In light of the arrogance of many textual critics in subsequent centuries or other traditions, we can see that the Hellenistic scholars, for all their petty rivalry, worked with admirable tact and skepticism. The Homeric poems had a place in their esteem that approached the sanctity of scripture in sacred traditions. Rather than make copies without the verses that one or more scholars considered redundant or unacceptable, they developed a system of marginal signs meant to alert the reader to the purportedly dubious status of these verses. For this we can be thankful, for while we can reconstruct their principles and to some extent their tastes, the text itself seems to have passed through this period without being diminished or recast. The Hellenistic scholars saw that many words and references in the Homeric poems were mysterious because of changes in the language and culture, but they had neither the historical linguistics nor the archeology to enable them to begin to reconstruct Homer’s world. At least they recognized that such a reconstruction would be the first step to an understanding of his poetry.
The Chronology, below, lists some significant moments in the history of Homeric scholarship. My interest here is to visit crucial stages in the Homeric question as it evolved, because our current hypotheses about the origins of the Homeric poems are best understood, again archeologically, as developments in an ongoing process. The Greeks—not only Classical and Hellenistic but their Byzantine descendants—focused on questions of Homer’s provenance, the authority of one reading over another, and the need to explicate his meaning. In some instances, the explication led rather far afield, particularly when an exegete adhering to one philosophical or religious sect or another, confident that his doctrine was true and that Homer was the supreme exponent of truth, used allegory to harmonize Homer with his truth. Thus for different interpreters the gods became physical elements such as air and water, gods and heroes became ethical virtues, and the nymphs human souls.
But throughout this period, and even after Homer was reintroduced into the literary culture of western Europe, he was less studied than evoked—as the original Western poet (along with other mere names such as Linus, Musaeus, and Orpheus), as the fount of Classical mythology, and as the first epic poet. As I have described, only in the fifteenth and sixteenth centuries did significant numbers of western European scholars have the opportunity to learn Greek and, with the advent of printing in the second half of the fifteenth century, the texts with which to work. Homer also began to circulate in translations, not only in Latin (at the time the universal language not only of the Church but of scholarship and international affairs) but in modern vernaculars, in English first in George Chapman’s rhymed Iliad and Odyssey (the pair were in print by 1616), then in Alexander Pope’s versions in his trademark heroic couplets (The Iliad, 1715–20; The Odyssey, 1725–26).
Pope’s Homer, great as it was, was more Pope than Homer, as contemporary critics were swift to point out. His historical pretensions as well as those of his critics are themselves significant, for they mark incipient scholarly concern with the historical Homer. Starting in the eighteenth and early nineteenth centuries, when, on the one hand, the study of history as a discipline was making rapid advances and, on the other, folktales and the heroic medieval literatures of the various European nations began to be studied and appreciated, substantial new light was shed on Homer. In this, scholarship and fashion went hand in hand. When, in the seventeenth century, the abbé d’Aubignac called Homer illiterate, he intended it as an insult; in the second half of the eighteenth century, which saw the popularity of the pseudomedieval songs of Ossian, the “illiterate bard” was an attractive figure.6 Working on a more serious level, Vico, Woods, Wolf, Lachmann, and others, with increasingly acute historical sense, led the way in reconceiving the cultural context in which the Homeric poems were created. All were right in this respect: that reconstructing that context was the key to understanding the origins of the poems, and thus solving the Homeric question.
By the end of the nineteenth century, it was not uncommon, at least among literary historians writing in English or German, to assume that the social and political conditions of the culture which produced the Homeric poems were comparable to those observed in “primitive cultures” (as these were then termed) around the world. Some also drew attention to the performance of oral poetry in such cultures and suggested that the Homeric poems were likewise the oral productions of a bard. Scholars were, however, uncertain how to connect this largely ethnographic insight to the analysis of the Homeric texts. At the opening of this century, the remarkable archeological discoveries in mainland Greece and the Troad (the area around Troy in the northwest corner of what is now Turkey) seemed to offer a more fruitful field for Homerists: surely here were the real objects, the material culture which Homer’s heroes would have known and which Homer was describing.
Soon, however, an approach based on both ethnographic parallels and a detailed analysis of Homer’s poetic language proved more fundamental to Homeric scholarship than the archeological discoveries of Heinrich Schliemann and his successors. In the 1920s a young American doctoral candidate, Milman Parry, wrote and defended a thesis at the Sorbonne.7 What emerged from this and subsequent studies by Parry and, after his death, by his assistant, Albert Lord, and others, we call the theory of oral-formulaic composition. Both Parry and Lord combined close analysis of the texts of the Homeric poems with on-site study of contemporary oral poets still performing in unbroken oral traditions. Of particular interest was the rich and then still productive bardic culture of the South Slavs.
Singers in this and other well-documented traditions learn their craft by hearing and doing. By manipulation of a repertory of formulae (stock phrases), the singer does not repeat but rather re-creates a certain tale afresh every time he sings it. Armed with the theory of oral-formulaic composition, Parry made sense of Homer’s oft-criticized repetitions and fixed epithets. But he did more. By careful analysis of the occurrence of various noun-epithet combinations (e.g., “swift-footed Akhilleus,” “wine-dark sea,” “rosy-fingered dawn”) at different metrical positions in Homer’s dactylic-hexameter line (a supple but strictly ruled system of anywhere from twelve long syllables—more normally eleven long and two short—to six long and eleven short syllables), Parry was able to show not only the extent of the formulaic system but its economy. In other words, the practice of oral poets over the centuries had undergone a sort of evolutionary “natural selection,” through which the bards’ artificial, i.e., conventional, language was so pared down that only rarely did more than one noun-epithet phrase for the same person or thing survive for any particular metrical slot in the line.8
This is highly technical, as are the many refinements of the idea that scholars have introduced over the years. Nonetheless, a grasp of even a simplified version of the thesis leads to a realization of its far-reaching implications. Does the theory of oral-formulaic composition undermine Homer’s poetic autonomy? That became the central question, for what above all provoked violent resistance to the theory was not only that it appeared to remove originality from the process of creating two of the greatest and certainly the first long poems in the Western tradition—“the bard uses inherited material”—but that it seemed to deny a place to creativity itself. If the singer were working with a strict metrical schema and his cache of formulaic phrases had for the sake of efficiency been reduced to one formula per slot, it seems that composition would have been well-nigh automatic.
But this is an extreme and untenable response. We can, I think, get a reasonable feel for both the constraints and the freedoms available to the singer of formulaic poetry by considering any number of familiar systems, for instance, music. Whether the musical idiom be Gregorian chant, Renaissance polyphony, late-eighteenth-century classical, ragtime, disco, or rap, given a certain bit of thematic material, the living practitioners in each style would think of harmonizing or developing it in only a limited number of ways. Later students of each style might argue that the system allowed little
or no choice. In contrast, practitioners and contemporary listeners, for both of whom the constraints are given and therefore virtually unconscious, focus solely on the range, however narrow, of possible variation. As with so many things, constraints permit freedom.
Indeed, if we analyze it, language itself is largely a formulaic system, albeit with a more generous range of possible variation. As competent speakers of the language, we know the rules of grammar and the range of each word’s meaning. We know that one will have failed to construct a comprehensible sentence if the words occur in an incoherent order or if the words themselves cannot be interpreted to make sense. Nor do we often create many of our own words: we take them from a preexisting vocabulary. Nevertheless, we feel entirely unconstrained in manipulating the language. Just as we draw on a vocabulary, the oral poet draws on a “formulary,” a repertory of formulae. And just as we sense the freedom rather than the constraint of the system—we do not spend much thought on what it is impossible or nonsensical to say—so the oral poet feels he has enormous freedom to manipulate his poetic diction.
The system works whether the constraints are largely unperceived—as with language, one’s native language at least—or they are observed self-consciously. As a living example of self-conscious observation, think of the oral-formulaic system of jokes. When we hear a joke, we don’t memorize it word for word. It is usually enough to retain the general context and the punch line. When we want to retell it, we re-create the structure and fill it in with what seem to be our own words but are in reality stock phrases of “joke-ese”: an X, a Y, and a Z—it’s always three—are at the Pearly Gates—a phrase hardly ever used today in any other context—requesting St. Peter to let them into heaven. Or an A, a B, and a C are in a rowboat….
Formulaic aspects, as Parry suggested and as other scholars have worked out in considerably greater detail, are not limited to the phrase, half line, or series of half lines. Certain “typical” scenes, such as leave-taking and offering sacrifices, have their formulaic rhythms and patterns. These two are of course actions which would have had their standard protocol in any event. However, many other actions, such as arming, which might vary from person to person, are presented according to very strict standards. Many jokes, as the examples in the previous paragraph suggest, have traditional themes and forms as well as a recognizably formulaic diction. We can think of whole families of contemporary jokes, for example, “How many X’s does it take to change a light bulb?”
Folktales provide another universe in which the very typicality of patterns makes retelling simple. Once you hear the story of a prince who must accomplish three tasks in order to win the hand of a princess, you can retell it with ease. You can retell it with any set of tasks, so long as there are three. You can tell it in a version in which it is a cruel stepmother rather than a fierce would-be father-in-law who sets the tasks (nor do you need to have experienced “stepmotherly cruelty” to make this substitution: you “know” from other tales that stepmothers can be cruel). And you can tell it in a version in which the princess herself sets the tasks, or asks riddles instead. The possibilities are endless, yet this is in every case the manipulation of traditional story elements. While the teller may feel he or she has complete freedom, the story that emerges will sound as if it were a thousand years old and had never been told in any other way.
The oral poet, then, has patterns at more than one level that guide him or her in the spontaneous re-creation of a story. It may well be that all popular culture is formulaic and generic. From our own experience we know just about all the possible story lines a western or horror movie can take. It is even easier with half-hour TV shows. (The half-hour slot, with commercial breaks, may be for screenwriters what the dactylic hexameter with mandatory breaks at half line—caesurae—was for Homer.) The genre and even the characters are known before the program begins, because they continue from one program to another within the series. So ancient audiences might have known what was likely to come if a tale from the voyage of the Argo, or of Heraklês’, or Theseus’ exploits, had been announced. As soon as the issue is presented, most intelligent viewers know exactly what is going to happen, and how and when the problem will be resolved. True aficionados can look at their watches and know whether, according to the protocol of the program, four minutes before the last commercial break is too late to introduce a new complication or plot reversal. None of this diminishes our enjoyment of our chosen entertainment; indeed, the particular proportion of predictability and variability, of familiarity and surprise, is the essence of such entertainment. I present this not as Homer’s equivalent—there are many reasons why the Homeric poems are and were for their day much, much more than television sitcoms are for ours—but as an example of contemporary experience that may help us understand the workings of a popular and formulaic system.
Most recently, it has been argued that a certain inevitability in the entire story set in as soon as a bard determined to sing of the wrath of Akhilleus or the homecoming of Odysseus. If the shapes of story patterns were set at this level, this would of course make it considerably easier for bards to sing a long story, not relying on memory but simply telling the story. While less familiar to us in narrative poetry, such patterned improvising is well-attested in groups of performers, for example, the troupes of commedia dell’arte comedians of the sixteenth and seventeenth centuries or, closer to our own time, music-hall vaudevillians.
These analogies are instructive. However, a peculiar and rather obvious problem presents itself when we analyze the Homeric poems as the products of an oral-formulaic tradition. They were written down. The essence of the oral tradition is that the participants, singers and listeners, believe in its inviolability, its unchanging nature, so long as it is oral. In the absence of recording equipment, it is not possible to compare two oral accounts by the same singer, and thus not possible to prove the fact of variation. Modern studies of singers in this century, conducted by scholars with tape recorders, have proved that there is a wide variation in subsequent as well as distant performances of the “same” story, just the sorts of differences the entire system of oral composition might predict. But the singers always believe that they are singing the story unchanged.
Two centuries ago F. A. Wolf, in his epoch-making Introduction to Homer (Prolegomena ad Homerum), tried to reconstruct the state of writing that could have been known to Homer and his contemporaries. And the introduction, or rather the reintroduction, of writing into Greece is also at the heart of the “Homeric problem.” Reintroduction because, as archeologists in this century have made it known, in the later Minoan and Mycenaean periods (roughly 1500–1200 B.C.E.), Greek was written widely around the Aegean, both on Crete and on the mainland, in the script known today as Linear B. Linear B is a syllabary, that is, it represents each syllable by a different character. While a considerable advance on earlier systems of pictographic writing—which represent entire words or concepts by ideograms or hieroglyphs—a syllabary is less efficient than an alphabet, which assigns to each sound, in no matter which combination, a single sign. Nonetheless, despite the comparative inefficiency, Linear B functioned well enough for the purposes of the Minoan and Mycenaean overlords who employed it, or trained scribes to employ it. Granted, the range of texts for which the script was employed was limited—inventories, public inscriptions, commercial documents; there are no literary texts. The very mode and material of writing would make the preservation of literary texts of any length difficult, although despite comparable difficulties a rich and highly evolved literature is extant in contemporary Sumerian, Babylonian, and Akkadian, also written in syllabaries on clay tablets.
Literacy was the province of a trained elite, a caste of scribes, as in ancient Egypt, China, and many other cultures. There was not a broad class of educated persons which could form a “reading public.” With the collapse of Minoan-Mycenaean culture, the needs for which the script had been used disappeared, and with them writing itself. The fall of this civi
lization is often treated as a fascinating mystery, something akin to (and at times linked to) the disappearance of Atlantis, but as in the case of the much later “Fall of Rome,” the transformation was likely more gradual. There were certainly wars, there may have been massive invasions and movements of peoples, there may even have been devastating earthquakes and volcanic explosions. But a healthy culture can rebuild and recover from any of these. What clearly happened, over several centuries and under the pressure of these occurrences, is that the old patterns of rulers and ruled, creditors and debtors, importers and exporters, changed.
Nor was the Aegean basin immune from changes in adjacent areas. Crete became less important relative to Egypt and to the new forces setting out from the eastern shore of the Mediterranean. There can hardly be more eloquent testimony to the importance of Phoenician culture and trade than the Greeks’ adoption of an alphabet clearly derived from the Phoenician characters. The first examples of writing in what we now call the Greek alphabet date from the last half of the eighth century B.C.E.
It is of course a long way from scribbling a few words to writing down the whole of The Iliad or The Odyssey. Moreover, the idea of doing so would have to occur to someone, and it would not be likely to occur to an oral poet at the height of an oral tradition. This is the gap over which bridges of various sorts have tentatively been built. According to one argument, in the late eighth or early seventh century B.C.E. an artist, perhaps called Homer (unless that be the name, then already traditional, of some outstanding singer at an earlier stage of the tradition), sang versions of the story of Akhilleus and the homecoming of Odysseus so remarkable, and recognized as such, that at once the practice of oral tradition changed. Instead of learning to re-create epics afresh at each new sitting, the younger generation, recognizing the monumentality of Homer’s versions, memorized the poems. Figurative “sons of Homer” (Homeridae) and then the rhapsodes described earlier continued to recite the Homeric poems by rote at least into the fifth century B.C.E., by which time writing had been so well established that scribes were able to take down the poems, the texts of which are the ancestors of ours today.
A Guide to the Odyssey: A Commentary on the English Translation of Robert Fitzgerald Page 3