The Priority of St. Matthew's Gospel

Daniel J. Castellano

(2014)

Part III: Ancient Use of Oral and Written Sources
   3.1 Faulty Form Critical Views of Oral and Written Transmission
   3.2 Studies of Oral Tradition
   3.3 Orality in Modern New Testament Scholarship
   3.4 Interplay of Oral and Written Communication
   Conclusion to Part III: A Model of the Oral Gospel Tradition
   Footnotes to Part III

Part III: Ancient Use of Oral and Written Sources

The two-source hypothesis was developed in a scholarly milieu with presuppositions about the transmission of oral and written traditions that have proven to be inapplicable to ancient history. The ancients did not use texts at all in the same way as we would, as their culture was still dominated by oral modes of communication. A more careful examination of how the ancients actually used texts, combined with comparative studies of modern oral cultures, has yielded fruitful insight into how the first-century Evangelists may have used sources at their disposal. In many respects, these findings are at odds with most standard solutions to the Synoptic problem.

Following the general overview of the relevant literature given by Terence Mournet,[1] we will recapitulate the faulty views of oral and written tradition held by the early form critics, which informed their hypotheses about the literary dependencies among the Synoptics. Then we will review some of the more important modern studies of oral tradition, including the cognitive psychology of memorization as discussed by David Rubin,[2] with the hope of finding some models appropriate to first-century Palestine and Syria. We will also look at how ancient use of written sources was informed by orality. We will move beyond Mournet and Rubin by occasionally pursuing the primary sources in some depth, and also give an analysis of the likely modes of composition used by the Evangelists, in light of the historical and manuscript evidence given earlier.

[Top of page]

3.1 Faulty Form Critical Views of Oral and Written Transmission

3.1.1 Early Modern Critics
3.1.2 B.H. Streeter
3.1.3 Bultmann and Dibelius
3.1.4 W.R. Farmer
3.1.5 H. Koester
3.1.6 E.P. Sanders
3.1.7 B. Gerhardsson
3.1.8 Summary

Hermann Gunkel (1862-1932), the father of modern form criticism, adopted the diffusionist concept of folklore transmission that was promoted by the Grimm brothers. In this view, each retelling of a story diverges more or less from an original, pure form. Stories are transmitted as old people retell them to youth, and in each iteration, something is added, changed, or omitted.

While early form critics quite sensibly assumed that the Gospel tradition was originally oral, they misunderstood how oral tradition really works. The idea that there is an original, pure form from which various retellings deviate turns out to be generally false, as is the notion that oral tradition is transmitted linearly, from one generation to the next. Even today, though we should know better, the transmission of oral history is often compared to the child’s game Telephone, where a whispered phrase is passed linearly from one person to the next. It is no wonder that those who hold this model take a dim view of the reliability of oral transmission.

A further assumption found among early form critics is a linearity in changes due to transmission. Some assumed that oral or written traditions gradually gathered accretions over time, while others assumed a linear process of deletion or paraphrasing. Such assumptions of linearity informed their judgment of which readings were more primitive, though we now know this to be a faulty heuristic.

Although the assumptions of the form critics are known to be wrong, their major theses are still widely accepted. Even Mournet does not challenge the two-source hypothesis, though his criticism of form critical assumptions should invite us to do so. The common excuse, both here and in Old Testament higher criticism, is that sometimes the wrong clue may lead to the right result. This is true enough, but it degenerates into special pleading when we find that practically all the original clues and heuristic assumptions have turned out to be wrong. It is far more likely that the theories of the form critics are widely retained because of their academic legacy, not because they have by far a superior fit to all available facts.

Form critical assumptions regarding the Evangelists’ use of written sources likewise reveal an ignorance of ancient modes of composition and transmission. In so-called redaction criticism, it is supposed that the Evangelist has copies of texts written by other authors laid out before him, and that he transcribes from these in order to compose his own work. Such a supposition presumes a model which separates the process of creating Gospels and the process of copying them.[3] This model applies only to the post-Gutenberg era, where the roles of author and publisher are distinct.

Without a good understanding of how texts were actually used in antiquity, we cannot properly interpret evidence of literary dependence among the Gospels. Most scholars take a purely literary approach to the Synoptic problem, however much they may pay lip service to the role of oral tradition. They justify this on the grounds that the verbatim similarities among the Gospels are far too great to be accounted for by purely oral recollection. While admitting this fact, we may nonetheless point out that the ancient use of texts was also mediated by orality, and it is perilous to form hypotheses about the content of the Gospels’ sources (e.g., that Q is a single written document) without understanding how texts were used.

The faulty views about oral and written transmission just outlined have led to a predominantly literary approach to the Synoptic problem. Nearly all form critics, whether they advocate the two-source hypothesis or some less popular view, assume that the Synoptic Gospels are related to each other by literary descent. While they may disagree on the direction of dependence, they generally regard it as mediated almost exclusively by written texts.

Even when they allow some role for oral tradition, form critics have tended to use the terms and concepts of textual redaction. The notions of copying, deleting, appending, etc. only properly apply to fixed texts, and are inapt characterizations of oral transmission. As we will see, they are also frequently poor descriptions of how the ancients used texts. This tendency to conceive of all transmission in literary terms is betrayed by the oxymoronic term oral literature, which Mournet criticizes at some length. The mistake of such thinking is to regard orality and written texts as merely two different media, rather than distinct modes of communication and transmission.

[Top of page]

3.1.1 Early Modern Critics

Nineteenth-century German critics widely accepted the two-source hypothesis of the composition of the Gospels, following its persuasive defense by H. Julius Holtzmann (1863). In this view, the verbal similarities of the Synoptic double and triple traditions were most parsimoniously explained if our editions of Matthew and Luke relied on Mark and a written collection of sayings. The latter was at first identified with St. Matthew’s logia mentioned by Papias, but later became denoted more agnostically as Q (Ger. Quelle, source).

German higher criticism of the Gospels was eventually taken up by British scholars, notably at an Oxford seminar in 1894. Among its attendees was J.C. Hawkins, who held that Matthew and Luke must be in some literary relationship due to shared unusual words or phrases, and shared unimportant details. Examples of such phrases are en gennhtois gunaikwn (occurring only five times in the Septuagint) and ikanos ina (found nowhere in the Septuagint). Hawkins thinks such phrases would have been replaced by more common wording in oral tellings. [John C Hawkins, Probabilities as to the So-Called Double Tradition of St. Matthew and St. Luke, in Oxford Studies in the Synoptic Problem (Oxford: Clarendon Press, 1911), 99.]

Mournet notes the irony that Hawkins looked for lack of literary parallels to show that a phrase is unusual and therefore not oral in origin! [Mournet, p.40] Hawkins assumed that people would have no reason to retain such phrases in oral tradition.

This assumption has not been borne out by later scholarship. As J.A. Draper has noted, oral tradition often uses metonymic markers. These words are retained by the oral tradition precisely because of their peculiarity, because they mark out a particular discourse unequivocally. [Jonathan A. Draper. The Sermon on the Plain (Luke 6:12-7:17) as Oral Performance, unpublished correspondence, 2000, cited by Mournet on p.40.] Draper finds peculiar phrases in Zulu praise poems, where epithets associated with particular historical figures are no longer understood but continue to be passed on verbatim because of their metonymic associations, i.e., the mere mention of the praise name conjures up the person even if its reference is not understood.

[Top of page]

3.1.2 B.H. Streeter

The two-source hypothesis assumed its modern form under B.H. Streeter (1874-1937), who acknowledged that the Gospels of Matthew and Luke also had their own unique sources besides Mark and Q, resulting in a four-document model. Streeter recognized that the older "two-source" hypothesis assumed everything could be explained by literary criticism, without studying historical conditions. Once it is acknowledged, however, that not all similarities are necessarily attributable to a common source text, it follows that there can be other sources besides Mark and Q. The probable use of non-textual sources makes it difficult or impossible to reconstruct Q, which in any case was written to supplement, not replace, oral tradition. Both the longer parables and the Passion story were easy to remember, and every one knew them… [B.H. Streeter, The Four Gospels: A Study of Origins (1924), p.229.]

Streeter considered it likely that Jerusalem and Caesarea, the two great Palestinian Churches, and Antioch, the original headquarters of the Gentile Mission, must each have had a cycle of tradition of its own. [Ibid., p.230.] These were church-based cycles, which continued even after the composition of the Synoptics. Despite this lip service to the importance of oral tradition, Streeter does not really incorporate orality into his theses.

[Top of page]

3.1.3 Bultmann and Dibelius

Rudolf Bultmann (1884-1976) and Martin Dibelius (1883-1947) are regarded as the cofounders of the form critical movement in New Testament studies. Bultmann defined form criticism in Synoptic Tradition (1931 ed.): the aim of form-criticism is to determine the original form of a piece of narrative, a dominical saying or a parable. In the process we learn to distinguish secondary additions and forms, and these in turn lead to important results for the history of the tradition.

Bultmann assumed that there is an original form to be distinguished from later accretions, thereby supposing a linear notion of transmission. The methodology of form criticism thus contains presuppositions about how content is transmitted, but this is something we ought to discover, not assume. Further, Bultmann's approach reduces history to a result that is posterior to literary criticism. This is the opposite of a historical method, which starts from actual testimonies or witnesses.

Form criticism also aims to rediscover the origin and intent of tradition, before it took written form in the Gospels. Dibelius (1919) began with presuppositions about the form of the early Church, informed by a Protestant low church ecclesiology. Dibelius thought traditions were handed down in isolation at first, and only later spread more broadly. He makes no reference to apostles and bishops as authorities of a more centralized tradition, as we find attested in history. Accordingly, he holds that the individual stories in the Gospels were originally independent. While it is reasonable to suppose that the Gospels drew upon various streams of tradition, we now know that oral stories are actually recounted in blocks. Importantly, Dibelius thinks that the development of a tradition involves growth and expansion.

Bultmann likewise thinks we must first account for the intent of individual units of the tradition. He also sees a tendency of expansion of an original saying by addition. It is not clear how he can discover this tendency to expansion from a purely literary criticism of the Synoptics, without presupposing the nature of their relationship. Like most early form critics, Bultmann thinks it relatively unimportant whether tradition in the Palestinian church was transmitted orally or by text.

Like Dibelius, Bultmann is informed by some contrived a priori views about ecclesial development. He discerns Palestinian and Hellenistic stages of tradition, attributing the miracle stories to the latter.[4] This is unsupported by historical and external manuscript evidence, as we find miracles in some of the earliest extra-canonical documents, including the Egerton Gospel. Bultmann accounts for the absence of evidence by saying that the editing of tradition began before it was in written form. Editing, however, is a textual term, and it is unclear how the concept should be applied, if at all, to oral tradition.

Both Bultmann and Dibelius think there are laws of transmission whereby Church traditions develop in some predictable fashion. This is a linear process, whether oral or written. These form critics retain a nineteenth-century way of thinking, assuming that everything follows natural, deterministic laws. A field of study became scientific by explaining everything in terms of such laws. Also note that they often try to deduce these laws from the results of form criticism, rather than study actual societies and historical testimonies about modes of transmission.

The early form critics thought that development is in the direction of expansion and creation of new material, and connecting previously disconnected material. As Mournet notes, this account is problematic if in fact there is an oscillation between oral and written tradition, rather than a linear development from oral to fixed written form. When they bother to address oral tradition, they misunderstand how it works, and consequently dismiss it as an unimportant side issue.

[Top of page]

3.1.4 W.R. Farmer

A purely literary mode of thinking is found in the work of W.R. Farmer (1921-2000), who followed a version of the Griesbach hypothesis upholding Matthaean priority. He held that Luke used Matthew, while Mark used Matthew and Luke. He considers that the degree of verbatim similarity among the Synoptics proves that they are in some relation of literary dependence, and then explores the eighteen possible sets of relations. In his analysis, he constantly speaks in terms of one source copying from another. He assumes that the literary dependence must have the form of copying or editing, and that there is direct access to the other Gospel text. He further assumes that they are related to each other in linear terms.

Farmer’s appearance of having exhaustively evaluated the possibilities is misleading, since he presumes a particular mode of literary dependence. He gives little consideration to the use of oral tradition as a distinct mode of transmission of source material. Further, he assumes relations of simple linear dependence on the grounds of parsimony, but this is not necessarily more parsimonious than a more complex relationship among the Gospels, Mournet notes, if the latter better fits the evidence. On the contrary, it is highly likely that Evangelists had access to living oral tradition.[5]

[Top of page]

3.1.5 H. Koester

We have already shown evidence from the Apostolic Fathers and non-canonical gospels that Christian oral tradition remained active until about AD 150. This concurs with the judgment of Helmut Koester (1926- ), who found that these sources were not linearly dependent on the canonical Gospels, but also drew upon parallel streams of oral tradition. He infers, sensibly enough, that the Synoptics themselves drew upon similar oral traditions.

Koester holds that most of the Apostolic Fathers stand on the same level as the Synoptics with respect to oral tradition. The only exceptions are 2 Clement and Didache, which show some dependence on the Synoptics, direct or indirect. This assumption of parity is evidently informed by Koester’s poorly founded belief that the distinction between orthodoxy and heresy only arose during the time of St. Irenaeus. We have seen that this contradicts early testimony about St. Polycarp and St. John, not to mention the repeated mention of false or foreign doctrines in the New Testament epistles. Here we see how literary criticism can be misled from the start by a false ecclesiology. Koester, like many critics, assumes that at first no versions of the Gospel or Christian doctrine were more authoritative than any other, ignoring the unanimous early testimony pointing to the distinctive authority of the Apostles. Indeed, 2 Clement itself cites the [O.T.] books and the Apostles (14:2) as authorities.

[Top of page]

3.1.6 E.P. Sanders

E.P. Sanders (1937- ), in The Tendencies of the Synoptic Tradition (1969), criticizes earlier form critics for fitting the evidence to presupposed laws of transmission without testing their hypotheses. Using rabbinic sources, Sanders finds that there is no law of expansion or abbreviation. Sometimes a piece of text is expanded; other times it is abbreviated. There is no general tendency. Accordingly, we cannot use the brevity or length of a particular reading as proof of priority. Further, he finds that similarities and differences among versions can result from processes other than lineal descent.

Nonetheless, Sanders believes that the less explicit version of a story is generally more primitive. Comparing the Synoptics, he finds:

Mt more explicit than Mk: 49 times
Mk more explicit than Mt: 23 times
Mt more explicit than Lk: 38 times
Lk more explicit than Mt: 26 times
Lk more explicit than Mk: 24 times
Mk more explicit than Lk: 21 times

Using Sanders’ assumption about textual dependency and priority, Mark and Luke would clearly be prior to Matthew, and Mark would be prior to Luke, though more closely followed. Following Goulder and Farrer, Sanders thinks that Matthew used Mark, and Luke used them both. There is no need for Q to explain the Matthew-Luke double tradition, though Sanders thinks there were some additional sayings sources.

Remarkably, Sanders approves Rudolf Bultmann’s statement that it is a matter of indifference whether the tradition were oral or written. (Tendencies of the Synoptic Tradition, 7) This naively supposes that the tendencies in transmission of oral tradition would be the same as for written texts.

[Top of page]

3.1.7 B. Gerhardsson

Birger Gerhardsson’s Memory and Manuscript, though written in 1961, did not engage the evidence presented in Lord’s The Singer of Tales (1960). Gerhardsson attempted to discuss oral transmission based on historical accounts instead of field study.

Gerhardsson postulated that Jesus formed a rabbinic academy of his disciples, who recorded teachings in notebooks as aids to memorization, as was done in rabbinic Judaism. Care was taken to preserve the ipsissima verba of the master.

This rabbinic model of transmitting the Gospel is open to several criticisms. First, the method described comes from the Tannaitic and Amoraic periods, several centuries after the Gospels, so the model is anachronistic. Second, even in the apostolic age, the Christian Church moved beyond its Judaic roots and became multi-ethnic. Third, rabbinic schools dealt with already existing sacred materials, whence their concern for preserving exact words, while Christian traditions arguably did not constitute a fixed sacred text until later. This last point is weaker than the first two, for we have no way of knowing that the Gospel material was not considered sacred from the beginning, and on the contrary, the earliest testimonies on the matter suggest that it was.

The fourth and strongest criticism is that there is no evidence that Jesus ever formed such a structured school. The rustic background of most apostles would argue against it. They may even have been illiterate, though the term agrammatoi [Acts 4:13] need not imply absolute illiteracy, only a lack of formal education. Further, as Pieter Botha notes, the Gospel of Mark is not high literature, but uses a plain, unpretentious style. It has no liturgical elements and no indications of a format. [P. Botha. Mark’s Story of Jesus and the Search for Virtue in The Rhetorical Analysis of Scripture, 1997.]

Like earlier NT scholars, Gerhardsson does not perceive much difference between oral and written modes of transmission. He assumes that memorization operates with the same aim of preserving a fixed oral text.

[Top of page]

3.1.8 Summary

The early form critics’ understanding of oral tradition is clearly deficient, and it is not altogether certain that more recent critics have made effective use of what has been learned. The arguments for literary dependence among the Synoptics rest mainly on near-verbatim agreement in many pericopes, agreement in the sequence of materials, and unusual phrases shared in double tradition. We have seen that the last element is not proof of literary dependence, and we will also find that agreement in sequence need not point to a written source. This leaves only verbatim similarity, which we will find is not nearly as extensive or consistent as commonly assumed.

Once it is accepted that many of the Synoptic agreements do not exclusively indicate literary dependence, there is far less reason to insist on the two-source hypothesis. A multiple-source hypothesis, with emphasis on oral tradition, must also be considered.

[Top of page]

3.2 Studies of Oral Tradition

3.2.1 Features of Oral Tradition
3.2.2 Parry-Lord Theory of Oral Epics
3.2.3 Transmitting Oral Tradition

In the last half century, studies in cultural anthropology and cognitive psychology have brought much more light on how oral traditions are composed and transmitted, discrediting the assumptions of earlier New Testament scholars. A synthesis of these findings has been presented by David Rubin in Memory in Oral Traditions (New York: Oxford Univ. Press, 1995). While some aspects of oral tradition depend on genre and culture, the following general properties have been identified.

Oral traditions

The fact that oral traditions are always narratives is especially pertinent to the Gospels. It suggests that the oral Gospel would contain narratives, such as the parables and deeds of Christ, contrary to the hypothesis that miracle stories were added later. In oral tradition, sayings and aphorisms are typically preserved in isolation, not as sets. Thus the organization of Gospel sayings would likely be a literary endeavor, undertaken either by the Evangelists or a written source. In fact we find that the narrative portions of the Gospel follow mostly the same order in the Synoptics, while there is great variation in the organization of sayings. All this is consistent with a primarily oral origin for Synoptic source material.

We are primarily concerned with stability and variation in oral tradition, since Synoptic analysis depends on how we interpret similarities and differences among the Gospels. The stability of a tradition depends on its genre. In epic poetry, the basic narrative is stable, but not the descriptive details, as singers are allowed poetic license here, without being considered to have falsified the story. Simple poems are highly stable even in their wording, since the constraints of meter and rhyme limit possible word choices, making recall easier. Ballads organized into stanzas have an intermediate level of stability, often preserving descriptive details and wording, but still with considerable latitude in retellings. Any assessment of the stability of the oral Gospel must attempt to identify its genre.

Transmission of oral tradition is not accomplished by verbatim recall, but by retaining the general meaning and form of a narrative. There are no documented cases of verbatim recall over fifty words long in oral tradition without writing. [Rubin, p.6] The lack of verbatim recall is not a liability, for in fact variability within limits aids the long-term stability of a tradition. Each singer can adapt the piece so that it will be easier for him to recall, and so that it is easily recalled by his audience, by following certain cultural norms. Contrary to the assumptions of early form critics, there is no single normative text or specific variant being transmitted, which is to say there is no primitive version or Neutral text. [Ibid., p.7]

Oral traditions are best considered as human behavior, rather than as fixed objects, such as a reified text. An oral tradition is recounted as a performance, and the rules of performance are pertinent to the telling. Any understanding of how oral narratives are composed and transmitted must take such rules into account.

[Top of page]

3.2.1 Features of Oral Tradition

Scholars of oral tradition have identified several features that help shape the composition and transmission of narratives. These include theme, scripts, story grammar, imagery, and sound patterns. [See Rubin, pp. 12-116]

A theme consists of a standardized list of actions and character attributes that are expected to be recounted in a given genre. This enables the singer to organize content and sequence more easily, without having to memorize a long story. In certain epics, for example, it is common to recount the noble ancestry of the hero, followed by some dispute in an assembly of elders, consultation with a mentor, and finding a patron, before embarking on some journey. [See Rubin, p.16 for fuller list.] A particular story of this type may omit one or more thematic elements, but familiarity with these helps the singer avoid omitting anything essential or telling the story out of order. The structure of a theme is more sequential than causal, being connected by a series of conjunctions such as and then…. This is exactly what we find in Gospel narratives, with the repeated use of a leading kai (and), believed to be a translated Hebraism.

Theme appears to be a special case of what is known in psychology as a schema, which F.C. Bartlett defined (1932) as an active organization of past reactions. A singer, by following a theme, allows each retelling to be informed by his and his audience’s attitude to past performances. Certain expectations develop about how a story should be told, and conformity to such expectations aids recall and adds stability to the tradition in order and content. If there are changes in recall over time, these will tend to make the story more like the schema, so we might regard non-conforming variants as more likely to be older. Further, the singer has discretion to omit parts that the audience already knows, since they are in the schema. Thus the existence of shorter versions does not prove the omitted parts are later additions.

Schematic or thematic conformity helps stabilize an oral tradition, but may be a liability in the preservation of accurate history. If we are dealing with a largely imaginative narrative, there is no problem with making it conform to a schema, since what should happen thematically coincides with what does happen. History, by contrast, is not so cooperative, so we should have reason to suspect the historical accuracy of elements that conform too neatly to a scheme, especially in oral traditions that are centuries old. This is less of a concern with the New Testament, which was committed to writing at an early stage, but we find a literary equivalent with later textual variants that attempt to correct irregular or difficult readings. From what we know of later oral tradition among Jewish Christians and Gnostics, it would seem that this too was modified to conform with the attitudes of the audience.

On a smaller scale, we find that parts of oral narratives are structured by scripts, which are stereotyped sequences of actions. We follow scripts in many routinized actions, such as ordering a meal at a restaurant or going to the dentist, so that merely mentioning such a task suffices for us to specify an entire sequence of events. Ancient peoples would have been familiar with other scripts, some of which became standardized in their oral narratives. A common example is the process of a hero arming for battle, which follows a generally fixed sequence, likely based on reality. The singer does not need to memorize the entire sequence as a story-specific text, but only needs to recall that the hero now arms himself, and then follows the standard script well known to all.

Narratives are also structured by a story grammar, which further aids recall. A story consists of a setting and an episode within that setting. We see that the Gospel episodes usually begin with a declaration of setting, such as a description of location, or the context of some proximate event or occasion. Episodes with action general consist of an attempt and attainment of some goal. In oral traditions, there is usually one such attempt or three. The use of three attempts helps build narrative tension before fulfillment, and may also be related to the well-known rule of three that aids memorization. When relating history, the number of actual attempts may be modified to fit this structure. In the case of the Gospels, no such artifice is needed, as Jesus never fails to accomplish his intent on the first attempt.

The various elements of a narrative, be they elements of themes, scripts or story grammars, are more easily recalled when they are associated with each other. Analysis of epics shows that certain thematic or schematic elements are more strongly associated with each other; i.e., stories containing one element are much more likely to contain an associated element. [Rubin, p.31] Parts of a story can be confused with each other, however, if they both fulfill the same thematic function. [Ibid., p.36] In such cases, one variant can supplant or replace the other.

Oral traditions are concrete rather than abstract in content, so they rely heavily on imagery. Modern psychology has shown that there is likely an imagery system distinct from the verbal system in the human mind. [Rubin, p.42] Images serve as analogs to external reality, with transformations of images corresponding to steps in the external world. In the Homeric epics, for example, warriors are depicted as moving or walking forward after each single combat. Imagery is spatial rather than sequential, as space represents sequence in action. Spatial location thereby serves as a cue to recall of sequence. [Ibid., p.46] Epics are structured by a hero’s travels, and in fact the Greek word for epic (οιμη) means path.

In the Gospels, narratives are frequently ordered in a path of known geographic locations. Sometimes this ordering is clearly artificial, as in Luke’s travel narrative, which is used to organize much of the sayings material. The synoptic apocalypse also uses spatial imagery, advancing the narrative as Jesus and the Apostles exit the Temple and its complex.

Imagery is good for advancing a narrative, since it is easier to detect changes in the concrete than in abstractions. Imagery, by its concreteness, can specify distinctive relationships, as opposed to the vagueness of abstract relations (e.g., Truth is good). Psychological experiments show that imagery with interacting objects (A X-ing B) is more memorable. [Rubin, p.54]

The imagery system is the same as that responsible for perceiving images in the first instance, as is shown by the fact that both perception and image-construction suffer deficits from the same injuries. Thus composing narrative imagery is functionally similar to the act of perception, a fact that is confirmed by the accounts of traditional singers. They use images to construct and to remember their songs. When performing, they need to see the story as a picture, then just follow the action and describe it verbally, as if they were watching a movie. [Ibid., p.59]

Oral narratives are also structured by sound patterns, called surface schemas, which include poetic elements such as rhythm, rhyme, alliteration, and assonance. Surface schemas are at least as important as thematic content in aiding recall.

The fact that all known oral traditions are sung might seem to imply inapplicability to the Gospels, but the Gospels have been sung in all the oldest liturgical traditions, and there is evidence of rhythmic structure in some sayings of Christ, when the Aramaic original can be reasonably inferred. Further, it is not necessary for all poetic elements, especially rhyme, to be present in every oral tradition.

Rhythm is the most ubiquitous sound pattern in human speech, found even in non-poetic orations and preaching. It fulfills an important function by integrating the units of speech. Speech is uttered sequentially, without the visual advantage of writing that enables construction of logically integrated long sentences. Accordingly, its idea units tend to be shorter, usually confined to a single intonation, typically a noun or verb phrase. While writing can use devices such as prepositional phrases and parenthetic clauses to integrate long sentences as a complex idea, speech merely integrates short units in sequence, accomplishing this by rhythm. One thing follows another as an integrated unit because they are joined by rhythm. Thus each rhythmic phrase constitutes a unit of memory, so we should not be surprised if variant tellings should preserve near-identical phrases in different order.

Rhyme, alliteration, and assonance also perform an integrative role, creating links between pairs of units. They assist memory by imposing constraints on the possible words that can be placed in a slot. Each of these elements acts as a cue to memory.

The various structural features of oral tradition act as cues to memory much more effectively in combination. With as few as two cues (e.g., meaning and rhyme), a high success rate in recall can be achieved. Other memory aids include organizing items in groups of three, and making gestures associated with language.

The constraints of oral traditions, especially surface schemas, induce a tendency to repeat words and syntax. This is congruent of a more general human cognitive tendency to repeat words recently produced. Repetition is strongly characteristic of oral composition, which makes it ironic that the old higher critics invoked duplications in Biblical text as evidence of literary redaction.

[Top of page]

3.2.2 Parry-Lord Theory of Oral Epics

The form critics’ deficient understanding of oral tradition as a distinct mode of transmission began to be remedied after Albert Lord’s The Singer of Tales was published in 1960. There had been earlier studies of oral traditions, but it was Lord’s work that first gained the serious attention of New Testament form critics.

Subsequent New Testament scholarship has been less naive about the differences between oral and written tradition. Shortly after 1960, the dominant paradigm of oral composition was the Parry-Lord theory, based on studies of singers performing South Slavic epics. The first such study was conducted by Milman Parry, who concluded that the Homeric epics were not originally composed ex nihilo by a literary genius, but were performed by an oral poet. (Mournet, p.68) Homer was not a literary author, but an oral composer working within a long-standing oral tradition. He stitched together prefabricated parts; only a small percentage of words were not parts of formulas necessary to oral composition. As Walter Ong said of Parry’s thesis, features of Homeric epic were due to the economy enforced upon it by oral methods of composition.

Parry died in 1935 after conducting only one field study in Yugoslavia, but his work was continued for decades by his student Albert B. Lord. In The Singer of Tales (1960), Lord compared Serbo-Croatian epic composition to Beowulf, The Song of Roland, and eventually the Synoptic Gospels. His findings are now widely used for studying oral epic composition throughout the world, and have overturned the supposed laws of transmission articulated by early form critics.

Oral epic performance is not the delivery of a memorized text, but instead a process of traditional formulas using mnemonic devices are presented by reflexive action. The singer uses set phrases to help in the performance of a given song. He draws upon these phrases to compose his song creatively as it is performed. Detailed content is not prepared in advance. Music helps him maintain rhythm and flow.

Some of Lord’s findings may be specific to the epic genre, but others have proven to have broader applicability. Among the latter, Lord challenges the notion of preserving ipsissima verbi, finding that word for word rendering is a foreign notion in oral cultures: …oral traditional narrative… has no fixed original. [Lord, The Gospels as Oral Traditional Literature, p.37.]

As an illustration of this principle, a singer told Lord that he could hear a song for an hour and then give every word and not make a mistake on a single one…. He claimed he could give the same word for word rendering twenty years from now. In fact, neither he nor other singers ever rendered the song word for word as we understand verbatim, since they did not have our text-based notion of words or lines. Their notion of a word is a complete speech-act, e.g., ‘he was drinking wine.’ (Lord, The Singer of Tales, pp.26-28) These bigger words are their units of composition. This agrees with what we find in ancient history. The Hebrews regarded the Ten Commandments as ten words, and the Greeks considered a rhema, which we might call a verb phrase, to be a word or unit of speech. The Latin verbum can also mean a saying or phrase rather than what we would call a grammatical word. In such cultures, a word for word rendering means something quite different from our understanding.

Recall of an epic does not result from the impossible task of memorizing a long text. When Salih Uglajanin sang the Song of Baghdad three times to Lord, it was evident that he had not memorized the passage.

He remembered, unconsciously, the elements that make it up and to some extent, the order of the elements…. Not memorized, not improvised either, not even exactly repeated either, but presented in more or less the same words, while expressing the same essential ideas. The text is not really fixed, yet because the essential ideas have remained constant, it is more or less fixed. [Lord, Memory, Fixity and Genre in Oral Traditional Poetries, in The Relationships among the Gospels: An Interdisciplinary Dialogue, ed. William O. Walker (San Antonio: Trinity Univ. Press, 1978), p. 453.]

This degree of variation resembles what we find in Synoptic parallels of a given pericope. There is usually not exact verbatim agreement, but many identical phrases, and the same significant elements, generally in the same order. If the Synoptics give different versions of an oral tradition, then it is generally meaningless or practically impossible to determine priority. We cannot say which written version of an oral tradition is more primitive, since there is no original fixed text.

Lord challenged the heuristic of previous literary scholars, who assumed that the longer version of a text is always later. With oral tradition, there can be expansion or contraction in subsequent retellings, so we cannot assume that a more detailed version is later.

Similarity of order within parallel Synoptic passages has been invoked as evidence of literary dependence, yet Lord demonstrated that the primary unit of composition was not an isolated saying, but a block or sequence of material. [Lord, The Gospels as Oral Traditional Literature, op. cit., p.59.] Units were interconnected from the beginning. Proverbs and sayings were always incorporated in blocks of material. [Mournet, p.73] Unsurprisingly, Lord concluded that Synoptic parallels were consistent with common use of oral tradition, without need for direct literary dependence on one another.

[Top of page]

3.2.3 Transmitting Oral Tradition

Oral tradition is preserved much more reliably than one might expect from the limitations of human memory. This is accomplished by techniques and modes of transmission that minimize the burden of memorization and provide safeguards to correct and stabilize narratives. Importantly, transmission is not simply serial, where one person relays a message to another, and the second person to a third, successively degrading the message. Instead, traditions are mediated by repeated performances in groups, following culturally specific norms.

The early experimental studies of oral recall by F.C. Bartlett used a model of serial transmission. Still, some of his findings match what is found in oral tradition. The titles of narratives tend to be lost quickly in oral tradition, much as we find in the serial transmission experiments. Likewise, both experimental and field studies show that the narrative tends to express more conventionalized opinions over time. In oral cultures, this means the narrative expresses those beliefs that are accepted by the group implicitly without questioning. In serial transmission experiments, proper names are quickly forgotten, but this is not the case in oral traditions, which have the safeguard of repeating names. [Rubin, op. cit., p. 131]

Oral tradition is not passed in a simple linear chain. Each singer has heard a tale several times, and in different versions, from groups of people. This variation actually enhances stability of the basic narrative. Variability within limits encourages the abstraction of patterns at recall. Audiences know which elements are essential to the narrative, and will correct someone who omits or falsifies such an element. Sometimes audience feedback is more subtle, as when responses from a congregation aid a preacher’s timing. [Rubin, p. 135.]

There are no conscious rules in learning to sing an oral tradition. Singers only need practice, with no formal training beyond repetition. A song version is composed swiftly, without advance preparation. The singer is not aware of composing anything, but thinks of himself as simply relating the story. The more prolific and talented singers tend to make much longer versions of the narrative, while believing themselves to be telling the story as it happened, in exactly the same form as they had learned it. [Rubin, p. 139]

The epic singers’ strange confidence that they are exactly recalling long narratives is mirrored in experimental studies. Although people cannot accurately recall sentences more than one idea long, they are actually more confident about their recall of longer sentences. [Rubin, pp. 140-41] This confidence is evidently grounded in the significance of the narrative, rather than its verbal form.

Recall can be frustrated by proactive or retroactive interference with similar content. Cueing can help release the memory from such interference. Recall is greatly enhanced by the number of cues, with dramatic improvement when there are three cues.

A running start helps oral recollection, especially at the start of a segment (e.g., a stanza or a sentence). Cues such as alliteration within a segment can decrease the importance of recalling the beginning. Rhythm can also help recall of fragments within the middle of a piece, as in parts of Psalm 23. [Rubin, p. 185]

Formulaic scenes, such as arming for battle, are almost always presented in the same order. This aids recall and transmission, since one does not need to memorize the order of actions for each story, if one only knows the standard formula.

When there is personal combat, there is always a single protagonist or a single antagonist, never a pair fighting a pair. This formula is also found in ballads. In The Murder of Poor Pearl, over time the story was simplified from two murderers to one murderer. [Rubin, p. 281] This may have important implications for the Gospels. In the past it was argued that the two demoniacs in Matthew as an amplification of the more primitive Marcan narrative, but from what we know of oral narrative conventions, it is more likely that the Matthaean version is older.

In oral tradition, a story’s structure is not revealed at once. The serial nature of speaking makes us focus on local meaning, since we do not have the luxury of scanning a text at once. Thus high-level structure is not nearly as important to oral recall and transmission as local meaning. Small-scale structures, such as rhyme, theme, scripts, and others discussed, aid recall, not only in isolation, but in feedback with the other factors, imposing sufficient constraints to narrow the range of possibilities.

Among these factors, rhythm or metre is central to the recall and transmission of oral tradition. Where there are oral sources to the Gospels, we should expect to find evidence of some metre, either in the Greek or in the inferred Aramaic original.

The study of epic poetry and other genres have shown that the basic story of an oral narrative remains stable indefinitely, even over periods spanning centuries. This is because there are cultural controls on traditional recitations that forbid alteration of essential elements, while permitting latitude and others. Since traditions are maintained by the community, not just the individual singer, there is no danger of losing track of which elements are essential.

Lord found that some South Slavonic singers could recite epics of Homeric length. This is not due so much to a prodigious memory, but to a mastery of the mnemonic devices of composition, combined with repeated practice. There is no obstacle then to admitting that the Iliad and Odyssey really are a particular singer’s recitation of traditional epics, committed to writing. As noted, the especially good oral poets tended to create lengthier versions.

Still, we should not be misled by the length of such epics to suppose that vast quantities of data can be preserved without writing. Most of an oral narrative is not memorized, but freely composed, with only some basic elements retained. The capacity of human memory greatly limits what can be preserved without writing. Supposing a singer can commit several dozen or even a hundred narratives to memory, this is still a small library in comparison to what literacy makes possible. Oral cultures must be highly selective in determining which stories shall remain in their canon, while others are neglected after a generation. The stories that are retained tend to be important to community identity.

[Top of page]

3.3 Orality in Modern New Testament Scholarship

3.3.1 C.H. Lohr (1961)
3.3.2 Albert Lord (1978)
3.3.3 W.H. Kelber (1983)
3.3.4 J. Dewey (1989)
3.3.5 P. Achtemeier (1990)
3.3.6 Ø. Andersen (1991)
3.3.7 K.E. Bailey (1991)
3.3.8 R. Horsley and J. Draper (1999)
3.3.9 J.D.G. Dunn (2000-03)

Terence Mournet [op. cit.] gives an overview of how the science of oral tradition has been applied to modern New Testament scholarship. We recapitulate and expand on his discussion here.

[Top of page]

3.3.1 C.H. Lohr (1961)

One of the first attempts to apply Lord’s theory of oral tradition to the Gospels is found in Charles H. Lohr’s Oral techniques in the Gospel of Matthew (Catholic Biblical Quarterly 23, 4, Oct. 1961, pp. 403-35). Mournet finds that Lohr still thinks in textual terms, so that the oral composition of Matthew’s sermons really resembles redaction. This misunderstands the actual process of oral composition discussed previously.

[Top of page]

3.3.2 Albert Lord (1978)

Lord himself applied his theory to the Synoptic Gospels in 1978, finding that they are three oral traditional variants of the same narrative and non-narrative materials. (The Gospels as Oral Traditional Literature, p.90) He started with the following suppositions of source criticism. First, correspondences in sequence between Synoptics indicate that two evangelists recognized this material as a unit. Fragmentation in sequence indicates that one of the three writers knows the elements, but not them as a unit.

One would be tempted to conclude that the relationship among the Gospels was one of written documents were it not for two decisive phenomena. First, there are many instances where elements of a sequence are scattered sporadically in one or more Gospels. Second, there is less verbal correspondence than I would expect in a manuscript tradition. I find it unusual for a writer to choose passages from several documentary sources as if from a buffet.

These last points are especially strong. There is abundant evidence of manuscript traditions lasting centuries, and these have far less verbal variation among them than is found among the Synoptics, written within decades of each other. Lastly, even modern writers would find it cumbersome to compile from written sources as if from a buffet. This would be vastly more difficult for an ancient writer, who could not conveniently lay out various texts side by side on a writing desk. (This criticism also applies to the Old Testament documentary hypothesis.) This does not preclude the method mentioned by Philo, following a single written text and then interweaving what one knows from oral tradition.

Apparent duplications (variants of same story) are characteristic of oral tradition, according to Lord. A singer may forget that he has already told a certain story element, or how he told it previously. The two tellings may be inconsistent, but this does not matter as long as they are within the crea tive license granted to the reciter. For those concerned with Scriptural inerrancy, however, it would be unacceptable to claim that one version of a duplication asserts an erroneous fact, unless it is admitted that the detail in question is just color added to the narrative per oral traditional convention, with no assertion of historical exactitude. Additionally, we may note that not all apparent duplications really are such, but could be distinct yet similar occurrences.

Unlike most New Testament scholars, who try to distinguish the primitive account of Jesus from later additions, Lord holds that there is no fixed original version of any Gospel material. As with all oral traditions, each retelling is really its own composition. There would have been verbal variations in recitations from the very beginning. Some versions may be older than others, but a shorter form does not prove comparative age, so the critical method of excising lengthier versions as later is fundamentally flawed. Such critics unwittingly assume that the Gospel was originally a text, to which words might be added. Yet if the Gospel is oral in origin, it is senseless to judge Synoptic priority based on conformity to some supposed primitive text.

The transmission of oral tradition is highly genre-dependent, so we should not be surprised to find that proverbs or sayings have a much more stable verbal form than narratives. The strong verbal similarity among Synoptic sayings, often considered evidence of literary dependence, is only consistent with what is known of oral transmission.

[Top of page]

3.3.3 W.H. Kelber (1983)

Following the modern critical thesis that the Gospel of Mark is the most primitive written Gospel, scholars have been most receptive to the idea that there is evidence of orality in Mark, if not in the other Synoptics. This perception is amplified by Mark’s rough Greek, evidently the work of a native speaker of Aramaic or Hebrew. Mark seems to be a link between Aramaic oral tradition and the written Greek Gospels.

Werner Kelber has held that Mark did more than merely commit oral tradition to writing. Though his Gospel has some oral character, Mark’s writing manifests a transmutation more than mere transmission. [W.H. Kelber, The Oral and the Written Gospel (Fortress Press, 1983), p. 91] Kelber thinks that oral tradition has been transformed by being transplanted into written medium. Mournet remarks that this does not address possible interplay between oral and written tradition even after writing. Kelber supposes a discontinuity between oral and written phases, rather than overlapping interaction.

Strangely, Kelber thinks Mark wrote his Gospel as a deliberate countersign to oral tradition, thereby accounting for the lack of sayings material. On the contrary, all the historical evidence we have seen shows that the Gospels were written to preserve oral tradition, which was revered even afterward. Mournet suggests that Mark likely wrote to preserve oral tradition in the wake of the Jewish War, which disrupted traditional communities. [Mournet, op. cit., p. 84] The notion that Mark was against oral tradition is undermined by the necessity that his Gospel was circulated among the churches orally. Further, if his Gospel had a negative attitude toward oral tradition, why was it supposedly incorporated into Matthew and Luke, which both embraced the oral sayings traditions?

Alternative explanations are available for the lack of sayings material in Mark. For one thing, short sayings would be well known by all members of an oral community, unlike longer narratives, which might have relatively few reciters. There would be less need to put sayings into writing than narratives. Another possibility is that there were already written collections of sayings material. These might include Q, Matthew’s logia, the Gospel of Thomas, or their sources. Another consideration is the difference in genre. Mark may have simply been concerned with one genre of Gospel material rather than others. It is not usual for oral traditions to assemble sayings into collections, and Mark’s refusal to undertake such an endeavor would actually be evidence of preferring oral over literary modes of composition.

[Top of page]

3.3.4 J. Dewey (1989)

In contrast with Kelber, Joanna Dewey has found that the Gospel of Mark as a whole, not just its parts, shows evidence of oral composition. [J. Dewey. Oral Methods of Structuring Narrative in Mark, Interpretation 43(1):32-44 (1989)] Writing at that time was highly influenced by oral techniques of composition.

One distinctive example of such technique is Mark’s constant use of kai to introduce pericopes in synchronic, not chronological fashion. This type of narration is additive, not subordinative. The latter is characteristic of written composition, where you have a main topics, and subtopics, which makes sense when everything is visually arranged. This is not so in oral composition, which is just one thing after another, much like what we find in Mark.

The absence of chronological order in Mark is a consequence of oral compositional technique, not evidence of use of multiple sources. The same singer may recount a story’s episodes in different order, as long as the basic structure is retained. Once this oral character is recognized, our interpretation of classic source critical evidence is transformed.

The orality of Mark helps account for why it does not have a well-defined theme or narrative trajectory. There is no clear beginning, middle, or end. As literature, it is rambling, but as spoken words, it is rhythmic and natural.

The Alexandrian ending at Mark 16:8 seems abrupt only with our literary expectation that a conclusion ought to tie up all loose ends. Yet an oral evangelist might have been content to end his story with the empty tomb, much like Homer ends his Iliad with the funeral of Hector. Every Greek already knew that the Argives would sack Troy, just as every Christian knew that the risen Christ appeared to various disciples. Homer’s work was completed when he reached the end of Achilles’ personal journey in defeating and burying Hector. Likewise, Mark may have stopped when he reached the destination of Christ’s empty tomb. Here the messianic mission is completed, while the subsequent appearances serve to establish the faith of the Church that will promulgate the Gospel.

[Top of page]

3.3.5 P. Achtemeier (1990)

In Omne Verbum Sonat: The New Testament and the Oral Environment of Late Western Antiquity (1990), Paul Achtemeier amplifies Dewey’s finding of oral composition in the Gospels by noting that orality informed the entire process of reading and writing. In antiquity, scribes either took dictation or spoke their own words aloud as they wrote them. This process lent itself to oral modes of composition, with its additive, localized structure. There was little in the way of visually oriented large-scale editing, as is commonly used by modern authors.

Reading was also a predominantly aural experience, as the ancients read aloud even in private. St. Augustine read silently, but this was considered exceptional by his contemporaries. In this milieu, how words sound remained important. Content was organized by audible cues instead of visual cues, as there was no paragraph structure or punctuation.

[Top of page]

3.3.6 Ø. Andersen (1991)

As Dewey and Achtemeier noted that written texts can have oral characteristics, Øivind Andersen observed that oral tradition, conversely, might adopt some literary traits. [Oral Tradition, i: H. Wansbrough (red.), Jesus and the Oral Gospel Tradition, Sheffield (Sheffield UP) 1991, 17-58 (Journal for the Study of the New Testament, Supplement Series 64)

Andersen stresses the importance of recognizing that there are different levels of deviation from verbatim rendering depending on genre. Problematically, in a literate society, oral tradition may take on literary characteristics, including greater emphasis on verbatim rendering. This makes it practically impossible to discern oral and written modes of composition in a given writing, at least not at a local level, such a pericope. Large-scale literary structure would remain impractical for an oral account.

These findings raise new problems in trying to identify oral modes of composition in the Synoptics. Since the Gospels were written in a quasi-literate age, where both oral and written tradition were important, it is possible that Christian oral tradition had some literary qualities from the beginning. This is especially likely, considering that most early Christian communities in Palestine, Syria, and Egypt consisted of Jews, who already laid great stress on preserving holy scriptures with verbatim accuracy. Thus we should not be surprised if even the orally composed parts of the Gospels have much more verbatim rendering than is typical of oral traditions. Still, such recall would have to be within the limits of human memory.

[Top of page]

3.3.7 K.E. Bailey (1991)

Any discussion of oral tradition behind the written Gospels should address how that tradition was preserved or controlled. Gerhardsson’s supposition that Jesus’s disciples formed a formal rabbinic school is ill-supported by the historical evidence, yet on the other hand, it is highly implausible that the Gospel tradition was completely uncontrolled. All oral traditions consisting of important, community-defining content are recited and transmitted under culture-specific norms. Kenneth Bailey described such norms based on his decades of experience living in the Middle East. [K.E. Bailey, Informal Controlled Oral Tradition and the Synoptic Gospels. Asia Journal of Theology, 5 (1991), pp. 34-54.]

Bailey’s observations contradict Bultmann’s supposition that informal oral traditions are always uncontrolled, like unreliable rumors and legends. On the contrary, Bailey consistently found that important cultural information was preserved by what he called informal controlled oral tradition. It is informal in the sense of there being no well-defined teacher-pupil relationship, yet it is controlled, in that only some are permitted to recite the tradition, and the audience corrects anyone who recites it wrongly.

Bailey acknowledges that other modes of oral tradition are found in the Middle East, both informal uncontrolled and formal controlled. The former consists of rumors, jokes, casual news, and atrocity stories. These are transmitted with total flexibility, and have no stability in content. Formal controlled oral tradition, on the other hand, strives for stability in content, and is mediated by formal instruction. Bailey cites Taha Hussein’s 1932 autobiography, which recounts how the author learned the Qu’ran from Shaykh Sayyed, who had memorized the entire Qu’ran and the thousand couplets of the Alfyat Ibn Malik. This is akin to the rabbinic memorization described by Gerhardsson. Even in societies with writing, there was still seen to be special value in committing tradition to heart.

Informal controlled oral tradition occupies a middle spot, balancing flexibility and stability, depending on genre. This tradition is transmitted by recitation in community meetings, usually by an honored elder, or else someone especially competent at recounting a particular narrative. The level of flexibility depends on genre. Proverbs and poems cannot be recited with so much as a word out of place without being corrected by a chorus of voices. Some flexibility, however, is permitted in the recitation of parables and recollection of historical persons and events important to the community.

…here was continuity and flexibility. Not continuity and change. The distinction is important. Continuity and change could mean that the story-teller could change, say, 15 per cent of the story —any 15 per cent. Thus after seven transmissions of the story theoretically all of the story could be changed. But continuity and flexibility mean that the main lines of the story cannot be changed at all. The story can endure a hundred transmissions through a chain of a hundred and one different people and the inner core of the story remains intact. Within the structure, the story-teller has flexibility within limits to ‘tell it his own way’. But the basic story-line remains the same. By telling and retelling, the story does not evolve from A to B to C. Rather, the original structure of the story remains the same but it can be coloured green or red or blue. [Bailey, op. cit., p. 44]

Informal controlled tradition guarantees a thematic fixity even while various incidentals may change, in accordance with what we have noted earlier regarding the discretion granted to epic singers. We will revisit Bailey later in more detail, when developing a hypothetical model of the oral Gospel traditions.

[Top of page]

3.3.8 R. Horsley and J. Draper (1999)

Richard Horsley and Jonathan Draper, in Whoever Hears You Hears Me (1999), have criticised the assumption that Q is literary. The use of metonymic referencing, where a part stands for the whole, is characteristic of orality. Horsley suggests that Q was a libretto, whose performance referenced extra-textual tradition. That is to say, one recites a concise part of a story or teaching, and that reminds people of the rest, which is not in the libretto. Draper thinks we can find residues of oral performance in written text, using Hymes’ method of measured verse on Q 12:49-59. They think Q is orally derived text, though a text nonetheless. Still, it does not follow literary mode of composition, and the Q portions of the Gospels are not mere copying from a written document, but amplified by the oral tradition, resulting in the variations we observe.

[Top of page]

3.3.9 J.D.G. Dunn (2000-03)

Following on Bailey’s work, J.D.G. Dunn has shown that the Synoptic double and triple traditions have fixity and flexibility, much like informal controlled oral tradition. Variation within pericopes is indebted to this oral process where traditions change during retellings but remain within bounds set by communities. Variation within the Lord’s Prayer (Matt. 6 / Luke 11) is due to living liturgy of community worship. The evangelists’ only access to the prayer need not have been through a written Q. [T]he combination of stability and flexibility practically cried out to be recognized as typically oral in character. [J.D.G. Dunn, Jesus in Oral Memory: The Initial Stages of the Jesus Tradition, in Jesus: A Colloquium in the Holy Land (ed. Doris Donnelly, New York: Continuum, 2001) 84-145.]

Variation in at least some cases is "due to knowledge and use of the same tradition in oral mode, i.e., community tradition familiar to both Matthew and Luke. It is unclear if this should imply that both evangelists were familiar with the same local church, e.g., that of Jerusalem.

And even if a pericope was derived from Mark or Q, the retelling by Matthew or Luke is itself better described as in oral mode, maintaining the character of an oral retelling more than of a literary editing. [Dunn, op. cit., p. 128]

Dunn here allows that there could be a written Q, but this was used orally. It was remembered, or heard recited, and altered either because someone else had made another oral retelling, or because the writer himself composed in an oral mode.

These inquiries make clear that we must face another set of problems beyond how oral composition and transmission work. We must also investigate how oral communication is manifested in written texts.

[Top of page]

3.4 Interplay of Oral and Written Communication

3.4.1 Cognitive Impact of Literacy
3.4.2 Normative Role of Writing
3.4.3 Establishment of Written Canon
3.4.4 Relative Authority of Oral and Written Christian Tradition
3.4.5 Orality and Text in Jewish Tradition
3.4.6 Oral Performance of Christian Tradition

Jews in the time of Christ were exposed to literacy, yet they retained a high residual orality. The Gospels were composed during a sort of intermediate period, transitioning from oral to written tradition. We have seen from historical evidence that the oral Gospel tradition likely persisted until the mid-second century, though the Jewish War (AD 66-70) disrupted or destroyed many oral communities. From the late classical period onward, Christian revelation became predominantly Scriptural, though oral tradition still shaped liturgical practices, interpretations of doctrine, and the histories of the Apostles and martyrs.

Since the Gospels were composed in a milieu where both oral and written tradition played important cultural roles, we need to address how both modes of transmission interacted with each other. In particular, we should examine whether there are different mentalities involved in the two forms of communication.

[Top of page]

3.4.1 Cognitive Impact of Literacy

The Jesuit historian Walter Ong (1912-2003) famously held that there are substantial psychological differences between people in oral and literary cultures. Once a person is exposed to literacy, there is no going back to a purely oral mode of consciousness. Writing thenceforth transforms our consciousness even when we are composing orally. We arrange our ideas spatially, facilitating more complex conceptualization and quantification. Although many modern media, such as telephone, radio, and television, make use of the spoken word, this is a secondary orality, which is self-conscious and presupposes literacy.

In a similar vein, the classicist Eric A. Havelock (1903-88) had earlier held that the development of the Greek alphabet (c. 700 BC) gave rise to higher conceptual thought processes. The Greek, rather than the Phoenician, was the first true alphabet in his estimation, because it had a letter for every linguistic unit, with no implied sounds. Thus orality could be fully supplanted by visualization. The alphabet, not Herodotus, was the founder of history.

David Rubin, by contrast, in his discussion of the cognitive psychology of oral tradition, is much more reluctant to assert that writing affects how people think and remember. Historical findings of such transformations seem to conflate the effect of writing itself with all its concomitant developments, such as formal schooling and urbanization. Anthropological studies prove that illiterate people generally have no concepts of formal logic or categorization, but it is unclear that writing as such is responsible for these.

Controls for these effects were possible in the Scribner and Cole (1981) study in Liberia. The Vai people had oral traditions in their native tongue, yet some had formal Western schooling in English, while some had Koranic schools in Arabic that stressed oral repetition and memorization. Researchers found that English formal schooling improved logic, but not categorization. Urban experience improved categorization. Thus the presence of writing does not immediately transform the human psyche (if it can be said to do so at all). Rather, the cultural developments it facilitates make possible this change in the way of thinking.

Even then, however, human mental capabilities as such have not really changed. Our memories are no more or less prodigious than those in oral cultures. We simply now have more prostheses and aids to our memory, making us effectively far more capable. Still, the way in which we think linguistically is changed by writing. We now may use our visual perceptual system for language, so memory can be aided by visual cues rather than auditory cues. This is especially helpful for categorization.

Human beings remain naturally oral/aural in their thinking and communicating, as cognition is still mediated by the brain’s speech center. Writing should be viewed as a device or instrument; a mnemonic or cognitive aid, rather than an aspect of our psyche. Even today, writing remains the exception and not the norm of communication in most cultures, outside the rich industrialized nations. Walter Ong noted that, among the 3000 spoken language groups that have survived into the modern era, only 78 of these have produced literature in the common sense. We should examine, then, what reasons may induce a culture to take up the exceptional task of writing.

[Top of page]

3.4.2 Normative Role of Writing

The earliest writing was used to record transactions, so it apparently originated out of economic necessity. This did not result in the immediate production of literature, as there was no perceived need to use writing to replace oral tradition as a means of preserving histories, myths, and customs.

As people settled down in non-nomadic groups, the need for greater social organization arose, hence a need for laws. Babylonians, Assyrians, Sumerians, and Hittites all had written law codes, which dealt with commerce, physical violence, and interpersonal relations. The numerous commerce laws covered agricultural transactions, buying or renting fields, hiring a donkey, interest rates on loans, and loan repayment. Laws also provided restitution for physical violence, loss or damage of body parts, even a slap on the face. This was usually resolved with financial compensation, or with the lex talionis, as in Hammurabi’s code, which also dealt with inheritance, adoption, and treatment of slaves. Babylonian laws were written on tablets displayed for all to see, even though few could read.

The Old Testament was written during this period, and likely served the same function as other written codes, namely to organize and unify a people. The Torah, though written by God Himself, evidently referred to already existing customs. Thus it was not a set of brand new laws coming from nowhere, but codified existing oral traditions, now with a fixed form, backed by divine authority. Sanctity and fixity of law were interrelated, and the fixed permanence of written law more fittingly expressed the eternity of divine law.

Still, the presence of written law did not abolish the need for oral laws among the Jews, which were sustained in parallel through the Second Temple period. Likewise, the Greeks sustained oral laws even after some of their laws were written. Writing served to supplement and record oral laws, but oral laws were not thereby abandoned.

Plato repeatedly speaks of written and unwritten laws, equal in authority and similar in content. [e.g., Politicus, 295e-296a; Republic, 563d-e; Laws 793a-b] He says that written laws are just summaries of ancestral customs, and therefore normative. This seems to imply that the authority of written law comes from its agreement with oral law. Aristotle, for his part, advocates enforcement of both written and unwritten laws. [Ethics, 1180b; Politics, 1319b-1320]

Andocides (c.440-390 BC), in De mysteriis, holds that written laws protect against abuse, as they are equally applied to all citizens. In other words, having a well-defined norm, with no variants, guarantees equity. He goes so far as to say that magistrates should never enforce a law that is not written. Andocides is not here denigrating unwritten law. He is imposing a restraint on magistrates, so they are not partial, but apply the same laws to everyone.

Thucydides distinguishes between written laws, which are enforceable (i.e., through law-courts), and unwritten laws, which result in disgrace. Yet the unwritten laws are not lesser on that account, for disgrace could ruin a man’s life no less than an unfavorable court verdict. The unwritten laws were not absolutely unenforceable, for not all power was vested in state institutions. Civil society or the people had the power to ostracize and punish. Even in the supposedly legalistic Roman republic, most justice was handled privately. This is not an ancient version of the liberal idea that you can’t legislate morality, for the modern liberal presumes that all coercive authority is in the hands of the state.

Xenophon says there is an unwritten law that is recognized by all nations. Such a natural law is also alluded to by Plato and Aristotle, and later elaborated by Cicero and Christian philosophers. This natural law also influenced Jewish thought during the Hellenistic and Roman periods, and the laws of Noah were interpreted as binding all men. Demosthenes contrasts written law with unwritten law and common mores of humanity. [De Corona, 275] Aristotle associates unwritten law with natural law. Unwritten law is 1) universally recognized; 2) based on nature; 3) inexplicit; 4) never changing. [Rhetoric, 1373-75] Aristotle says there is greater merit in following unwritten law, since this is not under compulsion. Evidently, this unwritten law is not same as the oral traditions specific to each nation.

In short, writing made laws visible, accessible, and uniform. Nonetheless, oral tradition continued to exist side by side with written law. Where its domain overlapped with that of written law, the written law generally prevailed in courts, for the sake of uniformity. Still, oral tradition was the font from which the authority of written law derived, and it continued to perform a community-defining role. Among the Greek and Roman philosophers, another notion of unwritten law arose, which was intuitively accessible to all men with a rational nature.

It is clear that the appearance of written law did not obviate the need for oral customs, nor did it necessarily diminish the authority of the latter. Oral tradition existed side by side with written law, as was case with the Mishnah until 200 AD.

[Top of page]

3.4.3 Establishment of Written Canon

Given what we know of ancient attitudes toward oral and unwritten laws, we should not assume that the mere act of writing down an oral traditional narrative established the written version as more authoritative. We have seen from historical testimony that St. Mark’s Gospel may have been merely tolerated by St. Peter as an acceptable instrument to record and transmit preaching. There was at first no sense that only the written Gospel was authoritative.

The very notion of a written document constituting a definitive rule or canon requires a consciously literary culture, according to Philip R. Davies in Scribes and Schools: The Canonization of Hebrew Literature. A work can become canonized only when it is(1) preserved by copying until it achieves status as a classic, and (2) is classified by belonging to a collection of some kind. Both activities—copying and archiving—presuppose a literary culture. We should not be surprised, then, that both the Old and New Testament canons were not formally defined until centuries after their constituent works were written.

I would impose important qualifications on Davies’ analysis. First, the authority of a written document does not derive from its inclusion in the canon, but generally antecedes the construction of the canon. Copying does not make the classic. With limited resources, you only copy a work because you find it important in the first place. Likewise, it was not archiving as such that made something worthy of canonization; rather, in the case of Christian books or epistles, it was the fact that it was read in all the churches. In turn, the reason something was read in all the churches is because it was believed to have apostolic, and therefore universal, authority. One need not have a purely literary culture to establish a canon; a culture where literature is used orally suffices.

Mournet thinks that scribes who copied documents witnessed the rise of the Jesus tradition from its humble origins as an orally proclaimed gospel to its status as an authoritative, inscribed text. This is a mistake. As Mournet himself elsewhere observes, oral tradition was considered no less authoritative than written tradition. The difference is in the fixity of verbal form and the mode of transmission. The authority of the oral tradition is that it is the words and deeds of Jesus, and its authenticity is guaranteed by the living eyewitnesses and the community. The authority of the New Testament epistles was immediate, coming from the universally recognized authority of the Apostles. They were authoritative even before they were copied and circulated. They lacked only to be gathered together in a new canon of sacred scripture.

[Top of page]

3.4.4 Relative Authority of Oral and Written Christian Tradition

Walter Ong claimed that the period from around 100 AD was a manuscript culture that began to interact with texts on new level. Mournet clarifies this was not a sharp transition, but gradual before and after 100 AD. This choice of date conveniently coincides with composition of the New Testament writings. Yet a broader study of the Roman world indicates that this textual culture developed around the third century, among pagans no less than Christians. [See: William V. Harris, Ancient Literacy (Cambridge: Harvard University Press, 1989); Mary Beard, Ancient literacy and the function of the written word in Roman religion, in Humphrey, ed., Literacy, 35-58.]

The vast majority of people in late antiquity, including Jews of Roman Palestine, were at best marginally literate. Less than 10% of Galilee and Judea were literate. [Harris, op. cit., p.114.] Still, Roman literacy compares favorably with the rest of the world, and the literary culture that would soon develop among aristocrats was without precedent. This unique development was favored by the use of an alphabetic script, the economic circumstances allowing production of writing materials, and religious interests in preserving a fixed written tradition, especially among Christians and Jews. Christianity not only antedates canonical Scripture; it helped shape our attitude toward the written word.

A predominantly religious interest in writing is found among the New Testament authors, who, except for perhaps Luke and the author of Hebrews, had no deeper acquaintance with secular Greek writing. [Mournet, op. cit., p.114.] Instead they came from the synagogue training of the Greek-speaking Jewish community. This might credibly apply to Ss. Matthew, Mark, and Paul, though it is less likely that the other Apostles would have had any formal schooling. Still, their scribes would have had such training, and the general point remains that most New Testament authors did not have a literary education as we know it. Their interaction with texts was via oral recitation and memorization, in a religious context.

Although Judaism was already a Scriptural religion, it does not seem to have occurred to early Christians that the Gospel should be a second set of Scripture. Christianity was spread mainly through preaching, discussion, and personal interaction, not literature, during the first century. Even after much of the Gospel was written, there was little sense that this form was preferable to oral tradition. Quoting the New Testament as Scripture did not become common until the third century, as Origen developed a school of exegesis and helped establish a library at Caesarea. It is only at this point that a truly literary Christian culture, copying and archiving manuscripts, was established.

Texts were owned and maintained communally, as very few people were wealthy enough to own or produce texts. Further, there was little perceived social need to own texts personally. In Judaism and early Christianity, texts were read in public for communal worship. Clarification of a text’s meaning was obtained from a viva vox, i.e., the Mishnaic oral tradition. Most Jewish students were literate only in the sense of hearing texts read aloud learning them, guarded in their memory. [Josephus Ant 4.210, 16.43; also Philo , Legat 115, 210; Martin Hengel, The Hellenization of Judea in the First Century After Christ (London: SCM Press, 1989), 55.]

In this context, we can make better sense of Papias’s preference for a living voice (zoses phones). Even though much of the Gospel had been written, a living testimony was necessary for interpretation. An abundance of document would not abolish the need for living testimony, as writing only supplemented oral tradition. Further, the authority of the written Gospels was grounded in that of the Apostles and other eyewitnesses. By speaking to those who knew the Apostles, Papias was going to the source.

Early Patristic authors ascribe an esteem for oral tradition to the Evangelists themselves. Clement of Alexandria says Mark followed St. Peter for a long time and remembered his sayings (memnemenon ton lechthenton). [Eusebius, Eccl. hist., VI, 14, 6] This need not imply formal instruction, but St. Mark, raised in synagogue school, would be accustomed to memorizing religious teachings.

Similarly, St. Irenaeus says Mark transmitted (paradedoken) that which Peter preached, and Luke wrote what Paul preached. [Eusebius, op. cit., V, 8, 2-4] The latter attribution may seem incredible, given that St. Paul never writes any of the Synoptic tradition in his epistles. This omission need not imply ignorance. Indeed, it is highly implausible that St. Paul, who visited so many Christian churches, could have been ignorant of the sayings, parables, and miracles of Christ. His reticence to mention these in his letters might be attributed to (1) a sense that a letter was not an appropriate context for recitation of oral tradition, or (2) an awareness that he was not a fitting reciter of such tradition, since he was not an eyewitness of Jesus’s ministry.

St. Luke, the only Evangelist who tells us something of how he composed his Gospel, says that he gives that which was handed down (paredosan) from eyewitnesses (autoptai). He relates this as an assurance of the accuracy and authority of his account. This implies that oral tradition was active and authoritative, and indicates that some witnesses were especially qualified to recite such tradition.

The second and third generations of Christians showed similar reverence for oral tradition mediated by qualified witnesses. St. Irenaeus says that St. Polycarp interacted with John and the others who had seen the Lord. Clearly this John was an eyewitness, even if he is not necessarily the Apostle. St. Polycarp remembered their words, and what were things concerning the Lord which he had heard from them, and about his miracles, and about [his] teaching. Like St. Luke, he received the oral Gospel from eyewitnesses. St. Irenaeus, in turn, listened to St. Polycarp, and made notes of these things not on paper but in my heart. This reflects a sensibility that the most precious teachings should be stored in memory rather than on paper. (Still, Irenaeus gives witness that the ancients did sometimes take notes on paper.) Just as Jewish students kept the law in their heart by memorization, so did Christians keep the Gospel within them.

Memorization of oral Christian tradition is not described as formal training, but as something undertaken on one’s own initiative. Thus there was no formal body of tradition mandated to be memorized, as in a rabbinical school. Still, this informally transmitted tradition was subject to some community controls, as evidenced by the preference for more qualified witnesses or reciters of tradition.

[Top of page]

3.4.5 Orality and Text in Jewish Tradition

Since the oral Gospel circulated initially among Palestinian and Hellenized Jews, we may turn to Jewish accounts to amplify how early Christians likely understood the relationship between oral and written tradition.

In the last centuries before Christ, the Jews established the importance of the written Law as foundational to their religion. This was emphasized in the late apocrypha, which portrayed Enoch as able to write, and purported to recount secret prophecies of Moses. Still, oral tradition was held in great esteem, and the oral Torah was considered to extend back to Moses on Mount Sinai. If a law comes to hand and you do not know its nature, do not discard it for another one, for lo, many laws were stated to Moses at Sinai, and all of them have been embedded in the Mishnah. (y. Hag. 1:7b) The laws of oral tradition are so revered, that even those of doubtful origin should be retained, lest you inadvertently discard an authentic law that was given to Moses. The Testament of Reuben says instruction comes through the sense of hearing, not sight (1:16), and urges Israelites to hear ordinances from Levi, who knows the Law (2:26).

A prophet is a mouthpiece of God, so Moses, as the greatest of prophets, certainly spoke much on God’s behalf. Thus the oral Torah comes from the mouth of the Almighty (b. Erub. 54b). Both oral and written Law must be kept. Oral law is an explication of the written Law; both are given by God at Sinai. Oral and written Law are passed on formally, in a teacher-disciple relationship.

We have scant evidence of how new Jewish texts were composed from oral accounts. The Letter of Aristeas (2nd cent. BC) says a trustworthy narrative has been compiled based on the author’s meeting with Eleazar the high priest and Philocrates, who had heard a personal account. Philo of Alexandria, in his Life of Moses, says his sources are the sacred books and what his elders told him. I always interwove what I was told with what I read. [De vita Mosis, 1.1.4]

[Top of page]

3.4.6 Oral Performance of Christian Tradition

Oral traditions are preserved through repeated performance in special social situations. This practice was also applied to writing in ancient cultures. Texts were usually read aloud publicly to a group, as most houses were too small for private reading. In Sumer, writing was taken by dictation and later read aloud, ritualistically invoking the king’s presence, thus says your king. The Book of Jeremiah indicates that it is the writing of previous oral performances by prophet. [Jer. 30:2, 36:2] The scribe Baruch later read aloud these prophecies. [Jer. 36:6]

We have only fragmentary historical evidence of how early Christians used written traditions mediated by orality. St. Paul’s citations of Scripture follow Septuagint and Masoretic readings erratically. Evidently, he did not have a text in front of him, but recalled verses from memory, either from having heard readings in synagogue, or from what he learned at the feet of Gamaliel. Wealthy individuals could own texts, as appears to be the case with the eunuch encountered by Philip, reading from Isaiah. [Acts 8:30] Even here, the text is read aloud in the open. Christian doctrine was also proclaimed by such means, as when a letter from the Council of Jerusalem was read to everyone in Antioch. [Acts 15:30-31]

Oral instruction retained pride of place among Christians well into the Patristic era. Clement of Alexandria, commenting on Romans 10:17, says faith comes from what is heard. [Stromata 2:6] St. Augustine says that he read Aristotle’s Categories when he was twenty, and understood it better than those who received oral instruction. [Confessions, IV, 15, 28] This remark indicates that it was ordinarily considered superior to learn a text by oral instruction than by private reading.

Loveday Alexander has surveyed uses of the term viva vox in 1) proverbs (e.g., Galen notes a saying among craftsmen that reading out of a book is not the same as learning from a living voice); 2) rhetoric (e.g., Quintilian says there is more nourishment from a living voice); 3) crafts; 4) schools (e.g., Seneca says you will gain more from a living voice than from a treatise [Epistulae morales VI, 5]); 5) esotericism and 6) Papias and Clement (already discussed). In schools, there was a strong tendency to see written texts as secondary to and subordinate to oral instruction. The living voice of a teacher had priority. [Loveday Alexander. The Living Voice: Scepticism Towards the Written Word in Early Christian and in Graeco-Roman Texts. In: The Bible In Three Dimensions, eds. DJA Cline et al. (Sheffield: Sheffield Academic Press, 1990), 221-47.]

In this context, we can see why St. Irenaeus says that the personal instruction received by St. Polycarp made him a more reliable witness than heretics. Although the heretical groups had their own books of written tradition, they were disconnected from the personal instruction of the Apostolic witnesses and their successors. St. Polycarp’s witness was more reliable not only because of his mode of education, but because of the identity of his instructors, who evidently were among the eyewitnesses (autoptai) qualified to recite the oral Gospel.

[Top of page]

3.5 Conclusion: A Model of the Oral Gospel Tradition

Using what we have learned of oral modes of composition and transmission, and the oral use of texts, we might attempt a hypothetical model of the oral Gospel tradition. The model that most closely fits the historical evidence is Bailey’s informal controlled oral tradition, though we should modify this to match the known organizational structure of the early Christian Church.

Albert Lord noted that an oral tradition attains a fair degree of fixity if the retelling is confined to a closed group of storytellers, and not just anyone can recite the tradition. Original eyewitnesses are also stabilizing influences, for as long as they are still living. The fixity of a tradition varies by genre. Short proverbs are much more likely to have a fixed form than a longer narrative. The mixture of genres found in the written Gospels implies that we should not expect a uniform degree of fixity in all parts of the Gospels.

Kenneth Bailey’s observations of Middle East oral traditions,[6] briefly discussed previously, give a credible model of how the oral Gospel may have had a relatively fixed, stable form, which would account for many similarities among the Synoptics without recourse to literary dependence. The consistency of this model with ancient historical accounts of oral tradition reduces concerns that it is wholly inapplicable due to anachronism. Still, any application of this model to the early Christian Church should take into account (1) the relatively high degree of literacy among early Christians, and (2) the authoritative precedence given to the Apostles, bishops, elders, and eyewitnesses. Thus we may expect to find more textual thinking in some of the oral Gospel, and a less egalitarian structure regarding the preservation and recitation of tradition.

Bailey contrasts his observations with two extremes proposed by earlier scholars. At one end, there is Bultmann’s view that the early Christian community was unconcerned with preserving the facts about Jesus, but freely constructed legends. This would be an informal uncontrolled oral tradition, akin to rumors. At the other extreme, some scholars have proposed a formal controlled tradition, like Gerhardsson’s theory that Jesus established a sort of rabbinic school among his disciples, who used standard mnemonic techniques to retain his teachings. Yet there is no evidence that Jesus ever did such a thing, nor that most Apostles were even capable of such learning.

From his observations in the Middle East, Bailey infers the existence of a third mode of transmission, which he calls informal controlled oral tradition. That is, there are few or no constraints on who is the teacher or the student. The group is collectively preserving its traditions by gathering together and reciting them. Anyone may recite, though in practice, the older, more gifted, and socially more prominent do most reciting. The choice of reciter often depends on who knows a particular story best. There are no official storytellers, such as the traveling bards among the Bedouin or the epic singers studied by Lord. Still, there is a restriction: only those who are from the village are fit to pass on the tradition. Even a man who was there for only thirty-seven years was considered unfit.

The mode of transmission, Bailey notes, is determined by content. Rumors of atrocities and gossip are transmitted in an informal, uncontrolled mode. Informal controlled tradition contains the following:

  1. Short proverbs or wisdom sayings
  2. Story riddles (e.g., the story of Solomon and the baby with two mothers), where a problem is solved by a wise answer; many Gospel stories fit this form.
  3. Poetry
  4. Parables
  5. Well-told accounts of the important figures in history of community. The key here is well-told, as opposed to the less reliable legends that crop up in uncontrolled settings.

With the exception of poetry, all these types of content are strongly characteristic of the Gospels. The problem of identifying the genre of the Gospels disappears when we consider that they have no literary genre, but are a pastiche of different genres of informal controlled oral tradition. As Jesus himself preached in proverbs and parables, this provided content in a mode that could be easily transmitted in an oral community. There is no mystery as to why Jesus never wrote anything. This showed no lack of concern for the preservation of his message, since this was well provided for by the mode of transmission.

Informal controlled tradition is informal only in the sense of there being no formally defined teachers. It is still transmitted in a formal setting, like all oral traditions. In the communities observed by Bailey, this was the haflat samar. Generally in Arabic, the term means a soirée, or night conversation party for the purpose of amusement. In the Christian communities observed by Bailey, however, this entertainment function was coupled with that of preserving local tradition. The group setting allows the community to impose controls on recitation. If the reciter deviates from acceptable norms, he is corrected by a chorus of voices. The shame-pride culture of the Middle East makes reciters do their best to avoid such correction.

The amount of discretion allowed to the reciter varies by genre. Poems and proverbs admit no flexibility, not even of a single word. The fixed structures of local poetry impose mnemonic constraints so that even those unfamiliar with a specific poem can recognize the structure. No one dares to recite a well-known poem unless he has it accurately memorized.

Parables and accounts of historical figures admit some flexibility. Exact verbatim recollection is not essential, except for key lines such as a proverb or punch-line, and names cannot be changed. If there are multiple scenes, in some cases the order of scenes may be changed if this is not essential to the story. The storyteller may vary the pitch of emotional reaction between characters, and may adjust the dialogue to his own style and interests.

Bailey calls this continuity and flexibility, as opposed to continuity and change. If the latter were permitted, a story-teller could change any 15 percent of the story, and after multiple iterations, the entire story might change. With flexibility, however, only some aspects of the story are allowed to change, while others must stay the same. So the fixed parts will remain constant. As C.S. Lewis said, a story is a pattern of events, not a pattern of words, like a poem. Thus stories could be preserved for many centuries.

Total flexibility is permitted to casual news of the day and atrocity stories. Here change and gross exaggeration are permitted. These stories are irrelevant to community identity and so not valuable.

Any notion that the Gospels would have been permitted total flexibility implies that they were irrelevant to the identity of Christian communities, which is absurd. The Bultmannian model of early transmission of Christian tradition is certainly wrong, if that tradition resembles what Bailey has observed in the Middle East, and may go back at least to the fourth century, judging from the antiquity of some histories recounted.

Bailey offers a test of the reliability of informed controlled oral tradition, by comparing independent oral and written accounts of the nineteenth-century Scottish missionary John Hogg, who founded churches in southern Egypt. He founded evangelical communities in various villages, each of which has its own stories of his words and deeds. These stories spread to other villages, but are primarily preserved in the village of origin in each case.

Some of the stories preserve exact words of Hogg’s replies over a hundred years ago. Hogg’s daughter Rena recorded some oral tradition in 1910, and Bailey found they were told the same way over fifty years later. Yet as Theodore Weeden has noted, Rena Hogg gives only cursory mention of these incidents, and fails to confirm significant aspects such as the resolution or conclusion.[7] Against this criticism, J.D.G. Dunn responds that Weeden is supposing that Rena Hogg’s version is a complete, authentic original edition against which others should be judged. This misunderstands oral tradition, which allows variation from the beginning, without a fixed first edition.[8]

Even supposing that informal controlled oral tradition is reliably transmitted, this would not address whether it is reliable in its origin. We would need to know how something enters such tradition in the first place.

As an example of the process of introducing content to tradition, Bailey offers a parable told by the Rev. Ibrahim Dagher, a community leader, in 1967. Eighteen years later, in Bailey’s estimation, the parable was remembered with about 80 percent verbal accuracy. It was memorable precisely because Rev. Dagher altered it from the expected ending to make a point about Lebanese nationalism. We can easily think of Gospel parables that have endings that would be unexpected to a Jewish audience, though still drawing upon familiar themes and stories.

Weeden remarks that Rev. Dagher’s alteration of a parable contradicts Bailey’s assertion that such content is controlled. Yet, as Dunn retorts, such alteration was not intended to distort tradition, but to tell a new story that assumed the audience already knew the traditional ending to the parable. This is an example of metonymy in oral tradition. Further, Dagher’s telling of a new version was itself a significant event (as a political statement), and so worthy of being added to tradition.

Historical events can be introduced into oral tradition by the following process. At a wedding in 1958, during celebratory gunshots, a rifle held by the groom’s friend Hanna failed to fire. Hanna lowered the gun, which then went off, killing the groom. Immediately afterward, multiple witnesses told Bailey the story, using several identical phrases at the climax. Already, the story was adopting a fixed verbal form at key points. They all said, The gun fired (passive), in order not to make Hanna the culprit. This reflects a community decision that this was an act of God, hence the use of the divine passive, often found in Luke, as Bailey notes. Community controls here involve the interpretation of the event, without falsifying facts.

As an interesting aside, the locals told the police that a camel had stepped on the groom. This resembles the apocryphal story of Judas being crushed by a passing chariot. Might that not be an example of a deliberate fiction later told as fact, via uncontrolled rumor? Yet even the fiction of the camel was something of a formality. The police gathered the real facts unofficially and privately, but no one would ever formally testify to these facts, as the matter was considered handled by the community, so they needed no police interference.

Bailey estimates that this story will not be told for more than a generation, since the figures were not important. Still, it is firmly memorable to those who heard it in that first week, with the constant repetition of the community condensation, which included the same core of information in nearly the same words.

Similar repetition of core information was observed after preaching a new story to the community.

At the conclusion of the telling of the story the attention of the congregation would literally break up in what I discovered was a form of oral shorthand. The elder on the front row would shout across the church to a friend in a loud voice, ‘Did you hear what the preacher said? He said…’ and then would come a line or two of the story including the punch-line. People all across the church instinctively turned to their neighbours and repeated the central thrust of the story twice and thrice to each other. They wanted to retell the story that week across the village and they had to learn it on the spot. The preacher was not allowed to continue until they had done so.

This is arguably the most impressive and relevant observation that Bailey presents, as it shows how noteworthy preaching enters oral tradition on the spot. Members of the congregation immediately recount the salient points, and begin to develop the essential oral form in which it is to be subsequently preserved. If a similar process occurred at the conclusion of each of Jesus’s discourses, then oral Gospel traditions may have been formed among hundreds or thousands of hearers, and need not have been confined to the Apostles or the seventy-two disciples.

The model proposed by Bailey is consistent with what early Christian authors, including St. Luke the Evangelist, tell us about the importance of hearing the Gospel tradition from eyewitnesses. If the oral Gospel were formally controlled, as in a rabbinic school, we should expect it to be confined to Apostles and their successors. If it were completely uncontrolled, we should expect much more variation in content and sequence than we observe among the Synoptics. Some form of informal controlled model best accounts for the Synoptic Gospels, once we recognize that the combination of similarity and variation is inconsistent with a purely literary relationship among them.

Bailey thinks that when St. Luke mentions eyewitnesses and ministers (autoptai kai huperetai, Lk. 1:2), the latter term is a translation of hazzan, the synagogue official who handles the Torah scrolls. This could mean that there were already written liturgical documents, or at least that certain members of the Christian churches had a special role in preserving the oral Gospel.

Bailey thinks that eyewitnesses and ministers of the word are not two distinct groups of people. Rather, only an eyewitness was qualified to be a huperetes tou logou. This is insufficently substantiated, as Luke may more naturally mean autoptai kai huperetai as two distinct categories. Still, even on that supposition, only certain people were considered competent to tell stories. The cultural preference for the wisdom of elders likely implies that these huperetai were contemporaries of Christ, even if they did not personally witness every single event they recounted. At any rate, the structure of the early Christian churches was non-egalitarian insofar as people had distinct roles, including regarding the preservation of oral tradition. This means that Christian tradition was not quite as informal as that observed by Bailey.

Still, early Christians would share with modern Arabs the need to continually retell their stories in a controlled setting, lest they should lose their identity as a community. Early Christians were certainly interested in the life and teaching of their Founder. The preservation of such information was essential to their identity, so they were certainly concerned with history. This does not mean they had our notion of history, for the Gospels are much more interested in relating religious meanings than chronologically ordered facts.

The notion that only some people were qualified to recite oral tradition may account for the absence of Gospel material in St. Paul’s writings. St. Paul, perhaps, was not a reciter of oral tradition, so he does not presume to recount Gospel material in his letters, though even the hardiest skeptic must acknowledge the well-traveled Apostle knew of such material. He recounts only the basic pattern of events, and makes a theological exposition of these. Here he undoubtedly relies on his readers’ presumed knowledge of the life and teaching of Christ, much of which can be seen behind the Apostle’s writing, even if not cited explicitly. This is no idiosyncrasy of St. Paul, as the other New Testament epistles likewise abstain from duplicating Gospel material, though several letters were certainly later than the Synoptics.

Bailey suggests that St. Paul may have known some formal controlled tradition, hinted in the expression: What I have received I delivered unto you…. (1 Cor. 15:3) Yet this expression is much too vague to indicate whether tradition was received by formal instruction. A similar expression is used with reference to the words of Institution at the Last Supper, one of the few places where St. Paul explicitly references Gospel material. Here, however, he says For I received from the Lord what I also passed on to you… (1 Cor. 11:23), which is hardly an explicit reference to formal instruction. Still, the presence of formalism is implied by the fact that St. Paul’s rendering of the Institution more closely resembles ancient liturgies than the abbreviated versions found in the Synoptics. Possibly, this text may have been more catechetical than liturgical, i.e., helping to understand the Eucharistic liturgy rather than recited during it.

While we cannot prove that early Christians really adopted the model of oral tradition discussed, it is at least a possibility within the capacity of people at that time and place. Accordingly, we are no longer compelled to invoke literary dependence as the only possible explanation for a well-preserved, yet variable tradition in the Synoptic Gospels, nor does the absence of written tradition or formal oral instruction necessary imply that the oral Gospel consisted of unreliable rumors or legends.

To examine which modes of composition and transmission best account for the Synoptic Gospels, we will have to examine the texts themselves in some detail. Here we hope to find clues as to the stability and variability of the underlying tradition. This may in turn have bearing on the historical accuracy and reliability of the text, though that is mostly beyond our scope. We are concerned principally with how the Synoptic Gospels were composed, in order to account for their similarities and differences.

Continue to Part IV


Footnotes to Part III

[1] Terence C. Mournet. Oral Tradition and Literary Dependency: Variability and Stability in the Synoptic Tradition. (Tübingen: Mohr Siebeck, 2005)

[2] David C. Rubin. Memory in Oral Traditions. (Oxford Univ. Press, 1995)

[3] David C. Parker, The Living Text of the Gospels (Cambridge Univ Press, 1997) pp.117-118. Cited in Mournet, op. cit., p.8.

[4] This hypothesis is grounded purely in Bultmann’s a priori bias against the possibility of miracles: It is impossible to use electric light and wireless and to avail ourselves of modern medical and surgical discoveries and at the same time to believe in the New Testament world of spirits and miracles. [Kerygma and Myth, I, A, 2] Setting aside the obvious falsity of this statement, it is ahistorical to project modern doubts onto the ancients. The Palestinian Jews, Christian and non-Christian, believed in miracles no less than the Hellenized. This is attested in rabbinic sources as well as the Gospel according to the Hebrews.

[5] Mournet, op. cit., p.25.

[6] K.E. Bailey, Informal Controlled Oral Tradition and the Synoptic Gospels. Asia Journal of Theology, 5 (1991), pp. 34-54.

[7] Theodore Weeden. Kenneth Bailey’s Theory of Oral Tradition: A Theory Contested by Its Evidence. Journal for the Study of the Historical Jesus, 7 (2009), 3-43.

[8] James D.G. Dunn. Kenneth Bailey’s Theory of Oral Tradition: Critiquing Theodore Weeden’s Critique. Journal for the Study of the Historical Jesus, 7 (2009), 44-62.


© 2014 Daniel J. Castellano. All rights reserved. http://www.arcaneknowledge.org