Over at Nick Norelli’s Rightly Dividing the Word of Truth, in the post Just Ordered…, he was describing how he’d purchased a copy of One Gospel From Two: Mark’s Use of Matthew and Luke, which I’d recommended here a few days ago. In the comments, I responded to some of the objections to Matthean Priority, which is a component of the Two Gospel Hypothesis presented in the above-mentioned book. One of the commenters to the post, John Poirier, brought up an interesting objection to Matthean Priority that I’d never heard before:
One problem with Matthean priority has to do with the patterns of agreement between Matthew’s quotations of the Old Testament, and Mark’s quotations. When Matthew quotes the OT in a Markan context, he quotes according to the LXX (as does Mark), but when he quotes the OT in a non-Markan context, he quotes according a non-aligned text (viz. proto-Aquila, kaige, whatever you want to call it). Supposing that he could know which contexts would be Markan when there as yet is no Mark simply defies the odds.
One may read my more superficial responses there, or continue reading here for the detailed examination of this idea.
Frankly, this conception is completely wrong. There is no correlation between Matthew and Mark and the LXX. Below I list all the instances in which a Markan quotation of the Old Testament is paralleled in Matthew, which also occasionally entails a parallel in Luke. I have not included Johannine parallels, as they are irrelevant to the Synoptic Problem and the specific “problem” in view. I used the (partly incorrect) list of quotations at the back of the UBS GNT4, along with the Septuagint Editio Altera (Rahlfs and Hanhart) and the Kurt Aland Synopsis of the Four Gospels, in order to get this done quickly. The passages are ordered according to their appearance in Mark. My notes are cursory. To understand what they mean, you’ll really have to be reading the texts involved. Here are the results:
1.) Malachi 3.1 in Matthew 11.10, Mark 1.2, Luke 7.27: None are LXX. In Mt, the first phrase matches LXX: “Ιδου εγω αποστελλω τον αγγελον μου.” Both Lk and Mk omit the εγω. The continuation of this allusion is as follows: Mt and Lk: προ προσωπον σου ος κατασκευασει την οδον σου εμπροσθεν σου. Mk: προ προσωπον σου ος κατασκευασει την οδον σου. Between LXX and these allusions lie a difference in pronoun in this second phrase, with LXX having μου, not σου.
2.) Isaiah 40.3 in Matthew 3.3, Mark 1.3, Luke 3.4: Almost LXX. In all three, the LXX τας τριβους του θεου ημων becomes τας τριβους αυτου. Mt and Mk end the allusion there. Luke continues through Isaiah 40.4 and into verse 5, with an almost identical text to the LXX, altering παντα τα σχολια to τα σχολια, and omitting the first phrase of verse 5, και οφθησεται η δοξα κυριου.
3.) Isaiah 6.9-10 in Matthew 13.13, Mark 4.12, Luke 8.10: Not LXX. The tripartite allusion in Mt (βλεποντες…ακουοντες…ουδε συνιουσιν) is shortened in Lk, with tenses adjusted, to a bipartite, balanced gnomic phrase through conflation of Mt’s 2nd and 3rd phrases. Mk follows Lk’s conflation of the 2nd and 3rd elements, and includes further allusions to the verse, but jumbled.
4.) Isaiah 29.13 in Matthew 15.8-9, Mark 7.6-7. Almost LXX. Mt’s ο λαος ουτος is LXX, not Mk’s ουτος ο λαος. LXX has τιμωσιν με, while Mt/Mk has με τιμα. LXX has διδασκοντες ενταλματα ανθρωπων και διδασκαλιας. Mt/Mk has διδασκοντες διδασκαλιας ενταλματα ανθρωπων.
5.) Exodus 20.12 in Matthew 15.4, Mark 7.10. Almost LXX. LXX has τιμα τον πατερα σου και την μητερα. Mt omits σου, while Mk inserts one after μητερα. (See #6 below for the importance of this.)
6.) Exodus 21.16 in Matthew 7.10 and Mark 7.10. Almost LXX. LXX has ο κακολογων πατερα αυτου η μητερα αυτου τελευτησει θανατω. Mt has ο κακολογων πατερα η μητερα θανατω τελευτατω. Mk is identical. Which is original? As Mt, in the Ex 20.12 quotation above, also displays the stripping of personal pronouns, while Mk inserts one, so Mt may be reckoned the source of this rendering of Ex 21.16 which is likewise stripped of its personal pronouns. Mk follows Mt in this, here.
7.) Genesis 1.27 (or 5.2) in Matthew 19.4, Mark 10.6. LXX. All read αρσεν και θηλυ εποιησεν αυτους.
8.) Genesis 2.24 in Matthew 19.5, Mark 10.7-8. Mk follows LXX, Mt is independent. Mt is not dependent upon Mk or LXX.
9.) Exodus 20.12-16 in Matthew 19.18-19, Mark 10.19, Luke 18.20. First part: Mt has LXX. Mk follows Lk’s forms in Mt’s order, with addition of μη αποστερησης. In Mt 19.19, τιμα…, again Mt lacks the personal pronoun, showing this lack as a characteristic of Mt. Lk supplies it and Mk follows Lk, both showing the LXX text. In Mt 19.19 is added και αγαπησεις τον πλησιον σου ως σεαυτον. Throughout, Mt follows LXX, Lk has an alteration thereof including the omission of και αγαπησεις…, which Mk follows, adding the anomalous (and non-Biblical) μη αποστερησης.
10.) Psalm 118.25-26 (LXX 117.25-26) in Matthew 21.9, Mark 11.9-10, Luke 19.38. Only ωσαννα comes from Ps 117.25, but this is not LXX. Both Mt and Mk follow LXX of 117.26. Lk has ερχομενος ο βασιλευς.
11.) Isaiah 56.7 in Matthew 21.13, Mark 11.17, Luke 19.46. All reflect LXX. Mt stops at κληθησεται, while Lk stops at the preceding προσευχης. Mk includes κληθησεται and continues the quotation from Isaiah: πασιν τοις εθνεσιν, a phrase of obvious interest to his Gentile audience. Lk is odd for omitting a verb here.
12.) Psalm 118.22-23 (LXX 117.22-23) in Matthew 21.42, Mark 12.10-11, Luke 20.17. LXX in all. Lk omits 117.23 though Mk includes it. (Note especially the conflation of Mt 21.46 and Lk 20.19 in Mk 12.12.)
13.) Deuteronomy 25.5 in Matthew 22.24, Mark 12.19, Luke 20.28. Not an LXX quotation at all, but a paraphrase. Again, however, Mk shows itself as dependent primarily on Lk, but adjusting to Mt. Note the expansion of Mt’s terse protaxis, which assumes the dead man to have had a brother, yet which for Gentile readers needs to be expanded to make that explicit.
14.) Exodus 3.6 (or 3.15) in Matthew 22.32, Mark 12.26, Luke 20.37. Almost LXX. Lk is a paraphrase. LXX lacks the definite article in all three positions before each θεος. Perhaps Mk does too, though this is textually questionable; in the UBS4 text, the article stands before the first, and is bracketed in the second and third position. Mt has definite article for each: ο θεος.
15.) Deuteronomy 6.4-5 in Matthew 22.37, Mark 12.29-30. Not LXX throughout, though Mk has that text for the first two items (καρδιας and ψυχης), and is consistent in using εξ, as does LXX. Mt uses εν throughout, and Lk uses εξ first, then switches to εν. Mt ends with διανοια rather than the expected LXX δυναμεως. Lk inserts a representation of the latter in the proper position as ισχυι, before Mt’s διανοια. Mk then switches the order of Lk’s last two items, as διανοια is better paired with ψυχης.
16.) Leviticus 19.18 in Matthew 22.39, Mark 12.31, Luke 10.27. Mt and Mk follow LXX. Lk removes the counting of first and second greatest commandments, and so elides the verb for this phrase, linking it to the former αγαπησεις with a simple και. Mk’s extended and repetitive agreement/reply is necessary to bring the dispute material to a close, which he does with the phrase about no one able to question Jesus anymore, wanting to get on with the story. Mt, however, has extended and important discourses following which Mk passes over, as these dealt primarily with halakhic matters, which are not of interest at all in Mk.
17.) Psalm 110.1 (LXX 109.1) in Matthew 22.44, Mark 12.36, Luke 20.43. Almost LXX. Lk has LXX υποποδιον, while Mt and Mk have υποκατω. All lack in this text the initial ο in the ειπεν ο κυριος of LXX.
18.) Daniel 7.13 in Matthew 24.30, Mark 13.26, Luke 21.27. This is only a paraphrase. However, the LXX (OG) phrase επι των νεφελων του ουρανου is in Mt. Lk has εν νεφελη, while Mk has εν νεφελαις. None is fully LXX, though Mt has a partial quotation.
19.) Zechariah 13.7 in Matthew 26.31, Mark 14.27. Not LXX. This is more of an allusion, with the imperative altered to a first person future indicative.
In the above, there are only three cases in which both Matthew and Mark and LXX reflect an identical text. The idea that Mk always follows LXX is simply wrong, as seen from the above. That Mt always follows LXX in a hypothetical use of Marcan quotations is therefore absurd. The patterns of utilization of these OT quotations may, however, demonstrate Mk’s use of Mt and Lk, as I touch on in the above notes in places. This investigation of quotations is something to look into further.