Right through antiquity literature was oral. Poems were sung, speeches delivered, and stories told, and even reading and writing in private could normally be heard: you read aloud to yourself or had a slave reading to you, authors dictated their works to scribes, and also for relatively informal writings such as letters to friends, slaves were regularly used for doing the actual job. Going to the marketplace you were able not only to buy goods, but also to attend lawsuits, participate in philosophical dialogues, or listen to the odd performance, for instance of epics. Public festivals for the gods included solemn processions, sacrificing animals, dining in common, and attending all kinds of entertainment. Literature was an integral part of life, performed officially at public and private occasions or informally to accompany work or sweeten leisure hours. Women sang or told stories at the loom, harvesters were entertained by musicians, rowers kept the rhythm by singing sea-shanties, and gentlemen sang or made speeches at drinking parties.
However, ancient literature has reached us in written form, either via unbroken transmission during the Middle Ages, or dug out by archaeologists and deciphered by philologists. Our access to the texts is in silent reading, often slowly with patient consultation of grammar and lexicon. If we want to get an impression of how the same texts were experienced by their contemporary audiences, we have to reconstruct the original settings in our minds. The sources for this are the literary texts themselves and other written or material sources. But in order to interpret them in a congenial way, interest has focused on the way oral literature works in our own times.
Interest in the performance of epic began in Homeric studies, with the oral-formulaic theory of Milman Parry and A. B. Lord. Their understanding of the formulas and themes of Homeric poetry linked these characteristics to the conditions under which the poet created his work: he was not sitting splendidly isolated at his table writing poetry, but composed as he performed, face to face with his audience. Accordingly, to understand this special kind of poetry, studies of the context of performing became of paramount importance.
Parry and Lord were concerned with the epic genre, but their theory claimed relevance for oral poetry as such since it was concentrated on how oral composition works. So for a start it is important to underline that oral poetry includes a broad spectrum of genres, and that these differ from one community to another. They are all traditional in the sense that poets rely on the experience of their craft, and memory always has an important role to
Play. Some texts or parts of them are more easily memorized than others: poetry is remembered better than prose, brief poems better than long ones, and in rhymed verse the end of a line is more stable than the beginning. Formulas may be retained even after poets and audiences have forgotten what they meant.
It makes a fundamental difference whether texts are supposed to be true or not. Variation may be appreciated in lyrics, but is considered erroneous in genres that have as their purpose to give reliable information of events in the past or other kinds of useful knowledge. Singers or storytellers who handle such material invariably claim that they are able to repeat exactly what they have been taught, and both they and their audiences feel confident that oral transmission may be precise and reliable. Variation is frowned upon as a sign that the performer is no expert in his field. This respect for correctness of data stands in a paradoxical relation to the fact that poems actually change, and not only in matters of detail. This was a main experience of Parry and Lord’s fieldwork, and it has been massively documented in fieldwork since then.
The present chapter is about epic, and based on the following definition of the genre: epic is a long narrative poem describing historical events. By ‘‘long’’ I mean: longer than other forms in a community’s spectrum of genres. ‘‘Narrative’’ means: mainly concerned with action, but allowing for both description and reflection. It is not always easy to define a ‘‘poem’’; it has for instance been contested that epic existed in Africa because scholars did not agree on what was poetry, what prose (Finnegan 1970: 108-9 vs. Biebuyck 1976: 5; Seydou 1990: 403). I take for poetry whatever is performed in a style distinguished fTom daily speech, whether by rhythm, rhyme, music, choice of words, pitch of voice, or whatever. By “historical events’’ I mean: events that both singer and audience consider to have actually taken place once upon a time, including such that scholars might call myth or legend. (Definitions of folklore genres: Ben-Amos 1976; definitions of epic: Blackburn and Flueckiger 1989: 2-5; Honko 1998b: 20-9; Beissinger et al. 1999: 16.)
Oral epic belongs to the wonders of humanity (see Chapter 13, by J. Foley). Singers are reported to be in command of huge traditional stories and ready to perform for hours on end. The Kirghiz bard Jusup Mamay performed his version of the Manas epic in 1979; it amounted to almost 200,000 lines and was published in 18 volumes 1984-95. Besides, he knew eleven other epics (Ying 2001). The Tibetan Gesar epic was sung by ‘‘Old Man Thrapa’’ in a version of 600,000 lines (Gyaltsho 2001: 281). The Mongolian Jangar epic was published in twelve volumes 1985-96; in this case more than a hundred singers contributed (Chao 2001b: 404). In northern Africa, Sirat Bani Hilal, an Arabic tradition taking its stories from the immigration of the Hilal tribe during the tenth to twelfth centuries, has been studied by several scholars since the last few decades of the twentieth century. A Tunisian singer, Mohammed Hsini, recited a 20-hour version during sessions between 1974 and 1980, and the recording ran to over 1000 manuscript pages (Saada 1985: 24). In lower Egypt, Shaykh Biyali Abu Fahmi sang a 32-hour version over eleven nights (Reynolds 1995: 42). In Andhra Pradesh in India a Telugu bard recited aversion of the Palnadu epic, the recording of which took place during two weeks in October 1974 and lasted about 30 hours (Roghair 1982: vii). And an epic of similar length sung by Gopala Naika during six days, 26 hours’ singing time net, has been published (Honko 1998a). These long texts, telling the story from beginning to end, were mostly produced at the request of scholars, while the normal performance was of single episodes. It is remarkable that in the two best described cases both singers were immediately able to conform to the scholar’s wish and during performance give form and structure to a poem of an unfamiliar kind (Reynolds 1995: 42 and Honko 1998a).
Analyzed as communication, oral epic differs from written first and foremost in the way contact is established (Jakobson 1960: 353-9). It is direct; addresser and addressee are face to face. They can see, hear, smell, and touch each other, and they mutually influence each other as the performance proceeds. The experience is shared, and joy, melancholy, fear, or aggression is contagious among the participants. The success of a singer depends on his ability to catch the interest of his audience and keep it. He is intent on meeting their demands and is all the time attentive to their reactions. If they show signs of being bored, he introduces something exciting or, on the contrary, abbreviates his narrative and hastens to the end. The more often he performs the same narrative, the better he knows what will make people laugh or cry, and the experienced singer is skilled at manipulating the feelings of his audience. But when asked, listeners say that they have been moved by the truth of the tale.
There are many variations from one tradition to another in the way songs are presented. Often singers accompany themselves on a stringed instrument, and there may be other musicians involved; the music is considered very important (Seydou 1983: 43-4; Saada 1985: 24; Reynolds 1995: 12-13). There may be extra singers, and parts of the narrative may be dramatically performed (Biebuyck 1972: 262-4). But there are also singers who use no other instrument than their voice (Honko 1998a). Perhaps the audience takes part; for instance, they may repeat lines or refrains (Honko 1998b: 548-57). Even the singer who performs alone may establish a dramatic effect by means of mimicry, gesture, and variation of voice. Some scholars talk of the epic text as a libretto, to remind us that the words we read are only part of the event (Foley 1998c: 80-5). The style varies with differences in aesthetic, but it is everywhere characterized by repeated phrases, parallelisms, adding style, type scenes, and patterned narrative.
The linguistic form tends to be solemn and characterized by archaisms (Quain 1942: 16; Biebuyck 1976: 28-30; Reynolds 1995: 44-6; Chao 2001b: 420). But little seems to have been done in the way of precise study. Epic traditions in the modern world typically exist in communities peripheral to the centers of wealth and power, and express themselves in languages that often do not even have an orthography. It is doubtful if the language and style of any living tradition will ever be analyzed in a way remotely similar to how Homeric language has been studied.
Occasions may be of many kinds, and in most communities who have an epic tradition they vary from big, formal events to private entertainment. Parry’s and Lord’s informants had their most important function during Ramadan, when many innkeepers hired a singer to entertain their customers at night (Lord 1960: 15). The singer Awadallah Abd aj-Jalil Ali, living near Aswan in Egypt, stated that he used to perform during pilgrimages, in cafes, and at private parties arranged by a patron, such as at the end of Ramadan, at weddings, or at the annual workers’ feast at the local sugar factory (Slyomovics 1987: 10). In communities where the interest in epic is sufficiently widespread to offer a basis for industrialization of performance, singers become good at delivering versions that fit precisely into a radio program or a tape (Wadley 1998: 155). Tulu epic makes an integral part of the cult at certain religious festivals, which may last 10-14 hours per night during two to four nights (Honko 1998b: 389-499, 2000a: 13). The Palnadu epic is regularly recited at the annual festival for the heroes of the narrative. The festival is formally called, sacrifice given, and during the week’s festivities epic is recited the first evening and the following four days. Certain episodes are prescribed for certain occasions, not following a chronological order of the narrative. On the fourth day there is a procession in which the protagonists of epic and festival are represented dramatically (Roghair 1982: 26-31). What is of paramount importance is that the occasion must offer a sufficient period of time for an audience to attend a rounded, harmonious performance.
As a rule, epic singers are male. Their social status varies from one community to another, and so does their education. They may learn just simply by attending other singers, as described by Lord (1960: 13-29), and often their teacher is their father or grandfather (Slyomovics 1987: 6-7; Harvilahti 2000: 218-19). There are communities in which epic singing is highly specialized and runs in certain families, as among the Bambara in the central parts of Africa. Here one and the same family recruits musicians, counsellors, negotiators, and epic singers, and small boys are under constant supervision so that the most gifted can already begin training as young children; apprenticeship lasts 10-15 years (Bird 1972: 278-80; Biebuyck 1976: 20). To learn the epic properly, they must make a tour from master to master and end at a recognized center at Krina in Mali. Here a special ritual takes place every seventh year at the sanctuary of the ancestors; the established bards participate in secret rituals and recite to each other, and young singers pass a kind of exam (Seydou 1990: 414). Sometimes singers state that they have been taught supernaturally, in a dream or a trance (Gyaltsho 2001: 280-6), and one tells of his grandfather that he had a book given in a dream and was immediately able to read it (Slyomovics 1987: 11-12). The above-mentioned Kirghiz singer was taught by his brother and by reading, but nevertheless maintained that he had been instructed by the heroes ofthe epic who had met him in a dream (Ying 2001: 223-8). Since the text is flexible, each performance is a recomposition of the narrative. Accordingly, in oral epic the terms ‘‘singer,’’ ‘‘bard,’’ and ‘‘poet’’ are synonymous.
The audience can be mixed or only male, but is seldom narrowly restricted. One instance is reported of an epic being regularly performed to a female audience, the Tulu tradition of Karnataka in India. Here one of the situations in which epic is performed is during work in the paddy-field. The singer is male, but the female workers join in as a choir, and some of them are also able to perform as main singers (Honko 1998b: 552-3).
The influence of the audience on the performance has been described as social control (Bogatyrev and Jakobson 1929: 901-3). What is not accepted by the audience will disappear from the tradition. In the case of epic, it is especially a question of truth: normally the story is known to the audience already, and they will accept only what they feel to be correct. Ancestors of persons who are powerful in the community will be important in the narrative; if their family dies out, they tend to disappear. This mechanism is called ‘‘telescoping’’ (Henige 1974: 5-6). In some cases performance is arranged by a patron, who as a special kind of audience exercises supreme authority over the text. This is not necessarily taking the form of open control, but may function as a self-censure on the part of the singer.
All kinds of performance to an audience have as one of their main functions to establish a feeling of group identity. But this is of special importance in epic performance, since it links participants not only to each other, but also to their ancestors. When attending the epic, the audience members feel that their own world is a continuation of theirs, and that they contribute to keeping the heroes alive by hearing of their great deeds. The performance reactivates the events of former times, and one scholar even speaks of ‘‘the intense tripartite relationship that obtains between the poets, their listeners, and the heroes’’ (Reynolds 1995: xiv). Often epics work as a means of legitimating affairs of the present. A narrative about how the tribe traveled until they finally reached the place where they now live, confirms their claim to be the legitimate inhabitants of the area in question (Vansina 1985: 91-3). There is also a didactic aspect of epic performance: people identify with the heroes, take them as their models, and learn from their misfortune or success. Among the Nyanga in Congo the hero even embodies features not appreciated in the community and thus tests the boundaries of good and bad (Biebuyck 1972: 272; Seydou 1990: 409). The group who is attending is inevitably led into negotiating their shared values.
In spite of this great, varied mass of examples, it is striking that a series of common traits link them all together: the special register of style and language, the respect for facts, the feeling of shared values among those present, including gods and ancestors, and of course the sheer entertainment of narrative. These characteristics are all based upon the event of performance, the face-to-face communication during which the singer recomposes his poem under the steady cooperation and control of his audience. It is his mastery of the tradition that renders his words authoritative and gains the acceptance of the listeners. This magic, creative, and potentially dangerous situation has been called an arena, in which word-power is established by means of the enabling event of performance and the enabling referent of tradition (J. Foley 1995: xiv and passim).
Even though each performance normally concentrates on a single episode, the full story lies behind, known in more or less detail to both singers and audiences. This has been called the ‘‘pool of tradition,’’ which furnishes the singer with stuff for his narrative (Honko 1998b: 66-74, cf. Biebuyck 1972: 266; Aloni speaks in the same sense of a ‘‘macrotext,’’ 1998: 25).
The word ‘‘text’’ is currently used in different ways by scholars. To some it means a written text as opposed to the ephemeral oral poem (e. g. Bakker 1997: 1), and a similar understanding is implied also in the new word ‘‘textualization’’ for the process of recording an oral poem (the idea is discussed in Honko 2000a). Others speak of text also in oral performance, and even of‘‘mental texts,’’ songs that exist as basic entities in the singer’s repertory. When learning a new song, the singer works on it in his mind and in performance until it finds a form with which he is satisfied. However much he varies his narrative from one performance to the next, he sticks to this mental text, being able to retain it in his memory over any length of time. Such mental texts are not, however, transmitted from one singer to the next (Honko 1998b: 92-9).
One may speak of pools of tradition on more than one level. Considering the similarities that exist among Near Eastern and Greek poems (Chapter 20, by Burkert), the stories behind them might be called a common pool of tradition extending over a period of two millennia and the whole area of the eastern Mediterranean and Mesopotamia. Similar musical instruments suggest that this pool of tradition included not only stories, but also the habits of how to perform them. Next come pools of tradition that unite more limited areas linguistically or geographically, and at the other end of the scale are pools that contain single cycles of stories such as the deeds of Gilgamesh or the Trojan War.
Only in one case is the performance context of a Mesopotamian epic actually known (see Chapter 15, by Noegel). Enuma Elish is mentioned in a ritual text which describes in great detail the new year’s festivals for Marduk (Pritchard 1955: 331-4). On the evening of the fourth day the urigallu-priest shall recite this poem to the god. This makes good sense with respect to the text we know since it is full of generous praise of the god. There are also passages of the poem in which orders are given about cult practice and recital (e. g. VI. 106-22), and it ends with an epilogue stating that it should be remembered and commented upon, discussed by experts, recited by father to son and heard by shepherds and herdsmen (VII. 145-8). The very last verses even state that the text has been dictated by an older man and written with the purpose of being heard by younger generations (VII. 157-8, according to West 1997b: 598). In the ritual text there is nothing to suggest that an audience should be present when the priest recites to the god, so the just mentioned passages point to other, less formal, performances. It is interesting that this written text so clearly counts upon an oral afterlife. The Hittite Kingship in Heaven opens with an address to various gods, a fact that suggests that this poem was meant for a similar cultic context (see further Chapter 17, by Beckman).
Mesopotamian epic shares with oral epic in the modern world the characteristic features of style, and it is composed in a special hymnic-epic dialect, different from common speech (Foster 1993: 3). Add to this the intriguing fact that there often exist more than one version of a poem. Experts do not agree, however, on the question of the relevance of the oral-formulaic theory (Vogelzang and Vanstiphout 1992). The Gilgamesh epic is impressive not the least for its great number of manuscripts in several languages from widely differing areas. The classic study analyzes the history of the epic exclusively in terms of written composition (Tigay 1982). But a more recent survey of the epic’s history tells a fascinating story of Sumerian and Akkadian oral traditions existing side by side, a great Old Babylonian poem composed and transmitted in writing and eventually lost, and a new written composition, the Standard Babylonian version, which was then preserved in written copies during many centuries (U. and A. Westenholz 1997: 23-54).
Poetry from the Near East is often accompanied by information about how it is to be performed. The biblical psalms are introduced by such notes, and Sumerian poems have comments on genres and instruments. Even epic is regularly opened with the phrase ‘‘Let me sing...,’’ and the texts are interspersed with instructions about how they are to be remembered, recited, and heard. At the end the scribe often added a colophon declaring that he has copied the text carefully. From Ugarit we even know the name of a scribe, Ilimilku; he says of himself both by whom he has been taught and that he works for the king, Niqmaddu (Parker 1997: 42, 141, 164). And when at the end of the Kirta epic he states that the text was donated - presumably by the king - we have what seems to be a reason for the writing of the poem, that it should be dedicated to a god.
Since oral epic has no need to be written down and both singer and audience feel convinced that memory is reliable, it has become a difficult question why texts were written at all. In the case of Near Eastern epic it has been suggested that a reason for writing might be found in the scribal milieus, and that the poems have been written so that scribes might have classics to study and copy (Alster 1992: 26).
Compared to the vivid Homeric representations of performing singers, Near Eastern epics are fairly reticent about singers in action. In two of the Sumerian Gilgamesh versions, however, the king’s singer plays a minor but crucial role (‘‘Gilgamesh and the Bull of Heaven’’ 103-20 and the Gudam epic 13-24), reminding king and people of their duties. It is remarkable that in both contexts the singer does not try to please his audience, but is on the contrary teaching them moral lessons.
The poet Enheduanna has described how she composed the poem Ninmesharra: she bore the song during the night, recited it to herself, and when the day had come called for a singer and taught him the poem (138-40, see Alster 1992: 29; West 1997b: 598). A similar passage is in the Poem ofErra: Kabti-ilani-Marduk, son of Dabibi, composed the tablet. God had revealed the poem to him during the night, and in the morning he recited it without leaving out or adding a single verse. The divine hero attended the recital and accepted it (42-6).
Such passages both testify to the experience of poetic composition as a divine gift, and argue for the authority of the poem in question. Scribes insist on the reliability of their work, whether they have copied fTom a written text or taken it down from dictation. The Standard Babylonian version of the Gilgamesh epic opens with such a guarantee of truth: the poem goes back to a stele erected by the hero himself, and the reader may even consult a book preserved in a shrine at the foundations of the city walls of Uruk (I.1-29). In a similar way Naram-Sin and the Enemy Hordes is framed by references to a tablet-box and a stele, written by the protagonist (J. Westenholz 1997: 266, 279, 301).
Cooperation between poet and scribe seems to have been normal. A fascinating description of the whole process of producing a book is to be found in the Bible (see Chapter 19, by Niditch), where it is said of the prophet Jeremiah that God wanted to use him for taking the impious king Jehoiakim to task (Jer. 36). The prophet was not allowed to visit the temple, however. Therefore God commanded him to call for a scribe, and a book was produced. The scribe brought it to the temple and read it aloud to a big audience. When the king heard of this, he asked for the book, had it read aloud, and as the recitation proceeded, the king cut the book-roll in slices and threw it bit for bit into the fire. But God’s words came back to the prophet, he called again for the scribe, a new book was written with the same content, only this time the prophet added more words. We may note, firstly, that the book was only needed because of the special problems the prophet had in gaining access to an audience. Next that he considered dictating to a scribe the normal way of writing. Finally that the king’s attempt at annihilating the text by burning the book was of no avail: the prophet had his mental text which he could reproduce, in even a fuller form than the first one.
While Near Eastern poetry is very conscious of its being written, archaic Greek poems hardly ever mention writing. And it is significant that in the Near East truth is ultimately guaranteed by reference to written texts: Gilgamesh’s monument and book, the tablet of fate that the Anzu-bird stole from the god Enlil, the ten commandments handed over in writing from God to Moses, and the book of law found by king Josiah when purifying the temple (2 Kgs. 22-3). Homer and Hesiod, on the contrary, base the authority of their poems on the Muses. This difference must have to do both with the fact that the Near Eastern poems belong to a community in which writing was much older and literacy more widespread than in archaic Greece, and with the different status of scribes: in Mesopotamia and Palestine they were respected professionals, while in Greece writing was slave work. Even in the classical period Sophocles’ Antigone holds king Creon’s written laws in contempt as something easily changed compared to the unwritten, eternal laws of the gods (453-7).
The Iliad and the Odyssey both begin with an invocation of the Muse in which she is quite directly asked for the story and told where to begin. The assumption is that she is in command of the story line and able to start at any given point in the sequence of events. The situation is parallel to a passage in the Odyssey in which the Phaeacian bard Demodokos is entertaining at a meal in the king’s court. One of the guests asks him, to sing of the Trojan Horse from the point when Epius constructed it with the help of Athena. And when he begins, the phrasing is revealing: he began with the Muse and showed the song (phaine d’ aoiden) (8.492-501). It is as if the story is materially there, ready to be taken forth just as you would take some precious object out of its cask and show it to your guests.
Nowhere is the Homeric poet as explicit about his relationship with the Muses as when in the Iliad he is beginning his long register of the warriors who participated in the campaign, the so-called ‘‘Catalogue of Ships’’ (2.484-93). Here he underlines that the Muses are there as his eyewitnesses. They are immortal goddesses, they were present when the war took place, and they know. But in the immediately following lines the poet’s worries shift from the problem of knowing to that of performing: so many were there that I could not mention them all, even if I had a hundred tongues and lungs of steel, if the Muses were not willing to support me. The problem of knowing shifts into a problem of physical capacity. An important trend in performance studies is concerned with interpreting the rhetoric of the poems as specially designed for oral delivery (e. g. Minchin 1996 and Bakker 1997).
In discussions of when and how Homer and Hesiod were performed, scholars have of course first and foremost looked at what the various poems say. The scenes in which singers perform to an audience have been taken as a kind of self-portrait, and so has Odysseus’ elaborate narration of his travels, commented upon by the Phaeacian king with the remark that he has described it all truthfully, just as if he had been a singer (Od. 11.368). Subtle analyses have been made of the speeches in the Iliad as performance: how in scene after scene performer and audience interact; how different kinds of performance are successful, or not; and how structures of power control the performance (Martin 1989, discussed and continued by Minchin 2002). It is by no means simple, however, to deduce from these texts’ internal scenes how performance of the poems themselves worked. In other matters the narrator is explicit about the distance between the world of the heroes and his own: at the time of the Trojan War princes were richer and heroes bigger and stronger than those he and his audience meet. Roughly speaking it seems probable that situations and protagonists may be idealized, but that ideas of aesthetics and procedure are those that the poet himself embraces.
Meta-passages are more reliable, but they are few. The narrator of the Theogony describes how Hesiod was called by the Muses to become a singer (22-34), but the syntax of the passage does not make it clear whether narrator and Hesiod are one and the same person or not, and anyway, how an epiphany is to be understood is not easy for us to know. Simpler to handle is a passage of the Works and Days in which the narrator tells of his experience as a singer: he once participated in a contest at the funeral of a Euboean magnate, an occasion at which he won a tripod which afterwards he consecrated to the Muses (650-9). The convention of the Homeric hymns is that the narrator begins - as was the rule already in Mesopotamia - by declaring: ‘‘I will sing,’’ and the object of his song is a god. At the end he takes leave of the god, asks for his or her support and states that he will proceed to another song. The help he asks for is often concerned with winning a prize. In a famous passage (3. 104. 4) Thucydides refers to the Hymn to Apollo and calls it a prooimion, a prelude, a terminology that suggests that such hymns have as their function to introduce epic performance, and that ‘‘the other song’’ mentioned at the end of the hymns is an epic narrative. The most detailed comment on the ‘‘poetic I’’ is a passage in the Hymn to Apollo in which the narrator addresses a chorus of young girls in Delos, asks them to tell everybody that the best songs are those sung by the blind bard from Chios, and promises to sing their praise wherever he comes (156-76). This is presumably where the portrait of Homer as the blind singer originated.
These passages all refer to the poet’s activity as a sung performance, deeply embedded in competition. The occasions for performance are a private funeral and a public festival for a god, and the singers involved are itinerant and presumably professional. The audiences mentioned in the Hymn to Apollo are the god and the young women, and it is a reasonable guess that men were present too.
From many other sources - literary texts, inscriptions, and decorated vases - it is known that epic was sung by rhapsodes; a portrait of such a rhapsode is given by Plato in the dialogue Ion. It has been suggested that the singers who perform in the Odyssey were creative ‘‘singers of tales,’’ as were those Parry and Lord met in Yugoslavia, whereas the rhapsodes learned the Iliad and the Odyssey by heart and recited them, and that such a development from creativity to memorization is expressed by the different terms, aoidos as against rhapsoidos (Sealey 1957). However, considering that even the most creative of singers maintain that they are repeating faithfully the song they have learned, it seems unlikely that singers and audiences were aware of such a distinction, and the difference in terminology is understandable instead as a difference in style, aoidos being poetic and rhapsoidos prosaic. (The meaning of the word rhapsoidos has been discussed in Patzer 1952; Ford 1988; Nagy 1996c: 60-6; Aloni 1998: 48-9.) In antiquity two different etymologies were given: from rhaptein, to stitch, or fTom rhabdos, a staff; the rhapsode is one who stitches verses or stories together, or he is one who is leaning on a staff. In modern times the rhaptein etymology is considered correct. Both understandings of the word are documented in Pindar’s poetry (Nem. 2.1-3, Isth. 4.37-8), and when in the Theogony the Muses hand a staff to Hesiod as a sign of his initiation this seems also to suggest that he was considered a rhapsode (30).
Much interest has been focused on the so-called Panathenaic rule. A couple of sources mention that either Solon or Hipparchus ordered that the Homeric epics should be sung at the Panathenaic festivals ‘‘in order, so that where one stopped the next should take over.’’ It has been argued that here we have the reason why the Iliad and the Odyssey were recorded in writing: in order to be able to perform like that the rhapsodes would need a fixed, authoritative text. Some have proposed that the purpose of the rule was to control the political message of the performance. There is also a discussion of the terms of the competition, and recently the competitive form of the Platonic dialogues has been used as a key to understand rhapsodic contests (Merkelbach 1952; Jensen 1980: 128-58; Nagy 1996c: 107-14; Bakker 1997: 31; Jensen 1999: 26-7; Collins 2001; Nagy 2002).
Oral performance of epic continued in the Greek world, as inscriptions testify; rhapsodic contests were a common element in many festivals of the city-states. However, singers hardly occur in our other sources, a fact that is probably to be explained not by disappearance, but by loss of social status. A sign of their existence is to be found in the Biographies of Homer and the description of the Contest between Homer and Hesiod, since they show that in imagining the way of life the great poet had led, the natural model was that of an itinerant bard.
However, a lot of names of epic poets have survived (Ziegler 1988). Typically they were employed by the great military commanders who dominated the Greco-Roman world after the death of Alexander. Little is known of them and their poetry. Nevertheless, there are two epic poets who have survived with a certain individuality, one historical and one fictitious, and both of them are described by Roman authors. One is the poet Archias whom Cicero defended in a case of citizenship. Among other things he had accompanied a Roman general in his campaigns. In his speech Cicero not only states that Archias is an excellent poet of written Greek epic, but also dwells on his exceptional ability to compose verses extempore (Arch. 8.18). The other one is a figure in Petronius’ Satyricon, carrying the Greek name of Eumolpus, but speaking and composing in Latin. Like Archias he is able to perform on the spur of the moment, and his longest work is an iambic poem describing the fall of Troy, inspired by a series of pictures he and the narrator contemplate in a portico. His performance comes to an abrupt end when passers-by begin to throw stones at him. Later we hear of a similar incidence in a public bath. But he is so dedicated to his art that he resumes his performance at the first possible moment, and even when at a certain point he suffers shipwreck he has no time to be saved because he is just finishing a verse (Petr. 89-93, 115). Again, such a figure must have been drawn after a living model.
The epics that have been preserved are not of this type, however. In literature as in so many other matters the Hellenistic period marked a fundamental change. At the museum in Alexandria, Callimachus and his contemporaries discussed whether it was possible at all to compose epic any more, considering that the Iliad and the Odyssey were the masterpieces nobody could compete with. Nevertheless, the genre continued and even seemingly unchanged, in hexameters and with mythic content. But an epic such as the Argonautica by Apollonius Rhodius is careful to handle traditional myths in untraditional ways, and there is nothing to indicate that it claims to offer a true narrative. When Roman poets began to imitate Greek epic, they had basically two concepts to choose between, the encomiastic form presumably embraced by poets engaged to celebrate military achievements, and the intellectualistic form as cultivated by Apollonius; the situation is best known fTom Ennius (Ziegler 1988: 63-92; Goldberg 1995: 111-34).
In the field of Latin literature, performance studies have mainly concentrated on questions of rhetoric (e. g. Vogt-Spira 1990). Poetry may be oral with respect to composition, transmission and performance (Finnegan 1977: 17), and in this last sense epic continued an oral life in imperial times. Reciting poetry or reading it aloud was a well-known feature of Roman cultural life. Virgil’s Eclogues are said to have been recited in the theater by cantores, singers, and the poet himself read aloud from his works to Augustus and Maecenas (Donat. Vit. Verg. 26-34). Horace is highbrow and speaks condescendingly of readings, and especially of such that take place in theaters or baths. He himself prefers to read only to friends, and only if they compel him to (Sat. 1.4.74-6). Martial envisages a great variety of places in which his book will be recited (7. 97 and 12. 2). It is clear that readings offered important means for a poet to gain publicity for his works, and also that patrons took an interest in the institution, as is demonstrated by the fact that both Nero and Domitian arranged for poetic competitions as part of their cultural policy. Of the poet Statius it is said that performance of his epic The Thebaid was a great success (Juv. Sat. 7. 82-6).
Pliny often refers to literary readings, and he sees them as a kind of pre-publication of works. By reading aloud to an audience the author has the chance of getting feedback, which is a useful stage in the finishing of a book (Ep. 7.17, cf. Cic. Off. 1.147). In such private readings the audience have presumably been polite and cooperative. But in one of the more entertaining descriptions, Pliny complains that listeners show little interest in such events. They prefer to sit chatting outside and only now and then have a slave report how far the reading has proceeded. They enter to attend only late and with hesitation, and many leave quite openly before the performance has been brought to an end (Ep. 1.13).
The coexistence in Rome of books, readers, and bookshops on one hand and oral performance of literature on the other has been discussed by scholars in some detail. Performance to invited friends seems to have been the most prestigious context for literature, whereas the very fact that a book left its author’s control and offered itself to anybody who might take an interest in it gave it a status similar to that of a prostitute (Habinek 1998: 103-21).
Arma uirumque cano, ‘‘I sing of weapons and a man’’ - such are the opening words of Virgil’s Aeneid. But Virgil was not singing. His choice of verb was a way of inscribing his epic into a tradition that was not his, but which he wished to call into life in the minds of his readers. And when a few verses later he says, Musa mihi causas memora, ‘‘Muse remind me of the causes,’’ his expression is metaphorical: he was not relying on memory in order to master his story. He built on a careful reading of his forerunners, preferably with commentaries, and if he needed a Muse she was not called for to furnish him with material for a performance, but to ensure inspiration in a much vaguer way. In Virgil, the singing poet and the reminding Muse are literary convention, and as such they were to have a long afterlife in western literature.
World History
