The epic flows like a river.
Where has water been born?
Gopala Naika

Lauri Honko, Textualising the Siri Epic. Folklore Fellows’ Communications No. 264.
Helsinki: Suomalainen Tiedeakatemia (Academia Scientiarum Fennica), 1998. 695 pp.
Hard (ISBN 951-41-0812-4), FIM 300,-
Soft (ISBN 951-41-0813-2), FIM 275,-

Lauri Honko in collaboration with Chinnappa Gowda, Anneli Honko and Viveka Rai,
The Siri Epic as performed by Gopala Naika. Part I. Folklore Fellows’ Communications No. 265.
Helsinki: Suomalainen Tiedeakatemia (Academia Scientiarum Fennica), 1998. lxx + 492 pp.
Hard (ISBN 951-41-0814-0), FIM 250,-
Soft (ISBN 951-41-0815-9), FIM 225,-

Lauri Honko in collaboration with Chinnappa Gowda, Anneli Honko and Viveka Rai,
The Siri Epic as performed by Gopala Naika. Part II. Folklore Fellows’ Communications No. 266.
Helsinki: Suomalainen Tiedeakatemia (Academia Scientiarum Fennica), 1998. x + 400 pp.
Hard (ISBN 951-41-0814-0), FIM 185,-
Soft (ISBN 951-41-0815-9), FIM 160,-
Available at the Tiedekirja bookstore
Kirkkokatu 14, FIN­00170 Helsinki, Finland
fax: +358 9 635017; e-mail:

Performing the Siri Epic took six days; during that period Gopala Naika, a possession priest and agriculturalist from Machar in southern Karnataka, India, composed a narrative of 15,683 lines, only five lines shorter than the most magisterial of Western epics, Homer’s Iliad. That performance was acoustically recorded and videotaped by a team of Finnish and Tulu researchers and is presented here in its entirety. The appearance of Naika’s poem in a bilingual, facing-page format ideal for consumption by specialist and nonspecialist alike is a triumph in itself, opening a window onto a world of living oral epic unglimpsed by many contemporary scholars and, more specifically, providing an opportunity to become acquainted with a moving and dynamic mythic universe and performance tradition that deserve attention in their own right. But the richness of the experience available in these three volumes is deeper yet. Also included are complete histories of the eight-year fieldwork project that led to this performance and its eventual codification, a cultural and religious context for its reception, and an evaluative chronicle of eleven other fieldwork projects and resultant publications on oral epic from various parts of the world. In discussing a few salient aspects of the three-volume series, I will naturally have a few queries or criticisms to bring forward, based on my own experience with South Slavic oral epic (both from the Milman Parry Collection and in my own fieldwork), as well as with oral-derived epic in ancient Greek and medieval English. But there is no question in my mind that the composite presentation of the Siri Epic in FFC 264-66 constitutes the single most thorough and most important resource of its kind. As an entrée into the complex, resonant experience of an oral epic tradition, it is unmatched.

One caveat before turning to my remarks. It is a commonplace in this genre of critical commentary that one cannot begin to do justice to the work under consideration in so limited a space. This is especially true for the Siri Epic series, of course. Rather than range widely and shallowly, then, I will be concentrating on a few prominent features of the first volume (Textualising) and then direct ing the remainder of this review where I believe it most clearly belongs – to the epic itself as viewed within a comparative context. I leave additional remarks and emphases to other reviewers and, at longer range, to the research and scholarship that will doubtless arise in spirited con versation with the entire project.

Textualising the Siri Epic (FFC 264) comprises four parts: “The Enigma of Long Epic” (A), which amounts to a comparative poetics of oral epic; “Textualisation of Oral Epics: Antecedents” (B), a history of fieldwork and publication; “Textualisation of Oral Epics: The Present Case” (C), focused on Gopala Naika’s performance in its traditional cultural context; and “The Siri Epic: A Synopsis” (D), a helpful overview of the main action of the tale. Early in section A, Honko rapidly establishes some of the most crucial dimensions of the frame within which the 1990 performance took place. We learn of the singer’s illiteracy and, more significantly, the positive advantage that his devotion to oral tradition entails. We start to get a sense of the bard’s religious status as the leader of the local Siri possession-cult, a connection that enlivens every last fibre of the song from individual lines and phrases to the overall performance arena.1 Not least, we read that it was Gopala Naika himself who initiated the textualisation of his epic, first by declaring “his willingness to dictate the epic” (264: 13; see further C.6) and then by following through with actual dictation in 1985-86, a prior textualisation of 8,538 lines that Honko labels “Homeric” because of the medium in which it was composed and taken down. In this and many other ways, the first of the three volumes enlarges and productively complicates prevailing views of long epic. In the process it weans us away from the parochial examples of Western narrative and provides firsthand evidence that oral epics can be and are created wholly without the technology of writing – in a word, that oral composition is a much more complex and many-sided process than has heretofore been appreciated.

Defining the epic

One step on the path toward a more realistically complex view of oral epic – an awareness of their differences as well as similarities – is the formulation of a definition that opens the door to as many narratives as possible while still maintaining the outlines of a generic model that will support comparison. Just such a comparative poetics is at the basis of Foley, Traditional Oral Epic: The Odyssey, Beowulf, and the Serbo-Croatian Return Song (Berkeley: UCP, 1990), which advocates attention to incongruities as well as congruities in prosody, phraseology, and narrative patterning. Here is Honko’s definition (264: 28):

Epics are great narratives about exemplars, originally performed by specialised singers as superstories which excel in length, power of expression and significance of content over other narratives and function as a source of identity representations in the traditional community or group receiving the epic.

As a test of this perspective, let us apply it to two well-studied traditions, one of them ancient and the other contemporary. Homer’s Iliad and Odyssey certainly fulfill the first criterion, with Achilleus and Odysseus (as well as many other characters) serving as exemplars. Their contem porary expressive power and significance are manifest from the literature and commentaries of the ancient world, where, according to Xenophanes of Colophon, “from the beginning all have learned according to Homer.” Plato’s philosophical writings, even as they often contest against Homer for dominance in worldview, constantly cite the epics as a cultural encyclopedia. The grip exerted by the Iliad and Odyssey on the evolution of European culture – and, very importantly, on the concept of epic itself – has been obvious for centuries. Although our knowledge of “specialised singers” in ancient Greece is fragmentary and sometimes contradictory, and although the rest of the epic tradition has vanished except for a few lines and summaries here and there, Honko’s definition generally fits what we know about Homer and his tradition. When we add the general consensus that the ancient Greek poems emerged from an oral tradition, though again in what particular manner we cannot say, the Iliad and Odyssey can be meaningfully approached via Honko’s model.

How about the living tradition of South Slavic Moslem epic?2 Once again there is little doubt about the role of exemplars; although the cast of characters from the glory days of the Ottoman Empire will be less familiar to Western ears than the celebrated figures of Greek mythology, a wide array of heroes and heroic women are prominent in the songs. Moreover, in this well collected tradition, the identity of the songs as “superstories” is more evident and compelling than it can ever be in the long-dead, partially extant traditions of the ancient and medieval worlds. The performances are complete in themselves, of course, but because of the resonant context in which they take place, they always imply more than they denote. In a sense they have no real boundaries, no absolute textual singularity, just as they are not repeated syllable for syllable in every version. Both their position in the overall network of oral epic tradition and their multiformity, a quality to which Honko returns again and again in these volumes, are fundamental aspects of their identity as “superstories.” Performed by specialized bards, called guslari, the Moslem songs demonstrate considerable if not overwhelming length (an average of perhaps 2500 lines, though master-singer Avdo Medjedovic’s performances reached 13,000-18,000 lines on two occasions), a power of expression that has been shown to be highly idiomatic and foundational for its interlocutors, and a significance of content that can be traced to the role of professional singers in Ottoman courts and until recently mirrored in the continuing activity of singers and audiences in Bosnia and Montenegro. Even the modern (and decidedly nonprofessional) singers of tales in these regions, whose first allegiance was to practical occupations such as farming, craftsmanship, and small-scale trade, preserved the heritage of South Slavic Moslem epic, keeping its encoded history and group-centered identity alive for themselves and their audiences. Once again, with allowance made for the idiosyncrasies of the individual tradition, Honko’s model applies reasonably well to this living epic tradition.

Epic register and epic idiolect

With such parameters in mind, Honko goes on to explore numerous aspects of performance and com position that deserve much more extended commentary than can be offered in the present format. I will concentrate here on three of them that are of primary importance for comparative studies: “epic idiolect and epic register” (A.7), modes of performance and dictation versus singing (A.9-10), and the concept of the “mental text” (A.12).

With the terms register and idiolect, Honko establishes perspectives on composition and reception that promise greater fidelity to oral epic because they promote the understanding of its performed instances on their own terms. (On the application of these terms to various traditions and genres, see e.g. Foley, Singer of Tales in Performance, pp. 49-53, 82-92.) Honko is careful to explain that he uses the term “register” in Dell Hymes’s sense of “major speech styles associated with recurrent types of situations,” thus identifying a specialized epic language that the singer learns to speak and in which the audience also (and crucially) gains a fluency. By transacting the verbal exchange within this medium, then, “what the performance brings about is essentially a community of reception” (264: 64). Within the shared dialect of the epic register, Honko also locates an individualized level of traditional language, the “idiolect.” Just as speakers of everyday language in any society share a dialect but develop their own personal versions of that more generalized language, so the individual epic singer carves out his own working language from the more generalized epic dialect. By directing our attention toward idiolect, and construing it as a flexible, multiform linguistic instrument, Honko is able to have the best of both worlds: he can speak of the traditional structure and meaning that are so much a part of the Siri Epic context, and he can describe Gopala Naika’s individual (and by definition inimitable) creation within that tradition.

This stereoscopic view – paying due attention to both individual and tradition – was precisely the original intent of applying such concepts to oral traditional performance. For further evidence of the necessarily paired contributions of individual and tradition, see Foley, “Individual Poet and Epic Tradition: Homer as Legendary Singer,” Arethusa, 31 (1998): 149-78 and, more generally, Foley, Homer’s Traditional Art (University Park: Penn. State Univ. Press, 1999).

Modes of performance

Especially informative for scholars whose main research area centers on Western epic is the discussion of modes of performance (A.9). Whereas many conceive of oral epic (on the model of Homer) as an extended, single-channel narration by the singer, Honko points out that worldwide epic must be more broadly conceived, taking into account such variations as the presence or absence of musical accompaniment, prose and poetry, dance, group rendition, drama, and other performative modes. Moreover, the mode may shift within a single performance, and the “same” epic may be realized in numerous different ways. As an example, he details four distinct modes of Siri Epic performance: “monovoiced singing with slow body sway and linear narration of the epic” (264: 76); a dialogue between the main singer and his male assistants; “the polyphonic overlapping solo singing by the male and female singers without any observable synchronising of expression” (264: 77); and alternating performance by two or three female cult members with back-channel reinforcement by the rest of the group. All of these may be responsibly and accurately understood as the Siri Epic, but each of them engages a different context and produces a different result. Such is the natural heterogeneity of oral epic in this Dravidian tradition, a quality that should encourage us to press for plurality in our idea of oral epic, no matter what our particular specialties may be, and perhaps to be aware of signals within oral-derived texts that may point in similar directions.

One is reminded, for example, of the poet’s invocations before the epic action begins (Homer’s Iliad and Odyssey; also the Anglo-Saxon Beowulf), of the interruptions of his story in the Iliad (before the Catalogue of Ships and Men in Book 2; thanks to Aaron Tate for this example) and the medieval English Andreas (lines 1478-91), or of the vocative address of the swineherd Eumaios in the Odyssey (e.g., 14.55), where the poet seems to speaking to one of his characters (on which subject, cf. Plato’s dialogue, Ion). Modes and performance styles of various kinds may persist into written texts as signals to be activated during the process of reception (see Foley, The Singer of Tales in Performance, chap. 3).

Just as an epic may be performed differently, so too it can be recorded in quite diverse media environments: dictation and singing. Here Honko’s analysis of the two possibilities via his research team’s experience with the Siri Epic plays out with special significance (A.10), and not only because of the light it sheds on the performance at hand. Since Albert Lord initially broached the subject in 1953, claiming a superiority in length and quality for the dictated text, Homerists and other scholars have jousted over the nature of ancient and medieval texts – artifacts that could not have been fashioned except by dictation for technological reasons – and the comparability of acoustically recorded and dictated performances from various parts of the world. Contrary to Lord’s findings, Honko demonstrates the artistic and structural superiority of the sung Siri Epic, and traces it to the greater elaboration of narrative multiforms in the sung as opposed to the dictated versions. As a parallel Honko notes that “Brenda Beck found the dictated version of the Tamil epic… contentwise and stylistically impoverished and much shorter than the sung version” (264: 82). To some extent, the question of sung versus dictated will rest on the individual properties of the given oral epic tradition (the variety of modes and performance styles, for example) and the particular circumstances that characterize the time and place of its recording (wax tablets, vellum, or paper; one or a team of scribes or various methods of acoustic and video recording). But Honko’s hard evidence makes it clear that we should not automatically privilege the medium nearest our everyday experience. In the Tulu case, the esteemed technology of writing is a positive impediment to the best epic performance.

Mental text

In section A.12 we come to one of Honko’s most suggestive and powerful concepts, the so-called “mental text.” It is also destined to be one of the most controversial features of his impressive composite theory – not, I must stress, because of its considerable explicative potential but rather because of the basic terminology employed. Working from the ideas of register, idiolect, and “pool of tradition” (espec. A8), he starts by shifting the perspective from the collective tradition, as instanced in song- performances by a variety of bards, to the individual performer. This redistribution of emphasis is apparent in the following summary of what the mental text contains:

… (1) storylines, (2) textual elements, i.e., episodic patterns, images of epic situations, multiforms, etc., and (3) their generic rules of reproduction as well as (4) contextual frames such as remembrances of earlier performances, yet not as a haphazard collection of traditional knowledge but, in the case of distinct epics of the active repertoire, a prearranged set of elements internalised by the individual singer. (264: 94.)

Few would argue with the gist of this description. “Storylines” correspond to the story-patterns and thematic sequences of oral-formulaic theory; what he calls “textual elements” are mirrored in various approaches to folklore and oral tradition; and “rules for reproduction” have been suggested as a way to rationalize repeated items as the products of a rule-governed process. (Cf. the concept of traditional rules as described in Foley, Traditional Oral Epic, chs. 3, 5, 8.) Once the transfer from collective tradition to individual singer is made, the organizing role of “contextual frames” becomes an inevitability. With these sound analytic principles it is hard to argue.

But one can argue with the choice of the critical term “text.” With all of its virtually inescapable overtones of fixity, literacy, and the technology of writing, text seems a risky label to employ in de scribing so central, and intangible, a concept. Given how much has to be overcome in pressing “text” into service for what may seem a counterintuitive purpose, would it not have been easier to use a less loaded word? On the one hand, I understand and admire Honko’s strategy: he seeks to explain the Siri Epic from this performance outward, through the composition of a single bard who harnesses a traditional register to his own idiolectal purposes. He seeks to show how the pool of tradition has taken shape within the mind of Gopala Naika, and his theoretical approach succeeds brilliantly in opening the epic to appreciation on its own terms. My only criticism is the choice of the term “text,” which presents a problem in reception for folklorists and general readers alike.

Section B of Textualising the Siri Epic begins with a discussion of Elias Lönnrot and the Finnish Kalevala, and continues with outlines of ten additional projects: the Manas epic and Wilhelm Radloff, the Mohave epic and Alfred Kroeber, the South Slavic epic and Milman Parry and Albert Lord, the Sunjata epic and Gordon Innes, the Anggun Nan Tungga epic and Nigel Phillips, the Annanmaar epic and Brenda Beck, the Palnaadu epic and Gene Roghair, the Son-Jara epic and Charles Bird and John Johnson, the Siirat Banii Hilaal epic and Susan Slyomovics, and the Paabuujii epic and John Smith. Each episode in the larger tale of oral epic projects includes comments on the idiosyncrasies of the given performance tradition and the circumstances of its collection, and offers evaluative consideration of the conclusions drawn by the investigator. This section needs little comment; by presenting a meticulous and fair-minded account of these other fieldwork projects on oral epic, Honko has compiled a unique resource that will undoubtedly prove extremely useful in itself even as it acts as a ready companion to the Siri Epic documented in these volumes.

For the record, I note that, contrary to the claim made here (264: 187), the Parry Collection’s major strength has always been the epic genre in South Slavic. Women’s songs (zenske pjesme) are quite numerous, it is true, but they are normally extremely short (in the range of 10-30 lines) and entirely lyric (non-narrative) in nature. The hundreds of epics collected in 1933- 35 and 1950-51 from six geographically distinct epic centers in the former Yugoslavia, on the other hand, are more than sufficient to provide a rich, multiform context for individual singers, individual regions, and so on; see the full accounting in Matthew W. Kay, The Index of the Milman Parry Collection 1933-35: Heroic Songs, Conversations, and Stories (New York: Garland, 1995). In addition, the extended conversations with guslari, conducted by Parry’s native assistant Nikola Vujnovic (himself an epic singer), help the investigator toward what Honko calls a “thick corpus” (264: 39, 512-13).

Epic archaeology

Section C combines a rich cultural contextualization of the epic performance with an insider’s history of the fieldwork. Instead of attempting to survey the twenty separate parts of this section, I will focus on one activity that links context and fieldwork: what Honko and his team call “epic archaeology” (C.10). On the practical level, the two-day pause in Gopala Naika’s performance, necessary to help the bard heal a throat and vocal apparatus overextended by singing, presented an opportunity to explore the dimensions of myth and reality in the Siri Epic. The team had already become well aware that the epic story was not simply entertainment or a reflection of ritual, but living truth for its practitioners. As Honko puts it (264: 322),

It was not a matter of ideological stance or mythical narrative only but the continuous construction of a tangible world in which you as a Siri-devotee or Kumara-devotee could, with the help of your prayer and epic song, touch and move the heroes-turned-into-gods and summon them to your festival, where in turn these could occupy your body and use it as their vehicle in the human sphere making you, a human being, a true Siri or Kumara for a transient moment.

This amounts to participation or epic co-creation at a much deeper level than Western epics seem to license, touching the sphere of religious identity in a personally important way. One aspect of epic archaeology was then the realization that the Siri Epic functions for those who perform and listen to it as a recurring rite of passage between two worlds, a retraceable path leading toward direct communication with divinity, a means by which gods enter the earthly realm and possess willing participants.

Moreover, this entry and possession constitutes not symbolic but literal truth. The events of the epic were considered historical, and the places where they occurred geographically real. The researchers were thus able to visit locations said to be those mentioned in the epic, bringing along with them the singer Gopala Naika, who had never before visited the actual sites himself. One example was the “fairly large, quiet pond called kandadi kaaya and known as the place where Siri took her bath and washed clothes” (264: 325). This modest body of water had taken on a legendary character as a taboo site, so that neither fishing nor bathing was permitted; its depth could not be measured and people said it never dried up. Local people even pointed out a particular tree near the pond’s edge as marking the spot where Siri gave birth to Sonne. In certain ways the more-than-mythic geography associated with the Siri narrative resembles the system of Western Apache place-names, which, far from simply indexing this or that topographical feature, encode traditional stories as embedded and emergent implications of the names. (See Keith Basso, Western Apache Language and Culture: Essays in Linguistic Anthropology (Tucson: UAP, 1990), pp. 138-73.) In both cases, mythic history is mapped onto real territory, which in turn takes on sacred dimensions.

Epic archaeology also culminated in increased awareness of another kind of context: the non-epic genres that subtend various aspects of the Siri story. Honko mentions a host of traditional forms that populate the epic universe: in addition to place-names, he identifies “belief legends, aetiological narratives, historical legends, prayers and incantations, proverbs and phrases, omens and taboos, rituals and customs” (264: 322) and the like, all of which cluster around the epic but were not part of the core narration as performed by Gopala Naika in 1990 and presented in these volumes. Such ancillary sources, some directly linked to geography and some not, are clearly important threads in the overall fabric of the Siri story, and their mention highlights the dangers inherent in isolating the epic performance from its natural context. Likewise, the whole networked constellation of genres begs the question of how to deal with oral-derived traditional texts from the ancient and medieval worlds, where there is no opportunity for ethnographic research to establish a “thick corpus.” At the very least, this aspect of epic archaeology should underline the recognized fact that traditional genres leak; that is, there is more interplay among and interpenetration between different genres than our analytical practices customarily assume. If a traditional phrase belonging to an Old English riddle turns up in Beowulf, if a story-pattern associated with the Odyssey also underlies the Homeric Hymn to Demeter, then perhaps this sharing of traditional forms, strategies, and content is the natural course of events. Only the peculiar and parochial text-centrism of Western scholarship, epitomized in the one-dimensional Homeric model for comparative epic investigation, has kept us from making these connections and hearing their resonance.

Expanding the idea of epic

Against the monumental background of Textualising the Siri Epic emerges its raison d’être – the epic itself (nos. 265-66). Several things strike the reader immediately. First, this is an enormous production on almost any scale: 15,683 lines performed over six days by a single bard, Gopala Naika. It puts the lie to untutored speculations that the achievement of epic length must involve the technology of writing. Second, it is a woman’s story, featuring a female divinity at its center and highlighting the activities and behavior of women and female deities throughout its course. Not coincidentally, it is a song of peace, social custom and rite, and the chain of creation. For those of us accustomed to the heroic contests of war – with male heroes competing for cities, women, and most of all for glory – as the major and defining activities of epic, this steady focus on the female world is an education in itself. From the very start, the subjects and dramatis personae of the Siri Epic force an enlargement or pluralization of our narrow view of epic; when we add the religious reality of the epic as a rite, we are truly in a different world from that of the Homeric epics, the Anglo-Saxon Beowulf, the Old French chansons de geste, the medieval Spanish Poema de Mio Cid, or even the South Slavic Moslem epics. The idea of oral epic begins to assume a realistic and engaging complexity. Third, this is a remarkable artistic achievement, and by that I mean a paragon of traditional art. Even for a non-Tulu speaker lacking a great deal of the cultural apparatus, the poetics of implication – the poem’s traditional referentiality – makes reading this epic a richly rewarding experience.

Some attention to performance units may help to place the Siri Epic in a comparative context (See Textualising, C.9, for a detailed study of the segmentation of the oral performance.) On the emic side, the singer himself separated his performance into 36 segments, each of these defined by the researchers as “a sequence of lines sung mainly without interruption” (265: xliv). On the etic side, the editors and translators decided to represent the text in some 56 cantos, primarily on the basis of textual logic and consistency. Here as elsewhere their intention to provide a readable poem remained foremost in their thinking; although no performance feature goes unreported or unexplained, their loyalty to a text that will stand on its own two feet is the operative policy. Although there is no opportunity to discuss these matters here, I recommend that comparatists scrutinize Honko’s comments on the discontinuation and resumption of singing and the “correction of errors” (265: xliv-lv). (On the former, see also Foley, Traditional Oral Epic, pp. 284-88.)

At the micro level, lines were established on the basis of the singer’s breath groups, very much an emic measure that mirrors what Dell Hymes has proposed in “measuring” verses in Native American narrative. Honko states straightforwardly that the metrical structure of the lines is not completely solved, and that an adequate analysis will require the expertise of an ethnomusicologist. Until such time as the inner prosody of the epic is established, one may observe that the emic lineation and the editors’ bias toward a readable segmentation of the text insure a worth while experience for the Tulu-less reader.

Verbal equivalence in translation

The Finnish-Tulu team offers some extremely interesting remarks about the decisions they made in adopting an overall style of translation. From the facing-page bilingual format through the insertion of punctuation, the aim at verbal equivalence in simple English, the handling of repeating phrases, and the discussion of particular challenges created by the idiosyncrasies of Tulu and/or the lack of parallel words or constructions in English, the investigators evolved a method based on clear choices about priorities.

Furthermore, as Honko notes, the translation of long epic has its own special problems, with consistency of rendering (over the 15,683 lines) high on the list of attainable but difficult goals. I would note that, however one may feel about the particular choices made and the results achieved, the English version of the Siri Epic reveals what might be called an “interactive” quality. That is, because the formulaic phrases are rendered exactly and consistently, the bilingual presentation allows the reader to track phraseology in both languages, whether he is fluent in Tulu or merely able to sound out the language.

Similarly, the decision to leave the metaphors in their literal form – “the head’s fire” rather than “bad headache,” for example (265: lxiii) – allows the poem to echo memorably as the reader proceeds from canto to canto, promoting a more genuine experience of networked meaning, of traditional referentiality. Not unrelatedly, the policies about punctuation bring out the adding or paratactic style of Gopala Naika’s composition.

A final feature of the translation strategy that also deserves mention is the uncluttered presentation of the text: in place of a cumbersome and distracting scholarly apparatus at the bottom of the page or at the back of the translation volume, the researchers depend chiefly on the freestanding commentary in section C of Textualising the Siri Epic (264). A brief glossary of proper names, place-names, and a few items of material culture appears in the last pages of the third volume, but otherwise the epic itself dominates the presentation in volumes 265-66.

In my opinion, the discussion of translation (265: lxi-lxix) should be required reading for all those involved in translating oral epic. Even if they choose other methods than the Finnish team selected, they would profit immensely from a careful examination and weighing of the alternatives.

Creating the narrative map

In order to convey an idea of the overall narrative shape of what I again emphasize is very much a woman’s epic, let me outline some of the major action in the first of five parts of the Siri Epic. The division into four “subepics” and a conclusion is of course the etic imposition of the editors rather than Gopala Naika’s own segmentation, but it makes sound narrative sense and increases read ability for a non-native audience. We will be sketching the broad superstructure of “The Ajjeru Subepic” (lines 1-3811).

This section covers the events leading up to the birth of the principal character and namesake of the epic, Siri herself. The narrative opens with a cosmic and etiological initiative – the creation of a Tulu story, as ordered by the god Iisvara and carried out by the god Naagaberamma, who starts the process by sending a serious illness to an aged widower, Ajjeru; Aarya Bannaaru Birmu Paalava. It is characteristic of this epic to maintain a variety of names, some simplex and some composite, for many of its characters. One soon becomes accustomed to this variety, however, much as one learns to navigate the approved noun-epithet pathways for naming people and deities in western European traditions like the Homeric poems.

In the highly lyrical fashion that frequently plays a role in this performance, the old man’s suf fering is epitomized in a tear that drops from his eye and eventually reaches the temple of the trinity gods. Naagaberamma then takes pity on Ajjeru and decides to visit him in the disguise of a “poor Brahmin man”. Even for the reader wholly unacquainted with the Siri epic tradition, this figure, who appears many times, will begin to take on certain implications, signalling some kind of unexpected change in the status quo. After all, he amounts to a disguised god entering the earthly realm.

Note that the formulaic, indexical phrase “poor Brahmin man” idiomatically identifies this figure together with the generic outline of the role he is to play. (For parallels to this kind of indexical phraseology in other oral epic traditions, see Foley, The Singer of Tales in Performance, chs. 4-6 and Homer’s Traditional Art, chs. 4, 7.)

After the temple at Lookanaadu is restored and ritual worship as stipulated by the “poor Brahmin man” is established, the infant Siri is born – parthogenetically and without direct agency, it appears, since her father Ajjeru simply finds her crying in the siri-shoot areca-bud that provides her name. Even as an infant, she inspires a suit by Kaantu Puuñja, and Ajjeru, seeking a male caretaker for his beloved home at Satyanapura, accedes to the would-be bridegroom’s (and bridegroom’s mother’s) proposal.

Before the wedding can be arranged, however, they must consult the astrologer about the propitiousness of the marriage and the most auspicious day for the actual wedding ceremony. This pattern of visiting the astrologer is another frequently recurring scene in the Siri Epic, a “theme” or “typical scene” in the terminology of the oral-formulaic theory; it follows a regular sequence of constituent actions and portrays them in formulaic language. Things proceed in an expectable order on each occasion: once the principals arrive, the astrologer asks why they have come and is answered straightforwardly. He then consults his almanac and determines the time at which the stars will be in the most promising position, whatever the actual event may be. The interlocutors, in the present case Kaantu Puuñja and his mother, then return to report the happy news to others concerned, in this instance to Ajjeru. Like the entry of the “poor Brahmin man,” the consultation of the astrologer is not only a predictable but, more significantly, an idiomatic sequence of actions: it “slots” the unique moment within a traditional frame of reference. Again, even the reader from outside the epic tradition grows familiar with the narrative map and configures his or her expectation accordingly. Imagine how powerful an effect is achieved, then, when in a single, unparalleled instance the map provided by this typical scene leads not to propitiousness but to foretold disaster. (See further the discussion of Abbaya and Daaraya below.)

All proceeds in a promising way with Siri’s wedding, however, and a pattern of events with which the reader will become familiar emerges: the procession, ceremony, onset of puberty (with attendant ritual), and pregnancy. In contrast to the usual and natural sequence, Siri’s puberty is delayed; she seems barren for a time (perhaps a measure of her “special” status), but eventually she comes of age and is able to bear her first child, Kumara. As in other epic traditions, delay in the development of a traditional pattern causes its own kind of suspense (cf. the interruption of the Feast multiform in the Odyssey, as discussed in Foley, Homer’s Traditional Art, pp. 180-81).

The puberty rite is itself traditional in its structure, with the bride’s passage into adulthood celebrated by a feast at her paternal home, the same place to which she will return for the actual birth. Between these crucial signposts on a woman’s anticipated life-path, the epic register places another significant event: the so-called “desire-feast.” Held at the husband’s home, this gathering has the ostensible purpose of soothing the unborn child’s hunger, but given the elaborate preparations (including a trip to the astrologer), we may also see it as another occasion for cementing the linkage between wife and husband and their families. Or so it customarily goes. In this unusual case, the wayward husband and father-to-be, Kaantu Puuñja, strays into an encounter with Harlot Siddu. Siri later recognizes her unwanted intervention by refusing to wear the sari and ornaments that Kaantu Puuñja had brought her, on the grounds that Harlot Siddu had tried them on first. When Siri dismisses her husband and wears another sari to the desire-feast, the reasonably well prepared reader cannot be surprised: in consorting with Harlot Siddu, Kaantu Puuñja has violated both his marital vow and the predictable pattern of the ritual. Cultural expectation and traditional referentiality are shattered by his actions. Such is the impact of wholesale reversal on the poetics of implication in the Siri Epic.

In due course Siri’s child is born, given the name of Kooti Kumara, and discovered to have divine characteristics. Ajjeru, the grandfather, builds a cradle for the infant and hangs it; with the assistance of maidens from four lineages, he spreads silk and deposits the child within. This sequence of actions forms a small unit in itself, helping to “slot” each instance of the care of and initial ministrations to a newborn throughout the Siri Epic. Meanwhile, a visit to the astrologer has established more than the usual details: in addition to the most suitable name for the new arrival, Ajjeru has been warned that if he looks at the child’s face he will die. Like all other such predictions, which invariably come true (with an idiomatic, traditional certainty), this prophecy is realized. Out of concern for the distraught infant Kumara, and with Siri off washing clothes with her servant maid Daaru, Ajjeru attempts to stop his crying: when the old man happens to catch sight of the child’s face, he immediately faints. Siri senses trouble from afar and hurries back to Satyanapura, but she can do nothing except comfort her father as he passes away. His last words consist of advice to his daughter to remain free, whatever the cost.

A poetics of implication

The remainder of the poem, which the editors divide into “The Siri Subepic” (lines 3812-9028), “The Sonne, Gindye Subepic” (lines 9029-12279), “The Abbaya, Daaraya Subepic” (lines 12280-15063), and “The Kumara Conclusion” (15064-15683), present a rich and diverse array of characters and events. Part of this richness, however, stems from the recurrence of traditionally significant features and actions: births, betrothals, weddings, puberty rites, desire-feasts, and funerals unify the overall narrative in more than a linear fashion, causing each “new” moment to resonate against the audience’s or reader’s prior experience. The genius of the traditional narrative lies not only in its impressive expanse and the epic scope of its concerns, but also – and perhaps most fundamentally – in the networked associations that enlarge its impact beyond any single text, no matter how lengthy or elaborate. For all that Gopala Naika actually manages to say during his performance of the Siri Epic, he implies a great deal more.

The same poetics of implication, the same traditional referentiality, can be glimpsed and heard at the level of individual lines or groups of lines. (Here one is especially grateful to the translators for their policies as outlined above.) In addition to such commonplaces as expelling the evil eye, normally accomplished in the same or similar language, and short catalogues that document building, digging, and the like, we come upon numerous smaller units of recurrent utterance.

One common example is the speech introduction, which can also follow the speech in question. Here are two instances of a frequently used phrase:

To the reply a reply, to the response a response the Brahmin gives:
(5213; speech follows)

Thus to the reply a reply, to the response a response
Siri of Satyamalooka is giving in the assembly court.
(4372-73; speech precedes)

Just as with analogues in ancient Greek, Anglo-Saxon, and South Slavic epic, this brief sentence has two parts: one provides the predicate in a stable, repeatable form and the other, variable part names the particular speaker. In such a case the recurrent phrase performs a basic, rather modest function by marking the onset of speech in a memorable, idiomatic way. It is perhaps among the simplest of signs at the phraseological level, bearing no apparent implication about the nature of the speech. One clear proof of this emphasis on function over resonance is the fact that this same phrase introduces both the most elevated speeches as well as Siri’s complaint to her mother-in-law Sankaru Puuñjedi, at the close of which Siri asks the following about her husband Kaantu Puuñja: “Has he gone to the end of town to watch Harlot Siddu’s bottom?” (4839). (On the semantic leavening involved in such introductory lines in other epic traditions, see the discussion of “traditional punctuation” in Foley, Homer’s Traditional Art, pp. 221-23.)

In the same category of small verbal signs with limited resonance we may place a phrase that signals the arrival of a person from another locale. With great regularity Gopala Naika describes “the beauty of his/her coming,” using this modest phrase as a kind of traditional index. Once again, there seems to be no special overtone associated with what we might understand as a composite “word,” a single unit; it is applied to a full spectrum of characters and events regardless of the nature and purpose of the journey. We read it best, that is, as neither merely literal nor simply structural, but rather as the modest traditional idiom it is. (For a similar kind of phrase in Homeric epic tradition, see Foley, Homer’s Traditional Art, p. 218 on “wine-dark sea.”)

More resonant than these first few examples are phrases or “words” such as “fire of the child’s mouth” (3020, e.g.) and the exclamation “Alas, what a sin! Alas, what a shame!” The former indexes the ravenous hunger felt by a pregnant woman due to the metabolic strain of gestation, as well as looks forward to the desire-feast that is socially prescribed as its ritual remedy. As such, the phrase also implicates the reality of pregnancy and the woman’s life-pattern of birth, betrothal, puberty rite, wedding, and childbirth in a powerfully economical way. Gopala Naika “speaks volumes” through these few syllables, drawn from the wordhoard of the traditional register. The same can be said for the latter couplet – “Alas, what a sin! Alas, what a shame!” – that recurs throughout the Siri Epic to signal cataclysmic changes of fortune. In the Ajjeru subepic, for example, Siri, returning from her clothes-washing to find Kumara’s cradle empty, marks the seriousness of the situation with this phraseological sign (3700). She repeats the same indexical exclamation a few lines later when she cannot locate Ajjeru, her child’s caretaker while she is gone, in his customary place on a sleeping cot (3714). Indeed, this formula recurs dozens of times in many different narrative situations, as when Siri’s co-wife Saamu Alvedi bemoans her “sister’s” lack of a place to give birth to her second child (7753). Whatever the particular circumstance, “Alas, what a sin! Alas, what a shame!” frames the problem as a life-threatening or at least socially disastrous one, embedding the uniqueness of the individual moment in a timeless traditional context.

We could pause over many additional phraseological signs, all with greater or lesser proverbial force in this traditional epic register and performance. Also deserving of study are the many and frequently deployed similes, comparable in their periodic profusion to the famous Homeric trade mark. In speaking of Saamu and Siri, co-wives to Kodsaara Alva, the singer lyrically pictures them as follows (7941-45); here Naaraayina is “a name of Vishnu, also used as a frequent refrain line” (266: 884).

…Saamu and Siri were,
Naraayina, like twin stars arisen in the sky,
like two young serpents born in a serpent-dwelling.
Naraayina oo Naaraayina oo.
Like Raama, Laksmana, like the children born in the belly of one mother.

Or we might explore slightly larger units, multi-line descriptions shorter than a full-blown typical scene, such as the summoning of a postman and delivery of a letter. Like the South Slavic oral epic tradition, the Siri Epic depends quite regularly on such loosely configured sequences, consisting in the Tulu case of engaging the services of the postman and sending him off, followed by the arrival at the prescribed destination only to require three calls to gain the attention of the person to whom he is to deliver the message, and finally his untying of the hem in which he keeps the letter, e.g., 10421-78. Expectable phraseology accompanies this small and familiar schema, assisting in the process of traditional indexing.

The game of fate

But, at least for this review essay, we must be content with a single last example of Gopala Naika’s mastery of the register that supports his epic tradition, one final feature of his epic idiolect. Late in the narrative, two girls named Abbaya and Daaraya, twin daughters of Siri’s second child Sonne, petition their father Guru Marla to be trained in the book-learning that oral epic so often celebrates but never itself employs. (See the discussion of the famous tablet of Bellerophon, mentioned by Homer, Iliad, 6.166- 80, in Foley, Homer’s Traditional Art, pp. 1-5.) Here are three short passages describing their request, its fulfillment, and their rejoicing over the success of the process. First comes their plea (12404-07):

“Naraayini Naaraayina, father,
we must now learn reading, writing, knowledge, wisdom,
we must learn! Such
a desire we have, father, father!”

Guru Marla then provides the necessary materials and sets the tutorial in motion (12411-13; 12416-17; 12421-22; 12429):

Guru Marla had slates, books brought, see.
He had bags, bags prepared.
The children’s bags he filled with books, slates…
Taking the children for reading, writing
to the gallery, to the writing hall,…
In the gallery, in the writing hall
The children now learn knowledge, wisdom…
with reading, writing the children filled their bellies.

Their education completed in an instant of narrative time, Abbaya and Daaraya happily celebrate their achievement (12444-49, italics added):

“we are, father,
in four kingdoms, in four regions,
in reading, writing, in knowledge, wisdom,
in playing cenne, today,
we are very skilful, skilful. Thus they say
in four kingdoms, in four regions, father!”

A traditional audience – and the reader attuned to the inimitable contribution of the epic register and idiolect – will hear the formulaic echoes among these passages, a resonance that indexes each of them in a larger frame of reference. As in other oral epic traditions, for the native audience this resonance presumably stems primarily from a deep fluency in the traditional language and an understanding of Gopala Naika’s personal use of it, and not simply from parallel instances within the linear expanse of a single performance-text. The twin girls ask for “reading, writing, knowledge, wisdom,” in short, and that is precisely what they receive and rejoice over once they have acquired it.

But the art of Gopala Naika does not end with function and structure. In the third passage an added detail (italicized here) is appended to the formulaic description of their learning: a surpassing skill at playing cenne, a type of board game with pebble counters. At first sight this seems an innocent enough discrepancy, especially within a multiform medium whose lifeblood is variation within limits. But consider the fact that soon Abbaya and Daaraya will beg their doting father for a silver cenne-board with golden pebbles, and that his ready compliance – which entails special forging of precious metals by the smith Ciinkrooji – will eventually lead to the twins’ death via the intercession of a “poor Brahmin man” begging alms. In a fit of pique, and encouraged by the Brahmin, Abbaya uses the board to strike her sister on the head, fatally as it turns out; later she commits suicide by throwing herself into the same well where she disposed of Daaraya’s corpse. These events are, however, not as final or disastrous as they first seem, (though the initial discovery that they are missing is indexed with the “Alas, what a sin! Alas, what a shame!” sign, 14092; 14106), since they allow the girls, who were never meant for this earth, to enter maaya, the other world (14047-52). In the end it is the telltale detail of expertise in cenne-playing, woven into the the otherwise entirely expectable tapestry of phrases documenting the girls’ learning, that signals their transformation-to-come. Gopala Naika has used the traditional medium to highlight that crucial detail, to put it into relief, to underline the modest-seeming discrepancy that forecasts the plot some 1,500 lines ahead.

An oral epic in traditional and scholarly context

In summary, any scholar interested in the worldwide phenomenon of oral epic, from the manuscript- based narratives of the ancient and medieval worlds to the still extant traditions around the globe today, must be extraordinarily grateful to Lauri Honko and the Tulu-Finnish team for their efforts. Thanks to their imagination and sustained work, the Siri Epic comes to us embedded in both its own traditional context and a unique scholarly context. For those familiar with the Siri stories and ritual from an insider’s point of view, this project offers an opportunity to study and share a remarkable heritage. For those who must adopt the outsider’s perspective (surely the far greater percentage of those who read these three volumes), there exists an unprecedented opportunity to enlarge one’s comparative vision. With this publication of the Siri Epic, specialists in oral-derived traditional narrative can juxtapose their theories of composition and reception, of phraseological and narrative multiformity, of the roles of individual and tradition to real, viable, meticulously documented analogues. Regardless of what our specialties may be, we owe a great deal to the research team, and not least to Gopala Naika himself, whose 1990 performance of the Siri Epic his fellow bard Homer might well have characterized heroically as “far the best of all things.”

John Miles Foley
University of Missouri


1. The term performance arena designates a virtual (rather than a geographical) space defined by the speech-act. Every time an oral traditional performance takes place, in other words, it occurs in what may be understood as “the same place,” in that the same or a similar context is summoned, the same or a similar form of the language is employed for communication, and therefore the same or a similar set of implications is active. See Foley, The Singer of Tales in Performance (Bloomington: IUP, 1995), pp. 47-49, 79-82.

2. I here focus on the Moslem epic because of its relative length, elaboration, and general “fit ” in genre with the kind of narrative that Honko is studying. The Christian songs from this same area (sometimes from the same singer), while in certain cases treating the same or cognate stories, are typically much shorter (100-400 lines on average) and display different characteristics. See Foley, Immanent Art: From Structure to Meaning in Traditional Oral Epic (Bloomington: IUP, 1991), chaps. 3-4.

FF Network No. 17
(June 1999): 13­23

Pin It on Pinterest

Share This