Son and Chosen. A Text-critical Study of John 1,34 105 to υ ς, not vice versa. Such a change clearly happens in ). This dem- onstrates a scribal desire to suppress such readings that could be used to support heterodox ideas (pace Haenchen). The transcriptional prob- ability is decisively against this variant as the “original”. This leaves us three contesters. Each of them includes κλεκτ ς. The variant readings (2) to (4) The variant reading κλεκτ ς το θεο is found in the D-text wit- nesses ()* b e ff2* syrs,c Ambrose Augustine1/4) supported by one B-text papyrus (P106vid) and five A-text minuscules (77 187 218 228 1784). It is a second century reading with diverse locales (Rome, Egypt, Syria). It is in disharmony with Johannes style and liable to heterodox ideas. The term κλεκτ ς appears also in other D-text (a vgmss syrpal-mss) and B-text (P75* copsa) witnesses47, albeit for different variants, so it is almost certain that one way or another such a term was found in the “original” text. The question is which one? Two rival readings exist. The reading υ ς κλεκτ ς το θεο is found in the D-text wit- nesses a bc ff2c vgmss syrpal-mss and in the B-text witness copsa. This variant is a late second / an early third century reading with roots in the second century. It is in disharmony with the usage of the Fourth Evangelist but not as much as the other two variants containing κλεκτ ς. It is geo- graphically widespread (Rome, Egypt, Syria) and less liable to heterodox ideas than the reading κλεκτ ς το θεο because of its sonship con- cept. The other variant is the singular reading υ ς κλεκτ ς found in P75*. It is a second century reading without a strict parallel in the Fourth Gospel or elsewhere. It is liable to heterodox ideas and this reading is lectio difficilior. If κλεκτ ς το θεο is the “original” reading, how does one explain υ ς κλεκτ ς? One might hypothesize that it is a mishap for υ ς κλεκτ ς το θεο , but as I showed, this is unlikely48. There are no good reasons why any scribe would change το θεο to υ ς and transpose the words. Transcriptionally such an alteration has to be intentional but it makes no sense. If, on the other hand, υ ς κλεκτ ς το θεο is the “original” reading, how does one explain the rise of variant readings 47 I assume in this study that copsa is in basic agreement with the B-text witnesses B), although there are also D-text variants in it. The work on textual affinities between Coptic and Greek witnesses is still largely undone. See F. Wisse, “The Coptic Version of the New Testament”, in B.D. Ehrman and M.W. Holmes (eds), Status Quaestionis, 131-41. 48 In case it is a mishap, the longest reading is probably a conflation supported partially by P75 and transcriptionally the variant reading κλεκτ ς το θεο would then most likely be the original.