^{page 273 note 1} Cf. Streeter, B. H., The Four Gospels (references to 4th (revised) impression 1930). Streeter's case has been strongly criticized by Abbot Christopher Butler in his The Originality of St Matthew (1951), pp. 62 ffGoogle Scholar. Mr Styler, G. M. discusses Butler's conclusions in an appendix to Professor C. F. D. Moule's The Birth of the New Testament (1962), Excursus iv, p. 223Google Scholar. Styler points out that even if Butler's case against the logic of Streeter's arguments is admitted, nevertheless a comparison of the style of Mark with that of Matthew and Luke still provides evidence for the priority of Mark. In effect this means that the whole case for the priority of Mark can now be seen to rest upon the kind of evidence produced by Sir Hawkins, John in Horae Synopticae (2nd ed. 1909, pp. 114 ff.).Google Scholar

^{page 274 note 1} DrFarrer, A. M., ‘On Dispensing with Q’, in Studies in the Gospels, ed. Nineham, D. E. (1955).Google Scholar

^{page 274 note 2} This phrase is used simply to describe material common to both Matthew and Luke, but not present in Mark. Its use is not of course meant to imply that there ever was a document of the kind that is sometimes labelled ‘Q’.Google Scholar

^{page 274 note 3} It was Butler, Abbot (op. cit. p. 109) who pointed out that between the publication of Oxford Studies in the Synoptic Problem and The Four Gospels, Streeter completely reversed his opinion on this matter.Google Scholar

^{page 274 note 4} ‘Parallel conflation’: this phrase is used to describe the theory that St Matthew and St Luke made use of a tradition parallel to that reproduced in Mark, and independently conflated the two accounts.Google Scholar

^{page 275 note 1} ‘Major agreements’: the inclusion of Q, material alongside certain of the minor agreements gives them an importance which distinguishes them from the other agreements against Mark.Google Scholar

^{page 275 note 2} For the moment we wish to leave open the question of whether that editor was identical with St Matthew or not. For the purpose of discussion it is convenient to assume that this was the case, but the question will be considered at greater length later on.Google Scholar

^{page 276 note 1} Since Luke here, as almost always, is further away from Mark than Matthew is, it follows that if we have to choose between Matthew and Luke in order to reconstruct the text of Q, Luke will most probably be nearer to Q and further from Mark. For the purpose of comparison therefore it is convenient to start from Luke rather than from Matthew.Google Scholar

^{page 278 note 1} Both Matthew and Luke also give the quotation a new conclusion, έμΠροσθέν σου.Google Scholar

^{page 278 note 2} Abbot Butler points this out (op. cit. p. 111): it was first noticed by Streeter himself (in Oxford Studies).Google Scholar

^{page 279 note 1} The Lucan baptism narrative is at first sight so different from that of Mark that it may be doubted whether St Luke could possibly have been using Mark as a main source at this point. However, the difficulty disappears when we see that St Luke has his own particular view of the baptism which leads him to modify the text of Mark. He saves the reference to John's baptism of the people (Mark i. 5) to this point in order to solve the difficulty presented by John's baptism of Jesus, which is thus included with that of ‘all the people’ (cf. Creed, J. M., St Luke, p. 57). This enables him to place the account of Jesus' baptism, together with a conventional reference to Jesus' praying, within a genitive absolute—virtually within brackets. The fact that the Western text of Luke iii. 22 assimilates the voice at the baptism to the text of Psalm ii. 7 demonstrates the tendency of St Luke (or even of later Christian scribes) to correct Mark's quotations. There is nothing here which requires us to suppose that St Luke is using a parallel source. It is significant that while St Luke uses none of the additional material introduced here by Matthew, he does make use of two of Matthew's modifications of Mark.Google Scholar

^{page 279 note 2} Streeter, , Four Gospels, p. 320, claimed that νομκός was interpolated into Matthew from Luke. Luke, however, normally uses the word in the plural in a rather formal manner (cf. Luke vii. 30; xi. 45, 46, 52; xiv. 3).Google Scholar

^{page 280 note 1} Krister Stendahl (The School of St Matthew, Uppsala, 1954, pp. 73 ff.) gives all the relevant textual evidence for this passage. He believes that Matthew is dependent on Mark, but does not draw any further conclusion in regard to Luke. Stendahl also draws attention to the fact that καρδα and δάνοα are translation variants (p. 75, n. 5). He denies that Mark xii. 30 has been influenced by II Kings, as suggested by McNeile (Comm. Mt. in loc.), on the grounds that it is very unlikely that a liturgical text would be subject to assimilation in this way. It may be, however, that Mark's form of the text indicates that its liturgical origins have been forgotten.Google Scholar

^{page 281 note 1} Direct borrowings from Mark are underlined with a single line, those from Matthew with a double one. Phrases borrowed from either of these sources which are reproduced in a modified form, are indicated by broken lines, single (Mark), or double (Matthew). It is also worth noting that έρημούν occurs in the New Testament on three other occasions only, all in the Apocalypse.Google Scholar

^{page 281 note 2} Streeter regarded Matt. ix. 34 as a ‘Western non-interpolation’ (Four Gospels, p. 170), and supposed that Matt. ix. 34 had been interpolated from Luke xiv. 15. The purpose of this interpolation remains obscure however, and we may well think that the scribe who was responsible for it employed a great deal of skill to very little purpose. For, once Matt. ix. 34 is deleted, there is little to connect this incident with the Beelzebub Controversy, and it is hard to believe that a scribe could have had any reason for making an interpolation of this kind. On the other hand it is possible to argue that St Matthew had a specific purpose in mind. Matt. xii. 22 and 23 is really a duplicate of the exorcism in Matt. ix. 32–4, and it is introduced in order to allow St Matthew to drop the unedifying story given in Mark iii. 20 and 21. Meanwhile Matt. ix. 34 provides a kind of remote preparation for the Beelzebub Controversy when it comes. This view of the matter is confirmed by the fact that St Matthew also introduces a reference to Beelzebub at x. 25.Google Scholar

^{page 282 note 1} Williams, C. S. C., Peake's Commentary (revised edn 1962), p. 749, section 654a, also denied the possibility of there being two editions of Mark, on the grounds that it is improbable that both redactions would have lacked the original ending. This is not altogether convincing however: apart from the question as to whether there ever was an ‘original ending’ of Mark other than the one we have, it is difficult to see why deutero-Mark should not have been constructed from the same mutilated edition of Mark used by St Matthew, St Luke and ourselves. The connexion between the Q, material and the major agreements is a more significant point. If we assume that St Luke read Matthew however, this connexion is easily explained on the theory that St Luke earmarked certain passages of Matthew for his own use, and then simply inserted this material (some of which would have been originally Marcan—hence the major agreements), into the non-Marcan sections of his own gospel. This theory also explains those coincidences in order between the Qmaterial in Matthew and that in Luke noted by Dr Vincent Taylor (J.T.S. n.s. iv, 1953, 27 ff.), while it allows for those dissimilarities which have led Dr C. K. Barrett (Expository Times, 1943, vol. 54, p. 320) and others to doubt whether Q could be regarded as a single document.Google Scholar

^{page 283 note 1} ‘The Indispensability of Q’, Theology LIX (1956), 182 ff. An interesting attempt to justify the Q, hypothesis is to be found in a recent article by Mr G. F. Downing. (‘Towards the Rehabilitation of Q’, N.T.S. xi, no. 2, Jan. 1965, 169 ff.). Mr Downing emphasizes the importance of the passages we have considered (along with others), but assumes throughout his study that there is a real possibility that Q, may have contained large sections closely parallel to Mark. He therefore sees in St Luke's divergences from Mark a sign that he is following Q, and gives a number of instances where Matthew follows Mark, and Luke does not. He regards these divergences not as evidence that St Luke was rewriting his sources, but as an indication that he was following an alternative source: (p. 176) ‘ …Luke seems deliberately to ignore just the Marcan material that Matthew has seen fit to reproduce exactly…. He seems to be reproducing Matthew's material independently of Matthew's use of Mark.’ But these two statements seem to be at variance with each other. Mr Downing does not deny that Matthew makes some use of Mark even where he does not quote him exactly. The fact that Luke sometimes seems to avoid those passages of Mark which Matthew quotes exactly does not suggest that he is following Q, but that he ‘deliberately… (ignores) just the Marcan material that Matthew has seen fit to reproduce exactly’!Google Scholar