Papyrus P52: a small fragment of John's Gospel (John 18:31-33) from the John Rylands Library
Papyrus P52, the earliest generally accepted surviving copy of New Testament text (public domain).

How the Gospels Were Written

Four books, no signatures, decades of retelling before ink. What scholars can say about who wrote the Gospels, when, and why three of them look so alike.

Nobody signed them

None of the four Gospels names its author in the text. The titles we use — “the Gospel according to Matthew,” “according to Mark,” and so on — do not come from the writers themselves. Most scholars hold that the Gospels were originally anonymous compositions, and that the present titles, though ancient, are not traceable to the evangelists (the Catholic Encyclopedia states this plainly, and even conservative scholars such as Craig Blomberg concede the point). A minority of scholars disputes this, arguing the names were attached from the start and never seriously contested — the disagreement itself is worth knowing about.

What is not disputed is the timeline of the attributions. By the early second century, manuscripts were being headed “according to Mark” and the like, and the early church writers preserve the traditions behind the names:

So the traditional names are the early church’s attributions, not the authors’ bylines. Modern scholars usually keep the conventional names for convenience while distinguishing them from what can be proven about authorship.

From telling to writing

The Gospels were not written the week after the crucifixion. The picture most scholars draw is of decades of telling — preaching, teaching, and liturgy — before the stories were fixed in extended written form. Luke’s Gospel opens with a preface that says as much: Luke notes that many had already undertaken to write accounts, that he investigated everything carefully from the beginning, and that he is writing an orderly account for his reader Theophilus (Luke 1:1–4, paraphrased).

When were they written? The evidence is scanty — no original manuscript survives, and the earliest complete copies date to the fourth century — so all dates are scholarly estimates, not certainties. The ranges most commonly given in reference works are:

Some scholars argue for earlier dates, some for later ones; the ranges above represent the broad middle of opinion, not a consensus.

Why three Gospels look alike: the synoptic problem

Matthew, Mark, and Luke — the “synoptic” Gospels, from the Greek for “seen together” — tell many of the same stories, sometimes in nearly identical wording. John stands apart, with different structure and much distinctive material. If the three were written independently, the verbatim agreements would be a remarkable coincidence. The question of how they relate is called the synoptic problem, and scholars have proposed several solutions:

Whether John knew the Synoptics is itself debated; most scholars treat his Gospel as an independent tradition.

What the manuscripts actually show

The image above is worth a long look. P52 is a fragment the size of a credit card, written on both sides — which means it came from a codex (an early book), not a scroll. It contains a few verses from Jesus’ trial before Pilate (John 18). It was found in Egypt, which means John’s Gospel had traveled far from its place of writing within decades of its composition.

Its date is conventionally given as c. 125, but that precision should not be overstated: the dating is palaeographic (based on handwriting comparison), and palaeography is not an exact science. The papyrologist Brent Nongbri has argued the evidence supports only a broad second-century range, and the John Rylands Library itself now notes that recent research points to a date nearer 200 — while also noting there is no convincing evidence of any earlier New Testament fragment. What P52 firmly establishes is that the Gospel of John was in circulation, in book form, in Egypt, early in the second century.

Source notes

Limitations