The following group of 14 doublets and formulas in Matthew and 8 in Luke includes the following doublets noted by Hawkins: Matthew 2, 3, 12, 18, and Luke 1, 8, 10. If the double tradition doublets in Matthew 10 (above) are included then this is approximately one doublet or formula every 68 words in these portions of the double tradition, three times the frequency with which they occur in the Main Sequence of the double tradition. Because of these differences it is reasonable to suggest that the ways in which these doublets were created were different to those in the Main Sequence. Nevertheless, each doublet/formula has to be considered on an individual basis before any overall conclusions can be made.
The order of these doublets below follows the order of the double tradition verses in Luke, and because these doublets are not part of the main sequence the corresponding parallel double tradition verses in Matthew are not in order. Six of the doublets have halves or parallels in Luke 11 and three have halves or parallels in Luke 16.
This doublet has no parallels in Mark, and is related to Hawkins’ Doublet in Matthew No. 20 (Above – Mk -, Mt 3:10 // 7:19, Lk 3:9) as it is part of the pericope of the good and bad trees and their fruit. With no Markan parallels the key issue here is whether Matthew and Luke both derive from a different source or whether either Matthew or Luke depends on the other. Hawkins comments:
Here Luke has been printed opposite Mt A and B, as it partly agrees with each of them: its agreement with A is chiefly as to the similitude of the grapes and figs, and with B chiefly as to the mouth speaking out of the abundance of the heart.
Hawkins identifies Mt 7:16-18 // 12:33-35 (Mt A // Mt B) as the doublet, with these verses being parallel to Lk 6:43-45, but he ‘splits’ the verses in Luke into two groups: “the grapes and figs,” (Lk 6:43-44) and “the mouth speaking out of the abundance of the heart,” (Lk 6:45). Although both groups are discussed here, the second (in Mt 12:35 / Lk 6:45a) is also discussed separately below under the heading ‘Treasure of the heart (Hawkins: Doublet in Matthew No. 3b, also Luke).’
Under the heading ‘The Tree and its Fruits’ Patton writes as follows regarding Mt 7:16-18 / Lk 6:43-44, suggesting that aMatthew may have had his own version (i.e. not from Mark) of a saying also in Q, or (less likely) two versions in “Matthew’s version of Q:”:
For this saying Matthew has a doublet in xii, 33-35. Mt vii, 20, is also an exact reproduction of vii, 16, with the particle ἄραγε prefixed. If Matthew found this saying in two of his sources, it is impossible to say what the second of these was, for it apparently was not Mark. In Matthew’s second report of the same saying he has used the words “generation of vipers,” which he has in iii, 7, ascribed to John the Baptist. The fact that both speeches in which the phrase occurs have to do with trees, and the fact of the repetition, not only of the saying twice in Matthew, but of the same sentence twice in one report, may perhaps indicate that Matthew found the saying only in his version of Q, and is himself responsible for the repetition. Or the saying may have been recorded twice in Matthew’s version of Q, with the variations shown in Matthew’s two citations of it. Upon either hypothesis the form of Mt xii, 35, is much nearer to Lk vi, 45, than is Mt vii, 19-20, or vss. 16-18. The writer assigns the section to QMt and QLk.
There appear to be three basic possibilities for how these passages were written:
aMatthew saw Lk 6:43-45 and ‘split’ it in his own version, resulting in doublets at Mt 7:17 // 12:33a and Mt 7:16a // 12:33b, with the latter having a parallel at Lk 6:44a;
Mt 7:16-18 and 12:33-35 have been ‘merged’ into Lk 6:43-45, with aLuke excluding both halves of the doublet Mt 7:17 // 12:33a while nevertheless including a version of Mt 7:16a // 12:33b, excluding Mt 12:34a, and reversing the order of Mt 12:34b and 12:35;
There was a common source for at least the material with parallels in Matthew and Luke.
The first possibility would require Luke to have preceded Matthew, the second would require aLuke to have used Matthew (e.g. as on the MwQH), while the third would suggest a source similar to Lk 6:43-45 seen by aMatthew, e.g. Q 6:43-45 or ELk 6:43-45. Hawkins seems to favor the third alternative, while suggesting a different ‘second source:’
Is it not possible that Lk may here give the passage of the Logia from which Matthew drew on both occasions, choosing and adapting its words in A so as to bring out the criterion of true and false teachers, and in B so as to bring out the importance of words as proofs of the state of all men’s hearts? The contexts seem to suggest this. There are no peculiarly Matthean points here.
Hawkins’ point is that there is nothing particular suggesting a Matthean origin for Lk 6:43-45, pointing instead to Luke having the more original wording (perhaps originating in Q or another source, e.g. an Early Luke), to which Matthew has added Mt 7:14 and 12:33-34a.
Hawkins connects this doublet with the doublet about Good and Corrupt Trees (See above, Hawkins: Doublet in Matthew No. 3a), because the Lukan parallel to that doublet is at Lk 6:43-44. However, this appears to be not only because the verses in Luke happen to be adjacent to each other, but also because Hawkins seemingly ignores the references to treasure and heart in Mt 6:21 / Lk 12:34. Although these are not examples in which the two halves of a doublet contain the same phrase, Mt 6:21 / Lk 12:34 and Mt 12:35 / Lk 6:45 are the only four verses in the Bible in which treasure and heart are mentioned together. Because these cannot be considered to be ‘favorite phrases’ it is appropriate to treat them as doublets.
As there is no text in Mark that could be a source for any of these verses then either Matthew is the source for Luke, Luke is the source for Matthew, or both used an unknown source such as Q or Early Luke. One issue to consider is the swapping of Mt 12:34b, 35 versus Lk 6:45b, 45a. As the Matthean parallels to the double tradition verses in Luke 6 (and so also the equivalent portions of Q) are in five different chapters of Matthew (5, 7, 10, 12, and 15) such a movement of text is not unusual for this part of the double tradition, but on the Mark-Q hypothesis (in which the order of Q-text in Luke is almost exactly that of Q) it does suggest a surprisingly fine degree of very localized shuffling of Q text by aMatthew.
If instead Early Luke was a source for these verses then it is possible that it differed in order from both Matthew and Luke, and that what we see is possibly the result of movement of text by both aMatthew and aLuke instead. As neither half of both doublets fits into the ‘main sequence’ of double tradition verses then this latter suggestion is possible. However, as Lk 6:41-44 and 46-49 are in the main sequence then it seems more likely that Mt 12:34b-35 were added ‘out of place’ in Matthew instead of where we see Mt 7:17-20 (Trees and fruit).
Hawkins has nothing to say regarding these doublets in both Matthew and Luke. It appears that Mt 29:39b / Lk 22:42 are derived from Mark, and while the while the overall meaning is clearly strongly related, there is little to directly connect the text of these verses to Mt 6:9b-10 / Lk 11:2 except “Father” and “what/as thou wilt,” and “thy will.” Some bibles have a short version of Lk 11:2, with the NET (for example) telling the disciples that: "When you pray, say: Father, may your name be honored; may your kingdom come,” but adds the following notes:
Most mss, including later majority (A C D W Θ Ψ 070 ƒ 33 M it), add ἡμῶν ὁ ἐν τοῖς οὐρανοῖς (hēmōn ho en tois ouranois, “our [Father] in heaven”) here. This makes the prayer begin like the version in Matt 6:9. The shorter version is read by P א B (L: + ἡμῶν) 1 700 as well as some versions and fathers. Given this more weighty external evidence, combined with the scribal tendency to harmonize Gospel parallels, the shorter reading is preferred.
Most mss (א A C D W Θ Ψ 070 ƒ 33 M it) read at the end of the verse “may your will be done on earth as [it is] in heaven,” making this version parallel to Matt 6:10. The shorter reading is found, however, in weighty mss (P B L), and cannot be easily explained as arising from the longer reading.
In other words, the majority of mss favor the longer readings (following Mt 6:9b-10), while the ‘critical text’ is much shorter. Given the differences between both the text and relative location of Mt 29:39b / Lk 22:42 and Mt 6:9b-10 / Lk 11:2, it is hard to see how Lk 11:2 (in either short or long form) could be in any way derived from Lk 22:42. It also appears very unlikely that that the long form (similar to Mt 6:9b-10) would later be shortened to what (for example) the NET has, so suggesting that the short form was original. However, in either short or long form it is hard to see the source of Lk 11:2 on the basis of the MwQH, Instead, it appears much more likely that Mt 6:9b-10 / Lk 11:2 have a different source to Mk 14:36 / Mt 26:39b / Lk 22:42.
As with the preceding doublet, Hawkins does not refer to the doublet Mt 6:15-14 // 6:12. This doublet is unusual not only because the verses are so close together, but in this case it is possible that Mt 6:12 / Lk 11:4a both have 1 Kings 8:50 as a source, i.e. that it was known to both Mark and Matthew. Some mss omit Mk 11:26, while a few add a copy of Mt 7:7-8 after this verse, perhaps suggesting that Mt 6:15 is original but Mk 11:26 is not.
It is not clear how much any of these verses depend on 1 Kg 8:50, but it is notable that Mt 6:12 and Lk 11:4a reverse the clauses in comparison with Mk 11:25b / Mt 6:14, so that two different sources appears likely even if aLuke knew Matthew.
Although these verses reflect two different sets of circumstances, they are clearly related, but equally clearly Mt 6:13 / Lk 11:4 are in a different setting to Lk 14:38 / Mt 26:41 / Lk 22:46. The former (double tradition) verses are part of the Lord’s Prayer, in which Jesus teaches the disciples how to pray, while in the latter (triple tradition) verses Jesus exhorts the disciples to do what he had previously told them. Both Lukan verses are shorter, and while there are no extant mss in which Luke has a parallel to Mk 14:38b / Mt 26:41b, the critical text excludes Lk 11:4c. Here Willker (Luke, TVU 193) writes:
P45: has a lacuna of about 7 lines here, but from space calculations it appears almost impossible that P45 contained all long variants of the Lord's prayer. Probably it read txt everywhere. [and] Again a clear harmonization to Mt and/or liturgical usage. There is no reason for an omission
The variant in Lk 11:4 and the lack of an equivalent phrase following Lk 22:46 suggests that Luke originally did not include Lk 11:4c.
Mk 11:24 and Mt 21:22 follow directly on from Mk 11:23 and Mt 21:21 in Hawkins’ Doublet in Matthew No. 11 (below), and so are in the same non-Lukan context (cursing the fig tree). Then, although Mt 7:7-8 and Lk 11:9b-10 do not follow Mt 17:20 and Lk 17:6, the same pattern exists here as in Matthew No. 11, with Mk 11:24 the source of Mt 21:22, and with the double tradition verse in Matthew preceding the verse having Mark as its source. As in other doublets, either aMatthew backtracked after having written Mt 21:22, or Mt 7:7-8 and 21:22 had different sources.
As with Hawkins’ Formula 1 in Matthew (above) the origin of the double tradition verses Mt 7:11 / Lk 11:13 is either aMatthew, Q, or Early Luke, with aMatthew later choosing to re-use the phrase at Mt 12:34 as a unique addition to the passage on trees and fruit.
In ‘Mark and Q,’ Fleddermann examines in detail (on the assumption of the Mark-Q hypothesis) ‘The Beelzebul Controversy,’ which he specifies as ‘Mark 3,22-27 (Matt 12,22-30.43-45 par. Luke 11,14-15.17-16),’ so indicating that Mk 3:22-27 has double tradition parallels (with Q as their source) in Mt 12:22-30.43-45 / Lk 11:14-15.17-16. It is not clear from this exactly which Matthean and Lukan verses parallel those in Mark, but this becomes clearer where Fleddermann breaks down his examination into five groups of parallel verses. His focus is on re-constructing the text of Q from Matthew and Luke, and as he mentions that “Matthew records a doublet of this section [Q 11,14-15] in Matt 9,32-34” it appears that we can align his parallels as shown below:
Mark Matthew Luke Content
Mk 3:21-22 Mt 9:32-33[,34] // Mt 12:22-24 Lk 11:14-15a Beelzebul
Mk 3:23 [Lk 11:15b] Satan
Mk 3:24-26 Mt 12:25b-26 Lk 11:17b-18a Kingdom and Satan divided
Mt 12:27-28 Lk 11:19-20 Beelzebub, Spirit/Finger of God
Mk 3:27 Mt 12:29-30 Lk 11:21-22 The strong man
Mt 12:43-45 Lk 11:24-26 The unclean Spirit
Mt 12:27-28, 43-45 / Lk 11:19-20, 24-26 have no Markan parallels and so are part of the double tradition. However, the presence of the doublet Mt 9:32-34 // Mt 12:22-24 suggests that Mk 3:21-23 / Mt 9:32-34 and Mt 12:22-24 / Lk 11:14-15 together form a Mark-second source overlap, with Mt 12:25b-30, 43-45 and Lk 11:17b-18a, 19-26 then having the same second source (Q or otherwise). The important issue is whether Lk 11:14-15 is a closer parallel to Mt 9:32-34 or to Mt 12:22-24, and selecting the correct alignment is crucial to an understanding of how these verses are related.
Fleddermann notes that: “Matt 9,32-34 … corresponds more closely to Luke 11,14-15 than Matt 12,22-24 does,” but that “Both Matt 9,32 and Matt 12,22 show signs of intense redactional activity.” He also writes: “When we turn to Luke 11,14 we also find signs of redactional activity, but Matt 9,33 agrees closely with Luke 11,14…” Hawkins suggests that it is not clear which half of the doublet is a closer parallel to Lk 11:14-15, writing:
Lk has not been printed exactly opposite Mt A [Mt 9:32-34] or Mt B [Mt 12:22-24]; for while on the one hand it corresponds to B in being given as the occasion for the ‘defensive discourse’ in Mt xii and Lk xi, on the other hand it more closely resembles A in wording.
Textually Lk 11:14-15 is closer to Mt 9:32-34, but its location in the Lukan narrative suggests that it is a parallel to Mt 12:22-24 instead, and on this Patton writes in his comments on Mt 9:32-34:
Vs. 34 is a doublet of Mt xii, 24; the latter is from Mk iii, 22, where Mark also appears to be following Q. Perhaps ix, 27-34, has been inserted at just this place, in order to warrant the statement of Jesus to John the Baptist that “the blind see and the deaf hear.” It is hardly necessary to assign it to a special literary source.
Hawkins is unsure as to whether Lk 11:14-15 ‘corresponds’ to Mt 9:32-34 or to Mt 12:22-24, while Patton suggests that Mk 3:22 is the source of Mt 12:24 even though Mark “appears to be following Q,” and does not see any need to assign Mt 9:32-34 “to a special literary source” (presumably not Q). As verses Mt 9:33,34 both have a large number of words in common with Lk 11:14,15 respectively it appears that the following alignment of verses is more likely to represent the situation correctly:
Mark Matthew Luke Content
Mk 3:21-23 Mt 12:22-24 // Mt 9:32-34 Lk 11:14-15a Beelzebul / Satan
Mk 3:24-26 Mt 12:25b-26 Lk 11:17b-18a Kingdom and Satan divided
Mt 12:27-28 Lk 11:19-20 Beelzebub, Spirit/Finger of God
Mk 3:27 Mt 12:29-30 Lk 11:21-22 The strong man
Mt 12:43-45 Lk 11:24-26 The unclean Spirit
This raises the question of why Mk 3:21-23 is so different to Mt 9:32-34, and why they are located in very different contexts. Do they in fact have “a special literary source,” or is there some other reason why they appear to not ‘fit’ here? Additionally, on this revised alignment, why do ‘Mt 12:25b-45 appear to alternate between Triple and Double Tradition verses? Fleddermann comments that:
from Mk 3,23 Luke basically re-writes Q. He does anticipate Mark’s concluding verse (Mark 3,30) in an addition to the end of v. 18, but within vv. 17-18a he edits Q… [and that in] the divided kingdom... Matthew conflates Mark and Q, and that “he will conflate Mark and Q’s sayings on the Unforgivable Sin (Matt 12,31-32).
Given these comments, it is reasonable to suggest that aMatthew is doing the same thing with the Strong Man, especially as the text of Mt 12:29 is so much closer to Mk 3:27 than to Lk 11:21-22.
The discussion immediately below summarizes just Beelzebul / Satan, with a more detailed discussion in Mk-3-19b-22 - Mad Jesus and Beelzebub, that includes a full analysis of the verses shown below in blue. “Mk 3:28-29, Mt 12:31 // 12:32, Lk 12:10 – The Unforgivable Sin” is discussed later. The other verses just discussed above do not have a full analysis here because none of them involves doublets.
Note: Lk 8:35a-36 are shown above to emphasize how ‘generic’ Mt 12:22b-24 are.
Although some see aLuke behaving uncharacteristically here regarding his use of material from Mark, it comes down to perhaps just Mk 3:21, 24-26, and 30. Of course, on the assumption that aLuke knew Matthew then (presumably) he saw both Mt 9:32-34 and 12:22-24 and in this situation could easily have chosen Matthew over Mark. It is also worth noting that in Lk 11:15 aLuke simplifies both Mk 3:22 and Mt 12:24, in Lk 11:16 adds text from Mk 8:11 / Mt 16:1, and in Lk 11:21-22 follows neither Mk 3:27 nor Mt 12:29, so it is much fairer to say that aLuke here sometimes follows Mark, sometimes Matthew, and sometimes neither. The biggest issue seems to be aLuke not following Mk 3:20-22, but instead using Mt 9:32-34 // 12:22-24 as the basis of Lk 11:14-15. However, it is not as simple as that, as witnessed by the number of variants in all three gospels: in Mk 3:20, 21; Mt 9:32, 34, 12:22; and Lk 11:14, 15. For whatever reason there seems to have been significant uncertainty in all four versions of this text, with perhaps the most interesting being the parallel to Mk 3:22 at Mt 9:34.
What reason could aMatthew have had to place Mt 9:34, a parallel to Mk 3:22, immediately after a collection of five healing stories: The man with the palsy – Mt 9:2-8; the woman with the issue of blood – Mt 9:20-22; the daughter of the ruler - Mt 9:18-19, 23-26; the two blind men - Mt 9:27-31; and last, the dumb man - Mt 9:32-33? Even though none of the four earlier healings were viewed as problematic, somehow curing someone of being dumb suggested the casting out of a devil, while the previous raising the dead and curing sickness and blindness did not. There is also the odd fact that none of Mt 9:34 is present in Western mss, and Willker comments that the transition from Mt 9:34 to 35 is abrupt. Quite so: Jesus is accused of being in league with the “prince of devils,” and then … nothing! Instead, “Jesus went about all the cities and villages” as if nothing untoward had taken place. There appears to be no reason why Mt 9:34 is here. So why is it?
Although Mt 9:34 is generally considered to be original (despite being not present in some mss), it would make much more sense for it to have been added after the close copy of Mt 9:32-33 was written at Mt 12:22-23, and Mt 12:24 was added to it. Either aMatthew then added Mt 9:34 himself (which would seem unlikely), or it was added later by someone else to match Mt 12:24. This of course raises the question of why Mt 12:22-24 even exists, the simple answer to which (as argued in Mk-3-19b-22 - Mad Jesus and Beelzebub) is that what aMatthew saw where we see Mk 3:20-21 made little or no sense. aMatthew then copied a previous healing story (Mt 9:32-33) to use in its place at Mt 12:22-23 and followed it by a version of Mk 3:22. He did not include a parallel to Mk 3:23 here, but instead incorporated some of this verse into Mt 12:26.
Later, aLuke created a copy of Mt 9:32-33 // 12:22-23 at Lk 11:14, and, like aMatthew, added a shorter version of Mk 3:22 at Lk 11:15 (a longer version of which appears in Western mss). All of the above is a direct result of problems with the text of what we see as Mk 3:20-21, which have left their mark in the form of very unusual variant readings in Mk 3:21, and (possibly) the lack of the beginning, and subsequent re-location, of what we see as Mk 3:30 but may have originally been text between Mk 3:20 and 22.
This is an example of possible damage to Mark in which what we see as Mk 3:21 (and possibly 3:30) was originally part of a healing story. aMatthew saw something here (possibly similar to what we see) but did not want to use it, so instead replaced most of it by a very slightly different version of what he had already written at Mt 9:32-33, and also replaced Mk 3:23 by Mt 12:25a.
If the damage to Mark in this area also led to the text surrounding Mk 3:21,30 being lost or becoming unreadable then it could have led to what we now see as Mk 3:21 (with its unusual variants) being thought to refer to Jesus. Then, on the assumption that both the end of Mk 3:21 and the beginning of what is now Mk 3:30 became lost or unreadable, moving what was left to after Mk 3:29 would make sense.
Depending on what damage aLuke saw in the copy of Mark he saw this could also explain both his lack of any parallel to Mk 3:19b/20a-22a, and the re-location of his parallel to Mk 3:22b / Mt 12:24b to Lk 11:15b instead of after Lk 6:19.
Hawkins notes: “For the words of Jesus which follow in each case, see above, No. 9.” Here this is The sign (of Jonah) (Hawkins: Doublet in Matthew No. 9), which overlaps with this doublet, Matthew No. 18. Hawkins continues:
Judging from the position and contexts of the two pairs of records, it would be natural to class Mt A [Mt 12:38-39] and Lk as Logian, Mt B [Mt 16:1-2] and Mk as Markan in origin. Probably we are right in doing so, notwithstanding the fact that as to two expressions, πειράζοντες [test/tempt] and ἐξ οὐρανοῦ [from heaven], Lk agrees with Mt B and Mk in having them, while Mt A is without them. But both expressions are so common, and so much in place here that, as in the case of No. 9, no importance can be attached to the insertion of them.
As Mt 12:40-42 and Lk 11:30-32 (the explanation of the sign of Jonas) are obvious parallel passages, it is clear that the preceding verses (Mt 12:39 and Lk 11:29) are also part of the same parallels. In addition, as Mt 12:38, 40-42 and Lk 11:29a, 30-32 have no parallel in Mark it is most likely that Mt 12:39 and Lk 11:29b also have no parallel in Mark, i.e. that Mk 8:12b is not part of a Markan version of this passage, and is therefore not in the same context as Mt 12:39b / Lk 11:29c. This makes it highly likely that Mt 16:1-4 depends on, and is an expansion of, Mk 8:11-12., but note that Mt 16:2b-3a (not shown above) is quite possibly not original:
When it is evening, ye say, It will be fair weather: for the sky is red. And in the morning, It will be foul weather to day: for the sky is red and lowering. [Mt 16:2b-3a]
Willker (Matthew, TVU 213) writes that this text is not in:
01, B, X, Y, G, f13, 2*, 157, 1424mg, pc, Sy-S, Sy-C, sa, mae-1+2, bopt, Or, Hiermss , acc. to Gregory 047 also omits
In addition he notes that: “579 did omit the passage here, but inserts it after verse 9!” There is no obvious reason for either an addition or an omission, but the text reads well without it. It is possible that this text was inspired by Lk 12:54b-55, although the words are very different.
Mk 4:21 has a parallel at Lk 8:16, and even though their wording is not the same these verses convey the same message, and both are part of common contexts in Mk 4:14-25 and Lk 8:11-18. Although Mt 13:18-23 is also in this context, after these verses the narrative in Matthew diverges and there is no corresponding parallel to Mk 4:21 / Lk 8:16 at this point. Instead, the only Matthean parallel is at Mt 5:15, with parallel at Lk 11:33, where there is no Markan context. Either:
aMatthew and aLuke used Mk 4:21 as the source of Mt 5:15 (instead of placing this after Mt 13:23) and Lk 8:16 respectively, and aLuke decided to add another version at Lk 11:33; or
aLuke used Mk 4:21 as the source of Lk 8:16, and there was a different version in a second source that aMatthew and aLuke used at Mt 5:15 and Lk 11:33 respectively.
Patton comments on ‘A Saying About Light (Mt v, 15; Lk xi, 33)’ as follows (paragraph breaks added):
This is another saying that stood in both Mark and Q. Mark has the saying in Mk iv, 21. His form of it is the apparently less natural one, “Does the lamp come in order that it may be put under a bushel?” etc. Weiss suggests that it has been given this form to make it refer to the coming of Jesus as the light of the world. Neither Matthew nor Luke has copied this feature of Mark’s saying.
By his context Matthew makes the saying refer, like the saying about salt, directly to the disciples. Luke has the saying twice: in xi, 33 and viii, 16. In both cases his context would indicate that he took the saying to refer to the teaching of Jesus. Matthew says the light is to give light “to all that are in the house.” Luke does not mention the house, but implies it in his statement that “those who are entering in see the light,” this form being found in both his reports of the saying. Mark says “under the bushel or under the bed”; Matthew, “under the bushel”; Luke once, “in a dish or under the bed,” and a second time, “in a cellar or under the bushel.”
Luke’s fondness for the same ending in his two uses of the saying can be explained only by the supposition that it so stood in one of his sources. The same idea in the conclusion of the saying as it appears in Matthew and Luke, and their common avoidance of the opening formula which is peculiar to Mark, would indicate that Matthew and Luke practically forsake Mark in this saying, and follow their other source. Luke, having a doublet for the saying, may be assumed to have taken it once from Mark and once from his other source; but he is evidently much more influenced by his other source than he is by Mark. The non-Marcan source in which the saying was found by Matthew and Luke was evidently an allied, but not an identical, one; the saying is therefore assigned to QMt and QLk.
What Patton misses (or perhaps did not know) is that there are two variants in the Lukan doublet (Lk 8:16 // 11:33) that bear upon the issue of the source of this saying, and so also on the synoptic hypothesis. As Pattton notes Lk 11:33 ends with words that are not in Mk 4:21: “that they which enter in may see the light,” and although these words are also present in Lk 8:16 in most mss, they are not present in P75 and B, and are in a different order in f1, 579 and 1241, as shown below (highlighting added):
And no one, after lighting a lamp, covers it with a jar or puts it under a bed, but puts it on a lampstand,
Οὐδεὶς δὲ λύχνον ἅψας καλύπτει αὐτὸν σκεύει ἢ ὑποκάτω κλίνης τίθησιν ἀλλ’ ἐπὶ λυχνίας τίθησιν
so that those who come in can see the light.
ἵνα οἱ εἰσπορευόμενοι βλέπωσιν τὸ φῶς [Not in P75 and B, and ending τὸ φῶς βλέπωσιν in f1, 579 and 1241]
Willker (Luke, TVU 121) writes:
Mt has different words, but the same meaning. Mk does not have the words. Is it possible that P75, B had the words in the f1, 579 order and then omitted them due to h.t. (σιν- σιν)? Otherwise the omission is difficult to explain. Weiss thinks (Textkritik, p. 190) that the omission is a harmonization to the parallel in Mk (so also Hoskier and Tischendorf). It is also possible that the words were adapted from Lk 11:33.
Willker notes that the omission at the end of Lk 8:16 in P75 and B is “difficult to explain,” and hypothesizes that perhaps these mss (surely here he means an ancestor of these mss?) had the words in the order seen in f1, 579, and 1241 (ending with βλέπωσιν), and that the final clause was omitted during copying due to the scribes eye jumping from τίθησιν directly to βλέπωσιν.
However, there is no evidence that any ancestor of P75 and B had these words in a different order, and instead it is more plausible that Mk 4:21 was the source of Lk 8:16 (which originally did not have the final clause), but that the source of Lk 11:33 (either Mt 5:15 or a second source such as Q) included a version of these words. Then, because they were included in Lk 11:33 they were added to Lk 8:16 in the later majority of mss (but note that P75 is the only early ms which is extant for Lk 8:16).
It is also worth noting that Tertullian comments on the version of Lk 11:33 in Ev (''Marcion's" Evangelion), but does so by stating that “a candle must not be hidden, but affirms that it ought to be set upon a candlestick, that it may give light to all,” so using a combination of the text of Mt 5:15 and Lk 11:33. This is one example that has led to questions regarding whether Tertullian was ‘quoting’ from memory, but if Ev preceded Luke then it is possible that “that are in the house” at the end of Mt 5:15 is an addition, and “under a bushel” led to the words from Ev and Mt 5:15 being conflated in Lk 11:33.
The doublet here is in the text regarding the best seats in the synagogues and the greetings in the markets, which occurs once in Mark, once in Matthew, and two times in Luke. Mk 12:37b-40, Mt 23:1-14 and Lk 20:45-47 all exist in the same context, and as Lk 20:46 is a very close parallel to Mk 12:38b-39 it is reasonable to assume that, as Hawkins states: “Lk B [20:46] is Marcan in character and presumably in origin.”
This makes the location of the other half of the doublet at Lk 11:43 hard to explain on the basis of a Markan source, as not only does it omit the references to long clothing/robes and rooms at feasts seen in both Mk 12:38 and Lk 20:46, but it occurs nine chapters prior to Lk 20:46. On the assumption that aMatthew saw Mk 12:28-39 either his version of the same passage has been ‘expanded’ in reaction to seeing both halves of the doublet in Luke, or aLuke took elements from Mt 23:2, 5-7a and spread then across his two versions.
On the MwQH aMatthew saw Mk 12:38-39, to which he added the reference to the Pharisees and detail regarding the long clothes. aLuke then saw Mt 23:2, 5-7 and created two versions of his own, the first featuring the Pharisees, omitting any mention of the long clothes/robes, and referring to ‘uppermost seats’ (conflating the uppermost rooms and chief seats), and the second featuring the scribes and being very similar to Mk 12:38 except for having ‘highest seats’ and ‘chief rooms.’ Although there is no problem with aMatthew expanding what he saw in Mk 12:38-39, both halves of the doublet in Luke contain very unusual differences from the versions in Mark and Matthew, with (for example) aLuke using the second half of Mt 23:5-6 at Lk 11:43 and the first half at Lk 20:46, omitting parallels to Mt 23:3, and moving his parallel to Mt 23:4 to Lk 11:46.
If instead there was a separate source for Lk 11:43 then it would appear that aMatthew created his parallel at Mt 23:2, 5-7 by combining elements of both Mk 12:38-39 and the other source, perhaps as Hawkins suggests:
It appears then that Mt’s use of the Marcan source here is affected and modified by the influence of that record (probably directly or indirectly Logian) in Lk xi, to which there are so many parallels in Mt xxiii.
Although Hawkins suggests that the source may be connected to the Logia, other sources are possible. On the Mark-Q hypothesis the source would be Q, with in this case the positioning of the two halves of the doublet suggesting that aLuke used material from Q in preference to Mark. On the MwEL hypothesis using Early Luke (both in order and content) as the basis for Luke would be a natural choice, with material from Mark being added later.
This doublet is not noted by Hawkins, perhaps because it could be viewed that this is simply a Triple Tradition passage. However, Mk 3:28-29 and Luke 12:10 are significantly different, with Mt 12:31 instead being close to Mk 3:28-29a. There appears to be no reason why aLuke would exclude a parallel to Mt 12:31 if he saw it, nor why aMatthew would add Mt 12:32 if he did not see a parallel to Lk 12:10.
The most reasonable explanation is therefore that aMatthew saw both Mk 3:28-29 and a parallel to Lk 12:10 in a different source and decided to merge them rather than choosing just one or the other.
Mt 6.33 / Lk 12:31 appear to be versions of Mt 7.2b / Lk 6:38b respectively, not including the use of the person’s own measurement system to determine how he or she would be ‘measured,’ i.e. judged, but instead suggesting he or she need to have a relationship with God. Assuming Markan priority Mk 4:24bc is clearly the source of Mt 7:2b / Lk 6:38b. However, there are a few variant readings in Mk 4:24bc, with the majority of mss adding “unto you that hear,” and D, W, and several other mss omitting “unto you that hear shall more be given.” Willker (Mark, TVU 87) comments on possible h.t. (homoioteleuton – ‘like ending’).
It is interesting to note the many occurrences of h.t. at this variation unit. It is possible that the words τοῖς ἀκούουσιν fell out due to h.t., too: ..ιν - ..ιν. The omission by D et al. is either due to h.t. or it is a harmonization to Mt. If it is h.t., which is probable, then the witnesses are indirect support for the txt reading.
Possibly the words have been added to make a connection with the Βλέπετε τί ἀκούετε. This is supported by the addition of the words by f13 at Lk 8:18 directly after ἀκούετε.
Nevertheless the construction sounds a bit awkward and seems rather unmotivated. It is also possible that the words have been omitted as confusing. Güting (TC Mark, 2005, p. 247): "the phrase appears difficult to understand and disturbing". He thinks that the promise of such an extreme reward was considered problematic.
As Willker comments, “the many occurrences of h.t.” are “interesting,” but he nevertheless appears to discount the possibility of “harmonization to Mt.” However, the multiple different variants (including that in Lk 8:18) do suggest that something other than h.t. has occurred, as does the fact that the parallels to Mk 4:24bc are not in Matthew 13 and Luke 8 (as in the surrounding verses), but instead in Matthew 7 and Luke 6 respectively. It should also be noted that apart from (possibly) Mk 4:24bc, Lk 6:20-49 has no parallels in Mark, and all the Markan parallels to Lk 6:1-19 are in Mark 2:23 – 3:19.
The implication is that Mk 4:24bc is not original, and was inserted here (with different variant readings in many mss) to provide a Markan parallel to Mt 7:2b / Lk 6:38b. If this is the case then Mt 7:2b / Lk 6:38b do not have Mark as their source, and so Mk 4:24bc / Mt 6:33 / Lk 12:31 cannot (on the Mark-Q hypothesis) be a Mark-Q overlap. Instead, Mt 7:2b / Lk 6:38b and Mt 6:33 / Lk 12:31 are then simply two different sets of double tradition verses.
If instead Mk 4:24bc is original then there is no satisfactory explanation for the variants in both this verse and Lk 8:18, nor for the fact that Mt 7:2b / Lk 6:38b are not instead located just before Mt 13:12 and in Lk 8:18 respectively.
These verses closely follow the triple tradition parable of the fig tree at Mk 13:28-31 / Mt 24:32-35 / Lk 21:29-33, all of which end with Jesus saying: Verily I say unto you, that this generation shall not pass, till all these things be done/fulfilled, and all refer to the time of the coming of the Son of Man. Hawkins notes the difference between the time given in Mt A [24:42] and Mt B [25:13]:
Mt A and Mk are very similar, and are found in the concluding part of the same discourse, though not exactly not in the same connexion. If that discourse comes from the Marcan source, Mt A may have been modified by ποίᾳ ἡμέρᾳ as a reminiscence of the τὴν ἡμέραν in the other or Logian version of the saying in Mt B, which may have been brought from the Logia with the parallel of the Ten Virgins. But this is only a doubtful matter of detail.
It will have been seen that the use of ἡμέρᾳ is a mark, though not an important one, of Mt A and B only.
It is clear that Mk 13:34-35, Mt 24:42, Lk 21:35-36 are triple tradition continuations from the parable of the fig tree, and hence do not favor any particular synoptic hypothesis. However, Matthew has no parallels to Mk 13:33-34 / Lk 21:34-35 and instead contains double tradition additions at Mt 24:37-41 that have parallels at Lk 17:26-27, 30, 35-36. On the MwQH aLuke saw these verses in Matthew and moved them forward in his gospel (to Chapter 17), or they came from another source and aMatthew and aLuke used them at different places relative to the material for which Mark was a source.
Mark | Marcion | Matthew | Luke
10:17-30 The rich young ruler | | 19:16-29 The rich young ruler | 18:18-21 The rich young ruler
But many that are first shall | | 19:30 But many that are first |
be last; and the last first. | | shall be last; and the last shall be first.
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
| | 20:1-15 The worker in the vineyard "
| | See discussion below | 13:22-29 The strait gate
| Again he falsified … “The last shall | 20:16 So the last shall be first, | 13:30 And, behold, there are last
| be first,” … (E) | and the first last: for many be | which shall be first, and there are
| | called, but few chosen. | first which shall be last
10:32-34 Jesus foretells his death. | | 20:17-19 Jesus foretells his death. |
| … and “The Pharisees came | 23:37-39 The lament over | 13:31-35 The lament over
| saying, Get thee out and depart, | Jerusalem | Jerusalem
| for Herod will kill thee.” … and, “Ye
| shall not see me until ye shall say,
| Blessed.” (E)
Hawkins notes the following regarding Mk 10:31, Mt 19:30 (Mt A), Mt 20:16 (Mt B), and Lk 13:30:
Here again Mt A and Mk, which are parallel in position, agree exactly, and probably come from the Marcan document.
Patton essentially agrees, noting that Mk 10:31 “is paralleled in Luke (xiii, 30) and has doublets in Matthew (xix, 30; xx, 16). It apparently stood in both Mark and Q,” and adds the following regarding Mt 20:1-16:
The parable is too long for Q, tho like the Q parables it has to do with the kingdom of God. The last verse is an apparently proverbial saying, for which Matthew has a doublet in xix, 30, and Luke a variant in Lk xiii, 30.
Not only is Mt 20:1-16 “too long for Q,” but the lack of any parallel to these verses other than Mt 20:16 / Lk 13:30 suggests that aMatthew had an unknown source (oral or written) for these verses.
As both Mt 19:30 and Mk 10:31 terminate passages about the rich young ruler (Mt 19:16-30 and Mk 10:17-31 respectively), then assuming Markan priority it is almost certain that Mt 19:30 rather than Mt 20:16 depends on Mk 10:31. However, there is no corresponding verse at the end of the parallel passage in Luke (Lk 18:18-21), and instead, there is a ‘gap’ at this point, with no parallel here in Luke to either Mk 10:31-32 or Mt 19:30-20:1. Both Mk 10:31 and Mt 19:30 have the order first–last then last–first, while Mt 20:16 / Lk 13:30 have the order last–first then first–last, and as neither of the latter two verses have any parallel context in Mark it is clear that they are part of the double tradition.
On the assumption of the Mark-Q hypothesis Mt 20:16 and Lk 13:30 therefore have Q as their source, and this would also appear to be the case for almost all of Lk 13:22-30 and their corresponding parallels in Matthew (all of which have no parallel in Mark). However, the parallels in Matthew are in a very unusual order when compared with Luke, and on any hypothesis in which both aMatthew and aLuke shared a non-Markan source here at least one of them heavily rearranged this passage. Using the order in Luke we have:
Lk 13:22 / Mt 9:35
Lk 13:23 -
Lk 13:24a / Mt 7:13a
- Mt 7:13b
Lk 13:24b / Mt 7:14
Lk 13:25 –
Lk 13:26 / Mt 7:22
Lk 13:27 / Mt 7:23
Lk 13:28a / Mt 8:12b (There shall be the weeping and gnashing of teeth)
Lk 13:28b / Mt 8:11b (when ye shall see Abraham, and Isaac, and Jacob, and all the prophets, in the kingdom of God,)
Lk 13:28c / Mt 8:12a (and you yourselves thrust out)
Lk 13:29a / Mt 8:11a (And they shall come from the east, and from the west, and from the north, and from the south, and shall sit down)
Lk 13:29b / Mt 8:11c (in the kingdom of God.)
Lk 13:30 / Mt 20:16
If the source was Q (so assuming that aLuke followed Q’s order) then aMatthew split up what he saw here in Q, spread the text over four different chapters of his gospel, and significantly edited and re-ordered the Q text he used as the source of Mt 8:11-12. In contrast on the MwEL hypothesis the order of the verses in Early Luke does not have to be the same as in Luke, and it is also allowed that aLuke may have added verses to what he saw in Early Luke. This means that on the MwEL hypothesis some of the text we see in Lk 13:22-30 may not have been either in the order shown above or even present in Early Luke, and there could have been two sets of changes to the text that we see in Luke:
1. aMatthew took the text of Early Luke and changed it as he saw fit, editing and/or rearranging to suit;
2. aLuke made a different set of changes as a result of seeing the parallels in Early Luke and Matthew.
While it may appear that this suggestion should be rejected on the basis that it is unnecessarily complicated and so violates Occam’s Razor, there is evidence to suggest that this did actually take place. In A Critical and Historical Enquiry Into the Origin of the Third Gospel, P.C. Sense writes regarding comments by both Tertullian and Epiphanius on the verses in Ev corresponding to Lk 13:25-35. He writes (paragraph breaks added):
The next verses 25-28, to the first clause, are clearly alluded to by Tertullian. He quotes intermittently verse 25: “When the master of the house has risen up … and hath shut the door, the wicked of course to be excluded, he will reply to those who knock: I know not whence ye are. And again to those who recount that they have eaten and drank with him, and he has taught in their streets, he adds, Depart from me, all ye workers of iniquity; there will be weeping and gnashing of teeth.” These allusions guarantee the text so far.
Tertullian proceeds to say: “Where? outside of course, where they have been excluded, the door having been shut by him: therefore there shall be punishment, by him by whom the exclusion is made for a punishment, when they shall see the just entering into the kingdom of God, but they themselves are detained outside.” These remarks guarantee verse 28 in a general way, but not its exact words.
Here Epiphanius gives us assistance (Sch. and Ref., x1.). He tells us Marcion cut out “Then ye shall see Abraham, and Isaac, and Jacob, and all the prophets in the kingdom of God,” and instead of these words he had, “When ye shall see all the just in the kingdom of God, but you (plural) thrust out.” And he further says, Marcion added: “Vanquished outside, there shall be weeping and gnashing of teeth.”
I propose from these data to re-construct the text of verse 28, after the first clause, thus: “When ye shall see all the just in the kingdom of God, but yourselves thrust out, vanquished outside, there shall be weeping and gnashing of teeth.” … Tertullian says not a word about the following verses, 29-35, but Epiphanius assures us (Sch. And Ref., xli.) that Marcion cut them all out, and as he quotes the initial words of the verses excised, there can be no mistake.
From the above it is clear that Ev contained parallels to Lk 13:25-28, but nothing of Lk 13:29-35. In addition, the parallel to Lk 13:28 read approximately “When ye shall see all the just in the kingdom of God, but you yourselves thrust out, vanquished outside, there shall be weeping and gnashing of teeth” i.e. having a different version of Lk 13:28b, ending with what we know as Lk 13:28a, and including nothing from Lk 13:29-35. So, here we have evidence from two sources that Ev had a shorter version of Luke 13:25-35, containing just text equivalent to these verses:
Lk 13:25 – No parallel
Lk 13:26 – Mt 7:22
Lk 13:27 – Mt 7:23
Lk 13:28b – Mt 8:11b (When ye shall see all the just in the kingdom of God,)
Lk 13:28c – Mt 8:12a (but you yourselves thrust out, vanquished outside,)
Lk 13:28a – Mt 8:12b (there shall be weeping and gnashing of teeth)
As can be seen, the order of the text in Ev 13:28 matches what we see today in Mt 8:11b-12, with nothing corresponding to Lk 13:30 / Mt 20:16 (last-first, first-last). Some people have suggested that Tertullian was not looking at Ev while commenting on it, but instead was looking at (or thinking about) Matthew, but Epiphanius confirms that in Marcion’s gospel this passage followed Matthew quite closely.
This of course raises the question of why Marcion would edit Lk 13:28-29 so that in 'his' Ev it looked more like Mt 8:11b-12. While of course Marcion could have done this, it seems highly unlikely that he would supposedly focus entirely on Luke but then for no apparent reason modify these verses to make them look more like Matthew. Instead, it is much more likely that here Ev reflects a text that precedes both Matthew and what we know as Luke, e.g. as Early Luke does in the MwEL hypothesis.
As there is no Markan parallel to these verses then on the MwQH (in which Matthew is a source for Luke) the originator of these verses can only have been aMatthew. However, there is no common context for either Mt 23:12 / Lk 14:11 or Mt 23:12 / Lk 18:14b:
Mt 23:12 immediately precedes the woes on the scribes and Pharisees, and if it had existed the parallel in Luke in the same context would have been between Lk 20:46 and 20:47;
Lk 14:11 is part of a Sondergut Luke passage at Lk 14:7-15;
Lk 18:14b ends a Sondergut Luke passage at Lk 18:1-14.
There is no indication that any of these verses were created specifically to be used in their own local context. Instead, they most likely represent three uses of a saying that existed prior to the writing of both Matthew and Luke. Whether this was simply a common oral saying, or it existed in a written form seen by aMatthew and aLuke, is not known. Neither Tertullian nor Epiphanius mention it with respect to Ev so there is no evidence of its existence at either corresponding location in Ev.
The International Q Project (IQP) considers that Q contained a parallel to Luke at Q 14:11 but not at Q 18.14. If this latter verse existed in Q then there is no specific reason to consider that Lk 14:11 is a double tradition verse but Lk 18:14b is not, except that Lk 14 contains other double tradition verses but Luke 18 does not. However, it is just as likely that this is simply a version of a saying known from Eze 21:16:
Thus saith the Lord God; Remove the diadem, and take off the crown: this shall not be the same: exalt him that is low, and abase him that is high.
In addition to the verses shown here in the synoptic gospels the phrase “the law and the prophets” can also be found in Jn 1:45, Ac 13:15, 24:14, and Rom 3:21. As a result, rather than this being just a favorite phrase of Matthew it appears that this phrase was in more widespread use. Nevertheless, aLuke only uses this phrase once, in the double tradition verse Lk 16:16. However, Lk 24:44 also has: “which were written in the law of Moses, and in the prophets” which is similar to Ac 28:23: “both out of the law of Moses, and out of the prophets.” On this basis it appears that aLuke preferred “the law of Moses” (also in Lk 2:22, 24:44) and “the law of the Lord” in Lk 2:23, 24, 39, to just “the law,” which he uses only where it is clear he is referring to the written form of the law, in Lk 2:27 and 16:17.
On the MwQH aLuke chose to only add a parallel to Mt 11:12-13, ignoring the other uses of the same phrases in Matthew. On the Mark-Q hypothesis aLuke did not know the parallels in Mt 5:17, 7:12 and 22:40, but included a parallel of Q 16:16. On the MwEL hypothesis aLuke kept what he saw in ELk 16:16, but did not add any parallels to Mt 5:17, 7:12 or 22:40.
The phrase “heaven and earth” appears over thirty times in the bible, but there are only five references to heaven and earth passing away, all of which are in the synoptic gospels, in which there are two different versions:
That heaven and earth shall pass away but God’s words shall not (Mk 13:31, Mt 24:35, Lk 21:33);
That God’s law will stand until heaven and earth pass away (Mt 5:18, Lk 16:17).
Mk 13:31 is the obvious choice for the source of the phrase used in Mt 24:35 and Lk 21:33, not only because of the similarities of the text, but also because Mk 13:28-31, Mt 24:32-35, Lk 21:29-33 are all very close triple tradition passages. However, Mk 13:31 is unlikely to be the source of Mt 5:18 and Lk 16:17 even though both precede the versions having a clear Markan source, as it is unlikely that aMatthew would write Mt 5:18 first and later add a much closer parallel to Mk 13:31 at Mt 24:35, or write Mt 24:35 first and then backtrack to create Mt 5:18.
If aLuke saw Matthew then it is reasonable to suppose that he saw Mt 5:18 and Mt 24:35 in the order in which we see them and simply placed his versions of these verses in the same order, but on the MwQH this does not explain why Mt 5:18 is placed before Mt 24:35. Instead, it is much more likely that Mt 5:18 and Lk 1:17 come from a different source, as on the Mark-Q and MwEL hypotheses.
For a detailed analysis of this doublet see Divorce and Doublets.
Mk 10:2-12 / Mt 19:3-9 are both part of long sections of text on divorce that have no parallel in Luke, while there are much shorter sections on divorce at Mt 5:32 / Lk 16:18 that have no parallel in Mark. In Mark the rules are simple: Mk 10:11 states that if you put away (divorce) your wife and re-marry you commit adultery, and Mk 10:12 applies the same rule to a woman divorcing her husband. However, Matthew and Luke contain some significant differences from both these rules.
Assuming Markan priority we can reasonably assume that both aMatthew and aLuke saw Mk 10:11-12, and these verses would therefore be expected to form the core of the corresponding rules in Matthew and Luke. Indeed, Lk 16:18a-c does follow Mk 10:11 regarding a husband committing adultery if he divorces his wife and re-marries, but Mt 5:32b and 19:9d contain an exception to that rule in the case of fornication by his wife. Mt 5:32 adds other changes that state that it is the act of divorce (except in the case of fornication), not the re-marriage, that causes the adultery but, strangely, in Mt 5:32b the Markan rule is ‘flipped’ so that a husband divorcing his wife causes her to commit adultery, presumably with no penalty for him if he re-marries. Additionally, in both Matthew and Luke Mk 10:12 is replaced by a rule that states that a man who marries a divorced woman commits adultery, although this rule is not present in Mt 19:9ef // 5:32d in some mss.
The contortions required by aMatthew (and to a lesser extent aLuke) virtually rule out any synoptic hypothesis in which Matthew was first or had only one source. There are more complex hypotheses (e.g. with more sources or stages of development) that can address some of these issues, but the less complex hypotheses most likely to be correct all assume Markan priority and that aMatthew and aLuke had access to a non-synoptic source (the most well known of which is Q). However, the Mark-Q hypothesis is not without its own problems, many of which are caused by the requirement that on that hypothesis aMatthew and aLuke did not know anything about each other’s gospel. Relaxing that requirement (e.g. as on the MwEL hypothesis) allows much greater freedom to solve problems such as the differences between Mt 5:32 and 19:9.
In ‘Mark and Q’ Fleddermann examines a Mark-Q overlap that he gives as “17. On Scandal: Mark 9,42 (Matt 18,6-7 par. Luke 17,1b-2),” following which he examines “14a. The Son of Man and That Man: Mark 14,21 (Matt 18,6-7 par. Luke 17,1b-2).” Initially it appears that here are two separate sayings: the first about the betrayal of the Son of man, and the second about more general “offences” against children. However, while the the key phrase “but woe to that man by whom” is present in both halves of the two doublets in Matthew and Luke respectively, it is only present in Mk 14:21, and not Mk 9:42.
From this it can be seen that Mk 14:21a / Mt 26:24a / Lk 22:22 and Mt 18:7 / Lk 17:1 are respectively Triple and Double Tradition parallels, so forming (on the Mark-Q hypothesis) a Mark-Q overlap, while Mk 14:21b / Mt 26:24b and Mk 9:42 / Mt 18:6 / Lk 17:2 are two separate statements containing different suggestions regarding “that man:” one that he should not have been born, and the other that he should be drowned. While the first is something that Jesus might well say, the second is much more likely to derive from something before Jesus, such as is referred to in Ex 10:19 and 15:4. Consequently, it is most likely that Mk 9:42 / Mt 18:6 / 17:2 are not part of this doublet. This is reinforced by Mk 9:41 having no connection to Mk 9:42. Indeed, Mk 9:38-41 seem to simply ‘interrupt’ Mk 9:37 and 42, as if this is part of a collection of individual sayings that have been strung together with little thought for how they read.
On the MwQH aMatthew wrote a direct parallel to Mk 14:21 at Mt 26:24, but also a looser parallel earlier in his gospel, at Mt 18:7. aLuke then appears to have written versions of both parallels at Lk 22:22 and 17:1 respectively, but omitting a parallel to Mt 26:24b. On any hypothesis in which aMatthew and aLuke saw both Mark and another source then Mt 18:7 / Lk 17:1 have the non-Markan source (e.g. Q, Early Luke) as their source. As indicated above Mk 9:42 / Mt 18:6 / Lk 17:2 are simply triple tradition parallels, and do not form a Mark-Q overlap as is believed by some.
It is clear that the source of Mt 21:21 is Mk 11:22-23, and although Mt 17:20 refers to a mountain and Lk 17:6 to a sycamine tree, they are obviously related, either with one depending on the other, or both depending on a different source. In the former case (e.g. as on the MwQH) there is the problem (already noted a number of times) of aMatthew either writing the significantly different variant of Mk 11:23 first (at Mt 17:20) and then later adding a close parallel to Mk 11:23 at Mt 21:21, or backtracking when adding Mt 17:20, making it again more likely that Mt 17:20 and Lk 17:6 have a different source to Mt 21:21, one that appears to have been aMatthew’s primary source, with Mark being his secondary source.
Patton notes that “There is a parallel for this saying [Mk 11:23] in Lk xvii, 6, and there are doublets for it in Mt xvii, 20, and xxi, 21. It stood in Mark and Q,” and then writes:
The parallel here is not close. But Matthew has a doublet in xxi, 21, and Mark a similar saying in xi, 22. The saying seems therefore to have been in both Mark and Q, and was taken by Matthew from both sources and by Luke from one. The connection of the saying in Luke indicates that he took it from Q; yet his saying is not the same as Matthew’s, in that he substitutes a sycamore tree for Matthew’s mountain, thus greatly weakening the comparison. The two sayings certainly cannot have been derived by Matthew and Luke from an identical source. It is only on the ground of their general logian character that they can be assigned to QMt and QLk.
As in many other instances the unlikely behavior of aMatthew regarding how he wrote Mt 21:21 both and 17:20 make it almost certain that there was a non-synoptic source seen by at least one of, and most likely both aMatthew and aLuke.
The table below records all seventeen of the doublets/formulas just discussed above. In each case in the ‘Matthew’ columns any doublet/formula half with a parallel in the same context in Mark is listed first, with the half in double tradition text second (and similarly in the ‘Luke’ columns). For identification purposes ‘D’ denotes a doublet and F denotes a formula. If a doublet or formula has a number then it is one identified by that number by Hawkins. In this group we have eleven doublets/formulas in Matthew and six in Luke, yet not only are there no corresponding doublets/formulas in Mark, in five instances there is no parallel text in Mark at all. There are three instances of doublets in both Matthew and Luke and yet, surprisingly, even one of those has no parallel text in Mark.
As in Summary - Double Tradition Doublets in the Mission or the Twelve and Summary - Double Tradition Doublets in the Main Sequence verses in the double tradition are shown with a colored background. Where they have a parallel in the same gospel (i.e. there is a doublet) and that parallel also has a parallel in Mark then this is a strong indication that these verses do not have a direct parallel in Mark, e.g. on the Mark-Q hypothesis the source of these verses is Q. However, other sources are possible, for example on the MwQH any double tradition verse in Luke can have Matthew as a source. Verse numbers in red mark doublets in which the doublet half with a direct parallel in Mark is located after the half in the double tradition.
The first five doublets/formulas shown above differ from the others in that they have no parallel in Mark, but still have doublets in Matthew and/or Mark. With one exception they have no known source, and so either both halves of the doublets/formulas came from a common source (e.g. Q), or one half came from an unknown source and the other is a ‘copy’ made by the author of the first half. Q supporters believe these verses came from Q even though there is no possible Mark-Q overlap here.
There are no doublets/formulas here in Mark, but with the above five exceptions the halves of the doublets/formulas that are not in double tradition text have a parallel in Mark in the same context. This means that on hypotheses in which Matthew and Luke had a non-Markan source there is a Mark-source overlap (e.g. a Mark-Q overlap on the Mark-Q hypothesis) in these doublets/formulas, 13 of which H.T. Fleddermann records as ‘Overlap Texts’ in Mark and Q:
Mark 3.22-37 (Matt 12,22-30.43-45 par. Luke 11,14-15.17-26) The Beelzbul Controversy
Mark 3:28-30 (Matt 12,32 par. Luke 12,10) The Unforgiveable Sin
Mark 4,21 (Matt 5,15 par. Luke 11,33) The Lamp
Mark 4,24cd (Matt 7,2b par. Luke 6,38c; Matt 6,33b par. Luke 12,31b) The Measure
Mark 8,11-13 (Matt 12,38-42 par. Luke 11,16.29-32) Demand for a Sign
Mark 12.38-39 (Matt 23,6-7 par. Luke 11,43) First Places
Mark 10,31 (Matt 20,16 par. Luke 13,30) The First and the Last
Mark 13,31 (Matt 5,18 par. Luke 16,17) Jesus’ Words
Mark 10,11-12 (Matt 5,32 par. Luke 16,18) On Divorce
Mark 11,22-23 (Matt 17,20 par. Luke 11,10) On Faith
Mark 11,24 (Matt 7,8 par. Luke 11,10) On Asking and Receiving
Mark 13,35 (Matt 24,44 par. Luke 12,40) Uncertainty of the Hour
Mark 14,21 (Matt 18,6-7 par. Luke 17,1b-2) The Son of Man and That Man
The direct correspondence between the doublets and the Mark-Q Overlaps provides very strong support for the view that Matthean and Lukan parallels to Mark in ‘Markan blocks’ (or in a ‘Markan context') have Mark as their source, while those in the Double Tradition have a different source. Whether that source was Q, an early version of Luke, something derived from an Aramaic source, or something else, is a different issue.
The doublets/formulas are listed in order of the double tradition verses in Luke, and in many (shown above in red) the half of a doublet/formula with a corresponding parallel in Mark occurs later in that gospel than the half with no parallel in Mark. This presents a problem for the MwQH / Farrer Theory since on that hypothesis Mark is the only known source for Matthew, making it hard to understand why in 8 instances out of 10 (80%) aMatthew placed his close parallel to Mark much later in his gospel than the (usually) looser parallel with no corresponding Markan context. aLuke (who saw both Mark and Matthew) then reversed the expected order in 4 instances out of 6 (66.7%), and while this would be easily understandable if aLuke was following the order in Matthew, in reality he only followed the order in Matthew in two instances. This reversing of order is much easier to understand on the basis of aMatthew and aLuke seeing two different sources, with each making their own decisions as to how to ‘prioritize’ the text from each when it came to creating doublets. Then, even if either aMatthew or aLuke also saw the other’s gospel it may have not made a difference to their choices here, depending on how they viewed the other gospel as a reliable source.
There are no doublets/formulas here in Mark, and despite the statement at the beginning of this study indicating that Markan priority is assumed, it is worth noting here just how unusual the actions of Mark would have been if Matthew was first and Mark last, as on the Griesbach or Two-Gospel Hypothesis (2GH). In this group we have eleven doublets/formulas in Matthew and six in Luke, yet not only are there no corresponding doublets/formulas in Mark, in five instances there is no parallel text in Mark at all. There are five instances of doublets in both Matthew and Luke and yet, incredibly, even one of those has no parallel text in Mark. If Mark was the last gospel to be written then (as well as excluding everything we see as Sondergut Matthew, Sondergut Luke, and the Double Tradition) aMark also deliberately chose to exclude multiple instances of text doubly attested in both Matthew and Luke. There really is no rational explanation for such behavior, even if aMark was attempting to produce a more concise record of some of the life of Jesus by taking extracts of both Matthew and Luke with virtually no additional material of his own.