I am making this post free to everyone, so that, if you’re not a member of the blog, you can see what you’re missing. Every week I make five posts on everything connected to the New Testament and Christianity of the first four centuries. Members can read it all, for a small fee. Every penny of the fee goes to support worthy charities. So why not join?
Recently you mentioned that your early work involved analysing patristic citations of the New Testament. I believe it has been said that virtually all of the NT could be recreated from such mentions if the need ever arose. Do you believe that this would, indeed, be possible, please?
This is a very interesting question. I may need to unpack what it means before giving an answer.
As most of you know, we do not have the original text of any of the books of the New Testament, only copies made many years (centuries) later. We have over 5600 copies in the Greek language in which it was originally written, and these are our primary sources for reconstructing what the authors originally wrote. But there are two other kinds of evidence: early versions and patristic citations.
The versions are the translations of the New Testament into other ancient languages. The earliest translations were Latin, Syriac, and Coptic, probably all beginning to be made in the 2nd century or so, and there are a bunch of later ones (Ethiopic, Armenian, Georgian, etc.). There are enormous problems with using these to figure out what the original Greek authors wrote. The manuscripts in these languages are themselves centuries after the originals, even if the translations themselves were centuries earlier. And so we cannot know how these later manuscripts have been changed over the years of copying away from what the original translations said.
But in theory, if you can use, say, the Syriac manuscripts that survive to reconstruct the Syriac “original” version, then you would have some purchase on seeing what readings were in the *Greek* manuscripts available to the original Syriac translator, and that would tell you what some of the Greek manuscripts in second century Syria looked like. That’s extremely helpful.
The problem is even bigger, though, because some elements of Greek cannot be detected by means of translation. For example, there are lots of instances where two different Greek grammatical constructions of a sentence could theoretically be translated the *same* way in many of the versions. And so if you have two different grammatical ways of expressing a sentence for a verse in, say, John’s Gospels – -that is, some Greek manuscripts read one way, others another – and both readings could in theory be translated the *same* way in Latin, then the Latin manuscripts will not tell you which of the two ways was found in the original Latin translator. Isn’t textual criticism fun?
The other source of information for reconstructing the text of the New Testament are the “patristic” citations. “Patristic” refers to the writings of the church fathers. These writers often quote the New Testament, and when they do so, we can tell what form of the New Testament (verse by verse) the authors had in the manuscripts available to them in their time and place. This was the topic of my PhD Dissertation (The Gospel Text of Didymus the Blind), and it is an unusually complicated matter. But I don’t need to go into all the problems here; I’ve talked about them on the blog before, if you want to pursue it: https://ehrmanblog.org/patristic-evidence-for-the-new-testament/
The question that I’m being asked here is whether it is true that even if we didn’t have any manuscripts or versions, we could reconstruct virtually the entire New Testament from these quotations of the church fathers, the “Patristic citations.” That is a claim you often hear from conservative Christian apologists who want to assure their readers/hearers that there is no doubt at all about what was in the original New Testament. And it certainly sounds impressive!
But I’m afraid the idea is rather naïve, and is almost always made by people who have no clue what is involved in undertaking the task.
As I’ve said, this was my original area of scholarly expertise – I spend many years working on it. In addition to my first book on Didymus, mentioned above (a published form of my dissertation), I did a second book, co-authored with Mike Holmes and Gordon Fee (well: co-researched with them; I actually wrote the book; but I couldn’t have done it without them – they did tons of the work with me), on the Gospel of John in the writings of Origen. In that book we tracked down and listed every quotation of the Gospel of John in the voluminous writings of Origen, the significant third-century theologian. From these quotations we reconstructed (and printed, in Greek) a huge chunk of the entire Gospel of John.
But here are insurmountable problems with saying that we could reconstruct the entire New Testament just from the Patristic citations:
- I should stress that Origen himself is highly exceptional. He quoted lots and lots of the NT and we have tons of writings from him. So yes, we could do what we did with his quotations of John.
- But we could do this because we ourselves have a Greek Gospel of John that we can compare Origen’s writings to. In other words, if we didn’t have a Greek text of John before us, in many places we would’t know that what Origien was writing actually was a quotation of John. Church fathers usually don’t say things like, “As is found in the Gospel of John”; they say something like “As we know from Scripture” or “As the Lord once said” or they just quote something without even telling us they’re quoting it.
I’m not sure if I’m explaining the problem well. But if you read a newspaper article that says, something like “As we all know, you must be born again to inherit the kingdom above” – there would be nothing in the sentence to make you think, “Oh, this author is quoting John 3:3.”
To use the quotation to see if it is an accurate quotation of John 3:3, you have to actually have a copy of the Bible with John 3:3 in it. Without that Bible, you can’t reconstruct that verse of the Bible. See what I mean? You wouldn’t know it was John 3:3 – or even from the Gospel of John, or even from the Bible.
- Relatedly, the church fathers never cited passages by chapter and verse, because they didn’t have chapters and verses. So if you have a church father like Tertullian, say, who quotes the Gospels a lot, you would certainly have his quotations, and maybe most of the time (?) you could figure out they were quotations of the Gospels, but you usually would have no way of knowing how the quotations were to be arranged, in what sequence, from beginning to end.
Let me illustrate the problem. Imagine you decided to cut up a Charles Dickens novel with scissors, cutting out whole sentences sometimes (never more than two or three at once), but far more often just clauses or phrases. You then shuffle together the tens of thousands (hundreds of thousands?) of scraps you have, and then you try to figure out how to arrange them to create the novel. You could do it if you already *have* the novel to compare the scraps to, but you couldn’t do it (or at least know that you had done it right) if you did *not* have the novel. So without the novel, you couldn’t reconstruct the novel.
- And what if you didn’t cut these scraps from a single book, but from hundreds of different copies of the novel, and each of the copies you used was different from each other?
That’s what we have with the patristic citations. The church fathers all quote the passages of the New Testament in *different* ways – either because their own manuscripts of it differed from one another, or because they were quoting it from memory and got a few words wrong (as people always do; and the church fathers absolutely did – no question about that one! Origen himself would quote the *same* verse in many different ways!), or because they were adjusting the quotation to the context of what they were talking about. If you’re alert, you will have noticed I did that very thing when I quoted John 3:3 above. It is not actually an accurate citation.
But suppose that my quotation of John 3:3 was the *only* quotation of John 3:3 that survives, and we didn’t have any manuscript or printed text of John to compare my quotation to. You would not know the quotation was from John; you would not know it came in what is now in chapter 3. You would not know that it followed what is now verse 2 and came before what is now verse 4. And you would not know if that’s what the author originally wrote or not. The problems are enormous and, I’m afraid, insuperable.
And so, in short, in theory, yes, the church fathers do quote most of the New Testament. But could we reconstruct the New Testament from their writings? No, I’m afraid not.