[1] This paper was read at the S.N.T.S. Synoptic Problem seminar at Durham in August, 1979. I owe the seed thought of this paper to the chance reading of Salmon, G., Some Thoughts on the Textual Criticism of the N.T. (London, 1897), ch. 5.Google Scholar

[2] NTS 25.4 (07, 1979), 523–5.Google Scholar

[3] Mark, 10. 17, 18Google Scholar and Luke, 18. 18, 19 are identical in sense here, though Luke reads ποιήσας for ποιήσω ίνα and has the word order εīπεν δέ αύτ ό 'Iησοūς. The texts of Mark and Luke at this point are virtually solid. The only variants of importance are: in Mark ίδού τις πλούσως is added by A W fam 13 565 700 sa; in Luke καί τί ηρωτς περί τ άγαθο is added to τί μέ λέγεις άγαθόν by cur; and ό θεός is omitted by ℵ B.Google Scholar

[4] The text of Matthew is that given in the Aland Synopsis, the UBS text (third ed.) and that agreed for the Nestle twenty-sixth ed.

[5] Metzger, B. M., A Textual Commentary on the Greek N.T. (UBS, 1971), p. xxviii.Google Scholar

[6] The complexity of the situation is illustrated by differences between B and ℵ, which users of the UBS text or Nestle would never suspect. In verse 16 where B reads пολήσω ℵα σχώ σωήν αιώνιον, ℵ reads ποιήσας ξωήν αιώνιον κληρονομήσω.

[7] The curetonian and harclean margin are the sole exceptions omitting ‘good’ at this point.

[8] This and subsequent lists are from the UBS text emended from Mr Johnston's transcriptions. It needs to be remembered that fam 1 or fain 13 does not necessarily mean the support of the whole family. The sign is used for the reading judged probably to represent the archetype, which may in fact be found in only one MS. In this case 1/131 are on one side and 118/209 on the other.

[9] It is tempting to date them as early as the time of Clement of Alexandria (died 215), who certainly knew τι με èρωτάς περì τοū άγαθοū; and of Irenaeus (died 202), who certainly knew τí με λέγεις άλαθόν; But it would be misleading to suggest that either bears witness to these precise proto-Alexandrian or pre-Caesarean texts. The form of proto-Alexandrian given here is that of, ℵ and the pre-Caesarean comes from fam 1. The latter is almost exactly the text of 118 and 209, except that their unusual πς for εις has not been followed. έχω which is found in fam 1 and σχώ which is found in B D Θ and 700 have been given as alternatives. (It should be noted that the term pre-Caesarean has been used for convenience and does not beg the question of the objective reality of a Caesarean text of Matthew.)

[10] Metzger, B. M., A Textual Commentary, p. xxix.Google Scholar

[11] There were of course minor variants which arose through the occasional and solitary eccentricities of particular copyists, but these are irrelevant to the major problem. In Mr Johnston's 17 texts, all the material is found in the two forms, except that D has λÉγει for ειπεν, cur omits άγαθόν, inserts γάρ and adds 'Ιησοūς; harc. marg. omits με; D and 700 omit τοū it(e) adds pater. (A number of fathers also know variants of the proto-Alexandrian form which πατήρ ± μον ό Έν τοīς ούρανοīς(Marcion, Clement, Ps-Clement, Juvencus).) In suggesting a simple solution to the textual problem I am only indicating the general lines along which an answer may be found. I am not claiming either that it is correct in all its details or that all the complex influences have been identified.

[12] J. Carmignac suggested in discussion the possibility of ambiguity in the interpretation of a Hebrew text (ℶ┐⊃л 'ℶℸ) lying behind the Greek. This could have been taken adjectivally: ‘Good Master’ or with (Doupt) regarded as an interrogative particle: ‘Master, What Good?’ - though admittedly the latter gives awkward Hebrew.