Spontaneity and Tradition: A Study in the Oral Art of Homer [Reprint 2020 ed.] 9780520320758

Citation preview

Spontaneity and 'Tradition

Red-Figured Calyx Krater by Euxitheos and Euphronios. Dead Sarpedon carried by Thanatos and Hypnos, framed by named Trojan warriors. PLATE 1.

2. Red-Figured Skyphos by Hieron and Makron. Abduction and return of Helen, with Aphrodite handling her kredemnon at both points. PLATE

Spontaneity and Tradition QA Study in the Oral oArt of Homer

M I C H A E L N. N A G L E R

U N I V E R S I T Y OF C A L I F O R N I A P R E S S

Berkeley

Los Angeles

London

University of California Press Berkeley and Los Angeles, California University of California Press, Ltd. London, England Copyright © 1974, by The Regents of the University of California ISBN: 0-520-02245-9 Library of Congress Catalog Card Number : 72-75520 Printed in the United States of America

To my ¿Mother and father

Contents Preface

*

*

*

*

*

*

*

*

*

*

ix

List of Abbreviations

*

*

*

*

*

*

*

xi

Introduction

*

*

*

*

*

*

*

xvii

*

*

*

1

2. The Traditional Phrase (II): Meaning and Significance * * * * * * *

*

*

27

3. The Motif

*

*

1. The Traditional Phrase (I) : Theory of Production * * * * * *

*

*

*

*

*

*

*

64

4. The Motif Sequence *

*

*

*

*

*

*

112

5. The "Eternal Return" in the Plot Structure of The Iliad * * * * * * * *

*

131

6. AYTAP AXIAAEYZ

*

*

*

*

*

*

*

*

167

*

*

*

*

*

*

199

Appendix A : Associative Patterns in Various Oral Literatures * * * * * * * *

*

210

Appendix B : Annunciation Phrases from The Iliad * * * * * *

*

*

*

214

Select Bibliography *

*

*

*

*

*

*

*

217

Index Locorum*

*

*

*

*

*

*

*

*

223

General Index *

*

*

*

*

*

*

*

*

233

Conclusion

*

*

*

It was more than ten years ago, at a party in San Francisco, that Professor Elroy L. Bundy succeeded in impressing upon me the importance for Western literature of the consolation speech delivered to Priam by Achilles in the twenty-fourth book of the Iliad. He had brought home to me as well the importance, for this question and others, of Milman Parry's work on the nature of oral poetry, and so for me was begun the line of inquiry that has lead to this book. May he smile graciously upon the fruits! In the meantime, the need for a book of its kind has become so keenly felt that I am glad to publish the present study, though sensible of its inadequacies and its immaturity relative to the greatness of the topic. The time is at hand, I think, to combine whatever has been gained in the relatively new field of oral poetry, both from anthropological fieldwork and philological inspection of the textual records, with a contemporary effort to appreciate the poetry of Homer, Beowulf, the great Sanskrit epics, and whatever else comes into this category. Chapters 1 and 2 of the present study are a revision and expansion of my article, "Towards a Generative View of the Homeric Formula," TAPA 98 (1967) 269-311, which I include here for the following reasons: There is a natural continuity between that work on the phrase and the new chapters dealing with higher levels of organization such that both are well brought under one set of covers. Furthermore, the earlier work has gained much since its original publication from my contact with the sphota theory of the Indian grammarians, to which my colleague Professor Barend A. Van Nooten guided me, and from written and oral discussions with fellow Homerists around the world, including Professors Albrecht Dihle and Wallace Macleod. My particular thanks in this connection go to Professor Robert Dyer and the Indiana University Summer Institute for Computer Applications in the Humanities, under whose auspices J. B. Hainsworth, Joseph Russo, Mark Edwards, and myself were brought under one roof (along with David Packard, Jr. and other computeroriented classicists). The translations of Homeric and other material offered below are not meant to be accurate in the ordinary sense of the word: They are frankly slanted toward bringing out the poetic interpretations that I am trying to ix

derive from the texts, and should not in themselves be treated as evidence. They are there because I hope this book may be useful for the growing number of students of oral poetry in many traditions outside of Ancient Greek, as their work and enthusiasm has been for me. Apart from chapter 1 and the notes, where the going is unavoidably technical, all the Greek has therefore been translated. Wherever possible, I have tried to write with everyone, from my fellow homérisants to the general littérateur, in mind ; I hope that the stylistic inconsistencies occasioned by this attempt, especially in the earlier chapters, will be compensated for by the broader usefulness of the book. A comparative literature student for example, if he doesn't know Greek, may want to start with chapter 2. Virtually all of my immediate colleagues at the University of California in the Departments both of Classics and Comparative Literature have helped me at some point. In addition to Professors M. B. Emeneau and B. A. Van Nooten, therefore, to whom I owe my knowledge of Sanskrit, and Professors Anne Kilmer and Wolfgang Heimpel, who have tried to relieve my utter ignorance of things Mesopotamian, I must thank Joseph J. Duggan for assisting me with Roland, Crawford H. Greenewalt for much help with Greek vases, Charles Murgia for the same with Serviana, W. G. Rabinowitz and W. R. Johnson for their encouragement at, respectively, the beginning and the end of this work, Albert Henrichs and J. K. Anderson for catching a variety of errors, Joseph Fontenrose who is still my unfailing oracle for myth, and, last but not least, Phillip Damon, Alain Renoir, and Thomas Rosenmeyer. I suppose that only those who have known and worked with these last will really be able to appreciate how much help they have given me over the past several years. In surveying all these resources of knowledge and understanding I feel worthy of taking credit for nothing in this book except its insufficiencies. It gives me pleasure also to thank my students at San Francisco State University and at the University of California, from whom I am still learning much, among them Mrs. Carol Flinders, who read a (very) rough manuscript, and Diana Thornton, who typed the final draft with exceeding care (correcting my Sanskrit the while). Plate 1 is reproduced through the courtesy of the Metropolitan Museum of Art, Bequest of Joseph H. Durkee, Gift of Darius Ogden Mills, and Gift of Mr. and Mrs. Buxton Love, by Exchange, 1972. Plate 2 is reproduced through the courtesy of F. Bruckmann Verlag, Munich. Finally, I thank my wife, who has been a helpmeet and friend since before this volume was conceived. x

List of Abbreviations Classical authors and other works are usually abbreviated according to

LSJ9.

Ameis-Hentze Karl Friedrich Ameis and C. Hentze, Homers Odyssee and Anhang zu Homers Odyssee (Leipzig 1868-71 ; reprinted 1894-1900)

ANET James B. Pritchard, ed., Ancient Near Eastern Texts Relating to the Old Testament (Princeton 1950)

Arend Walter Arend, Die typischen Scenen bei Homer (Berlin 1933)

Atra-hasïs Wilfred G. Lambert and A. R. Millard, Atra-jjasïs: The Babylonian Story of the Flood (Oxford 1969)

Bremond Claude Bremond, "La Logique des possibles narratifs," Communications 8 (1966) 60-76

Chantraine, Dictionnaire étymologique Pierre Chantraine, Dictionnaire étymologique de la langue grecque (Paris 1968- )

, Grammaire homérique Grammaire homérique vol. II, syntax (Paris 1953)

Dihle Albrecht Dihle, Homer-Probleme (Opladen 1970)

Ebeling H. Ebeling, Lexicon Homericum 2 vols. (Leipzig 1885)

xi

SPONTANEITY

AND

TRADITION

Fenik Bernard Fenik, Typical Battle Scenes in the Iliad: Hermes Einzelschriften Heft 21 (Wiesbaden 1968)

Flexibility J. B. Hainsworth, The Flexibility of the Homeric Formula (Oxford 1968)

Gesten und Gebärden Gerhard Neumann, Gesten und Gebärden in der griechischen Kunst (Berlin 1965)

Gilgamesh Alexander Heidel, The Gilgamesh Epic and Old Testament Parallels (2nd ed. Chicago 1949; reprinted 1963)

Gunn, Narrative Inconsistency David M. Gunn, "Narrative Inconsistency and the Oral Dictated Text in the Homeric Epic," AJP 91 (1970) 192-203

, Singer and Tradition The Singer and his Tradition: Aspects of Thematic Composition in Homer and Southslavic Heroic Song, unpubl. M.A., Melbourne 1966

HHT Cedric H. Whitman, Homer and the Heroic Tradition (Cambridge, Mass. 1958; reprinted New York 1965)

Hoekstra A Hoekstra, Homeric Modifications of Formulaic Prototypes (Amsterdam 1965). Verh. der kon. Nederl. Akad. van Wetenschappen, afd. Letterkunde, N.R., Deel 71, no. 1

Homer J. B. Hainsworth, Homer. Greece and Rome, New Surveys in the Classics Monograph no. 3 (Oxford 1969)

Homeric Craftsmanship Mark W. Edwards, "Some Features of Homeric Craftsmanship," TAP A 97 (1966) 115-179

HWHI Adam Parry, "Have We Homer's IliadV YCS 20 (1966) 175-216

xii

LIST

OF

ABBREVIATIONS

Kakridis Johannes Th. Kakridis, Homeric Researches (Lund 1949)

Language of Hesiod G. Patrick Edwards, The Language of Hesiod in its Traditional Context (Oxforc 1971)

Lanman Charles Rockwell Lanman, A Sanskrit Reader (Boston 1884; reprinted 1963)

Lesky Albin Lesky, "Mündlichkeit und Schriftlichkeit im homerischen Epos," in Festschrift für Dietrich Kralik (Horn, N.-O. 1954) 1-9

LSJ9 Henry George Lidell and Robert Scott, A Greek-English Lexicon; new ed. rev. by Henry Stuart Jones, et. al. (Oxford 1940)

Meister Karl Meister, Die homerische Kunstsprache (Leipzig 1921; reprinted, Stuttgart 1966)

MHV Adam Parry, ed., The Making of Homeric Verse: The Collected Papers of Milman Parry (Oxford 1971)

Monro, Homeric Grammar D. B. Monro, ed., A Grammar of the Homeric Dialect (Oxford 1882; 2nd ed. 1891)

, Iliad D. B. Monro, Homer Iliad 2 vols. (Oxford 1884-1888; rev. 4th ed. reprinted 1960)

Mycenae to Homer T. B. L. Webster, From Mycenae to Homer (London 1958; 2nd ed. London and New York 1964)

Onians Richard Broxton Onians, The Origins of European Thought (Cambridge 1951; 2nd ed. 1954)

xiii

SPONTANEITY

AND

TRADITION

Parry-Lord Milman Parry and Albert Bates Lord, Serbocroatian Heroic Songs (Cambridge, Mass. and Belgrade 1954)

Propp Vladimir Propp, Morphology of the Folktale: International Journal of American Linguistics 24:4 (1958 ; 2nd rev. ed. Austin, Texas 1968)

Python Joseph Fontenrose, Python (Berkeley and Los Angeles 1959)

Raja Kunjunni Raja, K., Indian Theories of Meaning (Madras 1963)

Reinhardt Karl Reinhardt, Die Ilias und ihr Dichter (ed. by Uvo Hölscher, Göttingen 1961)

Repetitions George M. Calhoun, "Homeric Repetitions," UCPCP 12 (1933) 1-25

Ruijgh C. J. Ruijgh, L'élément achéen dans la langue épique (Assen 1957)

Select Papyri D. L. Page, Select Papyri vol. 3, Literary Papyri, Poetry (Harvard 1941, rev. 1950)

ST Albert B. Lord, The Singer of Tales (Cambridge, Mass. 1960, reprinted 1964)

Stanford W. B. Stanford, The Odyssey of Homer (London 1947-1948 ; 2nd ed. New York and London 1958-1959)

Style and Meaning M. B. Emeneau, "Style and Meaning in an Oral Literature," Language 42 (1966) 323-345

Taylor Charles H. Taylor, Jr., Essays on the Odyssey (Bloomington and London 1963, reprinted 1967)

Usener Hermann Usener, Götternamen (Bonn 1896; 3rd ed. Frankfurt 1948)

xiv

LIST

OF

ABBREVIATIONS

Van Groningen, Proems B. A. Van Groningen, "The Proems of the Iliad and the Odyssey," Med. der kon. Nederl. Akad. van Wetensch., afd. Letterkunde: N.R., Deel 9, no. 8 (1946) 279-294

Van Leeuwen and Da Costa J. Van Leeuwen and M. B. Mendes da Costa, edd., Homeri carmina (Leiden 1906-1908)

Wace and Stubbings Alan J. B. Wace and Frank H. Stubbings, A Companion to Homer (London 1962)

Wege und Formen Hermann Fränkel, Wege und Formen frühgriechischen Denkens (Munich 1955, rev. ed. 1960)

West M. L. West, ed., Hesiod: Theogony (Oxford 1966)

Willcock M. M. Willcock, "Mythological Paradeigma in the Iliad," CQ 58, N.S. 14 (1964) 141-154

PERIODICALS

AJP American Journal of Philology

BICS Bulletin of the Institute of Classical Studies, London

CPh Classical Philology

CJ Classical Journal

CQ Classical Quarterly

cw Classical World

xv

SPONTANEITY

AND

TRADITION

GRBS Greek, Roman, and Byzantine Studies

HSCP Harvard Studies in Classical Philology

JAF Journal of American Folklore

JAOS Journal of the American Oriental Society

JHS Journal of Hellenic Studies

REG Revue des Études grecques

TAPA Transactions of the American Philological Society

TPS Transactions of the Philological Society

UCPCP University of California Publications in Classical Philology

UCPCS University of California Publications: Classical Studies

YCS Yale Classical Studies, especially YCS 20 (1966)

xvi

Introduction Every work of scholarship, indeed, every intellectual endeavor is limited by its prior assumptions. That limitation cannot be quite overcome, but it can be kept to a minimum when one's prior assumptions and prejudices are confronted as honestly as possible. The present study, too, has been conditioned by the position taken as its point of departure, which is as follows: Today's texts of both the Iliad and the Odyssey entered the stream of written transmission as two great oral dictated transcriptions originally taken down at two special performances given (perhaps for that purpose) by Homer, a traditional oral poet of the Ionian Greek territories, very likely the best of his time. Further, the remainder of the "Homeric" corpus and that of Hesiod are from the same singing tradition, albeit with certain differences of region, time, and especially of genre; they may well not be oral dictated texts, which would partially account for their inferiority of length and artistic quality relative to the great epics—the main difference, of course, being that they were not composed by Homer—but I believe they were also orally composed and then reduced to writing in some other way. Albert Lord's thesis that Homer was an oral poet who had dictated the Iliad and Odyssey directly to a scribe under conditions resembling but not identical to a normal composition-in-performance originally met with offhand disregard in the higher circles of Homeric criticism but is now gaining in popularity. 1 It seems to me to be the simplest explanation for the capturing of great oral epics in writing on the basis of the evidence available today, including that to be presented in the present volume. The 1 Cf. Lord, "Homer's Originality: Oral Dictated Texts," TAP A 84 (1953) 124-134; and now Gunn, Narrative Inconsistency, Tatiana Fotitch, "The Chanson de Geste in the Light of Recent Investigations of Balkan Epic Poetry," in Linguistic and Literary Studies in Honor of Helmut A. Hatzfeld (Washington, D.C. 1964) 150-155, 159. For my purposes, it actually matters little whether Homer dictated the two performances to be written down, or otherwise taught them to be preserved verbatim in oral tradition (as G. L. Huxley suggests in Greek Epic Poetry [London 1969] 194f); in either case one has an unusual but not a premeditated mode of composition. It does matter whether he had had any practice at such long compositions, which I consider very likely on internal and external grounds, respectively, the tight structure of both poems (within the style of oral tectonics), and the available setting of the Pan-Ionian festival.

xvii

SPONTANEITY

AND

TRADITION

following personal observation has also influenced my decision: When I was enabled by the National Endowment for the Humanities to collect some songs in Crete during the summer of 1966 my attitude to fieldwork was that I should make as little disturbance as possible on the environment I wanted to study, going even to the lengths of recording songs without people's knowledge. At the Research Center of the Anthropological Institute of the University of Athens, however, Professor Spyridakis and his assistant, Mr. Aikatherinides, showed me that time and again the best texts—meaning the most complete and coherent texts, those into which the singers put the best of their attention and enthusiasm—were collected when the singers had been acquainted with the marvels of modern recording apparatus and were fully aware that they were performing especially for it. After the initial shock of hearing their own voices had worn off (which required surprisingly little time) they took to the idea of singing for the recorder, and for posterity, with spirit. Furthermore, as Lord had already discovered (see note 1) South-Slavic singers needed only a little further habituation to adapt their normal mode of rapid composition to the slow pace imposed by dictation. Indeed, they quickly learned to take advantage of the new situation to produce longer and better performances. Oral poets combine the advantages of being performing artists by profession while remaining enviably unselfconscious by nature: they are at their best before an audience, even though their best always means their most spontaneous (see Cecil Sharp [chapter 2, note 48] xxvii). I regard the Iliad and the Odyssey, accordingly, not as typical of the average performance by the average singer, but as typical of what a great singer could do with his tradition before a highly appreciative audience. The transmission of this recorded text remains, no less than its genesis, a mystery. For the purposes of the present study. I shall make the operating assumption of most scholars today, that some kind of standard texts came through the Peisistratean period at Athens, through the Ptolemaic period at Alexandria, down to Van Leeuwen, Allen, and other modern critics, still bearing a reasonably good record of what Homer sang.2 This assumption may not be as wishful as it seems: although the 2 An excellent survey of the textual transmission as now known is Davison's contribution to Wace and Stubbings (pp. 215-266). Papyrus finds in the intervening time have not altered the picture substantially, although Dihle presents a reconstruction which differs in some particulars in Homer-Probleme (esp. Chapter IV).

xviii

INTRODUCTION

path that leads back from the Renaissance to Homer is even longer and more obscure than it is to the Classical period, internal and external evidence bespeaks an unbroken admiration of Homer as a great teacher, and the authorship of both epics as an integrity so powerful that it simply prevailed over the political interestedness of the Athenians, the creative enthusiasm of late classical school teachers, and all the other vicissitudes of this unstable life. There is this paradox in oral tradition, that, despite its great fluidity as a mode of composition, when a poetic creation is fastened upon as holy writ, like the Vedas in India or like the Homeric epics in pre-Christian antiquity, oral tradition can be a far more accurate medium of transmission than writing, probably because of the mnemonic effect of the sound medium which is an important part of the singer's whole creation. There must have been many "sons of Homer," both professional reciters and individual admirers of his, who may well have exercised a controlling influence over the idiosyncrasies of text compilers and copyists at least until Hellenistic times. Regarding the "Homeric Question," then, as restructured by oral theory, I confess myself, with Reinhardt, Lesky, arid others, an unregenerate unitarian (see my review of Dihle in CW65 [1971] 131f). When one considers both the artistic unity of the poems and the incertitude attending ancient and modern analyses of them, to attribute the bulk of both epics to Homer seems, again, the simplest working hypothesis, especially for the largely synchronic approach I shall be taking in the following chapters. For me, the long tradition of discomfort with the ancient attribution of the Iliad and the Odyssey to Homer does not shift the burden of proof to those inclined to accept that ancient testimony. The poems as they exist today are so well composed and have so much to teach that even where one senses anomalies in the texts one should search long and deep for the author's purpose before resorting to theories of interpolation or—as has become fashionable now—the awkward fusion of rigid traditional entities. In adopting an attitude of credo ut intellegam toward Homer I have never yet been disappointed. The special value of Homer lies mostly in his unrivaled poetic genius, but partly also in the fact that he was a true oral poet. This makes him a uniquely important figure for history, linguistics, comparative literature, and for anyone interested in creativity ; and as Verdenius has said, "Les greques ont toujours pensé plus oralement que nous," so that Homer's xix

SPONTANEITY

AND

TRADITION

habits of thought are for this reason, too, particularly important for every Hellenist.3 It is no longer necessary to describe how Milman Parry brought about today's understanding of "orality" in the field of classics and elsewhere, thanks to the great work rendered as a tribute to him by his son Adam, now also taken by a premature end. As a footnote to Adam Parry's work, however (especially The Making of Homeric Verse, hereafter cited as MHV, and its thorough Introduction), I will attempt to define the implications of Verdenius's statement a little more closely before beginning to formulate my answer to the latter's call for more attention to this important subject. Oral composition is not different from written composition in every way. Professor Lord's conclusion that there is no such thing as a transitional text (ST 128-132) was necessary at the outset and has remained an extremely useful precaution because of a natural tendency to assimilate the unfamiliar "echoes" in oral poetry into superficially similar features of works composed in writing. This tendency has caused us to judge oral poetry by the wrong standards and to overlook precisely those characteristics of the art which would be most revealing: "Were we to train our ears to catch these echoes, we might cease to apply the clichés of another criticism to oral poetry, and thereby become aware of its own riches" (ST 65). And yet, as is implied by Verdenius's comparative (plus oralement), there can be varying degrees of spontaneity and of other characteristics of orality, both in thought in general and literary style in particular. This is perhaps especially true in Greek, where there was a strong survival of everything Homeric among the later poets. Professor Lord's stricture, therefore, applies historically, in the sense that when writing is introduced into an oral tradition any accommodation of the two methods will prove highly unstable and the use of writing to compose will quickly supplant the oral habit; it does not necessarily mean that the resultant literary works will immediately have utterly different characteristics from their oral predecessors. The density of "formulaic expressions," by the criteria of Parry and Lord, is markedly greater in Homer than it is in Apollonius or Quintus, but there are stylistically transitional hexameter works in Greek, and the distinction between oral and written in other 3 Cf. "L'Association des idées comme principe de composition dans Homère, Hésiode, Théognis," REG 73 (1960) 346, also Mabel L. Lang, "Homer and Oral Techniques," Hesperia 38 (1969) 159-168. Note Plato's praise of spontaneous speech at Rep. 498 D-E.

XX

INTRODUCTION

traditions where this aspect of poetic diction has been studied is also anything but clear.4 For Greek, and perhaps for other literary traditions, one should really speak of not two but three "different literatures" in chronological succession, taking into account both the mode of composition and of appreciation: (a) oral-aural, primarily the great narrative epics, (b) written-aural, choral lyric and the drama, and (c) written-read, the prose genres after Herodotus and all Hellenistic literature.5 When all these provisos are taken into account, however, the most striking division still remains that between Homer and his fellow oral poets on the one hand, and almost all other literature on the other. And the most salient characteristic of his kind of poetry remains, not the absence of writing nor the statistical aspects of recurrence, nor any easily definable feature, perhaps, but the general fact that it is—as I should like to propose for inclusion into any definition of oral ^OQtry-spontaneoustraditional art. These terms are not mutually exclusive in the senses intended here, indeed, they are complementary: The oral poet is one who, at the moment of performance, makes spontaneous, and therefore original realizations of inherited, traditional impulses. As regards Homer's tradition, let me first consider how Parry's immediate predecessors, Witte, Meister, and Meillet (along with Arend afterwards), were dismayed by their discoveries of the extent to which Homer's verse was conditioned by meter, inherited diction, and other conventions, as is reflected in Meister's dictum that, "the Homeric art4 The modest statistical density of formulas in the fragments of Panyassis, as well as in Hesiod and the Homeric Hymns, could mean either that these works are in a transitional style or that their poets had some "unhomeric" formulas of their own, as is known to be the case among contemporary South-Slavic singers. Cf. Wallace E. McLeod, "Studies on Panyassis, an Heroic Poet of the Fifth Century," Phoenix 20 (1966) 95-110. I agree with G. P. Edwards's view that it is something about the use rather than the number of formulaic templates which makes Hesiod look like an oral poet; see Language of Hesiod, but compare Dihle, Homer-Probleme 123f. For other oral poetries cf. esp. Larry D. Benson, "The Literary Character of Anglo-Saxon Formulaic Poetry," PMLA 81 (1966) 334-341 (Benson's point is well taken, though there may be a statistical error in his calculations). 5 One tends to think of extemporaneous composition only in what C. S. Lewis called "primary epic" in his Preface to Paradise Lost (New York 1961), but it has also been found, complete with formulaic diction, in many lyric traditions; cf. M. B. Emeneau, Toda Songs (Oxford 1971); Robert C. Culley, Oral-Formulaic Language in the Biblical Psalms (Toronto 1967); Ching-Hsien Wang, Shih Ching: Formulaic Language and Mode of Creation (Diss., University of California, Berkeley 1971). This complicates the question of transitional styles and the question of premeditation versus spontaneity. For a glimpse of premeditated and completely extempore composition side by side in the same oral tradition, cf. Lee M. Hollander, The Skalds (Ann Arbor 1968), pages 63-65 and 211.

xxi

SPONTANEITY

AND

TRADITION

language composes for the poet." Absurd as it may sound, this hapless conclusion has been reached not only in Homeric studies but in virtually every area of oral poetry. 6 The trouble, as one can now begin to see, came from regarding the oral poet's tradition as a finished product, a repertoire of unalterable conventions already completed, with an existence of its own, separate from the minds of its bearers. This view may be illustrated by an extreme example from a sophisticated writer, one to whom such a concept might more properly be applied. In the first part of Don Quixote, Chapters XI-XIII, the demented hero and his squire fall in with a group of simple goatherds in a rustic spot in the Pyrenees. These entertain the Don with some local color concerning the recent death of a certain pseudo-peasant named Chrysostom who has succumbed to unrequited love. The object of his affections was the equally factitious shepherdess Marcella, because of whose great beauty and cold heart the entire forest is resounding with the groans and sighs of disappointed lovers-turned-rustics: 7 "Not far from here is a place where there are about two dozen great beeches, and every one of them has Marcella's name cut on its smooth bark. . . . Here one shepherd sighs, there another moans; from the distance you can hear songs of love; from near at hand dirges of despair." Cervantes' ironic thrust here depends upon both the distance and the recognizability of the two clearly defined sets of literary conventions he has drawn into the context of his work, namely Gothic romance (the world of Don Quixote's private delusions) and Latin and Italian pastoral (that of his equally lunatic environment). Now for Cervantes, these conventions were assuredly complete in themselves, something handed down to him from the literary past (in fact, largely from areas outside his own language), something he had learned from a relatively small number of finished examples and perhaps from secondary sources as well, something outside himself which he could treat 6 Meister's view can be traced back to the romantic notions about unlettered poetry which arose in the eighteenth century, as can be seen from Lewis's quotation from Goethe {op. cit. 25). It is taken up by Robert P. Creed, "The Making of an Anglo-Saxon Poem," English Literary History 26 (1959) 445f, and K. Kailasapathy, Tamil Heroic Poetry (Oxford 1968) 138f, among others. Although oral verse-making is special (cf. MHV 328), whether meter forms diction or diction meter is actually something of a chicken-and-egg problem. A very useful formulation, on the other hand, is that of Hainsworth: "In doi8f| the singing tradition there are necessarily already planted the seeds of noir|aiq poetic creativity" CFlexibility 128). 7 Tr. of J. M. Cohen (Baltimore 1950) 94. Note that I am not here comparing Homer to Cervantes, which would require a study of the latter's relationship to his own tradition, the picaresque novel.

xxii

INTRODUCTION with ironic distance and yet could not manipulate very much without risking loss of intelligibility. But the "tradition" in which Homer composed was not like that: for him it was a still-living stream which operated at a deeper level of consciousness. Despite appearances (as I shall argue below), it was essentially an inheritance of habits, tendencies, and techniques rather than of completed entities. Thus, once a singer learned to compose in his tradition it could not have occurred to him to step outside it, any more than it occurs to us in the ordinary course of events to speak something other than English. Thus also, the tradition itself composed no poetry—that was, as always, the work of poets. There is a world of difference, then, between the Kwstsprache of Karl Meister and the "art language" of Cedric Whitman—a concept more appropriate to oral-traditional poetics, which I shall discuss at greater length in the following chapter and indeed throughout this book. What of the concept of spontaneity? It is his unfamiliarity with the workings of an oral singing tradition that has made the idea of spontaneous composition so uncomfortable to the literary scholar of today. O f course, it is impossible to extemporize poetry of any value on a regular basis without a well-developed oral tradition; what the writing poet does in premeditating or revising his work has already been done by the oral poet in the long years of training he has undergone before he opens his mouth to sing (as Lord has rightfully stressed, cf. especially ST 24). What the writing poet does relatively consciously (perhaps) and relatively alone, the oral poet has already done at a deeper level of awareness and in more intimate cooperation with his teachers, his peers, and finally his audience. Incidentally, therefore, the much vaunted originality of the sophisticated writer, even if it were not illusory for the reasons to be considered below, has been purchased at the terrible cost of alienation—as can be seen today in art and in society at large. A t the present stage of our knowledge, we have no real reason to think that the oral poet's tradition confined him to a relatively narrow range of dictional or thematic choices compared to a writing poet or to the speaker of a natural language generally (if "choice" is, in fact, a concept applicable to linguistic creativity); or to think, on the other hand, that his spontaneity, his ability to compose in performance, was a kind of mantic creativity ex nihilo (which, in fact, the singers themselves like to think it is).

xxiii

SPONTANEITY

AND

TRADITION

Now spontaneity, however defined, operates from some rather obscure region of the mind; and the same is true, as I shall illustrate below, of the traditional component of the bard's creative process. Not only is it true, as Milman Parry pointed out, that to study Homer's language fully would be to study the full complexity of his mind (see MHV 307); but to study his language in any depth must involve a study of subconscious factors. What, then, can the literary scholar hope to achieve? The answer lies, I think, in what Professor Lord means by the "training of the ear," the patient but imaginative work whereby one can gradually come to understand Homer's bygone "art language" in much the same way that one learned to read his Greek. At a level between the denotations of his words and phrases, which can be learned quite well after some habituation, and the preverbal impulses from which they sprang, which are irrecoverable by nature, lies what a school of modern linguists would call the "multilayered semiotic system" of his total language, and this system should yield some of its secrets to careful and sympathetic study. That, at any rate, is the attempt I shall be making in the following pages. Beginning with phrases and moving along the spectrum of increasingly larger segments of the composition, I shall try to discern the patterns that underlie them. The question of meaning, and therefore the immediate and general context of the linguistic units under study, will enter the picture almost from the outset. In this regard, among others, I hope to try a somewhat different tack than that of most previous investigators: Bernard Fenik, for example, writes in his thorough and perceptive study of the "typical battle scene," "It seems that the pressure exerted by a familiar pattern could be so strong that it forced the inclusion of one of its standard elements into a particular scene where it is either inappropriate or could not be further developed" (Fenik 53). While this has unquestionably occurred many times in both poems, my provisional credo, and the work of certain modern anthropologists, encourage me to look at such an occurrence rather differently: The pattern, be it syntactic-phonological (for the "formula") or narrative-structural (for the "type scene"), need not be considered quite so autonomous, a view which is after all an extension of Meister's dictum. A pattern is an expression of the poet's conscious or unconscious purpose; it is part of his "semiotic system" and one can learn to read it. This part of his system lies below the surface of the narrative, which itself is often inconsistent and illogical; superficial illogicality was a natural by-product of the oral compositional situation XXIV

INTRODUCTION

and certainly did not cause Homer's audience any grief, although it dropped out of fashion with the introduction of composition by writing, to re-emerge only in our own century (where it seems to be cultivated for its own sake). Once one has grasped these underlying patterns one finds that minor inconsistencies in the speeches and actions of the poet's characters are not so much resolved as ignored in an awakened understanding of the real tale that he is singing. One then responds to the story in somewhat the way that Homer's audience did; both his original aural audience and his Greek listeners and readers for several centuries after the breakdown of the oral-composing tradition. I shall be leaning heavily for this investigation upon the work of many fellow Homeric scholars, for the current state of the field is a "mosaic of contributions," as Hainsworth says. However, in order to resolve the individual tessarae into a complete picture, one must also try to gain a wider perspective. The scope of this study is the totality of Homer's art, which is partly unfamiliar and partly the same as great poetry of any kind; there are insights to be gleaned from the special field of oral poetry, from literary criticism in general, and from disciplines at various removes from Homeric studies which cannot be left untapped. The first step is a reinvestigation of the phenomena of recurrent diction which have led to the concept of the "oral formula."

xxv

i. The Traditional Phrase (I): Theory of Production We may begin our study by reinvestigating a type of resemblance among phrases which afforded Parry some important insights into the production of "new" formulas. He called this resemblance calembour, and illustrated it with a pair of phrases, &Hcpf|X.uOev f|5u

-ÌKETO 5 f | | i o v

This fact raises as a rather startling possibility the notion that there may be formulas that are not made up of word groups at all, and, more generally, that considerations other than the present concept of word end may contribute to feelings of rhythmical subdivision within the epic hexameter. At the least, these corresponsions should suggest the operation of psychological cola or rhythmical groups of some sort bearing a hitherto undetermined relation to formulas, and based upon factors that are not 11

The term was used by Mezger and Bury for Pindar. This particular kind of corresponsion of vocalic and consonantal patterns is in fact present to some degree among all our examples and in the ST^IÓ; group as well; àpyéti Srmqi, SinXaKi 5T||I