Oral Tradition and Synoptic Verbal Agreement: Evaluating the Empirical Evidence for Literary Dependence

By Travis Derico

New Testament scholars routinely claim that verbal agreement among parallel Synoptic pericopae is a reliable indicator of literary borrowing by the Synoptic Evangelists. In Oral Tradition and Synoptic Verbal Agreement, T. M. Derico presents a critical assessment of that claim through a consideration of the most recent empirical evidence concerning the kinds and amounts of verbal agreement that can be produced among independent performances of oral traditions.

• Language: English
• Publisher: Pickwick Publications
• Release date: Jun 28, 2016
• ISBN: 9781532600432

Travis Derico

T. M. Derico teaches Bible and Religion at Huntington University in Indiana.

Book preview

9781620320907.kindle.jpg

Oral Tradition and Synoptic Verbal Agreement

Evaluating the Empirical Evidence for Literary Dependence

T. M. Derico

29395.png

ORAL TRADITION AND SYNOPTIC VERBAL AGREEMENT

Evaluating the Empirical Evidence for Literary Dependence

Copyright © 2016 T. M. Derico. All rights reserved. Except for brief quotations in critical publications or reviews, no part of this book may be reproduced in any manner without prior written permission from the publisher. Write: Permissions, Wipf and Stock Publishers, 199 W. 8th Ave., Suite 3, Eugene, OR 97401.

Pickwick Publications

An Imprint of Wipf and Stock Publishers

199 W. 8th Ave., Suite 3

Eugene, OR 97401

www.wipfandstock.com

paperback isbn: 978-1-62032-090-7

hardcover isbn: 978-1-4982-8558-2

ebook isbn: 978-1-5326-0043-2

Cataloguing-in-Publication data:

Derico, T. M.

Oral tradition and synoptic verbal agreement : evaluating the empirical evidence for literary dependence / T. M. Derico.

x + 354 pp. ; 23 cm. Includes bibliographical references.

isbn 978-1-62032-090-7 (paperback) | isbn 978-1-4982-8558-2 (hardback) | isbn 978-1-5326-0043-2 (ebook)

1. Synoptic problem. 2. Oral tradition. 3. Bible—Gospels—Criticism, interpretation, etc. I. Title.

BS2555.52 D33 2016

Manufactured in the U.S.A. 06/21/16

All translations are by the author unless otherwise noted.

Table of Contents

Title Page

Acknowledgments

Abbreviations

Chapter 1: Introduction

Chapter 2: Oral Tradition and the Problem of Synoptic Verbal Agreement

Chapter 3: Category 1 Anecdotal Evidence

Chapter 4: Category 2 Anecdotal Evidence

Chapter 5: Evidence from Transcribed Oral Texts

Chapter 6: Evidence from Experimental Psychology

Chapter 7: Verbal Agreement in the Synoptic Gospels and the Whitman Narratives

Appendix A: Transcripts

Appendix B: Quantitative Comparison

Bibliography

Acknowledgments

This book could not have been written without a great deal of help. Ron Piper and Chris Rowland gave invaluable guidance in the early stages of the project, and David Wenham patiently shepherded a version of it to successful completion as a doctoral thesis in the University of Oxford. Conversation and fellowship with Atsuhiro Asano, Steve Guthrie, Jon Norton, Jeff Schroeder, Robert Schumann, and Daniel von Wachter helped to clarify my thinking on various points, and made life better generally. In Jordan I received grace and hospitality from Fawaz Ameish, Peter and Kathy Clark, Dan and Mary Gibson, Jiryes Habash, Peter Law, Asma Ma’ayah, Jamil Ma’ayah, Wadia Qandalaft, Fayez Qaqish, Abdullah and Yasmin Rizq, Nabil and Carol Sweis, and Jalil Zureiqat. Sa’ib Ma’ayah gave crucial logistical assistance, especially in the arrangement of interviews. Muna Khashram made the initial transcripts of my recordings. Chris Dawson did much to help me orient to life in Amman, and also helped with the translation of several of my Arabic transcripts. Suheil Madanat generously provided much-needed background information about Roy Whitman and his role in the Jordanian evangelical community. I am deeply grateful to all these people, and most of all to Laura Derico for her unfailing love, endurance, and editorial support.

Abbreviations

Abbreviations in this book follow the SBL Handbook of Style, except the following:

ICOT 1991 Informal Controlled Oral Tradition and the Synoptic Gospels. Kenneth E. Bailey. Asia Journal of Theology 5 (1991) 34–54.

ICOT 1995 Informal Controlled Oral Tradition and the Synoptic Gospels. Kenneth E. Bailey. Themelios 20 (1995) 4–11.

OWG The Oral and the Written Gospel: The Hermeneutics of Speaking and Writing in the Synoptic Tradition, Mark, Paul, and Q. Werner H. Kelber. Voices in Performance and Text. Bloomington, IN: Indiana University Press, 1997.

1

Introduction

This book is an attempt to address part of a difficult problem in New Testament studies that, owing to certain advances in our understanding of the relevant issues, has recently got much worse. The problem can be adequately stated in three sentences:

1. Some first-century Christians remembered and transmitted oral traditions about Jesus.¹

2. The Synoptic Evangelists made some use of some of these traditions in the composition of their Gospels.

3. We don’t know very much about these traditions or how the Synoptic Evangelists used them.²

New Testament scholars have long been aware of this problem and of the crucial importance of its solution.³ The Synoptic Gospels contain the earliest and most detailed accounts of the life and ministry of Jesus of Nazareth now extant. To be able to accurately assess the data contained in these texts concerning Jesus, the primitive church, the Synoptic Evangelists, and the Synoptics themselves, we need to know something about the sources consulted in process of their composition. And since the Synoptic Evangelists were undoubtedly familiar with orally transmitted Jesus traditions, we need to know the extent to which their Gospels were derived from or influenced by those traditions. Unfortunately, we have very little unambiguous evidence to indicate the precise character of any part of the first-century oral Jesus tradition, or to illuminate the editorial policies of the Synoptic Evangelists with respect to it.

1.1 The Good News

However, it is widely supposed that if this problem has not been exactly solved, it has been largely circumvented. For more than two centuries, scholarly considerations of the first-century oral Jesus tradition have been based on a conception of oral tradition as a single, monolithic phenomenon possessing certain essential characteristics.⁴ On this view, every case of oral-traditional activity is just an instantiation of a globally uniform behavior—a universal orality—and suitably fundamental features of any such case can be reliably generalized to any other. Thus, even though we do not possess any detailed early descriptions of the methods employed by the first Christians to transmit their oral traditions about Jesus, we can learn most of what we need to know by observing the universal features of more accessible oral traditions.⁵ And indeed many New Testament scholars have claimed that by attending carefully to the essential characteristics of oral tradition it is possible to identify traces of those characteristics in the texts of the Synoptic Gospels.⁶

Admittedly, there remains a good deal of disagreement among New Testament scholars over just what characteristics are essential to oral tradition. A number of suggestions have been proposed: oral tradition has been portrayed as subject to the operations of certain definite laws,⁷ as being inherently formulaic,⁸ as constrained by various features of human memory,⁹ as entailing particular methods of composition, transmission, or performance,¹⁰ as differing in specific ways from written tradition or literacy in general,¹¹ and as being inspired and controlled by, or alternatively as inspiring and controlling, a specifically oral mentality;¹² and typically several or all of these are held to be essential at once.¹³ But these proposals have all been criticized on grounds that the phenomena they describe cannot be sufficiently distinguished from similar phenomena observed to occur in the production of written literature,¹⁴ that they do not adequately account for certain features which may be observed to exist in some known oral-traditional systems,¹⁵ or that they are misconstruals or misapplications of hypotheses originally constructed by folklorists and anthropologists.¹⁶ Consequently, none of them have achieved anything like a general acceptance among Synoptic critics.

Nevertheless, there is one characteristic that is now almost universally acknowledged by New Testament scholars as being essential to oral tradition: variability.¹⁷ Variability in this context is not a mere capacity, but a positive tendency to change. To describe oral tradition as essentially variable is to invoke every factor which is able to bring about any kind of change in an orally transmitted tradition (lapses of memory, intentional creativity, miscommunications, or whatever) and to claim that all such traditions must eventually be affected by at least one of these factors—and that this tendency to change is in some way distinctive of oral, as opposed to written, tradition.¹⁸

Sensitivity to this feature of oral tradition has enabled Synoptic critics to develop a rough test for determining the influence of oral Jesus traditions on the texts of the Synoptic Gospels, on grounds of the following argument: Since oral tradition is essentially variable, we should expect to see more variation among texts that have been directly influenced by oral traditions than among texts produced by literary copying. But then the more variation we observe among the Synoptic texts, the more likely it is that those texts were influenced by oral Jesus traditions; and the less variation we observe, the more likely those texts were produced by literary copying.¹⁹

This test has proved crucial to the methodological development of modern Synoptic criticism. The most celebrated feature of the Synoptic Gospels is, of course, their considerable similarity to one another in wording and narrative order. Without some knowledge of the oral and written sources available to the Synoptic Evangelists and their methods of dealing with those sources we have no way to tell how this similarity was produced. Repeated applications of the variability test, however, have demonstrated to the satisfaction of nearly all New Testament scholars that the similarity among the Synoptics is so great overall that they must have been produced primarily by reference to written sources. Moreover, the variability test provides a standard criterion against which claims about the oral-traditional provenance or flavor of particular Synoptic passages can be judged.²⁰ In effect, the variability test is the methodological starting point for source-, form-, and redaction-criticism of the Synoptic Gospels, none of which can get off the ground unless it is possible to distinguish the marks of deliberate literary redaction from those of independent reference to orally transmitted Jesus traditions.²¹ By treating the oral Jesus tradition as a specific occurrence of a global phenomenon, Synoptic critics have been able to construct an apparently stable foundation on which to establish the central methods of their field.

To be sure, important problems remain. But since we possess the means in principle to distinguish between tradition and redaction, and since by that means we already know that there is a substantial literary relationship among the Synoptics, we can continue to advance our understanding of the various literary and nonliterary relations among the Synoptic Gospels even if those problems and questions turn out to be completely intractable. For this reason the lack of detailed historical evidence with respect to early Christian oral tradition has not been viewed as seriously interfering with the progress of mainstream Synoptic criticism.

1.2 The Bad News

In the last few decades, however, it has become clear that the conception of oral tradition outlined above is false. It is no longer possible, in light of the accumulated field research on the composition, transmission, and use of oral traditions that has been carried out by ethnographers and folklorists in communities around the world, to credibly depict all oral-traditional activity as universally similar either in general or in regard to particular features.²²

Oral-traditional literature takes a multitude of forms: epic, lyric, elegiac, and panegyric poetry; ballads, odes, and work-songs; legends, tales, and sagas; parables, proverbs, aphorisms, and riddles; jokes; genealogies; laws and regulations; prayers, hymns, and liturgies; incantations, charms, and curses—all of these genres, and many more besides, are employed in oral-traditional contexts.²³ Oral traditions carry an enormous variety of information, including descriptions of family relations, accounts of historical or fictional persons or events, humorous anecdotes, technical knowledge, and ethical and sapiential instruction. Anything sufficiently interesting or important to a community or one of its members might be entered into its oral-traditional repertoire.²⁴ Oral traditions are employed to perform a broad assortment of functions: to entertain and to teach, to praise and to ridicule, to strengthen values, and to subvert them.²⁵ And oral traditions are composed and transmitted by a wide variety of methods as deemed appropriate to their forms, content, and functions.²⁶ Poetic traditions are often composed by different methods than prose traditions, and traditions which are designed to entertain may be transmitted using different techniques, and in different settings, than traditions which are meant to display or encourage religious devotion.²⁷

This is not to suggest that every oral-traditional system is completely unique, or that it is not possible to observe similarities between particular examples of oral literature. Substantial similarities can sometimes be observed among culturally unrelated oral-traditional systems and their products. But such similarities cannot be simply assumed to exist out of hand. Human culture is not reducible to a special oral variety with its own set of controlling laws, and human intellects are not constrained by a collective oral mentality until they acquire a sufficient degree of literacy.²⁸ No feature of one oral-traditional system or example of oral-traditional literature is universal to all—not even variability.

Variability is not a single, specific feature at all, of course, but only a rather broadly conceived phenomenon which may result from diverse causes. It is true that all sorts of changes may befall a tradition in the process of oral transmission: the opportunity for intentional or accidental addition, omission, or alteration is present at every recitation. But change is not a necessary correlate of oral transmission, and when traditions do change they are not all changed in the same ways or to the same extent.²⁹ It is impossible to predict how or how much an oral text will change in transmission apart from some knowledge of the oral-traditional process that produced it.³⁰

The consequences for Synoptic critics of this new awareness of the particularity of oral traditions are potentially severe. Since there are no invariable or universal standards of oral-traditional form or practice, we cannot appeal to any such standard as grounds for identifying Synoptic interrelationships. But appeals to such standards are the implicit evidential foundation of every currently held literary-source hypothesis of Synoptic composition.³¹ An ability to differentiate between features resulting from deliberate editorial activity and features resulting from transcription or revision of oral traditions is prerequisite to the current practice of source-, redaction-, and tradition-criticism, which methods remain central to the continuing debate over the authorship, compositional history, and theology of the Synoptic Gospels.³² But since we have no reliable way of distinguishing between texts produced by literary and oral-traditional means, we do not know whether our basic critical methods are capable of producing true solutions to the problems they are employed to solve.

Nor can we simply dodge the issue by claiming that we know a fair bit about Greco-Roman literary compositional practices, and that ancient authors could have produced most of the kinds of similarities and differences we observe among the texts of the Synoptic Gospels by redactional means.³³ These claims are neither untrue nor unimportant, but they are irrelevant to the issue at hand, for we also know that oral Jesus traditions influenced the composition of the Synoptics. But we do not know how those traditions were composed, preserved, or transmitted; so we do not know the extent to which the Synoptic Evangelists might have produced the kinds of features we observe among the texts of their Gospels by reference to orally transmitted Jesus traditions; and so neither do we know the extent to which they produced those features by strictly literary means. Under these conditions, to assume the primacy of literary explanations of Synoptic phenomena would be to consign ourselves to a kind of drunkard’s search, hunting diligently for our keys under the one lit streetlamp on an otherwise dark street, when they could just as well be anywhere between home and the pub.³⁴ We might find what we are after in the one area we know something about, but it is hardly a sure bet. We are almost certainly better off expanding our search. But to do this, we need more light.

1.3 A Way Forward

Fortunately, the obsolescence of the universal-orality model of early Christian oral tradition does not leave us completely without resources for addressing these issues. Although we can no longer describe the first-century oral Jesus tradition by reference to the characteristics of a generalized phenomenon, we can still legitimately ask whether it might have been similar in certain respects to some particular oral-traditional system. And this suggests the rudiments of a more critically defensible method by which we might begin to assess the influence of early Christian oral tradition on the Synoptic Gospels: we could hunt down and survey a large number of comparable parallel oral-traditional texts, and see whether they bear any relevant similarities to the Synoptics. This is one expedient that has not yet been tried, but is almost certain to produce useful results if carried out with sufficient rigor. Suppose we should find some oral-traditional system that tended to produce oral literature displaying certain features similar to those we observe in the Synoptics. Then we might have good grounds to think that the corresponding Synoptic features could have been produced without reference to written documents. If, on the other hand, after observing a substantial number of relevantly similar oral-traditional texts, we found that none of those texts contained certain features displayed among the Synoptics, we might have good reason to think that those features were not produced solely by means of the Synoptic Evangelists’ independent reference to orally transmitted Jesus traditions.

But so far this is all just a bit of hand waving. Similarity and relevance can be understood in various ways, and while it will be impossible (and perhaps undesirable) to completely eliminate the role of scholarly intuition from an investigation of this kind, it will not do to return to a de facto universal-orality model of early Christian oral tradition through lack of vetting. What we need first is a principled way to decide what kinds of oral-traditional systems might be properly comparable to the first-century oral Jesus tradition, and what kinds of oral-traditional literature are properly comparable to the Synoptic Gospels. And in fact there is a serviceable set of criteria for this purpose ready to hand, in the work of the late John Miles Foley.³⁵

The pioneering fieldwork of Milman Parry and Albert Lord in the former Yugoslavia and its application to the Homeric Question is the premier success story in the history of the comparative study of oral traditions.³⁶ Parry’s and Lord’s demonstration that the metrical demands of Serbo-Croatian Muslim oral epic poetry led to the production of structural and stylistic features analogous to the noun-epithet formulae of the Homeric epics radically altered the landscape of both Homeric studies and folkloristics, and became the foundation of a theory of folklore which identified the presence of formulae in written texts as a key signature of oral-poetic composition. This first great success inspired a large and diverse group of researchers to adopt and adapt the comparative method advocated by Parry and Lord, so that by the mid-1980s their Oral Theory had been employed in published studies of the oral traditions and written literature of more than a hundred different language areas.³⁷

In his 1988 history of the development of the Oral Theory, John Miles Foley hailed the huge expansion of oral-formulaic studies as nothing less than the making of a new discipline.³⁸ However, he also noted with some concern that enthusiasm for the past successes and future promise of the theory had sometimes been allowed to prevail over critical judgment. In particular, Foley observed that the oral-formulaic school had developed a tendency to emphasize the similarities between oral traditions and examples of oral literature to the exclusion of any differences. Warning his colleagues of the need for an increased awareness of methodological preliminaries in studies associated with Oral Theory, he proposed three principles that could serve as a corrective to some of the critical lapses of the past.³⁹ Foley termed these principles tradition-dependence, genre-dependence, and text-dependence. Because of their importance and their direct relevance to the problem at issue here I shall quote his discussion of them at length: ⁴⁰

By tradition-dependence I mean allowing each oral poetic tradition its idiosyncratic features and actively incorporating those features into one’s critical model of that tradition. In practical terms, this would include aspects such as natural language characteristics, metrical and other prosodic requirements, narrative features, mythical and historical content, and any other aspect that is peculiar to the given tradition and therefore a significant part of its definition. There is, I believe, no reason to suppose that traditional units which take shape under different, tradition-dependent systems must be exactly or even closely comparable. . . . By genre-dependence I mean demanding as grounds for comparison among traditions nothing less than the closest generic fit available, and, further, calibrating any and all comparisons according to the exactness of that fit. It has been one of the major shortcomings of Oral Theory in the past that it has not observed this simple criterion; especially in Old English, where the variety of genres is considerable and the number of texts in any one category usually rather small, the principle has been ignored again and again—setting lyric elegy beside epic or verse hagiography or riddle, building insupportable bridges between Beowulf and dissimilar genres from other traditions. Positively, on the other hand, genre-dependence encourages comparison of genres if a basic congruency can be established. . . . The third principle of comparison, text-dependence is simply the necessity to take into account the precise nature of each text: unquestionably oral or oral-derived, recorded from sung performance or dictated, audio record or manuscript, and so forth. Uncritical comparison of manuscripts and taped recordings, for example, can mislead more than inform. We should know, as far as possible, the circumstances of collection, the history of manuscript transmission, and the many other factors that would help to calibrate the comparison realistically. Where these details of contextualization are missing, we should admit the lacuna, even if it means leaving the actual oral provenance of the text in question. Finer distinctions, such as the diplomacy of an extant text or the editing process through which it might have or actually did pass, should also be made. This third principle, like its two counterparts, demands patient scrutiny of all of the problems involved before analytical research begins. And it also demands, of course, that all comparanda, even those briefly summoned, be consulted in their original languages and with philological precision, a practice unfortunately not uniformly observed in the past.⁴¹

Upon reading these recommendations, those familiar with scholarly discussions of the extra-literary history of the Synoptic tradition should be struck by at least two points. The first is that although Foley’s treatment of these principles is calibrated to a discussion of epic poetry, the principles themselves are eminently suitable for application to studies of the relationship between the Synoptic Gospels and the oral tradition of the early church. In order to construct a convincing model of the first-century oral Jesus tradition, one must be able to show that one’s proposed oral-traditional mechanism could have produced oral literature relevantly similar to at least some of the Synoptic materials, and that it (or something like it) was available to first-century Christians. This will not be possible without comparing the Synoptic materials to some known examples of oral literature, or without comparing the alleged circumstances of the transmission of oral Jesus traditions to the circumstances in which some known oral traditions have been transmitted. Given the inherent difficulty and complexity of such a project, and the enormous variety of modern and ancient oral traditions, Foley’s principles are well-constructed to help us determine not only which texts and which traditional mechanisms are properly comparable, but also how the resulting comparative data should be analyzed and displayed.

The second point is that there are very few comparisons of any part of the Synoptic materials with known oral-traditional literature that meet the standards of rigor and transparency demanded by Foley’s principles of dependence. Most of the factors Foley mentions as relevant to comparisons between oral traditions are never even mentioned by New Testament scholars; and the criterion which is at once most obviously relevant and most easily applied—a basic congruency with respect to genre—is routinely ignored. This is not entirely the fault of New Testament scholars: the study of oral traditions has never really been part of their training, and the universal-orality model that has so far dominated discussion of these matters does not encourage recognition of differences among oral traditions. Moreover, because the professional interests of anthropologists and folklorists do not widely overlap those of New Testament scholars, there are few published studies of orally transmitted prose traditions that are generically similar to the Synoptic materials and that include the kinds of precise native-language transcripts of independent parallel recitations of traditional texts required for a detailed comparison. Before we can launch a really illuminating attempt to determine the relationship between the Synoptic Gospels and the first-century oral Jesus tradition, we are going to have to lay hands on the right sort of oral-traditional data. In the meantime, we would do well to begin re-evaluating the received methodological apparatus in light of the considerations just raised, and it is with this task that the main part of this book is occupied.

1.4 The Problem of Verbal Agreement

Any attempt to determine the role played by oral Jesus traditions in the composition of the Synoptic Gospels must address the agreement in wording displayed among many parallel Synoptic pericopae. In particular, such an attempt will have to provide a plausible answer to the question: How much of the verbal agreement we observe among the Synoptics could be accounted for on the supposition that the Synoptic Evangelists referred exclusively to orally transmitted Jesus traditions? Although the responses to this question given by individual New Testament scholars often differ in detail, the consensus view is that a suitably precise answer to it has already been established—namely, very little.

Over the next five chapters I will expose the consensus view to serious scrutiny. In chapter 2, I present a detailed exposition of the principal axiom of mainstream Synoptic criticism that is thought to support that view: the claim that Synoptic verbal agreement is a good indicator of Synoptic literary relationships. This claim cannot be properly evaluated apart from a careful evaluation of the empirical evidence that has been adduced in its support. In chapters 3–6, therefore, I assess the most important non-rabbinic empirical evidence so far presented for or against any claim to the effect that a particular kind or amount of Synoptic verbal agreement can only have been produced by a process of literary copying. In light of that assessment, I argue that none of the empirical evidence on offer is sufficient to justify strong commitment to the axiom under consideration, and that we are therefore obliged to keep an open mind as to whether any particular case of Synoptic verbal similarity might have been produced without reference to written documents.

All this, I suggest, demonstrates the inherent weakness of the intuitive approach that has characterized modern scholarly debate over the character of early Christian oral tradition and its role in the composition of the Synoptic Gospels. If we are to make any real headway in our understanding of these matters, we must develop more effective methods for determining the sorts of verbal and sequential phenomena that might plausibly have been produced among reciters of oral Jesus traditions in the earliest period (and later), under particular oral-traditional circumstances. In a final chapter I offer a sample demonstration of one such method.

1. The term oral tradition is notoriously difficult to define with precision (note the different definitions proposed in Vansina, Oral Tradition as History, 27–29; Miller, African Past Speaks, 2; Henige, Oral Historiography, 2; Rosenberg, Complexity of Oral Tradition, 80; and the general suspicion with which such definitions and even the term itself are viewed in, e.g., Cohen, Undefining, and Finnegan, Oral Tradition). In this book the term oral tradition will be used broadly to mean information transmitted orally over the course of two or more generations (oral traditions or an oral tradition thus being particular cases of such information), or the process by which such information is transmitted.

2. Although the first two of these points have never really been a matter of dispute among New Testament scholars, I will just summarize the main grounds for their acceptance here. First, there is an extremely strong prima facie case for the existence of some sort of oral Jesus tradition in the first century. The self-identification of the earliest Christians as followers of Jesus the Christ and the very existence of the Synoptic Gospels suggest that the earliest Christians felt compelled to explain much of what they were doing and saying by reference to what Jesus said and did. But a large proportion of those early Christians who helped effect the rapid spread of Christianity throughout the Roman world by preaching to the uninitiated, teaching in the churches, bringing up their own children as Christians, and so on, were illiterate (see, e.g., Harris, Ancient Literacy, 22; Bar-Ilan, Writing in Ancient Israel, 33–34; Gamble, Books and Readers, 5–10). An even larger proportion of those early Christians will have been unable to write proficiently (see, e.g., Harris, Ancient Literacy, 4–5; cf. Youtie, βραδέως γράφων), and most of the people they were exhorting and instructing and bringing up were illiterate too. The notion that few stories about or sayings of Jesus were handed on orally in all of this seems a little implausible. Further, there will have been considerable practical incentive toward independence from written texts where possible and appropriate, even for literate Christian leaders. Apart from the inconvenience of having to carry around a lot of scrolls, there is evidence to suggest that teachers who knew their material by heart will have been more highly regarded (qua teachers, anyway) than those who had to rely on written notes (on this see especially Alexander, Living Voice). Second, there is direct early Christian testimony affirming the existence of an oral Jesus tradition immediately before (see esp. 1 Cor 11:23; 15:1–7) and after (see esp. John 21:25 and Papias in Eusebius, Hist. eccl. 3:39) the period when the Synoptic Gospels are generally supposed to have been written. Third, the prologue to Luke’s Gospel reports that a number of prior authors had made use of oral traditions about Jesus, and that Luke made use of them as well. Luke’s eyewitnesses and servants of the word [αὐτόπται κὰι ὑπηρέται τοῦ λόγου] are said to have "handed on to us [παρέδοσαν ἡμῖν] a narrative concerning the things that have been fulfilled among us" [διήγησιν περὶ τῶν πεπληροφορημένων ἐν ἡμῖν πραγμάτων], and Luke claims that he is equally well-placed to do what his literary predecessors have done (so ἔδοξε κἀμοὶ παρηκολουθηκότι ἄνωθεν πᾶσιν ἀκριβῶς καθεξῆς σοι γράψαι [it seemed good to me also, having followed everything from the beginning, to write for you an orderly account]). Luke thus places himself along with his predecessors in direct connection with the eyewitnesses of Jesus, and declares that he will be doing just what his predecessors had done (cf. Munck, Albright, and Mann, Acts, XV–XVI; Hengel, Four Gospels, 141–42). It is possible that some eyewitnesses might have produced written accounts of some of their traditions, and there is no difficulty about Luke referring to those accounts as tradition [παράδοσις] (cf. Bauckham, Jesus and the Eyewitnesses, 36–38). But it is hardly likely that the eyewitnesses disseminated their traditions primarily in written form, or that Luke had no knowledge of oral Jesus traditions purporting to derive from eyewitnesses. Fourth, it is the unanimous position of the Fathers (including Augustine, who is often misread on this point; cf. de Jonge, Augustine) that each of the canonical Gospels was independently composed on the basis of orally delivered apostolic testimony and tradition (thus Irenaeus, Haer. 3:1 [cf. Eusebius Hist. eccl. 5:8]; Tertullian, Marc. 4:2, 5; Origen, in Eusebius, Hist. eccl. 6:25; Clement of Alexandria, in Eusebius, Hist. eccl. 6:14; Ephraem Syrus, Diat. Tat. App. 1:1–3; Epiphanius, Pan. 51:5, 6, 11; Eusebius, Hist. eccl. 2:15; 3:24; 6:14; Jerome, Vir. ill. 3, 7, 8; ad Hed. (Ep. 120), 11; more controversially, see also the comments of Papias in Eusebius, Hist. eccl. 3:39; and those of Justin, Dial. 106:1–3). Furthermore, both Irenaeus and Clement claim to have participated in the oral transmission of Jesus traditions (see, e.g., Irenaeus, Haer. 3:2–3, and in Eusebius, Hist. eccl. 5:20; Clement, Strom. 1). The fact that all of these learned men, some of whom had firsthand experience of the oral transmission of Greek-language Jesus traditions, thought it unexceptional that the Synoptic Gospels should be produced entirely by means of a natural oral-traditional process is strong evidence that the Gospels were composed at least partially by such means.

3. Modern discussion of the role of early Christian oral tradition in the composition of the Synoptic Gospels began in earnest in the work of Johann Gottfried von Herder (see Vom Erlöser), and has continued with varying intensity ever since. Summary treatments of the roles assigned to oral tradition in various of the theories of Synoptic relationships offered since Herder are presented in Reicke, Roots of the Synoptic Gospels, 1–23 (cf. Reicke, From Strauss to Holtzmann); Keck, Oral Traditional Literature, 105–13; Culley, Oral Tradition and Biblical Studies, 52–56; and Mournet, Oral Tradition, 54–99. Recent decades have seen a resurgence of interest in this subject by New Testament scholars, as evidenced by the comparatively large number of books and articles that have been published in the last thirty years dealing with it at length, e.g., Riesner, Jesus als Lehrer; Wansbrough, Jesus; Bailey, Informal Controlled Oral Tradition and the Synoptic Gospels (1995) (hereafter cited as ICOT 1995); Crossan, Birth of Christianity; Horsley and Draper, Whoever Hears You; Byrskog, Story as History, 123; Dunn, Jesus Remembered; Mournet, Oral Tradition; Bauckham, Jesus and the Eyewitnesses; Baum, Der mündliche Faktor; Weissenrieder and Coote, Interface; Wire, Case for Mark; Dunn, Oral Gospel Tradition; Eve, Behind the Gospels; and the reprints of Werner Kelber’s The Oral and the Written Gospel (hereafter OWG) and Birger Gerhardsson’s Memory and Manuscript.

4. Until relatively recently this conception of oral tradition was shared by the majority of academic folklorists and cultural anthropologists, first on social-evolutionary grounds, then on more straightforwardly bio-evolutionary grounds, and eventually on somewhat more diffuse psychological grounds; see, e.g., Stocking, Race, Culture and Evolution, esp. 110–32; Barnard, History and Theory, 27–46; Finnegan, What Is Orality; Halverson, Goody.

5. So New Testament scholars have claimed to discover and describe features of early Christian oral tradition by appeal to features of the composition or transmission of Kuba myths, Lushootseed narratives, Oxbridge aphorisms, Russian fairy tales, Irish epic prose, Serbo-Croatian epic poetry, and the Jataka collection of the Buddhist canon (see, e.g., Abel, Psychology of Memory, 276; Horsley and Draper, Whoever Hears You, 167; Streeter, Four Gospels, 185–86; Henaut, Oral Tradition, 78; Crossan, Birth of Christianity, 50–52; OWG, 80; Bultmann, History, 7), to name but a few examples. A casual perusal of the literature will turn up many more; Bultmann’s History of the Synoptic Tradition by itself contains analogies between parts of the Synoptics and at least half a dozen different genres of ancient Jewish, Greek, Roman, Egyptian, Indian, and modern African, Arab, English, German, Greek, Indian, Russian, and Turkish oral literature (see, e.g., 6, 46, 67, 106–8, 184, 187).

6. This claim is at least as old as David Strauss’s mythical point of view (see Strauss, Life of Jesus, esp. 27–70), is central to the work of the form critics, and has recently been defended in various ways by, e.g., Richard Horsley and Jonathan Draper (Whoever Hears You, 150–94), J. D. G. Dunn (Jesus Remembered, 173–254), David Rhoads (Performance Criticism), and Antoinette Wire (Case for Mark).

7. As, famously, Bultmann, History; Dibelius, Tradition to Gospel; Schmidt, Rahmen der Geschichte Jesu.

8. E.g., Kloppenborg, Formation of Q, 45–46; Jacobson, First Gospel, 9.

9. E.g., McIver and Carroll, Experiments and Distinguishing Characteristics; DeConick, Human Memory.

10. E.g., Dewey, Oral Methods; cf. Achtemeier, "Omne verbum sonat."

11. E.g., Crossan, Birth of Christianity, 85–89; Horsley and Draper, Whoever Hears You, 183–86; Harvey, Orality.

12. E.g., Davis, Oral Biblical Criticism, 18–20; Mournet, Oral Tradition, 164–65.

13. As, e.g., OWG, 14–15, 24, 26–27, 30, 51, 59, passim (and indeed by most of the authors cited in notes 5–12 above).

14. E.g., Talbert, Oral and Independent; Wansbrough, Introduction, 12; Kloppenborg, Variation.

15. E.g., Grobel, Formgeschichte, 111; ICOT 1995; Gerhardsson, Memory and Manuscript.

16. E.g., Boman, Jesus-Überlieferung; Sanders, Tendencies, 8–26; Keck, Oral Traditional Literature, 108, esp. n13; Halverson, Oral and Written Gospel; Gerhardsson, Reliability of the Gospel Tradition, 85n56.

17. The notion that oral tradition is essentially variable has a long history in Synoptic scholarship (see, e.g., Marsh, Dissertation, 204: "the supposition, that our three first Gospels were molded in one form, is difficult to be reconciled with the opinion of a mere oral Gospel, which must necessarily have assumed a diversity of forms," his emphasis). But since the publication in 1983 of Kelber’s extremely influential book The Oral and the Written Gospel (reprinted with a new introduction in 1997), it has become common for New Testament scholars to discuss it explicitly in terms of the inevitable variability of all oral performances (Kelber, Generative Force, 18), viz.: in the oral stage variability within limits is the rule (Betz, Sermon on the Mount, 370–71); Here . . . we have an excellent example of the oral principle of ‘variation within the same’ (Dunn, Jesus Remembered, 212, quoting OWG, 54, quoting Havelock, Preface to Plato, 1, 92, 147, 184, passim); Oral tradition consists of varieties (Andersen, Oral Tradition, 29); and so forth.

18. So Kelber: Verbatim memorization as a key factor in oral transmission has been abandoned by the majority of experts, who now admit the inevitability of change, flexibility, and degrees of improvisation (OWG, 27).

19. Something like this argument is implicit in claims to the effect that so much of the order in which that common material is presented, and so much of the wording in which it is phrased are the same that dependence at the written rather than simply at the oral level has to be posited (Brown, Introduction, 112). But occasionally the argument is alleged explicitly, as by Terence Mournet: "This ‘verbal variability’ is one of the key characteristics of oral tradition and although it is not a definitive criterion capable of distinguishing oral from written tradition, it can be helpful in shifting the probability of a tradition’s origins in one direction or another . . . if redundant forms of a tradition exhibit no variability, it would be difficult to argue that in their current form they are direct transcriptions of an oral performance" (Oral Tradition, 180, his emphasis).

20. The value of this criterion has been pressed by J. D. G. Dunn, who argues with regard to a number of double- and triple-tradition parallels that the degree of variation between clearly parallel traditions and the inconsequential character of so much of the variations have hardly encouraged an explanation in terms of literary dependence (on Mark or Q) of literary editing. Rather, the combination of stability and flexibility positively cried out to be recognized as typically oral in character (Jesus Remembered, 253–54).

21. The variability test is not, of course, the only test so employed by New Testament scholars; there are the various generic tests applied by the form-critics and others, and there are Robert Stein’s nine canons for identifying Markan redactional seams (see Gospels and Tradition, 70–84), and so on. However, the variability test is the foundational test for determining the existence of Synoptic literary relationships. It is historically and methodologically the first and most widely accepted of such tests, apart from which the other tests would lose much of their plausibility.

22. So, e.g., Jan Vansina: It is not possible to achieve a universal cross-cultural classification [of oral traditional genres] even if it becomes convoluted to excess (Oral Tradition as History, 82); John Miles Foley: there is . . . no reason to suppose that traditional units which take shape under different, tradition-dependent systems must be exactly or even closely comparable (Theory of Oral Composition, 109); Lee Haring: Commonsense anthropology will have to agree that there are no universal or invariant themes, techniques, or devices of literary art, whether oral or written. Each culture operates differently in the realm of verbal art (True Comparative Literature, 37); David Bynum: So be exceedingly wary of anyone who suggests to you that oral traditional composition in any one genre is in principle the same for any other genre, much less for all genres generally. That would be a very good indication indeed that the person who tells you so really doesn’t know what he is talking about ("Antiquitates Vulgares").

23. See the more comprehensive lists of oral-traditional genres in, e.g., Finnegan, Oral Poetry, 135–57, and Dundes, Study of Folklore, 1–3.

24. To take just some Arab examples, note e.g., the diverse kinds of traditions described in Furayhah, Dictionary (proverbs); Caspi, My Brother (laments); Shryock, Nationalism (tribal history); Kressel, Haqq Akhu Manshad (legal knowledge); Nasr, Oral Transmission (religious and philosophical commentary); and the oral-traditional narratives transcribed here in Appendix A.

25. See, e.g., Vansina, Oral Tradition as History, 95–114. Of course, oral traditions also perform functions which are not directly intended by anyone.

26. This is particularly worth noting in the face of a claim like: "In orality, tradition is almost always composition in transmission" (OWG, 30, his emphasis). Ruth Finnegan observes: By now the diversities of oral literature are more widely recognized. Nor, contrary to what was once believed, does oral performance always emerge in the mix-and-match variability of composition in the moment of delivery. That is one form, certainly, famously attested in the Yugoslav heroic poetry studied by Parry, Lord, and other scholars in the ‘oral formulaic’ tradition. But it has now become clear that oral literature also includes cases of prior composition and of exactly repeated delivery (How of Literature, 168). For examples of verbatim transmission of oral-traditional materials see, e.g., Finnegan, Oral Poetry, 73–86; Goody, Domestication, 27; Jensen, Homeric Question, 40; Ong, Orality and Literacy, 62–64; Tedlock, Spoken Word, 235n5; Sherzer, Verbal Art, 240n1; Will, Oral Memory; Chesnutt, Orality. This point is explicitly taken by Riesner (see Jesus as Preacher, 207–8).

27. See, e.g., Vansina, Oral Tradition, 146–47; Finnegan, Oral Poetry, 135, 151.

28. Thus Jan Vansina: There is no specific ‘oral mentality’ (Oral Tradition, 218–19n53). See also e.g., Halverson, Goody; Finnegan, What Is Orality; Rosenberg, Folklore and Literature, 25–26; Street, Literacy.

29. So for instance Ruth Finnegan notes that "in performance [oral poems] are all subject at times to variation and adaptation—and perhaps it is this opportunity rather than actual variation that is ‘typical’ of oral compositions" (Oral Poetry, 79, her emphasis). However, similar opportunities are present in the production of written documents. Plausibly the salient differences between oral and written texts in general with respect to variability are to do with the respective characters of the storage media and retrieval mechanisms involved; i.e., with the distinctive properties of human memory on one hand and the artifacts of literary production on the other.

30. So Vansina, Oral Tradition as History, 42: The intent of the performance with regard to the faithfulness of the message it contains must be investigated for every separate circumstance. Whether innovation is appreciated or to the contrary word for word delivery is required should be investigated for every type of tradition in a given culture. Moreover, whether or not the intent to be faithful succeeds, and to what degree, will appear from the variability of the versions collected at various performances. Cf. Burch, More on the Reliability, 131.

31. So see, e.g., William Farmer, Synoptic Problem, 201: It is probable that there were other written sources and some kind of oral tradition also available to each of the Evangelists . . . but the use of hypothetical written sources (and/or oral tradition) by the Evangelists is not the best way to account for the major phenomena of similarity . . . among Matthew, Mark, and Luke; Christopher Tuckett, Q, 3–4: The theory of the existence of a Q source arose as part of the 2ST to explain the agreements between the texts of Matthew and Luke which were not to be explained by common dependence on Mark. These agreements are often so close, amounting at times to almost verbatim agreement in the Greek texts of the gospels (cf. Q 3:7–9; 11:9f.), that some form of literary relationship seems to be demanded; Goodacre, Synoptic Problem, 80, 16: It is a fundamental assumption of the study of the Synoptic Problem that the first three Gospels share some kind of literary relationship. . . . This consensus is based on the fact that there is substantial agreement between Matthew, Mark, and Luke on matters of language and order. The argument presented in each of these three quotations is implicit, but clear: there is a lot of similarity among the Synoptic Gospels; but oral tradition could not produce such similarity; so there is a literary relationship among the Synoptics.

32. This ability is also prerequisite to Synoptic textual criticism as currently practiced. Bruce Metzger notes in his Textual Commentary that the editorial committee of the UBS Greek New Testament employed the assumption of Markan priority as a criterion for determining what the author was more likely to have written (13–14). By my count, the second edition of Metzger’s Commentary (on UBS4) contains at least 14 references to text-critical decisions made wholly or partially on that ground (down from around 20 such references in his first edition commentary on UBS3; see Tuckett, Minor Agreements, 128, drawing on Wheeler, Textual Criticism, 373): at Matt 9:14; 14:3; 19:29; Mark 1:40; 2:5, 16; 6:47; 10:19; 12:23, 36; 15:12; Luke 8:43; 9:1; 11:33; and perhaps also 24:6. Whether this amounts to a serious and debilitating bias (as alleged in Dungan, Synopses of the Future) or not (as maintained in Tuckett, Minor Agreements, 127–32), it certainly presupposes the editorial committee’s ability to distinguish literary redaction from reference to oral traditions.

33. See, e.g., Derrenbacker, Ancient Compositional Practices; Derrenbacker, External and Psychological Conditions; Downing, Compositional Conventions.

34. For an application of the principle of the drunkard’s search to research methods in the social sciences, see Kaplan, Conduct of Inquiry, 11ff.

35. John Miles Foley’s work on the composition, transmission, and performance of oral poetry (e.g., Immanent Art;