Public space and public canon in early modern Madurai

In their most beloved rendition, the ‘Sacred Games’ were entextualized as a Tamil talapurāṇam (a legend of place); although composed by a human author, such a work, by the use of the word ‘purāṇam’, invokes the authority of the Sanskrit textual tradition. It may come as a surprise, then, that the ‘Sacred Games’ speaks to a subaltern audience as fluently as it does to Smārta Brahmans or literate Vēḷāḷas. Indeed, the ‘Sacred Games’ are replete with stories of self-actualization from a cross-section of society, as Śiva's grace enters the lives of a firewood vendor, a travelling bard, a young temple courtesan, and an impoverished Brahman poet hoping to raise enough money for his own wedding. Along the banks of the river are the docks where Śiva accepted the burden of an impoverished dock worker, and took upon his own back the lash of the foreman's whip, whose physical blows shook the foundations of the city as they fell upon Śiva's divine body. To this day the ‘Sacred Games’ are dramatized as annual calendrical festivals, preserved as the cultural patrimony of specific families and caste communities. During the famed Cittirai festival, diverse caste constituents, from Smārta Brahmans to the Kaḷḷar communities of Madurai,Footnote ¹⁸ participate in festival observances that permeate the same urban space of Madurai. In essence, the ‘Games of Śiva’ have permanently entered Madurai's public domain as a pillar of the city's public religious culture, as the city is annually resacralized by Śiva's Divine Play.

In early modern South India, publics—and religious publics no less—were invariably multiple, divided along the boundaries of caste, religious community or sect, and language. In contemporary discourse on the public and publicity in the West, the multiplicity of publics—or public spheres—has met with unproblematic acceptance in the years following the English translation of Habermas's seminal work The Structural Transformation of the Public Sphere.Footnote ¹⁹ This public sphere in the singular—itself a largely imagined construct—is made multiple through the emergence of the concept of ‘counterpublics’, a term employed by Nancy FraserFootnote ²⁰ and Michael Warner,Footnote ²¹ among others, to highlight the sites of subaltern resistance to a dominant cultural paradigm, as public discourse fragments to give voice to subordinated identities of gender, ethnicity, or sexuality. Likewise, in the South Indian context, to describe the ‘Games of Śiva’ in Madurai as a fixture of public culture in the singular immediately raises the question: of whom, precisely, is such a public constituted? In the seventeenth century, the answer to the question is by no means self-evident: is mass culture orchestrated by a king who owns no land and serves only as a regent of the goddess Mīnākṣī? Is public religiosity constructed by Brahman ritualists, the supposed religious elite, who hold no legal, political, or economic authority in South India?

Nevertheless, the conceptual difficulties—and indeed historical teleologies—implicit in the very idea of global early modernities have done nothing to discourage the pursuit of extra-European public spheres. In recent scholarship across disciplines, a veritable cottage industry of public sphere studies has emerged, seeking to supplement the canon of modernity studies with examples from the extra-European world. In virtually all cases, presuming that such a thing as a public sphere may exist in Tokugawa Japan,Footnote ²² the Ottoman empire,Footnote ²³ or Safavid Iran exerts a natural pressure on the original categories. As such, this calls into question the criteria of the Habermasian public sphere as definitive of modernity. Babak Rahimi,Footnote ²⁴ for instance, has recently offered the ‘carnivalesque public’ of early modern Iran—best captured by Muharram rituals—as an alternative to the communicative rationality so often viewed as a prerequisite for a democratic state and the key signpost of the emergence of political modernity. By locating a public sphere that need not stand squarely on the shoulders of the nation-state, the printing press, or a capitalist economy, such projects offer the potential to conceptualize early modernity as a truly transcultural phenomenon.

In the majority of scholarship, however, the public sphere, even from a global perspective, is presumed to be an elite phenomenon. Whether fuelled by rational debate, performance, or poetry, proponents of the global public sphere aim generally to articulate those minority discourses that leave an indelible impression on society and history. Beyond the European bourgeoisie, the term ‘public sphere’, one might argue, is only analytically effective when it excludes those subalterns who cannot speak. It is perhaps for this reason that the category of ‘public culture’, a less teleologically inflected alternative, has gained significant traction among scholars of pre-colonial South Asia. Appadurai and Breckenridge,Footnote ²⁵ for instance, define public culture as a ‘zone of cultural debate’, the amorphous spaces between state power and everyday life, amenable to acts of appropriation and resistance, ranging from modern mass culture to the bhakti poetry of low-class poet saints.Footnote ²⁶ Such a model, they argue, frees the concept of publicity from the constraints of the limiting conditions of bourgeois modernity, allowing for a strategic response from outside the spheres of print capitalism and the literate bourgeoisie. And, like many other regions across the subcontinent, in the early modern Tamil south, participatory media—from ritual performance to visual iconography—cultivated a non-literate readership, so to speak, that consumed—and responded to—the initiatives of elite social actors. Nowhere is this more evident, arguably, than in the publicization of the ‘Games of Śiva’. As we shall see, the calendrical festivals themselves were designed to strategically negotiate the interests of distinct caste, sectarian, and language communities.

In using the term ‘public culture’ in this article, I invoke the ambitions of Appadurai and Breckenridge to decentre publicity itself from the pages of a printed text or palm-leaf manuscript. And yet, dispensing with the loosely Habermasian public sphere—whether in the singular or the plural—would preclude the recognition, on a theoretical level, that often literate dialogue has acted as a crucible for systemic cultural change, articulating new performative media or cultural canons that set the stage for future generations across the social spectrum. Indeed, recent studies of Indic early modernityFootnote ²⁷ have highlighted systemic changes in the social dimensions of elite publicity, in some cases freeing intellectual production from the constraints of royal power. In the case of the ‘Games of Śiva’ as well, contrary to popular wisdom, the irruption of a new cultural narrative—now the foundation of Madurai's shared public culture—was not initiated by a decree of the city's regent, Tirumalai Nāyaka, but was cemented through public discourse—in this case, through interlinguistic literary exchange. Only after decades of active circulation among the poetic elite, writing in Tamil, Telugu, and Sanskrit, did the ‘Sacred Games’ pique the interest of the local monarch. In his mission to revitalize the Madurai Mīnākṣī-Sundareśvara temple, he emblazoned the temple walls and statuary with visual signifiers of what had become a cherished cycle of narratives in literary circles.

If the terms ‘public culture’ and ‘public sphere’, then, can be deployed in the service of distinct conceptual ends, it is their intersection that proves most provocative. Although both are ‘publics’, at least in name, the elite and the popular are often approached as non-overlapping magesteria, subject to entirely distinct analytical tools. In the study of early modern India, for instance, the accidents of history fail to mirror the traditional teleology of Europe, in which the vernacular translations of the Bible and the mass dissemination of illustrated pamphlets were purported to have served as causal mechanisms for democratizing religion in the Protestant Reformation. In the absence of mass literacy, however—whether in India or Europe—culture is constantly mediated by performative encounters in public space. Meanings are actively contested in public space when new cultural images are legibly inscribed in the public domain, whether on the walls of temples, on the side of the street, in dance performances in the bazaar, or on the very bodies of the people who inhabit it. It is perhaps no surprise that space has recently captured the attention of scholars of South Asian religions. From the politicization of Hindu festivalsFootnote ²⁸ to the religious ecologies of urban space,Footnote ²⁹ the visual and performative richness of South Asian religions—systems of orthopraxy rather than orthodoxy, as popular wisdom teaches—readily lend themselves to spatial analysis. In what has been termed the ‘spatial turn’ in the humanities at large, religion in South Asia has by no means been left behind.

But what is ‘public’ about public space, particularly at the dawn of modernity in India? No discussion of public space in South Asia can avoid mention, however briefly, of a seminal article on the subject by Dipesh Chakrabarty: ‘Of Garbage, Modernity, and the Citizen's Gaze’.Footnote ³⁰ The very idea of public space, Chakrabarty argues, readily gives rise to a certain scepticism in observers of the urban streets of South Asia, who read the casual disposal of waste as indicative of a sheer disregard for civic duty, public health, and the shared dignity of a public commons. In response to this apparent dilemma, Chakrabarty suggests that space must be read in South Asia not on the axis of public and private, but through an inside/outside structural binary. Perhaps, after all, the spatial practices of Europe translate so poorly from a global perspective that, particularly prior to colonialism and modernity, the term ‘public’ serves no adequate analytic function in the absence of a crucial term of reference such as ‘space’—which is itself implied by Chakrabarty in the spatial binary he proposes. By emphasizing the publicity of public space, then, prior to the advent of an Indic political modernity, my aim is to avoid replicating the assumptions of European civil society. Quite the contrary: from the legible embodiment of religious identityFootnote ³¹ to the visual oversaturation of open space, numerous spatial practices of seventeenth-century South India diverge substantively from their global counterparts while speaking, simultaneously, to the logic of a distinctly South Asian early modernity. As I aim to demonstrate, there is an undeniable publicity at work in the spatial contestation of religion in seventeenth-century Madurai.Footnote ³² As a process distinct from, and often quite independent of, state authority, cultural circulation in public space provides us with an ideal point of mediation for understanding how, in the domain of religion, cultural fashion can become public canon. The ‘Sacred Games of Śiva’ succinctly exemplify just how this process was enacted at the dawn of modernity in South India.

What, then, is a public canon, and what accords it its status of canonicity? The term ‘canon’ is most commonly used in the field of religious studies to refer to the codification of a sacred text, thereafter held as authoritative in either knowledge or practice by a particular community. A canon comes to be viewed, often, as inevitable, eternal, or beyond human control or contestation. In my use of this term, a public canon is not restricted to a particular signifying medium, whether to scripture, visual text, or otherwise. A public canon may have been polyvocal, as in the case of the ‘Sacred Games’, and read by a literate and non-literate public in both text and image.

By emphasizing its spatiality, in visual culture and performance, however, I aim to draw attention to the manner in which spatial practices are particularly amenable to contestation by multiple publics, religious or otherwise. Space can be inscribed by ecclesiastical or royal authority, as it often was in South Asia, but as a key mediator of embodied experience, it remains open to active contestation.

As public canon, however, the ‘Sacred Games’ did take on a canonical form, both in the structure of the narrative itself and through the accepted inevitability with which they came to permeate the urban space of Madurai—in fact, for many, they are themselves an icon that points directly to the city of Madurai. The ‘Sacred Games’, however, were never the property of a single religious community. Rather, we can locate the particular moment at which the ‘Sacred Games’ make the leap from elite cultural fashion to public religious culture, a moment that makes visible the intersection of the literate sphere of public discourse and the everyday experience of public space. In essence, they were forged by an active process of publicity and, in the process, were crystallized as a cultural product that was beyond ownership, but open to interpretation by all. When the ‘Sacred Games’ become part of the public canon, they are viscerally present in lived public space, in a way that cannot resist being read and interpreted but cannot easily be erased from public memory.

Inventing the public canon of Madurai: the ‘Sacred Games of Śiva’

When it comes to the ‘Sacred Games of Śiva’, an unparalleled prototype for a religious canon fashioned not through ecclesiastic authority but by public discourse, it is the interaction between these distinct publics that proved definitive for reshaping public religious culture. In the final decades of the sixteenth century, the ‘Sacred Games’ emerged first as an aesthetic fashion among a cultured elite writing in Tamil, Telugu, and Sanskrit. Only then, within a matter of decades, did the narratives begin to surface in temple murals, statuary, and calendrical festivals, irreversibly entering the wider public domain. It is at this point that public exchange, through literary and devotional discourse, departed from the multiple ‘bourgeois public spheres’, if you will, to become an established fact of what theorists have called public culture, visibly and spatially legible beyond the boundaries of any single discursive public. Thus, for example, the early seventeenth century witnessed the first complete sequence of mural paintings (a genre of representation quite popular in Nāyaka-period temple art) of all 64 ‘Sacred Games’, displayed quite prominently alongside the Golden Lotus Tank in front of the shrine of Mīnākṣī, the ritual heart of the temple.Footnote ³³ Furthermore, and perhaps most importantly, as the ‘Sacred Games’ began to enter the visual landscape of the Madurai temple, Tirumalai Nāyaka instituted a series of calendrical festivals showcasing several of the ‘Sacred Games’ in public performance.Footnote ³⁴ Previous scholarship, guided by an exclusive focus on dynastic political history and reading religious culture as political legitimation, has viewed the cultural innovations of seventeenth-century Madurai as a top-down act of royal authority.Footnote ³⁵ On the contrary, the fact of the ‘Sacred Games’ becoming public and transcending the text has quite a bit to do with the appeal it generated, first among reading publics in multiple languages, followed by its visual and performative enshrinement in public space. In this light, the publication of the text itself can only be ‘read’—figuratively speaking—through its evocations in temple art, architecture, and public festival, a fashion that followed, rather than preceded, its irruption into the spheres of South India's literary publics.

And yet, prior to the sixteenth century, these narratives were scarcely known outside the elite circles of the Tamil literati. One cannot help but wonder, then, how it happened that Madurai as a city came to be entextualized by a single work of Tamil literature. Indeed, the seventeenth-century Tiruviḷaiyāṭal Purāṇam Footnote ³⁶ of Parañcōti Muṉivar had come to be accepted by popular religious culture and temple authorities alike as the sole canonical instantiation of the 64 ‘Games of Śiva’. First premiered before a public audience in the Mīnākṣī-Sundareśvara temple itself, Parañcōti's TVP is a text inseparable from its context. Parañcōti composed his masterpiece, more than likely, during the reign of Tirumalai Nāyaka, Madurai's post-Vijayanagara regent, whose enthusiasm for temple renovation had radically transformed the visual contours of the city centre. Writing at such a pivotal moment in the city's history, Parañcōti had every reason to sing the praises of Madurai as a sacred centre of power, and his extended prologues on the incomparability of Madurai and Mīnākṣī’s temple leave little doubt as to his affection for his home town. His exposition of the ‘Sacred marriage’, the rhetorical high point of the epic and liturgical centrepiece of the temple's annual calendar, waxes eloquent for nearly 200 verses about the jewelled wedding pavilions and garlands of basil and campaka flowers, down to the very detail of the ceremony's ritual implements, as if to evoke a panorama that was, quite literally, lived reality to his readers. The TVP, succinctly, is a textual icon that points directly to the lived space of the city of Madurai.

A similar cycle of 64 ‘Sacred Games of Śiva’ was first compiled around the thirteenth century by Perumpaṟṟapuliyūr Nampi.Footnote ³⁷ Although some of the individual episodes we find in Nampi's work had surfaced on various earlier occasions in Tamil literary history,Footnote ³⁸ no evidence survives to indicate that a complete canon of Śiva's 64 divine sports had ever been compiled before. Writing from Cidambaram, the medieval seat of Śaivism in Tamil Nadu, Nampi fashioned his TVP in a register of verse that intersected seamlessly with the tail end of the more classicized and ornamentalizing Cōḻa period of literary culture.Footnote ³⁹ He claimed initiation under a Śaiva pontiff,Footnote ⁴⁰ a certain Paramañānacivaṉ operating out of the Māḷikai Maṭam,Footnote ⁴¹ a Śaiva monastery in the vicinity of Cidambaram. As a result, it may come as no surprise that Nampi's verse fuses a high Tamil literary idiom with the ethos of earlier Śaiva devotional (bhakti) hymns, in which both Madurai and Cidambaram were integrated into a network of Śaiva sacred sites spread across the Tamil landscape. Perhaps no aspect of the text better illustrates the divergence of Nampi's interest from that of Parañcōti than his treatment of the ‘Sacred marriage’. In Nampi's work, the ceremony itself is relegated to a mere eight verses. Śiva concludes the wedding ceremony by graciously taking political command of the Pandian kingdom, his in-laws’ estate, for he deems a mere woman, such as Mīnākṣī, obviously unfit to rule. The whole event, in fact, is entirely devoid of emotional affect. It is only in the following story, in which the sage Patañjali petitions Śiva to perform his cosmic dance in the city of Madurai, replicating the sacred centre of Cidambaram's Golden Hall in his second, colonized home—the Silver Hall of the Madurai Mīnākṣī temple—that Madurai becomes a truly sacred city. Himself a foreigner to the cultural heartland of southern Tamil Nadu, Nampi reimagines Madurai as an embodiment of a translocal Tamil Śaivism, with little intent to engage with either the landscape or the local populace of Madurai itself. As such, Nampi's work lends itself to interpretation as a novel and creative work of literature, synthesizing the scattered material of cultural memory into a textual artefact capable of entering, for the first time, the sphere of elite, translocal vernacular literature.

And yet, as we shall see, the ‘Sacred Games of Śiva’ boasts a lengthy history of creation and re-creation, making it perhaps the most fluid literary motif in South Indian history, remarkable for its facility to traverse boundaries of language, class, sect, and locality. Originally—as it was for Nampi—TVP had been quite simply a text, with no pretensions to achieving scriptural authority or becoming part of the fabric of everyday life. Whereas the legends had previously been known only to pre-modern Tamil literati through the work of Nampi, the ‘Sacred Games of Śiva’ irrupted quite suddenly into a more general popularity across the Tamil region in the late sixteenth century. This was a period in which the genre of the Tamil talapurāṇam (Sanskrit: sthalapurāṇa)—sacred narrative of place—surged in popularity in conjunction with the rising social, cultural, and economic prominence of the South Indian temple complex. Written for an entirely distinct literary and cultural milieu, a work like Parañcōti's thus speaks at once to an audience of literati as well as to popular enthusiasts already captivated by the cultural dynamism of the Madurai Nāyaka regime and the newfound social capital of the Mīnākṣī-Sundareśvara temple. Over time, Parañcōti's rendering of the work became such a fixture of Madurai's religious culture that it entirely eclipsed any public memory of Nampi's TVP and it remained an obscure fragment of literary history until (and perhaps even after) it resurfaced through the early twentieth-century edition of U. Ve. Caminataiyar.

Just how many Tiruviḷaiyāṭal Purāṇams are there? What little scholarship has been devoted to the subjectFootnote ⁴² is unequivocal. According to popular wisdom, there are two TVPs, the first authored by Perumpaṟṟapuliyūr Nampi in the thirteenth century and the second, celebrated seventeenth-century TVP of Parañcōti Muṉivar. In addition to these two primary Tamil variants, a single Sanskrit Purāṇic rendering has been attested—the Hālāsya Māhātmya (HM). The latter, by virtue of its Sanskrit Purāṇic pedigree, is by and large presumed to have preceded the TVP, with very little evidence to support this.Footnote ⁴³ Our textual archive, however, tells a rather different story: we meet with no complete retellings of the ‘Games of Śiva’ between Nampi's lifetime and the mid-sixteenth century, after which we witness a sudden explosion of variant narratives crossing linguistic and social boundaries. Indeed, within the span of a single century, the Tiruviḷaiyāṭal Purāṇam made the transition from a highly delimited legend of place, restricted to the classics of Tamil literary culture, to a canonical fixture of Śaiva religiosity across South India, visually re-enacted in sacred sites across the Tamil country. Over the course of a mere handful of decades, a narrative that had previously attracted little imitation prompted numerous transcreations in Tamil, Telugu, and Sanskrit, with Marathi and Kannada versions soon to follow.Footnote ⁴⁴ The legends even began to surface in temple murals, statuary, and public calendrical festivals in Madurai and beyond. In a far cry from the hallowed halls of elite literary societies, the ‘Games of Śiva’, one might argue, have permanently entered the public domain of the people of Madurai, and in fact have become a pillar of the city's public religious culture.

The most significant of these pre-Parañcōti variants is, in fact, not in Tamil but in Telugu: the Cokkanātha Caritramu (‘The Story of Cokkanātha’) of Tiruvēṅgaḷakavi (Telugu, circa 1540).Footnote ⁴⁵ This unique work was patronized by the pair of sub-chieftains Pedda Rāma and Cinna Rāma, who operated out of southern Tamil Nadu in the vicinity of Ramnad, significantly removed from Madurai's cultural orbit. Nevertheless, the Telugu Cokkanātha Caritramu is arguably the earliest example of a complete translation—or perhaps, more accurately, ‘transcreation’—of the complete 64 ‘Games of Śiva’, post-dating the thirteenth-century Nampi.Footnote ⁴⁶ Indeed, much like Nampi, as a member of the vernacular literary elite residing at a distance from Madurai itself, Tiruvēṅgaḷakavi held little interest—either regional or rhetorical—in sacralizing Madurai's landscape.

A number of fairly early works, dating to the first half of the sixteenth century, begin to show an increased fascination with the TVP narrative. The Tiruvuccāttāṉar Nāṉmaṇimālai of Nociyūr Paḻaṉiyappaṉ Cervaikkārar (Tamil, circa 1527),Footnote ⁴⁷ for instance, while largely concerned with other matters, includes a chapter that condenses Nampi's ordering of the ‘Games of Śiva’ into an easily digestible set of verses. Likewise, the Kālahasti Māhātmyamu of Dhurjaṭi (Telugu, circa 1509–1529), authored by a poet traditionally revered as one of eight literary celebrities (aṣṭadiggajulu) of the Vijayanagar court of Kṛṣṇadevarāya, incorporates—possibly for the first time—a cycle of the Tamil Tiruviḷaiyāṭal legends into a Telugu text.Footnote ⁴⁸ Within the domain of Madurai itself, the Maduraic Cokkanātar Ulā of Purāṇa Tirumalainātar (Tamil, early sixteenth century),Footnote ⁴⁹ belongs to the Ulā genre of Tamil literature, a literary form centred on the motif of the formal public procession of a ruler or deity in a particular locality—in this instance, Cokkanātar or Sundareśvara of Madurai. While recounting the procession of Sundareśvara through the streets of Madurai, the Cokkanātar Ulā sprinkles allusions to several of the ‘Games of Śiva’.

By the mid-to-late sixteenth century, we begin to encounter a number of complete renditions of the 64 ‘Sacred Games’ in multiple languages, including the Cokkanātha Caritramu, thus predating Parañcōti's own masterpiece. However, these often survive in fragmentary condition with numerous corruptions. Many have been all but forgotten by the scholars of Tamil or Telugu literary history. Take, for instance, the Cuntara Pāṇṭiyam (‘The Story of Cuntara Paṇṭiyaṉ’) of Aṉatāri (late sixteenth century), a virtually unstudied retelling of the ‘Games of Śiva’ in his incarnation as the Pandian ruler Sundareśvara, a work that, despite its intriguing digressionary discussions of subtle body yoga and Śākta devotionalism, has only barely survived to this day. How many other authors like Aṉatāri set out to retell the ‘Games of Śiva’ only to have the manuscripts of their compositions dismissed by their colleagues or lost to subsequent generations?Footnote ⁵⁰ We also encounter abortive attempts at alternative talapurāṇams of Madurai, which subsume the ‘Sacred Games’ within an entirely different narrative frame. The Katampavaṉapurāṇam (The Purāṇa of the Kadamba Forest) of Vīmanāta Paṇṭitar, for instance, claims to narrate the sacred history of the city from an entirely different stream of textual transmission, ostensibly adapted from a Sanskrit work variously referred to as the Katampavaṉapurāṇa or, synonymously, the Nīpāraṇyapurāṇa. Its author, Vīmanāta Paṇṭitar, refers directly to Parañcōti in the opening verses of the composition, adopting an almost apologetic tone for his audacity in putting forth another contender for Madurai's official talapurāṇam:

In short, the multilingual explosion of the narrative took place prior to its canonization as the accepted legend of place, or sthalapurāṇa, of the city of Madurai. While our pre-colonial textual archive often occludes the dynamics of extra-textual circulation, the ‘Games of Śiva’ may prove an exception to the rule, entering, as it did, public discourse with remarkable visibility by the mid-seventeenth century. We have a clear, succinct documentary trail of how literary voices across languages and religious communities worked and reworked a long-forgotten legend through a decades-long process of circulation and performance. None of these early narratives was identical to another: all were themes, variations, and experimentations on the initial theme provided by Nampi some four to five centuries earlier. It is only when we come to Parañcōti's Tiruviḷaiyāṭal Purāṇam, however, that the text itself took on a canonical quality, as the legends were frozen in a version with a fixed, invariable sequence. This canonization, moreover, was facilitated by the arrival on the scene of a ‘textual twin’: the Hālāsya Māhātmya,Footnote ⁵² the sole Sanskrit Purāṇic variant of the ‘Games of Śiva’, is virtually identical to Parañcōti's masterpiece in narrative and sequence. With an authoritative version having been produced in multiple languages, the legends began to shift from literary fashion to public religious canon.

Multiple languages, multiple publics

But precisely what sort of textual culture is it that we are encountering with the rise of the TVP from obscurity to preoccupying the literary elite of multiple language communities? Why were these texts written, and was their enunciative effect the same in Tamil, Sanskrit, and Telugu? In fact, by the sixteenth century it was not only the languages but also the institutions of literary circulation that had diverged radically in South India. This led to the emergence of distinct literary spheres that often intersected but also operated out of disparate commitments in the domains of literature, politics, and devotion.

By the very definition of the genre, the Tamil talapurāṇam (a narrative of place) deals with the unique soteriological properties and divine exploits associated with a precise locality in the Tamil country. As these legends, more often than not, owe relatively little to the pan-Indic corpus of Sanskrit purāṇas, one might expect that authors of Tamil talapurāṇams, composed primarily of narratives that are strictly Tamil in geographical and cultural origin, would look no further than the extensive literary and devotional archive accumulated in well over a millennium of Tamil textual history. Nevertheless, from the very inception of the Tamil talapurāṇam genre, poets evidently felt compelled to seek for these temporally and geographically delimited narratives a stamp of approval from the transregional Sanskritic tradition by framing their compositions as translations of, or perhaps transcreations from, original Sanskrit exemplars. Such was the case with Nampi's TVP, one of the earliest known examples of the talapurāṇam genre. However, despite the obviously Tamil origins of many of its episodes, Nampi informs us that it was not originally transmitted in Tamil at all. He claims, rather, to have drawn on an otherwise unattested Sanskrit ‘text’ known as the Sārasamuccaya, contained in the Uttaramahāpurāṇa.Footnote ⁵³ Based on the titles—even setting aside its absence in manuscript history—there is good reason to doubt that an excerpt entitled the ‘Compilation of Essences’ (Sārasamuccaya) in the ‘Other Great Purāṇa’ (Uttaramahāpurāṇa) ever existed at all.

To put it succinctly, the relationship between Sanskrit and vernacular in the early modern Tamil South may not be quite as cut and dry as it appears. Regardless of how strongly Nīlakaṇṭha Dīkṣita may have personally disapproved,Footnote ⁵⁴ the Nāyaka-period Tamil country belonged unmistakably to what Pollock has termed the ‘Vernacular Millennium’Footnote ⁵⁵ and, in fact, vernacular literature flourished there in abundance. Not only did the region continue to foster its vibrant and prolific heritage of Tamil literary production, but Nāyaka rulers, hailing from Andhra and formerly under the Vijayanagar empire, imported (along with their political rule) a predilection for Telugu literature. Through their continued patronage, it began to take root in the far South. Of course, the social and political functions of vernacularization had been fully present in the Tamil region since the height of Cōḻa rule, when Tamil literature began to assume the role of the primary medium for royal encomium, adopting numerous stylistic and tropic features from the pre-existing Sanskrit cosmopolitan tradition. Moreover, high Cōḻa literature was indubitably a courtly phenomenon, produced and publicized within the central networks of the empire's ruling elite and often directly underwriting the interests of royal power.

The vernacular of the Nāyaka period, however, took shape within a sphere of multiple competing cultural currents, creating a dynamic in which emulating and implementing received literary models did not flow unilaterally from the cosmopolitan to the vernacular, from the transregional to the language of place. In fact, literary classics were often adapted from one vernacular to another,Footnote ⁵⁶ and just as often from the vernacular back into Sanskrit. While the cosmopolitan vernacular, so to speak, often accompanies a certain documented social trajectory, much less is known about the sort of extra-textual environment that would support such a multidirectional sphere of literary influence.

Given this apparent fluidity of interchange between competing literary currents—that is, given the ease with which the content of the literary craft traversed the boundaries of language—should we presume an equally fluid social structure facilitating the production and transmission of literary texts across distinct language-based communities? Certainly, the answer to this question varies considerably by geographical region, even during the time frame we have been referring to as India's ‘early modern’ period (circa 1500–1800). Literary production in the Nāyaka-period Tamil country need not have operated within institutional frameworks equivalent to those of the seventeenth-century Rājput courts of Rajasthan or anywhere else in the Indian subcontinent. The situation in South India, however, is further complicated by the coexistence of multiple vernacular traditions within a shared geographical and cultural space. In such a context, Pollock's model of the Vernacular Age might suggest that the competing vernacular literatures of South India ought to have inherited equally certain constitutive features of the Sanskrit cosmopolitan paradigm. For instance, we might expect, in the present case, that Tamil and Telugu works of literature were patronized at the same Nāyaka courts, performed in the same venues, and influenced equally by the rhetoric and values of the Sanskrit literary tradition.

With its broad appeal across linguistic lines, the TVP and its numerous multilingual variants provide us with an ideal arena from which to explore the extent to which these expectations hold true in the South Indian case. Fortunately, the texts in question speak for themselves, providing information about their contexts of patronage and performance both explicitly and implicitly through the rhetorical tropes they invoke. Take, for instance, the following verse from the introduction to the Cuntara Pāṇṭiyam:

What precisely is this ‘six-legged seat’ that Aṉatāri so specifically foregrounds in his work? The remainder of the Cuntara Pāṇṭiyam provides us with no further clues, but fortunately Aṉatāri is not the only one of our authors to mention just such a six-legged seat and give it the same emphatic placement in the introductory verses. In fact, the first verse of Parañcōti's TVP is structured around a fourfold pun on the term ‘aṟukāṟpīṭam’, suggesting that it is more than an idiosyncratic turn of phrase:

Not only does Parañcōti inform us here of the location of the ‘six-legged seat’—that is, in the interior of the Madurai temple's shrine (caṉṉati/sannidhi) dedicated to Śiva as Sundareśvara or Cokkanātha—but he connects this particular ritual platform directly with the institution of the literary premiere or araṅkērram. The araṅkērram, as a literary-performative institution, survived well into the nineteenth centuryFootnote ⁵⁹ as a central pillar of pre-print-culture Tamil literary practice.Footnote ⁶⁰ Seventeenth-century evidence suggests unambiguously that the araṅkērram was an established institution of Tamil literary performance in the period; one notable example is an extant note written by the poet Antakakkavi to his patron inviting him to the araṅkērram of his forthcoming work.Footnote ⁶¹ What we learn here, however, is that in the Nāyaka-period literary sphere in which Parañcōti premiered his highly influential TVP, the araṅkērram of a talapurāṇam, and possibly other works relevant to the sacred sites of the Tamil Śaiva religious landscape, seems to have been directly facilitated by temple institutions. Thus, as Parañcōti informs us quite clearly, his TVP was debuted in the Madurai temple within the central shrine of Sundareśvara itself.

But just who were these sponsors of the literary works such as Parañcōti's that were publicly premiered at major temple sites? In some cases, temple officials or priests seem to have played an instrumental role in encouraging an author to embark on composing a sacerdotal literary work in the Tamil language, ostensibly translated from a Sanskrit original. Vīmanāta Paṇṭitar, author of the Katampavaṉapurāṇam, for instance, describes his impetus to begin his work in just such a fashion, claiming that the temple priests (talattōr) requested that he translate into Tamil the Sanskrit purāṇa on the greatness of the Kadambavanam, the Kadamba forest that preceded the urbanized landscape of Madurai:

That said, as the regional megatemples of South India (such as the Madurai temple) had by this time become significant centres of political and economic exchange, we should not underestimate the impetus for subordinate chieftains to participate as exhaustively as possible in this transactional network. Numerous other authors, such as Aṉatāri, author of the Cuntara Pāṇṭiyam, cite as individual patrons of their works not the Nāyaka kings of Madurai or Tanjavur, but more often their subvassals who had established smaller regional courts at various locations throughout the Tamil region. This decentralized form of patronage is a distinctive feature of what has been described, though not without some trepidation, as the feudal political structure of the Nāyaka regimes. From the Vijayanagar period onwards, the term ‘nāyaka’ was applied to describe a regional feudatory ruler subservient to the centralized authority of the empire. Even after the Madurai and Tanjavur had attained functional independence from the declining Vijayangar state, the term was retained as a key feature of political discourse, first perhaps as a rhetorical gesture of humility but later as a functional description of the similar political hierarchy that had emerged under the Nāyaka regimes themselves. Nāyaka vassals often referred to themselves by the title ‘Nāyaka’ as well, and breakaway states frequently emerged in competition with the generally prevailing authority of Madurai and Tanjavur. This increasingly decentralized political structure appears to have provided subchieftains and subordinate officers with a heightened incentive to engage directly in the patronage of Tamil literature, especially of works with a more overtly theological import which offered avenues for advancement in the competitive prestige economy that was centred on major temple institutions.Footnote ⁶³

Such was the case with Aṉatāri, who in the above verse describes his patron, Tiruviruntavaṉ of Kallur, as a subordinate officer. He then further elaborates the complex hierarchical chain that linked his direct patron with the centralized Nāyaka authority of Madurai under Kacci Vīrappa Nāyaka, apparently through the mediation of a tertiary figure, Cevvanti, who held some kind of official role at the Madurai court and evidently the personal favour of the Nāyaka as well:

A similar situation holds true for Purāṇa Tirumalainātar, the author of the Cokkanātar Ulā, who names as his patron one Vīramāṟaṉ, a functionary or ruler in the locality known as Mulaicai. He celebrates Vīramāṟaṉ’s anniversary of rule with the composition of the work in question, narrativizing the occasion as the impetus for Cokkanātha's public procession:

In short, whereas in some cases patronage may have derived from temple officials directly, in most cases it was more likely at the hands of various subvassals of the Nāyaka rulers or upstart rivals at minor courts who aimed to enhance their standing in the economy of ritual exchange centred around honours distributed by the Mīnākṣī-Sundareśvara temple. A third factor, however, which had a significant impact on the structures of literary circulation among Tamil Śaiva poets of the sixteenth and seventeenth centuries was participation in the devotional networks of prominent Tamil Śaiva monastic centres, such as Tarumapuram or Tiruvavatuturai. The economic and social prominence of these monastic centres had begun to increase dramatically in the preceding centuries,Footnote ⁶⁶ and by the sixteenth and seventeenth centuries seem to have provided a crucial venue for the circulation of literary worksFootnote ⁶⁷ and fostering of poetic talent among those who wished to participate in Tamil literary circles. Cultivating a distinctively Tamil Śaiva identity, in contrast to the Sanskritic lineages of the Śaiva Siddhānta,Footnote ⁶⁸ these monasteriesFootnote ⁶⁹ attracted lay participants from a largely Vēḷāḷa social background, a set of caste communities who were technically Śūdra in origin but had quite often attained a rather elevated social standing in this period as major landholders and managers of agricultural property.Footnote ⁷⁰

It is no accident, in fact, that the vast majority of Śaiva poets writing in Tamil who provide us with any biographical information during this period explicitly professed a Vēḷāḷa caste originFootnote ⁷¹ and an affiliation with the spiritual preceptors of the Tamil Śaiva lineages. Among the authors of TVP variant narratives, a prime example is Vīmanāta Paṇṭitar, author of the Katampavaṉapurāṇam, who directly links his poetic endeavours with his caste origin:

It was not merely caste alone, however, that provided a social foundation for the continued patronage of Tamil literature. Rather, it required the mediation of monastic institutions that structured their ideological self-representation on the Vēḷāḷa heritage of its founders and lay participants. It is perhaps no surprise, then, that Vēḷāḷa authors of Tamil Śaiva literature in this period often participated quite openly and actively in the development of these increasingly prominent devotional centres. The prototypic example of such a poet is Kumārakurupara, a seventeenth-century contemporary of Tirumala Nāyaka who authored numerous worksFootnote ⁷³ dedicated primarily to the goddess Mīnākṣī of Madurai. After a long-standing connection with the maṭams at Tarumapuram and Tiruvavatuturai, Kumārakurupara is believed to have been sent northwards by his lineage preceptors to establish a branch maṭam of the Tamil Śaiva tradition in Varanasi. From among authors of the TVP corpus, one highly specific reference speaks to the sectarian allegiance of the family of Purāṇa Tirumalainātar, author of the Cokkanātar Ulā. His son, in his grammatical work, the Citamparappāṭiyal, informs us of his family's close affiliation with the Tamil Śaiva lineage,Footnote ⁷⁴ referring unmistakably to Meykaṇṭar, the lineage's founder, and even suggesting that he composed the work in question at the behest of a later preceptorial figure, Tattvañānaprakācar (Skt. Tattvajñānaprakāśa):

A great deal of research remains to be done on the influence of Tamil Śaiva monasteries on both the literary sphere of early modern Tamil Nadu and its expression in public religious culture, despite their social influence and avid patronage of religious expression in diverse media. For instance, a significant portion of temple mural paintings produced during the sixteenth and seventeenth centuries was sponsored directly by highly ranked administrators or members of these same Tamil Śaiva monasteries.Footnote ⁷⁶ In short, currently available evidence suggests quite strongly that the Tamil literary sphere of the sixteenth and seventeenth centuries had become intimately intertwined with the Tamil Śaiva monastic lineages to form an institutional foundation for literary patronage and circulation. The dynamics of Tamil textuality in the early modern Tamil country, then, were markedly distinct from what we observe in the case of both Telugu and Sanskrit literature of the period: Tamil literariness, in early modern Madurai, was centred upon its production, performance, and circulation within the Śaiva monastery.

The patronage of Telugu literature, on the other hand, even within the same time frame and geographical region, diverges quite significantly from the Tamil case. One striking example, for instance, is the Cokkanātha Caritramu of Tiruvēṅgaḷakavi, a text that narrates the very same cycle of narratives but with a rhetoric that marks its social location as quite distinct from that of its Tamil counterparts. It is arguably the earliest example of a complete translation—or perhaps, more accurately, ‘transcreation’—of the complete 64 ‘Games of Śiva’ into a language other than Tamil. Dating to the mid-sixteenth century, it is likely to have predated the most influential renderings of the narrative, the TVP of Parañcōti and the Hālāsya Māhātmya. As a result, this previously unstudied work stands wellplaced to expand our perspective on the institutional foundations and linguistic media of literary circulation during this period.

In terms of patronage, the Cokkanātha Caritramu, much like a number of Tamil texts of the period, was sponsored by relatively minor chieftains from a sub-regional court to the south of Madurai. Its performative rhetoric, however, is quite different from its Tamil counterparts as it explicitly evokes the imagery and prestige of a courtly literary sabhā—a world where kings are attended with yak-tail fans and offered an uninterrupted flow of betel leaf. In this case one might even describe the setting as ‘secular’, as the work betrays no connection with any temple-based or monastic institution but rather emphasizes the aestheticized political power of its patrons. As we can glean from the following passage, Pedda Rāma and Cinna Rāma felt that their worldly prestige stood to benefit considerably from attracting skilled Telugu poets who hailed from long-celebrated literary families:

In this respect, the Cokkanātha Caritramu, unlike the Tamil texts we have examined, is undoubtedly an heir to the political, social, and literary values of the Sanskrit cosmopolis. Unsurprisingly, the linguistic register is also highly Sanskritized, and we encounter a celebration of cosmopolitan literary history in the guise of the traditional kavi praśaṃsā—praise of previous poets—not only of the great celebrities of the Telugu literary world but of the Sanskrit tradition as well.

Although an intriguing phenomenon in its own right, the multiplicity of institutional sites that supported literary production in the Nāyaka period also has significant implications for our understanding of how literary themes developed, circulated, and were disseminated into the domain of public culture. The TVP is simply one example of a narrative that grew to maturity and attained its now cherished place in cultural memory by navigating this multicentric, multilingual literary milieu. As a literary theme that received substantial attention throughout the sixteenth century across the boundaries of language, institution, and locality, the TVP appears to defy a number of our normative assumptions about how works of literature attain a position of social or cultural prominence, whether through the genius of an individual poet or through the direct patronage of a single political ruler or other social agent wishing to legitimize his claim to authority. In fact, its widespread dissemination throughout the sixteenth century—and this presuming a flawed and incomplete historical archive—defies the very possibility of reading its re-emergence in the Nāyaka period as a top-down act of political legitimation. On the contrary, Tirumala Nāyaka interventions coincide quite closely with the period of textual codification witnessed in the following decades, as the TVP began to circulate outside the boundaries of elite literary circles and enter the domain of popular literary culture.

And perhaps the most intriguing fact about the public canonization of the ‘Games of Śiva’ was the legends’ uncanny ability to circulate across the boundaries of language. At the height of South India's Vernacular Millennium, the vernacular of the Nāyaka period took shape within a sphere of multiple, competing cultural currents, creating a dynamic in which the emulation and implementation of received literary models did not flow unilaterally from the cosmopolitan to the vernacular, from the transregional to the language of place. In fact, literary classics were often adapted from one vernacular to another, and just as often from the vernacular back into Sanskrit.Footnote ⁷⁸ In short, while interpreters often assume that a singular South Asian ‘public sphere’—a Brahmanical ‘bourgeois public sphere’ which, in Habermas's terms, would exclude the majority of Madurai's population—public religious culture in early modern south India was forged at the intersection of multiple independent publics, inflected by language and sectarian affiliation and overlaid with each other in geographic space but not reducible to a single axis of hegemony or subalternity.

Twin texts: the canonization of the Tiruviḷaiyāṭal Purāṇam

Most importantly for our purposes, none of the above works appears to be indebted to either of the two exemplars of the TVP genre that were given historical primacy by existing scholarly literature, namely Parañcōti's TVP and the Sanskrit Hālāsya Māhātmya, which were allegedly the direct source for all representations of the ‘Sacred Games’ in the centuries after Nampi. In fact, two of the most interesting of these works, the Cuntara Pāṇṭiyam and the Cokkanātha Caritramu, have enough structural similarities to Parañcōti's TVP to suggest the genuine emergence of a shared template for narrative improvisation. Nevertheless, they diverge in crucial respects, bringing into question the presupposition that all of the texts could have been adapted unilaterally from a single point of origin. Reactions in the scholarly literature have varied considerably, ranging from voices such as that of Harman,Footnote ⁷⁹ who emphasized the purely rhetorical role of Sanskrit ‘originals’ in Tamil Purāṇic composition, to JeyechandrunFootnote ⁸⁰ and Wilden,Footnote ⁸¹ who virtually assume that Parañcōti translated the Hālāsya Māhātmya directly into Tamil. And yet, to date, I have not once encountered a single citation of the Hālāsya Māhātmya earlier than the late seventeenth century.Footnote ⁸² Internal textual evidence, on the other hand, speaks volumes about this issue, but only when Parañcōti's TVP and the Hālāsya Māhātmya are brought into dialogue with a much broader spectrum of contemporary literary production. As I will argue below, the suspiciously similar content of Parañcōti's TVP and the Hālāsya Māhātmya pair them as ‘twin texts’, so to speak, suggesting quite strongly that, at the very least, the Hālāsya Māhātmya could not have been known to any vernacular poets prior to Parañcōti.

Beyond its inclusion of three of Nampi's original ‘Sacred Games’, perhaps the most suggestively interstitial feature of the Cokkanātha Caritramu is its ‘elision’ of the prolific purāṇic frame narratives that feature prominently in both Parañcōti's TVP and the HM. The Cokkanātha Caritramu, much like Nampi's earlier TVP, undertakes a streamlined narration of each of the 64 ‘Sacred Games’, showing no predilection for mythological elaboration. However, the latter canonical narrative is scattered with mythological back stories and non-narrative materials—from ancient curses to applied religious observances (vratas) and spontaneous stotras (hymns)—as one would expect from the texture of a typical Sanskrit purāṇa. Some of these digressions, such as the apparently irrelevant Somavāravrata chapter in the HM and the stotra sung by Patañjali upon witnessing Śiva's dance after the sacred marriage in Madurai, feature only in the HM and no other known variants. Most mythological addendums, however, although preserved identically in both Parañcōti's TVP and the HM, appear in no other early rendering of the ‘Sacred Games’, including the Cokkanātha Caritramu, which otherwise conforms quite closely in narrative structure to the later TVP and the HM.Footnote ⁸³ Combined with his inclusion of Nampi's three original episodes, however, Tiruvēṅgaḷakavi's apparent unawareness of any of the later purāṇic frame narratives suggest quite strongly that he did not have either the HM or Parañcōti's TVP available as a model when he composed the Cokkanātha Caritramu. Moreover, given his deep respect for Sanskritic culture (shown in a lengthy digression on the virtues of sixteenth-century Varanasi), heavily Sanskritized vocabulary, and the purāṇic narrative style employed in his introductory frame, it is highly unlikely that Tiruvēṅkaḷakavi would have neglected entirely these new additions had he indeed ‘translated’ the HM into Telugu.

One prime example of such a mythological excursion, and a fairly controversial one at that, sets the stage for the origin story of the Tamil Caṅkam and is featured prominently in both Parañcōti's TVP and the HM, one of a series of narratives that eulogize the prehistoric efflorescence of Tamil literary culture in the city of Madurai. Although this particular narrative is unattested to prior to the TVP and the HM, its distinctive features in the HM have been cited as evidence for both the priority and relative antiquity of that work by scholars of Tamil literary history such as David Shulman.Footnote ⁸⁴ Our story begins with Brahmā and his three wives, who have set out on a pilgrimage to Varanasi to bathe in the Ganges together. Upon their arrival at the sacred river, Sarasvatī’s attention is suddenly diverted by the melodies of a celestial musician of sorts and she abandons the task at hand in pursuit of the unseen singer. When she returns to rejoin her husband and co-wives, Gāyatrī and Sāvitrī, Sarasvatī discovers that Brahmā and the others have already completed their ablutions and that Brahmā is distinctly displeased at her unexplained absence at a crucial moment of the ritual purification. Angered at her apparent irresponsibility, Brahmā curses her: she will have to undergo 48 mortal births in recompense for her lapse. Distraught, Sarasvatī begs Brahmā to relent, and he modifies the curse so that she will be born simultaneously as the poets of the Tamil Caṅkam, represented by the 48 letters of the alphabet. She will be accompanied by Śiva as the forty-ninth poet, the embodied form of the letter ‘a’. In Parañcōti's words:

And as similarly recounted in the HM:

Aside from the oft-noted confusion about the total number of letters, which may result in part from the ambiguities of cross-linguistic transmission,Footnote ⁸⁸ the most salient feature of this mythological prehistory is that the Caṅkam poets have been symbolically encoded as the incarnate letters of the Sanskrit alphabet, which together are said to comprise the body of Sarasvatī herself, the power of language. Shulman, for instance, argues that this esoteric imagery provides unambiguous evidence that the HM originated from an older, pan-Sanskritic Śākta theological system,Footnote ⁸⁹ which was later imperfectly transmitted into the Tamil cultural sphere in Parañcōti's TVP, resulting in a denuding of the HM's specifically Sanskritic Śākta vocabulary. It is true, in fact, that this episode, as well as numerous other passages in the HM, are heavily overlaid with Śākta terminology, from the reference to the saṅghaphalakam—the Caṅkam plank, the seat of the poets in the assembly hall—as a ‘vidyāpīṭha’ or ‘mātṛkāpīṭha’,Footnote ⁹⁰ to references to a set of navaśaktis, or nine fierce goddesses, who are somewhat less coherently integrated into the overall plot of the purāṇa.Footnote ⁹¹ Unfortunately, none of these terms is truly tradition-specific enough to evince a definitive origin in any pan-Sanskritic tradition of esoteric Śāktism, much less, as Shulman contends, within an unspecified Śākta lineage from the northwest of the Indian subcontinent.

We do, on the other hand, find numerous exact parallels to the Śākta terminology of the HM within the Tamil Śaiva canon itself, suggesting that we need not look as far afield for their origin as Shulman has contended. In particular, the Tirumantiram, which notoriously preserves numerous remnants of a proto-Śrīvidyā esotericism (which seems to originate in the Kashmiri Śākta-Śaiva traditions exported to the South) repeatedly invokes the set of 51 letters of the alphabet as central elements of its various yantras and other esoteric imagery. On several occasions, we also find reference to Śiva as embodying the foremost of these syllables, the letter ‘a’:

Although the Tirumantiram was most likely composed centuries before Parañcōti's TVP,Footnote ⁹⁴ as the tenth book of the Tamil Śaiva canon, its imagery quite understandably maintained a widespread popularity among Parañcōti's contemporaries, even surfacing in publicly available works of Tamil purāṇic literature. The trope of the 51 letters, for instance, makes an appearance in the Cuntara Pāṇṭiyam as well, entirely disconnected from any mention of the Caṅkam or its myth of origin:

Given these striking parallels, the very esoteric imagery that may seem to betray an extra-local origin for the Sanskrit HM in fact evokes the flavour of a distinctively Tamil Śākta-Śaivism, leaving little remaining doubt that the HM emerged not from any pan-Indic Sanskrit tradition but rather directly from the Tamil Śaiva textual culture of the early-to-mid second millennium. Although preserving a number of originally Sanskrit features—from the inclusion of the letter ‘kṣa’ in the alphabet to translocal yogic terminology such as nāda, bindu, and dvādaśānta—the imagery of the Tirumantiram had been adopted and reworked for centuries within the confines of the Tamil Śaiva tradition. Far from blending uneasily with Tamil Śaiva theology, as Shulman would have it, the 51 letters play a central role in a subtle cosmology that had been accepted centuries earlier into the core repertoire of Tamil Śākta-Śaiva tradition, remaining in circulation until the seventeenth century and beyond.

This being the case, the frame narrative of the Tamil Caṅkam cycle simply cannot serve to indicate an earlier, extra-Tamil origin for the HM. On the contrary, the fact that both the HM and Parañcōti's TVP preserve such a memorable and idiosyncratic purāṇic accretion in nearly identical form—one that is corroborated by no other known variant dating to the sixteenth century—establishes beyond doubt that the circumstances of their composition were directly linked, but within a much more delimited time frame than previously suspected. The twin texts appear to post-date the Cokkanātha Caritramu of the mid-sixteenth century, which closely resembles the later narrative structure but includes none of the purāṇic accretions and preserves Nampi's earlier episodes, which were forgotten by later audiences. When all evidence is considered, the HM was most likely re-Sanskritized directly from Parañcōti's fabulously successful TVP shortly after its composition in response to demands for a Sanskrit original. However, it remains possible that the Sanskrit purāṇic version was ‘found’—that is, commissioned—and employed as a model for Parañcōti's work. In any case, it is beyond doubt that the Sanskrit HM never circulated in South Indian literary venues before Parañcōti's TVP had left a substantial influence on the public culture of Madurai and the temple of Mīnākṣī and Sundareśvara.

Some decades later, however, we discover that Nīlakaṇṭha Dīkṣita, luminary of the seventeenth-century (circa 1637 ce)Footnote ⁹⁶ Sanskrit literary society of Madurai, had personally gained access to the HM, a fact that can be gleaned from a careful reading of Śivalīlārṇava, his own rendition of the ‘Sacred Games’ as a Sanskrit mahākāvya.Footnote ⁹⁷ In the course of the Caṅkam cycle of episodes, after the Goddess Sarasvatī had taken incarnation as the 48 Caṅkam poets in Madurai, the current Pandian ruler Campaka Pāṇḍya (so named for his well-known preference for the fragrance of campaka flowers) encountered a troubling dilemma. During an intimate evening with his newly wed queen, Campaka Pāṇḍya discovered that her hair had a rather distinctive fragrance and he began to wonder about its origin. The king was so troubled by his uncertainty that he promptly announced a prize of a purse of gold coins for any poet who could produce a compelling and eloquent verse explaining whether or not a woman's hair can produce such a fragrance without the presence of flowers or artificial perfumes. The prize-winning verse, which Śiva himself composed and entrusted to a young Brahman bachelor named Tarumi, was widely understood from the earliest attestations of the Caṅkam narratives to be a genuine Caṅkam-period verse preserved in one of the anthologies, the Kuṟuntokai:

In the course of adapting this episode, the necessity naturally arose for both Nīlakaṇṭha and the author of the HM to translate this verse into Sanskrit, in the process preserving a distinct linguistic texture from the surrounding narration. Beyond any possible coincidence, however, both the HM and the Śivalīlārṇava employ precisely the same verse,Footnote ⁹⁹ in āryā metre, as a translation for the Tamil of the second stanza of the Kuṟuntokai:

Nīlakaṇṭha's Śivalīlārṇava (circa 1625–1650), then, provides a rather definitive terminus ante quem for the twin canonical renderings of the ‘Sacred Games’—the HM and Parañcōti's TVP—which as a conjoined pair may have been composed a mere decade or two before. From a strictly literary historical stance, this dating exercise may appear somewhat inconsequential. From the standpoint of political history, however, that the publicly acclaimed versions of the ‘Sacred Games’ should have originated during this particular period demands a consideration of its role in Nāyaka statecraft and in the city of Madurai, a cultural capital rapidly transforming under the influence of the Madurai Nāyaka regime. Following the reign of Viśvanātha Nāyaka (1529–1564), who by the end of his career had achieved de facto independence from the declining Vijayanagar empire, the religio-political landscape of Madurai took on a newfound importance for the agenda of the Madurai Nāyakas. They may well have found it advantageous to highlight the rich cultural legacy of the ancient Pandian capital at the heart of their kingdom. Given the political, economic, and cultural significance of the South Indian temple complex during this period, the cultural renaissance instituted by the successors of Viśvanātha Nāyaka quite naturally began with an expansion of the most influential regional temples, particularly the Mīnākṣī-Sundareśvara temple, the geographical and cultural centre of Madurai.

From text to public religious culture: the Tiruviḷaiyāṭal Purāṇam in seventeenth-century Madurai

In the introduction to his edition of the Cokkanātar Ulā, U. Ve. Caminataiyar, father of the modern renaissance in Tamil literary studies, recounts a popular anecdote concerning Purāṇa Tirumalainātar, the text's author, and how he came to receive his rather peculiar nom de plume.Footnote ¹⁰¹ Far better known for the Citampara Purāṇam, his other surviving composition, Tirumalainātar is said to have been petitioned by the elders and devotees of the Cidambaram Śaiva community to translate into Tamil the surviving Sanskrit scriptural canon recounting the sacred history of Cidambaram. Without access to a suitable Sanskrit original, our would-be translator set off for the mountain country (malaināṭu), where he discovered a single, incomplete manuscript of the Sanskrit Cidambara Purāṇa and proceeded to translate the extant portion into the form of an equivalent Tamil talapurāṇam. Although he remained grievously disappointed at being unable to locate the entire Sanskrit original, the temple priests were so gratified by his efforts and the quality of his final product that they appended the prefix ‘Purāṇa’ to his title, in commemoration of the Citampara Purāṇam.

While we sadly lack any documentary evidence to confirm that sixteenth- and seventeenth-century Tamil poets actively sought out Sanskrit manuscripts on which to anchor the authority of their Tamil compositions, more recent accounts confirm that this was common practice in the nineteenth century. For instance, a similar anecdote recorded by U. Ve. Caminataiyar outlines the process by which his own teacher Minatcicuntara Pillai, a renowned scholar of Tamil literature, set out to produce a Tamil talapurāṇam of Kumbakonam at the request of local monastic authorities:

What is the significance, then, of Mīṉāṭcīcuntaram Piḷḷai's story to our understanding of public religious culture in seventeenth-century Madurai? What we witness in this vignette is nothing less than the creation of public canon, narrated from the perspective of onlookers who witnessed the debut of his Tamil Purāṇam and accepted its legends as an authoritative précis of Kumbakonam's sacred legacy. For Piḷḷai as well as for the seventeenth-century Parañcōti, the birth of a talapurāṇam was a social affair, imbricated with the monastic and temple institutions where the text was composed, debuted before a public audience, and commemorated by local elites with the bestowing of ritual honours. It is a process that takes place in time and space, fusing new meaning onto the sites it commemorates, which become legible for future generations of devotees. Space itself, in temple and monastic complexes, becomes entextualized with the emergent public canon—what previous generations of scholars have described as ‘sacred space’. And yet, this is anything but a hermetically sealed ‘sacred’, set apart from the phenomenal experience of the mundane realm. It remains, rather, a public space—a site for the reproduction of public religious culture, available for response, re-enactment, and contestation.

Indeed, religious spaces across the Tamil country were in the midst of a radical re-entexutalization in the sixteenth and seventeenth centuries: the materiality of religious sites itself was transformed through massive temple-building campaigns, and the newly refurbished temple complexes were indexed with fresh mythological narratives. Succinctly, we witness a remarkable upsurge both in the production and renovation of Hindu temples and the composition of talapurāṇams in Tamil to invest them with canonical meaning. In excavating the history of the talapurāṇam genre, David Shulman notes that while the vast majority of these texts lack the pedigree of the Tamil classics, they were composed primarily during the sixteenth and seventeenth centuries, the very moment at which the Nāyaka regents of the Tamil country were renewing their alliance with sacerdotal power by endowing new temple-building projects at the heart of their domains.Footnote ¹⁰³ The Madurai Mīnākṣī-Sundareśvara temple is no exception: after the succession of Viśvanātha Nāyaka to the throne of Madurai in the mid-sixteenth century, the early generations of Nāyaka rulers set out to remap the sacred landscape at the centre of Madurai. They did this by transforming the architectural visage of the temple with the addition of new temple towers (gōpurams) and pavilions (maṇṭapams), and imprinting it with signifiers of an entirely new mythology: namely, the 64 ‘Sacred Games of Śiva’.

Broadly speaking, the expansion of the Mīnākṣī-Sundareśvara temple under the Madurai Nāyakas took place in three main phases.Footnote ¹⁰⁴ Between 1570 and 1600, the temple took on its present shape with the construction of the external wall and four gopurams (gateway towers) of the outermost third prākāra (outer wall) along with the four gopurams of the second prākāra. Subsequently, the early decades of the seventeenth century witnessed further accretions such as the Thousand-pillared Pavilion (Āyirakkāl Maṇṭapam). The remaining structural innovations that grace the temple today were commissioned during the reign of Tirumalai Nāyaka (1623–1659), whose efforts have earned him the reputation of being the chief architect behind the entire programme of temple expansion. Tirumala Nāyaka's innovations include the Putu Maṇṭapam or ‘New Maṇḍapa’ (constructed circa 1626–1633),Footnote ¹⁰⁵ an external festival pavilion adjacent to the eastern side of the temple complex, as well as the towering Rāya Gōpuram which, although never completed, was intended to upstage all similar temple gōpurams across the southern half of the subcontinent.

The early Nāyakas did not restrict themselves, however, to expanding the physical edifice of the temple complex. Beginning around the early seventeenth century—that is, during the latter two phases of temple renovation—they began to enrich the symbolic face of the Mīnākṣī-Sundareśvara temple as well with sculptural and pictorial representations drawn from unprecedented literary sources, particularly the Tiruviḷaiyātal Purāṇam.Footnote ¹⁰⁶ In fact, no such image can be reliably dated to prior phases of temple construction.Footnote ¹⁰⁷ In short, the early to mid-seventeenth century witnessed an explosion of interest in graphic as well as performative portrayals of the ‘Sacred Games’ throughout the Madurai temple complex. Sculptural depictions of four of the ‘Sacred Games’ were displayed in the early seventeenth-century Āyirakkāl Maṇṭapam, and six in Tirumalai Nāyaka's Putu Maṇṭapam shortly thereafter. The seventeenth century also witnessed the first complete sequence of mural paintings—a genre of representation quite popular in Nāyaka-period temple art—of all 64 ‘Sacred Games’. They are displayed quite prominently alongside the Golden Lotus Tank (poṟṟāmaraik-kuḷam) in front of the shrine of Mīnākṣī, the ritual heart of the temple. Moreover, an intriguing detail of the ‘Sacred Games’ statuary reveals that the temple improvements took place simultaneously with, rather than subsequent to, the codification of the Tiruviḷaiyāṭal Purāṇam itself. In both the Āyirakkāl Maṇṭapam and the Putu Maṇṭapam, a statue portrays a game in which Śiva, depicted as a tiger, feeds a deer, a legend that appears in early versions of the ‘Sacred Games’ but has been elided in Parañcōti's TVP and the Hālāsya Māhātmya. The publicization of the ‘Sacred Games’, in short, was part and parcel of its canonization; it is very likely that Parañcōti composed his masterpiece while or after the statues were being erected, and his work certainly had not been fully accepted as part of the canon by the time of the construction of the Putu Maṇṭapam in circa 1630.

Furthermore, and perhaps most importantly, as the TVP began to enter the visual landscape of the Madurai temple, Tirumalai Nāyaka instituted a series of calendrical festivals showcasing several of the ‘Sacred Games’ in public performance. The Stāṉikarvaralāṟu, one of our most detailed sources of the temple's history, chronicles the changing ritual duties within various factions of the temple priesthood over the centuries. In the process it draws particular attention to the new centrality accorded to the performances of the ‘Sacred Games’ under the leadership of Tirumalai Nāyaka. In terms of the codification of public religious culture, however, Tirumala Nāyaka's most influential innovation was the Putu Maṇṭapam itself, an external temple pavilion designed to host many of the temple's festivals outside of the temple walls, thereby externalizing a previously internal, delimited space that may not have been physically accessible to the numerous caste communities who comprised Madurai's residents. Indeed, it is these new calendrical festivals that captured the attention of the temple's ritual officiants when describing the contributions of Tirumalai Nāyaka to the Madurai temple complex:

Spectacle, in fact, is just what the ‘Sacred Games’ had become by the mid-seventeenth century, as visual and performative media rendered the narratives of Śiva's miraculous exploits immediately accessible to a diverse and even non-literate public. Among the numerous games re-enacted in public ceremonies, however, it was the ‘Sacred marriage of Śiva’ and ‘Mīnākṣī’ that would leave the most visible imprint on Madurai's public religious culture. Tirumala Nāyaka's most radical adjustment to the festival calendar, undoubtedly, was to unite the wedding of Madurai's divine couple with the overwhelmingly popular Cittirai festival—in which the city's resident Vaiṣṇava deity, Aḻakar, made his annual procession to the river Vaikai, pausing in his journey to bestow temple honours on the dominant caste groups of the Madurai region. According to prior legend, Aḻakar had journeyed from his temple home several miles outside of town to the middle of the Vaikai river for the express purpose of liberating the sage Muṇḍaka from a curse that entrapped him in the body of a frog. This act of grace had previously been commemorated each year in the month of Cittirai (April–May) by Madurai's Vaiṣṇava residents. When the annual Cittirai festival was conjoined with the Sacred Marriage, Viṣṇu as Aḻakar, whom the ‘Sacred Games’ represented as the brother of Mīnākṣī, was understood to be entrusted with the task of performing the marriage rites for Śiva and his bride. As local legend has it, Aḻakar reaches the centre of the river Vaikai only to realize that the marriage has taken place without him and, in retaliation, returns to his temple without setting eyes upon the divine couple, pausing along the way to spend the night with his paramour, a Muslim courtesan.Footnote ¹¹⁰

By recreating the Cittirai festival, then, Tirumalai Nāyaka managed to draw unprecedented attention to the legend that best encapsulates the royal heritage of Madurai, whose kings are understood as the descendants of Śiva and Mīnākṣī themselves. Simultaneously, he positioned Madurai's most popular religious festival as a virtually unprecedented site for social and religious integration, creating a space that accommodated the interests of diverse caste communities, from Smārta Brahmans to the Kaḷḷar devotees of Lord Aḻakar, and blended the theologies of Śaiva and Vaiṣṇava devotionalism. Other favourites among the 64 ‘Sacred Games’ must have also quickly entered the repertoire of Madurai's residents, as by and large the same ones that were depicted in temple statuary—publicly available all year long as memory sites—were dramatized yearly with festival processions and even mimetic re-enactments of Śiva's divine interventions. Indeed, each of the minor festivals enacting Śiva's ‘Sacred Games’, managed by representatives of rival priestly families of the Madurai temple, provided substantive incentive among elite families to compete for the attention of a wider public.

Facilitated by Tirumalai Nāyaka's royal interventions, then, an increasingly popular literary motif quite rapidly achieved widespread circulation far beyond the literary domain. That the TVP legends did, in fact, circulate is evident from the rapidity with which sequences of the ‘Sacred Games’ began to appear in temple mural paintings across the Tamil region, demonstrating the broad appeal the narratives had achieved even outside Madurai, the city in which the miracles were originally enacted. For instance, mural sequences mirroring the 64 panels emblazoned on the Madurai temple outside the Golden Lotus Tank had appeared in the Naṭarāja Temple of Cidambaram by the late seventeenth century and in the Bṛhadīśvara Temple in Tanjavur by the eighteenth century. Footnote ¹¹¹ Likewise, the ‘Sacred Games’ soon became a common fixture of the material culture of religion in Madurai and beyond, replicated on temple carts across the region, festival textiles for chariots (tērcilai), and miniature paintings. Succinctly, the ‘Sacred Games of Śiva’ were legible for the majority of Madurai's seventeenth-century residents, not from the text of the Tiruviḷaiyāṭal Purāṇam but from the material culture of the Mīnākṣī-Sundareśvara temple and from participation in the collective re-enactments of the legends in temple festivals. In the words of Kim KnottFootnote ¹¹² on the production of religious space, the enactment of the ‘Sacred Games’ ‘does not take place, but makes place’, encoding the temple's visual facade with religious significance proper to multiple distinct communities.

Making religious space: placing publicity at the dawn of early modernity in India

What does religious space—what the classical history of religions has referred to as ‘sacred space’—have to do with public religious culture, with the religious values or frames of reference cultivated by individuals from diverse social and sectarian backgrounds across the city of Madurai? Looking at the temple as public space—a space in which publics move, a space in which publics are created—is a crucial step in transcending both the Western and modernist presuppositions that would unproblematically equate religion with the private sphere, the internal domain of affect or belief. Discounting the truly interior spaces of ritual worship, such as the garbhagṛhas, or inner ‘wombs’ of the temple accessible only to trained priests, the temple pavilions (maṇḍapas) in which festivals are performed serve as physical sites for public gathering and enactment of shared religious sensibilities. A temple in South India, it must be remembered, is constantly bustling with crowds in motion. Indeed, much of the physical space of the Madurai temple, as with the majority of sacred sites across South Asia, incorporates what the Western imagination has understood as non-religious, public space—from shopping complexes to casual gatherings for social conversation. As a result, the images inscribed on such a space, whether murals or statuary, by cultivating a collective public frame of reference, create an ideal readership of visual media. Such images, in other words, are nothing less than social agents, exerting an active influence on the human agents that move through spaces inflected by their signifying capacity.Footnote ¹¹³

To understand how new religious concepts and values come to be publicly accessible, or come to constitute a cornerstone of a particular public culture, then, depends fundamentally on theorizing the public as inhabiting a particular space, constituted in no small part by the visual signifiers that inhabit that space.Footnote ¹¹⁴ Space, in its very materiality, as recent theorists have come to recognize, is by nature a socially imbricated category. Indeed, the materialist turn in religious studies and the humanities at large has become sensitive to space not simply as the Kantian precondition for human cognition but as a site for the signification and contestation of cultural values. From the vantage point of early seventeenth-century Madurai, space—both within the temple and throughout the city at large—was in the process of being overlaid with new conceptual resonance. By visually inscribing the ‘Sacred Games of Śiva’ in temple art and architecture, devotees of Mīnākṣī who traversed the temple halls and circumambulated the Golden Lotus Tank were now confronted with contested claims as to the significance of their spatial practice—that is, their lived engagement with socially significant spaces. But at this particular juncture in history, the ‘Sacred Games’ did not yet fully belong to what Henri Lefebvre would call ‘representations of space’, that is, the normative and fully articulated conventions for how temple space ought to be interpreted.Footnote ¹¹⁵ On the contrary, the public space of the Madurai temple complex can best be characterized as ‘representational space’—a lived space of public contestation, capable of being contested by counter-hegemonic interest groups who aim to overlay shared space with localized layers of meaning.

By the late seventeenth century, the Tiruviḷaiyāṭal Purāṇam had become a household name. The legends were able to permeate Madurai's religious ecology without disrupting its delicate balance—that is, they appealed to the diverse religious publics of Madurai, whether Śaiva or Vaiṣṇava, whether Smārta, Vēḷāḷa, or Kaḷḷar. The temple itself, then, did not homogenize the publics of Madurai, separated as they were by the boundaries of caste, language, and sect, but facilitated the overlaying and intersection of parallel public spheres. Its explosion into literary fashion, as we have seen, took place across multiple literary publics, facilitated by media of performance and patronage centred on the temple site itself. Likewise, its departure from the textual form and entry into public culture was a process mediated by the transference of shared meaning from the written word to visual and performative text.

To this day, religion visibly inflects the public streets of South India—rarely evoking a unified orthodoxy, but subtly interweaving the networks and symbols of discrete religious publics, overlaid with one another in urban space. Extending our analysis to Chennai, a city with which I have become well acquainted over the course of my research, in Triplicane, one cannot walk down a major street without visibly encountering two distinct religious networks, with individuals dressed in either Muslim or Hindu garb, their foreheads bare or marked with ash and a bindu of kumkum, patronizing entirely distinct restaurants and shops that happen to be located a handful of feet from one another. While visibly distinguished by their embodiment of religious identity, these communities move in the very same public space and the street belongs to neither. Such urban pluralism has found a receptive audience in recent years among scholars of the global cityscape, uniting the experience of religious pluralism in contemporary India with the cultural and economic fragmentation of late capitalism.

But is such ‘disjuncture and difference’, in the words of Arjun Appadurai,Footnote ¹¹⁶ the distinctive property of global post-capitalism, a fragmentation produced by the schizophrenia of a modernist mass culture, as Jean BaudrillardFootnote ¹¹⁷ or Fredric JamesonFootnote ¹¹⁸ would have it? This multicentric cultural landscape, at least within the Indian context, has premodern precedents: the urban pluralism of contemporary India owes as much to its early modern antecedents as to the hegemony of economic globalization. By the late seventeenth century, the ‘Sacred Games’ had become a household name and the legends were able to permeate the religious ecology of Madurai without disrupting its delicate balance—that is, they appealed to the diverse religious publics of Madurai by cutting across the boundaries of caste, sectarian community, and language. What took place during this period in Madurai was more than the birth of a narrative or even the birth of a religious canon. It is the entextualization of public space that is at stake. The ‘Sacred Games of Śiva’ were not simply encoded in the memory of a collective populace, but canonized in physical public space, accessible in that it can be read, reinterpreted, and re-enacted over time. The temple became not a space outside of time but a nexus for the temporal and spatial encoding of meaning, a space in which people moved through and performed meaning—a spatial nexus for the overlaying of multiple parallel publics.