A few days ago someone raised a question about the focus of my article on 'The Text of P46: Evidence of the Earliest "Commentary" on Romans?' (in New Testament Manuscripts: Their Texts and Their World [eds T.J. Kraus & T. Nicklas; TENT 2; Leiden: Brill, 2006] 180-206). In brief, I suggest that some of the secondary readings (i.e., non-original textual variants) now preserved in P46 and/or B and/or 1739, on the one hand, and also in D and/or F and/or G, on the other, originated as 1 or 2 word “comments” or “notes” inscribed in the margins of an early copy of Romans by someone reading/studying the text (these “notes” would have entered the textual tradition when a subsequent scribe, mistaking these notes for corrections, copied them into the text). If we think of “commentary” as an activity rather than a genre, these notes or comments could represent the earliest surviving evidence of “commentary” on the text of Romans. Two examples:
1) in Rom 15:25, in place of διακονων, one finds in P46 DFG latt διακονησαι. Whereas the present participle is unexpected and ambiguous, the variant gives the anticipated final sense.
2) in Rom 15:31, in place of διακονια, “ministry”—by which Paul in fact meant his collection for the saints in Jerusalem—one finds in B DFG δωροφορια, the “bringing of a gift”: an excellent one-word explanation of the vague “ministry”.
Zuntz’s view of the history of the Pauline corpus—specifically, that readings with both early Alexandrian and ‘Western’ support are not later ‘Western’ intrusions into the Alexandrian text (as per the UBS Editorial Committee) but ancient survivors from a time before the textual traditions went their separate ways—is the foundation upon which my suggestion builds.
Hope this is of some help,