Early Traces of <em>Vacana</em>s in Kannada Literature

Many people see the vacanas, short devotional poems in the Kannada language that started to be composed in the twelfth century, as the major vehicle of the attitudes, values, and practices of one of the region’s most prominent religious traditions, known as “Vīraśaiva” or “Liṅgāyat” (the relationship between the two terms being historically complicated). The vacanas are central to the identity of this tradition, and as such they play a major role in the ongoing controversy about whether the tradition is a part of “Hinduism” more broadly or stands apart from it.¹ But even outside of this religious tradition the vacanas are enormously popular among the Kannada-speaking public.² The translations of selected vacanas into English by A.K. Ramanujan in 1973 presented them to a global audience as evidence of spiritual and social rebellion against oppression by the religious mainstream in a way that anticipates modern and Western values.³ Ramanujan did not invent this reception of the vacanas, however; since the late colonial period they had been mobilized for various social and political movements. Today, people quote them in support of a range of positions: for individual devotion and against blind ritualism and superstition; for personal spirituality unfettered by religious orthodoxy; for social upliftment and against the oppression of marginalized groups; for women’s rights, and more.

Examples of invoking vacanas as a premodern precursor to modern and progressive values abound.⁴ But there are also those who read vacanas differently. Against the movement to separate the Liṅgāyat tradition from Hinduism, some scholars quote vacanas to emphasize the tradition’s continuity with other Hindu traditions; some claim, too, that the vacanas quoted in support of progressive values are inauthentic.⁵

A handful of scholars have pointed out the open-endedness with which vacanas are read and the interests that have shaped the multiple interpretations mentioned above.⁶ But some part of the interpretive controversy of what the vacanas are “really” about—although clearly not the entire controversy—rests on the philological question of the “authenticity” of the texts themselves. When we experience a vacana, either in Kannada or in translation, either printed in a book, or performed to Hindustani or Carnatic music, how, if at all, do we account for the history of the text that we are experiencing? The uncritical attribution of the entire corpus to poet-saints of the twelfth century has led to a certain complacency, even on the part of scholars, as if the texts have come to us (or our eyes) directly from the twelfth century. Yet it is readily acknowledged that they have come to us through a long and winding journey across time and media. We are still learning about the vacanas’ textual history, and already it is clear that this body of songs has undergone heavy editing and reformulation at different points.⁷

To read vacanas critically means to consider their complex history of transmission and dissemination over roughly eight centuries or so. This history is punctuated by dramatic shifts in how vacanas were handled. For example, they were collected, edited, and written down as texts for the first time only in the fifteenth century. Until that important moment, the vacanas were transmitted from one generation to the next in oral recitations and performances, perhaps with the aid of written notes—a practice that inherently implies textual malleability and that apparently went on for a period of about three hundred years.⁸ Another important moment in the history of the vacanas started from 1870, with their early print publications: these publications involved the heavy-handed work of culling, reframing, and even rewriting that was guided by contemporary political, social, and cultural agents in the colonial and, later, the post-colonial, periods.⁹ Paradoxically, such historical interventions in the textuality of the vacanas—starting with recording them in manuscripts, then fixing them in printed books, and now making them available in digital media—have contributed to the widespread presumption that vacanas are and have always been an integral and natural part of what we think of as “premodern Kannada literature,” independently of cultural and religious practices, of performances, and of the various interpretations of and responses (or lack thereof) to the vacanas throughout this circuitous history.

Whether intentionally or not, vacanas are almost always presented to their readers and audiences without the apparatus of textual criticism and as the ipsissima verba of twelfth-century poet-saints. This is true not only with regard to translations of vacanas but also to publications in Kannada.¹⁰ In many of these publications, issues such as textual sources and manuscript variation are briefly dealt with in the introduction, if at all, while the vacanas in the body of the publication are rendered without the apparatus of alternative readings. This practice speaks of the exceptional popularity of the vacanas and their unique appeal to large audiences, but it also contributes to the general obfuscation of complexities in the history and textuality of the vacanas. A good case in point for this is the extensive edition of the vacanas edited by Eṁ. Eṁ. Kalaburgi, called Samagra Vacanasāhityada Janapriya Āvr̥tti (The Popular Edition of the Complete Vacana Literature), which contains fifteen volumes. This popular edition was reprinted several times and is commonly used for accessing vacanas. In the introduction to the first volume, Kalaburgi lists the sources he used for this publication and provides examples of text variations in fourteen vacanas, adding an explanation for why a specific variation is, to the editorial board, the “correct one.”¹¹ But the 1,414 vacanas in the body of this publication have no indication of these editorial choices. Furthermore, the historical possibilities embedded in the existence of the variations are never considered, and questions about the textuality of the vacanas before they were collected in manuscripts are absent altogether.

vacanas are thought of by many today as an outstanding form of Kannada literature. But this appreciation appears to be a late one. In the first centuries after the appearance of vacanas, very few Śaiva authors minimally refer to them and the literati outside the circle of Śaiva devotees never make any mention of them.¹² The idea that the vacanas were not considered a literary event in their own time was articulated in 2005 in an unpublished dissertation by Prithvi Datta Chandra Shobhi that locates the textualization of vacanas in the rise of the Viraktas of the fifteenth century, during the Vijayanagara period. Chandra Shobhi writes:

Chandra Shobhi argues that we do not have access to the vacanas before the fifteenth century—supposedly the first time they appeared in writing—and that the silence by earlier Śaiva authors is indicative of their lack of recognition or appreciation of the vacanas and of the vacanakāras, their composers, as such. Chandra Shobhi presents a radical alternative to the popular reading of the vacanas as authentic testimonies of the twelfth century, arguing that the absence of any textual source before the fifteenth century means there is no direct access to vacanas of the twelfth century. Accordingly, Chandra Shobhi goes on the analyze in his dissertation the massive textualization of the vacanas in the fifteenth century as the earliest moment for making a cultural sense of these poems, one which reflected communal anxieties of Vīraśaiva communities of that period.

In this article I probe a portion of the early Śaiva narratives that Chandra Shobhi refers to in the above quotation, specifically life stories of vacanakāras and their associates. I show that, contrary to Candra Shobhi’s claim, these texts do contain descriptions and quotations of what we recognize today as vacanas. Significantly, however, these references are infrequent in the examined set of stories and are sometimes incomplete, inconclusive, or obscure. These findings do not sit well with the popular wholesale embrace of the vacanas as products of the twelfth century nor with the categorical doubt that Chandra Shobhi casts regarding the possibility of having a “direct access” to the vacanas before the intervention of the Viraktas. Instead, the findings presented in this article lead to a more fine-grained understanding of the vacanas’ textual reception before the fifteenth century, according to which the vacanas, though certainly present in the early hagiographies, remained very much in the narratives’ margins, and were associated with only few of the saintly figures. Put differently, while the vacanas certainly reverberated in the early written texts about the local devotional culture, they did not receive a significant amount of attention from the early Śaiva authors who recorded in their works the emerging devotional (bhakti) tradition in the Kannada-speaking region.

A recognition of the liminal status of the vacanas in the earliest devotional texts opens up a new set of historical questions. Above all, it raises questions about the vacanas’ presence and role in the early history of Śaiva devotion in the Kannada-speaking region. If the vacanas were indeed the harbinger of a new local devotional culture, as commonly thought of today, why did the authors who wrote about it in the proceeding decades and centuries dedicate so little space to them in their writings?¹³ Given the fact that the saints of the twelfth century are today thought of primarily as vacanakāras, “composers of vacanas,” how do we account for the fact that the first authors who wrote about these saints did not think of them in this way? And what might this apparent lacuna indicate about the reception of the vacanas and the historical circumstances of the devotional community in its earliest stages?

The body of the earliest Kannada and Telugu narratives referring to the twelfth-century model devotees this article focuses on dates to the early thirteenth century. It contains a variety of authors and styles, but at its core we can clearly identify three poetic mavericks among the group of early hagiographers. Each of these authors introduces in his works radical departures from contemporaneous works in terms of literary practices and religious visions, departures that would have a lingering effect in succeeding centuries. The first two of the three—Hampeya Harihara and his nephew Rāghavāṅka—wrote in Kannada, and the third, Pālkuriki Sōmanātha, wrote mostly in Telugu (but also in Sanskrit, Kannada, and other languages). Considered together, these three poets represent a major shift in regional literature that undoubtedly echoed a larger historical process: the introduction of devotional religion as a major cultural and social force.¹⁴

In general, the vacanas were not what Harihara, Rāghavāṅka, Sōmanātha, and their immediate followers wanted to highlight about these saints. In their writings, the saints’ devotional sentiment was communicated not via lyrical utterances or public discourses but through religious action that included public and private worship, miracle making, and care for the devotional community. This article will focus on the work of Harihara. Judging by a casual examination of the other extant works, his treatment of vacanas is similar to that of all his near contemporaries.¹⁵

Harihara was a prolific author who composed his poetic works in Hampi in the early thirteenth century (close to two hundred years before the establishment of the Vijayanagara Empire in the same place).¹⁶ In terms of its content, his poetic oeuvre is devotional, and for this reason he can be considered an innovator against the prevailing trend of courtly narrative poetry. Stylistically, the genres and styles that he used involved both traditional practices and new ones. Harihara composed several śatakas (hundred-verse poems), one aṣṭaka (eight-verse poem), and one acclaimed mahāprabandham (court epic).¹⁷ Although these compositions were written in familiar pan-Indian styles, they nevertheless present stylistic and thematic innovations in a clearly articulated and self-confident devotional voice. They do not, however, directly address the historical appearance of the twelfth-century saints in the Kannada-speaking region, their remarkable exploits, or the poetry associated with them. It is another work by Harihara—a collection of stories written in the ragaḷe meter and conventionally referred to as the Śivaśaraṇara Ragaḷegaḷu (Ragaḷe Stories of Śiva’s Saints, eponymous of the meter) that presents for the first time in writing the early Śaiva cohort of devotees from the Kannada regions, including major figures such as Basava, Allama Prabhu, and Akka Mahādēvi.¹⁸

The fact that the Ragaḷe stories were the first written account of the Kannada saints contributes to this collection’s high ranking in the list of extant textual sources about these figures. Harihara composed these stories shortly—perhaps as little as a few decades—after the times of these saints, and in the same region of today’s north Karnataka.¹⁹ The lives of the saints are presented in a linear and straightforward manner, from before the figure’s birth to after his or her passing away, with obvious highlighting of significant moments in the saint’s religious career. The portrayal of the saints in the Ragaḷe stories is remarkably lucid, and it conveys the poet’s deep familiarity with the local, recently formed devotional culture.²⁰

Considering Harihara’s careful narrative crafting of these stories, one might expect to find a profusion of detail about the vacanas and the contexts in which they were created. However, the Ragaḷe stories have little to say about the vacanas, and what they do say is uneven and does not fully cohere with later configurations of the vacanas’ history. In this sense, Harihara’s treatment of the vacanas further problematizes our historical understanding of them. His view regarding the marginality of the vacanas is made evident in the generally minute space they occupy in the saints’ lives; the majority of the saintly figures in Harihara’s text, including those whose vacanas became their main claim to fame, are mostly appreciated for devotional exploits and not for composing poetry.

Above all other aspects of their lives, the Ragaḷe stories celebrate the saints’ unwavering determination to remain exclusively devoted to the god Śiva. This determination is expressed in myriad forms, of which the most apparent is their impassioned worship of the god. In addition, they are remembered for exceptional acts and miracles, and these occur mostly in the context of public competition against agents of other religions or in situations of social crisis. Another recurring feature of these life stories is the saints’ support of the local community of devotees by, for example, providing them with money, jewelry, clothes, and food, and organizing collective worship events.

Moreover, in the passages in which Harihara refers to the saints’ devotional poetry, he does not focus on their exceptional messages, their unique mode of expression, or their conversational contexts—all the distinguishing characteristics of vacana literature that came to be most recognized and appreciated and continue to be so today.²¹ Rather, Harihara usually mentions devotional poetry as a stock ingredient in a fixed grammar of ritual that takes place either in private worship or in public communal worship, but even this reveals very little about the performance of vacanas. Despite Harihara’s sensibility for composing devotional poetry in Kannada, attested to in the Ragaḷe stories and in his other works,²² his mentioning of devotional songs, with a few exceptions discussed below, is short and general, even offhand. Thus, for example, when Harihara describes the devotional gatherings in which Kōvūra Bommatande participates, songs are mentioned nominally to explain the saint’s excitement with no reference to their content or to their performative or dramatic context:

It is difficult to know from this passage whether the songs Bommatande heard were what later became known as vacanas, since Harihara refers to them as gīta, which means simply “song,” and attributes them to the purātanaru, or “elders,” a label that might refer to earlier devotees from other regions.²³

Harihara does occasionally use the word “vacana” in the eighteen Ragaḷe stories about saints from the Kannada-speaking region but never in relation to any recognizable vacana.²⁴ Read in context, the word “vacana” in the Ragaḷe stories most likely means an “utterance” and not “vacana” in the sense that we know the term today.²⁵ Another set of related terms Harihara uses are gadya (“prose”) and padya (“verse”), which he sometimes invokes as a pair when relating to verbal expression. In Kannada poetics, the former term overlaps with vacana in the general sense of “prose,” while the latter signifies a versified text.²⁶ The poems we call today vacanas are, with few exceptions, not metrically arranged, and therefore correspond to the general meaning of vacana and of padya in the sense of “prose,” although many are rhythmically patterned.

With these general observations in mind, we now approach specific Ragaḷe stories in Harihara’s collection that bear upon the question of the presence and significance of the vacanas in the early period of the tradition.

As noted, in specific cases in which Harihara mentions saints’ poems, he usually describes them in a conventional manner. Missing is any reference to the vacanas’ style and themes as well as any definite sense of how they came to be known with time. A case in point is found in the story about Koṇḍaguḷi Kēśirāja, a Śaiva poet who is described as a prominent leader of the Śaiva devotional community and an important political figure in Kalyāṇa, presumably several decades prior to Basava’s tenure at Bijjaḷa’s court. Kēśirāja’s poetry is not a central theme in the Ragaḷe about him. It is referred to in only eight verses in the opening section of the text. There, Harihara writes:

This passage is remarkable in the context of early sources about devotional poetry in Kannada in terms of the details it provides. Astonishingly, the rare passage is markedly not about the vacanas. The terms Harihara uses, such as “verse (padya),” “meter (chanda),” “grammar (vyākaraṇa),” and “ornaments of his verses (vr̥ttakk’ alaṅkāra[ṁ]),” are all conventional technical terms in Sanskrit that associate Kēśirāja’s songs, even if only generally, with conventional composition in traditional styles, which is very different from the way in which the style of vacanas has come to be considered.²⁷ None of Harihara’s descriptions in these verses is associated with what distinguishes the vacanas, such as the absence of meter and other poetic conventions, the poet’s signature line (aṅkita) at the end, the poem’s lyrical content, biting social critique, and so on.

Further evidence of the disconnect between Kēśirāja’s poetry and the vacanas is found at the beginning of the above passage with the phrase: “his meters were all made up of five syllables (pañcākṣarānubandhaṁ chandav’ anuv’ āge).” The five syllables are of course namaḥ śivāya, the most important mantra of Śaivism. In its six-syllable form (ōṁ namaḥ śivāya), it corresponds with the title of Kēśirāja’s most famous composition, the Ṣaḍakṣara Kanda (“Treatise of the Six-Syllable Mantra”).²⁸ The treatise, dated to the early twelfth century, is among the earliest Śaiva devotional works in Kannada, but unlike the vacanas it is lengthy and composed in a traditional style and meter, and therefore cannot be regarded a vacana. It should also be noted that no vacanas are attributed to Kēśirāja by the later tradition.

The only additional mention of Kēśirāja’s songs in this Ragaḷe can be found a few dozen lines further, in a passage that describes a devotional assembly led by Kēśirāja. Harihara writes:

It is difficult to determine whether Harihara refers in this passage to two different bodies of poetry—songs of the elders and Kēśirāja’s own verses—or whether the two belong to the same collection. The first reference is to songs of elder devotees (hāḍuva purātanara gīta[ṁ]), while the second reference is to verses to Śiva that Kēśirāja recites (śivana padyavan ōdi). But the use of “recites” in the second reference is noteworthy, because it implies that these verses were pre-composed—not extemporaneous, which is how the tradition usually understands the composition of vacanas. In addition, in both passages quoted here from the Kēśirājadaṇṇāyakara Ragaḷe, Harihara refers to “verses” (padyas), and this formally distinguishes them from vacanas, which, with the exception of a small subset composed in triplets (tripadi), are written without meter. Finally—and this is characteristic of Harihara’s reports of the saints’ songs or verses—we learn next to nothing about their outstanding content or message beyond devotional conventions that are ubiquitous in the text.

Thus, although the passages about Kēśirāja’s poetry are among the most pronounced sections of Harihara’s writing about devotional singing by the saints in the Kannada-speaking region, these are by any measure brief and do not shed light on the early reception of vacanas. They simply describe devotional poetry that was composed in traditional forms and was not recognized for groundbreaking messages or exceptional discursive quality, nor for the poetic features that distinguish the vacanas. Based on these passages, it is reasonable to assume that, in the devotional culture of the Kannada-speaking region of the period under discussion here, devotional poetry and its performance did not narrowly imply vacanas as they are imagined today: at the very least, devotional songs in more traditional forms were being composed and appreciated. The remaining sections of this article discuss the three most celebrated saint-poets of this tradition: Allama Prabhu, Basava, and Akka Mahādēvi. We will see that Harihara’s treatment of the poetic oeuvre of each of these saints complicates the historical understanding of vacanas in a different way.

As I argue elsewhere, Allama Prabhu, who is today considered among the most prolific and prized composers of vacanas, is not recognized as such in Harihara’s version of his life story.²⁹ Furthermore, in contrast to the later tradition, in which Allama is portrayed as a staunch and polemical spiritual leader of the nascent devotional community, Harihara’s portrayal of Allama as a saint focuses on his wandering as a reclusive mendicant who shuns the company of other. Harihara’s approach to Allama’s solitary sanctity leaves limited scope for presenting his celebrated lyricism or for describing at length his dramatic encounters with other spiritual figures, both of which are hallmarks of his later biographies.³⁰

In the Ragaḷe story about Allama there is only one passage that quotes a poetic utterance by this saint. This utterance, consisting of words of praise for Allama’s guru, is extemporaneous, just as vacanas are supposed to be, but in terms of form it is organically woven into the text using the ragaḷe meter and hardly resembles a vacana, which should be in rhythmic prose and include a “signature”:

Allama’s other direct speeches in this Ragaḷe are not lyrical and in general are terse and descriptive.³¹

In contrast to Allama himself, other figures in the Ragaḷe are quoted expressing their wonderment at the sight of him, but their appreciation focuses on Allama’s renunciation, his practice of carrying of the liṅga in the palm of his hand, and his spiritual merits, without making any mention of his poetic acumen or verbal skills.³² Although in this text Allama is claimed to have met with two other important devotees, Basava and Siddharāma, Harihara’s report of these important encounters is minimal, consisting in total of just six short lines, and he does not quote any direct conversation.³³ This can be put in sharp relief against the elaborate drama in the much later Śūnyasampādane (“Reaching Nothingness”) works about Allama’s stay in Kalyāṇa during his tenure there next to Basava, which is replete with vacanas.

The Ragaḷe story about Basava stands out in the collection as the longest and most developed piece, even in its extant form, which is about half the length of the original. The sheer length of Harihara’s version of Basava’s life offers a profusion of details about his remembered story. Fortuitously, this also includes references to Basava’s poetic compositions, although the space allotted by Harihara to this aspect of Basava’s life is quite limited:³⁴ Of the thirteen chapters, containing together about 1,200 lines, direct quotations of vacanas consist of only two short lines in chapter 12, and perhaps another short fragment in chapter 13.³⁵

To complicate things further, the actual task of identifying what might be considered a vacana quotation in the text is not an easy one. A few verses scattered throughout this long work quote Basava addressing his god or a fellow Śaiva with a devotional appeal, but these verses are difficult to identify as recognized vacanas. To illustrate: a passage in chapter 3 with a quotation of Basava praising Śiva reads more like traditional devotional poetry with stock descriptions of the god and a refrain at the end of every line; it does not meet the expectations of a vacana in terms of either structure or content.³⁶ Five other short quotations addressed to the god could be read like vacanas, but I was not able to find any similar statements in the published corpus of vacanas.³⁷ Three words of praise to Allama appear to correspond with the concluding line of a known vacana, although the verb is different.³⁸

Harihara’s own description of Basava’s songs in the end of chapter 9 and beginning of chapter 10 is telling in terms of Harihara’s appreciation of them. The passage comes immediately after a miraculous feat: Basava’s resuscitation after his life breath has left his body in his sleep to follow roaming devotees. The following passage celebrates Basava’s recovery:

Chapter 10 continues this theme in three opening verses that quote the devotees’ reciprocal eulogy:

In terms of understanding the reception of vacanas in the early period, this passage both reveals and obscures. We learn from it that Basava composed original songs of his own when inspired by the collective worship of the god and the company of fellow Śaivas. We also learn that the songs he composed were devotional in nature and focused on Śiva in his specific manifestation called “Saṅga,” which is a variant of “Kūḍalasaṅgamadēva,” Śiva’s name in many of the vacanas attributed to Basava. These descriptions can be easily read as direct testimony to the fact of Basava’s composing vacanas. Furthermore, according to this passage, the songs he composed affirmed the community of devotees, expressed hostility toward other religious traditions, and had a strong ecstatic quality to them. Such descriptions are corroborated by some of the vacanas attributed to Basava.

And there is more: the opening verses of chapter 10 tell us that Basava’s songs were disseminated widely, were possibly accompanied by musical tunes,³⁹ were sung in praise of Śiva, and purported to summarize all mainstream and traditional religious knowledge.⁴⁰ This passage therefore appears to confirm that already in Harihara’s time Basava was thought to have composed devotional poetry. Nevertheless, much remains unknown: What was the nature of this poetry and what was its flavor? In what style was it composed and how long was it? And what messages did it convey—were the messages radical and, if so, in what way? As much as the above passage extols Basava’s compositional craft in dramatic ways, it does not reveal much about the form and content of the poetry itself, and it leaves questions such as these unanswered.

There is one additional episode in Basava’s life story in which his songs appear, and it is here that the most pronounced reference to what we recognize today as Basava’s vacanas is found in the whole collection. The incident itself, located in chapter 12 of the work, revolves around a Śiva devotee who lives in Kaliṅga, in today’s Odisha. The devotee regularly attends assemblies of worship and recitations.⁴¹ In one of these, while listening to someone publicly singing devotional songs, the devotee hears the following song attributed to Basava:

Hearing this, the devotee, whose name suggests he is a merchant, concocts a ruse to extract gold from Basava based on the “blank check” the latter has rhetorically offered in his song.⁴²

The line in Harihara’s Ragaḷe might accord with what we think of in terms of a vacana, both syntactically and semantically. In terms of syntax, the quotation, which addresses the god as many vacanas do, culminates with the familiar signature line of Basava, which is the name of his chosen deity Kūḍalasaṅga. Semantically, this quotation presents a dramatic, indeed life-threatening, promise by the devotee to give up his life if he fails in his devotional commitment. Even Basava’s vow in this line, of cutting off his own head (taledaṇḍa), corresponds well with the local idiom of this devotional milieu, as a token that is repeatedly woven around the life of Basava and his fellowship.⁴³

The passage in which this verse appears identifies the poetic quotation as a song (gīte) by Basava that the devotee hears during a performance of “prose and verse composed by the elders.”⁴⁴ In the Kannada discourse of poetics, the term “prose and verse” (gadyapadya) describes two distinct forms of poetic expression. Indeed, gadya and vacana both mean “prose” (though the former term is usually associated with longer prose compositions). And, most significantly, even though the term vacana is nowhere to be found in the passage, the verse quoted above is a part of a published vacana attributed to Basava. The vacana is numbered 1053 in Kalaburgi’s publication of vacanas.⁴⁵ The full vacana in this edition reads:

Another quotation of a vacana soon follows. As the greedy devotee arrives at Basava’s place to test him, Śiva appears in disguise before Basava and worriedly admonishes him about the morass his song has generated. Basava, however, smiles, tells the god not to be afraid, and then declares unperturbedly:

Basava’s address to his god is striking in its boldness. We can note the saint’s proud refusal to wear the sacred thread, which is a Brahmanical sign of Vedic initiation and social privilege. We can also note Basava’s challenging of the god by calling him a coward. (At the end, Basava’s sharp words prod Śiva to arrange for the delivery of gold and precious stones to untangle the financial mess prompted by Basava’s earlier vacana). As in the case of the previous quotation, Basava’s speech here is almost identical to what is today recognized as vacana no. 701, with a slight variation in spelling and word choice.⁴⁷

The core message in the two vacanas discussed here is Basava’s unbounded commitment to provide for the material needs of any Śaiva devotee, and it corresponds well with Harihara’s understanding of Basava’s life goal as the author presents it in the framing story at the beginning of this Ragaḷe. The Ragaḷe about Basava opens with his prenatal life in Kailāsa, Śiva’s heavenly abode, where the divine attendant Basava, while passing out food graced by Śiva to all those present in the hall, mistakenly skips Śiva’s son Skanda. Śiva’s punishment for Basava’s failure to distribute the grace of the god to one and all is to spend a lifetime as a human on earth, where he will provide for all the devotees of Śiva, without exception. In Śiva’s own words:

Indeed, the story Harihara weaves around the life of Basava stresses the saint’s commitment to caring and providing for the community of devotees, and Harihara explicitly states this several times in the text. The vacanas quoted in Harihara’s story pertain to precisely this theme. By contrast, we encounter no vacanas on the themes for which Basava, as a composer of vacanas, was famous in later periods: his lyrical self-inquiry into the nature of devotional sentiments and his adamantine resistance to external and social constraints. These elements can be found in the fabric of Harihara’s literary version of Basava’s life but are not expressed in vacanas.

In chapter 13 there is yet another quotation that perhaps can be linked to a vacana. This chapter is the last in the Ragaḷe as we have it today, and it appears to be an amalgamation of disparate passages cobbled together. In the beginning of the chapter, Allama pays a quick visit to Kalyāṇa and teaches Basava about the śivaliṅga. Basava expresses his admiration to Allama by singing a song, from which Harihara quotes only a few words:

This short line is similar in meaning to the end of the four-line vacana numbered 1303, although the words are different.⁴⁸

The fact Harihara quotes only two, or perhaps three if we count a fragmented phrase in chapter 13, of what later will be recognized as Basava’s vacanas, and that these quotations narrowly focus on Basava’s care for the material welfare of Śaivas, speaks volumes regarding Harihara’s understanding of what is at the core of Basava’s historical importance, which is his exceptional sponsorship of devotees, and what is not, which is his criticism of social practices or religious rites.

Of Harihara’s portrayals of saints, that of Akka Mahādēvi (also known as Mahādēviyakka) has the most elaborate treatment of the subject’s original poetic compositions.⁴⁹ This exceptional attention makes sense in the broader context of Mahādēvi’s saintly persona, since the poetry attributed to her clearly played a central role in the spread of her fame well beyond the Kannada-speaking region.⁵⁰ In general, Mahādēvi’s vacanas express a woman’s rejection of worldly familial ties and longing for physical unity with the god, and the defiant voice in her vacanas corresponds with her traditional story—including in its earliest rendition by Harihara.⁵¹

The first mention by Harihara of Mahādēvi’s songs appears in chapter 5 of the Ragaḷe story about her, amid a long lyrical section describing Mahādēvi’s emotionally laden worship of the liṅga. Here, Harihara writes:

As in the case of other saints, here too Harihara describes his appreciation of the saint’s poetic oeuvre while divulging little about the nature of the songs beyond their devotional intensity. The four-line quotation at the end of the passage can be read poetically, although it is difficult to determine if this lyrical outpouring is a vacana: while it does carry the pleading and lyrical voice recognizable in many of Mahādēvi’s vacanas, it is in the ragaḷe meter (and not in prose as most vacanas are), lacks a concise message, and does not bear a signature line. Furthermore, a few verses earlier we were told about “auspicious songs sung with love by the supreme devotees”⁵⁴ to the sounds of conch, drum, cymbal, which is the conventional style of devotional singing to which Mahādēvi’s songs might have belonged.

In chapter 6, Harihara again quotes Mahādēvi singing:

In contrast to the previous quotation, this one is almost identical to and clearly identifiable as vacana no. 88 in the published version.⁵³ A few minor changes in that version stand out: (1) the pronouns appear in a different order, (2) the language register is slightly updated (e.g., from pēḷven to hēḷuven), and (3) the name of the deity whom Mahādēvi addresses has changed from “Śiva” to “Mallikārjuna.” This latter change is telling in the sense that it might point to an editorial need in modern times to fix the signature phrase of a vacana composer in more hermetic ways than in Harihara’s time (although in other parts of the text Harihara does identify Mahādēvi with the specific manifestation of Śiva as Mallikārjuna).

A bit further along in the same chapter Harihara again quotes Mahādēvi’s words:

Vacana no. 350 in the printed corpus is shorter than what we find here, and there are certain differences between the two renditions, but the two texts generally correspond in letter and in spirit.⁵⁵ One major difference is that in the version transmitted as a vacana, Mahādēvi asks the god to lead her to Srisailam, the famous Śaiva temple complex in Andhra Pradesh, while in the Ragaḷe version she asks to be taken to Kailāsa. As noted earlier, such a variation could be explained by a need in the later tradition to align more strongly Mahādēvi’s vacanas with her life story, in which she travels to Srisailam to unite with her god.

Another relevant passage appears toward the end of the same chapter, at the climactic moment of Mahādēvi’s desertion of the palace and her married life. Forced to remain disrobed in front of her husband, parents, guru, and fellow devotees, Mahādēvi formally announces the termination of her marriage agreement. She places her personal liṅga in the palm of her hand, bids farewell to her parents and guru, hands over her jewelry to the Śaiva devotees who stand there, and walks away naked. At this point, Harihara quotes Mahādēvi saying:

This passage can easily be read as vacana, among other reasons because of the concluding line in this utterance (“O beautiful Mallikārjuna…”), which is identical to the format of signature line found at the end of many vacanas. The signature line usually presents a reflection that brings home the preceding statements in the body of the vacana and is addressed to the devotee’s chosen manifestation of the god. In this passage the concluding line indeed captures the message that pervades in the body of the passage with the proclamation addressed to Śiva Mallikārjuna: “I have no fear.” In spite of this and other indications that what we have here is a vacana, I was not able to locate a vacana that resembles this passage in Kalaburgi’s publication.

In chapter 7, Mahādēvi is confronted by Kauśika, her non-Śaiva husband, who pretends to have had a change of heart and become a Śaiva devotee. Harihara quotes Mahādēvi’s response in the form of a reflecting song:

Here, Mahādēvi expresses her deep frustration and the difficulty she experiences in facing the lingering shadow of her previous married life. Her words conform to the vacana traits we mentioned earlier, and indeed vacana no. 303 in the popular edition is similar to what is quoted above.⁵⁶ There are also marked differences: after the first four lines, the two versions diverge, until they converge again with the concluding two lines (right after “Pray, O ocean of mercy!”). However, in Harihara’s version, Mahādēvi asserts “I fear your illusion!” (añjuven) while in the published version she states the opposite: “I am not afraid of your illusion!” (nān añjuvaḷ’ alla). Also, in the last line of the vacana in Harihara’s version, Mahādēvi beseeches the god for mercy (karuṇipud’), but in the published version she exclaims that this is the command of the god (in second person, nimm’ āṇe). Although the degree of variation between Harihara’s text and the published version is considerable, the two versions share enough of the vacana content and form to recognize them as the same vacana.

Finally, very close to the end of the chapter and of the Ragaḷe as a whole, Harihara mentions in two verses a meeting between Mahādēvi and another famous devotee named Kinnara Bommatande. Mahādēvi sings a song for the occasion, but the author does not provide any further details.⁵⁷

Unlike the Ragaḷes about Kēśirāja and Allama, but with some similarity to that about Basava, the Ragaḷe about Mahādēvi conveys Harihara’s genuine appreciation of her original songs and their inherent role in her life story. And yet, these vacanas appear in only a few, very specific moments. This limited appearance is nothing like the voluminous body of vacanas attributed to Mahādēvi in later texts. Considered together, Harihara’s sparse treatment of the vacanas in the lives of Allama, Basava, and Mahādēvi, as compared to what we might expect given their later reputation as “authors of vacanas,” compels us to reconsider our understanding of vacanas in their early history.

In concluding this article I would like to reflect on the vacanas as they were received during the first centuries of their appearance based on the evidence I have presented above.

The songs quoted in this article might appear substantial when read in sequence, but they are in fact a minute portion of Harihara’s voluminous Ragaḷe Stories. When examined against the full breadth of the work, which is a collection consisting of more than ten thousand lines in its current form, the references to what can be ascertained as vacanas are astonishingly brief, few, and casual.⁵⁸ Moreover, the only two figures to whom Harihara clearly attributes what can safely be considered vacanas—Basava and Mahādēvi—also have the longest stories in the collection (with 26 and 7 chapters, respectively). The fact songs by these two figures are not major in the respective texts again highlights the limited importance of vacanas in Harihara’s rendering of the saints of that period. Notably, we find in Harihara’s text nothing like the celebration of vacanas in the much later Śūnyasampādane, where long conversations between devotees (Mahādēvi included) are uttered in vacanas, and the story is itself a dramatization of the vacanas.

It is meaningful that Basava and Mahādēvi are the only model devotees in Harihara’s work who are directly linked to vacanas as we have them today, especially when considered against the absence of vacanas associated with Allama Prabhu, who since the fifteenth century has been regarded as a leading vacana composer and an important figure in the tradition but receives only limited attention from Harihara.⁵⁹ It is no less than astonishing that Harihara mentions nothing of this saint’s prolific vacana composition, the subject of much attention in later periods, including today.⁶⁰

When we look closer into the few instances of vacana quotations in the Ragaḷe stories, several additional features emerge. One is the fact that because Harihara does not label the quoted lines as vacanas and refers to them only as “songs” (gītegaḷu), it is difficult for the Ragaḷe listener/reader to tell them apart from traditional styles of devotional poetry that are referred to in the text, such as in the case of Kēśirāja or in descriptions of public singing by crowds of devotees. Harihara seems to acknowledge some newness in the songs of Akka Mahādēvi (nūtana), but he reveals little about their qualities beyond devotional merit, which he describes in very general terms. This point is significant, I believe, because it suggests a lack of appreciation of the vacanas’ uniqueness by Harihara or more broadly during this period, and even a lack of distinction between vacanas and other popular forms of devotional expression. The widely accepted understanding of vacanas as an ingenious and indigenous style in Kannada, unique in both its form and content, an understanding that has been so essential since the fifteenth century and is so even more strongly today, and which is captured by the label vacana, highlights the absence of such an appreciation in the early period.

A more technical aspect of vacanas as they appear in the Ragaḷe stories is that vacanas quoted in this text are not easily distinguishable from Harihara’s own prose. With the exception of one vacana by Mahādēvi, the vacana quotations appear in the prose chapters and not in those written in the ragaḷe meter.⁶¹ Harihara’s literary prose is often styled with repetitions, syntactical patterns, metaphors, and so on, and it is possible that Harihara felt that vacanas fitted better in his prose. In any case, this highlights the meter-less structure of vacanas and the fact that, as utterances, they were closer to “prose” than to “verse,” very much in accordance with one meaning of the word vacana as “prose.” Further, this coheres with the fact that until the nineteenth century and the advent of print, vacanas were written as continuous prose text without line breaks.⁶² Harihara’s referring to vacanas as “songs” complicates their labeling as “prose” (or even as “vacanas”), since the word “songs” traditionally connotes singing to music while the word “prose” does not.⁶³

A matter related to the absence of meter is the malleability of the language used in the vacanas, which becomes evident in a word-to-word comparison between Harihara’s quotations of vacanas and how they appear in current publications. The amount of language change found in these comparisons is, in my view, greater than that which would be expected in routine manuscript variation. It not only testifies to the importance of oral performance and of the performer in the transmission of vacanas but also foregrounds the textual malleability of these poems in terms of spoken registers through the ages. Above all, it reminds us that vacana literature, like any devotional poetry in India, is a lived genre that is experienced in performative and other live settings, which is very remote from the Western notion of premodern literature as frozen in its time. Put differently, we must change our expectations of “literature,” as a fixed form of text, when we deal with vacanas.

Returning to the starting point of this article and the larger question regarding the status of the vacanas in the early centuries of the Kannada tradition, it is possible to say that the authors trained in Kannada literature who first began writing about the devotional culture promulgated by the twelfth-century saints did not see the vacanas as something that required “textualization,” that is, their being written down and collected for posterity. In order to communicate devotional attitudes and behaviors to their audiences in written form in their own compositions, the authors chose not to record vacanas by saintly figures but to tell their life stories.⁶⁴ In light of this remarkable but quite apparent conclusion, we might ask: What was the status of the vacanas during this period?

The most radical possibility, both historically and politically, is that the vacanas were marginal for the emerging tradition during the early period and/or that the corpus of vacana literature was dramatically expanded after its moments of origins in the twelfth century. Although such a claim has been suggested by some historians, I find it farfetched, as it rests on the idea of a wholesale “fabrication of tradition” starting from the fifteenth century, an idea that seems improbable in light of the vastness of the vacana corpus, its originality, and its spectacular success and influence, even on the early authors, albeit indirectly.

Another unlikely possibility is that vacanas were performed, transmitted, and appreciated largely by a different audience than that of literary works. We do not know much about the public context in which literary works were composed nor about the vacanas’ performative context. The Kannada authors Harihara and Rāghavāṅka are only tenuously associated with political centers, and even less with religious ones, and we are ignorant about how their written texts, too, were performed.⁸² Notwithstanding this lacuna, one could argue that the vacanas in this period won appreciation elsewhere, perhaps in less literate circles where oral performance of vacanas was distinct from that of written literature. But making such claims on the basis of the canon of devotional literature as we have it today, or even as we inherited it in writing since the fifteenth century, seems precarious. In addition, the style with which Harihara composed his Ragaḷes betrays his fascination with oral and popular forms, and from the analysis in this article it is evident that he was not averse nor ignorant of (at least a few) vacanas.

Another explanation, and one that is most compelling in my view, is that the vacanas were circulated orally in the same communities for which the devotional authors composed their written texts, but that these authors did not feel compelled to write them down or to elaborate on them. The early authors saw the vacanas as part of conventional songs performed by devotional communities; they could not, and did not, recognize or acknowledge the vacanas’ unique significance as written literature. Such a hypothesis should not be read as radical in any way, because devotional poetry in a larger sense was circulating and available, and the poems we recognize today as vacanas were to some extent indistinguishable within a larger body of oral devotional poems, sung or recited in performative contexts in different parts of South India, perhaps as early as the sixth century in the Tamil region.⁶⁵ What was missing in this period was not the vacanas themselves but a literary recognition of their uniqueness and a public appreciation of their messages, vision, literary form, and performative context—all the elements that are signified by the much later label vacana. This possibility opens up new ways of imagining the development of an original literary form over time, in this case originating as an organic part of a pre-existing oral culture of devotional songs and evolving into a clearly identifiable and distinct textual corpus that is celebrated and argued over by different religious traditions, literati, and social agents in later periods.