005_Tam_co_570-588 13/02/15 12:48 Pagina 581 WHEN PAPYRI AND CODICES SPEAK: REVISITING JOHN 2,23-25 581 From the above table, two conclusions can be drawn. First of all, there is always a relative difference in the degree of breaks between 2,22-23 and 2,25-3,1. There is not a single manu- script showing that the breaks used in these two locations are com- pletely equal. This means that 2,23-25 never existed as a discrete sense-unit comparable to the preceding or the following pericope; it is always a unit closely linked to either its preceding context or following context. Such delimitation appears to require the readers’ judgment that takes 2,23-25 as more than something merely transitional (by contrast, for the transitional statement in 2,12, the textual evidence from these selected manuscripts is apparently more ambiguous and divided). Secondly, with the exception of î66, which has a new paragraph at 2,23 and not 3,1, the rest of the earliest available textual witnesses (î75, a, A, B, and W) all show otherwise: 3,1 represents a new section and a major break, but not 2,23. It has been commented that î66 is copied by a “careless scribe” based on the “numerous errors and a large number of singular editorial changes” 47. The fact that the initial draft of î66 has not been copied very accurately may also suggest the same for its sense-unit delimitations. Since the paragraphs have to be set during the initial transcription, they cannot be easily altered afterwards, unlike one or two letters/words. At any rate, even putting aside this possible explanation for the exception of î66, still, five out of six available witnesses support the view that 2,23-25 does not be- long to the Nicodemus pericope starting at 3,1 48. 47 P.M. HEAD, “Scribal Behavior and Theological Tendencies in Sin- gular Readings in P. Bodmer II (P66)”, Textual Variation. Theological and Social Tendencies? Papers from the Fifth Birmingham Colloquium on the Textual Criticism of the New Testament (eds. D.C. PARKER – H.A.G. HOUGHTON) (Texts and Studies 6; Piscataway, NJ 2008) 60. But note that subsequent corrections made by the same scribe show that it is still rather carefully checked. See J.R. ROYSE, Scribal Habits in Early Greek New Testament Papyri (NTTSD 36; Leiden 2008) 498. 48 These textual witnesses reflect traditions they inherited and can be reasonably treated as discrete regarding sense-unit delimitation. In terms of the manuscripts’ content, only î75 and B have a close textual affinity. Yet, the papyri, in general, confirm “diversification of the text associated with the copying process in early manuscripts” (Chapa), whereas the early codices are gathered from “previously independent and isolated codices” (Elliott). C.L. PORTER, “Papyrus Bodmer XV (P75) and the Text of Codex