinas, and their enlightened successors, had been incapable of writing Sanskrit. Siddhasena, in their opinion, was therefore implicitly doubting that very omniscience as well. The monks imposed a penance on Siddhasena, who had to wander incognito for many years—in other words, they rendered him symbolically speechless. This story does not deny the possibility of translation, but rather criticises an attempt to elevate Sanskrit to a position equal to Ardha-Magadhi. Granoff has rightly argued that the main point of the story is to deny the Mimamsaka position on language, which held that only Sanskrit is suitable for religious and ritual texts.10 Since Siddhasena was a Brahmin by birth, the story also ties language usage to socio-religious hierarchies. This is not primarily a story about translation (or the impossibility thereof), but certainly indicates one socio-religious context within which concerns about translation might arise.

Three Modes of Translation

At the centre of this essay is the analysis of “translation” as a text that in its intention is not independently authored, but explicitly brings into the new language a text previously authored in another language. If there is a known author to the first text, the name of that author is attached to the new text. This is what A.K. Ramanujan, borrowing terminology from Charles Sanders Pierce, has called an iconic translation, in which “Text 1 and Text 2 have a geometrical resemblance to each other, as one triangle to another”.11 Of course, a fully iconic translation is impossible, as theorists of translation in many cultures have argued for millennia. A fully iconic translation would in fact be nothing other than the first text itself. This is the conundrum explored by Jorge Luis Borges in his famous short story “Pierre Menard, Author of Don Quixote”—a text which George Steiner has described as “arguably… the most acute, most concentrated commentary anyone has offered on the business of translation”.12 The goal of Borges’s fictional author was not to “translate” Cervantes’s famous novel, but “to produce pages which would coincide—word for word and line for line—with those of Miguel de Cervantes”.13 Despite its logical impossibility, however, the “faithful” iconic translation has been the ideal of translators in European languages and Abrahamic religious traditions for 2,000 years.

The other two types of translation in Ramanujan’s schema do not aim for iconic faithfulness to the original. The first of these is indexical translation, in which the new text shares the same basic elements of plot, but differs in many specifics. This is what in Western classical music might be called a “variation on a theme”. The other is symbolic translation, in which the new text “uses the plot and characters and names… minimally and uses them to say entirely new things”.14 This takes us into literary texts and genres that push the boundaries of any definition of translation, and are more often termed by scholars as “adaptations”, “versions”, and “renderings”, and even, as I discuss later in this essay, “commentaries”. If one expands the definition of translation to include these latter two modes, then Indian literary traditions have long been filled with translations. But the practice of faithful iconic translation would appear to have been rare.

Anuvad

The absence of any sustained thinking about translation is indicated in the languages of India themselves. The word used most commonly in contemporary North Indian vernacular languages for “translation” is anuvad (anuvād, anuvāda).15 A look at the Sanskrit original indicates that this usage is a modern one—created from the English to “translate”, one can say, the word “translation” into Hindi and other vernaculars, to describe a concept that wasn’t as fully present in Indian languages as in European ones. In Sanskrit the primary meaning of anuvāda is “to say after, to say again, to repeat”, as a means of explanation.16 Monier-Williams cites Yaska’s Nirukta for this meaning. In the Brāhmaṇas, anuvāda means “a passage… which explains or illustrates a rule (vidhi) previously expounded”. Monier-Williams also cites “translation” as a meaning for anuvāda, but gives no examples. I suspect this was a modern Sanskrit usage that found its way into his dictionary as a back-formation from vernacular usage. The term was also used in Sanskrit poetics: something that was anuvāda-ayukta was an example that did not in fact adequately explain a rule.17

The modernity of this meaning for anuvad is seen by looking at its usage in several modern vernaculars. R.S. McGregor gives “translation” as the second meaning for the Modern Standard Hindi anuvad, with the first meaning being, as in the Sanskrit, the repetition of something already said.18 He also indicates that “translation” is now the prominent sense of the word; that a formerly secondary meaning is now the primary one indicates a shift—probably relatively recent—in its usage.

Gujarati also indicates the recentness of the use of anuvad to mean “translation”. Its first meaning in the Br̥had gujarātī koś is “to say again what has been said” (boleluṃ pharī pharī bolavuṃ), and then the second meaning is “translation” (bhāṣāntar, tarajumo).19 The two Gujarati synonyms for “translation” further indicate the relative recentness of the importation of the concept into Indian languages. Bhāṣāntar, literally “other (antar) language (bhāṣā)”, is a tat-sama derivation from the Sanskrit that sounds very much like a literal neologism. While it is also found in Monier-Williams, he again gives no example of its usage in classical Sanskrit literature.20 Tarjumo (Hindi tarjuma) entered the North Indian vernaculars from Central Asia with the Persianate literary culture of the Mughal and related courts.

Brian Hatcher has recently subjected the Bengali anuvada to a preliminary historical investigation by looking at the usage on the title pages of published translations of Sanskrit texts into Bengali in the nineteenth century. He found that early in the century, Bengali authors referred to their publications as saṃgraha bhāṣāte (“compiled in the vernacular”) or bhāṣā vivaraṇa (“vernacular exposition”). While these same authors when writing in English referred to the act of “translation”, it was not until the middle of the 1830s that Bengali authors began to use the term anuvada, and from then it rose to increasing prominence.21

It is clear that the vocabulary for translation as a distinct, nameable practice is recent in South Asia. Does this mean, however, that translation as a practice is something new? Is it yet another feature of modern South Asian life for which we must attribute responsibility, for better or worse, to the British colonialists?22 As I have argued elsewhere, before we attribute changes in India to colonial influence, we need a clearer understanding of India “on the eve of colonialism”, to borrow a phrase from Sheldon Pollock.23 To attempt even to begin a history of translation in South Asia is beyond the reach of this essay, for the materials for such a history have not begun to be collected. I do think, however, that we can see one significant moment in a history of translation in South Asia emerge in Agra in the seventeenth century. Let me now turn to the details of this specifically Jain setting.

Religious and Literary Performance among Digambar Jains in Seventeenth-Century Agra

The city of Agra was one of the centres of both political and economic power during the Mughal era.24 Thousands of Jains migrated to the city in response to the opportunities available there. Many of them occupied positions at all levels of the Mughal administration. Others used the city as a base for trading networks. They made Agra into a rich Jain cultural centre as well. In 1594 Pandit Bhagavatidas made a pilgrimage to Agra from his home in Ramnagar, and described the metropolis in his Argalpur jinvandanā.25 He said that there were forty-eight Digambar Jain temples in the city.26

Two of those temples were the seats of bhattaraks, the landed and domesticated Digambar pontiffs at the apex of the Jain religious community.27 Bhattarak Shubhkirti resided at the Tihuna Sahu temple, and Bhattarak Jagatbhushan resided at the Sahu Narayani temple.28 Many of the temples in the city would have been affiliated with one or the other of the two bhattarak seats, and the bhattaraks assigned ritual specialists known as pandits or pandes to the temples under their control. These specialists, who might be celibate (brahmachari) or married, oversaw and conducted devotional and tantric rituals in the temples, and organised and supervised the many annual celebrations of the Jain ritual calendar. They also engaged in the production of knowledge: they wrote and copied texts in Sanskrit and vernacular, and delivered public sermons on a regular basis. Almost every Digambar temple in North India has a large courtyard—sometimes covered, sometimes open—for the performance of congregational rituals and for delivering sermons. Many temples also have covered verandas off of the courtyards, where men or women can gather. There they perform rituals involving the worship of icons (either portable metal icons, or stone icons on subsidiary altars placed in the walls of the verandas). They gather to sing devotional songs. They also engage in the study of texts, either by themselves or under the guidance of a more learned person.

Balbhadra Jain has written of Agra during this period, “In the temples there were scriptural sermons both morning and evening, and also philosophical seminars [tattva-goṣṭhī]”.29 We can gain a better sense of these cultural performances through the writings of Banarsidas (1587-c.1643). In his autobiography, the Ardhakathānak,30 he wrote of a seminar that was held in 1635 under the leadership of Pandit (or Pande) Rupchand.31 Banarsidas wrote, “Pandit Rupchand was well-known, and an expert on the scriptures. He came to Agra. He stayed at the temple built by Tihuna Sahu. All the spiritual seekers held a seminar, in which he lectured on a text, the Gommaṭasāra”.32

This is the same temple mentioned by Bhagavatidas forty years earlier as the seat of a bhattarak. Pandit Rupchand, therefore, must have been a disciple of the bhattarak who oversaw the temple and its congregation. Banarsidas referred to a circle of spiritual seekers (adhyatmis); we will return to them shortly. This group was engaged in religious study. The term Banarsidas used for this was bichara; and the modern standard Hindi vicār-goṣṭhī is a term for seminar or symposium.33 The group took advantage of the presence of this intellectual, who was well-known (guni) as a scholar of the scriptures (agama-jana), and requested that he conduct a course of study (banchayau; cf. Hindi vanchan, bachna) on the Gommaṭasāra. This is a tenth-century text, written by Nemichandra, which provides a detailed summary of Digambar doctrine.34 The seminar was a key event in Banarsidas’s spiritual evolution. He described the effect of hearing Rupchand’s exposition: “Pande Rupchand was an authoritative teacher; my mind became happy from hearing the text”.35

Banarsidas gives us one more description of the way these seminars worked. The text for which he is best known among Jains is his 1636 Samaysār nāṭak, a Brajbhasha indexical translation of one of the foundational texts of Digambar philosophy, the Samayasāra of Kundakunda.36 Banarsidas had first encountered Kundakunda’s text three years earlier during a visit to his wife’s natal town of Khairabad. This had been another crucial turning point in his spiritual life. There he met a man named Arathmal Dhor, who was an enthusiastic proponent of the two-truths doctrine expounded in Kundakunda’s text. In Banarsidas’s words, Arathmal “spoke enthusiastically” on the subject.37 Banarsidas went on to say that Arathmal gave him a manuscript of a vernacular commentary on the Samayasāra out of a concern for Banarsidas’s spiritual welfare (hita). Banarsidas read (banchai) and studied (bhasha aratha bocharai chitta) the text on his own, but was unable to come to an adequate understanding of it. Only after studying the Gommaṭasāra under the guidance of Rupchand was he able to return to the Samayasāra. In the Ardhakathānak he simply said, “Now I had attained right faith, and knew the true nature of God. In 1639, with zeal and joy, I composed the vernacular Samaysār nāṭak”.38

In the conclusion of the Samaysār nāṭak, Banarsidas gives us a longer account of its composition:

The pleasant Samayasāra was read by pandits together with its Sanskrit commentary, so they understood it. But the common people could not understand its meaning. The Jain Pande Rajmall loved the Samayasāra, and so he wrote a commentary on it known as the Teacher of Children [or Beginner’s Textbook: Bālbodh] which was easy to understand. Thus conversation about this wisdom spread. This became the doctrine of those who follow the Spiritual [adhyatma] style. The Jina’s teachings became known everywhere, and people talked about the Samayasāra in every household. It became famous in Agra, and people became knowledgeable about it. Five skillful men began to talk about knowledge day and night. First there was Pandit Rupchand, and the second was Chaturbhuj. The third man was Bhagotidas [Bhagvatidas]. Then there was the virtuous Kaunrpal. Together with Dharamdas they were five men, who met and sat together. They would discuss the supreme truth, and nothing else. Sometimes they discussed the Samayasāra, sometimes other texts. Sometimes they would continue to discuss wisdom even after they had stood up [to leave]. … This work continued for many months—how many I can’t say. It became known in Agra that a man named Banarsidas had a little knowledge on the subject. The Samayasāra is beneficial, and the commentary on it by Rajmall makes it easy to understand. If it were composed in metre, everyone could read this vernacular text. So the thought came to Banarsi to publicise these teachings of the Jina. He received permission from the five men to compose it in metre.39

This passage gives us valuable insight into the cultural performance space in which Banarsidas worked. While the professional intellectuals—the pandits and pandes—were literate in Sanskrit as part of their training and occupation, they were also expected to transmit the contents of the classical scriptures in the vernacular, through both vernacular commentaries and sermons. In the cases of the Samayasāra and the Gommaṭasāra, the linguistic gulf between the original and the vernacular was even deeper, for both of these texts were composed in Prakrit. It is not clear how fluent even pandits and pandes were in Prakrit, and I suspect that in many cases they relied extensively, and in some cases exclusively, upon the later Sanskrit commentaries.40 The vernacular prose Bālbodh commentary on the Samayasāra by Pande Rajmall (also Raymall), for example, was not based on the original 415-verse Prakrit text, but rather on the Samayasāra kalaśa, a 278-verse Sanskrit metrical commentary composed by Amritachandra in about the twelfth century.41

Banarsidas then described a circle of five men who gathered to study and discuss these texts. This was a trans-sectarian group of Jains who were interested in the new style of religiosity known as Spirituality (Adhyatma). One of these, Pandit Rupchand, was another trained intellectual. He was not the same as the Rupchand who had lectured on the Gommaṭasāra. The other four were laymen. Banarsidas gave a striking description of the intense nature of their discussions, when he said that even as they got up to leave the seminars they would continue their conversations. Finally, Banarsidas indicated that it was within the setting of this seminar that he composed his own Brajbhasha rendition of the Samayasāra, basing his text on Rajmall’s version of Amritachandra’s Sanskrit commentary. While Banarsidas did not say this explicitly, I think we can read between his lines to see that the final product was the result of his own study and poetic skills combined with the critical comments and assistance of his five companions.

Banarsidas’s Samaysār nāṭak in turn became a textbook for seminars in Agra. In his colophon to the 1644 Banārsī vilās, the posthumous collection of Banarsidas’s shorter works, the editor Jagjivan—whose family were divans in the Mughal court42—mentioned a “circle of scholars” (jñānīn kī maṇḍalī) that engaged in the study (vicāra) of Banarsidas’s text.43 Jagjivan included in the Banārsī vilās the Jñān bāvanī, a poem written by the poet Pitambar in 1630 in praise of Banarsidas. In it Pitambar describes an assembly (sabha) that sat together at the Digambar temple of Kapurchand Sahu under the leadership of Kaunrpal and studied the writings (vachan) of Banarsidas: “They sit happily in the Kapurchand Sahu temple. They sit in the assembly hall (sabhā) where they meet for intellectual considerations, and study the writings of Banarsidas”.44

The Jains in Agra did not just conduct seminars devoted to the details of Digambar Jain philosophy. There were also poetic gatherings. In India at this time the performance of poetry was a musical event. In his autobiography Banarsidas recounted a time earlier in his life, in 1610 or 1611, when as a young man his father had sent him from his birthplace of Jaunpur to Agra to earn a living in trade. Banarsidas was not successful as a merchant, but did gain some recognition for his poetic and singing skills. He wrote that ten or twenty men would come to his room to hear him sing Qutban’s Mirigāvatī and Manjhan’s Madhumālatī, two narrative mystical poems composed in 1503 and 1545, respectively. These two classics of Hindavi verse were composed by sufi poets on the basis of a heterogeneous range of sources that included popular tales, and gave a mystical framework to express the soul’s yearning for God.45 In this we see a foreshadowing of Banarsidas’s later spiritual orientation. At this time, however, he was more interested in love poetry, and this was probably what attracted him to these texts.46 Mukund Lath has said of the Mirigāvatī and Madhumālatī that “though allegorical in intent, these poems retained the poignancy of a purely human passion, which accounted for their great popular appeal”.47

In the Ardhakathānak Banarsidas provides us with the details of his poetic education, for in the seventeenth century as today, the performance of poetry was a skill that required training and practice. It was common practice for the sons of merchant families to be given basic education in letters and numbers, as these skills were essential for their trade, so Banarsidas was sent to a local Brahmin in Jaunpur. Banarsidas wrote, “As a child of eight years I went to school to learn how to read. My guru was a Brahmin, from whom I learned letters, reading and writing. I studied for one year. Every day my knowledge increased, and I became proficient in my learning”.48 For most merchants this would have been sufficient, but at the age of fourteen Banarsidas studied with another local Brahmin teacher in Jaunpur, a Pandit Devdatt. He engaged in a concentrated course of higher studies. He studied astrology (jyotisha) and a work on mathematics, the Khaṇḍasphuṭa. The rest of his studies prepared him to be a poet. He studied two basic lexicographical texts: the Nāmamālā on synonyms, and the Anekārthakośa on words with multiple meanings.49 He studied the techniques of poetic embellishment (alankara) and also a text on erotics, the Laghukoka.50 Banarsidas said that he spent an entire year in these studies.

During this period Banarsidas also engaged in studies of a more religious nature, under two Shvetambar monks, Bhanchand and Ramchand.51 In addition to doctrinal and ritual texts, he studied many Jain hymns. He continued his literary studies at the same time. He learned metrics through the Śrutabodha and the Chandakośa. Under the two monks he also continued his study of other aspects of Sanskrit grammar. In the conclusion of his autobiography, Banarsidas summarised his literary skills: “I can recite both Sanskrit and Prakrit in a correct manner, I am skilled in various vernacular languages, and I understand the nuances of words and meanings”.52

The result of this extensive study was that Banarsidas became a trained poet. He composed a long Hindi poetic text of 1,000 verses in which he explored all the nine rasas, but by his own admission the main focus was only love: “I wrote a new book, in which there were a thousand verses. The theme on which I wrote was the nine sentiments, but mainly it just described love”.53

Banarsidas was sufficiently skilled that he was able to teach these subjects. Several years later the governor of Jaunpur, Chini Kilic, who was also an able vernacular poet, studied several of these texts under Banarsidas’s tutelage.54

It was these skills that Banarsidas later brought to religious poetry. None of his earlier secular poems survive, but there are a number of extant religious poems that were collected soon after his death in the Banārsī vilās. I think we can assume that these were composed and sung in the learned circles of Agra in a similar congregational, cooperative manner as the composition and study of his Samaysār nāṭak. Many of them come with an indication of the raga in which they were to be sung.55

The Performance of Translation: Banarsidas

In all of this, there was nothing unique about Banarsidas, although the rich detail he provides in his autobiography is largely unmatched for the period. But there was something else about his literary endeavours that was possibly ground-breaking: in addition to being a prolific author of original works, Banarsidas was a translator who produced iconic Brajbhasha versions of classical Sanskrit texts.

Mukund Lath has written of Banarsidas’s autobiography that the author “was evidently working without precedents” when he composed a text in this new genre.56 We can say something similar about his work in translation. Banarsidas composed five texts that we can describe as translations, two of which do not fit easily into established indigenous genres of translation in South Asia.

In several places in his autobiography, Banarsidas gave lists of poetic texts he composed, often as a result of the religious turns in his later life. The first was as a teenager in Jaunpur, while he was still studying with Bhanchand, when he composed two texts that are now lost: his thousand-verse erotic poem on the nine rasas, and a book on Sanskrit grammar titled the Pañcasandhi.57

The second period was in 1613, during his first residence in Agra, in his mid-twenties. He wrote two texts: the five-verse hymn Ajitnāth ke chand, addressed to the second Jina; and a grammatical text on synonyms, the Nāmamālā.58 While the first was an original composition, the latter was the first of Banarsidas’s compositions that we can call a translation, in this case an indexical translation.

In the period soon after 1623, and his first exposure to Rajmall’s Bālbodh on the Samayasāra, Banarsidas became more influenced by a spirit of aversion to worldly affairs (vairag bhav). During this time, he composed several texts: the Gyān pacīsī, twenty-five verses on spiritual knowledge; the Dhyān batīsī, thirty-two verses on meditation; several mystical songs (adhyatam ke git); several miscellaneous songs in the kabit (kavitt) metre;59 and a hymn entitled Sivmandir.60

Banarsidas’s text would lead the reader to think that all of these were his original compositions. The last one, however, is another translation. The Sivmandir is Banarsidas’s Brajbhasha iconic translation of one of the most popular of all Jain hymns, the medieval Sanskrit Kalyāṇamandira stotra of Kumudachandra.61 While its author was a Digambar monk, the hymn quickly became popular among Shvetambars as well. In forty-four verses, the poet praised Parshvanatha, the twenty-third Jina.

A few pages further on in his autobiography, Banarsidas gave another list of more than a dozen texts that he wrote between 1633 and 1645, when he was a mature writer living again in Agra.62 This list is also deceptive, as mixed in among a number of original texts are two more translations, one an iconic translation, the other an indexical one.

In 1633, Banarsidas finished the Sahas aṭhottar nām, or Jinasahasranām. As its name indicates, this hymn is a eulogy involving the 1,000 names of the Jina. Its ritual function as an auspicious benedictory text is indicated by its location at the start of the Banārsī vilās. This was another indexical translation. Also in this long list was the Sūktimuktāvalī, which Banarsidas and his colleague Kaunrpal (also spelled Kumarpal and Kanvarpal) composed in 1634. This was a Hindi iconic translation of a Sanskrit poem of 100 verses by the medieval Shvetambar monk Somaprabha. He composed the Sūktimuktāvalī, also known as the Sindūraprakara, in 1177. It covers a range of topics in Jain devotion, practice and belief, in the form of epigrammatic verses modeled on secular moral poems in the niti genre.63

Finally, between 1635 and 1639, after he had studied the Gommaṭasāra with Pande Rupchand, Banarsidas composed a number of works in the spirit of Digambar mysticism (adhyatma).64 Among these was his magnum opus of 1639, the Samaysār nāṭak. This was an indexical translation of the Samayasāra, the great masterpiece of Kundakunda, the foundational philosopher for the Digambar tradition. His last work was his autobiography, completed in 1641. While he optimistically entitled it “Half a Tale”, since he believed (despite all the evidence of seventeenth-century mortality in North India) that at fifty-five years he was only half-way through his life, he died soon thereafter, perhaps in 1643. Most of his poetic works were collected into the Banārsī vilās in 1644 by his friend, fellow poet, and fellow spiritual seeker Jagjivan.

Let me turn to a short analysis of Banarsidas’s five translations. My goal here is not to render a judgment on these texts as translations in terms of how faithful they are to their originals, nor on their qualities as original writings. Rather, I want to investigate what they can tell us about the practice of translation in the cosmopolitan literary cultures of North India in the first half of the seventeenth century.65

Nāmamālā

Banarsidas’s first translation was written in Agra in 1613. In his autobiography he wrote that he spent four months that year composing two texts. One of these was the Ajitnāth ke Chand, a five-verse hymn, and the other was the Nāmamālā. This was a lexicon of synonyms. He had earlier studied with his Sanskrit teacher in Jaunpur a Sanskrit Nāmamālā—probably the famous Nāmamālā of the ninth-century Digambar lay scholar Dhananjaya—and his vernacular text was in some manner based upon the Sanskrit predecessor. R.K. Jain and Mukund Lath note that Banarsidas’s lexicon was one of the earliest texts of its kind in Hindi, being predated only by two lexicons written by the Vaishnava poet and scholar Nanddas sometime in the latter part of the sixteenth century.66

To characterise Banarsidas’s Nāmamālā as a translation clearly stretches any definition of this genre near to breaking point. He made no direct reference to a Sanskrit original. He simply said of his text, “I compose the accessible Nāmāvalī, for the sake of educating the beginner. I make a river of vocabulary to illuminate the correct meanings. Banarsi makes it in the vernacular, according to his own development and thought. I have assembled various words from the vernacular, Prakrit and Sanskrit”.67 The notion that his text is based upon Dhananjaya’s is at best scholarly conjecture.68 Banarsidas’s text is only 175 verses, whereas Dhananjaya’s is 203 verses in length. Banarsidas perhaps inadvertently acknowledged his debt to Dhananjaya’s text when in his autobiography he mistakenly wrote that his Nāmamālā was 200 verses long.69 His statement that he made vernacular a subject from the Prakrit and the Sanskrit for the sake of educating the beginner (bāl vibodh) also locates this text within the field of other “translations” which I discuss below.

The order of words for which synonyms are given is unrelated in the two texts, and even the synonyms vary widely. For example, Banarsidas started his text, appropriately enough for a Jain author, by devoting two verses (4-5) to twenty synonyms for tirthankara. Dhananjaya, on the other hand, did not come to tirthankara until verse 116, when in a single verse he gave nine synonyms. Only five words are found in both lists.

Sahas aṭhottar nām/Jinasahasranām

In 1633, Banarsidas wrote another text that similarly stretches any definition of translation. This was his Sahas aṭhottar nām or Jinasahasranām, a hymn in which he gave 1,000 (or, more accurately, 1,008) names of the Jina. In the beginning of the hymn, in wording almost identical to that found in his Nāmamālā, Banarsidas wrote that the names in it are in vernacular, Prakrit and Sanskrit.70 He varied the metres in the ten sections, each of which contains 100 names and so is called a shatak. The metrical variety enhanced the performance of the text in a temple setting.

Just as he no doubt consulted Dhananjaya’s text (and possibly others) when composing his Nāmamālā, it is likely that he based his hymn in part on one or both of two very popular Sanskrit Jinasahasranāma Stavanas, and so it serves as an indexical translation, not an iconic one. Banarsidas characterised his text at its conclusion as bhāṣā-jinasahasranām, so he viewed it as the vernacularisation of an earlier textual tradition in classical languages. The first Sanskrit text was by the South Indian monk Jinasena (c.770-850), and was contained within his Ādipurāṇa, the most popular and influential of all Digambar texts on the Jain universal history. The Jinasahasranāma within this encyclopedic text has long circulated as an independent text. The second was by the thirteenth-century lay pandit Ashadhara.71 Each of the three texts consists of ten sections, but the titles of the sections differ, as do the lengths of the texts in total.72 In other words, the Sanskrit texts served as models for Banarsidas, and sources for many epithets, but he did not set out to make an iconic translation of either Sanskrit text.

Samaysār nāṭak

As I noted above, the text for which Banarsidas is best known among Jains is his Samaysār nāṭak. This is a translation of the Samayasāra of Kundakunda, one of the central texts of Digambara philosophy and mysticism. With this text we come a bit closer to what we might consider as a translation, even if still very much in the indexical sense. It is unlikely that Banarsidas could read Prakrit sufficiently to have used Kundakunda’s root text as his source. Instead, his classical source was the Sanskrit Samayasāra kalaśa, written in about the twelfth century by Amritachandra. As Bansidhar Bhatt has explained, Amritachandra originally wrote a Sanskrit commentary on Kundakunda’s text, which he entitled the Ātmakhyāti. The commentary consisted of both prose and 278 Sanskrit verses. In time, the verse portion of the commentary was separated from the prose, and became treated as an independent text with the name Samayasāra kalaśa. Amritachandra’s text was to a significant extent responsible for the widespread Digambar adoption of Kundakunda’s two-truth mystical philosophy, according to which from the absolute perspective (niścaya-naya) the self (atman) is the only existent that is “really real”, whereas everything else is only provisionally real from the relative perspective (vyavahāra-naya).73 Amritachandra’s text further influenced subsequent Sanskrit commentaries on Kundakunda’s text and also generated vernacular commentaries, starting with a fourteenth-century Kannada prose commentary (vachanika) by the monk Balachandra.

In this vernacular tradition was also the prose Bālbodh of Pande Rajmall, composed in the sixteenth century. As we saw above, this in turn was the inspiration and guide for Banarsidas’s Samaysār nāṭak. According to Bhatt, Banarsidas’s text followed Rajmall’s fairly closely, with the main difference being that the former “is more elaborate and poetic”, and the latter “philosophic and precise”.74 This is not surprising, given that Banarsidas by training and personality was a poet, while Rajmall by training and personality was a philosopher. Bhatt has also carefully delineated the relationship between Banarsidas’s Samaysār nāṭak and Amritachandra’s Samayasāra kalaśa:

The Samayasāra-nāṭaka contains 732 rhymes in Hindi for 278 kalasha-verses, that is, more than two rhymes have been allotted to each kalasha-verse, on the average. But the rhymes are not so evenly distributed. Strictly speaking, the rhymes are not a translation, nor do they have any explanatory character. They just touch on the kalasha-verses and run independently as a lucid poetry.

Param jyotī stotra/Kalyāṇamandira stotra

If Banarsidas’s Nāmamālā, Jinasahasranām and Samaysār nāṭak were vernacular texts that bore varying degrees of indexical translation relationship to their Sanskrit forebears, the other two texts were clearly translations in the most iconic sense.

Banarsidas’s 1633 translation of the Kalyāṇamandira stotra is also known as the Param joti (or Param jyotī) stotra, after the first two words of the first verse.75 This has long been a common way of “titling” hymns; for example, the Sanskrit original starts kalyāṇamandira, “auspicious temple”. For reasons of metre, Banarsidas translated this as śivmandir, also “auspicious temple”. The first verse of Banarsidas’s poem, a simple doha, is not found in the original, and so serves as an independent invocatory verse.76 Only with the second verse did Banarsidas begin directly to translate Kumudachandra’s text. Banarsidas then in forty-three subsequent verses in a variety of metres reproduced in Braj the forty-four verses (all in the Vasantatilaka metre) of the original Sanskrit.

The final verse is a slightly longer one, of six feet, allowing Banarsidas to provide a signature colophon, in which the translated nature of his composition was made clear:

The wise Kumudachandra made this Kalyāṇamandira.
Banarsi said it in the vernacular, for the sake of pure right faith.77

Sūktimuktāvalī

The joint translation of the Sūktimuktāvalī by Banarsidas and Kaunrpal, done the following year, in 1634, gives us a further indication of the ways in which textual practice in seventeenth-century Agra was a shared, perhaps even semi-public performance. By this time in his life Banarsidas was obviously a highly skilled and well-practiced poet. Kaunrpal, on the other hand, was a younger poet. He was the nephew of Dharamdas, one of Banarsidas’s business partners and a fellow participant in the spiritual seminar. Kaunrpal outlived Banarsidas, becoming the leader of the Adhyatma circle in Agra after him. It is likely, therefore, that he was younger by one or two decades. Mukund Lath describes him as “an occasional writer of verses on spiritual themes”.78 He is known to have composed only one other text, a short poem on Jain right faith called the Samakit battīsī. He composed it in 1630, four years before he and Banarsidas finished their joint translation.79

In the signature colophon to their translation, Banarsidas and Kaunrpal gave a hint of how they worked together. They wrote:

It’s called the Sūktimuktāvalī, and has twenty-two chapters.
In total extent the text has one-hundred verses.
The pair of friends Kunvarpal and Banarsi are like-minded (ikacitt).
They did the text in the vernacular, in verses of various meters.80

The key word here is ikacitt (pron. ikachitt), literally “of one mind”, and which I have translated as “like-minded”. It indicates that they saw eye to eye, or thought in the same way, and hence that this was a joint project. In Gujarati the word also has a range of spiritual meanings, of a person who is fixed in meditation (dhyan-stha), or who is absorbed in a spiritual state (tallin),81 so Kaunrpal and Banarsidas may have used it here to indicate a deeper spiritual harmony that suffused their task of collaborative translation.

Within their translation, however, we also see indications of a different modus operandi, as perhaps the two divided the task. Many of the individual verses contain a signature. In twenty-three verses the signature is of Banarsidas, and in only five do we find the name of Kaunrpal. This may indicate that Banarsidas did the bulk of the work, and that Kaunrpal contributed only a few verses. That the verses with Banarsidas’s signature all occur in the first half of the poem, and those with Kaunrpal’s signature all in the last half, may indicate still another way in which the labour was divided. But the use of ikachitt would appear to argue against a division of labour and in favour of a collaborative method of translation.82

Collaborative composition was a common practice of the time, and probably was part of the cultural practice of a circle meeting in a temple. One of the five companions whom Banarsidas mentions in the Samaysār nāṭak, and who also features prominently in the Ardhakathānak, was Dharamdas. He was Kaunrpal’s uncle, and for many years a business partner of Banarsidas. He and Banarsidas both settled in Agra in their later years, and evidently also shared in the practice of poetry. There is only one poem by Dharamdas that is extant, the seven-verse Guruśiṣyakathanī. In Mukund Lath’s description, it is “in the form of a discourse by a guru to his disciple, teaching the hollowness of all worldly things and the superior merits of the spiritual life”.83 To quote Lath further, in this poem “the metre and rhyme are interestingly unusual”. It maintains a strong rhyme scheme throughout the poem, with the final consonant in every case being a doubled retroflex -ṭṭ-. This results in a poem that is baroque in its artificiality. It is most likely not a coincidence that there is a poem by Banarsidas that exhibits a similar form, his Mokṣapaiḍī, “The Steps to Liberation”. In this poem the rhyme scheme involves the equally baroque use of a doubled -ll-. Banarsidas’s poem is significantly longer—twenty-two verses in addition to a two-verse colophon—perhaps indicating his greater skill at sustaining this difficult form. Lath comments, “the two friends had evidently decided to try their skill in a new, exciting metre together”. One can easily imagine the two friends meeting in the veranda of a temple and spending the afternoon engrossed in the shared composition of poems that displayed their literary skills.84

The Continued Performance of
Translation after Banarsidas

The performance of iconic translation that seems to have begun—or at the very least raised to a new level of visibility—with Banarsidas in the middle of the seventeenth century in Agra did not end with him. One of the leaders of the Adhyatma seminars in Agra after the death of Banarsidas was Hemraj Pande.85 He was the author of both indexical and iconic translations.

The first extant reference to Hemraj Pande is from 1644, when he was described as participating in a seminar (gyan mandali) along with Bhagavatidas (also Bhagotidas), one of the men Banarsidas had mentioned earlier in the context of the Adhyatma seminar.86 Hemraj was a disciple of Rupchand Pande. Premi argues that this was the same Rupchand who delivered the sermons on the Gommaṭasāra in 1635,87 and so Hemraj was in the tradition of professionally trained intellectuals working under the leadership of the bhattaraks.

In 1652 Hemraj composed two texts at the request of Kaunrpal, presumably for use in the seminars.88 One of these was a summary of eighty-four points of disagreement between the Digambars and Shvetambars, entitled Caurāsī bol. The other was a vernacular commentary (Bālbodh) on the Pravacanasāra, another central text on Digambar spirituality by Kundakunda. In his colophon Hemraj wrote: “Please listen to how I prepared this simple rendering of this great work. There is in Agra a prominent authority and learned man named Kanvarpal. At his request I rendered this extremely difficult text into the vernacular”.89

Hemraj’s vernacular version of the Pravacanasāra was in the long-standing tradition of vernacular commentaries. He explicitly modelled his version on the earlier vernacular commentary on the Samayasāra by Rajmall.90 Hemraj composed vernacular commentaries on at least five other important Digambar doctrinal texts, and a number of original works in Brajbhasha.91

Hemraj also composed an iconic translation of the c. sixth-century Sanskrit Bhaktāmara stotra of Manatunga, the most beloved of all Jain hymns. While there have been other Digambar translations of Manatunga’s text into Hindi, Hemraj’s remains the most popular; there are many manuscript copies of it in the Digambar libraries of North India, and it is included in almost all of the locally produced hymnals found in every Digambar temple in North India. Hemraj followed Manatunga’s text faithfully, reproducing in forty-eight Brajbhasha verses the forty-eight Sanskrit verses of the Digambar version of the hymn. In a manner similar to Banarsidas’s translation of the Kalyāṇamandira stotra, he began with a simple auspicious verse not found in the original, and then proceeded with the translation proper. Manatunga’s text concluded with a signature verse. Hemraj retained it, and then added his own signature, indicating clearly that he was translating Manatunga’s text:

Devotion will arise in the mind of any man who memorises this hymn.
Manatunga is absorbed in his soul, and has attained the wealth of liberation.
Hemraj rendered the Bhaktāmar in the vernacular for the sake of well-being.
Whoever recites it with the right spirit will attain the land of liberation.92

The Digambar practice of iconic translation has continued unabated ever since the work of Banarsidas and Hemraj. Many of the leading Digambar Hindi poets of North India composed translations. For example, in Agra, Dyanatray (1676-1726) translated the Sanskrit Svayambhū stotra of Samantabhadra (c. fifth century) and the Sanskrit Ekibhāva stotra of Vadiraja (c. tenth century), and Bhudhardas (fl.1724-1749) also translated Vadiraja’s hymn. Poets in other Digambar religious and cultural centres engaged in the constant practice of translation as well. Starting in the late nineteenth and early twentieth centuries we see the introduction of British colonial conceptions of translation, but for several hundred years before this it was an indigenous tradition.93

Iconic Translation in Agra: Something New?

To what extent was the Digambar Jain practice of iconic translation in Agra in the mid-seventeenth century something new? In the absence of explicit theorisation about translation on the part of the translators themselves, or other contemporary literary theorists, this is a difficult question to address.

As I have indicated above, the practice of translation was not new in South Asia. The Jains in particular were accustomed to working in and between multiple languages. Their sacred scriptures and other early authoritative texts were in various dialects of Prakrit, while from early in the Common Era the Jains had been full participants in the intellectual and literary cultures of Sanskrit (and later Apabhramsha). The many Sanskrit commentaries of Prakrit texts often included a literal word-for-word trot or chāyā, in which the Prakrit was as directly as possible rendered into Sanskrit. This was also a common practice for the Prakrit passages in Sanskrit dramas.

Then there was the practice of vernacular commentarial translations (the difference between the two genres usually being one of degree, not kind), often known as bālbodh or bālāvabodh, which Phyllis Granoff has translated as “Instructions for the Unlettered”. While this was an especially popular genre among Jains, it was also widely found among Hindu authors. Jnaneshvar’s thirteenth-century Marathi version of the Bhagavadgītā is one of the best-known examples in this genre, although we have no direct evidence that Banarsidas and the other Jain Brajbhasha poets in seventeenth-century North India knew it. Banarsidas’s Samaysār nāṭak fits firmly within this practice of translation-as-commentary.

There was also a long and well-known tradition of indexical and symbolic translations in South Asia, especially of narrative texts. The Jains for centuries had been among the most prolific at composing such texts, with their many versions of Jain Puranas and other narrative texts existing in all the languages in which they wrote. Mughal North India saw other examples of the vernacularisation of classical texts, such as Tulsidas’s Rāmcaritmānas. In his introduction, Tulsidas explicitly said that he was retelling the story of Rāma in the vernacular.94 Poets in most of the languages of North India composed vernacular versions of the Hindu classics such as the Rāmāyaṇa, Mahābhārata, and Bhāgavata-purāṇa.95 Some of these were indexical and symbolic translations, and hence most scholars in English have characterised them as “versions”, “retellings”, “reworkings”, or “adaptations”, not as “translations”.96 Some of these, however, such as Nanddas’s Brajbhasha Bhāṣā dasamskandh, an unfinished translation of the tenth book of the Bhāgavata-purāṇa, more closely approximated iconic translations.97 How aware Banarsidas, Kaunrpal, Hemraj, and the other Jain authors were of these contemporaneous developments in “Hindu” (both Brahminical and non-Brahminical) circles is difficult to know, although the Jains have always been avid readers, as is evidenced by the number of non-Jain texts found in Jain manuscript libraries. As we saw in the case of Banarsidas, Jains have also been dependent upon Brahmins for “secular” education in both basic literacy skills and the more advanced skills needed for literary production. The intellectual and literary boundaries between these communities were highly porous.

There was also a vibrant culture of literary translation in the Mughal court itself, as well as in other Islamicate courts. Classics of Indic narrative and science were translated into Islamicate languages, and in other ways Persianised and Islamicised, as in the example of the Mirigāvatī, Madhumālatī, and other Indic folk texts that were recast in sufi and courtly settings. There were also translations of Islamicate texts from Persian into Sanskrit and Indic vernacular languages.

Here again we run into silences in the literary and historical record that limit our ability to speak of how aware Banarsidas and his fellow poets were of these literary developments in Islamicate circles. As Rupert Snell has recently observed, “An often-felt frustration for those concerned to read Hindi literature against its own historical backdrop has been precisely the lack of connectivity between the literary texts on the one hand and historical data on the other”.98 While the work of scholars such as Allison Busch and Audrey Truschke certainly allows us to say much more about literary and social interactions within and without Mughal court circles, we are still somewhat in the dark concerning where the Jains fit into those interactions.99 This problem of silence is exacerbated in the case of Jain Brajbhasha poets, who are invisible in the writings of their Hindu and Muslim contemporaries.100

Banarsidas himself makes no mention of any literary activities in Agra except those of the Jain seminars. The Jains (both Digambar and Shvetambar) played crucial economic and administrative roles in the Mughal court, so we have to accept that his silence may indicate a lack of interaction on literary matters. He was explicit about such interactions in the provincial court of Jaunpur where, as we have seen, he taught and practiced poetry with the governor, Chini Kilich. Social and literary interaction might have been easier in the provinces than in the imperial centre. It was presumably also in Jaunpur that he learned the Mirigāvatī and Madhumālatī, which he publicly sang during his first residence in Agra as a young man. His very knowledge of these Hindavi sufi classics bespeaks a literary culture that was not hindered by religious boundaries. In his Banārsī vilās we also find an eleven-verse Śāntināth Jin stuti, with the notation that it “is in the style of the Candavā of Vaki [Baqi?] Mahammad Khan”.101 While this author and text are unknown to me, this is evidence of further knowledge on the part of Banarsidas of Islamicate literature. Banarsidas’s curiosity about matters literary and spiritual (and, in his youth, erotic) meant that he was clearly widely read.102 But we cannot judge whether or not his possibly innovative approach to translation was in any way influenced or shaped by translation practices in the Islamicate courtly circles of Agra.

It is possible that Ramanujan’s tripartite classification of translation is less helpful than it seems at first. What makes sense in twentieth-century Chicago from a Piercean semiotic stance may be less applicable in the literary circles of seventeenth-century Agra, as the concerns about different modes of translation were different in the two settings.103 Nonetheless, Ramanujan’s typology does allow us to see much more clearly that questions of accuracy and faithfulness that are central in the long-standing Western practice of iconic translations tend to disappear in communities that put more emphasis on the practice of indexical translation. Ramanujan allows us to see that South Asian concerns about “translation” have for many centuries simply been different from those in the literary cultures that collectively make up the “Western” tradition. Theorists of translation from the times of the Greek translations into Latin, and of Biblical translations into Greek, Latin, and European vernaculars, have been concerned primarily with issues of “faithfulness”. Winand Callewaert and Shilanand Hemraj, in their wide-ranging study of translations of the Bhagavadgītā, wrote, “It would appear that in the Indian tradition translators have been more concerned with the communication of the message rather than with the scrupulous preservation of the form”.104 While scholars almost always want to ask why certain issues do not appear to be matters of significant concern in a given culture, nonetheless it may very well be the case that anxiety over the faithfulness of premodern translations was rare in South Asia.

The issue of genre is also significant here. The three iconic translations by Banarsidas, Kaunrpal, and Hemraj which I have discussed are of Sanskrit poems. In translating a narrative text it is easy, and even expected, for the translator to expand, contract, and in other ways modify the source as he brings it across into a new linguistic home. The translation is less of a set “text” as found on paper or in oral memory than of the entirety of the narration itself. The same is true of a philosophical or theological work such as Banarsidas’s Samaysār nāṭak, in which his goal was to bring across into Brajbhasha the concepts of the original, not necessarily the words and style. Here we see again how “translation” in many ways is simply a mode of “interpretation” in another language, and hence blends into the genre of “commentary”. We must also be mindful of the ways that techniques of mechanical reproduction, coupled with the growth of concepts of property rights as extended to literary activity, have profoundly shaped both the practice and theory of translation in the West in recent centuries. What a literary critic or a lawyer might today decry as plagiarised and even actionable was in many other historical and cultural settings simply a matter of standard practice.

I argue that a poem, however, involves a higher degree of unity between form and content than does a narrative or philosophical text.105 Theorists throughout the world have often advanced poetry as a primary example of untranslatability, as a poem in many ways has no content or meaning outside of the very linguistic form itself. In translating a poem, therefore, whether a religious stotra or a more “secular” poem in a niti genre, we should expect a stronger degree of “faithfulness” to be displayed by the translator. The exact number of verses of a longer poem in Sanskrit is often well-known; to alter the length too significantly would be seen as an act of infidelity.106

Translation and Society

Let me conclude with some comparative observations on the practice and the framework of translation in seventeenth-century Agra and after. The same time that Banarsidas and his colleagues were (perhaps) initiating a new practice of translation in North India also saw the golden epoch of translation in England, as well as the widespread practice of and theorisation about translation elsewhere in Europe. Comparing the contemporaneous situations in North India and England is quite revealing. By looking at this alternative setting we can see the extent to which the local situation in seventeenth-century Agra could have been different. Comparing the two cases shows us ways that translation in India did not have the same implications that it did in England, and thereby gives us further insight into cultural and social practices and structures in seventeenth-century North India. In particular, I briefly discuss two facets of the practice and social location of translation that were dominant in England and largely missing in India.

The first concerns philosophical and theological reflection upon the very practice of translation. As I indicated in the beginning of this essay, translation had been the subject of extensive theoretical reflection since the time of the Roman translation of Greek classics into Latin, and then the Christian translation of scripture into first Greek and then Latin.107 Sixteenth- and seventeenth-century English authors gave much thought to central issues of translation, such as the relationship between language and meaning, and the tension between faithfulness and creativity in translation. They asked the basic question of whether translation is even possible, or if meaning is so deeply imbedded in any linguistic form that essential elements of meaning are inevitably lost when a text is transported into another linguistic frame.

One way in which this second-level reflection upon the practice of translation can be seen is through the many metaphors advanced by authors for the translator and his work.108 The well-known and oft-cited Italian play on words, traduttore-traditore, which implies that the translator is merely a traitor to meaning, indicates the depth of anxiety in European thinking on translation. The translator is someone who follows in the footsteps of others. He engages in a contest that he inevitably loses. He is a mere labourer, or even a slave, in the service of a master. He offers glass and imitation jewels in place of diamonds. He shows only the back of a tapestry, not its front. The reader of a translation dons a poor garment, not a rich one. The translation is a pale star that only dimly reflects the sun. It is only a shadow, not a body.

The richness of the metaphorical language in Europe is evidence of widespread thinking about the nature of translation. This stands in stark contrast to the very lack of a word to describe the act in seventeenth-century India. Banarsidas, Kaunrpal, Hemraj, and their successors simply said that they were “making it vernacular”, using the noun bhasha (bhāṣā) and a form of the verb √kar. Nor do the Jain poets exhibit any anxiety concerning the relative merits of their vernacular translations vis-à-vis the Sanskrit or Prakrit originals. Translators in Europe and North America ritually express the inadequacies of their translations in their prefaces; the Jain poets express no such angst.

This is not to say that vernacular poets did not make value judgments concerning the importance of classical and vernacular languages. The very language of the balbodh or balavabodh, in which the vernacular text is expressly aimed at an intellectually less-advanced audience, indicates that the truly educated reader or hearer would be able to access the classical original, and would not need the prop of a vernacular version. Banarsidas, as we have seen, said in his Samaysār nāṭak that educated people (pandits) could read the Prakrit original and Sanskrit commentary, but that common people with little understanding (alpmati) required the vernacular versions of Rajmall and Banarsidas himself. A similar value judgment is found in the Bhāṣā dasamskandh of Nanddas, who, in the opening verses of his translation, wrote, “This is the Daśam skandh, in which I have given some descriptions in the vernacular. The words are like the Sanskrit, but that is difficult to understand. I have therefore made it easy by making it vernacular”.109

Deven Patel has discussed an articulation of the value of Sanskrit over the vernacular by the fifteenth-century Bhalan in his Nalākhyān, a Gujarati translation of the Nala episode of the Mahābhārata. Bhalan’s comments bear striking similarity to the metaphors used by European translators that I mentioned above. He wrote that he was presenting “glass set in cheap metal” in place of “real diamonds, embedded in gold ornaments”, that his poem was the “coarse millet” eaten by a poor man instead of the delicacies enjoyed by a rich king, and that his poem was a “leaf hut” of the down-trodden in lieu of a seven-storey high-rise building inhabited by the successful.110 Here again, I think that the anxiety expressed by Bhalan is not the same as expressed by his contemporaneous European translators. Bhalan was bemoaning having to compose in the lowly vernacular instead of the truly polished and sophisticated Sanskrit in which Vyasa had been able to compose the Mahābhārata. Bhalan was making a twofold hierarchical judgment: Sanskrit authors are superior to vernacular authors, and an audience that understands Sanskrit is superior to one that only understands the vernacular. He was not, however, raising a philosophical or theological question concerning the very possibility of translation. Bhalan wrote in the vernacular with the full confidence that by doing so “even a child could understand”.111

While there does not appear to have been any reflective discourse on the practice and possibility of translation on the part of the Jain authors, the sort of anxiety that has long underlaid European thought on the subject was not absent.112 As George Steiner has noted, one source of anxiety about translation is theological: “So far as speech is divine and numinous, so far as it encloses revelation, active transmission either into the vulgate or across the barrier of languages is dubious or frankly evil”.113 This is a religious theory of untranslatability. We find such a theory in India as well. The best-known case is perhaps the Brahminical Vedic literature, which is untranslatable due to the way that the “meaning” of the text inheres to the very sounds themselves at a level beneath the surface sense.

We find a similar mantric rejection of the possibility of translation among the Jains. Due to the Jain theory of mantras, the vernacular translations of hymns such as the Bhaktāmara stotra and Kalyāṇamandira stotra never supplanted the Sanskrit originals in Jain ritual culture.114 Jains understand that a Sanskrit (or Prakrit) stotra is not simply a communication event. It is also a performance event. A stotra is understood to be mantra-maya, i.e. “made of mantras”. While the expectation is that the person who recites or sings it is full of faith, and that the intention embodied in that faith is important for the efficacy of the performance of the stotra, there is a significant extent to which by its very mantric nature the stotra is efficacious all on its own. This efficacy does not translate into the vernacular. The vernacular versions of Banarsidas and Hemraj are widely acknowledged to be beautiful, and they are found in most contemporary Digambar hymnals. But they are not mantra-maya, and so the number of people who sing or recite them daily pales in comparison with the numbers who perform the Sanskrit originals.

A second key aspect of translation in sixteenth- and seventeenth-century England also appears to have been missing in India. As Michaela Wolf has succinctly put it, “Any translation, as both an enactment and a product, is necessarily embedded within social contexts”.115 As a result, translation needs to be placed within these contexts; there needs to be what Wolf and others term “a sociology of translation”. Translators live in specific social locations, and so the very act of translation has social implications.

This is clearly seen when we look at the English contemporaries of Banarsidas, Kaunrpal, and Hemraj. Translation in England was part of two larger social processes. One was the discovery and appropriation of the classical past. In the words of F.O. Matthiessen, “A study of Elizabethan translations is a study of the means by which the Renaissance came to England”.116 Translation was also part of the broader Protestantising of European Christianity. The process of vernacularising the Bible began before the Reformation, but in the early sixteenth century any attempt to translate the Christian scripture was inevitably done within a charged religious context—and in an era when religion and state were thoroughly intertwined, it was therefore done within a political context as well. This was a lesson William Tyndale learned all too well when he was tortured and executed for heresy in 1536 for daring to translate the Bible into English.

In seventeenth-century England translation as a cultural practice was still deeply implicated in contemporary social and political processes, for it was inseparable from the production of literature more broadly. This was not a period in which freedom of expression was taken for granted. Rather:

the ideological policy of the English monarchial, ecclesiastical, and patriarchal state during and after the reformation was to maintain political and institutional control over individual and group reading. Reading, here, encompasses interpretations and translations of the texts of Scripture, the law, and classical antiquity […] all directed reading and translation was to redound to the good of the commonwealth.117

Especially in the contentious decades of the Civil War and the Restoration, all publications were seen by whichever party was in power as having social and political implications. This was a time of printed pamphlets, so the lag between composition and publication was very short. Printed literature had a social and political immediacy that nowadays we associate with media such as newspapers, radio, television, and now the blogosphere. All publication, therefore, was done within the social context of government censorship.118

In such a setting, many authors turned to translation of the Greek and Roman classics as a way of saying things for which they would be jailed if it were an “original” work.119 The loyalist John Dryden, for example, found it safer to express many of his pro-monarchy opinions in the words of classical authors than as his own thoughts.

Here again by turning to seventeenth-century England we see features of translation as a social practice that were missing in Agra. The Jain poets were clearly aware of the political conditions within which they lived and worked. Banarsidas’s Ardhakathānak is full of discussions of the doings of the Mughal rulers and their agents. The colophon to his Nāmamālā includes praise of the Mughal Emperor Jahangir: “Every day his glory and majesty is victorious, and there is always bountiful food. The emperor is the source of steady radiance, that Emperor Jahangir”.120 Many colophons of manuscripts include explicit reference to which Mughal emperor was on the throne.121

The Jain poets’ translations, however, were not politics by other means. They did not turn to translations of classical and medieval Prakrit and Sanskrit works to say things they otherwise could not say in Hindi. A very significant difference was one of technology: Banarsidas, Kaunrpal, and Hemraj operated in an economy of the hand-written manuscript, whereas their English contemporaries operated in an economy of the printed, mass-produced pamphlet. Banarsidas made no mention of how he obtained the paper on which he wrote his texts, but we know for certain that writing for him involved ink and paper, and that texts were physical objects on paper. We saw above that Arathmal Dhor gave him a manuscript (likhi, pothi) of the Samayasāra in 1633. Many years earlier, in 1605, Banarsidas had repented of the infatuation with love represented in his long poem about the nine rasas. In a fit of regret at what he perceived as misspent teenage years, he threw the manuscript (pothi) into the Gomti River in Jaunpur, where the pages (patra) floated away.122 Banarsidas’s seeming ready access to paper may have been a sign of his middle-class merchant economic status, and paper may not have been as readily available in seventeenth-century North India as it was in seventeenth-century England. C.A. Bayly indicates that it was not until the later seventeenth and especially the eighteenth centuries that North India began to see the widespread production of paper manuscripts.123

Banarsidas’s English contemporaries also lived in a society in which the ruling authorities expressed greater concern to control their subjects, as evidenced by the active presence of official censors. While the Jain translators would have been in trouble had they pronounced anti-Mughal sentiments in the public sphere, the Mughal (and more broadly late-medieval Indian) culture of allowing religiously-defined communities a fair degree of “private space” meant that their translations did not inherently carry political consequences. The Mughals sought to gain the knowledge of their subjects that they needed to rule effectively; but Bayly has estimated, “the detailed descriptions of royal intelligence we encounter may have represented more of an aspiration than a reality”.124 The Mughals did not exert detailed control over the religious lives of their subjects, especially outside of the most visible public sphere, as “the close bureaucratic control of heresy and witchcraft which forged the early modern European state had no equivalent in India”. Religious communities such as the Jains were expected to police such matters internally, and only if matters “passed beyond the bounds of propriety”, or involved major public shrines, did the officials intervene. Banarsidas gave no evidence of Mughal intrusion into his religious life; only as merchants (and therefore sources of ready revenue for a cash-poor government) were he and his fellow Jains subject to close scrutiny and interference.125 The Mughal court did not have a high ranking post of official censor, and the acts of reading, writing, and translating were not seen as inevitably tied to the ruler and the legitimacy of his rule.

Translation was not innocent of all social implications, however. Nathuram Premi has argued that the hundreds of translations of Prakrit and Sanskrit (and, I should add, Apabhramsa as well) texts into the vernacular undermined the traditional authority of the bhattaraks. Premi said that this was an intentional program on the part of Terapanth scholars, who were opposed to the domesticated bhattaraks of the Bispanth.126

The past three centuries have indeed seen a dramatic social and theological transformation in the North Indian Digambar community due to the rise of the Digambar Terapanth sect.127 While many of the roots of this transformation go back to the Adhyatma movement of Agra and other urban centres in the sixteenth and seventeenth centuries, its more immediate origins are found in the Jaipur area in the early eighteenth century. The Terapanth is a lay movement that has argued against the worship of deities other than the Jinas, against the use of flowers and other living (or formerly living) substances in rituals, and against the authority of the bhattaraks. As a result of this movement, many Digambar congregations and temples in North India (and especially in Bundelkhand) now follow Terapanth ritual practices, and have rejected the bhattaraks. The movement has been so successful that even Bispanth congregations and temples, that maintained their adherence to the older Digambar ritual culture, have adopted aspects of Terapanth style. In particular, the Terapanth criticism of the bhattaraks as false Jain monks has resulted in the extinction of the bhattarak tradition throughout North and central India; it survives only in South India, in Kannada and Tamil speaking areas. Premi argued that the translation of classical texts into the vernacular was part of a Terapanth agenda of shifting literary and intellectual authority away from the bhattaraks and their trained assistants, the pandes, into the hands of the laity.

I am reluctant to make such a strong claim in the absence of any explicit evidence that this was the intention of the translators. Many translations also came from the hands of Bispanth authors. There is no denying, however, that the traditional authority of the bhattaraks, which was based on their control over the Sanskrit literary culture of the Digambars, and was institutionally personified in the pandes who were their disciples and whom they deputed to various temples to conduct rituals, give sermons and lead seminars, was lessened as Digambar textual culture moved out of their control and into the broader Digambar public sphere. This process was, to a significant extent, connected to the practice of translation.128

Making it Vernacular:
Translation in South Asia

In this essay, I have employed A.K. Ramanujan’s tripartite categorisation of translation to bring into focus some of the distinctive aspects of translation as literary practice among Jains in seventeenth-century Agra. Translation was nothing new. Just as multi-linguality is a universal aspect of human culture, so is translation, as people have never allowed the existence of multiple and even mutually unintelligible languages to be an impenetrable barrier to linguistic communication. Translation has a deep history in South Asia, and Jains have always been vital participants in that history. They have translated among Sanskrit, Prakrits, Apabhramshas, and other classical and vernacular languages for two millennia.

The medieval and pre-modern period in North India was also a time and place where the need and practice of translation was highlighted. While Islamicate cultures and languages had been present in South Asia for many centuries, the Mughals and their courtiers brought into South Asia a renewed emphasis on Persian, Turkic, and Arabic literary cultures. The Mughal courtly culture was also deeply intertwined with indigenous cultures, to the extent that the distinction of foreign and native, imported and vernacular, was often meaningless on the ground. Sanskrit and related classical Indic languages had a vital place in the cosmopolitan urban culture of North India. While Prakrit was clearly waning, and relegated to restricted social settings, it was not totally absent. Fifteenth-century Gwalior saw the final flowering of Apabhramsha.129 While Brajbhasha and other vernaculars did not always carry the same prestige as the older literary languages in courtly circles, this period saw their rise to a position of prominence.130 In other words, urban North India in this period was home to many linguistic literary traditions. Many authors were fluent in multiple languages, in continuity with older court traditions that a true poet, to be one of the jewels of the court, should be able to compose in as many as six languages at once.

In all of this linguistic richness and diversity, translation was an inevitable activity. Translation went in multiple directions. Many of these translations were in the time-honoured South Asian tradition of indexical and symbolic translations, as earlier narratives, theological treatises, and philosophical discourses were brought into new languages with varying degrees of innovation and faithfulness. Some of these bore clear traces of their indebtedness to the texts and languages being translated, and others were translations only in the loosest understandings of the term. There were many vernacular commentaries, aimed at a putatively childlike or unlettered audience that was fully comfortable only in the vernacular. While far too few of these translations have received the attention they are due, as scholars have preferred to study texts deemed “original” instead of translated, translation in various forms was ubiquitous. All of these genres of translation also fit easily into long-established modes of translation in the subcontinent.

But there may have been something new, as well. Banarsidas, Kaunrpal, and Hemraj all produced iconic translations of Sanskrit poems. In these translations, the translators explicitly acknowledged the authorship of the original text by a Sanskrit poet. They faithfully adhered to the length of the originals in their translations. They did not claim to be composing original texts, but rather simply to be “making it vernacular”. Whether or not these translators saw what they were doing as something new and innovative is open to question; there is no evidence that they did. But this is part of a larger silence of theorising about translation in the pre-modern South Asian literary tradition. Just how new this mode of iconic translation was is also open to question. Arguments from silence are always hazardous. This is especially the case when the silence concerns vast archives that have barely been studied, and a subject that has been marginalised by generations of scholars.

This essay is a highly tentative foray into a topic that has hitherto received little scholarly attention. Almost all discussions of translation in India are from a postcolonial framework, and concern the power disparities between English and vernacular writing. Translation, however, even faithful iconic translation, is not something that did not exist in India until the coming of British colonialism. This essay, therefore, serves as an introduction to one particular, local instance of translation as a cultural and social practice. It is also an invitation to other scholars to explore other literatures in pre-colonial India using the framework of translation. Among the Jains of seventeenth-century Agra, at least, the performative aspects of North Indian literary culture involved translation in addition to other genres.