105 Reconsidering the Text of Jude 5, 13, 15 and 18 // al K:Sms A. The external evidence is decisively on the side of the reading πάντα. It is a widespread second century reading, found in Egypt, Rome, Syria and Ethiopia. It has the support of the corrector of the earliest manuscript P72 and the two “best” manuscripts, codices B and 81. Most witnesses supporting the singular τούτο are of the A-text (Byzantine), though L is a later witness to the B-text (Alexandrian) and the Armenian text perhaps relates to the D-text but it could also be a witness to the A- text38. Its existence in a single Sahidic manuscript may be accidental. In any case, πάντα is more widespread and earlier and should be preferred on external grounds. The internal evidence also favours reading πάντα, because both read- ings πάντας and τούτο appear to be stylistic embellishments. Transcrip- tionally πάντας harmonises with εἰδότας39. Because P72 lacks ὑμᾶς, the change takes πάντας as the subject of εἰδότας with the text meaning “all of you know once and for all”. With πάντα the text reads “you know eve- rything once and for all”. Mayor points out that it is hardly expected that every reader knew the examples the author turns to in his argumentation. This probably makes πάντας contextually unsound. In contrast, there are statements about addressees knowing all things elsewhere in the NT (Rom. 5,14; 1 Joh. 2,20 v.l.; 27)40. The τούτο is common in Greek literature with a collocation of εἰδώς with the demonstrate pronoun plus ὅτι-clause that has an argumenta- tive context. The πάντα would be ambiguous in such context41, for which B.M. Metzger, The Text of the New Testament. Its Transmission, Corruption, and 38 Restoration. 3rd ed. (Oxford 1992) 82-83, notes that the question whether the Armenian text relates to the C-text (Caesarean) is still not settled. If it does, it is a distant relative to the D-text. J.M. Alexanian, “The Armenian Version of the New Testament”, in B.D. Ehr- man and M.W. Holmes (eds.), The Text of the New Testament in Contemporary Research. Essays on the Status Quaestionis (SD 46; Grand Rapids 1995), 157-72, divides the Arme- nian text into two subgroups. He states that the Arm 1 relates to the Old Syriac while its revision Arm 2 relates more to Latin Vulgate and the Byzantine Greek text. If so, it could be a distant relative to D-text. However, in this case it is also possible that the Armenian reading was taken directly from the A-text (Byzantine) source used in the revision, so it may not carry an independent value. Wikgren, “Some Problems in Jude 5”, 149; Kubo, P72 and Codex Vaticanus, 85-86, 39 141, argues for the “originality” of πάντας on the basis that the final σ could have eas- ily dropped out accidentally. He finds support in 1 Joh. 2,20 where many witnesses read πάντα instead of πάντες. Transcriptionally this is possible but the external evidence is too strongly against reading πάντας. Mayor, The Epistle of St. Jude, clxxxiv. Wasserman, The Epistle of Jude, 261, notes 40 that 1 Joh. 2 is particularly interesting because it also has opponents, who do not know the truth. Bauckham, Jude, 2 Peter, 48, argues that the apostolic faith the addressees had received at the time of conversion was complete and not in need of new information. Wasserman, The Epistle of Jude , 261; Wikgren, “Some Problems in Jude 5”, 149. 41