The Christian poetry, too, represented the ideals of the governing class in the Middle Ages—I am thinking, for example, of the times of the emperor Dušan in the fourteenth century—and of the rebellious leaders of the eighteenth and nineteenth, as parts of Serbia began to be liberated from the Ottoman empire by the emerging kings and princes who were not ousted until World War II. Both the Moslem and the Christian poetry expressed the ideals of the governing class even when it was out of power.

Although much talked about in negative criticism, living oral-traditional literature is still not very well known, and I try over and over again in the course of this book to acquaint the reader with some of the best of what I have had the privilege of experiencing and to demonstrate the details of its excellence.

The first article in this volume was the keynote address at a conference in 1985 at the University of Natal in Durban, South Africa, “Oral Tradition and Literacy: Changing Visions of the World,” and while focusing on Homer and South Slavic, it deals with some general questions that plague those of us who work with oral-traditional literature. In it, for example, I stoutly defend the legitimacy of the term “oral literature,” and I also emphasize that the method of composition of oral-traditional narrative poetry is not “improvisation” but a special form of “composition in performance” that uses units called “formulas” and “themes,” to which I have more recently added the highly important {2|3} concept of “blocks of lines.” In one guise or another, this topic runs through all the papers.

The third paper, “Homeric Echoes in Bihać,” dates from about the same time as “Homer’s Originality: Oral Dictated Texts.” It is an early example of my interest in mythic patterns and their persistence in time and across language barriers in contiguous traditions. It appeared two years after The Singer of Tales and was one of my first attempts to study mythic patterning. The conclusion that elements in the patterns involved had survived, without the intervention of the poems of Homer themselves, since Homeric times and earlier was almost irresistible.

Chapter four speaks for itself and needs no special explanations other than to stress the significance of Avdo Međedović in Parry’s experience. Unfortunately, Milman Parry did not live to give his own account of Avdo. {4|5}

With the fifth article, “Homer as an Oral-Traditional Poet,” read at a classics symposium at the University of Pennsylvania that, unofficially, I seem to remember, had the idea of evaluating Homeric scholarship “after the oral theory,” I attempt to correct some of the misunderstandings of Parry’s “oral theory.” The paper has not been published before. In it the subjects I spoke of earlier, such as improvisation and the quality of South Slavic oral-traditional poetry, come to the fore. It was a memorable symposium, with papers by, among others, Jasper Griffin, Richard Janko, Joseph Russo, and Laura Slatkin. In my contribution I took Albin Lesky gently to task for judging “The Wedding of Smailagić Meho” on the basis of a summary of the poem that I had given to C. M. Bowra, without Lesky’s ever reading the poem himself. I also have taken issue with George Goold over his interpretation of the passages in the Homeric poems which are repeated verbatim, or almost so. He has claimed that such nearly exact repetitions are characteristic of written style and are thus proof that Homer wrote his poems. Since it is not unusual to find such passages in pure oral-traditional songs, and since editors, and even collectors, tend to “standardize” repeated passages, I do not find that his arguments agree with the facts.

“The Kalevala, the South Slavic Epics, and Homer” was read at an exciting International Folk Epic Conference held at University College, Dublin, in September 1985 to celebrate the hundred and fiftieth anniversary of the publication of the Kalevala (1835) and the fiftieth of the establishment of the Irish Folklore Commission (1935), which was to become the Department of Irish Folklore, the organizer of the conference. The Homeric poems and the Liedertheorie of K. Lachmann played a part in Lönnrot’s conception of the Kalevala, a concatenation of shorter poems of various kinds to form a long epic.

 

When I returned with Milman Parry from Yugoslavia in September 1935, it was decided that I should work for a doctorate in comparative literature, with a major in English and minors in ancient Greek and Serbo-Croatian. My graduate adviser during my first year was George Lyman Kittredge, whose place was taken after his retirement by Fernand Baldensperger. Sitting with me on the steps of Warren House, Kittredge assigned me to take Anglo-Saxon and Middle English. The following year I was to add Old Norse with F. Stanton Cawley, and eventually Middle High German with Taylor Starck. The first semester {5|6} in Anglo-Saxon was taught by Francis Peabody Magoun, Jr., and the second, devoted entirely to Beowulf, by Kittredge himself; this was his last class before retiring. For Middle English, including Chaucer, my teachers that year were first John Livingston Lowes and second Fred Norris Robinson. Magoun was one of the readers of my doctoral dissertation, “The Singer of Tales”; the others were Roman Jakobson for Slavic and John H. Finley, Jr., and C. M. Bowra, who was then visiting professor at Harvard, for Greek. I cite all this to explain how someone who was essentially a classicist found himself studying Germanic, particularly medieval English, epic poetry and Old Norse sagas in addition to South Slavic.

The seventh, eighth, and ninth chapters in this book are concerned with Beowulf and other Anglo-Saxon poems and continue an interest in them that first saw publication in the section on Beowulf in the chapter titled “Some Notes on Medieval Epic” in The Singer of Tales in 1960. Students in my seminar on medieval epic very often analyzed formulas and themes in Anglo-Saxon and Middle English poetry, and I regularly lectured on Beowulf and Maldon in my courses in comparative literature, and even included in later years the Alliterative Morte Arthure. Both the seminar and the course on the comparative study of oral-traditional epics dated from the early 1950s. In 1965 I was asked to contribute an article to a festschrift for Francis Peabody Magoun, Jr., and “Beowulf and Odysseus” was written for that volume. “Interlocking Mythic Patterns in Beowulf” was read at the Old English Colloquium at Berkeley in the late 1970s and published in 1980.

The ninth paper, “The Formulaic Structure of Introductions to Direct Discourse in Beowulf and Elene,” has a different history, although it too goes back to my dissertation and to The Singer of Tales. I was invited to give a paper at the annual International Congress on Medieval Studies at Kalamazoo, Michigan, in 1982. At the time of the Old English Colloquium at Berkeley, I was also asked to give a public lecture in which, for the first time, I attempted to answer the criticisms that Larry Benson made in an article in Publications of the Modern Language Association in 1966. In 1981 the tape of my lecture was transcribed. Even though I was not satisfied with my comments, I decided that at Kalamazoo I would return to formula studies of Beowulf. I had checked Benson’s statistics and could not fault them. On the other hand, I was thoroughly convinced that the formulaic style originated in the “oral period”—Benson had not questioned that—and was characteristic of oral-traditional narrative song. I set out then to see if I could differentiate between the formulaic style of Beowulf and that of other Anglo-Saxon poems.

As I indicate in Chapter 9, a certain amount of published research had continued to concern itself with the study of the formulaic style. In my own thinking, I had been expanding the boundaries of the formula so that they included phrases of more than one hemistich or of more than one line. I also became interested in larger syntactic units, even whole sentences, as the units of composition. In the Kalamazoo paper I began to apply those concepts to the structure of introductions to speech in Beowulf and Elene. I had also accumulated information about Andreas, Christ, Phoenix, and other poems as well. I have continued to deepen and widen my investigations of introductions to speech and of sentences beginning in the b) verse. Those studies will be reported on at a later date. The paper in the present volume, therefore, represents only the beginning of an ongoing research project that began in 1982.

The Milman Parry Collection was made for purposes of experimentation {7|8} with the intent to learn how a living oral tradition operated. Chapter 10, written for a volume of papers to honor Roman Jakobson, was the result of research, using materials in the Parry Collection, into the effect of a fixed text on the compositional technique of oral-traditional singers of epic narrative. What happened when a published, fixed text came into the possession of traditional singers? When they began to memorize that text, what were the results? In the collection we had songs that had been gathered and published in the nineteenth century and had become known in fixed texts to traditional singers in the 1930s. One summer I experimented with comparing the texts that Parry had collected with those written down in the early 1800s by Vuk Stefanović Karadžić from one of his best singers, Tešan Podrugović of Gacko. This paper, “The Influence of a Fixed Text,” is an example of one of the kinds of experimentation that the Parry Collection was intended to make possible.

The final three papers in this volume are concerned with the relationship between the South Slavic oral-traditional epics and other modern traditions, Greek and Asiatic. The first of the three dates from the early 1950s, when I was deeply involved in the Slavic department while Michael Karpovich was its chairman and Roman Jakobson and Dmitri Čiževsky were very active. Volume 2 of the Harvard Slavic Studies (1954) dealt partly with Byzantine-Slavic relations, since it honored Father Francis Dvornik, historian and professor at Dumbarton Oaks. This paper differs somewhat from the other comparative papers because it focuses attention not on mythic patterns—it was written before that time in the development of my ideas—but on a motif. Moreover, it makes use of written authors, such as Vergil and Lucretius, as well as traditional poems such as the Homeric and Serbo-Croatian epics and the Byzantine epic of Digenis Akritas. I have revised the paper to some extent (this is true of most of the papers in this volume), but I have not changed its general thrust.

In contrast to “Notes on Digenis Akritas and Serbo-Croatian Epic,” the last two papers in the volume stem from the 1980s and treat of relations with Central Asiatic epic traditions. The first of these is a kind of “archaeology” aimed at digging out long-forgotten and far-distant meanings from Bulgarian songs of this century. It started with my curiosity about an anomaly, as was the case with several other papers. In Beowulf it was the anomaly of a guest who was gratuitously insulted, as was Beowulf in the court of Hrothgar, an episode that called to mind the {8|9} scene in Phaeacia in the Odyssey in which Odysseus was insulted by one of Alcinous’s sons. There is another instance of anomaly in the Odyssey, when Odysseus’s young companion, Elpenor, fell from the roof and broke his neck. This death seemed to make little sense unless one fitted it into a mythic pattern in which a companion’s death led to the hero’s journeying to the land of the dead seeking answers to questions of mortality. This occurs in the case of the Bulgarian song in which an eagle brings the wounded hero water in its beak in return for the hero’s once having saved the eagle’s children from death. What is that all about? I chanced upon an incident in the Turkic epic of Er Töshtük which I believe suggests a plausible answer and in turn sheds light on one of the sources of the Balkan epic tradition.

The final paper also looks to Central Asia and, among other things, focuses on horses and types of stories. The latter is, of course, related to my concern with sequences of incidents in narrative and mythic patterning. The former arises from my long-time interest in the importance of horses in epic poetry in the Balkans, where horses are not especially native, and in Central Asia, where they are at home on the steppes. Coupled with this anomaly is the fact that a great many words for horses of many kinds are of Turkic or Persian or Arabic origin, and our journey to Central Asia via Anatolia and points in between is not only a natural one but also inevitable.

It has been a strange experience to review my past writings, some of them going back more than fifty years. The earliest papers in this volume {10|11} were published more than thirty-five years ago; others represent research in progress. They were addressed to a variety of audiences in places as distant from one another as Berkeley, California; Philadelphia; Dublin; and Durban, South Africa. Sometimes, needless to say, they reflect controversies pertinent to the times and places where they were delivered. Much has been written on most of the subjects treated, and I have tried to learn from what I have read of both positive and negative criticism of Parry’s ideas as well as of my own attempts to follow them. I feel it necessary to move further, in directions that the epic singers and their songs indicate, toward an understanding of the meaning of oral-traditional epic poetry. Naturally enough, I have found it necessary to edit and to revise the papers, at times considerably, because I did not want to say anything with which I no longer agreed. For the most part, however, the revision has been stylistic or concerned with a shift of emphasis; I have made very little substantive change.

As I point out in the notes, my paper from 1953, “Homer’s Originality: Oral Dictated Texts,” is reprinted nearly exactly. It has been included in several anthologies of articles on Homer, notably those by Geoffrey S. Kirk in England (1964) and Joachim Latazc in Germany (1979). I include it here in spite of the fact that I have some fear, not entirely unjustified, that it may continue to give rise to misunderstandings of what the specific advantages may be that can accrue to the dictating singer. Under the circumstances of a “dictating performance” the singer can expand his song, as I have demonstrated. He also, it appears, can enter into a different rhythm of composition from that of his singing performance. This facet of dictating, strangely enough, has not been thoroughly investigated, possibly because collectors at the present time do not write down songs from dictations but record them on tapes, including videotapes or cassettes. The rhythm of dictation may induce, or enhance, structures that are endemic to traditional poetics, such as parallelisms and other paratactic arrangements of ideas and words. Just as the singing brings with it sequences of rhythmic and melodic patterns, so the dictating may have its own distinct structures.

As I reconsidered very recently the stylization of a passage from Salih Ugljanin’s “Song of Bagdad” which was found in a dictated version but not in two sung texts, I was suddenly aware of the experience of listening to Salih dictate. In my memory I could envisage a room in Novi Pazar on a series of rainy days, and I could see him sitting at the table at which Nikola Vujnović was writing and hear him speak the lines rapidly {11|12} but with a deliberate tempo; a line, a pause, then another line, as he watched Nikola set it down on the paper. But the pauses did not interrupt either his thought or his syntax. The measured delivery made the parallelisms in the following lines of the passage I just referred to almost inevitable:

This kind of construction may emerge more clearly in the mannered dictating than in the necessary flexibility of composition in the hurly-burly of the live singing performance.

One might think that dictating gave the singer the leisure to plan the words and their placing in the line, that the parallelism was due to a careful thinking-out of the structure. First of all, however, dictating is not a leisurely process; neither the singer nor the scribe has patience for long pauses for deliberation. The hours pass and the lines accumulate as the song is set down. But more especially, a mood and a tempo, as I have suggested, are established which produce the balanced utterances of the poet. Not conscious planning but the rhythm of that particular process of composition calls forth the structures. I might add that not all singers can dictate successfully. As I have said elsewhere, some singers can never be happy without the gusle (a one-stringed bowed instrument) accompaniment to set the rhythm of the singing performance.

Scope alone, however impressive, and performance alone, however spectacular it may be, constitute but the outward trappings of the study of oral-traditional epic song. It is the singer and what is sung that count. They are strongly affected by the traditional setting of performance and by the traditional audience, but those alone do not create the singer, or the words, or what they relate. It is the mind of the composing traditional singer that we must seek to comprehend both at the moment of performance and even when it is seemingly at rest. The epic tradition lies in the myths and tales stored in the minds of all the singers past and present, the least as well as the greatest. It is this that one learns “in the field,” that is, in listening to and talking with singers. I had the great privilege of apprenticeship, short as it was, with Milman Parry; it was from him I absorbed a feeling for the epic songs of the Slavic Balkans, and it was with him that I came to know and to listen to many singers telling urgent stories of olden times.

On another level I learned from Parry the necessity of careful field recording so that the texts thus obtained would reflect accurately the words of the singer rather than those put there by the collector. To reach valid conclusions about the process of composition of oral-traditional narrative song, conclusions that could be used for comparative research, one had to have reliable texts. I hope that I acquired from Parry some degree of fastidiousness in that regard.

Parry came to know many varieties of singers, skillful and otherwise, and at least one of genius. I think that perhaps he found two Homers in those years. One, whom he has remembered in the title of his field writings, namely, Ćor Huso Husein of Kolašin, an almost legendary figure from the past who had died before Parry’s collecting years, but of whom Parry heard from those who had listened to his singing and had learned songs from him. Had Parry had time, he would have written more about Ćor Huso. The other great singer was Avdo Međedović, whose name reverberates everywhere in this volume; for him Parry lingered in Bijelo Polje during the summer of 1935. It is fitting that the {13|14} two generations should have been represented so well among the South Slavic peoples in the 1930s, because any tradition’s chances of immortality rest upon the desire of the younger singers to continue to tell the myths and histories of their elders, and upon the ability of the older generation to inspire that yearning for knowledge.

Before I close, I should like to pay tribute to Gregory Nagy’s volume in this present series, Greek Mythology and Poetics. In emphasizing the significance of myth as a source of the Homeric epic, he points directly to what, I believe, gives the Iliad and the Odyssey both deep structure and deep meaning. Reading them as traditional poems expressive of sacred realities, we can appreciate their religious—in the broadest sense—dimension, extending back to their origins.

In another book, appearing at the same time, Pindar’s Homer, Nagy gives a wide view of archaic Greek poetry and song, redefining these terms in an innovative way. His vision of the formation of the Homeric poems through panhellenization and their “textual fixation” orally, with all its ramifications, is challenging. The differentiation of the Iliad and the Odyssey, through panhellenization, from the poems of the Cycle, which represent local traditions, is a view worth serious consideration. Most significant, however, is Nagy’s concept of the Homer Pindar knew.

These two books, together with The Best of the Achaeans, present a rare holistic view of Archaic Greek poetry, including both lyric and epic, and its background in the polis and at the panhellenic festivals. {14|15}