Although most of the Double Tradition doublets discussed above have a corresponding parallel in Mark (so that on the Mark-Q hypothesis there is a possible Mark-Q overlap), none have a doublet in Mark that potentially could be the source of both halves of the doublets in Matthew and/or Luke. Because in all instances in this group there are doublets in Mark then on the assumption of Markan priority there are two possible places in Mark that could be the source of parallel text in Matthew or Luke, and hence there is no immediate reason to suggest a non-synoptic source for any doublet in either Matthew or Luke in this group. However, there still may be differences in the parallels or related text that do appear to point in that direction.
Hawkins notes five formulas in Mark that have no corresponding formula in either Matthew or Luke (‘Peculiar to Mark’ in his terminology). On the assumption of Markan priority these are all just short phrases that aMark used in two places in his gospel, but that neither aMatthew nor aLuke included in their respective gospels. Some of these phrases appear so unexceptionable as to make it hard to understand why they would not have been included in Matthew and Luke, suggesting that their omission could be attributed to aMatthew or aLuke not seeing a portion of the text due to damage to their respective copies of Mark. In addition, Hawkins notes two closely related formulas (Nos. 2 & 3 in Luke) that have parallels in Matthew and Mark, and a single doublet in Mark that he discusses together with his doublet in Matthew No. 13. In addition, the 16th formula in his list of formulas in Matthew with parallels in Mark and/or Luke includes a formula in Mark.
Mk 3:20-21 contains the suggestion (only in Mark) that Jesus’ family believed that he was mad, but in their place Matthew and Luke contain the double tradition verses Mt 12:22-23 / Lk 11:14 about Jesus casting out a devil. In all three gospels these verses act as an introduction to the controversy over Beelzebub at Mk 3:22-26 / Mt 12:24-28 / Lk 11:15-20, suggesting that aMatthew and aLuke saw Mk 3:19b-21 but both chose to reject it and replace it by a common passage about casting out a devil. For details see Mk 3:22, Mt 9:32-34 // 12:22-24, Lk 11:14-15 - Casting out devils (Hawkins: Doublet in Matthew No. 17) above.
As Mk 6:30a and 32 have parallels at Mt 14:12b-13 and Lk 9:10 it would appear likely that aMatthew and aLuke would have also seen Mk 6:31. However, neither Matthew nor Luke contain parallels to Mk 6:16b, 17c, 20c, 21b, 22b, 24b-25a, 27b, or 30b either, so it is possible that this portion of Mark was damaged in some way and that neither aMatthew nor aLuke saw Mk 6:31, or that it was unreadable.
Mk 4:2b-3a is the only portion of Mk 4:1-6 that is unique to Mark, and Mk 12:37b-38a is the only portion of Mk 12:35-40 that is unique to Mark. διδαχῇ (teaching) is used in all three synoptic gospels, but these are the only uses of ἔλεγεν … ἐν τῇ διδαχῇ αὐτοῦ (saying … in his teaching), a phrase that both aMatthew and aLuke appear to have chosen to not include in their respective gospels.
Hawkins notes the following two formulas in Luke that are also used in Mark and/or Matthew (realigned below for emphasis). Although Hawkins Formula 3 is related to Formula 2, it is sufficiently different to require its own analysis.
F2a: Your faith has saved you - ἡ πίστις σου σέσωκέν σε Lk viii. 48 = Mk v. 34; Mt ix. 22 ; (and) Lk vii. 50
F2b and Lk xviii. 42 = Mk x. 52 : also Lk xvii. 19
F3: Go in peace - πορεύου εἰς εἰρήνην Lk viii. 48 = Mk v. 34 (ὕπαγε) : also Lk vii. 50.
Hawkins notes Jesus’ repeated reference to a person’s faith having saved them. In the KJV there appear to be three distinct parallel phrases related to Formula 2, two of which are repeated: “thy faith hath made thee whole/saved thee.” However, these are just differences in translation, as all seven examples have identical Greek: ἡ πίστις σου σέσωκέν σε. In Lukan order they are:
One at the end of the unique Lukan ending to the triple tradition passage of the woman with the alabaster box at Lk 7:39-50;
Three parallels at the ends of the triple tradition passage of the woman with the issue of blood at Mk 5:25-34, Mt 9:20 -22, Lk 8:43-48;
One at the end of the unique Lukan passage of the ten lepers, at Lk 17:19;
Two at the end of the Markan and Lukan versions of the healing of the blind man at Mk 10:46-52, Lk 18:35-43.
The healing of the blind man in Mark and Luke both also have a parallel at Mt 20:29-34 that in most details is very close, except that in Matthew Jesus:
Heals two blind men instead of one, and;
Heals them by touching their eyes instead of telling them that their faith had saved them.
In Matthew there is also a unique earlier healing of two blind men at Mt 9:27-31 (following the healing of the daughter of Jairus) close to the end of which Jesus heals them by saying: ‘According to your faith be it unto you.’ Κατὰ ὑμῶν τὴν πίστιν γενηθήτω ὑμῖν .
On the MwQH aMark was first to use the phrase, at Mk 5:34 and 10:52. aMatthew then used the phrase in his parallel of Mk 5:34 at Mt 9:22 but not in his parallel to Mk 10:52, but did use a similar phrase at Mt 9:29. aLuke saw both Mark and Matthew, created parallels to Mk 5:34 and 10:52, saw and ignored Mt 9:29 and 20:34, and added unique uses of the phrase at Lk 7:50 and 17:19.
On the Mark-Q hypothesis aLuke did not know Matthew, so he did not use Mt 9:29 and 20:34 because he did not know them. Although none of the verses shown above are in the double tradition it is possible that an ’extended’ version of Q could have contained a version of those verses that have parallels in Matthew and Luke, i.e. Mk 5:34a / Mt 9:22b / Lk 8:48 and Mk 10:52a / Mt 20:34a / Lk 18:42. However, as in both cases there is a Markan parallel this alone could explain the parallels in Matthew and Luke, but not the healing at Mt 9:29-34 that is unique to Matthew.
On the MwEL hypothesis Early Luke could have contained the passages that we see as unique to Luke, i.e. Lk 7:39-50 and 17:7-21, although as on that basis aMatthew would have also seen and rejected them it is perhaps more likely that they are additions to Luke that aMatthew did not know. Similarly, if Lk 18:42 had a parallel in Early Luke then it makes it hard to understand why aMatthew rejected what he saw in both Mk 10:52a and ELk 18:42 in favor of what we see in Mt 20:34a.
Overall there is nothing here that points decisively to any of these synoptic hypotheses. Because all three synoptic gospels contain the same phrase using identical Greek, it seems likely that all three authors knew this exact phrase from somewhere else. However, the two unique additions in Luke suggest a source (whether oral, written or tradition) known to aLuke but not to aMark or aMatthew.
Hawkins also notes the repetition of ‘go in peace’ (ὕπαγε εἰς εἰρήνην) in Mk 5:34 and Lk 7:50 // 8:48 (Formula 3). As this expression appears many times in the OT (Old Testament) it is almost certainly a common phrase that aMark chose to use at Mk 5:34, that aLuke then used in the parallel at Lk 8:48, and also at the end of the unique Lukan conclusion to the passage of the woman with the alabaster box at Lk 7:50. It may or may not be significant that Jesus only says this to the women, and instead just says ‘go [thy way]’ to the men at Mk 10:52a and Lk 17:19.
Because ‘go in peace’ appears many times in the OT it is tempting to suggest that the OT is the source of the whole phrase, but nowhere in the OT is anyone saved by faith. Instead, as Eph 2:8 states: “For by grace are ye saved through faith; and that not of yourselves” it appears that this concept was more generally known and used by all three synoptic gospel writers, perhaps via access to Paul or his writings.
Assuming Markan priority neither aMatthew nor aLuke chose to include the first half of this doublet, or perhaps they did not see either half in their copies of Mark. However, on this point Mk 6:20c is in a section of Mark not present in Luke, and only present in significantly shorter form in Matthew (Mk 6:17-29 – Herod and John), which could indicate damage to this area of Mark, only a few verses earlier than those omitted in The Great Omission in Luke (See Luke's Great Omission: Luke 9:9-10 Another Omission). Otherwise there is no obvious reason why both aMatthew and aLuke would exclude these words. It should also be noted that that Mk 6:20c contains an odd variant involving one word, with ‘did (or doing) many things’ making no sense, and so also suggestive of damage at this point. Alternatively, on the MwQH it is possible that there is no parallel to Mk 12:37b in Luke because the parallel at Mt 22:46a is so different, even though this is the end of a triple tradition passage.
Hawkins notes οὐδένα ἤθελεν γνῶναι in Mk 7:24 and οὐκ ἤθελεν ἵνα τις γνοῖ in Mk 9:30. Although these are not exactly the same they express the same desire by Jesus for secrecy that is seen elsewhere (predominantly in Mark) with regard to healings. As in the previous formula both verses are in the area of the Great Omission in Luke, so there may be a common explanation for the lack of parallels in Luke.
Mk 9:35 // 10:43-44 is the only doublet Hawkins records in Mark, about which he notes only the following:
Mk ix. 35 with x. 43, 44 : for this see Matthew No. 13, on p. 73 above.
f1, pc22 omit ‘last of all, and,’ from Mk 9:35, and the majority of this verse is omitted in D, d, k, so this is possibly a Western non-interpolation. In addition as neither Matthew nor Luke contain any parallel to this verse there is reasonable doubt as to its originality.
For his doublet in Matthew No. 13 Hawkins has Mk A = Mk 9:35, Mk B = Mk 10:43-44, Mt A = Mt 2:26-27, Mt B = Mt 23:11, and Lk = Lk 22:26. He writes:
Here again the identity of language in Mt A and Mk B is almost complete, and points decidedly to a common source, which would be generally held to be a Marcan one.
In this one case a doublet in Mark is entered, πρῶτος and ἔσχατος being used in both passages although not in the same order. The combination πάντων διάκονος occurs only in Mk A: it is one of the very few expressions peculiar to Mk which are found in sub-apostolic writings, being applied to Christ in Ep. Polycarp. v. 2.
The passage from Lk has a link to Mt A and Mk B in its context and opening, but to Mt B (a probably Logian passage) in ὁ μείζων; so its origin is very doubtful.
πάντων is used only in Mk A and B: the Matthaean saying have nothing peculiar to them.
There is little here to point to a second source for these verses, except that in the triple tradition passage about the chief seats and uppermost rooms the Matthean version at Mt 23:1-14 is much longer than the parallels at Mk 12:38-40 / Lk 20:46-47, with only Mt 23:1-2, 5b-7a, 14b having parallels in both Mark and Luke. The majority of the other verses are unique to Matthew and so suggest that aMatthew knew a longer version of this passage than did either aMark or aLuke. However, Mt 23:4 has a parallel at Lk 11:46bc but none in Mark, and Mt 23:11 has no parallel in Luke, but does have a parallel at Mk 9:35, as noted by Arthur Peake in his Commentary on the Bible:
In his account Jesus does not embrace the child (cf. Matthew 19:15, Mark 10:16), and the saying of Mark 9:35 is omitted, or rather reserved till Matthew 23:11.
However, Peake is only partially correct as only the last portion of Mk 9:35 is paralleled in Mt 23:11, which may be the result of aMatthew not seeing the whole verse, a possibility raised by Thomas Francis Glasson in ‘An Early Revision of Mark’:
I should prefer to think variant readings which arose in different ways, some of them just scribal blunders. To give an example: in Mark 9:35 the Western authorities D k omit the words of our Lord, and the only part of this verse then retain is the opening: “And sitting down he called the twelve and.” They then continue with vs. 36. “(And) taking a child …” avoiding the repetition of the word “and.” … If something like this Western text of Mark lay before Matthew and Luke, we can understand their remarkable agreement in passing from vs. 34 to 36, and in making the child an object lesson…
Willker (Mark, TVU 202) notes that D, d, and k omit Mk 9:35b while f1, pc22 omit just ‘last of all,’ and that this could be considered a Western non-interpolation, adding:
Mt and Lk use a different wording and omit this sentence (Minor Agreement?).
Perhaps the omission is due to parablepsis?
9:35 καὶ λέγει … 36 καὶ λαβὼν
et ait illis … et accepit
He then notes that “A similar sentence appears later” at Mk 10:44 / Mt 20:27 / Lk 22:26 (as above), but does not note the parallel in Mt 23:11, perhaps because it is not in the same context as any of the other verses. It is notable that there is no ‘disconnect’ between Mt 23:10 and 12 if Mt 23:11 is discarded even though it is related to the issue of the Pharisees wanting to be called Rabbi to show their superior status:
Neither be ye called masters: for one is your Master, even Christ. [Mt 23:10]
And whosoever shall exalt himself shall be abased; and he that shall humble himself shall be exalted. [Mt 23:12]
As Mt 23:11 has the only reference to ‘servant’ in Mt 23:1-14 it appears likely that, as Mk 9:35b has no parallel in Matthew or Luke in the same context (Mt 18:1-2, Lk 9:46-47), and as it is one of the very many small pieces of text in Mk 9 that have no parallel in the same context in either Matthew or Luke (c.f. Mk 9:10b, 14a,c, 15-16, 20b-21, 22b-25a, 25c, 26b-28a, 30b, 33-34a, 36c, 39b), Mt 23:11 is likely to have its origin in Mk 9:35b. The lack of parallels to all this Marcan text supports the damage hypothesis explored in The Not So Great Omission, and on this hypothesis it is reasonable to suggest that there was damage to the text of Mk 9:35 that resulted in some mss having different text here. Then, aMatthew may have seen a copy of Mark in which some of the text of Mk 9:35b was loose, causing him to insert a parallel to what text he did see at what he considered to be the most appropriate position, i.e. as Mt 23:11.
As stated at the beginning of this section there is nothing here to suggest a non-Markan source for these verses in Matthew and/or Luke. Although servants are mentioned many times in the OT, the phrase “servant of all” is not, and so the source of Mk 9:35b // 10:44 is not known. On the assumption that aMatthew knew Mark then these verses are the source of Mt 23:11 // 20:27 respectively, and either Mark or Matthew is the source of Lk 22:26. However, if aMatthew did not see Mk 9:35 in situ then it is possible that aLuke did not see this text at all, suggesting that aMark may have been aLuke’s primary source.
Mk 9:36 contains καὶ ἐναγκαλισάμενος αὐτὸ while Mk 10:16 has καὶ ἐναγκαλισάμενος αὐτὰ. On the assumption of Markan priority these appear to be Markan expressions that both aMatthew and aLuke ignore in otherwise triple tradition passages. However, neither Matthew nor Luke contain parallels of Mk 9:33-34a, 35, 36c, 39b, nor parallels of Mk 10:16a,c and 10:17a,c.
On the MwQH the simplest explanation for the lack of this text in both Matthew and Luke is that aMatthew omitted it and aLuke followed Matthew.
On the MwEL hypothesis the simplest explanation is that the author of Early Luke omitted the text and aMatthew and aLuke both used the shorter form.
On the Mark-Q hypothesis it would seem impossible for both aMatthew and aLuke to have chosen to omit exactly the same portions of text unless Q contained the shorter versions of both passages, in which case they would both have to be considered as additional Mark-Q overlaps.
Hawkins records this formula as follows:
ὄψονται τὸν υἱὸν τοῦ ἀνθρώπου ἐρχόμενον Mt xxiv. 30 = Mk xiii. 26 ; Lk xxi. 27 : and
ὄψεσθε κ.τ.λ. in Mt xxvi. 64 = Mk xiv. 62 :
also Mt xvi. 28 ἴδωσιν τὸν υἱὸν τοῦ ἀνθρώπου ἐρχόμενον (where Mk ix. 1 and Lk ix. 27 mention only the kingdom as being seen).
This can be re-written as: Mk 13:26 // 14:62 = Mt 24:30 // 26:64 = Lk 21:27 (but not Lk 22:69 as it mentions neither seeing nor coming), and Mk 9:1 / Mt 16:28 / Lk 9:27. As Hawkins notes, Mk 9:1 / Lk 9:27 do not include the formula used in the other verses because they refer only to the kingdom being seen instead of the Son of man being seen coming/sitting.
In addition, because Mt 16:28 refers to the coming of the kingdom it seems more likely that aMatthew saw Mk 9:1 rather than Lk 9:27. It is then conceivable that aMatthew also saw Mk 13:26 and/or 14:62 before writing his gospel and chose to refer to the Son of man in his parallels to these places. As these are all triple tradition passages they do not provide any other insight into the synoptic problem.
None of the doublets noted by Hawkins as ‘Peculiar to Mark’ (Here identified as P1-5) have a parallel in either Matthew or Luke, although in three instances there is different text at the corresponding locations in Matthew and/or Luke:
P1: Mk 3:20-21 vs. Mt 12:22-23 / Lk 11:14
P2: Mk 12:38a vs. Mt 23:1 / Lk 20:45
P3: Mk 12:37b vs. Mt 22:46a
On the assumption of Markan priority the existence of this different text shows either that aMatthew and/or aLuke did not see the Markan text and recognized that something was missing at these points, or did see text but found some or all of it unreadable, or that is was not acceptable to them in other ways. In the case of P1, P4 and P5 there is no evidence that either aMatthew or aLuke attempted to ‘fill in the gap’ with other text.
None of these doublets/formulas in Mark provide support for the Mark-Q hypothesis, because none of them have parallels in Matthew or Luke that are either in double tradition text or are one half of a double tradition doublet. In addition. some of these pieces of text in Mark appear to be so innocuous that assuming Markan priority it seems very unlikely that both aMatthew and aLuke would omit them unless there was a common reason for the omission. However, some have seen this as evidence that these same pieces of text are additions by aMark instead of exclusions by aMatthew and aLuke.
In ‘The Synoptic Problem: A Critical Analysis’ William Reuben Farmer comments on the formulas that Hawkins’ identifies, first referencing those “peculiar to each Synoptist” (here P1, P2, etc.). He counts 15 in Matthew, 6 in Mark and 12 in Luke (although in Horae Synopticae Hawkins actually lists only 8, 5, and 5 respectively), and then notes those which Hawkins states are:
… used once (or in a few cases twice) by a Synoptist in common with one or both of the others, and are also used by that Synoptist independently in other parts of his narrative.
Then, regarding those in Matthew Hawkins notes that:
A careful examination of such cases certainly leaves the impression that the mind of [the author of] Matthew was so familiar with these collocations of words that he naturally reproduced them in other parts of his narrative, besides the places in which they occurred in his sources. It is to be observed that these apparent reproductions often occur earlier in the Gospel than do the apparently original occurrences of the formulas, which seems to indicate that Matthew drew them from his memory of the sources and not from documents before him. So far as it goes, then, the drift of this section is in favour of the oral theory.
Of course, here Hawkins appears to be assuming an oral theory, since an obvious reason why “these apparent reproductions often occur earlier in the Gospel than do the apparently original occurrences” would be that they were present in another source that was used later.
Of these (here F1, F2, etc.) Farmer notes there are 7 in Luke and 19 in Matthew (Hawkins lists 6 and 17 respectively) while for Mark he writes: “He [Hawkins] recorded none for Mark presumably because he had been unable to find any clear cut instances in that Gospel,” and continues:
Of the 7 formulas which appeared to Hawkins to be “favorite or habitual expressions” of Luke, 5 are found at least once in Mark in parallel passages, and 2 occur also in Matthean parallel passages.
Of the 19 formulas which appeared similarly to Hawkins to be characteristic of Matthew, 14 are found at least once in Mark in parallel passages, and 10 occur also in Luke.
This means that formulas which appear to be “favorite or habitual expressions” of Matthew are found frequently both in Mark and Luke in parallel passages where there is evidence of copying. Such formulas of Luke also occur fairly frequently in Mark. There seem to be no such expressions characteristic of Mark, however, which show up in either Matthew or Luke. This fact is particularly difficult to understand on Streeter’s theory concerning Marcan priority… This would seem to be especially true in the case of Matthew where the amount of verbatim agreement between Matthew and Mark is so great that if Matthew copied Mark it would seem to be highly unlikely that he would have averted all characteristic expressions of Mark… Hawkins had no ready explanation for this evidence on the Marcan hypothesis, and since the Griesbach hypothesis was no longer under consideration because of the arguments of Abbott and Woods, Hawkins was constrained to find some other explanation.
Farmer contrasts the fact that “favorite or habitual expressions” in Luke are found in Mark and Matthew and that similar Matthean expressions are found in Mark and Luke, but that similar “favorite or habitual expressions” in Mark do not “show up in either Matthew or Luke.” This might be interesting if there actually were a significant number of such “favorite or habitual expressions” in Mark, but there are not, as Farmer himself notes. Instead, we are left with just those 5 (or 6) expressions ‘Peculiar to Mark’ none of which can be termed “favorite or habitual expressions” because each occurs in just two places in Mark. Farmer then points out that “Hawkins had no ready explanation for this evidence” “that if Matthew copied Mark it would seem to be highly unlikely that he would have averted all characteristic expressions of Mark,” but how could he, when those “characteristic expressions of Mark” appear not to exist?