Who wrote the Gospels, and when?

Four short books about one man are the most studied documents in the history of the world. Who wrote them, and when, matters because Christianity is a claim about history: if the Gospels are late legends written far from the events, the claim weakens; if they are early and rooted in eyewitness testimony, it stands on firmer ground.

This article sets out the evidence as it is, including where scholars sharply disagree. For the non-Christian evidence that Jesus existed at all, see Did Jesus really exist? in this library.

Two kinds of evidence

Historians use two kinds of evidence for the authorship of an ancient book.

Internal evidence is what the text itself shows: its language, its knowledge of places and customs, its interests, any claims it makes about its author.

External evidence is what other writers say about it: who quotes it, who names its author, and how early.

The Gospels are unusual in one respect: none of the authors gives his name in the text. Luke and John come closest — Luke writes in the first person in his preface (Luke 1:1-4), and John's Gospel ends with an editorial note identifying "the disciple whom Jesus loved" as its source (John 21:24). So external testimony carries more weight than it would for, say, Paul's letters.

The external testimony

Papias (c. AD 110–130)

Papias was bishop of Hierapolis in Asia Minor. His five-volume work on the sayings of the Lord is lost, but the church historian Eusebius quotes from it around AD 325 (Ecclesiastical History 3.39).

Papias says he preferred to learn from "a living and abiding voice" rather than from books, and that he questioned people who had known the disciples. He reports what "the elder" told him: that Mark, having become Peter's interpreter, wrote down accurately — though not in order — what he remembered of the things the Lord said and did. He also says that Matthew arranged the sayings (logia) in the Hebrew dialect, and each person interpreted them as best he could.

The Papias evidence is early — some scholars date him as early as c. AD 100 — and he stands only one link from the apostolic generation. But it raises puzzles. Our Greek Matthew does not read like a translation from Hebrew or Aramaic, so what did Papias mean by the logia? And who was "the elder John" whom he mentions alongside the apostles? Scholars are divided on both.

Irenaeus (c. AD 180)

Irenaeus, bishop of Lyon, had as a boy heard Polycarp of Smyrna, who in turn was said to have known the apostle John. In Against Heresies (3.1.1) he gives the clearest early statement: Matthew wrote among the Hebrews in their own language; Mark, Peter's disciple and interpreter, handed on in writing what Peter preached; Luke, Paul's companion, set down the gospel Paul preached; and John, the disciple who leaned on the Lord's breast, published his Gospel while living at Ephesus.

Irenaeus also insists that there are four Gospels — no more and no fewer. By his time the fourfold Gospel was settled across much of the church, from Gaul to Asia Minor.

The Muratorian fragment

A damaged Latin list of New Testament books, discovered in Milan in the eighteenth century. Its opening is lost, but it calls Luke's Gospel the third book, written by Luke the physician, Paul's companion, and John's the fourth. Most scholars date it to the late second century, because it describes the Shepherd of Hermas as written "very recently, in our times"; a minority argue for the fourth century.

The manuscripts

The titles themselves — euangelion kata Markon, "the gospel according to Mark", and so on — appear in our earliest manuscripts that preserve the beginnings or endings of the books. Papyrus 75, usually dated c. AD 175–225, has "according to Luke" and "according to John".

The anonymity question

Here the debate is real.

The sceptical case, made forcefully by Bart Ehrman, runs like this. The Gospels are anonymous in the text. Early Christian writers such as Clement of Rome, Ignatius and the Didache echo Gospel material without naming authors. Justin Martyr around AD 150 calls them "memoirs of the apostles" but still does not name them. The titles may therefore be second-century attributions. And Jesus' disciples were Aramaic-speaking Galileans, described as "uneducated" in Acts 4:13, unlikely to have composed sophisticated Greek prose.

The traditional case, argued by Hengel, Richard Bauckham and others, replies:

• Formal anonymity was normal for ancient biographies, whose authors were often known to the original readers without signing the text.
• The uniformity of the titles points to an early origin, not scattered guesswork.
• Papias names Mark and Matthew as authors long before Justin.
• Nobody inventing authority would choose Mark and Luke, who were not apostles, nor Matthew, a hated tax collector, over Peter or James.
• "Uneducated" in Acts 4:13 means untrained in the rabbinic schools, not illiterate. A tax collector like Matthew worked with documents, and ancient authors often used secretaries (compare Romans 16:22 and 1 Peter 5:12).

Simon Gathercole has argued further that the Gospels are not strictly anonymous at all, since their texts give real signals about authorship. On balance, the traditional names are well supported for Mark and Luke, reasonably supported for John, and most debated for Matthew.

When were they written?

The arguments for post-70 dates

The main one is Jesus' prophecy of the temple's destruction (Mark 13:1-2), especially Luke's version: "when you see Jerusalem surrounded by armies, then know that its desolation has come near" (Luke 21:20). Many scholars read this as written after the event. Matthew 22:7 — a king who "burned their city" — is read the same way.

The arguments for pre-70 dates

• The ending of Acts. Acts finishes with Paul under house arrest in Rome, preaching "with all boldness and without hindrance" (Acts 28:30-31), around AD 62. It says nothing about the execution of James, the Lord's brother, in 62 (which Josephus records), Nero's persecution in 64, Paul's own death, or the fall of Jerusalem in 70. If Acts was written around 62, Luke — its first volume — is earlier, and Mark, which Luke probably used, earlier still.
• Silence about AD 70. No Gospel says the prophecy came true, though that would have been a powerful point to make.
• Details that assume the temple still stands, such as Matthew's teaching on offerings at the altar (Matthew 5:23-24) and the temple tax (Matthew 17:24-27).

John A. T. Robinson, a liberal Anglican bishop, argued in Redating the New Testament (1976) that the entire New Testament could have been written before 70. Few have followed him all the way, but the book showed the case for late dates was weaker than often assumed.

The Synoptic problem

Matthew, Mark and Luke are called the Synoptic Gospels because they can be "seen together" — they share large blocks of material, often in the same order and sometimes word for word. Explaining that relationship is the Synoptic problem.

Markan priority is the majority view: Mark was written first, and Matthew and Luke both used it. The reasons include:

• Matthew reproduces the great majority of Mark, and Luke over half of it.
• Where they share material, Matthew and Luke rarely agree together against Mark's order.
• Mark's wording is often rougher and his sayings harder, and Matthew and Luke seem to smooth them. Mark says Jesus "could do no mighty work" in Nazareth (Mark 6:5); Matthew says he "did not do many" (Matthew 13:58).

Q (from the German Quelle, "source") is the proposal that Matthew and Luke also shared a second, now lost source — mostly sayings, such as the Lord's Prayer and the Beatitudes — accounting for some 200-odd verses they share that are not in Mark. Q is a hypothesis: no copy has ever been found. Mark plus Q is the "Two-Source Hypothesis", still the most common view.

The Farrer hypothesis, associated with Austin Farrer, Michael Goulder and Mark Goodacre, dispenses with Q: Luke used both Mark and Matthew. Its supporters point to the "minor agreements" where Matthew and Luke agree against Mark in a shared passage. A smaller group defends Matthean priority (the Griesbach or Two-Gospel hypothesis), close to the order the early church assumed.

None of this threatens the Gospels. Luke tells us directly that he used earlier accounts (Luke 1:1-4). Using sources was what responsible ancient historians did.

John and the beloved disciple

John's Gospel never names John son of Zebedee. Instead it speaks of "the disciple whom Jesus loved", who leans on Jesus at the Last Supper (John 13:23), stands at the cross (John 19:26), runs to the tomb (John 20:2-8) and recognises the risen Lord by the lake (John 21:7). The closing note says: "This is the disciple who is bearing witness about these things, and who has written these things, and we know that his testimony is true" (John 21:24).

Three main views:

1. John the son of Zebedee, the traditional view from Irenaeus onwards. The beloved disciple is a member of the inner circle, closely linked with Peter, and the sons of Zebedee are strikingly absent from the Gospel by name.
2. John the Elder, the figure Papias mentions. Bauckham argues this was the beloved disciple — an eyewitness disciple from Jerusalem, but not one of the Twelve — which would explain the Gospel's focus on Judea.
3. Another figure, such as Lazarus, described as the one Jesus loved (John 11:3), or an idealised figure created by a later community, as many critical scholars hold.

What the text itself claims is clear: it presents itself as resting on eyewitness testimony (John 19:35; 21:24). Dating it around AD 90–100 fits both the ancient tradition and modern scholarship, and the early papyrus fragment P52, usually dated to the second century, makes much later dates hard to sustain. For more on John himself, see John — the disciple who outlived them all in the Apostles section.

Bauckham's eyewitness thesis — and its critics

In Jesus and the Eyewitnesses (2006, second edition 2017), Richard Bauckham challenged the long-standing assumption that the Gospel traditions passed through decades of anonymous community storytelling. He argues that they remained attached to known eyewitnesses who were alive and available. Among his arguments:

• Named minor characters. Why name Bartimaeus (Mark 10:46), or note that Simon of Cyrene was "the father of Alexander and Rufus" (Mark 15:21), unless the readers knew these people? Bauckham suggests they were living sources.
• Name frequencies. The relative popularity of names in the Gospels — Simon, Joseph, Judas, John — closely matches what we know of Jewish Palestine at the time, and differs from the Jewish diaspora. The Gospels also add qualifiers (Simon Peter, Simon the Zealot) exactly where common names needed distinguishing.
• Literary markers. Peter is the first disciple named in Mark (Mark 1:16) and the last (Mark 16:7), which Bauckham reads as a signal of Peter as main source — fitting Papias.
• Papias's own method, deliberately seeking living witnesses.

The critics are serious too. Bart Ehrman's Jesus Before the Gospels (2016) draws on memory research to argue that eyewitness recollection is malleable, that stories change in the retelling, and that we cannot assume the eyewitnesses controlled what was written. Others accept that name statistics show Palestinian roots for the traditions without proving direct eyewitness authorship.

What follows

The case for the traditional authors is not airtight, and Christians should not pretend it is. But it is a good case — stronger for some books than for others — and the alternative theories have problems of their own.

And the Gospels themselves claim to be testimony, deliberately written so that readers can know what happened and decide: "these are written so that you may believe that Jesus is the Christ, the Son of God, and that by believing you may have life in his name" (John 20:31).