The Quotidian Revolution: Vernacularization, Religion, and the Premodern Public Sphere in India

A commonplace understanding in scholarship and popular memory about the Yadava century is that the state supported the production of Marathi literature. As we will see, this is a claim made largely through circumstantial associations: the assumption is that because Marathi literature emerged during the latter half of this period, it must have enjoyed the support of the Yadava state. On the contrary, we have reasons to believe the Yadavas were quite indifferent to Marathi, which they appeared to consider a language of the quotidian world, though that indifference may be better described as a benign ambivalence. They may have enjoyed Marathi, even literary Marathi, at court or in private, but they have left no evidence of state funds directed toward its production or patronage, and it has no place in the Brahminic ecumene discussed in chapter 1. There may be precursors here to what will become, in the modern period, “linguistic nationalism” linking Marathi and the Yadava state, and we may see here rudiments of the way in which an “imagined community” will be formed in Marathi in the centuries to come, particularly under the Marathas, the Peshwas, in the colonial period, and in the formation of the modern state of Maharashtra. Yet here, to read too much of this history into the past will be to risk missing the truly startling innovations of this period, to see this period only as a trough for the present moment. My aim in this chapter is to scrutinize how Marathi was viewed through Yadava materials—inscriptions primarily. In this process, I hope to show how and why the Yadavas evinced a benign ambivalence toward Marathi, but also to show how the writing of Marathi and the culture it accessed did serve the Yadavas and their political-social aims. In particular, we will see how what I view as a fear of the public was present in inscriptions that gave gifts of revenue and land to particular individuals. At the same time, we will note how this fear of the public could be expressed also as a valorization of it, as is particularly witnessed in inscriptions by the Yadavas around the famous Vitthal Temple in Pandharpur.

This chapter will show that, precisely because the Yadavas did not officially patronize Marathi as a literary medium, vernacularization could flourish outside royal courts and within the public cultures of the Yadava realm, leading to the ironic assertion that by not engaging in patronage of Marathi literature, the Yadavas enabled its development within the relatively stable context of their rule. Here I will pursue the question of the place of Marathi within the Yadava century, for in understanding how Marathi was viewed in the elite spheres of court and the Brahminic ecumene, we can better understand the cultural politics that surrounded using Marathi as a literary language in this period.

Our primary archival materials are Marathi inscriptions from the eleventh to early fourteenth centuries. This is, of course, a written archive, but it is a very unique form of writing. Inscriptions, I believe, are public declarations, for the most part. By this I mean that they are social texts, expressed outwardly and propelled into time and space. In this sense inscriptions are less like manuscripts and more like genres of public performance; they are examples of a self-conscious literary public performance. Though inscriptions may be intended as communication among literate elites, they are still available for public observation. Even if they are situated in private spaces, they are composed with materials meant to endure over time and certainly transcend the lives of individuals and even the spans of kingdoms. They are “workly texts,” to use Pollock’s phrase, in terms of the actualization of their proclamations, but they also do work well beyond the plans of their authors or agents.¹ They may share with manuscripts a workly conveyance of information, but they are situated in space in a much different way. Inscriptions do their work because they are social texts and public performances.

What makes them “public” is the way in which they are made available for the attention of an audience that happens to be present by virtue of social forces in a given region or space. For example, an inscription may be outside a temple, directed toward the “outer” world of people that might be oriented, for any number of reasons, to a given temple at a given moment. And an inscription might be inside a temple, a space circumscribed by sect, gender, caste, and class, yet also available to the public formed by admission to the temple—those who may enter a temple or care to do so. Alexis Sanderson, in his recent study of inscriptions and what they yield about the “mundane realities” of the world of classical Shaivism refers to inscriptions as within “the public domain.” In a sense, an inscription is a text without a cover.

Sanskrit texts are circumscribed by a tight circuit of attention, the “Sanskrit cosmopolis.” This is a literary sphere spread wide by geography, but narrowly construed by sex, caste, and class. Conversely, the field of the “vernacular language” is tightly circumscribed by region, but heterogeneous by sex, caste, and class in that region. It is easy to imagine that, given the individual empowerments of these two language spheres, a thriving group of amateur Sanskritists has always existed on the subcontinent and elsewhere, ready to “vernacularize” the “cosmopolitan” on the fly and orally.² As Pollock notes, “listening to rather than reading literature long remained the principal mode of experiencing it,”³ and this engenders what I think of as a kind of aural literacy. Indeed, it is possible that just such an informal culture of orally translating Sanskrit inscriptions and other texts into regional languages is part of the impulse felt by figures like Jnandev. In any case, it is my assumption that inscriptions, in whatever language they may appear, were not consumed simply by reading but by being heard and being translated, in situ, throughout the subcontinent, including within the Yadava realm. And when inscriptions are in a language largely intelligible to everyone in a given region, like Marathi in Maharashtra, then it is likely that the public performative nature of inscriptions is routine and the cultures of interpretation and conveyance that accompany them are expansive.

I do not think this fact was lost on the people who composed inscriptions in thirteenth-century Maharashtra. I believe they recognized the general public that attended to their decrees, and they used language to signal particular kinds of symbolic capital and not necessarily to exclude particular audiences. In other words, a public proclamation in any language is ultimately conveyed in content and affect to the world around it, one way or another, and often by means of aural restatement. For these reasons, I treat all inscriptions as publicly intelligible, even if they are in Sanskrit, a language unknown to most of the population of the quotidian Yadava world. I also understand these inscriptions to be intended for a public audience and so I read in them one layer of address: to the public around the inscriptions themselves. I am not claiming that an imagined public is the primary subject of address—in most cases it is certainly other elites. But I am asserting that those who composed or ordered these inscriptions knew that they could not control who would encounter this message and thus always wrote also with this general public in mind. To search for this recognition of the open nature of inscriptional discourse is one important way to write the history of publics in the premodern world and, in particular, to search for vernacular publics and even a nascent public sphere well before the printing press or the Irani café in India. It is the ever present force of an observing public that is one factor in the composition of inscriptions. And while most inscriptions do not explicitly address this public, I will examine those, in Marathi for the most part, that do, for if Yadava-era public culture was primarily a Marathi-speaking culture, then Yadava-era inscriptions in Marathi speak to that public culture explicitly. And at the earliest layer of the public address in inscriptions in Marathi, we find the curious case of the donkey curse.

Among the Yadavas, as I mentioned in chapter 1, Marathi was not the primary language, but rather the tertiary language of expression, trailing Kannada and Sanskrit in number of inscriptions. Yet we can glean from the use of Marathi how the language was understood by elites within the Yadava century. The use of the language clearly indicates a population that aurally understood Marathi (as well as Kannada and perhaps Sanskrit), and this strongly suggests a Marathi public culture in place, and perhaps even a Marathi literary sphere as well—these are, after all, inscriptions, even if most of the population of the Yadava realm would not have read them but rather heard them read aloud. In addition, the way in which Marathi appears in inscriptions is often to reiterate a proclamation already articulated in Kannada or Sanskrit. Thus the use of Marathi is, in inscriptions, often a replication of those parts of the inscription in Kannada or Sanskrit. However, one interesting exception to this practice is to be found in the realm of the curse. It is here that we get a glimpse of the way elites within the Yadava court and the Brahminic ecumene viewed Marathi.

The curse in general is an ancient feature of Sanskrit, and other inscriptions throughout the “Sanskrit cosmopolis,” and it appears in multiple vernaculars, as it does in Marathi.⁴ For example at the conclusion of an inscription from Amba Jogai from 1144 CE, granting funds for a Brahmin community of ascetics, a rather typical curse in Marathi is given—“upon him who severs or removes this agreement the lightning bolt of the Yogini will fall.”⁵ Another grant, at Savargav, for the upkeep of an Amba Devi Temple, likely in the care of Brahmins, in 1164 CE ends with this threat: “Remove [this decree] and you’ll become a dog, a village beast.”⁶ At the conclusion of a grant of 1202 CE (the “Jaitra Savanta” copperplate inscription perhaps of Jaituga I’s reign, 1191–1210 CE) reinforcing the taxation rights of a Brahmin agrahara, we have two curses. The first declares, “The one who takes even one cow, one coin, or one finger’s worth of land will attain hell until the end of the world,” and the other states, “whoever defiles this agreement, is a dog, a donkey, a Chandal [Untouchable].”⁷ An inscription in Velapur of 1300 CE, which records the donation of funds for the upkeep of a Brahmin monastery, states succinctly, “the protector [of this decree] wins heaven, the one who doesn’t protect [this decree] goes to hell.”⁸ Such curses are preeminent examples, particularly within the field of Yadava-era inscriptions in Marathi, of Pollock’s idea of a “workly text” that functions as a “speech act” rather than as a simple act of documentation.⁹

The most often repeated curse in Marathi in these inscriptions is also the single most repeated phrase in Marathi in any inscriptions of the Yadavas, and so it demands some attention here. This is the perennial “ass” or “donkey” curse. The curse is not unique to Marathi or Maharashtra, and is a mainstay of Indic epigraphy and the Sanskrit cosmopolis in general, where it appears in Sanskrit and other languages. But the curse has an important and unique place in Yadava-era inscriptions. It is not only the most common curse, it is actually the single most common phrase in Marathi across the inscriptional record.

The donkey curse warns that a donkey would have forced sexual intercourse with the mother of anyone who would contravene a gift of state.¹⁰ In the Marathi-speaking regions of the twelfth to fifteenth centuries, the donkey curse appeared regularly in inscriptions across at least four polities, from the Shilaharas and the Yadavas, to the Bahmani Sultanate, and in the Vijaynagar period under Harihara and Devaraya.¹¹ Often along with the curse would also appear a graphic illustration etched into the stone, called a gadhegal. In Marathi, gāḍha or gāḍhav means “donkey” and gal indicates either a “stone” (gaḷa)¹² or a “profanity,” (gāḷa) related to the common term for a “swear word” in northern India, gāli. These gadhegal then are “donkey curses” etched on stone, and so they are often called “donkey curse stones” in English. They are found throughout Maharashtra and Goa (see figures 2.1 and 2.2). Importantly, as S. G. Tulpule points out, this curse was invariably in Marathi, not Sanskrit or Kannada or any other language of inscription. The fact that this curse appeared only in Marathi, and with such regularity, is a key fact in understanding the value of Marathi in the Yadava century. Sometimes the verbal curse would be absent, and only the graphic illustration would stand as implied imprecation: further evidence of the visual, aural, and material performative work of inscriptions.¹³ In these contexts it would appear following a certain decree in an inscription.

Our oldest example of the donkey curse is implied rather than written. The Akshi inscription (which Tulpule, as noted, places in 1012 CE, against the opinion of some other scholars) is the oldest extant example we have of the visual depiction of the donkey curse (figure 2.1). Importantly, this also is probably the oldest example we have of Marathi, according to S. G. Tulpule, as noted earlier. This point is important to highlight, for while the Akshi stone does not record the donkey curse textually, it depicts the curse graphically along the bottom half. Because we know that the verbal curse in Marathi will become de rigueur in the subsequent century, we should note here the link between written Marathi (probably the first instance of it here in this inscription) and the curse (the first instance of its visual depiction in the region). There seems to be an important link between Marathi and the donkey curse.

A similar stone, which carries an inscription in Marathi with the image but not the language of the curse, is found on a stone in Paral (better known as Parel, in central Mumbai), attributed to the Shilahara reign of Anantadeva (figure 2.2).

The first time the written donkey curse is joined to its visual depiction is also from a Shilahara inscription in Parel, but of 1184 CE, and it is important to note that the curse here is the only section of the inscription in Marathi.¹⁴ The inscription in Marathi reads: “Now, whomever contradicts this order, the blade of Lord Vaidyanath will fall upon his family. A donkey will have sex with his mother.”¹⁵

An example of this link between curse, image, and Marathi in the era of the Yadava king Ramachandra can be found on a second Akshi stone, with a date of 1291 CE, and this stone bears both the image and the donkey curse as verbal formula (figure 2.3).

Indeed, Maharashtra has many such stones or gadhegal, especially along the Konkan coast in the former areas of the Shilahara polity, which became a vassal state to the Yadavas in the thirteenth century.¹⁶ In most cases a curse of any sort, and especially the donkey curse, would follow a donation to a Brahmin or for a Brahminic activity, that is, a donation in support of the Brahminic ecumene, as the examples above indicate.¹⁷ A recently discovered gadhegal (without the verbal curse) in the Raigad region of Maharashtra, dated to 1254 CE, bears a gift of the Shilahara king, likely of land, to two Brahmins (Ram Mandalik and Ganapati Nayak).¹⁸ The frequency of the verbal appearance of this curse increased significantly during the reign of Ramachandra, which is the period during and after Marathi literarization. The donkey curse, then, is the most commonly repeated single phrase in Marathi inscriptions across four centuries, especially in the later phase of the Yadava century, generally invoked to protect a donation to the Brahminic ecumene. It is not by chance that the most elite and prestigious of royal gifts is protected by the most vulgar of curses.

The donkey curse—as inscription and graphic depiction appended to donations to Brahmins (in the main)—both suggests the importance of such donations and reveals an anxiety over the stability of the social, political, and economic order those inscriptions fueled.¹⁹ It also tells us something about how Marathi was understood in this period with relation to gender and “class” in a sense. I should note that the penalty for contravening such grants, particularly those offered by the state or its agents, was likely something other than what is portrayed in the donkey curse. We have records of the penal regimes of the Yadavas, and the donkey curse is not mentioned as actualized consequence.²⁰ The curse is purely formulaic, as curses are, yet also indicative of anxiety. But anxiety about what?

Several theories have been presented about the function of the donkey curse and the gadhegal image. Scholars who study them most commonly refer them to as “boundary markers.” These scholars note that the stones exist usually at the periphery of a territory given as a gift in the inscription.²¹ R. C. Dhere understood the image to serve a semiotic function within a folk-belief system, understanding the female figure to represent divine nature and the donkey to represent the Goddess of smallpox and pestilence.²² Dalal, Kale, and Poojari, in their recent study of four newly discovered gadhegals suggest that the stones, while they may bear such a symbolic weight, also conveyed the threat of an actual punishment. However, as noted, there is no evidence of this kind of punishment in the Yadava period in other texts, and it seems very unlikely to be a real punishment.²³ It also seems unlikely that these stones were worshipped, while vīragal or “hero stones” were objects of veneration, as Dhere has so amply demonstrated. While both theories are possible, they do not provide a general theory for the imprecation (that could include those noted previously regarding yoginis, worms, etc.), nor do they account for the visual and aural power of the donkey curse.

As an alternative to these two ideas, I situate the donkey curse within the context of the larger “evil eye” prophylaxis, a feature of the global history of human culture, but one with particular relevance here to our examination of a medieval Marathi quotidian “public.” The evil eye and the various prophylactic responses to it, from talismans to rituals and invocations, appear to be an almost panhuman convention from ancient times to the contemporary.²⁴ One essential feature of the evil eye belief complex is the idea that a public display of one’s good fortune—such as a donation or other acquirement—risks attracting the envy of others. This envy then causes something bad to happen to the recipient of good fortune. The evil eye prophylaxis wards off both the initial envy as well as the resulting bad fortune. The way the prophylaxis works is by diverting the envious gaze, with a distracting phrase or charm, but often with a vision of something else, something out of place, unexpected, or even outrageous. In India today this is commonly done by dressing baby boys as girls, giving the boys temporary female names, or simply marking a spot of “lamp black” or kājal on their person or as an outline around their eyes—all of which are intended to divert the general, free-floating malevolency of “envy” and hence the “evil eye” from alighting on a boy child.²⁵ In the various recipes for evil eye prophylaxis, the oral, aural, and visual combine to ward off the threat.

I argue that the gadhegal depictions and verbal donkey curse inscriptions are a form of evil eye prophylaxis.²⁶ These are “outrageous” depictions in image and word that both seek to distract the attention of the reader (or more likely listener) from the material of the donation to the possible curse that awaits the person who envies or acts with envy. The punishment depicted in the donkey curse is shocking and peculiar, but not actualized. It is an “empty” threat, and so we should treat it as a symbolic one, as evil eye countermeasures are. Such countermeasures exist in the register of the semiotic, even if they are believed to have real-world effects as well.

The donkey curse and the gadhegal ward off envy, and, in so doing, these practices reveal a recognition of the threat of the public that surrounds inscriptions. All evil eye cultural forms exist in this context of the fear of the public, of public envy and desire. It is in social spaces, out in the open, that a person is susceptible to the evil eye, and this is the same social space for the inscriptions the donkey curse—text and image—accompanies. The donkey curse, as a representative example of an attempt to ward off the evil eye, indicates not only a recognition of an attentive public, but a fear of that public, in the mind of the originator of the inscription.

There is more to the donkey curse, other social effects that are also related to evil eye prophylaxis. As a device that draws away envious attention and intent, evil eye prophylaxis must also invite interest and catch a person’s attention. The visual depiction of the gadhegal similarly invites interest, and thus the material inscribed on the stone is communicated and “advertised.” Though I find nothing humorous in this image or curse, the figure of the donkey, and the idea (threat, description, etc.) of someone having sex with another person’s mother are common features of South Asian humor today, as they are in Western humor, and likely over several thousand years.²⁷ The “humor” is of course operationalized through the violent objectification of women and a world of threat directed toward a man’s female relatives, the object of his patriarchal “protection,” but also possession. This aspect of the function of the curse perhaps also reveals a predominantly male audience, a gendered audience that the works associated with Chakradhar and Jnandev will seek to transcend. It is possible to read in the donkey curse, and in other curses as well, the intention of crude humor (though the image depicted is violent and not humorous, in my opinion). Taken this way, the donkey curse image and text with its shocking (or humorous, to some) message serves the dual purposes of deflecting the envious gaze by distraction, while at the same time drawing attention to the material of the main proclamation. As Melanie Dean has pointed out in her work on contemporary evil eye (tiruṣṭi) practices in Tamil Nadu, the evil eye prophylaxis not only protects against the natural and supernatural effects of the evil eye but also serves to display the relative benefit and gain of the person in the first place—it attracts and repels simultaneously and by the same rationale. As Dean puts it, such practices represent and resolve a “dialectical tension between the desire to display one’s wealth to one’s peers and the fear that the visual attention of one’s peers will cause harm.”²⁸ The visual and written donkey curses similarly resolve this dialectic as they both attract and repel attention; they mediate the attention of the public, channeling it in culturally appropriate ways. The fear of the public is challenged by the violence of this graphic curse, while the curse also serves to index the value of the gift—for why would one need a curse to ward off envy if it was not an enviable gift in the first place?

Given the “folk” nature of the evil eye belief complex, it seems likely that the historical origins of the donkey curse in South Asia are similarly within the field of the “folk,” of the quotidian world. Though the first examples of the ass curse outside of Marathi are in Sanskrit and from around the turn of the Common Era, we might assume that Sanskrit absorbs into its inscriptional authority this “vernacular” tradition of warding off “evil.” While Sanskrit may exist “as the language of the Gods” within an elite literary sphere above and beyond the quotidian, the inscription is the exemplary moment when Sanskrit comes down to earth, when a verbal formula is expressed to the public, etched on a stone for all to see and hear read aloud, likely drawn from the cultural idioms of a region, not the literary tropes of the Sanskrit cosmopolis. For the work of the Sanskrit inscription, like all inscriptions, is to produce a monodirectional proclamation into the quotidian sphere, into the space of a region, amid a local people, and within everyday life where the rule of law that undergirds gifts of state is always tenuously maintained. The donkey curse was perhaps a way to translate yet further, from language to idiom, to put the “sacred” or inviolable aspect of the content of the inscription into terms a local, non-Sanskritic population could better understand. The inscription speaks to many audiences, such as other class and caste elites and members of the Brahminic ecumene, as well as elite aspirants to such gifts of state. And it speaks into the future as a “legal” record of a gift meant to endure for years. But inscriptions also exist within the world of “mere attention,” as Michael Warner described the sphere of publics, and they occur in social spaces that are difficult to mediate or that go almost entirely unmediated.²⁹ The donkey curse, preserved only in the regional language in the case of Maharashtra, and in its highly colloquial register bordering on slang, shows us that those who inscribed this curse also sought to direct this speech act to the general population going about their everyday life. The public is unmistakably one of the objects of address in the donkey curse as text or image, and it seeks to ward off a fear of the public. Bourdieu, in his analysis of state formation, makes a perceptive and relevant observation: “Force is always on the side of the governed, only opinion can sustain the governors.”³⁰ The evil eye that the donkey curse sought to ward off is none other than the surveillance (and possible ill opinion) of the state by its subjects, the governed.

If we search for any kind of “special” use of Marathi in the Yadava period by the state or the Brahminic ecumene, we find virtually nothing in the spheres of the literary, philosophical, poetic, or aesthetic. So the fact that the donkey curse in Yadava-era inscriptions is only in Marathi, never in Sanskrit or Kannada, suggests to us that the Yadavas and the Brahminic ecumene they supported regarded Marathi as a language suitable for threatening and cajoling the general population, the masses of quotidian life.³¹ We see that they patronized Marathi as an inscriptional language, a tertiary linguistic medium to convey the most pungent of their curses because it was the language of the everyday. But, aside from the inscriptional record, do we see evidence for the support of Marathi in the Brahminic ecumene, at the Yadava royal court, or in any other context that might indicate Yadava patronage?

One of the enduring assertions in modern historiography about the Yadava era—and one that provides the clearest link between that era and the contemporary period—is the idea that the Yadavas supported and encouraged Marathi literary production. Many scholars—writing in Marathi and English³²—have made this claim, and it becomes common by the mid twentieth century when, for example, the famous scholar of Maharashtra, Irawati Karve, writes in the Maharashtra State Gazetteer issued by the state Government of Maharashtra in 1968 that Marathi was patronized by the Yadavas.³³ Yet the historical sources of this idea are fuzzy at best. Karve, for example, cites only the fact that Jnandev praises Ramachandra at the conclusion of his Jñāneśvarī, which actually tells us nothing of what Ramachandra or any other Yadava potentate may have supported or why Jnandev might have included such a tribute. In the absence of any definitive inscriptional or textual record issued from the Yadava state, scholars have tended to rely on such circumstantial evidence to substantiate the claim that the Yadavas were patrons of Marathi and Marathi literary production.

Perhaps the scholarly origin, or at least one prominent origin, of this assumption is found in the pioneering work of V. L. Bhave (1871–1926), whose Mahārāṣṭra Sāraswat is the first grand history of Maharashtra and Marathi literary culture and a text that is foundational in Marathi cultural historiography, a kind of source text and touchstone for all the major historians of Marathi literature who followed, both Indian and non-Indian. Bhave began publishing the work that would form Mahārāṣṭra Sāraswat around 1898–1899 in serialized form and this culminated in its first publication in full as a monograph in 1924.³⁴ To my knowledge, Bhave’s text is the earliest and most powerful (and most often cited) articulation of the idea that the Yadava court supported the production of literary Marathi.

In this work Bhave recalls secondary accounts and proofs of Marathi’s literary existence through texts and references to children’s folktales in Marathi as well as citations of women’s folk songs in Marathi in such texts as Someshwar III’s Sanskrit Mānasollāsa (c. 1138 CE).³⁵ While none of the texts mentioned in such secondary sources are available to us today, I think Bhave was right to take references to such compositions as evidence that a kind of oral literary Marathi was likely in place before the actual literarization of the language, before its written form, in the late thirteenth century. He asserts that among the quotidian masses, whom he calls “Vaishyas, Shudras, and others” (a phrase that he may have drawn from Jñāneśvarī, as we will see), a vibrant Marathi folk-story culture existed in the century before the first historical record of Marathi literary work. Through such references, Bhave establishes that Marathi was a spoken-poetic language, likely even a literary language, that contained “texts,” either oral or written, well before the advent of the Līḷācaritra or the Jñāneśvarī, which is a historical observation I believe is likely true.

After establishing the existence of some kind of Marathi literary-oral sphere by the twelfth century, Bhave connects this sphere to the royal court of the Yadavas by reference to a story in the Līḷācaritra that recalls a chance meeting between Chakradhar and the Yadava king Krishna at Lonar.³⁶ In this meeting, the king attempts to give money to Chakradhar, but Chakradhar refuses to accept the money or indulge the king, leaving the king’s presence.³⁷ Krishna then uses the money to build a wall at the Kumareshwar Temple in Lonar. Bhave implies that Krishna’s interest in such figures moved his son, Ramachandra, to patronize Marathi at his court.³⁸ Here, as with the reference to Jnandev’s vote of thanks to Ramachandra, we have circumstantial evidence at best, and none that suggests actual patronage expressed by a Yadava ruler. Even in the case of Chakradhar, if we accept this encounter with the Yadava king Krishna as historical fact, it merely establishes a connection between a king and a particularly prominent saintly figure. The endeavors of many such saintly figures were indeed financed through royal patronage—a phenomenon we will note in greater detail when we discuss the Līḷācaritra. Furthermore, Chakradhar was not a figure of literary Marathi in this story: we should recall that he did not compose any Marathi literature of any kind—the Līḷācaritra was the work of his followers within a decade after Chakradhar’s departure from Maharashtra. And so there can be no question that the story of the Yadava king Krishna’s attempt to give Chakradhar money had any relationship to Marathi literary work, or at least extant work. This was an attempt to give a gift of state to a Brahmin mendicant, not a Marathi literary figure.

The only connection Bhave makes between the Yadava court and Marathi literary production during or before the time of the Līḷācaritra and Jñāneśvarī is through recourse to Mukundaraja and his Marathi texts, the Vivekasindhu in particular. Marathi scholarship long held this text to be the first example of literary Marathi. However, by the mid-1960s a general consensus formed around a later date for this text. The argument that Mukundaraja lived in the twelfth century and that the colophon of his Vivekasindhu gave the date of composition as 1188 CE are highly suspect and are discounted by most scholars; the only manuscript with a colophon gives the date of transcription as 1656 CE. Indeed, there is ample evidence that Mukundaraja and his text come after Chakradhar and appear to constitute a kind of offshoot of Chakradhar’s own teacher lineage (guru paraṃparā). Thus Mukundaraja’s dates would be either coterminous with Chakradhar’s dates or, far more likely, at least one generation later.³⁹ Despite the fact that most linguistic historians understand Mukundaraja’s text to be from the fourteenth century at the earliest (which better matches its style of Marathi), it is still common to find this text cited as the “first” instance of Marathi literature.⁴⁰ Bhave’s text thus situates Mukundaraja, whom he considers the first Marathi literary author, producing the first Marathi literary work, at the court of the Yadava general and feudatory ruler Jaitrapala (r. c. 1192–1202 CE)⁴¹— both this connection and this time frame have been challenged and emended by successive Marathi scholarship over the last five decades. It is highly unlikely that the Vivekasindhu is a text that precedes either the Līḷācaritra or Jñāneśvarī.

A second claim made by some scholars involves another story from the Mahanubhavs that remembers a Brahmin poet named Narendra who, in 1291 CE, composed an early work called the Rukminī Svayaṃvara or “Rukmini’s Choice of Groom.”⁴² Indeed, the story of Rukmini’s elopement and marriage to Vitthal/Krishna forms a minor literary genre, not just among the Mahanubhavs, but among the Varkaris as well (and outside of Maharashtra), likely signaling a popular topic of public performance in the Yadava period.⁴³ The fourteenth-century Mahanubhav text Smṛtisthaḷa recalls that Narendra and his two brothers were poets who recited their poems for the Yadava king Ramachandra.⁴⁴ Narendra composed a poem on the subject of Krishna’s wife, Rukmini, and her choice (svayamvara) of Krishna for a husband, and this poem, in Marathi, so pleased Ramachandra that he offered to buy the rights (essentially to buy “authorship”) of the poem from Narendra. However, Narendra declined, saying that to take money from the king would tarnish the reputation of his community of poets (kavikula).⁴⁵ The story tells us that it is after this experience that Narendra goes to Bhatobas and becomes a Mahanubhav, bringing his poem with him.⁴⁶

As with all hagiographical work, there is reason to be suspicious of its historical claims, particularly claims that portray relatively ordinary figures meeting kings and other temporal rulers.⁴⁷ However, as I will detail in the next chapter, the Mahanubhavs in particular sought to record the lives of their leaders—Chakradhar, Bhatobas, and Gundam Raul especially—with great historical detail, in the mode of literary realism to be sure, and the Smṛtisthaḷa verse that records this episode is just such a text, wedded to at least the trope of historical accuracy. If we take this as historical fact, even heuristically, it is still unclear what “courtly patronage” might entail. We have no corroboration in inscriptional or other evidence issued by the Yadava state, though one can imagine, in an informal context of courtly entertainment such moments might have happened with a high degree of frequency but with little documentation. Indeed, my argument that there is no evidence for Yadava courtly patronage of Marathi even approximating their lavish patronage of Sanskrit is not also an argument that the Yadavas did not speak Marathi or consume courtly entertainment through the medium of Marathi. On the contrary, I think there is every reason to believe the Yadavas entertained themselves in every way possible, including with performances of Marathi plays, poetry, and other pleasures, as well as perhaps the same in Kannada, Sanskrit, or even other languages.

However, this recollection in the Smṛtisthaḷa does not record state patronage of Marathi, but rather the impulsive offer of a highly impressed Yadava king. Furthermore, Narendra rejects Ramachandra’s offer—the “courtly patronage” actually does not happen, even in the story. Obviously, scholars who invoke this passage do so to suggest that if Narendra was offered money for his poem, though he refused it, it is likely other Marathi composers were likewise offered such rewards and accepted them. Yet Narendra’s response is curious, and it undercuts such an assumption for his rejection is not based on an idiosyncratic position but on a position he attributes to all his fellow poets as well. For Narendra tells the king that he must reject his reward, for to take it would denigrate all Marathi poets, it would “cast blame” upon them—that is, people would begin to speak badly about his kavikula, his “community of poets.” Narendra is not yet a Mahanubhav either, though the text tells us he had heard Bhatobas speak. Still, his community of poets must reference other professional poets, perhaps limited to his family, but more inclusive of all poets composing Marathi poetry in oral public culture (and perhaps written literary culture) at the time.⁴⁸

What is it exactly that Narendra refuses to sell? The text of the Smṛtisthaḷa tells us that it is authorship of the poem, the work itself—“the work’s ascription,” the entire copyright, as it were (granthācā abhaṅgu)—that Narendra refuses to sell to Ramachandra.⁴⁹ Therefore, if this story tells us something in general about Marathi poets and kings of the Yadava era, it is that Marathi poets refused to accept courtly patronage and sell their authorship of poems to kings. So if anything is corroborated by the lack of inscriptions offering support for Marathi literary production, it is this: that Marathi authors rejected courtly patronage even if (when) it was offered. This story, rather than describing courtly patronage of literary Marathi, is a story that tells us that writers of Marathi literature may have functioned outside the context of court as a matter of principle; they sought their “patronage” elsewhere. Whether or not this was the case historically, my point is that to use this particular example to establish that the Yadava court patronized literary Marathi is also to misread this very example.

My discussion here involves the literary record of Marathi, but it does not answer the question of what might constitute the first “work of Marathi literature” in some absolute sense. The first work of Marathi literature that we have is the Līḷācaritra, but that text, alongside the Jñāneśvarī, suggests to me, in any case, that Marathi has long had a literature, well before these two texts were composed. And this literature may have been oral primarily—perhaps even a fixed oral form in various cases, an aural literary form—or it may have been written, but we no longer have that written record. Yet it is hard for me to imagine the Līḷācaritra or the Jñāneśvarī emerged sui generis, and I am very suspicious of such declarations of “firsts” in history in any case. However, it is beyond my ability to prove the existence of any text before these two or to establish a field of oral Marathi literature before the late thirteenth century. Despite this, it is my own conviction that such literature—likely not only orally preserved but also written—did exist. Yet I am restricted by my sources here, and so I treat the Līḷācaritra and the Jñāneśvarī as the first two works of Marathi literature, as a statement of current historical contingency.

Another common claim that the Yadavas supported literary Marathi points to the inscriptional evidence of gifts for the maintenance of the Vitthal Temple in Pandharpur made by various rulers of this dynasty. These inscriptional gifts are preserved in a set of three inscriptions from: a) 1189–1190 CE; b) 1273–1277 CE; and c) 1311 CE.⁵⁰ None of these inscriptions denotes funds to support Marathi literary production, it should be stated.⁵¹ However, they do reveal the cultivation of the veneration of the deity Vitthal at his temple in Pandharpur by the Yadava state and establish the recognition by the state of a large group of devotees who gathered there, and thus the state’s sense of the power of Vitthal devotionalism among the general population of the region. Yadava interest in the Vitthal temple may have been a strategy to unsettle the claims of their rivals, the Hoysalas, over the same temple and community. We can see that the Hoysala king Someshwar in 1237 CE and at other times supported the Vitthal temple with gifts of state, making it clear that Vitthal devotion was not the sole purview of the Yadavas (or of the Marathi-speaking world). As R. C. Dhere has noted, the broad swath of polities taking on the mantle of “Yadava” in this period and region all appear to worship the deity of Pandharpur—another commonality with, in particular, the political rivals of the Yadava dynasty.⁵²

In order to explore with care the relationship between the Yadavas, Vitthal worship, bhakti, and literature, we must parse from the materials we have the role of literature, particularly the bhakti literature that will come to form a core of Marathi literature in the centuries to come, through the figures of Jnandev, Namdev, Eknath, and Tukaram, among many others. While the Yadavas made gifts to the Vitthal Temple, there is no mention of patronage for literary works, nor support of Jnandev and Namdev, the two earliest literary figures claimed by the Varkari faith. What do we find in these inscriptions, I will argue, is the recognition of what I have called a “bhakti public,” a social sense of a general devotion shared by many people in the region yet not confined to any given sect, religious order, or explicit community. Though later the religious community we now call the Varkaris will form, at this period—in these inscriptions and also in the Līḷācaritra and the Jñāneśvarī—there is no mention of such a community. Still, a public culture of worship clearly surrounded the Vitthal temple, and this is what we glean from the inscriptions of gifts in Pandharpur. Observing these gifts to Vitthal’s temple provides a rubric for understanding the devotional public culture at large, the general sphere from which our key texts will emerge.

Scholars will often connect the support of Pandharpur, Vitthal’s temple, and Vitthal’s devotees by the Yadava court to the devotional literature that marks the Varkari religion in the centuries that follow. This is a flawed assumption for several reasons. The only work attributed to Varkari figures that can be even conjecturally applied to this period or to Pandharpur would be the work attributed to Jnandev and Namdev. Namdev makes no mention of the Yadavas, and the Yadavas make no mention of Namdev, and so it is far too tenuous to draw a link between these two, quite aside from the significant problems with locating Namdev in history.⁵³ What is more, as I’ve pointed out in other work, Namdev disavows literacy, the conceits of literature, and the very idea that writing is useful.⁵⁴ In addition, the dates suggested for Namdev’s life—1270–1350—overlap with the outer limit of the historical period covered here. For these reasons, and for others, it does not seem that Yadava patronage of the Vitthal Temple can be connected to what will later become the literary song-poem corpus associated with Namdev and his companions, even if we accept his traditional dates of life as historical fact.

The case of Jnandev, likewise, carries significant difficulties. Jnandev makes no mention of Vitthal or Pandharpur in the Jñāneśvarī, and we hear of Vitthal only in later abhangs associated with him, which many scholars consider significantly later than the Jñāneśvarī, perhaps by a century or more.⁵⁵ In other words, the Jnandev of the Jñāneśvarī does not present himself to be a Varkari, much less a devotee of Vitthal.⁵⁶ I am not stating that Jnandev was not a Varkari—he is presented since at least the sixteenth century as a Varkari and devotee of Vitthal, Varkaris revere his texts and memory, and hence Jnandev is undoubtedly a Varkari. Instead, I am simply stating that he did not explicitly present himself as such in the Jñāneśvarī. Varkaris and Jnandev’s own followers acknowledge that between the time of the Jñāneśvarī and the later abhangs of Jnandev, a time when he comes to encounter key figures, like Namdev, and other low-caste and female bhaktas, Jnandev undergoes a change of heart and a conversion to the worship of Vitthal. The story of this “conversion” is attributed to Namdev, in a set of biographical and autobiographical songs called Tīrthāvaḷī or “the journey.”⁵⁷ In part, this story and its effect on Jnandev’s public memory will be noted in the conclusion. I should also state here that Jnandev also never mentions Namdev in any text attributed to him, to my knowledge.⁵⁸

My point is neither to challenge the strong Varkari belief that Jnandev was a Varkari (which he was, without a doubt) nor the idea that Jnandev and Namdev were companions (they certainly were), but instead to show that it is anachronistic to read direct Yadava support for the Vitthal Temple in Pandharpur throughout the thirteenth century as support for literary production in Marathi in the idiom of Vitthal bhakti that will flourish in the fourteenth century under these names. It is quite possible, even probable, that the support shown for the Vitthal Temple by several states and polities in the thirteenth century, especially by the Yadavas, spurred the creation of Marathi literature in subsequent centuries. Indeed, the root of this support is important to note here, for it registers a keen awareness by the Yadava state (and others, such as the Hoysalas) of the vitality of an emerging quotidian Marathi public, one major focal point of which was the bhakti world that surrounded Vitthal and Pandharpur. And so while support for the Vitthal Temple and the worship of Vitthal by the Yadavas from 1189 to 1311 CE cannot indicate support for the Marathi bhakti literature that would later be associated with Pandharpur, Vitthal, and the Varkaris, it does tell us what made it possible, in part, for the Varkari tradition to flourish in social and literary worlds, and for Jnandev to be enfolded into the Varkari religion.

The three sets of inscriptions that we have from the Yadava era recording gifts to Pandharpur all indicate an interesting sociopolitical story, one that connects the Yadava state with public culture, bhakti, and religion, drawing the Vitthal Temple and Pandharpur into a dynastic vision of society in a way quite unique in this period. The historical beginning and end of the Yadava dynasty are literally marked by state donations in Marathi to the Vitthal Temple in Pandharpur, and so it seems inevitable to read these Marathi inscriptions as gifts to a temple for the sake of “religion” as also statements about statecraft, soft power, and the recognition of a public culture both active and powerful within the Yadava century.

The first donation recorded was given in 1189 CE, attributed to the Yadava king Bhillama—often considered the founder of the sovereign Yadava dynasty, as mentioned earlier.⁵⁹ Here is a translation of the readable portions of the inscription:

The donors in addition to Bhillama are listed as the mahajans or Brahmin elite of the region. The donors, including the king Bhillama and Brahmin elites, claim this gift to the Vitthal Temple is for the devaparivār, the “family of the God” as a whole, which would include themselves, as donors, but also implies the general public of devotion.⁶¹ The inscription refers to the Vitthal Temple in Pandharpur as a “small temple” (lān maḍu), and the grant is given to enlarge the temple, support the devotees of Vitthal, and expand the worship of Vitthal.⁶² This inscription, like many others, extolled the virtue of giving to the temple, where one’s karma would increase by such a gift or decrease in the absence of giving. Many scholars have referred to this inscription in order to demonstrate that the worship of Vitthal and the temple in Pandharpur grew from a small, pastoral deity site to a major temple complex under the care of the Yadavas (but also the Hoysalas).

Why invest in this “small temple”? The reasons are several. Politically, Pandharpur exists in a region of contention, especially between the Yadavas and the Hoysalas, and the town has suffered the wrath of warring polities because of its location through the centuries. Bhillama’s support for the temple and for Pandharpur is also a political signal to the Hoysalas and others, a way to claim territory that is both geographic and cultural. Indeed, another inscription some decades later in 1237 CE describes the temple as lying within the realm of the Hoysalas, suggesting that Pandharpur was at the center of land under dispute and unrest throughout the thirteenth century, as it was in the century after.⁶³ Yet it is also a reflection of its popularity in the everyday world of the emergent Yadava realm that this modest temple had a sufficiently grand number of devotees to attract such an investment of state funds. The bhakti sphere that surrounded the Vitthal Temple, marked by this inscription, is one piece in the larger growth of a devotional, Marathi-based field of public religious practice, a field in which Jnandev and Chakradhar, though neither expressed devotion to Vitthal in the texts of this study, are yet to be situated. And though the word bhakti does not appear in a Marathi inscription anywhere at this point in time, to my knowledge, we can still see that the devaparivar circumscribes a devotional social sphere of elites and non-elites alike who shared devotion to Vitthal.

As I have argued elsewhere, bhakti fundamentally produces publics, contexts of social “sharing” (a root meaning of the word), and interaction.⁶⁴ As Jack Hawley has argued, bhakti also produces “networks” that link people, places, and politics.⁶⁵ The social work of bhakti is apparent here as a fulcrum around which social cohesion adheres, and thus it is an opportunity to invest both cultural and economic capital in the growth of a potent symbol. But we should also note that the Yadavas supported many other temples with far more lavish funds. And, as we will see, this was not a one-time investment, but a first installment in a relationship of financial and cultural capital that would endure until the end of Yadava rule.

Beginning in 1273 and lasting through a long series of inscriptions of gifts, the last dated to 1277, we have our second major donation in Pandharpur. This is a Yadava-era donation, from the reign of Ramachandra, but unlike the inscription of 1189 CE, issued by the Yadava king Bhillama, this one has a vast assembly of donors, with each name and date of donation listed serially. This is important to note. Often, when scholars invoke this set of inscriptions, it is referred to as a Yadava gift, or the donation of Ramachandra or his minister Hemadri. Yet the names of these two august figures are rather muted in the litany of donors. The donation is actually recorded in a substantial set of inscriptions—seven in total—with long lists of donors offering funds for the growth and maintenance of the Pandharpur Temple to Vitthal.⁶⁶ The name of Ramachandra is but one of the many names given here, preceded by the name of Hemadri. Hemadri is donor number twenty-three in the list, and Ramachandra number twenty-four.⁶⁷

The inaugural statement of the inscription that precedes the long list of names is not clearly legible, but it appears to contain names and specifics about donations. However, the second portion in Sanskrit is clearer and it reads: “I pay joyful homage to Vitthal, who is endlessly praised by sages and Gods, known in the Vedas as supreme, the lotus-eyed beloved of Shri about whom celestials happily sing pleasing songs, who is celebrated by the fervent populace of devotees (bhaktajana) with bristling hairs and eyes filled with happy tears as they honor the son of Nanda, who looks after the joy of his devoted attendants (bhaktānugam).”⁶⁸ The Marathi inscription follows immediately after this and it reads: “In the auspicious holy year of 1195 [1273], called Shrimukh, details are given here of various [amounts of] money [donated] by the leaders of devotion (bhaktimāliāṃ) to Lord Vitthal of Pandhapur for flower garlands [to be given] three times a day for as long as the sun and moon [exist].”⁶⁹ As the first inscription of 1189 CE suggested, giving to the temple and to the devotees who come to Pandharpur—here in 1273 called bhaktajana in Sanskrit—was a way to accrue merit and, to be sure, having one’s name inscribed in stone ensured both a public and a personal register of that merit.

After a long line of names and gifts, we arrive at Ramachandra’s gift, which reads: “King Ramachandra, who is the chief of Vitthal’s congregation in Pandharpur (pāṇḍrīphaḍmuṣya), who is the bearer of tradition, of the powerful majestic auspicious Yadava Narayana dynasty, who is the universal monarch (cakravārttiṃ), he will bestow a well for the All Powerful, Victorious, King, who is the Chief Deity of the Holy Temple, Shrimangala Mahashri Lord Shri Vitthal.”⁷⁰ Even Ramachandra’s terms of self-aggrandizement, the very bread and butter of the prashasti or royal panegyric form, are still balanced by an equally august set of titles for Vitthal. And included in Ramachandra’s lofty description, first and foremost, is not the idea that he is a king among men but that he is a chief among devotees. What is interesting about this inscription is the way the presence of primary figures of the Yadava state—Hemadri and Ramachandra—is relatively downplayed, nestled as their gifts are within a long list of others. These two key figures of Yadava rule are enumerated among many other elites and enfolded within the general world of the bhaktajana, the people of devotion to Vitthal. It seems the prime reference for donation in these inscriptions is devotion or bhakti itself.

To my knowledge, these are some of the first instances, in 1273, of the words bhakti and bhakta appearing in inscriptions in the region, in Marathi or Sanskrit.⁷¹ These instances conjoin bhakti and bhakta in compounds that express companionship and sociality, as in mali or “leader,” in the Marathi inscription, and anuga or “attendant” and jana or “people” in the Sanskrit one. Thus the very first explicit appearances of the idea of bhakti in inscriptions in the Yadava era invoke social organization.

The donors are identified in the Marathi inscription as bhaktimali, the leaders of devotion. The phrase is odd—the text may read bhaktamali instead—but the point is essentially the same: the donors who have given to the Vitthal Temple are the luminaries and benefactors not only of the temple but rather of bhakti itself, of public devotion. The individuality and singular attention usually given to a donor in inscriptions is not a convention we find in the donations inscribed to the Pandharpur temple around 1273 CE, and this is unique. One gets the impression that patronage of the devotionalism, the bhakti, associated with Vitthal, is the driving force conveyed in the inscriptions. This suggests a conscientious effort to valorize a general public, the bhaktajana, or the people who make up the field of bhakti, the devotees of Vitthal, in whose name, in some fundamental sense, these inscriptions exist.

The growth of the Vitthal Temple between 1189 and 1273 reflects not only an increase in the popularity of Vitthal bhakti in the Yadava realms and throughout the region but also the rise of the political usefulness of Pandharpur and its large quotidian populace of devotees.⁷² These inscriptions align the political state with public devotionalism, and while this alignment does not indicate support for literary Marathi production, it does reveal support for the quotidian world of bhakti, the field in which the Marathi public sphere will first emerge. Though offered to a private temple, this gift identifies a kind of “public good,” a benefit to which all are entitled. I say this not to assert that the Pandharpur temple was public in a modern sense. It is not. Access was restricted in all the usual ways, and the funds given would have been in the control of the temple’s Brahmin proprietors.⁷³ But there is something unique here. Most gifts to temples in the Yadava period carefully name and enumerate the people and jatis that received funds and rights to ceremonies. The invocation of a general bhaktajana in this massive inscription is both anomalous and suggestive.

A final inscription of a donation to the temple in Pandharpur is made in 1311 CE, several years after the Delhi Sultanate had captured Devgiri (1294) and the year that Malik Kafur finally made subject the Yadava dynasty, the same year in which Ramachandra died and his successor, Singhana III, mounted an ill-fated rebellion against the Sultanate army in the region.⁷⁴ It is unclear who has given this last donation of the Yadava reign to Pandharpur, however.⁷⁵ The inscription records a gift to Pandharpur and especially, again, to aid in the service of the bhagatajana, the “people of devotion.”⁷⁶ Here is a translation of the relevant portions of the inscription:

This inscription does a lot of work. We have some continuity from our first two inscriptions in that a devotional public is again invoked. We hear of the bhagatajana or bhaktajana, “the people of devotion,” that surrounds Vitthal and Pandharpur, and a “family of God,” a devaparivar, is cited again as well, likely indicating the community of devotees. We also have the peculiar description of the first donor as a bhaktikinkar, that is to say, a servant of devotion or bhakti itself, though this may be an orthographic mistake. Yet it would reinforce the reading in the inscription of 1273 of a bhaktimali, which may also contain a reference to bhakti itself. Bhakti, here, appears as a fully formed concept linking theology and sociology, a belief attached to a population making of it a social unit. The gift also invokes the memory of Pundalik, who is a hoary figure of Pandharpur credited with convincing Vitthal into staying in Pandharpur by displaying his deep and unending devotion (bhakti) to his parents.⁸⁰ Pundalik is the quintessential figure of the common man in bhakti spheres, a son whose filial devotion demands he ignore even the deity Krishna standing at his threshold, whom he tells to wait until his interminable piety to his own parents is done. Finally, the inscription also notes that it devotes its funds to the “constant happiness [and] support of all people” (sadāprasanna samujjīvailoka), a direct invocation, to my mind, of the public at large.

The last lines of the inscription are particularly curious to note in relation to the outline of a concept of a public culture in these edicts. The inscription invokes the term Maratha. We have the word Maharashtra attested several decades earlier, in the Līḷācaritra and elsewhere, as a designation of place, the Marathi-speaking region, as we will see again in subsequent centuries. But here we have something more: a designation of place that is also a designation of social distinction. Here Maratha means more than just Maharashtrian, but instead also implies Marathi-speaking, and, by extension, devotee of Vitthal. It does not yet indicate a caste or jati, as it will in later centuries and as it does in the present. The final line equates the Maratha parivanda, or the Maratha devotee, with the Vitthal sevak, the servant of Vitthal. We find a direct correlation between being Maratha and serving Vitthal. To my knowledge, this is the first instance of a term—Maratha—that designates the people of Maharashtra as a social collective, a group beyond simply a language or geographic unit, but as a distinctive social group, made coherent by their shared worship, the shared bhakti, of Vitthal.⁸¹ It reminds me of a line from Irawati Karve’s well-known essay about traveling with the Vitthal pilgrims in 1950. She wrote, “I found a new definition of Maharashtra: the land whose people go to Pandharpur for pilgrimage.”⁸² We now have a state-issued inscription that articulates a region, a polity, a language, and bhakti itself all intertwined around the idea of being Maratha, being Maharashtrian.