HAVE WE THE GOSPEL OF MARK IN ITS ORIGINAL FORM?
The number of instances in which Matthew and Luke agree in their changes of Mark has given rise to the theory that Matthew and Luke did not use our Mark but an earlier form. A certain number of such agreements might be passed over as merely accidental. A certain number more might be assigned to assimilation. But if the agreements of Matthew and Luke in their corrections of Mark are so numerous and so striking as to be quite beyond accounting for in these ways, the assumption would be justified that Matthew and Luke used, not our copy of Mark, but one in which the text ran as it now does in those passages where Matthew and Luke agree against Mark.
There are some indications that we do not have the Gospel of Mark in its original form. The conclusion is lacking. This however throws no light on an Ur-Marcus, since the conclusion was lacking in the Mark used by Matthew and Luke.[50]
There are many signs of apparent transposition in our Mark. The insertion of one miracle into the midst of another, as in the case of Jairus’ daughter and the woman with the issue of blood (v, 21-43), might be held to be such a transposition. The incident of the Beelzebul dispute (iii, 20-30) is inserted between the coming of the family ofJesus (iii, 21) to take him home with them, and Jesus’ statement (iii, 31-35), which is the sequel of their coming, about his true brotherhood. The speech about the cursing of the fig tree (xi, 20-26) intervenes between the cleansing of the temple (xi, 15-19) and the demand of the scribes (xi, 27-33) as to the authority by which Jesus has done so unwonted a thing. After this question about authority, and before Jesus’ reply to it, or before the description of the discomfiture of the scribes at the reply, seriously interrupting the connection, comes the parable of the Wicked Husbandmen.[51]
After the story of the transfiguration the prediction of Jesus’ sufferings comes in between the Scribes’ question about Elijah and Jesus’ answer to that question (Mk ix, 11-13). Loisy thinks Mk xiv, 28, out of place. It certainly disturbs the connection. Jülicher considers Mk xiv, 25, to be later and less original than its parallel in Mt xxvi, 29. The saying in xiv, 9, about the name of the woman being known wherever the story of Jesus is told has been suggested as the remark of some preacher or commentator à propos of the occurrence, and not a saying of Jesus. Wellhausen has even suggested that the whole story in xiv, 3-9, may be a later addition. The saying, “Ye shall say to this mountain” (xi, 23) should probably be placed in Galilee, presumably at Capernaum, where with a wave of his hand Jesus could point to both mountain and sea—not in Jerusalem where Mark gives it. Schmiedel considers Mk xiv, 58, secondary. It has been argued, or almost assumed, that the second feeding of the multitude could not have been written by the same hand that described the first, nor the events narrated in the first thirty-four verses of chap. iv have beenwritten in their present order. If one is at liberty to subtract what he will from the Gospel of Mark, and to rearrange its parts somewhat, he can undoubtedly make a much more readable and better arranged Gospel of it than it now is.
DISCUSSION OF THE ANALYSIS OF MARK BY WENDLING AND VON SODEN
Two attempts have recently been made to resolve our Gospel of Mark into its constituent elements, which are sufficiently successful to be noticed here. The first is that of von Soden, in hisDie wichtigsten Fragen im Leben Jesu, and the second Wendling’sUr-Marcus.[52]
Von Soden[53]begins by distinguishing two strands of narrative, easily separable from each other by matter and style. The great differences between these two strands betray two different authors. As the clearest instance of the earlier strand, he takes Mk ii, 1-iii, 6, which he contrasts with iv, 35-v, 43. In the first, all the interest is centered in the words of Jesus; in the second, in the events themselves. “Let one compare the story of the Gadarene demoniac with its twenty verses and the debate about fasting with its five verses, and estimate the weight of the religious value of the thots expressed in the two sections.”
Von Soden next separates Mk vii, 32-37, and viii, 22-26 (the healing of the deaf man and the blind man), as quite distinct in character from such stories as those in ii, 1-12, and iii, 1-6. “In the former, the miracle of healing is itself the subject of the representation; in the latter, the miracle is merely a part of the story, whose real subject is Jesus’ forgiveness of sins and his violation of the Sabbath laws.”
In this way von Soden picks out hisKernstücke. To theseKernstückecertainly belong the group of narratives in i, 21-39; ii, 1-iii, 6; xii, 13-44; iii, 20-35; vi, 1-6; iv, 1-8; iv, 26-32; and x, 13-31; perhaps also vii, 24-30; vi, 14-16; i, 4-11. To these narratives which go back to Peter may also belong the brief notices concerning the stages of growth of the apostolic circle, in i, 16-20; iii, 13-19; vi, 7-13; viii, 27-ix, 1; and ix, 33-40.[54]To these passages von Soden adds xiii, 1-6, 28-37. He says that at the basis of the story of the days in Jerusalem, xi, 1-xii, 12, and the passion narrative in chaps. xiv and xv, lie narratives of a similar style; but these latter he does not include in hisKernstücke.
Von Soden then prints the passages which he thus refers to Peter (or the Petrine tradition), “undisturbed by all that our Gospel of Mark has interwoven with them.”[55]The result presents the Petrine nucleus of the Gospel as follows: John the Baptist and the Baptism of Jesus; a Sabbath in Capernaum; the offense of the Jews at Jesus’ forgiving of sins, his association with sinners, his breaking of the Sabbath, and the fact that his disciples do not fast; how the Jews attempt to take him; how Jesus meets the general misunderstanding; parables about the kingdom of God; the question as to who shall enter that kingdom; the development of the apostolic circle; glimpses into the future.
This makes (with the readjustment in the order of some of the sections) a remarkably straightforward and connected narrative. Von Soden’s remarks concerning it are well worth quoting:
These narratives are without any embellishment or secondary interest. They are plastic and concrete in every feature. Thelocal coloring is strikingly fresh and yet in no way artificial. No edificatory remarks are inserted, no reflections, only deeds and striking sayings. No story requires its secret meaning to be explained by symbol or allegory. In no one of them does one feel any occasion to inquire for the meaning, which lies clear upon the surface. Situations and words are too original to have been invented. Everything breathes the odor of Palestine. There is no reminiscence of Old Testament stories. Miracles appear only here and there, and incidentally.... The christological or soteriological question never constitutes the motive of a story. Not once is there any expression from the language of the schools, especially from that of Paul. Words and sentences are reminiscent of the Aramaic. The figure of Jesus itself bears in every reference a human outline. He is stirred and astonished, he is angry and trembles, he needs recuperation and feels himself forsaken of God, he will not have the thotless, conventional designation “good” addressed to him, and confesses that he does not know when all which he sees to be approaching shall be fulfilled. His mother and his sisters fear that he may be out of his mind. This and much else is told with the greatest naïveté. So Jesus lived; so he expressed himself; thus they received him; thus the apostolic circle was formed and developed—this is what the writer intends to tell.[56]
These narratives are without any embellishment or secondary interest. They are plastic and concrete in every feature. Thelocal coloring is strikingly fresh and yet in no way artificial. No edificatory remarks are inserted, no reflections, only deeds and striking sayings. No story requires its secret meaning to be explained by symbol or allegory. In no one of them does one feel any occasion to inquire for the meaning, which lies clear upon the surface. Situations and words are too original to have been invented. Everything breathes the odor of Palestine. There is no reminiscence of Old Testament stories. Miracles appear only here and there, and incidentally.... The christological or soteriological question never constitutes the motive of a story. Not once is there any expression from the language of the schools, especially from that of Paul. Words and sentences are reminiscent of the Aramaic. The figure of Jesus itself bears in every reference a human outline. He is stirred and astonished, he is angry and trembles, he needs recuperation and feels himself forsaken of God, he will not have the thotless, conventional designation “good” addressed to him, and confesses that he does not know when all which he sees to be approaching shall be fulfilled. His mother and his sisters fear that he may be out of his mind. This and much else is told with the greatest naïveté. So Jesus lived; so he expressed himself; thus they received him; thus the apostolic circle was formed and developed—this is what the writer intends to tell.[56]
These sections of Mark certainly have a very primary character; so far as their contents is concerned, they may well go back to the Petrine tradition.
With these sections von Soden contrasts the remaining parts of the Gospel, in which he finds not only much interruption of the primary narrative, but much interpretation, much allegorizing, much absence of actual situations, much reminiscence of Old Testament stories, much influence from Paul, and many reflections of the experiences of individual Christians and the Christian church.[57]No one can work thru this analysis of vonSoden’s without feeling that it is easy to distinguish between primary and secondary elements in the Gospel of Mark, and that von Soden has at least pointed out many of the junctures between these two.
The attempt of Wendling in hisUr-Marcus[58]is still more thorogoing. The basis of his discussion is Mark’s 4th chapter, where he considers the two strands most easily separated. To the original belong iv, 1-9, and vss. 26-33. Vss. 10-25 are later; they have been inserted mechanically, yet so as to respect the older text; they have no organic connection with the rest of the chapter, and even contradict its situation. Jesus is teaching from a boat (and other boats are with his); then suddenly, in vss. 10-25, he is alone with his disciples who ask him the meaning of the parable of the Sower. He gives his explanation, and again without any indication of change of situation he is in the boat surrounded by the other boats, with the people still on the shore, and the storm comes up and is stilled.
This little insertion (iv, 10-25) also contains theories of the writer, quite contradictory to those of the writer of other parts of the Gospel. In other places, Jesus speaks to all the people in parables “as they were able to hear him”; he stretches out his hand over the multitude of his disciples and says, “These are my mother and my sisters”; he is the teacher of the crowd, who understand him better than his own family; there is nothing in his parables that needs explaining. But in this insertion (iv, 10-25) the theory of the writer is that the parables are “mysteries,” enigmas, which not only need to be explained (by the allegorical method), but which are spoken for the express purpose of preventing the peoplefrom understanding. Without the key which Jesus gives, even the disciples do not understand them. The section is also marked by Pauline influences.[59]
Two clews are thus given, aside from interruptions in the narrative, by which the work of a second writer may be detected. He has theGeheimnis-Theorieof the parables, and he has in thot and vocabulary reminiscences of the Pauline school. Applying these tests to another section which seems to interrupt the narrative where it stands, Wendling adds a second insertion—iii, 22-30. This is the section about the dispute with the Pharisees, which comes in inaptly between the introduction (iii, 20, 21) and the continuation (iii, 31) of the story of Jesus’ family who have come to take him home. It seems to have been inserted in this place because the Pharisees also said “he hath a devil.” By repeating in vs. 30 the ἔλεγον ὅτι which he found in vs. 21, the redactor preserves for the continuation of the original story precisely the same connection it would have had without his interpolation; and by the use of the same words in vs. 22 he connects the interpolation with the opening narrative. His hand is seen in the superfluous repetition of words, especially of the subject, as in iii, 24, 25.[60]
To these two insertions should be added a third, iii, 6-19. The motives for it seem to be copied from narratives in other chapters. It consists (in part) of generalization and interpretation, both marks of the redactor’s work. It also contains hisGeheimnis-Theorie.
To these should be added i, 34b(“he suffered not the demons to speak, because they knew him”), because ofthe presence in it of this same theory. Nor does i, 45, fit where it is; the connection without it is good; it also contains the favorite theory of the redactor that the more Jesus told people not to proclaim him, the more they did so, and the more he tried to seclude himself the more they found him.
To these again, on somewhat other grounds and not so securely, should be added the little groups of loosely strung logia which are found in vi, 7-11; viii, 34-ix, 1; ix, 40-50; x, 42-45; xi, 23-25; xii, 38-40; xiii, 9-13. The ground for asserting these to be additions is that these logia are not closely connected in the passages in which they occur, and that they share this characteristic with the similar group of disconnected sayings in the first and best attested interpolation, iv, 21-25.
In i, 1-3, 14b, 15, the word εὐαγγέλιον arouses a natural suspicion. The same word also occurs in four other places (viii, 35; x, 29; xiii, 10; xiv, 9), all of which are in passages which are suspicious upon other grounds; consequently with the three instances in chap. i, they are ascribed to the redactor.
With the exception of the interpolation in iv, 10-25, the section i, 16-iv, 33, appears to be a unit, and belongs to the oldest stratum. But with iv, 35, says Wendling, begins a new section, easily distinguished from that just mentioned. It copies the motives and the characteristics of other sections.[61]The writer is to be distinguished, however, not merely from the writer of the earliest stratum, but from the author of the insertions already identified. None of the criteria of the latter’s manner appear in the section beginning at iv, 35. It showsno trace of Pauline conceptions, has none of Jesus’ prohibitions to the demons, itsHeimlichkeitis of a different sort, and goes back to Old Testament exemplars. And since the insertion in iv, 10-25, presupposes the story of the storm on the lake in iv, 35-v, 43, this latter is older than the former. The writer of this section (iv, 35-v, 43) therefore stood between the writer of the original strand, and the evangelist or redactor. The last writer (Wendling calls him Ev) worked over the combined work of his two predecessors.
To the author who is intermediate between the first writer and the Evangelist, Wendling assigns twenty-nine different sections, some of considerable length and some of only a verse or part of a verse. They are as follows: i, 4-14a; iv, 35-v, 42; v, 43b; vi, 14, 17-30, 35-44; ix, 2-8, 14-27; x, 46-xi, 10; xiv, 12-20, 26-35a, 36-37, 39-41a, 42, 47, 51-56, 60-62a, 63, 64, 66-72; xv, 16-20, 23, 24b, 25, 29-30, 33, 34b-36, 38, 40-43, 46-xvi, 7a, 8—about two hundred verses or parts of verses in all.
The contributions of the author of the Gospel are more extensive than those of his predecessor. They comprise i, 1-3, 14b-15, 34b, 39b, 45; ii, 15b-16a, 18a, 19b-20; iii, 6-19, 22-30; iv, 10-25, 30-32, 34; v, 43a; vi, 1-13, 15, 16, 30-31, 45-viii, 26, 30b-33a, 33c-35, 38-ix, 1, 9-13, 28-50; x, 2-12, 24, 26-30, 32b-34, 38-40, 45; xi, 11-14, 18-25, 27a; xii, 14b, 32-34a, 38-44; xiii, 3-27, 30-32, 37; xiv, 8, 9, 21, 35b, 38, 41b, 57-59, 62b; xv, 39, 44, 45; xvi, 7b, in all about two hundred and seventy verses or parts of verses.
This leaves to the original writer the following sections: i, 16-34a, 35-39a, 40-44; ii, 1-15a, 16b-17, 18b, 19a, 21-iii, 5, 20, 21, 31-iv, 9, 26-29, 33; vi, 32-34; viii, 27-30a, 33b, 36, 37; x, 1, 13-23, 25, 31-32a, 35-37, 41-44; xi,15-17, 27b-xii, 14a, 14c-31, 34b-37; xiii, 1-2, 28-29, 33-36; xiv, 1-7, 10, 11, 22-25, 43-46, 48-50, 65; xv, 1-15, 21, 22, 24a, 26-27, 31-32, 34a, 37, in all about two hundred and twelve verses or parts of verses.[62]
Wendling calls the writers of these three strands M1, M2, and Ev. Printing the text of M1 and M2 without rearrangement, but with the omission of all matter assigned to Ev, he finds them to make a continuous story, well connected and without breaks. Whether M1 alone makes such a story, he is in doubt; and therefore as to whether M2 found M1 as a connected discourse, or himself first assembled the sections of it in connection with his own additions, the same doubt exists. The passion-story of M1 by itself seems to be a connected account; it may therefore be assumed that so much of M1 was found by M2 as a whole and in its present order. Further, since the work of Ev in the passion-story is so slight, it is to be assumed that the combination of M1 and M2 in this story was more carefully done than in many other parts, and also that for this part of the gospel history Ev possessed very few traditions which had not already been embodied in M1 + M2. This would agree with the natural assumption that the earliest part of the gospel tradition to be carefully treasured would be that relating to Jesus’ death, and that it was only later that the attempt was made to preserve with equal care the story of his whole public career.
When one remembers the fine-spun analyses of the historical books of the Old Testament, which, long ridiculed for their elaborateness, have finally been acceptedby most scholars, one hesitates on this account alone to pronounce an adverse judgment upon Wendling’s theory. Yet his analysis certainly seems over-elaborate. It is a great advantage to be able to distinguish the more obvious work of the redactor from the earlier document upon which he worked. All students will feel this with reference to chap. iv, and the advantage in chap. iii is perhaps only less great. Still more welcome is the assignment of vi, 45-viii, 27, to the redactor. The great stumbling-block of this section is its feeding of the four thousand, so obviously copied from the feeding of the five thousand. That one and the same author should have written both these accounts has seemed strange to many readers. But this duplication is as easily disposed of upon von Soden’s theory as upon Wendling’s. Von Soden’s analysis into two strata (without the assumption of two writers) is much simpler than Wendling’s analysis into three, with three writers. Wendling’s theory is more secure where it goes with von Soden’s, and less convincing where it goes beyond it.
Some distinction has in any case to be made between the final writer of the Gospel and the earliest tradition upon which he worked; and Wendling has indicated the criteria which such a distinction must employ. Von Soden’s division of the Marcan material into a Petrine and a later source amounts to the same thing. The two critics do not differ greatly about the passages they regard as secondary. Von Soden’s Petrine narrative does not differ greatly from Wendling’s M1 + M2. But the line of demarkation between M1 and M2, and Wendling’s reasons for drawing this, are not as self-evident as the line which Wendling and von Soden agree in drawing between the earlier document, or source, and the work of the Evangelist.
CONCLUSIONS OF VON SODEN AND WENDLING COMPARED
A tabulation of the results discloses the following agreements and disagreements between von Soden’s Petrine narrative and Wendling’s M1+M2.
The comparison shows Wendling’s analysis to be much more complex than von Soden’s. This results from his separation of his groundwork into two strands. It also shows that Wendling assigns considerably more to M1 and M2 than von Soden to his Petrine source. This Wendling can afford to do, since he supposes two documents instead of one. The matter assigned by von Soden to the Petrine source is in part assigned by Wendling to M1 and in part to M2. E.g., i, 4-11, is assigned by von Soden to the Petrine source, and by Wendling to M2; but i, 16-39, is assigned to the Petrine source, and (with the exception of two parts of verses) to M1. Thepassage ii, 1-28, is assigned by von Soden to the Petrine source, by Wendling to M1 (again with exception of a few parts of verses). Of the one hundred and seventy-seven verses assigned by von Soden to his Petrine source, up to and including xiii, 37 (after which he so assigns nothing), Wendling assigns about one hundred and twenty-four to his M1, and only ten to M2. Tho he assigns some verses to M1 which von Soden does not give to the Petrine source, and omits some (assigning them to the redactor) which von Soden does so assign, up to xiii, 37, the M1 of Wendling agrees very closely with the Petrine source of von Soden. The material assigned to M1 and M2 after xiii, 37, is about equally divided between them. Wendling makes no claims for the Petrine origin of his M1 or M2, but after these are subtracted from the whole Gospel there is a smaller amount left for the work of his redactor than remains after the Petrine source is subtracted. Since Wendling distinguishes between two sources and the work of the redactor, and von Soden only between the Petrine tradition and other matter, this result also is what would be expected.
The relatively great agreement of the results of these two investigations seems to prove that it is possible to distinguish an earlier and a later tradition in the Gospel. Beyond this, the difference between von Soden and Wendling is that the former makes no assertions concerning the identity of the final editor with the writer who recorded the Petrine tradition, while the latter asserts that the redactor is quite another person than the writer of either M1 or M2. Is this latter position of Wendling’s susceptible of proof or disproof?
Perhaps the simplest criterion, and the one to be most safely applied, is that of vocabulary. Sir John Hawkinscompiled a list[63]of forty-one words which he regards as characteristic of Mark. Do these words occur indiscriminately in M1, M2, and Ev, or are they confined some of them to M1, and some to M2, and some to Ev? Or is there sufficient difference in the frequency with which these words occur in the three strata to justify the assumption of three different authors, and especially that Ev was distinct from the writers of the two documents? If not, the division between earlier and later material in Mark may still stand, but it may have been one and the same writer who put the whole Gospel together out of these earlier and later materials.
Characteristic of Mark[64]is the historic present. Hawkins finds one hundred and fifty-one examples of this use in Mark against seventy-eight in Matthew (twenty-one of these taken from Mark),[65]and four in Luke. Of these one hundred and fifty-one historic presents in Mark, forty-nine occur in passages assigned by Wendling to M1, sixty-nine in M2, and thirty-three in Ev.
Of the peculiarly Marcan words, some prove nothing in this connection. Εὐαγγέλιον is used only by Ev (seven times); but since Wendling uses the presence of this word as a criterion of Ev’s work in six out of the seven passages where it occurs, this adds nothing to the proof. Ἄλαλος is used once by M1, twice by M2, and not by Ev. But since Ev adds no story of a dumb man, he has no occasion to use the word. (He does add a story of a stammering man, where he uses the word, μογιλάλος.) Κλάσμα, used once by M2 and three times by Ev, signifies little; sincethe three uses in Ev occur in the same passage, and this passage is a copy of the passage in M2 (the feeding of the multitudes). Στάχυς occurs three times, all in M1, but this also signifies nothing, since no passage in which it could occur is assigned to M2 or Ev. Ἐκπορεύομαι is used twice each by M1 and M2, and seven times by Ev; but since five of these seven occurrences are in the same passage, they cannot establish any particular fondness for this word on the part of Ev as against the other two. Εἰσπορεύομαι looks a little more favorable for Wendling’s hypothesis, since it is used once by M1, twice by M2, and five times, in separated passages, by Ev. Ἀκάθαρτος, found three times in M1, four in M2, and three in Ev; ἀπὸ μακρόθεν, three times in M2 and twice in Ev; διδαχὴ, used three times by M2 and twice by the redactor, and φέρω, five times used by M1, eight times by M2, and twice by Ev, do nothing toward establishing a distinct vocabulary for any one of the three. Only two words, διαστέλλομαι, used four times by the redactor in four different chapters, and not by M1 or M2; and ἐκθαμβοῦμαι, used only by M2, four times in three different chapters, point in the direction of distinct vocabularies. But the absence of the third of these words can certainly, and of the second probably, be accounted for by the subject-matter.
There is here practically no evidence of distinct vocabularies. Even if there were, it would be fully offset by the use of words having no necessary connection with any particular subject-matter, and therefore equally likely to occur in any part of the Gospel. Five such words are the adverbs εὐθὺς, πάλιν, πολλὰ, οὐκέτι, and οὔπω. Of these, the first (Mark’s most characteristic word) is used seventeen times by M1, fifteen by M2, and ten by Ev.Considering the relative amounts of narrative matter ascribed to the three, this usage seems to indicate an equal fondness for this word among them. The second (πάλιν) is used ten times by M1, eight times by M2, and nine times by Ev; the third (πολλὰ) is used adverbially three times by M1, six times by M2, and three times by Ev; the fourth (οὐκέτι), twice by M1, twice by M2, three times by Ev; the fifth (οὔπω), once by M1 and four times by Ev.
Characteristic of Mark also is his use of the imperfects ἔλεγεν and ἔλεγον. They are found fourteen times in M1, fifteen times in M2, and twenty-one times in the passages ascribed to Ev.
Of the forty-one verses listed on p. 246 as standing in both Mark and Q, thirty-four are in passages assigned by Wendling to Ev. This would seem to tell in Wendling’s favor, since the last writer who had a hand in the making of the Gospel of Mark would naturally be the one most likely to make use of Q. Three verses, however, occur in passages assigned to M1, and four in M2. This would indicate that all three writers, besides having the same favorite words, were acquainted with and made some use of Q. The item of the relation of the various writers to Q, however, has little or no significance; since it is the sections having the greatest amount of logian matter and the least narrative, that are assigned to Ev.
The cumulative effect of these considerations is very much to the discredit of Wendling’s assumption of three different writers for our Gospel of Mark. It cannot, to be sure, disprove that assumption; but it at least shows a lack of proof where proof would be most easily found and most convincing.
MATTHEW AND LUKE USED OUR MARK AS A SOURCE
Even if Wendling’s analysis had been capable of substantiation on linguistic grounds, his division of our Gospel of Mark into three strands from three different authors would not help us toward an Ur-Marcus lying behind our Gospels of Matthew and Luke. For Matthew or Luke or both of them follow Mark in all the transpositions, dislocations, and other misarrangements of his Gospel. Whether these features stood in the original Mark or not, they evidently stood in the Mark used by Matthew and Luke.
Matthew and Luke also used a Mark which contained the story of Jesus in the same order given by our present Mark. Tho both of them deviate from this order for assignable reasons, one or the other of them is found following it all the time. If these deviations go back to an Ur-Marcus, there must have been one Ur-Marcus in the hands of Matthew and another in the hands of Luke.
THE HYPOTHESIS OF A PRIMITIVE MARK SUPERFLUOUS; SIMPLER EXPLANATIONS
Can the verbal agreements of Matthew and Luke as against Mark, or their deviations from him without apparent reason, be explained upon any simpler hypothesis than that of Ur-Marcus? It appears to the writer that they can.
A certain number (tho no one can say exactly what proportion of the whole) of the agreements of Matthew and Luke against Mark may be allowed to be accidental. Many of them, like the substitution of εἶπεν for λέγει, or of an occasional δέ for Mark’s invariable and monotonous καί or the substitution of a common for an uncommon word (like κλίνη for κράβαττος) require no explanation.
Agreementsof Matthew and Lukein their omissionsfrom the Marcan narrative do not stand upon the same plane with the agreements insubstitutions, and may all be accounted for on the ground of accident, or by the same desire on the part of both writers to be more concise, or to avoid anything derogatory to Jesus or the apostles; or by some other similar motive at work separately in the minds of the two later evangelists. It is only the agreements in corrections and substitutions that require accounting for. I believe these can be explained chiefly on two grounds:
1. It is not necessary to assume that Matthew and Luke both worked upon the same identical copy of our Mark. If they used two copies, these two would not be expected to agree absolutely with each other in the wording of every passage. This would account for some of the slight deviations in the wording of either Matthew or Luke where the other agrees with our present Mark. These two copies that Matthew and Luke used may neither of them have been the original (since in both of them the conclusion at least was gone); or at least not the original in its original form. One of them may have been a copy of the original, and the other a copy of this copy. Or they may both, as Sanday argues, have belonged to a type later, and not earlier, than our present Mark. This would account for the agreements, and for such deviations as have not already been accounted for, or cannot be accounted for, by the known literary peculiarities of Matthew and Luke. Since the text of Mark that has come down to us is more corrupt than that of either Matthew or Luke, various words in which Matthew and Luke now agree against Mark may have stood in the text which both of them used, and may later have dropped out, before the copy was made to which our present textsgo back. Or the two copies of Mark, assumed above, may both have been made from the original copy which Mark made with his own hand. Upon this supposition even, they would not always agree, and so deviations in Matthew and Luke from Mark, and occasional agreements in such deviations, would be explained. Or these agreements may be explained, as is obvious in many instances, by the working of similar motives in the minds of Matthew and Luke, even assuming them to have made their extracts from one and the same copy, or from two practically identical copies, of Mark.
Dr. E. A. Abbott, in hisCorrections of Mark(London, 1901) gives an exhaustive list of the deviations from Mark in which Matthew and Luke agree. Many of these are such as to suggest that Matthew and Luke used not an Ur-Marcus, but a text of Mark later than the one that has come down to us. E.g., in twelve instances Matthew and Luke agree in supplying the subject or object which our Mark omits. In fifteen, they agree in correcting abrupt constructions, supplying a connecting word. In thirteen (exclusive of λέγει) they agree in correcting Mark’s historic present. In twelve they agree in replacing Mark’s relative clause or his subjunctive by a participle. In twenty-three they agree in substituting εἶπεν for λέγει. In thirty they agree in the use of δέ for καί. It is not impossible that Matthew and Luke, independently bent on improving Mark’s style, have accidentally agreed in making these same improvements in the same places (especially since there are other improvements of the same sort in which they do not agree). But it is a much simpler and more adequate hypothesis, that they both used a text of Mark in which these corrections had already been made.
Yet even of this text they probably did not use the same identical copy. And as the copy used by one or both of them may have been two or three removes from the text from which it started, many changes may have crept into the copy used by one of them, not contained in the copy used by the other. This would account alike for the agreements in deviations from our present Mark, and for the fact that these corrections are not all of them found in both Matthew and Luke. This last item is further accounted for by the freedom of Matthew and Luke in making their own corrections in the copy that lay before them.[66]Allowance should also be made for the fact that we cannot be sure that we have yet recovered the true text of either Matthew or Luke.[67]
2. The agreements of Matthew and Luke against Mark can further be accounted for by the hypothesis of assimilation. Matthew made certain changes of his own in the wording of Mark; Luke apparently made many more. The various texts still extant show many efforts of copyists to bring the deviations of Matthew and Luke in small verbal items into an agreement. If this same process went on during the period covered by our earliest manuscripts, it is probable that it went on to a much greater extent at an earlier date, before our Gospels had acquired the sacredness which they later came to possess. A fine illustration of this process and its results is to be seen in the Matthean and Lucan versions of the Lord’s Prayer, in which the probably original “Let thy Holy Spirit come upon us and purify us,” of Luke, has been assimilated to “Thy kingdom come,” and in many manuscripts also to “Thy will be done as in heaven so upon earth,” ofMatthew. The extent of this sort of assimilation can never be determined; but it seems quite sufficient to account for agreements of Matthew and Luke against Mark not easily accounted for on other grounds.
A more general reason against the assumption of an Ur-Marcus in the hands of Matthew and Luke is the comparatively small number and importance of their agreements against Mark, as compared with the very large number of the deviations in which they do not agree, and as compared also with the vastly greater number of instances in which both Matthew and Luke follow Mark faithfully. In other words, if Ur-Marcus differed from our Mark only in those words and phrases in which Matthew and Luke agree against our Mark, then Ur-Marcus was at the most not a different Mark from ours, but only a different copy or text of our Mark. The assumption of an Ur-Marcus was a natural one for the explanation of the phenomena in question; but it is a cumbersome hypothesis, and insecure; further study seems to discredit it. Matthew and Luke used our Mark, not another.
It has often been suggested that the Marcan material covered by the “great omission” of Luke (Mk vi, 45-viii, 26) was absent from the copy of Mark used by Luke, tho present in that used by Matthew. Reasons for Luke’s omission of this long Marcan section have been given, and seem sufficient without the assumption of its absence from Luke’s copy of Mark. But the theory of its absence has also important items directly against it. The section has the general Marcan characteristics. Mark has one hundred and forty-one historic presents; eighteen of them are in this section. He uses εὐθὺς thirty-four times, five in this section; πάλιν twenty-six times, five in this section. He is partial to the imperfects ἔλεγεν andἔλεγον, which he uses fifty times (against Matthew’s twenty-three and Luke’s nine), six times in this section. The same habit of duplicate expression which occurs in other parts of his Gospel appears here. ὅ ἐστιν in the sense of “i.e.,” peculiar to Mark among the evangelists, appears here twice (four times elsewhere in the Gospel). Seven out of the nine sections begin with καί. The section seems to be too homogeneous with the rest of the book to be from a different hand.[68]
The foregoing considerations seem to render the hypothesis of Ur-Marcus superfluous. The phenomena for which it was designed to account are more easily and naturally explained by other suppositions.
SOME REMARKABLE VERBAL RESEMBLANCES
In the preceding pages sufficient consideration has been given not only to the fact, but to the manner, of the use of Mark by Matthew and Luke. Visual illustration, by the printing of a few passages in different kinds of type may serve to enforce some of the more general facts already brot out. The words (or parts of words) common to the three Synoptics, in the following passages, will be printed in heavy-faced type.
Here the evangelists differ each from the other in the words ascribed to Jesus, but when they come to the parenthetic explanation injected into the midst of the sentence, ἵνα δὲ εἰδῆτε, etc., they agree exactly, not only in the wording, but in the awkward placing of the clause. The three accounts agree in the first five lines, except for the presence of γὰρ in Matthew, the insertion of τῷ παραλυτικῷ in Mark, and a slightly different form of the verb ἀφίημι in Luke. In the fourth line Luke also inserts σοι, after which come seven consecutive agreeing words (tho with slight rearrangement in order by Luke). Mark then has a clause of six words which Matthew and Luke omit. The latter agree in substituting περιπάτει for ὕπαγε, and two (different) words from the same root for Mark’s κράββατον. Luke has preserved the σοὶ λέγω which Matthew has dropped.
Few brief passages in the triple tradition will better repay study than this. Note that the three introduce their question with three different particles. Matthew and Luke omit the apparently superfluous words of Mark, χρείαν ἔσχεν, but Luke retains the αὐτὸς of Mark which Matthew has dropped. Luke adds ὄντες, perhaps in deference to Mark’s οὖσιν, used in a similar phrase but different connection. He substitutes ὡς for the πῶς of Mark and Matthew. Mark and Luke both have the statement that David “gave” the bread to those that were with him, Luke adding that he “took” it. All three have in conclusion the phrase “to those with him,” but each has inserted it in a different place. Matthew follows Mark more closely than does Luke, the latter transposing one or two clauses. Both Matthew and Luke have omitted the reference to Abiathar, either because they (or Luke at least) had no interest in it, or for its historical difficulty. In spite of these changes there is a most remarkable verbal agreement thruout. Except for Mark’s superfluous “had need,” and his reference to Abiathar, nothing can be found in either account that is not duplicated, practically word for word and almost letter for letter, in one or both of the others.
This passage contains the striking addition of the parenthetical explanation ἦσαν γὰρ ἁλεεῖς. That this should occur in a narrative portion, and not in a saying of Jesus, is the more significant. For the rest, the saying ascribed to Jesus runs word for word (tho its brevity in this case robs this fact of any very remarkable significance); in the narrative portion Matthew mentions that Simon was called Peter (a remark which Mark saves till he comes to the formal naming of the twelve), and in the conclusion he says “they left the boat and their father,” while Mark says “they left their father in the boat,” adding, “with the hired men.” Mark says Jesuscalledthe two “immediately.” Matthew says theyleft“immediately.”
USE OF A COMMON DOCUMENT BY MATTHEW AND LUKE
The document used by Matthew and Luke as the source of their common non-Marcan material was for some time generally identified with the “Logia” which Papias says Matthew, the disciple of the Lord, wrote in Hebrew, undoubtedly meaning Aramaic. Until some sufficient justification for this identification has been given, it seems better to refer to the common non-Marcan source of Matthew and Luke under the more colorless symbol Q.
The common non-Marcan tradition of Matthew and Luke consists almost exclusively of logian material. It contains a few parables, brief, and dealing usually with the “kingdom of heaven,” and one or two sections (such as that concerning the centurion from Capernaum, and the Temptation) which may quite properly be regarded as narrative, but which also contain large logian content and may have been introduced for the sake of the sayings.
The proof that the source of the common non-Marcan material of Matthew and Luke was a document and not an oral tradition lies in the extent and character of the agreements between the two Gospels; it cannot be summarized in a paragraph, but comes out only in a detailed examination of the double tradition such as is undertaken in the following pages.
Before the theory of a common documentary source for the non-Marcan material in Matthew and Luke canbe accepted, it must defend itself against two apparently simpler hypotheses, viz., that Matthew copied from Luke or Luke from Matthew.
Did Matthew copy from Luke? His genealogical tree does not agree with Luke’s.[69]He betrays in his story of the birth at Bethlehem no knowledge of the fact that Joseph’s home was originally at Nazareth. This latter place he first mentions in ii, 23, as the place to which Joseph went upon his return from Egypt. Matthew has a greater interest in John the Baptist than has Luke, as is indicated by his fuller treatment of the fact and circumstances of his death, contrasted with Luke’s leaving him in prison undisposed of. Yet Matthew does not employ the material concerning the preaching of John, which Luke has embodied in his iii, 10-14. Matthew makes a specialty of the sayings of Jesus, yet omits many that Luke contains. In short, the reason for denying that Matthew copied from Luke is the impossibility, upon that hypothesis, of explaining the omissions of Lucan material from Matthew’s Gospel, and the very great divergences between the two Gospels where such divergences would not be expected with either one using the other as an exemplar.
The same argument which refutes Matthew’s use of Luke refutes Luke’s use of Matthew.
But it may be added, that upon either of these hypotheses it becomes impossible to explain the changes which appear to have been made by both Matthew and Luke in the material common to them, both in its wording and its order. If Matthew copied from Luke, he wouldnaturally have followed his order, which he does not do. Or, deviating from that order for obvious reasons, he would naturally return to it when those reasons no longer prevailed, which he does not do. Or if Luke copied from Matthew, he could hardly have inserted a genealogical tree which is at variance with Matthew’s, in the unnatural place where it now is, as against the natural place in which he found it in Matthew. Nor could he, when he had the Sermon on the Mount before him in the form in which Matthew gives it, break it up into little pieces and scatter it up and down thruout his Gospel. Moreover, in the sayings common to Matthew and Luke it is now one and now the other who preserves what we must consider the most original reading; as when Matthew says, “Cleanse first the inside of the cup,” and Luke in place of this says, “Give alms of that which is within.” But again it is not Matthew but Luke who gives the more original form of a saying; as when Luke says “Blessed are ye poor,” and Matthew says, “Blessed are the poor in spirit.”
The phenomena of peculiar words, ninety-five characteristic of Matthew and one hundred and fifty-one characteristic of Luke, is also impossible of explanation upon the theory that either writer copied from the other. If either one were copying from the other, they would certainly agree against Mark in some really important matter, and not merely in an occasional word or phrase. If Luke were copying from Matthew, he would certainly have incorporated some one of those numerous additions which Matthew makes to the narratives of Mark.[70]
In addition to any of the more general considerations which have suggested the possible use of Matthew byLuke, a recent writer has evolved an ingenious and somewhat elaborate proof for this use, which it may be well to consider in some detail.
A RECENT ATTEMPT TO PROVE MATTHEW A SOURCE FOR LUKE
Mr. Robinson Smith[71]attempts to dispose both of Ur-Marcus and Q by maintaining that Luke copied from Matthew. His argument rests upon the deviations which Matthew and Luke make, respectively, in their common abbreviations of certain of Mark’s narratives. “Where a choice from two or more Marcan expressions has been made, the first choice falls to Matthew and the second to Luke.”
As examples of these first choices by Matthew and second choices by Luke, Mr. Smith instances (with the parallel passages in Matthew and Luke) Mk i, 32; iii, 7, 8; x, 29, 33, 34; xii, 3; xiv, 1, 12, 65; xv, 42. The argument seems to be that Luke having both Mark and Matthew before him, and seeing that in each of these instances Matthew has chosen a certain part of Mark’s phrase and rejected the rest, himself avoids using that part of the phrase which Matthew has chosen, restricting himself to the part which Matthew has left unused. We will take up first the particular instances, and see whether other, perhaps simpler, reasons suggest themselves for these deviations; after that we will consider the general argument.
Mk i, 32 (Mt viii, 16; Lk iv, 40): Mark’s phrase runs Ὀψίας δὲ γενομένης, ὅτε ἔδυ ὁ ἥλιος. Of this phrase, Matthew takes the first three words as they stand. Luke appropriates the remainder, changing into Δύνοντος δὲ τοῦ ἡλίου.Mark’s phrase is here redundant, and Matthew and Luke (as usual) both reduce the redundancy. But Matthew has omitted the point of Mark’s phrase, since in Matthew’s account the events described did not happen on the Sabbath. Luke has retained the essential part of the phrase.[72]
Mk iii, 7, 8 (Mt iv, 25; Lk vi, 17): “Mark gives in order and by name six districts from which the multitudes came. Matthew mentions all save the last, Tyre and Sidon. Luke omits the first, fourth, and fifth, but does mention the last, Tyre and Sidon.” The changes in these lists seem to be more various than Mr. Smith suggests. Matthew adds Decapolis and omits Idumaea.[73]The thing hard to account for in Luke’s list is his omission of Galilee, not his inclusion of Tyre and Sidon. These latter regions would interest him especially, with his universalistic tendency; we should hardly have been surprised to find him adding them if he had not found them in Mark. A simple explanation of the changes made by both Matthew and Luke may perhaps be seen in Matthew’s Judaistic tendency, which led him to omit Tyre and Sidon, and in Luke’s universalistic tendency which made him include them. To make Mr. Smith’s argument hold in this case, Luke should certainly have come much closer than he does, to preserving the parts which Matthew rejects, and rejecting the parts which he retains. It appears that Luke has no great knowledge of nor interest in Palestinian geography, but Tyre and Sidon suited his purpose.
Mk x, 29 (Mt xix, 29; Lk xviii, 29): Mark here has ἕνεκεν ἐμοῦ καὶ ἕνεκεν τοῦ εὐαγγελίου. Matthew hasἕνεκα τοῦ ἐμοῦ ὀνόματος, and Luke εἵνεκεν τῆς βασιλείας τοῦ θεοῦ. But Matthew’s “my name’s sake” is not the same as Mark’s “my sake,” and seems to bespeak Matthew’s later date of writing. Luke’s “for the sake of the kingdom of God” has a more primitive sound than the latter part of Mark’s phrase. It probably represents the original words of Jesus which Matthew has everywhere changed into the “kingdom of heaven.” Since all the passages in Mark where the word εὐαγγέλιον occurs are on independent grounds suspected of being later additions, it seems probable that the reading of Mark which Matthew and Luke had before them here was merely ἕνεκεν ἐμοῦ, that both Matthew and Luke changed this phrase as they would, and that the ἕνεκεν τοῦ εὐαγγελίου of Mark is later than either Matthew or Luke. At all events, it does not seem to be true in this instance that Matthew takes the first part of Mark’s phrase and Luke the last.
Mk xii, 3 (Mt xxi, 35; Lk xx, 10): Matthew’s account here is quite different from Mark’s (which is followed much more closely by Luke). According to Mark, only one servant was sent, whom the vineyard-keepers “caught and beat and sent away empty.” According to Matthew several servants were sent; the vineyard-keepers caught them, beat one, killed one, and stoned another. This form of the story indicates the times of persecution in which it was worked over by Matthew—when more than one man had suffered more than one kind of indignity. Luke sticks close to the story of Mark, and merely omits the λαβόντες which Matthew retains. Perhaps Luke had reflected that the servant had to be caught if he was to be beaten, and so regarded the item as superfluous. It does happen to come before the items that Luke retains,but there is no reason to suppose Luke would have had any greater antipathy to omitting it if it had stood last or if Matthew had also omitted it. It is not only hard to detect any influence of Matthew upon Luke here, but much harder to see, if Luke were copying Matthew, why he should not have preferred his several servants to Mark’s one. Later in the same story, Luke again omits Mark’s λαβόντες where Matthew retains it (Mk xii, 8), tho here both Matthew and Luke change the order of the incidents in the verse, probably to make them conform more exactly to the experience of Jesus. The omission of the participle by Luke and its inclusion by Matthew is most simply explained by Luke’s greater interest in stylistic improvement. The instance seems to be barren for Mr. Smith’s purpose.
Mk xiv, 1 (Mt xxvi, 2; Lk xxii, 1): Matthew’s account is here very different from Mark’s. He introduces it with the words, “And it came to pass when he had ended these sayings.” This is a formula which Matthew uses five times,[74]and which is found in Matthew alone. Since the construction ἐγένετο followed by a finite verb is found in these five passages alone in Matthew, the formula appears to have stood (once, at least, if not in all five instances) in Q.[75]It also seems to be used by Matthew to mark his transition from one of his sources to the other.[76]The remark which Mark here makes about the approach of the passover, Matthew puts into the mouth of Jesus as a part of the speech which Mark does not have. Luke follows Mark in making the statement a part of his narrative and in omitting the speech which Matthew gives.These facts would seem to indicate that Matthew is here following Q, while Luke follows Mark. Luke’s looser statement (omitting the μετὰ δύο ἡμέρας, and substituting his own favorite ἤγγιζεν,[77]and adding his ἡ λεγομένη πάσχα) would seem to go back to his desire not to trouble his Greek reader with too exact details, and yet to supply him with a little information about the Jewish feast. Here again, as in the last instance, it seems especially strange to suggest Matthew as a source of Luke where he shows such an absence of any influence from him.
Mk xiv, 12 (Mt xxvi, 17; Lk xxii, 7): Here, says Mr. Smith, Matthew gives the first and second parts of Mark’s phrase, Luke the second and third parts. The fact seems to be that Matthew here, with his usual habit of condensing Mark’s narrative, omits (what his Jewish readers would know without his stating it) the statement that on the first day of the feast of unleavened bread they “killed the passover.” Luke changes this from a particular to a general statement, so (as above) conveying to his Greek reader some information about the custom of the occasion (ἔδει θύεσθαι τὸ πάσχα). Luke here shows the influence of Mark and not of Matthew; since he follows Mark (Mk xiv, 13; Lk xxii, 10) in including eleven words which he copies very closely and which Matthew omits. He also agrees with Mark in the ascription of supernatural knowledge to Jesus upon this occasion, whereas Matthew’s narrative does not carry this implication.
Mk xiv, 65 (Mt xxvi, 67, 68; Lk xxii, 63, 64): Mr. Smith finds the influence of Matthew upon Luke in this passage, in the fact that while Mark says that they “spatupon Jesus, blindfolded him and smote him, Matthew records the first and third of these actions, Luke the second and third.”[78]Why Luke omits the spitting, may not be easy (or necessary) to say. But that Luke here shows the reverse of any influence from Matthew is indicated in the fact that whereas Matthew follows Mark in relating, first the examination of Jesus, then the mockery, and third the denials of Peter, Luke rearranges the Marcan narrative to make it run, first the denials, second the mockery, and third the examination. He has received his suggestion for this rearrangement from the fact that Mark, just before he begins the story of the mockery, has mentioned that Peter was outside the hall, warming himself by the fire.[79]It has seemed (quite naturally) to Luke that this is the place where the story of the denials should be related, tho Mark inserts the story of the mockery before he goes on[80]with the denials. In a passage where Luke has so thoroughly rearranged Mark it seems unnecessary to account for his omission of one word, especially by such a remote theory as that of Mr. Smith; and in a passage, too, where his rearrangement of Marcan material contradicts Matthew’s slavish following of it.
Mk xv, 42 (Mt xxvii, 57; Lk xxiii, 54): “Where Mark says, ‘When the even was come because it was the preparation, that is the day before the Sabbath,’ Matthew says, ‘When even was come,’ and Luke ‘the rest.’”[81]But Luke does not quite say “the rest.” He says,[82]“It was the day of preparation, and the Sabbath was dawning.” And this he says, not in the same connection, nor with the same purpose, as Mark (and Matthew). Mark and Matthew use their statement about the evening having come as an introduction to their story about the requestof Joseph of Arimathea. Luke tells his story of Joseph without any such introduction, and mentions the time only after he has finished that story, apparently with reference to the story of the women which follows, rather than to that of Joseph which precedes. The argument of the last paragraph will apply here.
It will not be necessary to go with equal care thru the five other instances in which Mr. Smith detects in a similar way the influence of Matthew upon Luke.[83]He admits “two, or three, or at the most four, cases of Marcan expressions” of which (without explanation) it might appear that Luke uses the first part and Matthew the last. His willingness to push his theory to the extreme may be inferred from his general estimate of the character of Luke as a writer: “He blurs, obliterates, blunders, fabricates, falsifies, flattens out, mutilates, murders.”[84]
The secondary interest of the writer would also seem to have influenced his work somewhat too strongly. That interest is indicated in the following statements: “If Acts was written inA.D.62, ... and Luke was written before Acts, then Matthew, slipping in between Mark and Luke must throw Mark still further back.... We thus would come very close to the resurrection, perhaps to within fifteen years, and the possibility of legendary and controversial elements having entered into the gospel story would accordingly be reduced to a minimum.... With our understanding of Lucan derivations from Matthew, as well as from Mark, theghost of a chance of existence belonging to postulated common sources, such as an earlier or a later Mark and a Q is frightened away, and we are left with the Gospels Mark, Matthew, Luke, written in that order,” etc.[85]
Passing from the details of Mr. Smith’s statement to the general argument upon which they rest, the present writer can see no cogency in that argument. Even if the use of Matthew by Luke were not contradicted by so many characteristics of both those Gospels, the writer cannot see how the choice by Luke of the second part of a phrase of which Matthew has taken the first part should prove the use of Matthew by Luke. Why should not Luke feel free to take precisely that part of a Marcan phrase which Matthew has taken—if he wanted it? Why should his finding it in Matthew make him feel that he was not at liberty to use it? Why, indeed, if Luke was copying Matthew, should he not havefollowedhim in his quotation of a certain part of a Marcan phrase, instead of putting himself every time to the trouble of going back to his Mark to pick out that part of the phrase which Matthew had left? It does not quite appear why the facts cited by Mr. Smith (so far as analysis of the passages from which they are cited leaves any of them standing) might not just as well be turned against his theory as for it.