The Making of the New Testament

by Benjamin Wisner Bacon

This Papias of Hierapolis, the friend and colleague of Polycarp, had undertaken in opposition to "the false teachers, and those who have so very much to say," to write (probably after the utter destruction of the community of 'apostles, elders, and witnesses' at Jerusalem in 135), an Exposition of the Sayings of the Lord. He based the work on authentic tradition of the Jerusalem witnesses, two of whom (Aristion, and John 'the Elder') were still living at the time of his inquiries. In fact, this much debated "John the Elder," clearly distinguished by Papias from John the "disciple of the Lord," may be identified, in our judgment, with the John mentioned by Eusebius and Epiphanius midway in the succession of 'Elders' of the Jerusalem church between a.d. 62 and 135. Epiphanius dates his death in 117. Papias gives us practically all the information we have regarding the beginnings of gospel literature. He may have known all four of our Gospels. He certainly knew Revelation and "vouched for its trustworthiness," doubtless against the deniers of the resurrection and judgment. He "used testimonies" from 1st John, and probably the saying of Jesus of John xiv. 2; but he seems to have based his Exposition on two gospels only, giving what he had been able to learn of their history from travellers who reported to him testimonies of 'the elders.' Papias' two gospels were our Matthew and our Mark, whose differences he reconciled by what the Jerusalem elders had reported as to their origin. Matthew, according to these authorities (?), represented in its Greek form a collection of the Precepts of the Lord which had formerly been current in the original Aramaic, so that its circulation had of course been limited to Palestine. The original compiler had been the Apostle Matthew. Various Greek equivalents of this compilation had taken its place where Aramaic was not current. Thus Papias, in explicit dependence on "the Elder" so far as Mark is concerned, but without special designation of his authority for the statement regarding Matthew. It is even possible that his representation that the primitive Matthew was "in the Hebrew tongue" may be due to rumours whose real starting-point was nothing more than the Gospel of the Nazarenes, a product of c. 110-140 which misled many later fathers, particularly Jerome. We cannot afford, however, to slight the general bearing of testimony borne by one such as Papias regarding the origins of gospel composition, and particularly the two branches into which the tradition was divided. For Papias had made diligent inquiry. Moreover his witness does not stand alone, but has the support of still more ancient reference (e. g. 1st Tim. vi. 3, Acts i. 1) and the internal evidence of the Synoptic Gospels themselves. The motive for his statement is apologetic. Differences between the two Gospels had been pointed out on the score both of words and events. Papias shows that Gospel tradition is not to be held responsible for verbal agreement between the two parallel reports of the Lord's words. The differences are attributable to translation. So, too, regarding events. Exact correspondence of Mark with Matthew (or other gospels) is not to be looked for, especially as regards the order; because Mark had not himself been a disciple, and could not get the true order from Peter, whose anecdotes he reproduced; for when Mark wrote Peter was no longer living. Mark has reproduced faithfully and accurately his recollection of "things either said or done," as related by Peter. But Peter had had no such intention as Matthew of making a systematic compilation (syntagma) of the sayings of the Lord, and had only related his anecdotes "as occasion required." If the tradition regarding Matthew, as well as that regarding Mark, was derived from the Elder, he, too, as well as Papias, knew the Greek Matthew; regarding it as a "translation" of the apostolic Logia, he naturally makes Matthew the standard and accounts as above for the wide divergence of Mark as to order.

The Jerusalem elder who thus differentiates the two great branches of gospel tradition into Matthæan Precepts and Petrine Sayings and Doings, is probably "the Elder John"; for this elder's "traditions" were so copiously cited by Papias as to lead Irenæus, and after him Eusebius, to the unwarranted inference of personal contact. Irenæus even identified the Elder John with the Apostle, thus transporting not only him, but the entire body of "Elders and disciples" from Jerusalem to Asia, a pregnant misapprehension to which we must return later. In the meantime we must note that this fundamental distinction between syntagmas of the Precepts, and narratives of the Sayings and Doings, carries us back as far as it is possible to penetrate into the history of gospel composition. The primitive work of the Apostle Matthew, was probably done in and for Jerusalem and vicinity—certainly so if written in Aramaic. The date, if early tradition may be believed, was "when Peter and Paul were preaching and founding the church at Rome." Oral tradition must have begun the process even earlier.[16] Mark's work was done at Rome, according to internal evidence no less than by the unanimous voice of early tradition. It dates from "after the death of Peter" (64-5) according to ancient tradition. According to the internal evidence it was written certainly not long before, and probably some few years after, the overthrow of Jerusalem and the temple (70). At the time of Papias' writing, then (c. 145), all four gospels were probably known, though only Matthew and Mark were taken as authoritative because (indirectly) apostolic. At the time of prosecution of his inquiries the voice of (Palestinian) tradition was still "living and abiding." If, as tenses and phraseology seem to imply, this means Aristion and the Elder John (ob. 117?) it is reasonable to regard it as extending back over a full generation. The original Matthew was even then (c. 100), and in Palestine itself, a superseded book. It had three successors, if not more, two Greek and one Aramaic, all still retaining their claim to the name and authority of Matthew[17]; but all had been re-cast in a narrative frame, which at least in the case of our canonical first Gospel was borrowed from the Roman work of Mark. So far as the remaining fragments of its rivals enable us to judge, the same is true in their case also, though to a less extent. It is quite unmistakably true of Luke, the gospel of Antioch, that its narrative represents the same "memorabilia of Peter"; for so Mark's gospel came to be called. Thus the Petrine story appears almost from the start to have gained undisputed supremacy. But side by side with this remarkable fact as to gospel narrative is the equally notable confirmation of the other statements of 'the Elders' regarding the Precepts. For all modern criticism admits, that besides the material of Mark, which both Matthew and Luke freely incorporate, omitting very little, our first and third evangelists have embodied, in (usually) the same Greek translation but in greatly varied order, large sections from one or more early compilations of the Sayings of Jesus.

It is indispensable to a historical appreciation of the environment out of which any gospel has arisen that we realize that no community ever produced and permanently adopted as its "gospel" a partial presentation of the message of salvation. To its mind the writing must have embodied, for the time at least, the message, the whole message, and nothing but the message. Change of mind as to the essential contents of the message would involve supplementation or alteration of the written gospel employed. No writing of the kind would be produced with tacit reference to some other for another aspect of the truth.

It was not, then, the mere limitation of its language which caused the ancient Matthæan Sayings (the so-called Logia) to be superseded and disappear; nor is mere "translation" the word to describe that which took its place. The growth of Christianity in the Greek-speaking world not only called upon Jerusalem to pour out its treasure of evangelic tradition in the language of the empire, but stimulated a sense of its own increasing need. That which could once be supplied by eye-witnesses, the testimony of Jesus' mighty works, his death and resurrection, was now fast disappearing. And simultaneously the appreciation of its importance was growing. It was impossible to be blind to the conquests made by the gospel about Jesus. Enclosed in it, as part of its substance the gospel of Jesus found its final resting-place, much as the mother church itself was later taken up and incorporated in a catholic Christendom. So it is that in the Elder's time the church of the 'apostles, elders and witnesses' have done more than merely supersede their Aramaic(?) Syntagma of the Precepts by "translations." They had adopted alongside of it from Rome Mark's "Memorabilia of Peter" as to "things either said or done by the Lord." We can see indeed from the apologetic way in which 'the Elder' speaks of Mark's limitations (Peter is not to be held responsible for the lack of order) that Mark's authority is still held quite secondary to Matthew's; but the very fact that his work is given authoritative standing at all, still more the fact that it has become the framework into which the old-time syntagma has been set, marks a great and fundamental change of view as to what constitutes "the gospel."

No mere syntagma of the Precepts of Jesus has ever come down to us, though the papyrus leaves of "Sayings of Jesus" discovered in 1897 at Behneseh in Egypt by Grenfell and Hunt had something of this character.[18] It was impossible that any community outside the most primitive one, where personal "witnesses of the Lord" still survived "until the times of Trajan," could be satisfied with a "gospel" which gave only the precepts of Jesus without so much as an account of his crucifixion and resurrection. And, strange as it may seem, the evidence of Q (i. e. the coincident material in Matthew and Luke not derived from Mark), as judged by nearly all critics, is that no narrative of the kind was given in the early compilation of discourses from which this element was mainly derived. After the "witnesses," apostolic and other, had begun to disappear, a mere syntagma of Jesus' sayings could not suffice. It became inevitable that the precepts should be embodied in the story. And yet we have at least two significant facts to corroborate the intimations of ancient tradition that this combination was long postponed. (18) When it is at last effected, and certainly in the regions of southern Syria,[19] there is even there practically nothing left of authentic narrative material but the Petrine tradition as compiled by Mark at Rome. Our Matthew, a Palestinian Jew, the only writer of the New Testament who consistently uses the Hebrew Bible, makes a theoretical reconstruction of the order of events in the Galilean ministry, but otherwise he just incorporates Mark substantially as it was. What he adds in the way of narrative is so meagre in amount, and so manifestly inferior and apocryphal in character, as to prove the extreme poverty of his resources of oral tradition of this type. Luke has somewhat larger, and (as literary products) better, narrative additions than Matthew's; but the amount is still extremely meagre, and often historically of slight value. Some of it reappears in the surviving fragments of the Preaching of Peter. To sum up, there is outside of Mark no considerable amount of historical material, canonical or uncanonical, for the story of Jesus. This fact would be hard to account for if in the regions where witnesses survived, the first generation really took an interest in perpetuating narrative tradition. (2) The order of even such events as secured perpetuation was already hopelessly lost at a time more remote than the writing of our earliest gospel. This is true not only for Mark, as 'the Elder' frankly confesses, but for Matthew, Luke and every one else. Unchronological as Mark's order often is (and the tradition as to the 'casual anecdotes' agrees with the critical phenomena of the text), it is vastly more historical than Matthew's reconstruction. On the other hand Luke, while expressly undertaking to improve in this special respect upon his predecessors, almost never ventures to depart from the order of Mark, and when he does has never the support of Matthew, and usually not that of real probability. In short, incorrect as they knew the order of Mark to be, it was the best that could be had in the days when evangelists began to go beyond the mere syntagmas, and to write "gospels" as we understand them, or, in their own language, "the things which Jesus began both to do and to teach" (Acts i. 1). From these two great outstanding phenomena of gospel criticism alone it would be apparent that the distinction dimly perceived in the tradition of the Jerusalem elders reported by Papias, and indeed by many later writers, is no illusion, but an important and vital fact.

A third big, unexpected fact looms up as we round the capes of critical analysis, subtracting from Matthew and Luke first the elements peculiar to each, then that derived by each from Mark. It is a fact susceptible, however, of various interpretation. To some it only proves either the futility of criticism, or the worthlessness of ancient tradition. To us it proves simply that the process of transition in Palestine, the home of evangelic tradition, from the primitive syntagma of Precepts, framed on the plan of the Talmudic treatise known as Pirke Aboth, or "Sayings of the Fathers," to the Greek type of narrative gospel, was a longer and more complex one than has commonly been imagined. A cursory statement of the results of critical efforts to reproduce the so-called "second source" of Matthew and Luke (Mark being considered the first), will serve to bring out the fact to which we refer, and at the same time, we hope, to throw light upon the history of gospel development.

The mere process of subtraction above described to obtain the element Q offers no serious difficulties, and for those who attach value to the tradition of 'the Elders' it is natural to anticipate that the remainder will show traits corresponding to the description of an apostolic syntagma of sayings of the Lord translated from the Aramaic, in short the much-desired Logia of Matthew. The actual result is disappointing to such an expectation. The widely, though perhaps somewhat thoughtlessly accepted equivalence Q = the Logia is simply false. Q is not the Logia. It is not a syntagma, nor even a consistent whole, and as it lay before our first and third evangelists it was not (for a considerable part at least) in Aramaic. True, Q does consist almost exclusively of discourse material, a large part of which has only topical order, and is wholly, or mainly, destitute of narrative connection. Also we find traces here and there of translation at some period from the Aramaic, though not more in the Q element than in Mark. But to those who looked for immediate confirmation of the tradition the result has been on the whole disappointing. Some, more particularly among English critics, have considered it to justify a falling back upon the vaguer generalities of the once prevalent theory of oral tradition. In reality we are simply called upon to renew the process of discrimination. Most of the Q material has the saying-character and is strung together with that lack of all save topical order which we look for in a syntagma. But parts of it, such as the Healing of the Centurion's servant (Matt. viii. 5-10, 13 = Luke vii. 1-10), or the Preaching of the Baptist and Temptation Story (Matt. iii. 7-10, 12; iv. 2-11 = Luke iii. 7-9, 17; iv. 2-13), obstinately refuse to be brought under this category. Moreover, the latter section has the unmistakable motive of presenting Jesus in his character and ministry as "the Son of God," precisely as in Mark. It begins by introducing Jesus on the stage at the baptism of John, after the ancient narrative outline (Acts i. 22; x. 37 f.), and cannot be imagined as forming part of anything else but a narrative having the conclusion characteristic of our own type of gospel. Other considerable sections of Q, such as the Question of John's Disciples and Discourse of Jesus on those that were 'Stumbled' in him (Matt. xi. 2-11, 16-27; Luke vii. 18-35; x. 13-22), share with the Baptism and Temptation section not only the doctrinal motive of commending Jesus in his person and ministry as the longed-for Son of God, but in a number of characteristics which set them quite apart from the general mass of precepts and parables in Q. We can here mention only the following: (1) the coincidence in language between Matthew and Luke is much greater in these sections of Q, often even greater than in the sections borrowed from Mark, showing clearly the existence of a common document written not in Aramaic, but in the Greek language. (2) This material, unlike most of Q, has served as a source and model in many portions of Mark. (3) It is for the most part not included in the five great blocks into which Matthew has divided the Precepts by means of a special concluding formula (vii. 28; xi. 1; xiii. 53; xix. 1, and xxvi. 1) but appears outside, in the form of supplements to the Markan narrative (iii. 7—iv. 11; viii. 5-13, 18-22, xi. 2-27; xii. 38-45, etc.). Finally (4) the Q material of this type seems to be given more copiously by Luke than by Matthew, and with something more than mere conjecture of his own as to its historical occasion. In fact, since it appears that at least this element of Q was known to Mark, there is nothing to justify exclusion from it of such material as the Transfiguration story, though in this case it would be needful to prove that Mark was not the source. Similarly it would be reasonable to think of Luke's wide divergence from Mark in his story of the Passion as occasioned by his preference for material derived from this source. Only, since Matthew has preferred to follow Mark, we have no means of determining whence Luke did derive his new and here often valuable material.

The existence, then, of an element of Q which quite fails to correspond to what we take the Matthæan syntagma to have been by no means proves either the futility of criticism or the worthlessness of the ancient tradition. It only shows that our synoptic evangelists were not the first to attempt the combination of discourse with narrative, but that Luke at least had a predecessor in the field, to whom all are more or less indebted. Criticism and tradition together show that there are two great streams from which all historically trustworthy material has been derived. The one is Evangelic Story, and is mainly derived from Mark's outline of the ministry based on the anecdotes of Peter, though some elements come from another source, principally preserved by Luke, which we must discuss in a later chapter devoted to the growth of Petrine story at Rome and Antioch. The other stream, "Words of the Lord," comes from Jerusalem, and is always associated in all its forms with the name of Matthew. We have every reason for accepting the statement that as early as the founding of the church in Rome (45-50) the Apostle Matthew had begun the work of compiling the Precepts of Jesus, in a form serviceable to the object of "teaching men to observe all things whatsoever he had commanded." Our present Gospel of Matthew, however, is neither this work nor a translation of it; for the only three things told us about the apostle's work are all irreconcilable with the characteristics of our Matthew. The compilation of "Words of the Lord" was (1) a syntagma and not, like Mark, an outline of the ministry. It was (2) written in Aramaic; whereas our Matthew is an original Greek composition. It was (3) by an apostle who had personal acquaintance with Jesus; whereas our first evangelist is to the last degree dependent upon the confessedly defective story of Mark. Still if we take our Matthew as the last link in the long chain of development, covering perhaps half a century, and including such by-products as the Gospel according to the Hebrews and the Gospel of the Nazarenes, we may obtain a welcome light upon the environment out of which has come down the work which an able scholar justly declared, "the most important book ever written, the Gospel according to Matthew."

The language in which it was written was alone sufficient to place the Greek Matthew beyond all possible competition in the larger world from Aramaic rivals. But its comprehensiveness and catholicity still further helped it to the position which it soon attained as the most widely used of all the gospels. Matthew is not only in its whole structure a composite gospel, but shows in high degree the catholicizing tendency of the times. Just as it frankly adopts the Roman-Petrine narrative of Mark with slightest possible modification, so also it places in Peter's hand with equal frankness the primacy in apostolic succession. Almost the only additions it makes to Mark's account of the public ministry are the story of Peter's walking on the sea (xiv. 28-33), and his payment of the temple tribute for Christ and himself with the coin from the fish's mouth (xvii. 24-27). The latter story introduces the chapter on the exercise of rulership in "the church" (ch. xviii.), beginning with the disciples' question: "Who then is greatest in the kingdom?" Peter is again in it the one salient figure (xviii. 21). An equally important addition, connected with xviii. 17 f. is the famous committal to Peter of the power of the keys, with the declaration making him for his confession the 'Rock' foundation of "the church." This addition to Mark's story of the rebuke of Peter at Cæsarea Philippi, is one which decidedly alters its bearing, and seems even to borrow the very language of Gal. i. 16 f. in order to exalt the apostleship of Peter. In fact, the Roman gospel and the Palestinian almost reverse the rôles we should expect Peter to play in each. Matthew alone makes Peter "the first" (x. 2), while Mark seems to take special pains to record rebukes of the twelve and the brethren of the Lord, and especially the rebukes called down upon themselves by Peter, or Peter and John.

In respect to the primacy of Peter we can observe a certain difference even among the Palestinian gospels which succeeded to the primitive syntagma of Matthew. Little, indeed, is known of the orthodox Gospel of the Nazarenes, beyond its relatively late and composite character; for it borrowed from Matthew, Mark and Luke in turn. Its list of apostles, however, begins with "John and James the sons of Zebedee," then "Simon and Andrew," and winds up: "Thee also, Matthew, did I call, as thou wert sitting at the seat of custom, and thou followedst me." The anti-Pauline Gospel according to the Hebrews shows its conception of the seat of apostolic authority by giving to "James the Just" the place of Peter as recipient of that first manifestation of the risen Lord, which laid the foundation of the faith. Why then does the Greek Palestinian gospel, in contrast with its rivals, lay such special stress on the primacy of Peter?