Ancient Narratives of South Asia

The Encyclopedia of the Novel - Peter Melville Logan 2014

Ancient Narratives of South Asia

Lawrence McCrea

South Asian literature, in the broadest sense of that term, begins with the Vedic scriptures, the earliest of which is the Rigveda (ca. 1500—1000 BCE). The earliest portions of the Vedic corpus consist of versified hymns, typically addressed to one or more divine beings and invoking their blessings. These hymns contain frequent allusions to what must have been well-known stories of the deeds of the gods, but these are generally brief and cryptic. A few hymns take the form of dialogues, but these too are opaque and allusive; there is little in the way of straightforward narrative. The later Vedic canon, particularly the prose texts called Brahmanas (ca. 1000—500 BCE) that grew up around the earlier collections of hymns, contains many, mostly brief, narrative sections, some providing context for the hymns themselves, and others explaining or providing justifications for elements of the elaborate rituals prescribed by the texts.

The Mahabharata and the Ramayana

Full-fledged narrative literature comes into its own for the first time with the emergence of the two great Sanskrit epics, the Mahabharata and the Ramayana. These texts would appear to have developed out of an oral EPIC tradition, and presumably underwent a long process of development, but took on something close to the current shape around the beginning of the first millennium CE. These two vast texts—the first consisting of roughly 25,000 verses, the latter of 100,000 even in its shortest versions—are remarkably similar in their form and language, and represent a major break with previous Sanskrit compositions. Both texts, in addition to presenting stories of far greater length and complexity than any produced up to that time in South Asia, have highly sophisticated NARRATIVE STRUCTURE, involving multiple-FRAME stories and self-referential descriptions of their own composition and transmission.

The Ramayana is traditionally considered to be older than the Mahabharata, though most modern scholars, on linguistic and stylistic grounds, consider it to be somewhat later (see LINGUISTICS). It is regarded as the “first poem,” and the poem itself describes its own author's discovery of the poetic form, which he produces spontaneously upon witnessing a bird's sorrow at the loss of its mate. This author, the brahmin sage Valmiki, having discovered the verse form, seeks a human subject worthy to be memorialized by means of it and finds one in Rama, the prince of Ayodhya. Rama is presented explicitly as a paragon of all human virtues and is said in fact to be an incarnation of the supreme god Vishnu. Deprived of the rulership of the kingdom and forced out into the wilderness through the machinations of a scheming stepmother (who desires the throne for her own son), Rama selflessly accepts his exile to defend his father's honor. During his sojourn in the forest, Rama's wife Sita is abducted by the ten-headed demon Ravana, but, after fighting a war to destroy the demon and free his wife, Rama refuses to accept her back, again for ostensibly selfless motives: as king, he fears that the scandal of remaining with a wife who has dwelt in the house of another will pose a threat to public morals. Pregnant and forlorn, Sita is taken in by the brahmin Valmiki—the inventor of poetry and author of the Ramayana. She gives birth to twins, whom he trains to recite his composition. The frame story of the Ramayana describes its own initial public performance, in which Rama's own tale is told to himself by his own (as yet unrecognized) twin sons.

The Mahabharata has a similarly elaborate and recursive frame story and a similarly involved author—Vyasa, who is grandfather to the main protagonists and antagonists of the story. But, it presents a far darker image of the realities of rulership and the human quest for power. The central story concerns the struggle between two sets of cousins for control of the kingdom of Hastinapura. It culminates in a massive war which wipes out nearly the entire warrior class and brings despair to both winners and losers. Its heroes, the five Pandava brothers, are, like Rama, idealized figures, but the text goes to great lengths to show the moral compromises they are forced to make in their struggle for power. The Mahabharata also resembles the Ramayana in that it is centrally concerned with the deeds of a human incarnation of the god Vishnu. Yet the narrative role of this divine manifestation is quite different. The Krishna of the Mahabharata is not the hero of the epic and is not presented as a moral exemplar and a model for human, and specifically royal, emulation, as Rama explicitly is. He appears as a friend and adviser to the Pandavas and is presented as something of a trickster, manipulating events from behind the scenes and aiding the Pandavas in their quest for power, often by pressuring them to adopt underhanded means against their enemies.

Into and around the central story of the Pandavas' struggle for power the Mahabharata weaves a great deal of peripheral material: supplementary narratives of both human and divine action, as well as extensive legal and moral instruction on, for example, the duties of kingship. Indeed, the Mahabharata comes to be seen as something of a cultural encyclopedia, and famously says of itself that “what is not found here, is found nowhere” (1.56.34).

A somewhat later text which shares many features with the two principal Sanskrit epics, and which bears a similar literary and cultural destiny, is the Harivamsha, or “Lineage of Hari.” This text is traditionally regarded as an addendum or appendix to the Mahabharata, and is concerned principally with relating the life story of the incarnate god Krishna, in particular the story of his early life in disguise as a cowherd and his killing of his uncle Kamsa, who usurped the throne of his family's ancestral kingdom of Mathura (events which are occasionally alluded to in the Mahabharata, but not related at length, despite the central position of Krishna in the main narrative). The Harivamsha provided the model for later accounts of Krishna's early life (a major focus of later Hindu devotionalism), most notably that of the (ninth-century) Bhagavatapurana.

Modern scholarship on the Sanskrit epics has made much of the supposed GENRE distinction between the two texts: the Mahabharata is classified as itihasa or “history,” in part because of the large amounts of legal and didactic and supplementary narrative material it contains. The Ramayana, because of its account of its own origin through the genesis of the verse form, is classified as kavya, or “poetry.” But this distinction is both overstated and highly misleading. It is abundantly clear from the later references to and discussion of these works by poets, commentators, and literary theorists that they were each regarded as both itihasa and kavya—both accurate accounts of historical events and works of literary art. And it is this dual character that is most strikingly evident in the literary legacy of these texts.

Didactic and Poetic Narrative

The Sanskrit epics come to serve as inspirations and as models for two rather different streams of literature in premodern South Asia. In both of these streams narrative remains a central preoccupation but is seen as serving different purposes in each. Viewed as itihasa (“history”), the epics become the archetype for the large body of texts known as puranas (“ancient texts”). These voluminous compendia of traditional lore cover a wide range of topics, conventionally grouped under five headings or lakshanas: sarga (the creation of the world), pratisarga (“secondary creation” by lesser gods or demiurges), vamsha (lineages of kings), manvantaras (the ages of the world), and vamshanucarita (the deeds of the royal dynasties). As this list suggests, these texts devote a great deal of attention to both mythical and dynastic history (see MYTHOLOGY). In addition, they carry on much of the religious and didactic function of the epics, relating further narratives of the gods and their incarnations, and prescribing modes of worship. Like the Mahabharata before them, they often contain long instructional passages describing proper legal procedures and even rules for the construction of buildings or the writing of poetry.

But, in addition to providing the model for the Puranas, the two Sanskrit epics also come to be seen as the prototypes for a quite different sort of narrative literature: kavya (poetry or belles-lettres). Kavya is defined not by its form—there are both prose and verse kavyas—but by its function: it is conceived of by the indigenous literary tradition as a type of text concerned primarily not with providing information or instruction (whether religious or worldly), but with directly producing a pleasurable experience for the reader. Kavya is recognized as comprising a wide variety of prose and verse literary forms, as well as drama (usually consisting of a mixture of prose and verse). Diverse as it is, the kavya tradition looks back to a single work, the “first poem,” the epic Ramayana, as its origin and archetype.

The various genres of kavya differ in the extent to which they depend on narrative for their aesthetic effectiveness. Some, e.g., lyric poetry, lack any continuous narrative thread. Others, such as drama and the later court epic (mahakavya—literally Great Poem) are built around at least a minimal narrative frame but rely to a considerable extent on extended description and elaborate figuration (including puns and other forms of complicated wordplay) for their literary effectiveness. Yet there was also a substantial genre of narrative-driven texts (some prose, some verse) which came to exercise a major impact on South Asian and ultimately world literature.

The seminal work in this tradition of story literature was the Brihatkatha (Great Story) of Gunadhya. This vast collection of stories was probably composed or compiled in roughly the second century CE. The original version of this work is lost, with only a few quotations preserved in the works of later literary critics. What we know of the work is based chiefly on several later TRANSLATION and ADAPTATIONs. There are three such adaptations in Sanskrit—the Brihatkathashlokasamgraha (eighth century? Summary of the Brihatkatha in Verse) of Budhasvamin, the Kathasaritsagara (eleventh century, Ocean of Rivers of Story) of Somedeva, and the Brihatkathamanjari (eleventh century, Garland of Great Stories) of Kshemendra—as well as a partially extant Tamil version and a condensed version (fifth century?) in the Prakrit work Vasudevahindi of SanghadImagesa.

The Brihatkatha was composed not in Sanskrit, the dominant literary language of the time, but in Paishachi (Demonic), a dialect of the Middle Indic Prakrit language mentioned in several early grammars but lacking any extant literature apart from the few surviving fragments of the Brihatkatha. The frame story found in several of the later adaptations explains that the author's choice of language was determined by his loss of a bet with the grammarian Sarvavarman over who could most effectively teach the king to speak proper Sanskrit. Gunadhya, as the loser of the bet, was prohibited the use of Sanskrit, and therefore composed his story collection in Paishachi.

Like the Ramayana and especially the Mahabharata before it, the Brihatkatha is structured as a frame story with multiple embedded narratives. The central story concerns the sexual and political adventures of the Udayana, the king of Kaushambi, and of his son Naravahanadatta, whose romantic pursuit and marriage of a series of semi-divine princesses known as vidyadharis (bearers of knowledge) form the primary narrative thread. But the text, at least as far as one can judge from later versions, appended a great many digressive and peripheral narratives to this central thread, becoming something like an encyclopedia of stories and providing the most important model for later story collections.

Of the many South Asian story collections composed on the model of the Brihatkatha, the one that exercised the greatest overall influence on world literature was undoubtedly the Panchatantra. Like many works from this period in South Asia, the Panchatantra has a complex textual history and exists in several widely divergent rescensions (see EDITING). It was probably first compiled in the fourth or fifth century CE. The author's name is given as Vishnusharman. The work is a collection of didactic tales, grouped into five tantras (systems), and is designed primarily to impart lessons in political policy and morality. Most of the tales it contains are in the form of fables with animal characters. Many of the stories are drawn or adapted from earlier works, such as the Mahabharata and the Buddhist scriptural canon, which contains a collection of stories of the Buddha's jatakas (prior incarnations), many of which depict the lives of the Buddha in animal form.

The Panchatantra proved to be extremely popular and spread rapidly beyond the confines of South Asia. It was translated into Pahlavi (medieval Persian) in the mid-sixth century and into Arabic in 750 CE. Many stories from the Panchatantra found their way into later story collections in Arabic and Persian, as well as in European collections, for instance, in medieval versions of Aesop's Fables. Perhaps more importantly, it seems to have popularized the form of the frame-linked story collection in the Islamic world and beyond, providing the model for later works such as, most famously, the Thousand and One Nights (ninth century) and, indirectly, European works such as The Decameron (ca. 1350), The Canterbury Tales (ca. 1400), and The Manuscript Found in Saragossa (1810).

While no other South Asian story collection exercised so great an impact on world literature, there were several others that spread in similar ways, most notably the Vetalapanchavimshati (Twenty-five Tales of the Vampire) and Shukasaptati (Seventy Tales of the Parrot), which was translated/adapted into Persian in the fourteenth century as the Tutinama.

Poetic Narrative in Theory and Practice

However, despite the existence of this extensive and influential body of story literature in premodern South Asia, the literary status of this narrative-driven prose genre is somewhat ambiguous, at least in the view of Sanskrit literary critics and aesthetic theorists. These theorists draw a sharp distinction between instructive or useful literature and art literature (kavya), and place texts in which narrative content is the primary concern in the former category. This attitude is summed up in an often quoted remark from the (lost) Hridayadarpana (Mirror of the Heart) of the tenth-century Kashmiri theorist Bhattanayaka: “Among the types of literature, people distinguish ’scripture' as that which depends on the preeminence of word; that in which meaning is the essential element is ’narrative' [akhyana]; but, where both word and meaning are secondary, and the expressive process is the main thing, people consider this to be kavya.” In scripture it is the precise wording that matters most—the text must be retained and recited exactly as it always has been to be ritually efficacious. In narrative literature (and here Bhattanayaka is thinking primarily of religious/didactic narrative such as that found in the Puranas), it is the meaning, the informational content, and not the precise wording, that matters above all. But poetry, for Bhattanayaka, differs from both of these, in that both the precise sounds and the meaning are secondary; it is the specifically literary and aesthetic mode of expression that matters most. Thus, for him, poetry, that is, aesthetically oriented literature, is defined in part precisely by its deemphasis of narrative. This general attitude is linked to the tendency, already noted, to see kavya as primarily intended to produce pleasure in its audience, rather than to serve any informational or didactic purpose.

Thus at the theoretical level there is taken to be a clear divide between narrative and aesthetically oriented literature or kavya; in the former, informational content—both story elements and any explicit or implicit didactic message they convey—should predominate, whereas in the latter this content should be deemphasized and priority given to the development of a pleasing or aesthetically compelling mode of expression. In reality what we see is less the sharp division presupposed by Bhattanayaka's typology than a continuum running from more explicitly didactic works such as the Panchatantra to more aesthetically minded treatments of narrative. This range is reflected perhaps most clearly in the variety of Sanskrit prose works produced from the mid- to late first millennium. Some of these works, like the story collections alluded to earlier, are very much content driven, with the primary emphasis on plot and character. In others, however, the narrative is a fairly minimal frame, serving primarily as a vehicle for the deployment of elaborate description, figuration, and other literary devices (see RHETORIC).

The Indian poetic theorists generally recognize two genres of prose kavya: katha (story—the genre to which the Brihatkatha and the Panchatantra belong) and akhyayika (biography). According to the earliest surviving work on poetic theory, the (seventh-century) Kavyalamkara (Ornament of Poetry) of Bhamaha, a “biography” ought to be narrated by the hero himself, and a “story” by someone other than the hero. But, in his own Kavyadarsha (Mirror of Poetry) Bhamaha's near contemporary Dandin (an important prose writer in his own right) rejects this distinction. In general, works labeled as “biographies” are historical, often dealing with contemporary subjects, whereas “stories” typically have invented plots (see LIFE WRITING).

Many agree that the most outstanding exemplars of these two genres are the two great works of Bana, the celebrated master of Sanskrit prose kavya. Bana was the court poet of Harsha (r. 606—47 CE), king of Kanauj and ruler of most of North India. Harsha was a great patron of the arts as well as a major poet and playwright in his own right. Bana was the author of a massive prose biography of his patron, the Harshacharita (Deeds of Harsha), generally regarded as the archetype of the “biography.” (It is not, following Bhamaha's prescription, narrated by Harsha himself, but Bana does devote a large portion of the work to an autobiographical account of his own early life and his first encounter with Harsha.) The central story concerns Harsha's accession to the throne after the death of his father and brother. The action is quite limited, and large stretches of the text are given over to extensive descriptions, many of them involving elaborate puns and other figures of speech, as well as long lamentations of Harsha and his family members over their losses. It is a work as much or more concerned with the play of language as with the development of its story, and seems to illustrate very well the emphasis on the “process of expression” praised by Bhattanayaka.

“Biographies” are on the whole far rarer than stories, and most would seem to be deliberately modeled on the Harshacharita. Like it, they are often the work of poets praising their own patrons. Notable examples include Vidyachakravartin's Gadyakarnamrita, written in praise of the thirteenth-century Hoysala king Viranarasimha II, and the (twelfth-century) Vikramankabhyudaya, written by the Western Chalukya king Somesvara III in praise of his father, Vikramaditya VI.

Bana's other great prose masterpiece is his Kadambari, one of the most celebrated examples of the “story” genre. Far more narrative-driven than the Harshacharita, the Kadambari is a sprawling romance with an intricate plot involving multiple sets of separated lovers, past births, talking parrots, apparent deaths, and miraculous resurrections. In its literary artistry, it is on a par with the Harshacharita, but here given the intricacy and careful development of the story, narrative is necessarily a more prominent concern here than in the former work. While certainly the orientation of the work is more aesthetic than didactic, it still shows more of a balance between narrative and figurative/linguistic concerns than the theorists' typology might lead one to expect. The same would seem to be true of many of the most celebrated and critically acclaimed examples of the “story,” such as the Vasavadatta of Subandhu (one of the most important precedents and models for Bana's Kadambari) and the Dasakumaracharita and Avantisundarikatha of Dandin (a near contemporary of Bana's, who seems to have modeled his own prose works on the Kadambari). All of these works combine an intense devotion to figuration and the play of language, with careful development of plot and character, and seem to belie any simple opposition between the art literature and narrative as such.

This blurring between the functional categories envisioned by the theorists can likewise be seen in various traditions of religious narrative. Both the Buddhist and Jaina traditions, for example, produced substantial bodies of straightforwardly didactic hagiographic narrative, often deriving ultimately from scriptural sources, but, in both cases, efforts were made by later figures within these traditions to produce more aesthetically ambitious, “poetic” versions of such narratives. Prominent examples include works such as Aryashura's Jatakamala (fourth century, Garland of the Buddha's Past Lives) and Hemachandra's Trishashtishalakapurushacarita (eleventh century, Deeds of the Sixty-Three Great Persons). Here too we see works that appear to be pursuing aesthetic and didactic aims simultaneously.

Over time, at least in the field of literary prose, there seems to have been a gradual drift toward a more intense focus on plot and character development and away from the elaborate and loving play of language that had been so central to Bana's work. Toward the end of the first millennium, there was a major surge of interest in extended prose narrative, centered mainly in Western and central India, much of it produced by Jaina authors. Prominent examples include the ninth-century Prakrit Lilavai of Kouhala (ninth century), the Tilakamanjari of Dhanapala (tenth century, Madhyapradesh), and the Udayasundarikatha of Soddhala (eleventh century, Maharashtra). These works are not anthologies or groups of tales embedded in a frame story, like the story collections described above. They develop a single, sustained plot at great length. And, unlike earlier katha literature such as Bana's, the telling of a compelling story is plainly their primary objective. Thus they would appear to shift the balance of literary interest away from the “expressive process” and toward the presentation of compelling narrative content.

In addition to this resurgence of art-prose narrative, this period also witnessed the rise of a new genre: the champu, or mixed prose/verse composition. The champu was known to writers as early as Dandin, but no examples of the genre survive prior to the tenth century, and it appears that it was only after this time that the genre rose to prominence. Some of these, such as Trivikrama's Nalachampu (915) and King Bhoja's Ramayanachampu, were treatments of epic themes or other preexisting narratives, but others, such as Somadeva's Yashastilakcampu (951 CE, Karnataka), set forth original stories, often very intricately plotted. The mixed prose and verse format provided authors with a useful device for balancing the imperatives of narrative development and literary artistry, allowing them to embed the usually more ornate and linguistically playful verses in the more straightforward prose passages.

So one can see that, while the tension between the imperatives of crafting ornate literary language and of creating compelling plot-driven narratives persisted throughout most of the ancient and into the early medieval period in South Asia, there was a gradual shift, at least in prose kavya, toward more content-centered and narrative-driven modes of expression, attaining something close to a proto-novelistic form in the great katha and champu works of the late first and early second millennium.


1. Aryashura (2009), Garland of the Buddha's Past Lives (Jatakamala), trans. J. Meiland.

2. Bana (1897), Harsha-charita, trans. E.B. Cowell.

3. Bana (2009), Princess Kadambari, trans. D. Smith.

4. Budhasvamin (2005), Emperor of the Sorcerers (Brihatkathashlokasamgraha), trans. J. Mallinson.

5. Dandin (2005), What Ten Young Men Did (Dashakumaracharita), trans. I. Onians.

6. Goldman, R.P. et al., trans. (1984—), The Ramayana of Valmiki.

7. Hemachandra (1998), Lives of the Jain Elders (Trishashtishalakapurushacharita), trans. R.C.C. Fynes.

8. Krishnamachariar, M. (1937), History of Classical Sanskrit Literature (repr. 1970, 1989).

9. Lienhard, S. (1984), History of Classical Poetry, vol. 3, fasc. 1 of History of Indian Literature, ed. J. Gonda.

10. Somadeva (2007—), Ocean of Rivers of Story (Kathasaritsagara), trans. J. Mallinson.

11. van Buitenen, J.A.B. et al., trans. (1973—), Mahabharata.

12. Vishnusharman (2006), Five Discourses on Worldly Wisdom (Panchatantra), trans. P. Olivelle.

13. Warder, A.K. (1989—), Indian Kavya Literature, 2nd ed., 7 vols.