Gospel of Thomas

Printed in 1897, Identified in 1952

On the second day of their first season digging the rubbish mounds of Oxyrhynchus, in January 1897, B. P. Grenfell and A. S. Hunt turned up a leaf carrying sayings of Jesus that were not in any Gospel. They rushed it into print the same year as a shilling pamphlet, Logia Iesou: Sayings of Our Lord from an Early Greek Papyrus, and dated the hand not much later than the year 200. Two more fragments of the same work appeared in the fourth volume of The Oxyrhynchus Papyri in 1904. Nobody knew what the work was. It was a collection of unplaced sayings with no title and no context, and it stayed that way for half a century.

The Coptic manuscript found at Nag Hammadi settled it. Working in the Coptic Museum in Cairo alongside G. Quispel in about 1952, H.-C. Puech recognised that the Oxyrhynchus fragments were pieces of the complete text in front of him; he published the identification in 1957. A. D. DeConick draws out why that mattered: it "meant that the Coptic was a translation from an earlier Greek version of the Gospel, and since one of the Greek papyri had been dated to 200 CE, this meant that there was manuscript evidence of this Gospel 150 years older than the Coptic."

The three Oxyrhynchus pieces are scattered. P.Oxy. 1 is a codex leaf at the Bodleian Library, Oxford; P.Oxy. 654 is a reused roll at the British Library, with an unpublished land survey on the back; P.Oxy. 655 is six small fragments of a two-column leaf at the Houghton Library, Harvard. They do not agree with the Coptic in detail, which is itself informative — R. Cameron judges the differences "best explained as variants resulting from the circulation of more than one Greek edition" in antiquity, and concludes that "the text of Gos. Thom. must be regarded as unstable."

A Book With No Story In It

S. J. Patterson describes what a reader actually meets. In contrast to the canonical gospels, he writes, this book "has no narrative. It does not tell a story of Jesus' life. It contains only a few brief anecdotal stories. In it there are no stories of Jesus' birth, baptism, his preaching at Nazareth. There are no stories of his entry into Jerusalem, his arrest, crucifixion, or resurrection." What there is instead is a list, most items beginning "Jesus said". It opens by promising secret sayings and by making the stakes of interpretation the whole point: whoever finds their meaning "will not experience death".

About half the sayings have parallels in the first three Gospels. The other half were unknown before 1945. Some are close to the canonical wording, some are recognisable but turned, and some are strange — the kingdom "spread out upon the earth", a two-word saying, "Become passers-by", and a closing exchange in which Simon Peter demands that Mary Magdalene — so she is generally identified, though the text gives only the first name — be sent away, and Jesus answers that he will lead her "in order to make her male". That last one is the book's most-quoted sentence and the one most often used as a cudgel in both directions. It is worth reading beside the earlier saying about making the two into one, which is how specialists on every side of the dating question generally take it; and it is worth knowing that DeConick, who argues for a very early core, assigns it to the collection's latest layer rather than to anything traceable back to Jesus.

There is no evidence this book was ever a candidate for the New Testament. Hippolytus of Rome, writing between 222 and 235, quotes a saying "occurring in the Gospel inscribed according to Thomas" and rebuts it by pointing out that it is really a line of Hippocrates. Eusebius puts books of this kind below even his rejected category. No ancient writer argues for its inclusion; no canon list contains it; no council is recorded as debating it. One honest caution, and it comes from a scholar in the early-dating camp: Cameron warns that bare title-only references in the fathers "must be treated with care, since they might refer to the Infancy Gospel of Thomas", a completely different work. Hippolytus is the one citation that quotes text we can match.

How Old Is It? The Field Is Genuinely Split

Everything turns on one question: does this book draw on the Synoptic Gospels, or does it draw on the same older stream they drew on? Cameron states the dependency between the two questions with unusual candour, and it explains why the argument never ends: "If Gos. Thom. is a sayings collection based on an autonomous tradition, and not a gospel harmony conflated from the NT, then a date of composition in, say, the last decades of the 1st century would be more likely than a mid-to-late-2d-century date." The date follows from the judgement about sources, not the other way round.

The case for independence is real and should be met at its strongest. Patterson gives three arguments: that where the same saying occurs in both, the form in this book is often the more primitive; that there is "between them virtually no shared order", which is not what literary borrowing usually looks like; and that its genre, the bare sayings collection, belongs to an early period "contemporaneous with Q". H. Koester went further, calling its material a "direct and almost unbroken continuation of Jesus' own teaching" — a sentence quoted here as DeConick prints it from his 1971 essay, not from a copy of Koester this article has opened. DeConick's position is more careful than either label allows: she reads the book as a "rolling corpus", an old core which, in her words, "appears to have contained five speeches of Jesus" whose contents "point to their origin in the Jerusalem mission prior to 50 CE" — onto which a Syrian community's later reinterpretations accumulated over generations — so that parts of it may be very early and parts of it certainly are not. She also rejects the "Gnostic gospel" framing outright, finding "no distinctive feature of Gnostic mentality" in the sayings.

The case for a second-century book that knows the Synoptics has gained substantial ground since 2012. M. Goodacre argues that the verbal agreement is sometimes too close to be anything but a literary relationship — including, in his own words, "a 13-word verbatim agreement (Mt. 7.5 // Lk. 6.42 // Gos. Thom. 26 …)" and "a phrase that is never found in Greek literature until it appears in the parallel between Lk. 17.21 and Gos. Thom. 3" — and that the decisive test is whether the redactional fingerprints of one text show up in the other, what he calls "diagnostic shards". His most intuitive argument is about coverage: this book has parallels to material from "every single strand of Synoptic material" — the double tradition, the triple tradition, and the matter peculiar to each of the three — so that if it did not know them, "the author was able to access material from every pool of tradition that fed them". S. Gathercole argued in parallel that the book was composed in Greek rather than translated from Aramaic or Syriac, which cuts against the reconstructions the early-dating case has leaned on; this article has not read Gathercole's volumes directly and reports his thesis at second hand, which is worth saying plainly.

Where does that leave a reader? Not with a settled answer. An early, independent book was the more common view among specialists through the 1990s and 2000s; since 2012 the case for a second-century book dependent on the Synoptics has gained ground far enough that the Society of Biblical Literature's own public reference work now presents a date of roughly AD 135 to 200. The early camp has not conceded, and Goodacre himself grants that he and Gathercole "may sometimes give the impression" that the relationship "is as straightforward as the lines that students draw" — adding, in the same breath, that "the maps we draw and the models we make are necessarily simpler than the reality that they are attempting to represent." This app's own position is that the four Gospels are the trustworthy record, and that is a theological commitment held before the evidence rather than a result read off it. What the evidence itself supports is narrower and worth having on its own terms: whichever date wins, this is a real second-century document that tells us a great deal about how some Christians in Syria remembered and reworked the words of Jesus — and, as the Bible Odyssey article puts it, "a later date for the Gospel of Thomas is not a negative judgment."

Scripture references

• Luke 17:20-21 — The kingdom not coming with observation — the closest canonical parallel to this book's central theme
• Matthew 7:5 — The beam and the speck: the thirteen-word verbatim agreement at the centre of the dependence argument
• John 20:30-31 — A Gospel that states its own purpose, and names it as belief rather than secret knowledge
• 1 Timothy 6:20-21 — "What is falsely called knowledge" — the apostolic warning this book is usually read against

Sources

• Holding institutionThe Oxyrhynchus Papyri, University of Oxford: Location-List — The Oxyrhynchus Papyri project, University of Oxford · The project's own tabular location list, read directly rather than through a summary. It is the source of all three shelfmarks and holdings: P.Oxy. 1 at the Bodleian as Ms. Gr. Th. e 7 (P), 654 at the British Library as Inv. 1531 (its key notes the transfer from the British Museum), and 655 at the Houghton Library, Harvard, as SM Inv. 4367Supports: Where each Greek fragment is and under what shelfmark
• ScholarshipA. D. DeConick, "The Gospel of Thomas" — April D. DeConick · The Expository Times 118/10 (2007), open access in Rice University's institutional repository and read in full. Source of the rolling-corpus model in her own words, of the Kernel dating and its Jerusalem setting, of her rejection of the Gnostic framing, of her placing of saying 114 among the accretions, and of the account of H.-C. Puech's identification in about 1952 and its significanceSupports: The identification of 1952, the rolling-corpus position, and the layering of the sayings
• ReferenceS. J. Patterson, "The Gospel of Thomas and Jesus" — Stephen J. Patterson · Dialogue: A Journal of Mormon Thought 28/2, open access and read in full; his three arguments for independence and his description of what the book is are quoted from it. His dating of roughly AD 70-80 is from his monograph The Gospel of Thomas and Jesus (Polebridge, 1993), 120, which this article has not read directlySupports: The case for independence, stated by one of its principal advocates
• ScholarshipM. Goodacre, "Did Thomas Know the Synoptic Gospels? A Response to Denzey Lewis, Kloppenborg and Patterson" — Mark Goodacre · Journal for the Study of the New Testament 36/3 (2014), 282-293, DOI 10.1177/0142064X14523523. Free PDF on the author's own site, downloaded and read in full; the diagnostic-shards definition, the thirteen-word agreement, the every-pool argument and his own concession are quoted from it. His book is Thomas and the Gospels: The Case for Thomas's Familiarity with the Synoptics (Eerdmans, 2012), published in Britain by SPCK as Thomas and the Gospels: The Making of an Apocryphal TextSupports: The case for second-century dependence, stated by its principal advocate
• ReferenceBible Odyssey (Society of Biblical Literature): When was the Gospel of Thomas written? — Simeon Burke, for the Society of Biblical Literature · Cited for the field's current public reference position — a date of about AD 135 to 200 — and for its own statement that the question "remains highly controversial". Also the source of the closing observation that a later date is not a negative judgmentSupports: The current state of the question, from a body with no stake in this app's position
• Primary textHippolytus, Refutation of All Heresies, book V (Ante-Nicene Fathers, vol. V) — Hippolytus of Rome, translated by A. Roberts and J. Donaldson · Public domain, hosted at Wikisource, read for the passage in question. The one patristic citation that quotes matchable text: Hippolytus reports a saying from "the Gospel inscribed according to Thomas" and answers it by attributing the thought to HippocratesSupports: The earliest datable external notice, and its hostility
• ReferenceR. Cameron, "Thomas, Gospel of", in The Anchor Bible Dictionary — Ron Cameron · Anchor Bible Dictionary, vol. 6, 535-540. This article has not opened the printed dictionary; every sentence quoted from Cameron here was read on the page linked, which reproduces the relevant passages with page references (dating at 536, attestation and the Hippolytus point at 535). Source of the judgement that the Greek and Coptic differences point to more than one Greek edition and that the text "must be regarded as unstable", of the statement that the date follows from the judgement about sources, and of his warning that title-only patristic references may mean the Infancy Gospel instead. H. Koester's sentence about unbroken continuation is NOT on this page; that one was checked in DeConick's article, which prints it from his 1971 essaySupports: The instability of the text, the logic of the dating argument, and the caution about patristic references
• EncyclopediaWikipedia: Gospel of Thomas

Further reading

• The Oxyrhynchus Papyri, University of Oxford: Location-List
• A. D. DeConick, "The Gospel of Thomas" (open access, Rice University)
• M. Goodacre, "Did Thomas Know the Synoptic Gospels?"
• Bible Odyssey (SBL): When was the Gospel of Thomas written?