Mark’s endings and the sources of the Gospels

A source guide to the manuscript evidence, the different resurrection narratives, and the proposed relationships among the Gospels.

Mark’s shorter ending announces that Jesus has been raised but stops before narrating an appearance. Additional endings circulated in antiquity, and ancient readers already knew that copies differed. These textual questions belong alongside another chronological fact: Paul’s surviving letters report resurrection and appearances before the canonical Gospels were written.

Understanding this history requires distinguishing the contents of surviving manuscripts, reconstructions of earlier texts, and interpretations of what their writers intended. This guide follows those distinctions through the evidence. It also explains why the proposed Q source belongs to a different category from documents that survive.

Mark’s ending

Mark was composed in Koine Greek, the common Greek of the period. There is no surviving Hebrew original behind the canonical Gospel’s abrupt ending.

The age and character of Sinaiticus and Vaticanus make them important witnesses. Describing them simply as the “most reliable” manuscripts can conceal the work of textual criticism: evidence must be evaluated for each disputed reading. Their agreement here is significant without settling every question about how Mark first ended.

The Sinaiticus manuscript page shows Mark ending and Luke beginning. The critical Greek edition distinguishes the additional endings. See the University of Birmingham news article on the Sinaiticus ending for the unresolved question of intentional versus lost ending, and Peter Head’s manuscript study for the surviving Greek evidence.

The longer ending is secondary but ancient

Mark 16:9–20 is widely judged a later addition to the Gospel. Its precise date, author, and use of earlier sources remain matters of reconstruction. The evidence for its secondary character does not identify a particular copyist who composed or inserted it.

The secure chronological point is that Irenaeus quotes Mark 16:19 as part of Mark’s conclusion around 180 CE. The material therefore circulated well before the fourth-century manuscripts Sinaiticus and Vaticanus. Its early attestation and its secondary character can both be true. Irenaeus, Against Heresies 3.10.5.

The ending resembles traditions found elsewhere: Mary Magdalene, two travelers, the Eleven, commissioning, and ascension. Its vocabulary, transitions, and reintroduction of Mary contribute to the case for a different hand. A short passage’s count of otherwise unused words is not a self-sufficient proof. James Kelhoffer argues for composition in the first half of the second century and use of other Gospel material. This is a reasoned reconstruction rather than an event whose circumstances are directly recorded. Kelhoffer’s account of his research.

The testimony of Eusebius and Jerome

Eusebius, To Marinus 1. The discussion offers two ways of resolving an apparent discrepancy in the Gospel accounts. One would set aside the disputed passage, appealing to copies ending at 16:8, including copies described as accurate and nearly all copies. The other would retain it and place a pause after “having risen,” so that the subsequent time expression refers to Jesus’ appearance. These are alternative solutions. Their presentation does not amount to an unconditional declaration that the passage must be rejected, or to a modern census of all early manuscripts. Greek text and English translation.

The timing question. Matthew 28:1 explicitly dates the women’s visit, not the instant of resurrection. “Late on the Sabbath” reflects an ancient interpretation of its wording; many translations render it “after the Sabbath.” The difficulty Eusebius discusses therefore includes an interpretation of Matthew’s chronology. Matthew does not directly state that Jesus rose on Saturday evening. Matthew 28:1–6.

Jerome, Letter 120.3 to Hedibia. Jerome repeats both alternatives, including the statement about almost all Greek copies lacking the passage. His account depends closely on Eusebius’s discussion and is not demonstrably an independent survey of manuscripts. As Kelhoffer explains, Jerome may have known copies ending at 16:8, but the letter neither establishes nor denies that independent knowledge.

The Vulgate includes the longer ending, whose circulation and recognition predate Jerome. Textual variation at Mark’s conclusion was therefore known in antiquity; it was not first discovered by modern critics. Kelhoffer, pp. 99–100 — university publication record.

The freely accessible English rendering of Jerome’s letter is identified as a rough, machine-assisted translation from French. It provides a route into the letter, but exact quotations require a scholarly edition or the translated passages in Kelhoffer’s study.

Reading Kelhoffer’s study

James A. Kelhoffer, “The Witness of Eusebius’ ad Marinum and Other Christian Writings to Text-Critical Debates concerning the Original Conclusion to Mark’s Gospel,” Zeitschrift für die neutestamentliche Wissenschaft 92, nos. 1–2 (2001): 78–112. Uppsala University publication listing; publisher DOI: 10.1515/zntw.92.1-2.78.

The passage used here is on printed pages 99–100 (pages 23–24 of the university PDF, which includes a cover). Kelhoffer examines Jerome’s use of Eusebius. Jerome’s close dependence means his letter cannot be treated as a demonstrated independent survey. The letter also does not exclude his having independent knowledge of manuscripts ending at Mark 16:8. This is why the essay uses the qualified wording “not demonstrably an independent survey.” This reading note summarizes the relevant argument; it is not the full article.

The university listing opens as a normal webpage. Under “All publications,” look for the title beginning “The Witness of Eusebius.” The listing includes the citation and download option. The linked DiVA PDF has been intermittently unavailable, so the essay links to this university listing rather than opening that download directly.

What the manuscripts preserve

Sources: Sinaiticus project history, holding institutions, Head’s study, Vatican catalog, discussion of Vaticanus’s blank column and additional endings, Regius catalog, Smithsonian’s Washingtonianus record.

Comparing the Gospel narratives

The Gospels give different narrative forms to resurrection proclamation. Comparing their actual scenes makes those differences clearer than arranging them along a simple progression from fear to certainty.

• Matthew retains fear and doubt. The women leave with fear and great joy (28:8), and doubt or hesitation remains at the Galilean encounter (28:17). Matthew includes appearances and commissioning, but does not simply remove Mark’s unsettling elements. Matthew 28.
• Luke’s Emmaus encounter is the first appearance narrated as a scene. By the travelers’ return, an appearance to Simon is also reported (24:34). Peter’s visit to the tomb and Jesus’ later invitation to touch, together with eating fish, belong to Luke’s emphasis on embodied continuity. Jerusalem and its vicinity are the narrative focus; the ascension scene takes them toward Bethany. Luke 24.
• John includes personal and group encounters. The scene with Mary is intimate, but John 20 also narrates appearances to the assembled disciples. Jesus invites Thomas to touch; the text does not say that Thomas actually does. The blessing of those who believe without seeing is explicit. John 20.
• John’s giving of the Spirit and Acts’ Pentecost are distinct presentations. John 20:22 depicts Jesus breathing on the disciples on the evening of the resurrection day. Acts 2 places its public outpouring at Pentecost. Their theological relationship requires interpretation; calling one a “Johannine Pentecost” is a scholarly label. Acts 1–2.
• John 21 links Peter’s love to care. The repeated questions plausibly recall his denials, while Jesus’ repeated charge is to feed or tend his flock. Describing this as rehabilitation is an interpretation well grounded in the narrative. The chapter follows what looks like a conclusion at 20:30–31. Its compositional history remains debated; the transition does not identify a particular editor or establish exactly when the chapter took shape. John 21.

These differences do not establish that Matthew and Luke added appearances specifically because Mark’s ending troubled their communities. That explanation assigns a motive the surviving evidence does not document and leaves aside other traditions they may have used. Appearances already belong to the tradition Paul transmits in 1 Corinthians 15, before any of the canonical Gospels was composed.

Markan priority, Q, and competing explanations

Markan priority is the widely held view that Mark preceded Matthew and Luke and was used by them. It is not identical to the Two-Source hypothesis.

Under the Two-Source hypothesis, Matthew and Luke independently used Mark and another source conventionally called Q. Their independence is part of this proposed explanation, rather than a separately established fact that rules out their use of one another. Q accounts for material shared by Matthew and Luke but absent from Mark.

The Farrer hypothesis also accepts Markan priority but proposes that Luke used Matthew, making Q unnecessary. Other models exist. The competing explanations differ in how they account for shared wording, narrative order, and editorial decisions. Their continued discussion does not mean that every proposed model has equal scholarly support. Goodacre’s discussion of competing models; Hägerland’s defense of Q against an argument for direct dependence. These links provide abstracts rather than unrestricted access to the full chapters/articles.

No independent manuscript of Q has been recovered. Its wording, extent, date, layers, and community are reconstructed. A date in the 50s is a proposal, not an established point on a documentary timeline. Standard reconstructions do not contain a passion-and-resurrection narrative, but Q is not literally only isolated sayings: reconstructed material includes narrative settings and episodes.

An inferred collection lacking a resurrection story cannot establish that its users lacked resurrection belief. A sayings collection need not state everything a community believed, and a reconstruction cannot recover everything a hypothetical source might have contained. James Robinson on reconstructing Q.

An approximate chronology

Accessible scholarly introductions: Mark, Matthew, Luke, John. For the ongoing question of John’s relationship to Mark, see Wendy North’s research summary.

Paul gives no continuous biography of Jesus or account of discovering an empty tomb. Nevertheless, he does preserve a supper scene (1 Corinthians 11:23–26), attributes instructions to the Lord (7:10–11; 9:14), and refers to Jesus’ ancestry and birth within Jewish life (Romans 1:3; Galatians 4:4). His letters therefore preserve material about Jesus’ earthly life even though they are not biographies. Their purposes and occasions help explain what Paul chooses to recount. 1 Corinthians 11; 1 Corinthians 7.

What this evidence means for the inquiry

The shorter ending of Mark announces that Jesus has been raised but stops before narrating an appearance. Additional endings supplied further closure, and ancient readers already knew that copies differed. Whether Mark’s abrupt conclusion was intentional remains debated. The surviving material permits us to identify variation more confidently than to recover every decision that produced it.

Paul’s earlier letters already transmit resurrection and appearance traditions. The subsequent Gospels give those claims fuller and differing narrative forms. A hypothetical sayings source without a resurrection narrative cannot establish a period when the movement had no resurrection belief. Genre, selection, and the limits of reconstruction all matter to what can be inferred from an omission.

This history invites close attention to the sources and their differences. It does not establish that resurrection began as a suppressed account of ego dissolution. Second Humanity can explore transformation in the present while acknowledging that Paul also hoped for the resurrection of the dead—and required the community remembering Jesus to care for the people sharing its table.