Bring up the Gospel of Thomas with a churchgoing relative and there’s a decent chance you’ll get it waved off in three words. “Those are forgeries.” Sometimes it’s “late forgeries,” said with the confidence of someone repeating a sermon.
I’ve heard it enough times that I wanted to take it apart properly. The phrase bundles two separate claims. One is about when these texts were written, and the other is about who claimed to write them. They have different answers, and one of those answers is a lot less comfortable for the person making the charge than they might expect.
How Anyone Dates an Ancient Book
Nobody has the original copy of any early Christian writing. That goes for the New Testament too. Scholars work backward from the copies that survive, and they lean on four kinds of evidence.
- The physical copies. The Nag Hammadi books were bound in the fourth century. We know because the binders stiffened the leather covers with scrap papyrus, and some of that scrap was dated receipts. The latest one, according to Bart Ehrman, is from 348.
- Older fragments. Bits of Greek papyrus sometimes turn up that preserve the same text from earlier copies.
- Outside mentions. If an ancient writer names a book, the book existed by then.
- The content. What a text argues about, and whose writings it seems to know, can place it in a period.
The first point is where the “late” charge usually goes wrong. A fourth-century copy tells you when that copy was made. It doesn’t tell you when the text was written, any more than a 2026 paperback of Plato tells you when Plato lived.
The Opponents Keep Receipts Too
Some of the best evidence for early dates comes from the people who wanted these books gone. Irenaeus, the bishop of Lyon, wrote his five-volume attack on the heretics around 180. In it he mentions that the Valentinians have a work called “The Gospel of Truth.” One of the Nag Hammadi tractates happens to open with those exact words, and most scholars think it’s the same book, which would mean the Gospel of Truth was already circulating before 180.
He also denounces a Gospel of Judas, and his short summary fits the Coptic text that resurfaced in the 1970s and was published in 2006. Marvin Meyer, introducing it in The Nag Hammadi Scriptures, puts its Greek original around the middle of the second century. My post on the Gospel of Judas gets into what it actually says.
The Gospel of Mary survives in Greek fragments from the early third century. Meyer’s volume reports that scholars are unanimous in dating it to the second century. The Gospel of Philip comes a little later, somewhere between the second half of the second century and the early third.
So most of these texts are second-century works. That’s later than Mark, which Ehrman’s timeline puts around 65, and later than John in the 90s. It’s also within a hundred years or so of the canonical gospels, which is a long way from the medieval inventions the word “late” tends to suggest.
Thomas Is the Hard Case
Thomas is where scholars really argue, and I’d rather show you the disagreement than pick a winner.
Three Greek fragments of Thomas turned up in the ancient rubbish heaps of Oxyrhynchus, a town in Egypt. They’re from the first half of the third century at the latest. Their first editors, Grenfell and Hunt, reasoned that the original must have been composed by around 140.
Meyer goes further. He notes that Thomas often gives Jesus’s parables as bare stories, without the allegorical explanations that Matthew, Mark, and Luke sometimes attach, which looks like an older stage of the tradition. His conclusion is bold:
“The textual evidence for an early date for the Gospel of Thomas thus may rival that of any of the New Testament gospels.”
Ehrman reads the same evidence more cautiously. He agrees that Thomas doesn’t seem to copy from the New Testament gospels, and that some of its sayings may go back to Jesus himself. But other sayings, in his view, presuppose a Gnostic myth that can’t be reliably dated before the second century. His best guess is that the book as we have it took shape in the first half of the second century.
Notice how close those positions are. Both scholars think Thomas holds very old material, and they’re arguing over whether a first collection existed in the first century. Neither one thinks it’s a late fake. For the sayings themselves, the 114 sayings post is a good place to start.
What Counts as a Forgery
Now for the second half of the charge. Ehrman has a handy way of sorting early Christian books into four kinds.
- Genuine books, actually written by the person named.
- Anonymous books, later attributed to someone famous.
- Books by someone who just happened to share a famous name.
- Pseudonymous books, which he’s blunt enough to call forged, because they claim an author who didn’t write them.
Run the Gnostic texts through that sorter and they scatter. The Gospel of Truth has no title in the manuscript at all, and it never claims an apostle. The Gospel of Judas, Meyer points out, is titled the “Gospel of Judas,” and not the “Gospel According to Judas,” so it doesn’t pretend Judas wrote it. The title of the Gospel of Philip appears only at the end, and Meyer’s volume suggests the Coptic translator may have added it.
Thomas does make a claim. Its first line reads:
“These are the hidden sayings that the living Jesus spoke and Judas Thomas the Twin recorded.”
By Ehrman’s definition, that puts Thomas in the fourth group. I don’t see any point in dodging this. He calls many of these texts forgeries without flinching, and he points out that forgery was condemned in the ancient world, even though it was hard to detect. Tertullian tells of a church elder in Asia Minor who was caught writing the Acts of Paul and confessed.
The Charge Cuts Both Ways
Here’s the part your relative probably hasn’t heard. Ehrman applies the same sorter to the New Testament.
The four canonical gospels go in the anonymous group. None of them names its author in the text, and the names Matthew, Mark, Luke, and John were attached later. Of the thirteen letters that carry Paul’s name, he says six are heavily disputed by scholars. He lists 2 Peter as a book forged in an apostle’s name, right alongside the Gospel of Peter.
Ehrman’s account of why anyone did this is the same for every side:
“They wanted their views heard, and they wanted their views to be accepted as authoritative, and so they wrote down their views in the names of apostles.”
I’m not saying this to score a point against the Bible. I love a lot of what’s in it, and plenty of this blog leans on it.
The honest conclusion is narrower. “Forgery” can’t be the line that separates the canon from the Nag Hammadi library, because by the strictest standard it runs straight through both. If a pseudonymous book can still be worth reading in your Bible, the same has to be possible for one buried in a jar.
What the Date Can’t Tell You
There’s one more thing worth saying. Even a confirmed first-century date wouldn’t make Thomas true, and a confirmed second-century date wouldn’t make it false.
What the dates do tell you is that these were living books for living communities. People read the Gospel of Truth aloud in the same decades Irenaeus was trying to settle which gospels counted. You’re looking at one side of an early argument about who Jesus was, and that’s valuable whether or not you end up agreeing with it.
Questions to Ask the Next Time You Hear It
When someone tells you a text is a late forgery, these questions usually sort out what they mean.
- Which text? Thomas, Truth, Judas, and Philip have different dates and different claims about authorship.
- Are they dating the manuscript or the composition? A fourth-century copy says nothing about when the text was written.
- Does the text claim an author at all? Several of these don’t.
- Who says so, and what’s their evidence? A specialist like Ehrman and a specialist like Meyer can read the same fragment differently, and both will show you why.
- Would the same test pass every book in the New Testament?
If you’re new to these texts and want a way in, I wrote up how to read a Gnostic text for the first time. This post belongs to the same series as Isn’t Gnosticism Elitist?, which takes on another objection you’ll hear.
I keep thinking about that receipt from 348. Somebody wrote it for a dull business reason, and later somebody else stuffed it into a book cover because papyrus was too useful to throw away. It’s the most ordinary document in the whole library, and it’s the one that keeps everybody honest about dates.
The Gospel of Thomas is quoted from Marvin Meyer’s translation in The Nag Hammadi Scriptures (HarperOne, 2010). The dating discussions for Thomas, the Gospel of Truth, the Gospel of Mary, the Gospel of Judas, and the Gospel of Philip come from the introductions in that volume, including Meyer’s comparison with the New Testament gospels. Bart Ehrman’s four categories of books, his dating of Thomas and the canonical gospels, the Nag Hammadi receipts, Tertullian’s report, and the quotation on why Christians forged documents come from the course guidebook for his Teaching Company lectures, Lost Christianities: Christian Scriptures and the Battles over Authentication (2002).
