Major and minor agreements.
Matthew and Luke agreeing against Mark.
Also refer to the list of agreements of Matthew and Luke against Mark.
Potential solutions to the synoptic problem may be broadly divided into two groups:
To the second category belong the Augustinian theory (Matthew-Mark*), the Farrer theory (Mark-Matthew*), and the Greisbach theory (Matthew-Luke*), among others less well-known. The first category covers very many theories, of course, but by far the best-known is the two-source theory (Mark-Q*).
* Refer to my page on synoptic nomenclature for an explanation of this system, which is derived from Stephen Carlson.
It is only in theories of the first category that agreements between any two of the synoptic gospels against the third become an issue; in theories of the second category the third evangelist knew both of his predecessors and was therefore free to agree with one or the other as he saw fit. But, in cases in which it is theorized that the second and third evangelists wrote in ignorance of each other, it becomes essential to explain agreements between them against the first evangelist, whom both knew and followed.
Since by far the dominant theory of the first category has been Mark-Q, the very term agreements has in synoptic studies come to refer to those instances in which Matthew and Luke agree against Mark. To this term the adjective minor is regularly applied, as indicating that these agreements against Mark are usually quite insignificant and probably coincidental. In most standard treatments of the synoptic problem you will find at least one chapter dedicated to the minor agreements.
It is with great care that I employ the word usually in that description. For there are instances in which the agreement simply cannot be chalked up to coincidental editing of Mark by both Matthew and Luke. In those instances, few as they may be, it seems unwise to apply the adjective minor. So I myself like to refer to some of the agreements as major agreements.
To be sure, there will be a degree of subjectivity in deciding which agreements to classify as minor and which as major, but my basic procedure is to separate out those agreements which on the Mark-Q theory must be categorized either as the result of slimly attested (or even unattested) textual corruption, or as an overlap between Mark and Q in the triple tradition, or as an overlap with some other source, including perhaps a strong oral tradition. These cases are the major agreements. Those cases that remain once the major agreements have been sifted out ought to be attributable to coincidental editorial work based on tendencies discernible elsewhere in Matthew and Luke, or perhaps elsewhere in Hellenistic Greek.
I am posting these agreements as I go further along in my synopses, so by no means are all of the agreements represented yet. I am also making an ongoing and cumulative raw list of agreements of Matthew and Luke against Mark.
These agreements I do not think can be explained away on the Mark-Q hypothesis without recourse to either textual corruption or the influence of Q, or to some other early text or tradition. In other words, they are probably not the result of independent editing. I have not finished excavating these agreements, so no need to panic if you do not see your favorite on the list yet.
The region of the Jordan.
From the introduction of John the baptist. Matthew 3.5b = Luke 3.3a, respectively:
The Marcan parallel would be 1.4-5.
The sermon of John, part 1.
At Matthew 3.7-10 = Luke 3.7-9 the first and third evangelists agree against Mark in giving a speech by John the baptist to the Pharisees and Sadducees (Matthew) or to the crowds (Luke). Within the speech proper, after the introduction, Matthew agrees with Luke in 60 out of 63 words and Luke agrees with Matthew in 60 out of 64 words:
The Marcan parallel would be 1.7-8.
The sermon of John, part 2.
In the sermon of John the baptist Matthew 3.11 and Luke 3.16 agree in the present tense βαπτιζω against the Marcan aorist εβαπτισα in Mark 1.8. Both Matthew and Luke also have John the baptist speak of his own baptismal rite before speaking of the one to come; Mark 1.7-8 has him speak of the one to come before describing his own baptismal rite. Furthermore, Matthew 3.11 and Luke 3.16 have υμας βαπτισει instead of the Marcan βαπτισει υμας.
Both Matthew 3.11b-12 and Luke 3.16b-17 extend the second part of the sermon of John the baptist by some 28 (Matthew) or 27 (Luke) words, of which 25 are verbatim between them:
The three temptations.
Matthew 4.1-11, Mark 1.12-13, and Luke 4.1-13 all place the tempting of Jesus in the desert at the same narrative locus; Matthew and Luke, however, actually describe three temptations at the hands of the devil, while Mark gives no content to the testing at all. The relevant section is Matthew 4.2-11a = Luke 4.2b-4, [9-12], [5-8], 13:
In close conjunction with the preaching of Jesus Matthew 4.13 has Jesus leaving Nazara (Ναζαρα) to settle in Capernaum, while Luke 4.16 has Jesus entering Nazara (Ναζαρα) to teach in the synagogue there.
The relevant parallels would be Mark 1.14-15, 21; 6.1; but this special name for Nazareth appears nowhere in the rest of the gospel, either.
Given that the town name is so differently applied by Matthew and Luke, this agreement would not mean much except that Nazara (Ναζαρα) is not the usual name of the town. The usual name is Nazareth (Ναζαρετ or Ναζαρεθ).
Fallen on the sabbath.
The setting of the great sermon.
Matthew 3.1-4.22 parallels Mark 1.2-20 very closely. For Matthew 4.23-25, however, we have to skip around a bit to pick up the Marcan parallels:
In Matthew notice of the sermon on the mount follows immediately (in 5.1). Notice that, although our global narrative point is Mark 1.21, the local Marcan context is actually 3.7-8, the description of the great crowds coming to follow Jesus. After Mark 3.9-12, which goes almost completely unparalleled in Matthew, Mark now describes the commission of the twelve disciples. Matthew 5.1, then, is parallel to Mark 3.13, a fact to which we shall return later.
Luke has led up to his sermon on the plain a bit differently. With the exceptions of the transposed call of the first disciples (in Mark 1.16-20 = Luke 5.1-11) and rejection at Nazareth (in Mark 6.1-6a = Luke 4.16-30), Luke 3.1-6.11 parallels Mark 1.2-3.6 very closely. At this point Mark and Luke run as follows (in Lucan order):
The unit missing in Mark is the introduction to the sermon on the plain. Notice, though, that Luke 6.12 is parallel to Mark 3.13.
So both Matthew 5.1 and Luke 6.12 are parallel to Mark 3.13 (and let us also look at Luke 6.13a):
In all three gospels at this point Jesus is ascending a mountain and his disciples are coming to him. The local transpositions ought not to obscure the fact that these separate passages are parallel:
I might mention that at this point in Mark Jesus goes into a house (Mark 3.20), which probably means Capernaum (see Mark 2.1). Matthew, after Jesus has finished his sermon, has him heal a leper on his way down the mountain (Matthew 8.1-4) then proceed into Capernaum (Matthew 8.5). Luke, also after the sermon, has Jesus go into Capernaum (Luke 7.1).
But this parallelism means that the context for the sermon on the mount or plain is actually Marcan. That Matthew and Luke would both choose to insert their respective sermons into this same Marcan context is a major agreement against Mark.
Wine poured out.
Several minor agreements toward the end of the controversy over fasting accumulate to the extent that I think it is justified to view them as a major agreement:
The healing of a mute man.
This healing is a sort of preface to the Beezebul controversy pericope that Matthew and Luke contain but Mark does not; however, the healing in Matthew 12.22-23, which actually precedes the Beezebul material, is not as close to Luke 11.14 as Matthew 9.32-33 happens to be:
There is no parallel in Mark (refer especially to Mark 3.22).
The Beezebul controversy, part 1.
This is the first major agreement of Matthew and Luke against Mark in the Beezebul controversy pericope proper. Matthew 12.25 has ειδως δε τας ενθυμησεις αυτων (but he, realizing their thoughts), and Luke 11.17 has αυτος δε ειδως αυτων τα διανοηματα (but he himself, realizing their reasonings), neither of which is paralleled in Mark 3.23.
The Beezebul controversy, part 2.
This is the second major agreement of Matthew and Luke against Mark in the Beezebul controversy pericope proper. Matthew 12.25 has πασα βασιλεια μερισθεισα καθ εαυτης ερημουται (every kingdom divided against itself is laid waste), and Luke 11.17 has πασα βασιλεια εφ εαυτην διαμερισθεισα ερημουται (every kingdom divided up against itself is laid waste), while Mark 3.24 has και εαν βασιλεια εφ εαυτην μερισθη, ου δυναται σταθηναι η βασιλεια εκεινη (and, if a kingdom is divided against itself, that kingdom cannot stand).
The Beezebul controversy, part 3.
This is the third major agreement of Matthew and Luke against Mark in the Beezebul controversy pericope proper. Matthew 12.26 has πως ουν σταθησεται η βασιλεια αυτου (how then will his kingdom stand), and Luke 11.18 has πως σταθησεται η βασιλεια αυτου (how will his kingdom stand), while Mark 3.26 has ου δυναται στηναι (it cannot stand).
The Beezebul controversy, part 4.
This is the fourth major agreement of Matthew and Luke against Mark in the Beezebul controversy pericope proper:
There is no parallel in Mark (refer especially to Mark 3.26-27).
The Beezebul controversy, part 5.
This is the fifth major agreement of Matthew and Luke against Mark in the Beezebul controversy pericope proper:
Matthew and Luke place this saying right at the end of the Beezebul pericope, while Mark, paralleled by Luke, places a similar saying elsewhere.
A word against the son of man.
In Matthew 12.32 Jesus says concerning the sin against the spirit: Και ος εαν ειπη λογον κατα του υιου του ανθρωπου, αφεθησεται αυτω (and, whoever should say a word against the son of man, it will be forgiven him). Luke 12.10 reports these words as: Και πας ος ερει λογον εις τον υιον του ανθρωπου, αφεθησεται αυτω (and everyone who shall say a word against the son of man, it will be forgiven him). Mark 3.28-29 offers no parallel.
To know the mysteries.
Calming the sea.
Matthew 8.23-27a = Luke 8.22b-25a evince several striking agreements against Mark 4.36-41a. First, Matthew 8.23 has εμβαντι αυτω εις το πλοιον ηκολουθησαν αυτω οι μαθηται αυτου (when he embarked onto the boat his disciples followed him) and Luke 8.22 has αυτος ενεβη εις πλοιον και οι μαθηται αυτου (he himself also embarked onto a boat, and also his disciples), but Mark 4.36 has παραλαμβανουσιν αυτον ως ην εν τω πλοιω (they take him along with them, just as he was, in the boat). Second, Matthew 8.24 has εν τη θαλασση (on the sea) and Luke 8.23 has εις την λιμνην (upon the lake), while Mark 4.37 has no parallel. Third, Matthew 8.25 has και προσελθοντες ηγειραν αυτον, λεγοντες (and they came toward him and woke him, saying) and Luke 8.24 has προσελθοντες δε διηγειραν αυτον, λεγοντες (but they came toward him and woke him up, saying), but Mark 4.38 has και εγειρουσιν αυτον και λεγουσιν αυτω (and they wake him and say to him). Fourth and finally, Matthew 8.27 has οι δε ανθρωποι εθαυμασαν, λεγοντες (but the men wondered, saying) and Luke 8.25 has φοβηθεντες δε εθαυμασαν, λεγοντες (but they were afraid and wondered, saying), while Mark 4.41 has και εφοβηθησαν φοβον μεγαν, και ελεγον (and they were afraid with a great fear, and they said).
Lord of the harvest.
Luke 10.2 (from the mission discourse):
There is no corresponding saying in Mark 6.7-13.
The kingdom has come near.
From the mission discourse. Matthew 10.7-8:
The corresponding Marcan verse, Mark 6.8, has no such phrase.
No staff, no shoes.
When Jesus is instructing his disciples on what they may and may not take along on their mission to Israel, Matthew 10.9 (μηδε αργυρον, nor silver coinage) and Luke 9.3 (μητε αργυριον, nor silver) agree on forbidding silver, while Mark 6.8 is silent on that commodity. Furthermore, Matthew 10.10 (μηδε υποδηματα, nor shoes, and μηδε ραβδον, nor a staff) and Luke 9.3 (μητε ραβδον, not a staff) and 10.4 (μη υποδηματα, no shoes) agree on forbidding a staff and shoes. Mark 6.8, on the other hand, allows nothing except a staff only (ει μη ραβδον μονον) and allows the disciples to wear sandals (υποδεδεμενους σανδαλια).
The worthy worker.
From the mission discourse. Matthew 10.10:
Mark 6.9-10 has nothing to match this saying.
Peace on the house.
From the mission discourse:
Mark 6.10 lacks any such instructions.
From that city.
From the mission discourse:
Sheep amidst wolves.
From the mission discourse. Matthew 10.16a:
Casting out the salt.
Healing the crowds.
Matthew 14.13b-14 = Luke 9.11, respectively:
Mark 6.33-34a lacks mention both of following after Jesus and of the healings:
Scandals will come.
Mark 9.42 has no equivalent.
Jesus and the high priest.
Matthew 26.63b-64 = Luke 22.67-70 evince several striking agreements against Mark 14.61b-62. First, Matthew 26.63 has ημιν ειπης ει συ ει ο Χριστος (might say to us if you are the Christ) and Luke 22.67 has ει συ ει ο Χριστος ειπον ημιν (if you are the Christ say it to us); Mark 14.61 has only συ ει ο Χριστος (you are the Christ or are you the Christ). Second, Matthew 26.64 has απ αρτι (from this moment) and Luke 22.69 has απο του νυν (from now on) where Mark 14.62 has no parallel. Third, Matthew 26.64 has συ ειπας (you [singular] said) and Luke 22.70 has υμεις λεγετε (you [plural] say) where Mark 14.62 has no parallel.
Who hit you?
Mark 14.65 lacks these five words. The usual reconstructions of Q lack a passion narrative, so recourse to the hypothetical document will not work for this agreement.
He wept bitterly.
I have no intention of listing each and every minor agreement of Matthew and Luke against Mark. Rather, I intend to categorize these agreements and handle them broadly.
In Greek, as sometimes in colloquial narrative English, a writer can use the present tense where the past would be expected. An example in English:
So he says to me....
Clearly the incident being narrated occurred in the past, yet the narrator uses the present so as to make the narrative more vivid. In Greek this phenomenon is called the historic present.
Quite a few of the minor agreements between Matthew and Luke against Mark consist of the former two using a verb in past tense or in participial form where the latter uses an historic present. It turns out that Mark tends to use the historic present far more often than Matthew and Luke put together. It is one of his more noticeable stylistic traits.
The historic presents on these lists come in the main from John S. Hawkins, Horae Synopticae, pages 144-149. His standard text was Westcott-Hort. By the numbers, he lists the following historic presents:
It is this distribution of the phenomenon that renders these particular agreements minor. For, if both Matthew and especially Luke desired to reduce the number of historic presents that they would have found in Mark on the Mark-Q theory, then it follows that they will occasionally, perhaps even frequently, have changed the same verb. Indeed, Luke has so avoided the historic present that it would be difficult for Matthew not to create a minor agreement by changing any given historic present in Mark.
This list presents only the 93 historic presents in Matthew. None of them corresponds with an historic present in Luke. The asterisked * instances are the 21 that correspond with historic presents in Mark. Refer to the Marcan table below for the corresponding words in Mark and Luke.
Hawkins notes that the following 15 instances occur in parables:
Matthew 13.28, 29, 44 (×3); 18.32; 20.6, 7 (×2), 8; 22.8, 12; 25.11, 19 (×2).
An amazing 68 out of the 93 historical presents (or ~73%) in Matthew are verbs which narrate speech (λεγει, λεγουσιν, φησιν).
This list presents the 151 historic presents in Mark, as well as the corresponding verbs, if any, in Matthew and Luke. The asterisked * instances in the Marcan column are those in which Matthew, 21 times, retains the historic present in Mark. A double asterisk ** marks the single instance in which Luke retains the historic present in Mark.
Hawkins notes that none of the Marcan instances occurs in a parable.
Some 72 out of the 151 historical presents (or ~48%) in Mark are verbs which narrate speech (λεγει, λεγουσιν, φησιν).
This list presents only the 11 historic presents in Luke. None of them corresponds with an historic present in Matthew. The single asterisked * instance is that which corresponds with an historic present in Mark.
Hawkins notes that the following 5 instances occur in parables:
Luke 13.8; 16.7, 23, 29; 19.22.
Luke 24.12 and 24.36, the last two instances of the Lucan 11, are textually doubtful. The first has Peter running to the tomb and looking in, and the historical present βλεπει is identical to that in John 20.5. The second has Jesus speaking peace to his disciples, and the historical present λεγει is identical to that in John 20.19. Both are examples of the shorter western readings.
Hawkins lists the following 13 instances of historical presents in the Acts:
Acts 8.36; 19.35; 22.2; 23.18; 25.5, 22, 24; 26.24 (φησιν).
The use of δε and και.
Matthew and Luke often agree against Mark in using the conjunction δε instead of the more Marcan και. I here list the 34 instances in which Matthew and Luke agree in using δε where Mark has used και, according to chapter 4 of Robert Stein, The Synoptic Problem: An Introduction. All references are from Mark:
Stein points out that Matthew and Luke simply like to use δε more than Mark, who in turn likes to use και more than Matthew and Luke. The distribution of δε and και Stein gives as follows:
Matthew and Luke, therefore, are very nearly equivalent in their use of και (once per 15.6 and 13.3 words, respectively), while Mark uses it more often (once per 10.2 words). Matthew and Luke are also very nearly equivalent in their use of δε (once per 37.3 and 35.4 words, respectively), while Mark uses it much less often (once per 68.9 words).
Granted the Matthean and Lucan stylistic preference for δε and the opposite Marcan stylistic preference for και, it is to be expected that Matthew and Luke might at times both independently change και to δε.