The archive: 300 years of commentarial writing in Hindi

Modern scholarship's neglect of precolonial commentarial literature in Hindi is striking given that the modern inheritor of that commentarial writing tradition is hiding in plain sight. On the platforms of every major train station in north India, one finds copies of the religious classics in Sanskrit for sale; these publications invariably include a ṭīkā (commentary), bhāṣya (exegesis), or anuvāda (translation) in Hindi.Footnote ⁷ Editions of Sanskrit works of literature, philosophy, and science that are read by university students in north India often include a chāyā (lit. ‘shadow’, a literal or rough translation) in Hindi. The publications of various religious communities and organisations, yoga schools, and gurus present excerpts from Sanskrit works with translation and commentary in Hindi. Though no quantitative study has been undertaken in this regard, it could easily be argued that the majority of today's readers of Sanskrit in the north access the language and its works through the Hindi paratexts, translations, and commentaries that accompany printed copies of Sanskrit works. This has been the case since the technology of printing was popularised across north India in the early nineteenth century by British missionaries and administrators. Popular publications such as the Rāsapañcādhyāyī (Five Chapters on the Rāsa, 1828–1829), Muhūrtagaṇapati Saṭīka (The Annotated Auspicious Moment of Gaṇapati, 1894), Bhaktirasāmr̥tasindhu (Ocean of the Nectar of Bhakti of Rupa Gosvāmī, 1900), and multiple editions of the Bhāgavata Purāṇa published from the mid-nineteenth century onward are evidence of the demand for such vernacular commentaries on Sanskrit works.Footnote ⁸

These modern publications were not wholly an invention or product of print capitalism, but rather the inheritors of a thriving commentarial tradition in the manuscript culture of precolonial north India. Even a cursory survey of the manuscript catalogues of archives and libraries such as the Lalbhai Dalpatbhai Museum, Rajasthan Oriental Research Institute, Nagari Pracharini Sabha, Hindi Sahitya Sammelan, and Vrindavan Research Institute reveals dozens of such commentarial works, copies of which number in the thousands. Colophons and other evidence tell us that these copies were produced and circulated among monks, merchants, princes, kings, and amateur intellectuals across north India.

The history of Hindi commentarial writing—and, in particular, commentarial writing on Sanskrit works—begins with Sanskrit itself. As Gary Tubb and Emery Boose have written, commentary constituted a flourishing literary tradition within Sanskrit; far more than just an ancillary appendage to the source text, the commentary was understood to be the discursive space in which intellectuals engaged with past thinkers and proposed new ideas.Footnote ⁹ As a whole, the Sanskrit intellectual tradition placed a high value on engagement with prior and authoritative sources; whatever new idea or innovation an author might propose had to be grounded in either a confirmation or refutation of the ideas of recognised authorities. In such an intellectual culture, it makes sense that the commentary was the form in which many thinkers chose to put forward their ideas—even radically new ones.Footnote ¹⁰ As Whitney Cox has argued, commentarial literature is one of the primary domains in which an Indian practice of philology was accomplished.Footnote ¹¹ Consequently, composing a cogent and influential commentary became a way of establishing one's intellectual credentials. Writers in Sanskrit employed different terms for different classes of commentarial writing: a ṭīkā (lit. ‘explanation’), for example, tended to be a commentary that supplied glosses for terms and passages in the source text (which was referred to by the term mūla, lit. ‘root’). In contrast, a bhāṣya (lit. ‘speaking’) generally provided an extended elucidation and exegesis of the source text. A bhāṣya could sometimes be an explanation of a work in ‘common’ language, which would appear to have influenced the later nomenclature of commentarial writings in the vernacular of Hindi. Yet another type of commentary, the pañjikā, provided an analysis of each individual word in the source text (a helpful resource in a language such as Sanskrit, which makes extensive use of compounds and employs various technical lexicons and metalanguages). Other terms, such as avacūrṇikā, nibandhana, and prabandha were also used to refer to commentaries but, as these terms were used for either very specific types of commentaries or to refer to a broader range of scholastic writings, we can set them aside for the moment. It is important to note the salient techniques used across different types of commentaries, including padaccheda (the separation of words joined through sandhi), padārthokti (indicating the meaning of words), vigraha (grammatical analysis of compounds), vākyayojanā (syntactical analysis), akẹspa (raising a question or objection), and samādhāna (resolving a question or objection).Footnote ¹² Many of the composers of Hindi commentaries in the sixteenth, seventeenth, and eighteenth centuries were apparently familiar with the tradition of commentarial writing in Sanskrit and appropriated its techniques and terminology—though not without making substantial modifications.

The second source of the Hindi commentarial tradition is the multilingual literary culture of the Jain community. As John Cort has argued, beginning in the eleventh century of the Common Era, the ability to compose works in multiple languages was the norm rather than the exception for Jain religious scholars. During the first few centuries of the second millennium, this multilingual intellectual production was limited primarily to the languages of Sanskrit, Prakrit, and occasionally Apabhramsha but, beginning in the fourteenth century with the emergence of the Maru-Gurjar language, Jain authors increasingly composed in local vernaculars as well. By the eighteenth century, the list of languages in which Jain scholars were writing had expanded to include the languages that we now recognise as Gujarati, Marwari (or Rajasthani), and Brajbhasha (the predominant literary dialect of Hindi in the sixteenth to eighteenth centuries).Footnote ¹³ Cort suggests that ‘cosmopolitanism’ in the Jain literary and cultural milieu was defined by multilingualism—in contrast to the (predominantly Hindu, Brahmanical) ‘Sanskrit cosmopolis’, in which only Sanskrit could make a claim to the universal and cosmopolitan (per Pollock).

In such a milieu, interlingual commentary writing played an important role in bringing various strands of scriptural and exegetical writing in different languages together: the sizeable Jain manuscript archive is filled with manuscripts containing a root text in one language and one, two, or sometimes even three commentaries in other languages. These commentaries are often called bālāvabodha (lit., ‘instruction [for] the young’) and consist primarily of what a modern-day reader might call a verse-by-verse or sentence-by-sentence ‘translation’ of the source text. Despite their name, these commentaries were not only used for educating neophytes, but were in fact consulted by senior monks and lay scholars.Footnote ¹⁴ This mode of commentary, and Jain multilingual culture more generally, would supply a model for Hindi commentarial writing in the seventeenth and eighteenth centuries.

The tradition of commentarial writing in Hindi begins at the turn of the sixteenth century with the Bhagavad Gītā Bhāṣā (1500 CE) of Theganāth. The previous century and a half had seen the inauguration of literary writing in the vernacular of north India, which was known by various names: Hindī, Hindavī, Hindukī, or, most often, simply bhāṣā (lit. ‘speech’). This early literary activity had occurred primarily at the urban centres of the north Indian sultanates but, by the mid-fifteenth century, the Hindu royal court of Gwalior (particularly under Raja Mān Siṁha, r. 1486–1516) had begun to patronise the production of vernacular literary works, including retellings of the Hindu epics, Jain doctrinal works, and musicological treatises in a literary register that Imre Bangha has identified as Madhyadeśī.Footnote ¹⁵

It was in this milieu of vernacular literary production across multiple religious and literary traditions that Mān Siṁha's uncle, Prince Bhānu, commissioned the monk Theganāth (evidently an associate of the Nāth ascetic tradition) to compose a versified commentary on the Bhagavad Gītā at the turn of the sixteenth century. The single available copy of the work, which dates to the mid-eighteenth century, contains only the commentary, omitting the source text. As Akshara Ravishankar has noted, through various allusions, Theganāth projects the dialogic structure of the Bhagavad Gītā onto his own work and his own relationship with his patron: just as Krishna recited the utterances recorded in the Bhagavad Gītā to Prince Arjuna, Theganāth dictates the meaning of those utterances to Prince Bhānu in his commentary.Footnote ¹⁶

The second known example of commentarial writing in Hindi was also produced in a courtly context: the Vivekadīpikā (Lamp of Discernment, circa 1600), written by Indrajit, the Bundela Rajput prince of Orchha, not far from Gwalior. Like the earlier Tomar kings, Indrajit patronised literature and music in the vernacular and was a poet himself.Footnote ¹⁷ Despite the recognition of Indrajit's role as a poet and patron in modern scholarship, his authorship of an extensive commentarial work has been virtually forgotten. His Vivekadīpikā is a prose commentary on the Sanskrit poems of Bhartr̥hari (seventh century?). Arranged in three śatakas (collections of 100 verses) devoted to the themes of nīti (ethics and statecraft), śr̥ṅgāra (eros), and vairāgya (renunciation), Bhartr̥hari's muktaka (independent) verses had played a role in educating courtly elites in literature and comportment for the previous several centuries.Footnote ¹⁸ Indrajit's ambitious prose commentary provides Hindi glosses for almost all of the terms and phrases in the Sanskrit text, and occasionally gives explanations for particularly obscure or ambiguous references or allusions. Extant manuscript copies of the Vivekadīpikā—which can be traced back to an archetype that most likely dates to the lifetime of the author—include Bhartr̥hari's Sanskrit ślokas.Footnote ¹⁹

After these two initial experiments with the genre—separated from one another by roughly a century and distinct from one another in terms of style—commentarial writing in Hindi experienced an efflorescence. The seventeenth century witnessed the composition of dozens of commentaries in Hindi on all manner of Sanskrit works as well as commentaries in Hindi on works originally composed in Hindi. Two large-scale changes in the political economy and literary culture of northern India drove this rapid development, the first of these changes being the consolidation of Mughal rule over northern India at the end of the sixteenth century under Akbar. The establishment of the Mughal empire brought both political stability and economic expansion, both of which facilitated the patronage of literary works in the vernacular and their circulation across great geographical distances. The consolidation of Mughal sovereignty also led to the formation of a composite ruling class (consisting of Mughals, Rajputs, Afghans, Iranian immigrants, and others) and culture in which the cultivation of literature and knowledge in the vernacular constituted part of courtly comportment and the right to rule. This led to an ever-increasing demand among Mughal and Rajput elites for material in the vernacular, including not only ‘literature’ proper (kāvya in the aforementioned rīti genre), but also works on all manner of sciences, religion, and philosophy.

The second driver of the growth in commentarial writing at the turn of the seventeenth century was the formation and rapid expansion of religious communities associated with the devotional mode of bhakti (loosely translated as ‘devotion’) during the sixteenth century. These communities, which included the Vallabha Sampraday, Gaudiya Sampraday, Ramanandi Sampraday, Dadu Panth, Niranjani Sampraday, and Kabir Panth, among dozens of similar sects, used the vernacular as their primary medium of hymnody, liturgy, scholarship, and religious instruction. Most of these communities also evinced a desire to ground their theologies and ritual practices in textual sources of authority, which inevitably led to an engagement with Sanskrit religious works and genres such as the upaniṣads, purāṇas, śāstra (treatises), stuti (liturgies), and even religiously themed kathā (narratives) and kāvya (poetry). As I have argued elsewhere, many of these communities also attempted to make themselves legible as religious sects or even as independent religions (sampradāya, panth, maẕhab) under the gaze of Mughal and Rajput rulers in order to secure patronage and/or fiscal benefits from the state; appropriating authoritative works in Sanskrit was a particularly effective strategy for establishing one's community as a legitimate religious tradition.Footnote ²⁰

In this context, some Sanskrit works gave rise to entire traditions of commentary in Hindi; one such work was the Bhāgavata Purāṇa—an important source for the literary narratives, ritual cultures, and aesthetic regimes of the aforementioned religious communities including both those of the so-called saguṇa, Vaiṣṇava persuasion, and those of the nirguṇa, aniconic tradition. Most vernacular commentaries specifically address either the tenth book (skandha) of the Bhāgavata Purāṇa, which details the exploits of Viṣṇu and his avatars, especially Kr̥ṣṇa, or the eleventh book, which deals with yoga, renunciation, the duties of a householder, the nature of spiritual knowledge, and bhakti. Though at least a dozen such commentaries were composed during the seventeenth and eighteenth centuries, I will draw examples from only two in this article. The first, the Bhāgavata Purāṇa Ekadaśamaskandha Ṭīkā (Commentary on the Eleventh Book of the Bhāgavata Purāṇa), was composed in 1635 CE by Caturadās, a monk of the Dadu Panth in what is now Rajasthan. Caturadās composed his commentary in verse and, as Monika Horstmann has noted, took much of his hermeneutical and interpretive inspiration from an earlier Sanskrit commentary on the text—the Bhāvārthadīpikā of Śrīdhara Svāmī (fl. fourteenth century CE).Footnote ²¹ The second example, the Bhāgavata Purāṇa Daśamaskandha Ṭīkā (Commentary on the Tenth Book of the Bhāgavata Purāṇa), was composed in 1704 CE by Bhagavānadās, a monk of the Niranjani Sampraday, also located in Rajasthan.Footnote ²² An accomplished poet, Bhagavānadās composed his commentary in a variety of poetic meters.

Like the Bhāgavata Purāṇa, the Bhagavad Gītā prompted a tradition of commentary that developed across theological and sectarian distinctions, with monks and even lay devotees of both Vaiṣṇava and nirguṇa sects composing commentaries on the work. This commentarial tradition technically begins with the aforementioned Gītā Bhāṣā Ṭīkā of Theganāth; however, there is no evidence that any of the commentators of the seventeenth and eighteenth centuries were aware of Theganāth's work. Winand Callewaert lists no fewer than 30 commentaries on the Bhagavad Gītā in Hindi in his study of ‘translations’ of the Gītā; in this article, I will draw examples from only four such commentaries.Footnote ²³ The first is Theganāth's Gītā Bhāṣā Ṭīkā. The second is the Paramānanda Prabodhā (1704), which is a combined verse–prose commentary by Ānāndarāma, who was likely a functionary at the court of Anūp Siṁh of Bikaner (r. 1669–1698). The third and fourth are anonymously authored prose commentaries on the Bhagavad Gītā, of which copies began to appear in the early eighteenth century.Footnote ²⁴

A surge in interest in the Advaita Vedānta tradition of non-dualist philosophy among pandits in religious centres such as Banaras and among political elites at Mughal and Rajput courts during the seventeenth century led to the production of many works on the subject, including dozens of works in the vernacular.Footnote ²⁵ Some of these vernacular treatises are, in fact, commentaries on Sanskrit works. I draw examples from two such commentaries here: the Vedānta Mahāvākya Bhāṣā (Commentary on the Great Sayings of the Vedānta, 1660 CE) and the Ṣaṭpraśnī Nirṇaya Bhāṣā (Commentary on The Resolution of the Six Questions, n.d.), both composed by Manoharadās of the Niranjani Sampraday.

Vernacular commentaries on works of Sanskrit kāvya (poetry) and itihāsa (epic) are less common but certainly not absent from the literary archive. For example, the muktaka poems of Bhartr̥hari—the subject of Prince Indrajit's aforementioned Vivekadīpikā—were the topic of at least two more commentaries, one of which—the Vairāgya Vr̥nda (Collection on Detachment)—was composed by the Niranjani monk Bhagavānadās in 1673 CE. Bhagavānadās also composed a commentary on the episode of the horse sacrifice recorded in the Mahābhārata epic, which is titled the Aśvamedha Bhāṣā (Commentary on the Horse Sacrifice, 1698 CE). I draw examples from both in order to highlight salient aspects of literary commentaries in Hindi.

Scholars who were composing in Hindi commented not only on Sanskrit works, but also upon works composed in the vernacular itself. Famous among early Hindi commentators is Sūrati Miśra (floruit 1673–1744), a pandit at the court of Jorāvar Siṁh in Bikaner; Miśra produced several commentaries on classics of the rīti genre, including two of Keśavadās's seminal poetic handbooks (the Rasikapriyā and Kavipriyā).Footnote ²⁶ As is the case with Sanskrit, some works in Hindi gave rise to entire traditions of commentary: foremost among these is the Satasaī (Seven Hundred Verses) of Bihārīlāl, poet at the court of Jai Siṁh of Amer (r. 1611–1667). Bihārī's highly complex couplets occupied the minds of commentators from the seventeenth century into the early twentieth; at least 37 individual commentaries have been identified.Footnote ²⁷ The Vaishnava poet Tulsidas's Rāmacaritamānas (Lake of the Deeds of Ram, circa 1600 CE), a retelling of the epic of Rāma, produced multiple traditions of commentary: treated by many Hindus of north India as a sacred scripture, the Rāmacaritamānas gave birth to distinct ‘schools’ of commentary with their own guru-disciple lineages and hermeneutical approaches.Footnote ²⁸

While the production of commentaries in Hindi on Hindi works is inextricably linked with the production of Hindi commentaries on Sanskrit works, I deal only with the latter here. I focus on the latter type in order to understand what the production of commentaries in the ‘vernacular’ on works in a ‘classical’ language can tell us about the changing relationships between those languages within the language order of early modern India and the (re-)construction of Hindi into a medium of literary and intellectual discourse. This relationship influenced the particular hermeneutic and exegetical techniques and strategies that commentators deployed, as well their works’ stated raison d’être. All of these elements suggest that these works are indeed what they claim to be—that is, commentaries and not translations.

Commentary, not translation

Modern scholarship tends to characterise vernacular commentaries on non-vernacular works in precolonial north India—not just those in Hindi, but also commentaries in other languages—as acts of translation.Footnote ²⁹ There are multiple reasons why the composers of commentaries in Hindi would not have understood their literary and intellectual activity to be one of translation. Most of these reasons relate, in one way or another, to the ‘language order’ of precolonial South Asia—a phenomenon recently discussed by Andrew Ollett. As Ollett argues, ‘what’ a language was understood to be was determined, to a great extent, by its relationships with other languages; the uses to which a given language could reasonably be put were correspondingly determined by its position within that web of relations.Footnote ³⁰ It appears that most literary composers in early modern north India (and their audiences) understood that Hindi and Sanskrit simply could not do the same things, at least not in the same manner, and therefore they could not serve the same purpose. If they were to serve the same purpose, they would have to do so in very different ways.

One aspect of this difference is reflected in contemporary notions regarding Sanskrit's status as a sacral language: Sanskrit was still, at least in some circles—and certainly among the vernacular pandits commenting on Sanskrit works—understood to be saṁskr̥ta, ‘purified’, ‘refined’, or ‘consecrated’, the ‘language of the gods’ (devavāṇī).Footnote ³¹ The fact that Sanskrit was also used for mundane or worldly (laukika) purposes, such as kāvya and praśasti (encomium), did not detract from its character as a medium for thinking about and speaking of the alaukika—the supramundane, the other-worldly, the transcendent. It was precisely the alaukika that concerned the majority of Hindi commentators discussed here, working as they were with texts on metaphysics, theology, and soteriology.

The authors of commentaries were acutely aware of their role in mediating between the realms of Sanskrit and Hindi, between the alaukika and the laukika. Bhagavānadās dramatises this role at the beginning of his commentary on the Aśvamedha Parva by first addressing the poets and sages of yore, Sarasvati (the goddess of speech), the Supreme Himself, and his guru, before ‘turning’ to address his audience—to whom he promises to convey the fame of the Pāṇḍavas as it was told by the sage Jaimini to Janamejaya, at least as much as his intellect will permit him (mati jesa / budha anumāna).Footnote ³² Bhagavānadās verbally places himself between the world of the eternal gods and sages from where the source text originated and the world of his audience in which he will make the meaning of the source text ‘apparent’ (pragaṭa) in the vernacular.

Even if authors did not necessarily recognise the sacral character of Sanskrit, they nevertheless appear to have acknowledged Sanskrit's status as the preeminent medium for thinking about and articulating alaukika matters and for pursuing scientific discourses in general. Mirza Khan, a noble at the Mughal court writing in Persian, wrote in his Tuhfat ul-Hind (Gift of India, 1676 CE) that Sanskrit is the language in which ‘books on all sorts of sciences and various arts’ are composed (kitāb-hā dar aqsām-i ʿulūm va anvāʿ-i funūn); in contrast, he reports that bhākhā, the language of ‘the world in which we live’, is used for ‘colourful poetry and description of the lover and beloved’ (asha'ār-i rangīn va vasf-i 'āshiq va ma'shūq). Commentators writing in Hindi often tackled this epistemological problem head-on by explaining why complex and subtle discourses in Sanskrit must be explained in the vernacular and qualifying exactly which element of their respective source texts they were attempting to convey (or reproduce) in their commentaries.

Caturadās presents the problem in a relatively pragmatic manner at the end of his commentary on the eleventh book of the Bhāgavata Purāṇa. He writes that the knowledge (jñāna) that Kṛṣṇa himself conveyed to the ascetic Uddhava and others was then recited by the poet Vyāsa (the composer of the Bhāgavata Purāṇa) in Sanskrit; however, ‘the meaning [of this discourse] is not illuminated’ by Vyāsa's recitation because ‘one knows [Sanskrit] if they are a pandit’ while ‘no one else understands it’. Apprehending this problem, Caturadās's guru, Santadās, bestowed upon his disciple the insight necessary to ‘expatiate the commentary’ (bhāṣā bistārī) for the ‘benefit of all people’ (saba-lokani-hita).Footnote ³³ Caturadās maintains a set of critical distinctions between the discourse itself (saṁvāda) that took place between Kṛṣṇa and his interlocutors, its meaning or significance (artha), and the knowledge (jñāna) contained within it. He does not purport to convey the discourse itself in all of its subtlety, but rather to illuminate its meaning by supplementing it with a commentary. The supplementary nature of the commentary is captured in the very idiom used to articulate its composition: bhāṣā bistārī, in which the verb bistār- (Sanskrit vistāra-) connotes ‘to spread’, ‘expand’, ‘expand upon’, ‘flesh out’. The complex and nuanced discussions of devotion (bhakti), liberation (mokṣa), and divine manifestation (avatāra) in the Sanskrit work cannot necessarily be reproduced in Hindi, but they can at least be explained.

Broader notions regarding the metaphysics of sound and language also inflected the relationship between Sanskrit and Hindi, again producing obstacles to any generalised concept of ‘translation’. This problem was particularly acute in the context of soteriological, yogic, and tantric discourses. Mantras and verbal formulae used to initiate religious disciples, induce gnosis, secure liberation, or effect change in the yogic or ‘subtle’ body were understood to achieve their ends through their sounds (śabda) and not simply through their meaning (artha). Such conceptions of ‘natural language’, to borrow a phrase from Robert Yelle, assume that language possesses ‘a direct and immediate connection to, and is therefore capable of influencing, reality’.Footnote ³⁴ Since such language is not merely denotative or referential in nature, one cannot simply substitute one sign for another—say, for example, substituting the Hindi tadbhava term kānha for Sanskrit kr̥ṣṇa in the bīja mantra or initiatory mantra of the Pushtimarg, śrī-kr̥ṣṇa śaraṇaṁ mama (‘Kr̥ṣṇa, [is] my refuge’)—even if their putative referent is the same object. (For this very reason, Sanskrit mantras appear untranslated in manuscripts of Hindi works in general, including commentaries.)Footnote ³⁵ Several of the Hindi commentators considered in this article (particularly those associated with the Dadu Panth, Niranjani Sampraday, and Ramanandi Sampraday) participated in religious traditions that emphasised the spiritual and metaphysical effectiveness of the spoken word, including in utterances spoken (or sung) in the vernacular, not just Sanskrit.Footnote ³⁶ Consequently, Hindi commentators tended to emphasise the soteriological effects of reciting their respective source texts while offering their commentaries as a means of accessing the knowledge (jñāna), semantic meaning (artha), or essence (sāra) of those source texts. For example, at the end of his commentary on the Bhagavad Gītā, Ānāndarām explains that ‘those who recite and listen to the Bhagavad Gītā with concentration achieve continuous devotion and become the servants of Hari’ and specifically enjoins his audience to recite the Gītā aloud daily (gītā pratidina uccarai). At the same time, he tells his audience to study his commentary as ‘those who recite, meditate upon, and listen to the Paramānanda Prabodhā will be made to cognise the Abode of the Lord’ (paḍhai gunai yākau sunai saucāvai prabhu dhāma).Footnote ³⁷ Ānāndarām (and his audience) recognise that reciting and listening to the Sanskrit Gītā sonically confer liberation; studying and reflecting on his commentary bring liberation through gnosis. In the logic of this and similar commentaries, the soteriological (or gnostic, or purifying, or sacralising) potential of a work and its commentary were imagined to be complementary, somewhat in the manner of the relationship between śabda and artha themselves.

The epistemological challenges inherent in the act of commenting on Sanskrit works in the vernacular were not limited to metaphysical notions regarding sound or Sanskrit's privileged status as a language of revelation; grammar itself posed an epistemological problem. Aspects of Sanskrit grammar that were not present in Hindi such as the dual number, neuter gender, and morphological cases (precolonial Hindi distinguishes between only two cases whereas Sanskrit distinguishes between eight) presented problems of both explanation and interpretation. When arguments in Sanskrit source texts proceeded from observations or arguments about grammar, Hindi commentators were forced to find a way to make those arguments intelligible.

Manoharadās faces such a challenge at the very beginning of his Vedānta Mahāvākya Bhāṣā: following the practice of some earlier commentators in Sanskrit, Manoharadās frames the question of the identity of the individual soul and the supreme soul (or God), ātmā and paramātma, as a problem of logic encoded within grammar itself. He cites the locus classicus regarding this question in a phrase from Chandogya Upaniṣad verse 6.8.7: tat tvam asi (literally, ‘you are that’). What is the nature of predication indicated by this Vedic (and therefore axiomatic) utterance? How can one thing (you) also be something else (that)? Following the example of earlier commentators, Manoharadās (who never actually translates the verse) takes recourse to the concepts of vācyārtha and lakṣyārtha, respectively the ‘expressed meaning’ and the ‘implied meaning’ of a given utterance. Here, Manoharadās is potentially faced with two problems: the first being that, if his audience is unfamiliar with Sanskrit, he must first explain the referents (artha) of the three lemmata (tat, tvam, asi) using Hindi before he can distinguish between the expressed and implied meanings of the utterance. The second potential problem concerns predication: can the elegant (but paradoxical) simplicity of ‘you are that’ be recreated in Hindi, and with similar rhetorical force? Manoharadās, a sharp thinker and gifted poet, uses this challenge as an opportunity: structuring his exposition as a fictional dialog between a guru and disciple, Manoharadās first discusses the distinction and relationship between a word (pada) and its meaning (artha).Footnote ³⁸ Having established these terms, he is able to launch directly into an analysis of the literal, expressed meaning of the phrase and its implied counterpart, using the caupaī meter:

Manoharadās cleverly manages to make multiple hermeneutical and exegetical moves, if sometimes only by allusion, within this single verse: he emphasises the ostensible duality of the referents (artha) of tat and tvam because the words (pada) themselves are clearly two; he introduces the distinction of the expressed meaning and implied meaning; he provides a (preliminary) interpretation of the expressed meaning of tat and tvam as God and the individual self; he indicates that the present-tense, second-person conjugation of the stative verb, asi, denotes a type of predication that is one of complete identity; and he introduces the idea that tat must therefore refer to a kāraṇopādhi (something ‘inflected as a cause’). Leaving aside for the moment the details of Manoharadās's hermeneutical claims, what is noteworthy is that differences in the grammatical structures of Sanskrit and Hindi compel Manoharadās to adopt particular hermeneutical techniques—techniques that render the significance of an utterance in Sanskrit legible without taking recourse to grammar—the most elevated of all sciences in Sanskrit and the most utilised tool in the Sanskrit commentator's toolbox.

The difference between Sanskrit and Hindi also generated aesthetic challenges. The question of how to reproduce aesthetic effect was no less important than the aforementioned epistemological and soteriological problems because aesthetics, and particularly the aesthetics of rasa, were foundational to the epistemological, soteriological, and/or didactic imperatives of a literary work. By the early modern period, the concept of aesthetic rasa (lit. ‘juice’, ‘essence’) had been extensively described and debated in the domains of poetics and dramaturgy in Sanskrit; in Hindi as well, purveyors of all types of literature from the late fourteenth century onward invoked rasa as the aesthetic logic behind their works or as the epistemological ‘key’ to unlocking the secrets within a text.Footnote ³⁹ Those secrets could be poetic inferences (e.g. a double entendre) or esoteric truths (e.g. the identity of the individual soul with God) and were revealed through the evocation of specific aesthetic/affective states (rasa), such as the erotic (śrṅgāra), the tragic (karuṇa), and the heroic (vīra). Within poetics and dramaturgy, these states were understood to be produced through certain essential narratological and characterological elements (alambanavibhāva), the perceptible expression of emotional states within characters (anubhāva, etc.), contributing or ‘stimulant’ factors of setting (uddīpanavibhāva), and even through the employment of particular phonemes (e.g. ‘soft’ versus ‘hard’ consonants). In the domain of musicology, rasa was understood to be evoked through the employment of particular rāgas and certain melodic techniques; in some devotional traditions (including the Dadu Panth and Niranjani Sampraday, in which several of the commentators considered here wrote), rasa was understood to itself constitute a spiritual state that was brought about through the verbal art of poetry when performed musically and within specific ritual contexts.

The priority of rasa in literary composition and reception often meant that a commentator's first responsibility was to reproduce, transmit, or otherwise evoke the rasa produced by the source text in his or her commentary; this requirement trumped considerations of narrative and even, in a strict sense, semantics. This is not to suggest that Hindi commentators ignored the content of an utterance in Sanskrit and focused exclusively on the aesthetic and affective dimensions; often, the narrative and rhetorical elements of an utterance in Sanskrit could indeed be reproduced in Hindi, in more or less the same arrangement, to produce the same or similar aesthetic effect. Yet, we repeatedly find what might be termed ‘limit cases’ in which the aforementioned isomorphism of content, form, and aesthetic effect could not be maintained across the two languages, or in which commentators deliberately argued for the ‘reading’ of a particular rasa while minimising other considerations. Take, for example, Bhagavānadās's treatment of an erotic poem by Bhartr̥hari: in his commentary, the Vairāgya Vr̥nda, Bhagavānadās first presents Bhartr̥hari's original muktaka verse:

After providing an extended exegesis of the Sanskrit verse in Hindi using the caupaī meter, Bhagavānadās summarises the ‘message’ of the verse in a couplet that that evokes, if anything, the rasa of fear (bhaya) rather than eros:

This is not a misreading; Bhagavānadās's treatment of other verses makes it clear that he possesses sufficient facility in Sanskrit to grasp the import and tone of the original verse. Indeed, in whichever recension of Bhar̥trihari's poetry Bhagavānadās was consulting, this verse would have been most likely anthologised in the śataka on the erotic (śr̥ṅgāra) rather than the śataka on detachment (vairāgya). The pleasure to be derived from the verse is generally understood to lie in Bhar̥trihari's indirect manner of praising women's glances, thus enhancing the verse's erotic effect. Yet, Bhagavānadās argues for a wholly different reading: he suggests that we should read the verse in a manner that awakens not attraction, but rather repulsion. He comments by pointing the reader toward the correct affective response (i.e. to the correct rasa); he does not translate. He makes a hermeneutical claim—in this case, an admittedly radical claim—about what the poem ‘means’ in aesthetic and affective terms.

Such techniques and strategies do not jibe neatly with theories of translation—even those that are broad and inclusive. John Cort has recently challenged scholars to reconsider translation as a topic of analysis in the context of South Asian literary history; borrowing a tripartite schema first proposed by A. K. Ramanujan, Cort's definition of translation is capacious. It includes 1) ‘iconic’ translation, in which ‘Text 1 and Text 2 have a geometrical resemblance to each other, as one triangle to another’; such translation produces ‘a text that in its intention is not independently authored’, but rather is attached to the name of the author (if any) of the ‘original’ text.Footnote ⁴¹ This, in Cort's words, is the logically impossible yet pervasive ideal of translation in European languages for the past 2,000 years. It also includes 2) ‘indexical’ translation, in which Text 1 and Text 2 share fundamental elements of plot but differ in many details of their rendering, and 3) ‘symbolic’ translation, in which Text 2 uses elements of Text 1 (such as plot and characters) to ‘say entirely new things’.Footnote ⁴² One would be hard-pressed to fit the Hindi commentaries considered here into any one of these three categories of translation. All make a clear distinction between the authors of their Sanskrit source texts and the authors of the commentarial apparatus. Plot is not a fundamental element of the majority of the source texts, which consist primarily of dialog and/or exposition.

It is therefore erroneous to speak of these and similar works in other South Asian languages as works of ‘translation’—not because their composers and their audiences supposedly ‘lacked’ a concept of translation as it is conceived in our current episteme, but because their very notions of language and language order made such a concept illogical. Cort, along with Brian Hatcher, points out that the usage of the Sanskrit-derived neologism anuvāda (lit. ‘speaking after’) for translation in several contemporary South Asian languages is of modern vintage; as Thibaut d'Hubert has argued, its earlier hermeneutical sense referred more to a process of ‘semantic unfolding’ (vr̥tti).Footnote ⁴³ Contemporary theories of translation can help us to identify and understand practices of translation in the precolonial subcontinent, yet I would caution against extending the use of translation as an analytic category to the archive of precolonial commentaries. The preceding discussion should make it clear that commentators writing in Hindi did not believe that the type of isomorphism (of language, meaning, aesthetic, etc.) alluded to in Ramanujan's definition of translation was necessarily possible or even desirable in the context of languages such as Sanskrit and Hindi.

So what did these commentators think they were doing? Put succinctly, they thought that they were commenting upon or responding to an existing text, which is to say that they were supplementing the original text. I have already noted the supplementary nature of the commentary as encoded in idiom and usage: a composer ‘expands’ or ‘fleshes out’ a commentary (bhāṣā vistār-). Sometimes, commentators used the metaphor of illumination to characterise their labour: for example, Bhagavānadās writes that a desire arose within him to ‘illuminate the meaning’ of Bhartr̥hari's work (grantha aratha parakāsa kūṁ antara upajī prīti). If we look closely at the grammar and syntax of commentators’ descriptions of their works, we find that sometimes the phrases cited by modern scholars as indicative of translational activity (in the idiom of ‘making [something] vernacular’) are, in fact, assertions of commentarial activity. The phrase in question is bhāṣā kar-, which is sometimes read as a phrasal verb meaning ‘to (re)make in the vernacular’, with the source text (in Sanskrit, or Prakrit, or other language) as its direct object; for example, the phrase tinahiṁ grantha bhāṣā kiyo is rendered in English as ‘they did the text in the vernacular’ (i.e. ‘they rendered the text in the vernacular’).Footnote ⁴⁴ I suggest that, in commentarial works composed in precolonial Hindi, the phrase bhāṣā kar- should often be understood in a different manner. In such contexts, the term bhāṣā denotes a ‘commentary’. This appears to be an assimilation of two terms: Sanskrit bhāṣya (‘an explanatory work, exposition, explanation, commentary’) and bhāṣā (‘common or vernacular speech’).Footnote ⁴⁵

This bhāṣā should be understood to be the direct object of the verb kar-; in other words, bhāṣā kar- means ‘to make a commentary’—specifically a commentary in the vernacular. Evidence for the usage of bhāṣā as bhāṣya is found in scribal colophons, which often alternate between using bhāṣya and bhāṣā to refer to a commentarial work. Yet, the most convincing evidence is found in the composers’ own utterances regarding their labour. For example, though Theganāth appears (to the modern reader) to render the Bhagavad Gītā in its entirety into Hindi in the manner of an iconic translation, he describes his work in the following words:

One could possibly take ‘the eighteen chapters of the Gītā’ to be the object of the phrase bhāṣā kari but the syntax and poetic line break make such a reading exceedingly unlikely: the ‘eighteen chapters’ are part of a relative–correlative pair concluded in the same line (‘are difficult to recite’).Footnote ⁴⁶ The phrase ‘thegunātha bhāṣā kari kahai’ consequently reads as ‘Theganāth, having made a commentary, recited [it]’.

Similarly, Caturadās's description of his treatment of the eleventh chapter of the Bhāgavata Purāṇa leaves little room for doubt regarding his meaning:

No mention of the Bhāgavata Purāṇa precedes or follows the verse, making it unlikely that the term yaha functions as a demonstrative pronoun; instead, it functions as an adjective: ‘this commentary’. This reading is supported by the opening and closing verses of the work, in which Caturadās outlines the origin and transmission of his source text, the Bhāgavata Purāṇa, as well as the origin of his own work, and gives a sense of the relationship between the two texts. He writes that the contents of the Bhāgavata were originally told by Śrī Bhagavān (Viṣṇu), the supreme godhead, to the creator deity Brahma, from whom it passed through the divine sage Nārada, the sage Vyāsa, Vyāsa's son Śuka, and on to King Parīkṣit. ‘It is that very thread (sūtra) that I draw out now,’ writes Caturadās, making a pun of the terms sūtra (thread, aphorism) and bisatārai (to draw out, expand) to suggest that he is both carrying on the thread of transmission of the Bhāgavata Purāṇa while ‘expanding upon’ its meaning—literally ‘drawing out’ its meaning through commentary.Footnote ⁴⁸

Commentarial strategies

If we assume that the purpose of a Hindi commentary was not to ‘stand in’ for its Sanskrit source text, but rather to respond to it, then we can begin to differentiate between different types of responses—in other words, between different commentarial strategies and techniques. A given utterance in the source text could invite one or more types of responses from the commentator—which is why different Hindi commentaries on the same Sanskrit source text can appear so distinct from one another in terms of form and purpose. Hindi commentaries are, in a certain sense, in dialog with their source texts: they not only explain or interpret the source text—‘this is what the source text means’—but also amplify it, defend it, validate it, mollify it, or even mock it. The particular techniques that a commentator deploys (gloss, paraphrase, summary, example, exegesis) depends on the type of response that he or she is articulating.

Commentaries on Bhartr̥hari's muktaka verses provide a revealing example. Indrajit opens his commentary with a verse in Sanskrit stating that he engages ‘in thoughtful intellectual discourse’ comprising both ‘Bhartr̥hari's own utterances and glosses (ṭippaṇī)’ for ‘the benefit of others’; his commentary is thus ostensibly an explanation of Bhartr̥hari's utterances.Footnote ⁴⁹ He accordingly adopts a structure in which he presents each muktaka verse followed by glosses of individual words and phrases in the vernacular. Take, for example, his treatment of a verse on the futility of reasoning with fools:

Indrajit was reasonably proficient in Sanskrit—there is ample evidence of this proficiency in the Vivekadīpikā itself—but, here and elsewhere in the work, his ‘glosses’ (ṭippaṇī) are not always straightforward statements of the semantic meaning of the terms and phrases to which they correspond. On the one hand, Indrajit is careful to gloss salilaṁ (water) using the oblique case in Hindi (jala-hi), making clear the syntactical relationship between water and the optative verb pibeta (‘may drink’). On the other hand, he glosses the adverbial yatnataḥ (‘with effort, assiduously’) in the manner of a noun: ‘a person who labors’. Indrajit treats the adjectival pipāsārditaḥ (afflicted by thirst) and pratiniviṣṭa (obstinate) in a similar fashion. This apparent discrepancy makes sense when we observe the structure and purpose of the commentary as a whole: Indrajit does not gloss every word and phrase of the original verse, meaning that one would require at least some familiarity with Sanskrit grammar and lexicon in order to understand the verse. Keeping his Sanskrit-literate audience in mind, Indrajit neither parses terms nor gives their definitions, strictly speaking; instead, he does a little bit of both in the process of stating the import or significance of the utterance. In this example, Bhartr̥hari employs his characteristic deadpan irony and the optative mood to suggest (through contrast) the impossibility of changing the mind of a fool: ‘Squeezing with great effort, one may even obtain oil from sand.’ Indrajit gives his audience a sense of this irony (and contrast) by using the subjunctive: ‘A man who exerts great effort shall obtain oil from sand.’ The difference is subtle but important: Indrajit's gloss is not so much an analysis of the individual parts of the original utterance as it is a restatement of the utterance in different terms. In form, the gloss resembles the homiletic style of exegesis employed by contemporary religious poets in the vernacular.Footnote ⁵⁰

Bhagavānadās responds to Bhartr̥hari's verses in a different manner. The poet comments only on Bhartr̥hari's verses regarding detachment—for the most part, those anthologised in the Vairāgya-śataka—and tells his audience that his primary concern is with ‘illuminating the meaning’ of those verses as they relate to the three degrees of detachment (i.e. manda, tīvra, and taratīvra). Though Bhagavānadās denies having any poetic skill (humility was traditionally a sign of a good poet), he insists that he has ‘wrought delightful aphorisms without doing harm to [Bhartr̥hari's] original’; those among his audience who are wise and possess a superior intellect ‘will perceive the light of the original work in the commentary’ (mūla hāṁni kīnhī nahī karyau suvāka vilāsa / vāsa vudhi bhāṣā lakhaiṁ paṇḍita mūla prakāsa).Footnote ⁵¹ Bhagavānadās's didactic imperative is thus to inculcate in his audience a state of detachment from worldly pleasures, but he accomplishes this by producing an experience of aesthetic pleasure (vilāsa). The poet accordingly dispenses with glossing or parsing and instead responds to each of Bhartr̥hari's muktakas with several of his own verses, which summarise, expound upon, and amplify the original.

By way of example, let us return to his treatment of Bhartr̥hari's verse regarding the eyes of women. As discussed above, Bhagavānadās responds to Bhartr̥hari's erotic verse with the following couplet:

This is hardly a summary of the contents of Sanskrit verse, much less a translation. Bhagavānadās is instead performing a reading of Bhartr̥hari's verse: though ostensibly an erotic poem that expresses wonderment at the power of women's glances, Bhagavānadās suggests that it should be read as an indictment of women's ability to generate destructive desires in the hearts of men. Bhagavānadās's stated purpose is, after all, to ‘illuminate’ the meanings of Bhartr̥hari's verses as they relate to detachment. Bhagavānadās also amplifies what he understands to be the didactic import of Bhartr̥hari's words by composing three caupaī verses in which he extends Bhartr̥hari's list of the things that women's eyes accomplish into an exhaustive pathology of erotic desire:

Whereas Bhartr̥hari 's verse lists six things that women's eyes do and closes with a rhetorical question (‘What indeed do the eyes of women not do?’), Bhagavānadās lists eight men—or, rather, eight scenarios in which a woman seduces men. The subject has shifted from the woman's eyes to the woman herself (her eyes are only one of the instruments through which she effects her charms) and the conclusion is not a rhetorical expression of awe, but a blunt diagnosis of damnation. True to his stated purpose, Bhagavānadās attempts to inculcate in his listener a revulsion toward sexual pleasure and women through offering a different kind of pleasure: that of well-wrought poetry (suvāka vilāsa). Leaving aside the question of whether Indrajit's and Bhagavānadās's respective readings of Bhartr̥hari are ‘good’ readings, we can clearly see that their different purposes and aims shape the strategies and techniques that they employ in their commentaries. Each adopted the analytical and rhetorical tools and textual structure that suited his immediate task.

Even when commentators are silent regarding their didactic, critical, literary, or aesthetic aims, the structure and style of their commentaries reveal important aspects of their intellectual projects. This is reflected in the diversity of commentarial approaches to the Bhagavad Gītā. The earliest commentator on the Gītā, Theganāth, does speak of the aims and context of his work, telling his audience that he has composed his commentary to address the questions and doubts of his patron, Prince Bhānu of Gwalior. His commentary takes the form of a verse-by-verse restatement of the text in Hindi; however, as Parmeswaran has noted, Theganāth dilates on certain verses (using multiple caupaī verses to restate the argument of a single verse of the original) while glossing over the details of certain other verses, reflecting his own pedagogical concerns as a Nāth monk teaching a Rajput prince.

In contrast, Ānāndarām (who does not explicitly articulate any intellectual or interpretive programme) presents each verse of the original work followed by a prose exegesis in which he makes explicit the dialogical context of the verse (i.e. who is speaking to whom), identifies the referents of deictic terms in the source text, and provides an explanation of epithets, titles, and references to Puranic terms. At the end of this exegesis, he composes a couplet that restates the main ideas of the verse. Ānāndarām's commentary circulated among vernacular devotional communities of north India and the structure of his commentary would appear to fit this pedagogical context: the guru or monk reciting it would first recite the original verse, then give a detailed explanation of its meaning and significance, and, once his audience had understood it, conclude with a pithy and memorable restatement of the verse in a meter used for vernacular aphorisms (the dohā).

The anonymously authored commentary in Ms. 14901 of the Rajasthan Oriental Research Institute takes yet another approach: its author quotes only the first verse of the Gītā, after which they launch into a condensed prose summary of the dialog between Krishna and Arjun. Paratextual indications of the speaker (e.g. kr̥ṣṇa uvāca, ‘Krishna said …’) and of the chapter make it possible to collate the prose summary with the verses of the original work while the synoptic character of the commentary makes it unlikely that the commentary was used on its own; one would still need to read the Gītā itself to learn its essential ideas and arguments but this Hindi commentary would explain the significance of the more complex of those arguments for a practising Vaishnava in simple, unadorned language. Finally, the anonymously authored commentary in Ms. 16052 of the Rajasthan Oriental Research Institute contains a verse-by-verse restatement of the source text in dohā meter; like Theganāth, the commentator expands upon certain verses while omitting the details of others, reflecting an overriding concern with Vaishnava theology and soteriology.

Intellectual programmes explicit and implied

Even this synoptic account of the diversity of commentarial structures, strategies, and techniques will give a sense of how different intellectual imperatives led to the adoption of distinct commentarial styles and modes. Yet, there is more to be said about the broader intellectual, literary, and aesthetic programmes of which these commentaries were just one expression. Some commentators gesture toward these programmes in their discussions of their labour. We saw earlier how Caturadās declared in his commentary on the eleventh book of the Bhāgavata Purāṇa that he had composed the work at the behest of his guru ‘for the benefit of the world’ (articulated as both lokahitārtha and saba lokāni hita). This world is imagined in both cosmological and social terms: the loka is both ‘this world’ (saṁsāra) and the world of the volk or ‘folk’—the undifferentiated mass of all social classes and creeds. The problem, as articulated by Caturadās, is that the wisdom of the Bhāgavata Purāṇa is locked away in Sanskrit, inaccessible to anyone who is not a ‘pandit’; by composing his commentary in the vernacular, he makes the knowledge contained within this scripture available ‘to all’. This would indeed appear to support the scholarly argument that the rise of vernacular literatures in South Asia was driven by an egalitarian impulse to democratise access to knowledge systems.

Yet, this is not the case for all the commentaries considered here. For example, though Prince Indrajit writes (in Sanskrit) that he composed his commentary on Bhartr̥hari's śatakas ‘for the benefit of others’ (paropakāraya), those ‘others’ constitute a delimited community of discerning listeners (vivekanāṁ śrotr̥manaḥ); indeed, as noted above, a listener or reader would have required at least an elementary knowledge of Sanskrit in order to make use of Indrajit's commentary. Most often, commentators begin (or end) their works with an appeal to the kovid (learned) and the kavi (poets) to fairly evaluate their writing and amend any errors or infelicities; Indrajit, Bhagavānadās, Manoharadās, and even Theganāth all address the kavi and kovid directly in their commentaries. This invocation of the community of the learned reminds us that, in early modern north India, the appreciation of literature—and indeed the right to access literary and scholastic works—was a matter entailing substantial literary training, socialisation into the community of connoisseurs, and the cultivation of specific modes of intellectual and affective behaviour. In more ‘courtly’ milieus, this was the domain of the rasika or sahr̥daya (connoisseur); in monastic and religious contexts, this was the space of the sādhu jana (true people) or satsaṅga (the company of the good). Commentators sometimes made explicit the qualifications required of their readers; for example, Bhagavānadās ends his Vairāgya Vr̥nda with a lengthy discussion of the intellectual and behavioural traits that a student must demonstrate in order to have the adhikāra (qualification, authority, claim) to read his commentary.

These commentators were part of a broader project that sought to re-form the vernacular of Hindi into a language capable of conveying intellectual and literary discourses. The strongest evidence of their participation in the construction of a ‘cosmopolitan vernacular’ is perhaps found in what they did not do—or, more specifically, in the terms and concepts that they chose not to ‘translate’ or explain in their writings. For example, what is most striking about Manoharadās's Vedānta Mahāvākya Bhāṣā and Ṣaṭapraśanottara—both commentaries on works dealing with metaphysics—is the commentator's resistance to translating or explaining the abundance of technical terms that appear. The lexicons employed in the source texts of these works draw from the philosophical traditions of Sāṁkhya, Advaita Vedānta, and Mīmāṁsa, as well as the highly technical lexicons of philology and linguistics in Sanskrit. And yet, we have seen earlier how Manoharadās uses terms such as vācyārtha, lakṣyārtha, and kāraṇopādhi in his Vedānta Mahāvākya Bhāṣā without providing any glosses: the tatsama terms, taken directly from Sanskrit and without any type of phonological or morphological change, simply appear in the Hindi commentary without comment. Authors such as Manoharadās (as well as Theganāth, Caturadās, Bhagavānadās, Harirāmadās, and Ānāndarām) imported technical lexicons into Hindi by employing these lexicons in their commentaries whilst, in the process, making new types of discourse possible in the vernacular. At the turn of the seventeenth century—the moment that the efflorescence of commentarial writing in Hindi began—the Hindi literary corpus was rich in the genres of the lyric, romance, epic, and epigrammatic poetry but only beginning to produce scientific works (śāstra) on topics such as statecraft, philosophy, rhetoric, economics, agriculture, and sex.Footnote ⁵² The writing of Hindi commentaries on scientific works in Sanskrit helped to quickly grow these emerging genres in the vernacular.

How was the emerging class of vernacular intellectuals—a community that included courtly poets, ministers, gurus, monks, and even merchants who were familiar with Sanskrit but more comfortable composing in the vernacular—to produce the various technical lexicons that would be necessary to discuss such topics? The short answer is that they borrowed those lexicons from Sanskrit (and, in some cases, Persian), deploying them in their Hindi commentaries. When Caturadās composed his commentary on the eleventh canto of the Bhāgavata Purāṇa in 1635—a conceptually dense section of the work dealing with cosmology, metaphysics, theology, and soteriology in terms taken from Sāṁkhya and yoga—he adopted a clear and unadorned style but used nearly all the same technical terms as the original: this included ontological terms such as prakr̥ti, mahātattva, and sāttvika, and compounds such as bhogāsakti (self-indulgence) and indriyasukha (pleasure derived from the senses). Although occasional instances of some of these terms may be found in preceding Hindi literature, this is the first instance of their being used in a consistent and technical fashion—and is indeed the first instance in which a number of these terms appear in a written Hindi work at all. Just as the lexicon of, say, literary theory in English is being constantly broadened and enriched through the adoption of terms from other languages (e.g. différance, gestalt, parataxis, mise en scène, dramatis personæ, and so forth), commentators of the sixteenth century imported terms and concepts from preceding ‘classical’ traditions, broadening the conceptual and imaginary horizon of the vernacular in the process.Footnote ⁵³ By the turn of the eighteenth century, thanks to the efforts of commentators and their audiences, intellectuals working in Hindi could boast of a robust literature and lexicon in which matters of philosophy, theology, literary criticism, medicine, and statecraft were being discussed in the vernacular.

The production of Hindi commentaries on Sanskrit works was thus not only about producing new lexicons, but also about producing new genres and discourses that had not previously existed in the vernacular. The poets, scholars, monks, and aficionados that gradually and collectively constructed the traditions of rhetoric, metaphysics, sexology, veterinary science, and the like did not create these traditions ex-nihilo; most often, they created them by explaining and responding to these traditions as they already existed in the classical language of Sanskrit—but using the vernacular. The awkwardness of characterising this project as one of translation should now be clear: how were the Hindi writers of the sixteenth and early seventeenth centuries to translate works for which the necessary vocabulary did not exist in Hindi (let alone the generic and stylistic conventions critical to helping readers make sense of the text)?

What made possible the importation of lexicons, conventions, and genres from Sanskrit into Hindi was the bilingual literary and intellectual culture of the writers themselves: by expanding the discussion of Sanskrit works and knowledge systems beyond the limits of the Sanskrit language itself, they ‘pulled’ those elements into the domain of the vernacular. This phenomenon relates to, but is somewhat distinct from, Pollock's concept of ‘literarization’ as a process in which the vernacular becomes a language capable of supporting literary and intellectual discourse through ‘written uses of language for expressive purposes that came into being by emulation of superposed models of literature’.Footnote ⁵⁴ In this formulation, the pioneers of writing in a vernacular emulate the genres, forms, styles, and conventions of the formerly hegemonic literary culture, recreating these elements in their works. In the case of Hindi commentaries on Sanskrit works, emulation is certainly at work; Hindi writers were adapting the commentarial genre itself from Sanskrit and often mimicked the conventions of Sanskrit commentary. Yet, these writers were not reproducing or recreating discourses from Sanskrit in the vernacular in such a manner that the latter necessarily displaced or ‘stood in for’ the former; in the case of several works considered here, the commentary was unintelligible without its Sanskrit source text. The meaning and significance of the vernacular text were always to be understood in relation to an antecedent Sanskrit work, and the task of explicating that relationship was the job of the commentator and whoever recited or performed the commentarial text. As Karin Barber has written, the use of unfamiliar terms and references in a text are often structural features that are intended to cue or prompt hermeneutic and exegetical performance.Footnote ⁵⁵ In the case of precolonial Hindi commentaries on Sanskrit works, the appearance in the Hindi commentary of an unglossed and unfamiliar technical term from Sanskrit such as lakṣyārtha or kāraṇopādhi or bhogāsakti acted as a cue for the reciter—most often a guru, teacher, or ritual professional—to expound upon the term and its attendant conceptual context for his or her audience.Footnote ⁵⁶ This was the oral, performative labour through which the lexicon and conceptual universe of Hindi language and literature were expanded from the late sixteenth to early nineteenth centuries and through which an audience or reading community for these works was gradually produced.

An early modern renaissance

The explosion of commentarial activity in the vernacular during the seventeenth century is part of what one might call north India's early modern renaissance. I use the term ‘renaissance’ in two specific senses: as a ‘revival’ of classical models of thought and literature, and as a (distinctly contemporary) paradigm for understanding a historical period.Footnote ⁵⁷ The efflorescence of literary, scholastic, and artistic activity that occurred from the sixteenth century (the late Sultanate period) until the early nineteenth (the end of the Mughal imperium and beginning of the colonial episteme) was characterised by an engagement with ‘classical’ works of the Sanskrit cannon; however, this engagement took place in the vernacular. Classical traditions in north India—including those of Sanskrit and Prakrit, but also Arabic and Persian—enjoyed greater continuity over the medieval period than their European counterparts (Latin and Greek) and so the early modern north Indian renaissance was less about ‘recovering’ knowledge systems from antiquity and more about adapting and redeploying those systems in the vernacular. Not unlike Dante Alighieri, Guido Cavalcanti, and Francesco Petrarch, who studied classical languages while composing new works in the vernacular, all of the Hindi commentators considered here were scholars of a classical language who composed in the vernacular. In addition to writing commentaries, Indrajit, Manoharadās, and Bhagavānadās (such as Dante, Cavalcanti, and Petrarch) composed vernacular works in several genres, including lyric poetry, narrative poetry, and scholastic treatises.Footnote ⁵⁸ Merchants and mercantilism played a critical role in the north Indian renaissance: the merchant-monks of the Niranjani Sampraday and Dadu Panth (the same orders of which Manoharadās, Bhagavānadās, and Caturadās were members), patrons such as the famed Jagat Seth of Murshidabad, and laymen litterateurs and philosophers such as the Jain merchant Banārasīdās (1595–1663) reflect the engagement and investment of merchant classes (and call to mind figures such as Giovanni Boccacio and Cosimo de Medici of the Italian renaissance). Manuscript evidence tells us that the Hindi commentaries considered in this article were read as much by wealthy merchants as they were by princes or monks.

The value of calling this movement in north Indian literature and scholarship a ‘renaissance’ lies in the way in which we characterise early modern Indian intellectuals’ relationship with the past. The study of works in Sanskrit never disappeared from north India but, beginning in the mid-sixteenth century, the character of that study began to change. Reading in Sanskrit and writing in the vernacular became the norm for pandits, gurus, monks, and even merchants, and this new practice provided a fillip to vernacular literary production and a new reason to study the classics: in the hands of a skilled commentator, old works could be put to new uses. An anthology of epigrams such as the śatakas of Bhartr̥hari could be used as the basis for a theory of renunciation, the Aśvamedha Parva of the Mahābhārata could be mobilised as the foundation for an excursus on non-dualism, and the Bhagavad Gītā could be made to speak to the exploits of the puranic Kr̥ṣṇa of Braj or even to the guru-worship of the Nāths. The spread of literacy in the vernacular ultimately produced more potential readers for Sanskrit. This overlapped with the increased interest among Persophone elites in Sanskrit works during the Mughal period.Footnote ⁵⁹

The monks, merchants, and princes who composed Hindi commentaries on Sanskrit works cultivated a knowledge of the classical language and studied ancient works in a mode that was not merely antiquarian in character—they were deeply concerned with how past thinkers and texts could provide answers to contemporary questions and provide inspiration and material for emerging vernacular genres. Bhagavānadās wrote at the end of his Vairāgya Śataka that his commentary turns the stagnant well water of scholarly discourse (śāstra-artha so kūpa-jala) into a flowing river (silatā syandha). This sentiment was shared by many of his fellow commentators and reflected an awareness of the fact that, even as they were enriching the conceptual and literary lexicon of the vernacular with their commentaries, they were also bringing new life to earlier works and ideas. This was not ‘emulation’ in a derivative sense, but rather a fresh response to the classics that had occupied their predecessors for centuries.

The term ‘renaissance’ (especially ‘the Renaissance’) is not without conceptual baggage: historians of the European Renaissance have widely critiqued the analytical usefulness of the term for understanding the period now more commonly known as ‘early modernity’, pointing out that its associations with rebirth and revival have as much, if not more, to do with the anxieties and aspirations of nineteenth-century historiographers as they do with the character of the period itself. Yet, ‘the Renaissance’ has proven to be a remarkably durable concept in modern scholarship.Footnote ⁶⁰ This is also the case for the period of Hindi literature considered here: beginning in 1889 with the publication of Indologist George Grierson's Modern Vernacular Literature of Hindustan, scholars of the late nineteenth and early twentieth centuries consistently referred to the sixteenth and seventeenth centuries as the ‘Augustan’ or ‘golden age’ (svarṇa-kāl) of Hindi literature.Footnote ⁶¹ Writing on the Vaishnava poetry of the period, the twentieth-century literary historian Rāmacandra Śukla emphasised the role of Hindi authors in ‘awakening’ Hindus and Muslims alike through their fresh expressions of ancient devotional (bhakti) themes and motifs, and their success in ‘bringing Hindi literature to maturity’.Footnote ⁶² Śukla and his fellow nationalist historians were explicitly concerned with recovering the ‘classics’ of Hindi literature from obscurity; these classics were themselves imagined to reflect a recovery of earlier traditions in Sanskrit, especially the purāṇās.Footnote ⁶³ This characterisation of the sixteenth and seventeenth centuries as a kind of (implicitly Hindu) renaissance has persisted in Hindi scholarship.

Early modern north India did witness a renaissance, but not exactly of the sort imagined by Śukla and his contemporaries. This renaissance was not limited to Vaishnava authors and material; nor was it limited to the language of Hindi. Driven by ‘middle castes’ such as merchant and scribal communities and by other groups that had previously stood at the peripheries of Sanskrit intellectual and ritual culture, this renewal of classical literature and learning took place at the meeting of Sanskrit with Hindi, Persian, and other languages used by these upwardly mobile groups.