Here is an extract regarding the controversial issue of dating Christian papyri:
In support of his contention that the widely accepted number of second-century Christian papyri is too high, Bagnall points to the slightly later dates of early papyri assigned by the great papyrologist/palaeographer Eric Turner, rightly observing that Turner’s expertise was unsurpassed. (It must be noted, however, that in general Turner’s dates differ by only a few decades, e.g., dating several items to the early/mid-third century instead of the late second century.) Bagnall also offers an argument from probability. Essentially, he contends that we should expect that the percentage of Christian papyri among extant second-century papyri should correlate with the likely percentage of Christians in the population of Egypt in that time. In the absence of hard data on either the population of second-century Egypt or the number of Egyptian Christians then, Bagnall adopts Rodney Stark’s “guesstimates” of the number of Christians in the early centuries. This leads Bagnall to propose that Christians comprised as much as 1 percent of the Egyptian population only by “the late 220s” (19). So, he reasons, Christian manuscripts from the second century should comprise no more than 1 percent of the total extant, or about one or two manuscripts. Consequently, he concludes, the widely accepted view that we have as many as eight second-century Christian manuscripts must be wrong.
I share Bagnall’s high regard for Turner and am also reluctant to base much on any dating of manuscript that conflicts with Turner’s judgments. But I am less swayed by Bagnall’s attempt to mount his argument from probability. It all seems to me too much guesswork to form the basis of anything compelling.
See Peter Head's evaluation of Bagnall's book here and here.