What is the two-source hypothesis?

4 The Gospels called "Synoptic," i.e. as together representing the same general view (cf. note 3 on Mark), can have been composed under one or other of the following conditions. I. To each Evangelist was separately revealed that which he had to write, independently of the rest (cf. note 10 on Mark). Very few Biblical scholars since Greswell have committed themselves to this theory, but it was that of the Expositor himself, and very much the view of Dr. Harvey Goodwin, when writing contemporaneously as Dean of Ely, before becoming Bishop of Carlisle. II. Each Evangelist reduced oral tradition to a written form in the light of his special purpose and in his own way. Such has been the view of English scholars who followed Gieseler: since the early sixties it has been represented by Westcott, in particular, and by Abbott, Wright and Salmon; in Germany by Zahn, etc. III. Some written record long defunct lies behind all the Gospels in their existing form — the view of Eichhorn and his followers. IV. One of the Gospels is a main source of the rest; the second and the third used one or more additional sources no longer separately extant. During the last preceding generation this has been the view most accepted. The first, the ecclesiastical or "traditional" view, which satisfied most minds from the fifth to the eighteenth century, means that "all the differences between the Gospels were taken as individual variations of a divine type, each variation perfect in its kind" (Nash, pp. 61, 155). The second will be considered here in sections A., B. Whatever is true in the third is generally deemed to be included in the last of the views above stated: these together rely on the notion of dependence as governing the development of the Gospel records (sections C. — F.).

Wellhausen (pp. 73-89) assigns priority in time to Mark. But Weiss, dating "Q" in the year 67, thinks it earlier than Mark's record (so Harnack, p. 193 f.; cf. Sanday "Life of Christ in Recent Research," p. 157). From the fact that it includes no account of the Lord's death, etc., Salmon (p. 247) and Ramsay have revived a suggestion of Paley, the celebrated writer on evidences, that some record was kept in Christ's lifetime; as to this, Loisy ("Gospel Studies," p. 11) remarks that the Twelve were "more accustomed to the fisherman's rod than to the pen of a scribe"; but this would be inapplicable to Matthew, who, moreover, would know both Greek and Aramaic (Zahn). As far as concerns Greek, knowledge of that language must, it seems, be credited to Philip and Andrew (John 12:20 f.) as well as to Matthew. A reason given by some who discredit Paley's idea is, that the vivid expectation which the disciples had of the speedy "completion of the age" would keep them from so recording the Lord's utterances (cf. Westcott Introduction, etc., p. 163). It has to be remembered, however, that the Lord's words about this came only at the end of His ministry. Whatever view be adopted, it must of course take into account the bearing of His words in John 14:26. There can be little doubt that the prevalent impression is that any record was oral only.

Godet (pp. 48-53), whilst admitting use of documents, questioned their being common to the different Evangelists, and Zahn does not support the view that any such document as "Q" was a dominant factor in the composition of Matthew and Luke, whilst, from what has been said in section A. of Salmon's judgment, the reader will be able to appreciate the Dublin scholar's opinion already recorded, that, although he joins others in thinking that Luke was dependent on "Q," our Evangelist obtained his knowledge of it, not by study of any document, but by having heard it read at the weekly gatherings of the Assembly in Antioch. A "Synoptic Table," according to the Two-source Theory, would be found in Holtzmann, Introduction, pp. 376-382. Although Schmidt, describing his own opinion as "in harmony with early tradition," holds that "the attempt to solve these problems by the so-called two-source theory cannot be regarded as successful" (p. 228 f.). Similarly Bartlet in "Oxford Studies," Burkitt ("Gospel History," p. 37), and Allen (Preface to St. Matthew, 7), following B. Holtzmann, as already stated, accept the theory as established before the recent works of Harnack and Weiss appeared. Moreover, currency has been given to it in this country by the Angus-Green popular "Handbook" which is put forth by the Religious Tract Society, so that it is no longer confined to the fraternity of the learned.

When material from the sources dealt with in the last two sections has been discriminated in Luke's Gospel, there is still a residue of its contents to be accounted for, which must be considered in the next section.

The hypothesis that now holds the field is that our canonical Mark is the common source of Matthew and Luke, so far as regards the matter which these Gospels have in common. In his "Manual Commentary" on the Gospels, H. J. Holtzmann already, seventeen years ago, wrote of it as proven (p. 3). This professorial writer is possessed by the idea that "Christianity is a book religion"; so that for him it does not so far, rise above the level of Islam. Not so, happily, his fellow-critic Bernhard Weiss, for whom it is "a life."

Prof. Schmidt in America, and Prof. Orr of Glasgow ("The Resurrection," pp. 63-72), are amongst writers of repute who repudiate the Marcan part of the documentary hypothesis, which Burkitt and Allen in this country fondly imagine to be "among the most assured results" of investigation of the Synoptic problem. Orr has written, "None of the critics defending dependence are able to do more than elbow out the difficulties created by the phenomena set out in Alford's 'Prolegomena."' For the view of Thirlwall and De Wette, see note 6d below.

As it is, those who contend that Luke was dependent on Mark for his narrative portion have, on their own principles, to account for the large amount of matter, chiefly of course sayings, which is common to the first and the third Gospels, but without any equivalent in Mark. This presents itself for consideration in the section which immediately follows.

E. In an earlier section was foreshadowed that portion of the theory now in favour which, in particular, is supplementary to the portion of it dealt with in section D. Matthew and Luke also often agree closely when they do not follow Mark, that is, they have the same material within the limits of our second Gospel, which, nevertheless, has nothing corresponding to their parallels inter se. But neither Luke nor Matthew could have borrowed from the other, because the literary peculiarities of each remain the same as when following Mark (see section C. above).

The matter common only to Matthew and Luke is found to be a record chiefly of the Lord's sayings: see λόγων, verse 4 of the Prologue, R.V. marg. "words," and Luke 24:19, where "things" is divisible between "deed" and "word," for which last cf. John 17:8. The two are combined again in Acts 1:1, "to do and to teach." The deeds of our Lord, as has been seen, are supposed to be accounted for by the Marcan material in Luke (section D.); for the sayings, Weisse propounded a theory, that in the composition of the Gospel of Matthew use was made of another document, which H. Holtzmann, Weizsäcker, B. Weiss, Harnack, etc., think Luke also has turned to account, each compiler being independent of the other in the process. This further phenomenon has already been introduced in the last paragraph of note 10 on Mark (p. 238). It is discussed in, amongst recent books, Hawkins' "Horae Synopticae," pp. 88 ff.; Wernle, "Synoptic Question," pp. 80-91, 224-233; Wellhausen, "Introduction," pp. 65-68; Harnack, "Sayings," and B. Weiss, "Luke's Gospel," chapter 2; "Sources of Synoptic Tradition," chapters 1, 2: and Moffatt, "Introduction," pp. 194 ff.

The critics' belief was suggested by a statement preserved by Eusebius, of Papias, Bishop of Hierapolis (see note 1), which acquires importance for the study of the Gospels of Matthew and Luke. Papias speaks of a collection made by Matthew of Logia (λόγια) of our Lord in "the Hebrew dialect." From the use of this by the compiler of our first Gospel the name under which that goes is supposed to have been derived. Harnack, whilst himself of opinion (pp. 115, 171) that it is probable that this collection, which with B. Weiss he conceives lay primarily behind Luke's Gospel, was (as Salmon supposed) identically the same as that spoken of by Papias, thinks that it is a point which "can neither be proved nor disproved" ("Journal of Theological Literature," 1907. No. 5, p. 13 ff.). Weiss regards the compiler of the Logia as an eyewitness. To this hypothetical document the designation of "Q" has been given, as first letter of the German word for "source" (Quelle).

There remains, however, the question whether any other document, no longer existing, once furnished materials for the Gospels of Matthew and Luke. Another statement of Papias, also reproduced by Eusebius, appears in all the modern literature upon this topic. According to Papias (as well as Irenaeus and Origen), Matthew wrote λόγια (Oracles: cf. Rom. 3:2, Heb. 5:12) in the "Hebrew dialect," by which is understood Aramaic. Each, Papias says, made what he could of this record of "sayings." Its existence is not questioned by W. Kelly, or those like-minded (cf. "Irrationalism of Infidelity," pp. 102 note, 294). This is the document called "Q" by Wellhausen. Bishop Lightfoot ("Essays on Supernatural Religion," p. 173 f.) and Zahn (as Burkitt; p. 135, giving parallels with Mark at p. 147 f.) have thought that it recorded both acts and words of the Lord. Most are of opinion that, while Matthew's Gospel in its Greek form, and Luke may have drawn from this collection, Mark did not; but Ewald, B. Weiss, and Schanz extend its use to the second Gospel also. Of course, the existence of such a collection in Aramaic would carry with it the possibility of a corresponding Greek document. Words of the Lord not in our canonical Gospels seem to have been in circulation. In Acts 20:35 "remember" indicates that Paul was not speaking by revelation. "There can be no doubt," wrote Neander, "that Paul made use of written memoirs of the life of Christ" (p. 7). Some of the Apostle's own writings have not been in like manner rescued from oblivion (1 Cor. 5:9; Col. 4:16).

It is commonly believed that "Q" was written originally in Aramaic, the colloquial dialect (Wellhausen, p. 14, etc.), in which form Pfleiderer and Nestle think that it was used by both Evangelists; but Abbot, Resch, Briggs, and the Jewish scholar Dr. Gaster are more probably right (such also was the opinion of Delitzsch in his later years) in holding that Hebrew was the language of its composition. Aramaic does not seem to have been used in Palestine for literature so early.

The late Bishop Goodwin gave as his judgment in the sixties that "the common written materials used by the Evangelists were not originally in Greek, but were translated into Greek by different hands" (p. xxxii.). It is clear that some record in Greek would early be needed by the Greek-speaking believers of the Dispersion, when it is remembered that there are said to have been no less than three hundred synagogues for the Hellenists (cf. Acts 6) at Jerusalem alone. Although the question of variations arising from different translations (see Jülicher, p. 359) may present itself, as Harnack says (p. 92), in places like Luke 15:4, "wilderness," compared with Matthew 18:12, "mountains," yet this distinguished scholar supports in the main the view of Weiss, that one and the same Greek version of "Q" was used by Matthew and Luke, because their verbal agreements in this element of each Gospel are generally so close.

It will be observed that in chapters 5, 18, 21, 23 f. nothing from "Q" is found noticeable. That document is thought to begin to be clearly discernible in the record of John's Baptism. According to Wernle's analysis, about 48 per cent. of our Gospel consists of discourses.

The text of "Q" as conceived by B. Weiss appears in chapter 1 of his critical commentary. Harnack similarly furnishes a translation of it according to his own analysis. Weiss has always contended that it extends to narrative also; until lately his view has been supported by very few writers; but Harnack now finds in "Q" seven narratives, two of them miracles of healing (p. 163). Orr also has recently expressed his agreement with Weiss on this point. Cf. Schmidt, pp. 219 f., 227 f.

The question as yet remains, had the writer of our third Gospel seen the one going under the name "Matthew." The answer of Hug was that Luke made use of it, and the Roman Catholic professor was followed in this by Greswell, who held that Luke must have seen Matthew's record. A few writers, such as E. Holtzmann, Wendt, and Allen, think that he used it to some extent. The esteemed Roman Catholic professor Schanz adheres to the ancient view that Luke's Gospel was to a large extent derived from both Matthew and Mark.

But the view of most moderns, amongst them Ewald, Meyer, B. Weiss, and now Harnack ("Sayings," p. 112), is that the independence of Luke and of the present "Matthew" was established by Weisse (1838), after the great English scholar Thirlwall (Introduction to Schleiermacher's Essay) in 1825 had expressed the belief that neither of them could have known the other's account of the Infancy and the Resurrection. Current criticism favours the theory that it is the use by Luke of a document lying behind our Matthew which explains a considerable amount of matter being common to these two Gospels alone. This point will be discussed in section E. below.

G. A succinct account having thus been given of the development to the present time of the "Synoptic Problem," a few words may be added by way of summary and appraisement of critical results. We have found that the hypothesis now prominently before students of the Gospels assumes the existence of at least two, most scholars think three, main sources of "Luke." Amongst indications of the synoptist's use in general of various sources is that afforded by what are deemed duplicate records (Weiss, "Sources of Luke's Gospel," p. 119); that is, two records of the same incident or discourse noticeable in Matthew, and even in Mark, in regard to which our Evangelist is believed, like John, to have exercised something like modern "criticism," at least discrimination: as to these "doublets" so-called, see Hawkins, pp. 80-107, if not Wernle, pp. 99-101. Luke 4:16-30 might be compared with Mark 6:1-6, or Luke 15:4-7 with Matt. 18:12-14: see Bennett, p. 141, Hawkins, loc.cit., or Wernle, p.99 ff. Some, as H. Holtzmann (followed by Wernle), find "doublets" even in Luke, e.g., the mission discourse in Luke 11 compared with that in Luke 10; 19:26 with Luke 8:18; Luke 21:14 with Luke 12:11. Such writers discredit the supposition that the Lord spoke or acted in the same or a similar way at different times, in diverse connections, with distinct purposes; but see Godet, l., p. 69.

9 The characteristic portions of Matthew referred to are Matt. 5 - 7; of Luke: Luke 6:20 - 8:3, and Luke 9:51 - 18:14.

10 The natural solution — from the literary standpoint — is doubtless that the Synoptists either copy one from another, or make use of a common source (Burkitt, p. 34; cf. Westcott, pp. 184-186).

As common source, Lessing (1785) supposed (it is all a question of hypothesis, note 6) an Aramaic Gospel of the Nazarenes; and so Eichhorn (1794), whose work was translated into English by Bishop Marsh. Eichhorn held that no Synoptist had used either of the two others. Greswell makes the natural remark (p. 36) that "a primary common source might account for verbal agreements, but not for supplemental arrangement of facts"; and, the more helpful observation (p. 39 f.), that "if you believe in the inspiration of the Gospels, the supposed existence of sources from which the Evangelists derived their materials is not more precarious than unnecessary" (cf. note 13 below). Such is precisely the standpoint of these Lectures.