1. Words Heard and Words Seen*

It seems superfluous to remark that in the history of mankind words were heard before they were seen. For the majority of people, as a matter of fact, words still are heard rather than seen, and even those who have learned to visualize words as containing particular letters in a particular sequence continue to operate much of the time with the heard, and hence the spoken, word. In our individual experience we share in varying degrees in both worlds. We have gone in our individual development from orally conceived words, without visible representation, existing within boundaries defined by utterance rather than spelling, to a sense of words with rigid, visual characteristics; cultures, like individuals, moved from one world to another through a series of gradual adaptations. Although the two worlds, the oral and the written, of thought and its expression, exhibit some striking and important differences, they are not really separate worlds.

As my title indicates, I intend to discuss words rather than orality, words in their oral form, as it were, and words in their written form. [1] I {15|16} want particularly to treat the artistic use of words in what we rightly call literature, and to stress the positive and creative qualities of oral literature, the oldest literature in the human world, and its significance for "written literature.” [2]

We use the word "literature" in at least two senses. When the automobile salesman tells us that he will give us the "literature" about a given model of car, he is not using the word in the same sense as the Department of English Literature at a university. When scholars say that they have read all the "literature" on Beowulf they are not speaking of belles-lettres. The salesman's "literature" means "something in writing," and the scholars' "literature" indicates "what has been written" on the Old English epic. In this case, scholars and salesman are using the word with the same meaning. [3] The English department, on the other hand, has made a qualitative judgment on part of the vast amount of written documents. Some people, stressing the etymology of the word "literature," make a distinction between the written and the nonwritten, thus viewing all literatures as written, by definition, as the origin of the word implies. At the same time, the same people might hesitate to subscribe to the idea that everything that is written is literature, although that is the automobile salesman's attitude. They would insist that literature means belles-lettres. This is a well-attested use of the word. Surely you have heard someone say that a given piece of writing is not "literature." Sometimes writings that contain many obscenities are condemned because they are not literature, and sometimes they are defended because they are literature. Whichever side may be in the right, they are both speaking about the quality of what is written, not whether it is written or not. In that use of the word "literature," therefore, we make a distinction of quality among various expressions in words. It is to that meaning of literature that I turn, for under it we can speak of both an oral and a written literature, products of verbal expression of high artistic quality. In sum, words heard, when set in the forms of art, are oral literature; words seen, when set in the forms of art, are written in literature. {16|17}

These considerations lead us to the question of what the role of writing is in literature. Written literature is dependent on writing. That sounds axiomatic, but the type of literature that I think of as "written literature" par excellence, historically, was created in writing and was impossible without writing. Let me illustrate by way of explanation. Can you imagine James Joyce's Ulysses being created without writing? Or a poem of e.e. cummings, whose very name must be seen to be recognized? Or a short, graphic example from Ezra Pound's Canto LXXVI: This is visual poetry; its very placement on the written or printed page indicates a phrasing and emphasis in meaning; and its lack of punctuation is a purposeful element put there by the author to convey a message. You must see it to understand it fully. The Italian quotation was taken from Guido Cavalcanti's Donna mi prega. [5] This is real borrowing from a thirteenth-century poet, impossible without a written text. This kind of poetry requires writing. These words have to be seen.

Even such lines as the following from Yeats's "The Wanderings of Oisin," which exhibit some of the characteristics of oral literature, are inconceivable without writing:

In spite of the adding style of this lovely passage—balanced as it is, nevertheless, with necessary enjambements—this poetry must be seen as well as heard, so that one may go over it again and again to appreciate its subtleties. If Yeats's lines were really oral-traditional lines, and if you were in the traditional audience or its equivalent, you would not need to go back over them to savor them. The traditional diction would be familiar, known, understood, and appreciated on first hearing, because words and word-clusters or configurations like them had been heard before. They were "just right." On the other hand, the phrase "sooty fingers" has no traditional resonances, and the same can be said for the sentence "many a tree rose out of the warm sea." This is neither traditional diction nor traditional imagery. It is individualistic in an individualist's milieu. Its particular style, its striking choice of words and ideas and poetic combinations are purely Yeats. Song-birds cling to every branch "like swarming bees," which just might be traditional, but a million of them stand on the shore "like drops of frozen rainbow light," which I wager was not. These delights are in a tradition of written poetry, but are not in an oral traditional Hiberno-English poetry. The technique here, indeed, is to seek a striking nontraditional image.

Few cultures with which I am acquainted have developed writing from within their own society. For many of them writing was brought to them from outside, from a "more advanced" culture, or at least from a culture with writing. But writing does not always imply written literature. The ancient Greeks first presumably devised a script called Linear B, probably adapted from one called Linear A, in the second {18|19} millennium B.C.; and in the eighth century B.C. they borrowed and modified the Phoenician alphabet for writing the Greek language. One of the noteworthy facts about Linear B is that it seems not to have been used for writing down Mycenaean oral-traditional literature or even for creating a written Mycenaean literature. Mycenaean literacy served the interests of trade or religion. In Mycenaean times, to be literate had practical mercantile or cult implications but none concerning the culture of literature.

The Greeks themselves then developed a literary culture from within their own ranks. There may have been outside models from ancient Near Eastern cultures with writing and with a literature in writing, be it written down or primary, which influenced the Greeks in that development. I have often wondered whence the idea came to someone in the eighth century B.C. to write down the Homeric poems, since whatever had been written up to that point had been aimed to further commerce or administration. There is the possibility that the writing of the earliest Hebrew scriptures, or the terracotta tablets of other Near and Middle Eastern peoples, may have become known to the Greeks from their contacts with the Near East and that they may have given the Greeks the idea of writing down their own myths.

Greek literature was already formed when it was first written down. The earliest written texts, such as the Homeric poems, could not be transitional, because oral literature was highly developed and so far as we know, written literature, as written literature, was nonexistent when they were recorded. It might be said that on the basis of the oral-traditional Homeric poems and other archaic Greek poetry ancient Greek written literature was created. The oral period must have lasted for a long time and true written literature must have been worked out very gradually; oral literature satisfied all requirements. [7]

Writing did however, provide an opportunity for Homer to dictate—or write, if you wish, although I find the idea incongruous—a song, or songs, longer than a normal performance. [8] It took away one set of time {19|20} limits, that of performance before an audience, the circumstances under which the traditional epic was usually sung. It imposed another set of time limits, more flexible, but artificial and probably difficult for the poet/singer without the accustomed type of audience. Yet the flexible time limits held great potential for more prolonged composing than occasion usually afforded, a different kind of performance, as it were. At this stage, that potential was for greater length, for more of the same, nothing else.

The case of the Germanic peoples is in part much the same as that of the Greeks. They had a runic alphabet with a restricted and nonliterary use. [9] Like the Mycenaeans, they did not have a literature written in that, their own, alphabet. Like the Mycenaeans, too, they had an oral-traditional literature. That it was highly developed we know, because when it eventually came to be written down, it was revealed to be of a complexity in its structure that argued a formative period of generations before writing recorded it. Words heard were sufficient for literature, for ritual utterances, for the recounting of myths, and for the telling and singing of tales, just as words heard were sufficient for everyday communication. Literary language, oral or written, after all, differs from everyday language in its function, particularly in its association with the sacred world. It was characterized by repetitions of sounds, and by parallelisms of structure, for example, which had the function of rendering magic utterances more powerful and hence more surely effective. Writing was not needed for those devices. In fact, when you come to think of it, the written word becomes operable as sound only when spoken aloud! Many basic rhetorical devices of written literature do not depend on words being seen, but come to life only when they are words heard.

In the case of the Germanic tribes in continental Germany, Scandinavia, Iceland, and England, writing was not used to record and eventually really to write literature, until the peoples were converted to Christianity. The Church, moreover, brought these tribes not only an alphabet, but a developed literature in a hieratical language. The alphabet that came with Latin was used to write down some of the oral literature. Since that literature was pagan—non-Christian—in its sacred ambience, however, the Christian Latin texts were translated or paraphrased in the vulgar tongue so that the teachings could be understood by those who did not know Latin. And the pagan oral-traditional literature {20|21} of the people was sometimes adapted, when possible, to Christian ideology. And thus gradually a new phase began in those literatures and a real written literature in those languages began to appear. It was an amalgam of two cultures, the vernacular with its own developed oral-traditional literary style and the new Christian Latin culture. The first effect of the latter was on ideas, on content, rather than on style, because, especially in poetry, Latin written style was not easy to reproduce in the metrical and alliterative schemes of Germanic verse. The oral-traditional vernacular style continued for some time to be the backbone of the new vernacular written literatures. Only then among the Teutonic peoples did words heard become literally words seen. Yet, except for a certain small and limited group of people, the literate—not only those who could read and write, but, more specifically, those who actually read literature—the vision of the world of orality changed not one iota.

Latin brought with it not only religious writings and the works of the Church Fathers such as Augustine, but it also made available the great writers of ancient Rome, such as Vergil and Ovid, and the new non-religious Latin literature. All these writings eventually played a decisive role in the development of the new literature in the vernacular. And a new secular Latin literature appeared, which for a while dominated the learned world as well as producing a medieval Latin literature of great distinction.

Oral literature did not need writing to become literature, and it continued long after writing was invented. Walter J. Ong has given us a useful term, "oral residue," referring to the characteristics of orality which remain in the world of literacy after the introduction of writing. The term applies very well to literature. But his "characteristics of orality," given in his book Orality and Literacy, were really intended to cover many other areas than literature. [10] Accordingly, they apply more widely than "words heard and words seen," to encompass a psychology of the "oral mind" and many facets of the world of the unlettered, including their literature and its interaction with the written word.

At this point it would be helpful to distinguish oral literature from oral history and also to place oral literature in respect to that vague but useful term “oral tradition.” [11] Sometimes that term is synonymous with oral literature; it may be another way of saying "oral-traditional literature." {21|22} For example, one may hear, or read, that a particular story is found in "oral tradition." More often, the term is used to designate oral report, which shades into oral history. Oral tradition in that case covers what one hears of what has happened in the past, distant or recent Although it can exist in a casual form, when it takes on a formal aspect it is oral history. Literacy has little or no effect on oral history, except that eventually, when literacy becomes widespread and begins to be used for recording, and finally for writing, literature, the writing of history is an important part of that larger development.

Oral literature is varied. It includes a number of genres, and each has its own role. In it, stories are told, songs are sung, riddles are posed, proverbs are wisely expounded, and in Africa praises are "performed." Stories and songs entertain and instruct, as do also the more humble, shorter forms of proverbs and riddles. Each has its time and place. Certain genres of wisdom literature—proverbs and riddles—sometimes, in fact quite frequently, are contained within stories and songs. Genres are not watertight compartments.

Among the shorter forms of song is a group that includes ritual songs of several kinds, such as lullabies, wedding songs, laments, and keenings at funerals. Here, too, belong the praise poems for which Africa is justly famous. [12] Laments and praises tend to be pure improvisations. Literacy has little effect on the shorter forms of oral literature. Except for an occasional collector, no one would think it useful to write down proverbs, riddles, or sayings, and, by definition, improvisations do not require writing.

Prose stories in oral-traditional literature, that is, anecdotes or more complex folktales, do not have set texts, except that there are "more or less stable" introductory formulas, such as "once upon a time," and concluding words, as well as some short set "runs," for frequently recurring passages.

Excellent examples of such runs can be found in Irish storytelling. In his relating of the long hero story Eochair Mac Ri in Eirinn, Eamon Bourke used the following run whenever the hero came to a giant's castle and "struck the challenge pole."

Another run in the same story—there are many of them—is a description of the beheading of a giant:

Here is another example of the same run from Myles Dillon's translation of a text of "The Giant of the Mighty Blows" that he recorded by dictograph from Joe Flaherty in County Galway in 1932.

These last two quotations illustrate individual variants of the same common "theme." The storyteller may vary the run a little or not, or he or she may omit it, but in individual practice it is comparatively stable, although not actually memorized as fixed. When writing comes, these set passages tend to disappear, because in written literary style variety, rather than repetition, is sought after. Yet the stories, as narratives, remain the same, though the written style of them changes. {23|24}

The storytellers continue to tell their stories as before. Even when they become literate, this is true. It is what the literate person other than the storyteller—usually a collector from either within or outside of the traditional community—does with the tale, that brings about a real change. Through such persons a written literary folktale genre is created. [16] The tales in it are often compilations of elements belonging to several traditional tales, told in written literary style. The literacy in the community and the presence of an already developed literature create a new genre, which lives its life in literary circles, parallel to the continuing oral-traditional stories in their usual setting, as long as that setting itself remains unchanged.

The case of oral-traditional poetry, specifically epic song, is somewhat different, because with it we are dealing with stories in long poems in verse, or rhythmic periods of some sort. The question then arises: What is the impact on the individual singer of the introduction of writing into the oral epic poet's community? If the singer did not personally learn to read or write, it had no direct effect at all, of course. There might, however, be indirect effects. Someone might read a song to the singer from a printed or written source. Other than possibly bringing him an epic tale which he might otherwise not have known, this reading would not trouble the waters of his oral-traditional literary world.

As printed material increases in the community and more and more people learn to read and write, and to read literature, prestige may become attached to the literate, and literary, members of the society, and consequently the unlettered may lose prestige. As a result, their cultural activities, such as singing traditional songs and telling traditional stories, may also lose prestige and eventually be lost. There would be pressures, of course, for the unlettered traditional singer to join the prestigious group of the literate. This would not necessarily mean at first that he would immediately become a person knowledgeable in written literature. If he succumbs to the pressure, nothing may happen to him or to the songs or stories, provided the society continues to foster the traditional culture as well as the newer written culture, to listen to and to sing and tell traditional tales and songs.

If the traditional singer/poet is composing in that special oral formulaic style which came into being to make rapid composition in performance possible, and which he has learned from previous generations {24|25} of singer/poets, he does not need writing to compose lines and tell stories. But in a literate, more particularly a literary, society, a singer might get the idea of writing down an epic song from his repertory in the words and manner in which he usually sang it. This would be the same as if someone else wrote it down from his dictation. If, however, a singer made changes in the way in which he wrote from the way in which he sang, then his knowledge of writing would have played a role in the composing of his text. If, for example, he uses some new, non-traditional, phrases or constructions, nevertheless still keeping mostly to the traditional diction, he would be moving in the direction of written literature. It could be argued that he is already a practitioner of written literature or that he is writing in a transitional style. Such a singer's text, therefore, could be considered legitimately as either a written literary text or a transitional one. One must, however, be cautious. Not every "new" word used necessarily constitutes a breaking of the traditional formulaic style, for some new words quite normally find their place in the traditional formulaic systems. The singer without the pen, including the beginner at one end of the scale, the highly gifted singer at the other end, and the unskilled singer in between, breaks the system from time to time, making unmetrical or inept lines, or even lapsing into prose. These are the aberrations of performance, be it before a live audience or in dictating to a scribe. The breaking of the new structure of the formulaic systems themselves is more important than are new words. Donald Fry was right in stressing the system in his definition of the formula in Old English. [17] As long as the systems continue, it does not matter whether the singer composes with or without writing. In fact, the "oral residue" expressed in the systems, themselves formed in orality, would persist in the world of literacy, in the usage of the literate traditional singer until such time as the nontraditional-minded writer with a pen in his hand should rearrange the words and traditional patterns in the basic systems. Thus would a written literature be born from an oral literature.

How can one distinguish an oral-traditional text from one of written literature? It must be said at the beginning that one must know something—the more the better—about the tradition in question to which a singer belongs as well as his own habits of composition in order to make the judgment. By that I mean that one must know what the specific characteristics of a given tradition are in order to tell whether they are {25|26} present or not in the text under consideration. One needs, also, as many texts of a singer or storyteller as possible.

Milman Parry distinguished three stylistic characteristics of oral style: (1) the presence of a large number of formulas; (2) the presence of "themes"; and (3) the presence of many cases of unperiodic enjambement. [18] During his lifetime he wrote much about formulas, something about enjambement, and very little about "themes." Enjambement is useful as a rule-of-thumb measurement if other characteristics are also present, but is not in itself decisive as a criterion of orality. It is a manifestation of the adding style. That style is a sign of oral-traditional composition, but it is easily imitated.

Formula density, the presence of a substantial number of true formulas in a text, is still a reliable criterion for oral composition under certain circumstances which need further review. Formula density should, however, be tempered by an additional investigation of the specific formulas used in a given work vis-à-vis the traditional formulas as they are known.

Let me give two examples that may help in determining whether a given poem is an oral composition, or, more specifically, was composed in the oral-traditional style. The first is the poetry of Peter II, Petrović Njegoš, Prince-Bishop of Montenegro in the first half of the nineteenth century. [19] He was brought up as an (oral) traditional singer of epic, but after his education he became a well-known written literary poet. One can trace in his early written poems his gradual departure from the style in which he was brought up. One can note phrases and patterns that were not "traditional," and thus one can document in his case the moving away from tradition, even while keeping the traditional meter. The formula count, in such a case, would in the course of time naturally reflect the emerging written literary style of the author. In addition, an {26|27} analysis of specific phrases and structures would show marked differences from his known tradition.

The other case is of an author, a poet, brought up in an educated milieu as a Franciscan monk in the eighteenth century. Andrija Kačić-Miošić became so immersed in the traditional style that he could write poems which were, to the uninitiated, indistinguishable from the real thing. In other words, he used well-known traditional formulas. He sought no new stylistic effects. Occasionally, as in expressing a date, he broke the tradition, not only with new words but with new structures. Thus his style betrayed the writer. But he could also write, still in the traditional meter, poems, which by content and even genre (e.g., an epistolary genre from the written literary tradition in which he was brought up) were clearly not oral-traditional compositions. If one were to analyze all his work together, the formula density would be fairly high. It is necessary, however, to analyze each song separately, or to segregate those which are clearly written literature from those "in the style of" oral-traditional poetry. His chief and most influential work was a history in prose and "popular" verse of the South Slavs, written in a manner which they could readily understand. [20] He could write in both oral and written styles.

The "theme" in oral-traditional epic, a repeated passage, is as characteristic of oral-traditional composition as is the formula and for the same reason, its usefulness in composition. [21] There are several important things to note about the "theme." First, it is not simply a repeated subject, such as a council, a feast, a battle, or a description of horse, hero, or heroine. It is that, but it is more than that. All those subjects occur repeatedly in written literature as well. The "theme" in oral literature is distinctive because its content is expressed in more or less the same words every time the singer or storyteller uses it. It is a repeated passage rather than a repeated subject.

Second, the degree of variation of text and of detailed content among occurrences of a "theme" in the usage of a single singer or storyteller differs considerably from individual to individual and from "theme" to "theme." In general it is clear that a "theme" which he or she uses frequently tends to be more stable in its text, as well as in its content than one used infrequently. It is also true that a short "theme" is more {27|28} stable than a long one. One singer I have known had the opening scene of his favorite song down pretty much word for word—not quite, but the text for ten or fifteen lines was fairly stable. This comparative fixity of text in a "theme" is not a mark of a written text. The singer involved, Djemail Zogić of Novi Pazar, was illiterate; he had not consciously memorized those lines. [22] He simply remembered them. Here are the pertinent lines from his dictated version of July 1934 (on the left) and his sung version of November 1934 (on the right). The lines common to both versions are printed in the center column.

"Osamnula" in the second line, wrongly repeating the verb of the preceding line, is a slip of the tongue (or of the recording pen) for "udarila." In the fifth line the only difference is in the prefix of the verb, namely, "zapevala" and "prepevala." In the last line, in addition to the difference in prefixes in "okahri" and "zakahri," there is the difference between "moga" (mine) and "našeg" (our). The main differences between the two versions are the expansions in the singing of the second version.

As further evidence that the lines were not memorized, I present the pertinent lines from Zogić's version sung for the tape-recorder in the summer of 1962, nearly twenty-eight years later. [23]

There is a popular misconception that oral literature is crude, formless, unstructured, and that without writing one cannot create intricate structures of verbal expression. A corollary to this belief is the idea that any work of literature with a complex structure must be a product of the written word, the word seen, rather than the word heard. Those intimately acquainted with an oral-traditional literature, however, are cognizant of the fact that this is a false impression, arising from a lack of experience with that type of literature.

Cope and Opland have demonstrated the stylistic and artistic excellence of Zulu praise poems and folktales and of Xhosa praise poetry, respectively. [24] Douglas Mzolo has done the same for Zulu clan praises, and Daniel Kunene has been especially painstaking in analysis and eloquent in describing the heroic poetry of the Basotho. [25]

For an illustration of larger orchestration in a long epic song by a talented South Slavic singer I offer here a brief sketch of ring-composition at the beginning of Avdo Međedović's 12,311-line song, “The Wedding of Smailagić Meho,” collected by Milman Parry in 1935 and written down by Nikola Vujnović from Avdo's dictation. [26]

The song opens with an assembly of the nobles with Hasan pasha Tiro at their head. He notices that young Meho, son of Smailaga, is unhappy, and he sends Meho's uncle, Cifrić Hasanaga, to inquire the reason for his sadness. Meho responds that he is sad because his elders {30|31} will not allow him to engage in warfare. He plans to rebel and go over to the enemy. His uncle tells him that he has been the darling of all, that they have been waiting for him to grow up so that he may assume his father's and uncle's position as leader of the Border warriors. The pasha agrees to send Meho to Buda to receive his credentials from the vizier there. The nobles sign the petition and say farewell to Meho.

This opening council scene—a typical “theme,” by the way—is a good example of ring-composition. Here is the scheme: This is a perfectly acceptable ring. Its pattern is inherent in the narrative itself, and its focus, the speech of Meho to the assembly, is significant in the story; for in it is contained the background for the whole plot. The dramatic confrontation between uncle and nephew with its centerpiece of the nephew's angry speech, which is to provide motivation for the entire poem, is framed in a setting of hierarchical social organization and a statement of heroic values.

After a brief linking theme which takes Meho and his uncle back to Smailaga's house to report what has happened in the assembly, of which the father had as yet no knowledge, the scene for the next ring begins. It extends from the report of Meho's uncle to his brother Smail to the completion of preparations for the departure of Meho and his companion Osman for Buda. The first ring is the conversation between the brothers. The second circle of the ring is the scene between mother and son. This is a scene of elaborate ritual adornment of Meho prior to his appearance before his father. The center of the ring is that appearance. Moving outward in the circle we find a ceremonial preparation of the hero's companion, Osman. This balances the ceremonial preparation of Meho himself. The outermost circle reveals Meho, Osman, and their alter egos, their horses, also ceremonially prepared. Schematized, the ring looks like this: {31|32}

There is not space to follow the intricate structure of rings for the whole poem. Suffice it to say at the moment that each scene can be analyzed in this way, as well as the whole narrative, in terms of rings or chiastic constructions, resembling Cedric Whitman's analysis of Homer's Iliad. [27] Some would doubt that oral-traditional poets would have the ability to construct their scenes, and perhaps even an entire poem of some length, in this manner. Here is proof that they not only can do but actually do just that.

There is a tendency for us in the European tradition to forget how extensive and how basic our literary heritage from the world of orality has been, and there is a corresponding tendency to believe that the world of literacy invented some of the characteristics of literature, which in reality originated in oral literature. Among them is a sense of form and structure, as I have just illustrated, and many devices, later termed “rhetorical” and attributed to the schools, actually were created in the crucible of the oral world. The world of orality gave us anaphora, the use of the same word at the beginning of each series of lines, epiphora, the use of the same word at the end of each of a series of lines, alliteration, assonance, rhyme, both internal, medial, and final, and the sense of balanced structure as typified by parallelisms in sentences and other forms of parataxis. In short, our poetics is derived from the world of orality, with some later additions and modifications introduced by the world of literacy.

Consider as an example of anaphora, alliteration, and parataxis the following passage from one of the South Slavic epics from the Milman Parry Collection collected in 1935. The setting is an assembly of leaders of the Border. All are boasting except one who keeps his head down (my translation reflects the structures of the original). [28] {32|33}

Note that after two lines of introduction, there are six lines beginning with “ko” (who), followed by a summary line. The six lines are paratactic, in addition to their alliteration and anaphora. Moreover, the next five lines, this time beginning with “neko” (someone), are in parataxis with the preceding group of five lines, and they too end in a “coda.” They repeat in substance the previous group, but with a slightly different construction, both grammatically and alliteratively, the “neko” appearing not only at the beginning of the line, but also, in three lines, after the caesura.

The play of "ko" and "neko," which is joined by the neuter "nešto" (something, or somewhat) continues in the scene in the negative. The text goes on then with a rhyming couplet: