David L. Dungan, Constantine’s Bible: Politics and the Making of the New Testament.
Fortress Press, 2007.
Reviewed by Robert M. Price
One of the many things Dan Brown got wrong in the supposedly historical background of his novel The Da Vinci Code was that the emperor Constantine chose, more or less ham-handedly, a set of books henceforth to count as Christian scripture. As often, there is the merest element of truth amid Brown’s distortions, and if one wants to know what it is in this case, one might read David L. Dungan’s Constantine’s Bible: Politics and the Making of the New Testament. (For the full story one ought to read this fascinating book along with David Trobisch’s The First Edition of the New Testament.) Dungan begins with the very helpful observation that “scripture” and “canon” do not mean the same thing; then he demonstrates how Eusebius’ famous classification of New Testament books was a discussion of scripture and not an attempt to define a canon of scripture, i.e., some official list of what is in and what is out. Dungan shows that Eusebius was merely doing what the legatees of all philosophical schools (or even librarians) did: organizing the writings ascribed to founders of the school, dividing them into authentic, debatable, and spurious, according to general opinion. There was no hint that some of the writings were evil or to be shunned, just zeal to safeguard the outlines of a particular tradition by distinguishing authentic texts from pseudepigrapha or misattribution, and by keeping accurate copies available. One also looked to heirs of the founders to try to maintain the original interpretation. It wasn’t that no one had freedom to reinterpret old traditions; one simply wanted to keep the historic originals extant and available.
Eusebius knew who among his predecessors and contemporaries regarded which writings as orthodox in content and apostolic in authorship, and on this basis he reckoned twenty books as genuine or acknowledged (our four gospels, Acts, thirteen or fourteen Pauline epistles, 1 Peter, and 1 John). Revelation seems formerly to have belonged in this category but had begun to lose its position because of an anti-Roman bias that ill-comported with Constantinian patriotism. The second group was the disputed: James, Jude, 2 Peter, 2 and 3 John (and, according to some, the Epistle to the Hebrews). These were held in good repute by very many, but some had second thoughts about their authenticity. Third came the spurious: nothing sinister, just not authentically apostolic (Shepherd of Hermas, Gospel according to the Hebrews, Acts of Paul, Apocalypse of Peter, Apostolic Constitutions, and Epistle of Barnabas. Some place Revelation here, deeming it the work of the heretic Cerinthus). This third group would appear to be constituted by the books that some Catholic bishops/scholars accepted, but which Eusebius and/or the majority rejected. The fourth group of texts were rejected as heretical forgeries pure and simple: Gospel of Peter, Gospel of Thomas, Infancy Gospel of Matthew, Acts of Andrew, Acts of John, of Thomas, of Peter (all ascribed, along with the Acts of Paul, to a Gnostic named Leucius). Naturally, not everybody rejected these books, as they continued to be used and copied for centuries. But they were unwelcome in the “catholic, orthodox” circles, the only ones whose votes mattered for Eusebius and Constantine.
Let me pose a modern analogy for Eusebius’s four categories: the four color voting system of the Jesus Seminar. The Seminar Fellows discus the issues and then vote a gospel passage red if they feel very sure Jesus said it or did it. They vote it pink if it looks quite likely to be authentic but with a lingering grain of doubt. It is rated gray if the probability is low but the door is not shut completely, while the black vote denotes a spurious saying.
Dungan’s excellent point about “canon” not covering the same ground as “scripture” illuminates things right at this point: different sects may (informally) prefer and use certain writings and not others, ignoring the favorites of other sects. Enochic Judaism and Pharisaic Judaism had different books, as did Samaritan versus Pharisaic Jews. The Sukhavati Sutras were central to the Pure Land Buddhist sects, irrelevant to others. But none of these groups was trying to make a list of books which were alone were to be used by the faithful. That is where Constantine came in. As is well known, he inserted himself into the theological affairs of the church like a Papal Bull in a China shop. It was he who convened the Council of Nicea to endorse and enforce the Christological views of Athanasius. And it was he who, in the case of scripture, made the descriptive into the prescriptive.
I mentioned David Trobisch earlier. Trobisch showed to my satisfaction that the 27-book New Testament dated from long before these debates. Marcion had compiled the first New Testament, consisting of the Evangelicon (Ur-Lukas) and the Apostolicon (Galatians, Romans, 1 and 2 Corinthians, Laodiceans [Ur-Ephesians], 1 and 2 Thessalonians, Philippians, Colossians, and Philemon (?). The Pastorals had not yet been written (being rejoinders to him). Our 27 book New Testament was Polycarp’s counter-blast to the Marcionite Sputnik. Polycarp turns out to have been the Ecclesiastical Redactor of Luke and John, and the author of Acts, the Pastorals (and Hebrews?). This was slightly after the middle of the second century. All our manuscripts attest this selection, much earlier than Eusebius’ list or the Easter Encyclical of Athanasius in 367. It would appear, however, that this edition of the New Testament, while widely influential, was not binding on anybody but merely enjoyed great favor among the circles who copied it and used it in church. Some were less certain about some of the shorter works Polycarp had included, and that debate is enough to account for the second and third categories of Eusebius. Likewise, some thought Polycarp should have included a few more. It would have been such people who picked the contents of Codex Sinaiticus and Codex Vaticanus (adding Barnabas, Hermas, etc.), which would then have been pre-Constantinian, though some have suggested they were among the fifty shiny-new pulpit Bibles Constantine ordered up. But such flexibility abruptly ended as of Constantine’s intervention. From now on, Polycarp’s edition became the one and only Catholic, Orthodox New Testament. The monks of St. Pachomius had to burn or bury their other books, and they decided on the former.
So I would supplement Dungan’s excellent presentation with Trobisch’s. It is the latter who explains, as the former does not, what ancient agendas led to the specific contents of the New Testament.
My main quibbles with Dungan’s history of canonization occur in his discussion of Jewish and Islamic canonizations. Dungan avers that the Rabbis did not focus on the books of scripture, but on the oral traditions of Torah interpretation, the Mishnah. With other scholars, he dismisses the long-standing belief that there was a decision on the canon of scripture at the Yavneh Sanhedrin. Well, I am missing a piece here. It seems clear that there had been no one Jewish canon in the first century of the Common Era, as a glance at the table of contents of the Dead Sea Scrolls will reveal. But surely the Pharisees had a definitive list? Wasn’t one of the reasons they rejected the Septuagint that the latter contained certain books they considered apocryphal? Dungan dismisses Jospehus’ insistence that Judaism accepted only the 22 books (= the Protestant 39 book OT), commenting that we don’t know that Josephus is speaking for Judaism in general here. If not, however, what’s he doing? At least he seems to want us to think he is, right? And why say he was lying? He certainly shared the Rabbinic dogma of the cessation of prophecy, itself a kind of canon-closing maneuver. The Qumran sect, the Samaritans, and the Sadducees all perished with the century, and so did whatever scripture collections they revered, and wouldn’t that leave the Pharisees’ version as normative by default?
And when it comes to the canonization of the Koran, I wonder if Dungan is not conflating two different scripture collection projects. Tradition tells us that already in the caliphate of Abu-bekr there had been an initial gathering, charged to Zaid ibn-Thabit, of disparate fragments of revelation, pictured as individual but complete Surahs. These were only arranged by length, from longest to shortest, with no further work required. But the in the reign of Uthman came the text-standardization. Complaints arose that theological debates went deadlocked because, when one side appealed to the Koran, the verse did not read the same way in the other’s copy. To solve the problem, Uthman ordered the collection of all known copies. He had, again, Zaid collate all manuscript evidence and produce a critical text. Uthman directed that all the older texts be burned. Now all would be playing on the same game board. The trouble here is that both of the collection traditions appear to be fully as spurious as the old legends of the verbatim, inspired translations of the Septuagint (as well as, one might add, the modern scholarly legend of the Yavneh canonization). As for the collection of Abu-bekr, this seems an attempt to bridge the fiction of an oral origin of the Koran in the preaching of the Prophet and the written volume called the Koran, which, however, repeatedly refers to itself as “this book.” As Günter Lüling has shown, the Ur-Koran was an Arabic Christian hymnal.
As for the Uthmanic collection, as John Burton and others have argued, this collection is a fiction, too. It appears to have been an invention to allow various ulemma to claim a Koranic pedigree for this or that favorite (and spurious) hadith. If modern preachers, in a moment of enthusiasm, exclaim, “It ain’t in the Bible, but it oughta be!” these hadith-mongers were saying, “It ain’t in the Koran, but it used to be!” They claimed they had derived so-and-so reading from a family copy of a Koran predating the Uthmanic collation/standardization. It was a way of pretending that the standard Koran is not the real Koran. The real one used to contain this reading. Thus I question whether either the Jewish or Islamic parallels Dungan discusses are relevant, at least in the way he thinks they are.
Finally, I have to admit that the evidence of Irenaeus, apparently defending the outlines of Polycarp’s New Testament, appear already to be a definite move toward an official, exclusive canon. He does not want anybody using any but the four gospels. He marshals every cockamamie rationalization he can think of to clinch it that there must be four and only four gospels: Matthew, Mark, Luke, and John. No Gospel of Truth, etc. To me that sounds an awful lot like the intolerance for scriptural pluralism that Dungan doesn’t expect till Constantine some centuries later.