Why is this important? For whatever reason, early Christians chose to copy their Scriptural books, both Old Testament and New Testament books, into codices (with pages, like a modern book) rather than onto the more common format of a roll or scroll. There are only a few New Testament manuscripts that are written on the back side of used rolls (these are called opisthographs), a practice adopted probably when other writing materials were scarce or unaffordable. Of the forty or so pre-fourth century representations of one of the four Gospels, not one is copied onto an unused roll. Only one is an opisthograph. By comparison, of the ten non-canonical Gospel fragments, only five are in codex form. This means that, currently, in 50 percent of the cases, there is indeed something that distinguishes contemporary non-canonical Gospels from canonical ones: their basic physical form. (See Figure 2.)

Figure 2. Formats of 2^nd and 3^rd Century Gospel Manuscripts

Further, those five non-canonical Gospels that are left, that were copied into codices, tend to lack other physical or scribal characteristics that set almost all the New Testament manuscripts off from others. These characteristics include careful handwriting, as we saw above, but also standard codex size, and the use of readers’ aids like punctuation, paragraphing, and the like.²⁴ It comes down to one or maybe two that have even the same outward form and basic scribal characteristics comparable to the mainstream of canonical-Gospel productions. That is before we even talk about their contents, which, in most cases, is where the real differences begin!

This suggests either that the scribes who copied canonical and non-canonical Gospels either belonged to different socio-religious groups or scribal networks, or if they were in the same groups or networks, they were making conscious distinctions between Scriptural and non-Scriptural documents, even at the stage of copying.

Now, one of the most significant things about the evidence just cited is that all the papyri we have been talking about are dated to the period before the establishment of Christianity in the fourth century. That means they all come from a time before there could have been any suppression of competing Gospels by state-sponsored Christianity.

Another point of interest is that the papyri all come from Egypt. Why is this important? Egypt is consistently portrayed by historians of early Christianity as the hotbed of “alternative Christianities” in the early period. “[H]eterodoxy,” says Epp, “was the mark of the earliest Egyptian period.”²⁵ Ehrman concurs: “the earliest Christians in Egypt were various kinds of gnostic.”²⁶ Even if these statements are somewhat exaggerated, no historian would deny that the tide of theological diversity in the second and early third centuries was perhaps at its highest in Egypt. What that means is that if there is any place in the Mediterranean world where we should expect the “heterodox” Christian books to outnumber “orthodox” ones, it is in Egypt. The fact that heterodox Gospels do not even come close to doing so (even some of the non-canonical Gospels are probably not heterodox), that they are in fact currently outnumbered four to one, and that most of them have distinctively different physical properties from the canonical Gospels are, I think, ponderous problems for the political interpretation.

II. PRAXIS

While the “political” approach just described seems to have emerged as the dominant one both in academia and especially in the popular culture—so much so that it has contributed to our current cultural mythology—it is also fair to say that not all scholars of early Christianity are driven towards primarily political explanations. These scholars realize that churches were mostly busy with their own internal affairs: corporate gatherings for worship, instruction, and fellowship; mission and evangelism; inter-church relations; cultural engagement of various kinds, including, in many instances, trying to avoid harassment or martyrdom. Churches were not fixated on the effort to eliminate rival Christian groups. Even those scholars who favor political explanations usually give some attention to other factors leading to the construction of the Bible as we know it. Key among those factors was the alleged development of certain criteria of canonicity applied to the books available.

One of the most widely-published writers on the formation of the New Testament canon working today is Lee McDonald. McDonald reports that it is generally acknowledged among scholars “that the churches used several criteria, often unequally, in order to determine the contents of their New Testament.”²⁷ As these scholars see it, the churches needed “guidelines … to determine which books were to be included in their scripture collections and which were not.”²⁸ “The most common criteria employed in the process,” McDonald says, “include apostolicity, orthodoxy, antiquity, and use,” with the last one being the most determinative.²⁹ He offers the following explanations:

The writings eventually incorporated into the New Testament apparently met the worship and instructional needs of the churches, while the others did not. The writings that did not remain in the church’s sacred collections were those that did not meet the needs of the greater church and had more difficulty being adapted to the churches’ changing needs.³⁰

Ultimately, it appears that the writings that were accorded scriptural status were the ones that best conveyed the earliest Christian proclamation and that also best met the growing needs of local churches in the third and fourth centuries.³¹

[T]he key to understanding the preservation and canonization of the books which make up our current New Testament is probably usage, especially usage in the larger churches during the third through the fifth centuries.³²

This model tends to picture the churches as faced with a nearly undifferentiated mass of similarly-credentialed books, and applying to these books a number of tests: “is it apostolic? Is it orthodox? Is it ancient? Do we find it useful?” Compared to the political approach, especially in any of its more extreme, popular forms, the emphasis here might seem more reasonable, and more appealing. It is only common sense that the church should authorize—canonize—the books it found most useful, those that ministered to its membership in worship, preaching, catechizing, counseling, conversation. And it is undeniable that the early church also regarded its New Testament Scriptures as apostolic (either written by an apostle or by an apostolic assistant, like Luke or Mark), as orthodox (consistent with the witness of the rest of the Scriptures and the church’s creedal expressions), and as catholic (valid for and recognized in not just one localization of the church but across its broad expanse). Yet, there are also empirical problems with this way of conceiving of the canon as the result of the churches’ “praxis,” its practice of using a set of criteria for determining its authoritative Scriptures.

First, scholars who stress the role of criteria usually characterize what happened as a “selection process” and try to deduce what criteria churches used in that process. Yet we do not find in the ancient writers any lists of criteria, or any discussions of criteria for selecting books. That is, while Scriptural books are sometimes explicitly acknowledged to be (among other things) apostolic, orthodox, and catholic, and other books might not be, there is no evidence that the books used as Scripture by the church had first been subjected to a list of qualifying criteria before they were used and acknowledged as Scripture.³³ We simply find them being used as such. And if objections to such use arise, or in the case of a “newcomer,” if a book is put forward to be treated as Scripture, then points might be made in response about its having or not having certain qualities that Scriptural books manifest themselves to have. And for most of the books of the New Testament, there is no suggestion that any “tests” were applied at all.