Two Endings, One Received Gospel: Eusebius, Jerome, and Burgon’s Ledger of Mark 16

The Witnesses They Cannot Count — Part I: The Gospels the Church Received — Article 4

The longer ending of Mark cannot be defended by pretending that Eusebius and Jerome did not exist. They did exist. They knew of copies which ended Mark at 16:8. Their words belong in the case, and a defense which hides them deserves to fail.

But the reverse mistake is just as serious. Eusebius and Jerome did not write the sentence often placed in their mouths: “Mark 16:9–20 is a late, uninspired addition.” They describe a textual problem, report one way of disposing of it, and then provide a way to retain and harmonize the passage. Their testimony is substantial; it is not simple. It must be read as historical evidence rather than used as a slogan.

This is where Dean John William Burgon remains useful. He did not win every point by force of assertion, and his nineteenth-century catalogues must be checked against modern records. Yet he understood something many modern discussions forget: the ending of Mark is not a contest between two famous codices and a mass of later copies. It is a historical question involving Fathers, versions, lectionaries, scribal practice, local transmission, and the public reading of the Gospel in the churches.

The question is therefore not whether the evidence is mixed. It is. The question is whether the mixed evidence justifies removing twelve verses from the church’s received Gospel.

What Eusebius actually says

Eusebius’s discussion appears in his answer to Marinus, who asked how Mark 16:9 could be reconciled with Matthew’s account of the resurrection. The issue is grammatical and chronological. Mark says that Christ, “being risen early the first day of the week,” appeared first to Mary Magdalene. Matthew speaks of the events connected with the end of the Sabbath. How should the accounts be read together?

Eusebius gives two possible answers. The first is textual: someone who wishes to remove the difficulty may say that the disputed section is not found in every copy of Mark and that the more accurate copies end at “for they were afraid” in verse 8. The second answer does not dismiss the passage. It retains the words and resolves the apparent tension by punctuation and interpretation: “having risen” may be taken with the preceding thought, while “early on the first day of the week” governs the appearance to Mary Magdalene.¹

Both elements matter.

The first tells us that Eusebius knew an issue in the transmission of Mark’s ending. A confessional account must acknowledge it. It is no use claiming that no ancient Christian knew of copies ending at 16:8. Eusebius did. The surviving texts of Vaticanus and Sinaiticus show that such an ending was not a theoretical possibility only.

The second tells us why it is wrong to turn Eusebius into a simple witness against the longer ending. The question is introduced by a man who cites the passage. Eusebius then offers a solution which keeps the passage and harmonizes it with Matthew. The fact that a textual solution is possible does not prove that Eusebius personally rejected the verses. Nor does it prove that the churches generally regarded them as false. It shows that Christians confronted a textual and harmonistic problem and that Eusebius could discuss more than one way to answer it.

That is a more demanding conclusion than either party’s slogan. Eusebius cannot be erased. He cannot be made to say more than he says.

Jerome is a witness—but not a second census

Jerome’s Letter 120 to Hedibia is frequently cited as though it supplies an independent fifth-century count of Greek manuscripts. His answer closely parallels the Eusebian discussion. Jerome states the twofold solution: either one refuses the testimony of Mark because the final portion is lacking in most copies, “almost all the Greek codices” being without it, or one retains both Matthew and Mark and harmonizes the accounts. He then immediately gives the harmonizing reading of Mark 16:9.²

The statement is important. Jerome plainly knew the claim that many Greek copies lacked the longer ending. It cannot be dismissed because it inconveniences the received-text case. His words also tell us that the question endured across the centuries from Eusebius to Jerome.

Yet Jerome’s statement must not be counted twice as though he had gone through the Greek world independently taking a fresh numerical survey. His answer follows the Eusebian form closely enough that it belongs to the history of the same scholarly question. It is powerful evidence that the report was transmitted and known; it is not an arithmetical second witness which allows a modern writer to say, “Eusebius plus Jerome proves two separate majorities of Greek codices.”

More importantly, Jerome’s own answer does not end by excluding the text. He gives the reader a way to retain it. That does not settle authorship; it does settle the false claim that Jerome simply ruled it out of the Gospel.

This lesson has wide use. The Fathers must not be treated as counters which can be added together whenever they agree with our conclusion. Dependence, quotation, context, and purpose matter. When Jerome preserves Eusebius’s answer, he bears witness to an ongoing discussion. When Irenaeus independently attributes Mark 16:19 to the conclusion of Mark in the second century, he bears another kind of witness. The historian who wants a verdict must first learn how to tell those witnesses apart.

The historical problem created by the omissions

The omission of Mark 16:9–20 in Vaticanus and Sinaiticus remains the central difficulty. It should be stated plainly. These are important fourth-century Greek manuscripts, and both end the Gospel at verse 8. Eusebius’s discussion makes clear that they were not alone in representing a form of the text without the longer ending.

No confessional doctrine of preservation requires us to call that stream imaginary. It was real. The question is whether it preserves the authorial form of Mark or a mutilated form of the Gospel.

The modern eclectic answer commonly turns the question into a rule: because the two early codices agree, the longer ending must be removed. But agreement between two witnesses does not explain its own genealogy. If the longer ending was known in the second century as Mark’s conclusion, as Irenaeus expressly says, then its absence in some fourth-century copies needs explanation no less than its presence in the church’s wider tradition.

The two major accounts are familiar.

The first says that Mark originally ended at verse 8, and that someone later supplied verses 9–20. If this is so, the addition must have entered the church’s text early enough to be quoted as Mark’s conclusion by Irenaeus, accepted in ancient versions, read publicly in churches, and preserved across broad lines of transmission. The theory must also explain why the source of so consequential an addition is nowhere clearly identified by the ancient church as an addition.

The second says that Mark originally had the longer ending and that some copies came to lack it. This account must explain why the omission appears in witnesses of great age and why Eusebius could mention copies ending at verse 8. It cannot simply invoke “accident” and move on. It must give a plausible account of how a section at the end of a Gospel might be lost, omitted, or set aside in a particular line of copying and use.

The received-text position does not deny its burden. It argues that the second account better fits the whole historical record: early explicit patristic reception, widespread translation and church use, later Greek manuscript transmission, and the lack of a credible early account of a wholesale twelve-verse fabrication. The omissions are part of the history. They are not the whole history.

Burgon’s ledger: why no witness may be isolated

John William Burgon’s The Last Twelve Verses of the Gospel According to S. Mark was written against an emerging critical consensus which treated the question as effectively closed. Burgon’s rhetoric can be sharp, and some of his conclusions are phrased more absolutely than a modern historical essay should phrase them. Yet his method contains an enduring lesson: collect the evidence before announcing that it is settled.

He begins with the early Fathers, then turns to versions, manuscript testimony, notes and scholia, internal considerations, and lectionary use. His argument is not that every line of evidence is equally clear. It is that a theory which depends on two manuscripts cannot be accepted while it leaves the rest of the church’s evidence unexplained.³

That is precisely the right question. A manuscript may be ancient. A lectionary may be late in its surviving form. A Father may quote only a verse. A version may be difficult to date. None is therefore useless. Each has a different historical function.

The lectionaries are especially important because they record public reading rather than private copying. A lection containing Mark 16:9–20 does not, by itself, prove that Mark wrote the words. Liturgical use does not create inspiration. But a passage repeatedly appointed for Easter and Ascension shows that the churches did not handle it as a marginal speculation. They heard it as Gospel in worship. That fact belongs to the doctrine of preservation because God preserved Scripture not for a museum but for the public instruction of His people.

Burgon also repeatedly draws attention to Irenaeus. On this point his instinct is exactly right. A direct second-century attribution to Mark cannot be balanced away by merely saying that two fourth-century codices are earlier than most surviving manuscripts which include the passage. The chronological categories are different. Irenaeus’s witness reaches behind the surviving manuscripts because he testifies to a text already read in the second century.

Still, Burgon must not be invoked as a substitute for research. His lists should be verified in the modern manuscript catalogues; his explanations of individual omissions should be examined rather than copied; and his confidence about the universal lectionary should not be confused with a complete reconstruction of every local practice. To follow Burgon faithfully is not to repeat every sentence he wrote. It is to share his refusal to let a verdict be purchased by ignoring inconvenient evidence.

The church’s reading of the ending

There is a theological reason why the historical question matters. The longer ending does not stand alone as a detached proof of some favorite doctrine. It continues the Gospel’s own proclamation: the risen Christ rebukes unbelief, sends His disciples to preach to every creature, confirms their proclamation, ascends, and works with them as the word goes forth. The church read these verses in connection with the resurrection and the ascension because the verses themselves were read as the completion of Mark’s Gospel.

This does not authorize reckless practice. The church is not commanded to manufacture snakes, poison, or miracles in order to prove faith. Scripture must interpret Scripture, and narrative promises must be read in the context of the apostolic mission and the whole canon. But difficulties of interpretation are not evidence of spuriousness. Every part of Scripture requires faithful interpretation; no difficult passage is therefore disposable.

The more basic point is that Christ’s church did not discover the longer ending in the nineteenth century. The church had already read it, cited it, translated it, and appointed it for worship over many centuries. The modern editorial brackets are themselves a historical judgment about that inherited text. They do not arrive as a voice above history.

A rule which applies in both directions

The received-text advocate must allow Eusebius and Jerome to count as evidence of omission. The critical-text advocate must allow Irenaeus, the versions, the lectionaries, and the church’s public reception to count as evidence of inclusion. Neither may call only his own preferred kind of witness “real evidence.”

This is the recurring point of the series. A church father may bear witness to a physical copy now lost. He may bear witness to the public text of a church now gone. He may preserve a quotation whose source manuscript has perished. His testimony does not become infallible, but it also does not become unreal because it is not a Greek codex on display today.

Eusebius tells us that the longer ending was absent from some copies. Jerome preserves the same difficulty and gives a way to retain the text. Irenaeus tells us that Mark’s conclusion contained Mark 16:19. The ancient church’s public use shows that the ending was not a furtive interpolation known only to a sect. Burgon reminds us to place these witnesses in one ledger rather than arrange them in separate rooms.

That is not an evasion of the hard evidence. It is the refusal to call a partial history the whole history.

Part Two will turn from the major Gospel passages to readings in Acts, John, Luke, and the Lord’s Prayer. The same question will follow: when a passage is called late because current Greek support is thin, what do the church’s own witnesses say about the text she actually received?

Sources cited