107 Son and Chosen. A Text-critical Study of John 1,34 κα μεμαρτ ρηκα τι ο τ ς στιν υ ς κλεκτ ς in John 1,3449. This reading was lost by early second century scribal harmonisations and managed to survive only in P75* for us to see. Implications A few notes are in order. The reading I am arguing for poses a prob- lematic question for the text-critical method. Just how likely is it that a singular reading is the “original”? Kurt Aland would not consider it a real possibility. His text-critical “rule” number 7 states that singular read- ings are “original” only theoretically, not in practice, because “it will only confirm the view of the text which it presupposes”50. Ernest C. Colwell used singular readings as a basis for determining the level of corruption in the early papyri in order to establish scribal habits. He rejects singular readings as “original” in toto and states that a principle of disregarding the singular readings merits the most rigorous observance51. James R. Royse followed suite and argued that even though a singular reading may preserve the “original” variant, the amount of witnesses make such pres- ervation highly unlikely52. The reasons are that singular readings are the result of scribal additions, omissions, transpositions, alterations, spelling mistakes, hearing problems etc. Yet some are more cautious. Michael W. Holmes has stated that: … the mere occurrence of a reading does not give it equal status with all other readings; when a reading occurs only in a few late witnesses, for example, it must be demonstrated, not assumed, that it could be an ancient survivor rather than a scribal correction or emendation53. This admission leaves room for singular readings if it can be de- monstrated that other readings derive from it. J. Keith Elliott states more 49 It should be understood that this study does not deal with the other text-critical problem, whether the verse should read ρακα or ρακα. I follow the NA27 at the begin- ning of the verse for the sake of convenience. 50 Aland and Aland, The Text of the New Testament, 281. This is an interesting state- ment since there are a number of singular readings accepted as “original” in the NA27. For example, the editors of NA27 accept ε παν as original in Mark 16,8; John 4,52; 6,60; 9,22; 11,37 on the basis of D, D, D, ), ) alone, respectively. They accept λθαν as “original” in John 4,27 on the basis of B* alone. 51 Colwell, “Method in Evaluating Scribal Habits”, 107-24; idem, “Hort Redivivus: A Plea and a Program”, in idem, Studies in Methodology, 148-72. 52 Royse, Scribal Habits in Early Greek New Testament Papyri, 33-34. 53 M.W. Holmes, “The Case for Reasoned Eclecticism”, in D.A. Black (ed.), Rethinking, 82, n. 15.