How Early Was the New Testament Written? The Case for Closeness to the Events

Imagine a rumor trying to spread through a city while the people it is about are still alive to deny it. A story that grows wild in the telling needs two things above all: time, and distance from anyone who was there. Take away the time and the eyewitnesses, and a legend has nowhere to grow. This is the quiet, unglamorous engine behind one of the most important questions in the study of the New Testament: not whether its claims are true—that is a separate fight—but how close its documents stand to the events they report.

It is a question that cuts both ways, and honesty requires saying so at the outset. If the Gospels were composed a century or two after Jesus, by people far removed from Galilee, then a critic’s instinct that we are reading developed legend is reasonable. But if the central claims were already circulating, in fixed verbal form, within a few years of the crucifixion—while hostile witnesses in Jerusalem could still object—then the “legend had time to grow” explanation loses much of its grip. The early-dating argument is a wager that the second picture is closer to the truth.

Stating the Argument Carefully

The argument is usually run in three moves. First: a document produced within the lifetime of eyewitnesses is too early to be a fully developed legend that could circulate unchallenged. Second: the core New Testament material falls inside that eyewitness window—and its earliest layers fall startlingly early. Therefore: the New Testament must be assessed not as late mythology but as first-generation testimony.

Two cautions are needed before we go further, because the argument is often overstated. The conclusion is modest: it tells us the material is early, not that it is accurate. And the second premise does not require the most aggressive possible dating of the four Gospels. As we will see, the strongest version of the case rests on the layers that almost everyone—believer and skeptic alike—already grants to be early: Paul’s undisputed letters of the 50s, and the pre-Pauline tradition embedded inside them.

How the Argument Developed

For much of the twentieth century, the regnant view—shaped by the form critics, above all Rudolf Bultmann—treated the Gospels as the deposit of decades of anonymous community storytelling, with the historical Jesus all but unrecoverable beneath the layers. The most dramatic challenge to that consensus came from an unexpected quarter. In 1976 the liberal Anglican bishop John A. T. Robinson—no friend of conservative apologetics, and famous for the radical theology of Honest to God—published Redating the New Testament, in which his fresh examination of the internal evidence led him to the startling conclusion that the entire New Testament may have been written before the fall of Jerusalem in AD 70.¹ Robinson’s specific dates remain a minority position, but his deeper point landed: the textbook datings, he argued, rested on a foundation thinner and more circular than their confidence suggested.² In parallel, work on the pre-Pauline creeds (reaching back to Joachim Jeremias) and, more recently, Richard Bauckham’s case for eyewitness sources behind the Gospels, has pushed the conversation toward closeness rather than distance.

Premise One: An Early Date Starves a Legend

Why should earliness matter at all? The classic statement comes from the Oxford Roman historian A. N. Sherwin-White, who in his 1960–61 Sarum Lectures (published 1963) turned the historian’s eye on the Gospels. Working from how legends actually behave in the ancient sources—he used Herodotus as his test case—Sherwin-White argued that even two full generations was too short an interval for legend to wipe out a solid historical core; the mythologizing tendency, he observed, takes longer than that to triumph over hard fact.³ For the Gospels the interval is shorter still. Where living memory remains, an invented account does not float free; it collides with people who were present.

That collision was not hypothetical. According to Acts, the earliest preaching of the resurrection happened in Jerusalem, weeks after the crucifixion, to crowds that included the very authorities who had engineered Jesus’s death. Claims about an empty tomb or public events could in principle have been falsified on the spot. Early dating keeps that adversarial check in force, which is exactly what a slow-growing legend needs to escape.

The premise should not be oversold, and a careful critic will press it. The skeptic Richard Carrier and others have noted, rightly, that Sherwin-White’s “two-generation rule” was an aside about a particular body of Greek material, not a demonstrated law of folklore; rapid legendary embellishment is not impossible, and the rate at which stories mutate varies enormously.⁴ The honest reply is that Premise One does not need an iron law. It needs only the comparative point that more time and more distance make undetected legend easier, and that the New Testament window is unusually tight by the standards of ancient history—where our best surviving biographies of Alexander the Great, by Arrian and Plutarch, were written some four centuries after him and are still mined for a historical core.

Premise Two: The New Testament Really Is Early

The Creed Behind 1 Corinthians 15

Here the case finds its firmest ground, and it must lead the way—because the popular objection that “the Gospels were written decades later” quietly assumes the narrative Gospels are the earliest evidence we have. They are not. In 1 Corinthians 15:3–7, writing in the mid-50s, Paul hands on what he himself had “received”: that Christ died for our sins according to the Scriptures, that he was buried, that he was raised on the third day according to the Scriptures, and that he appeared to Cephas, then to the Twelve, then to more than five hundred, to James, and to all the apostles. The technical language of “delivering” and “receiving” tradition, the non-Pauline vocabulary, the rhythmic structure, and the Aramaic name Cephas all mark this out as a pre-formed oral creed that Paul is quoting, not composing—a formula already circulating before he wrote it down.⁵

How early? Paul’s own conversion is usually placed within a few years of the crucifixion, and in Galatians 1:18–19 he describes going up to Jerusalem to meet Cephas (Peter) and James—two of the named witnesses in the creed—three years later. The tradition he relays must predate that meeting. This is why the dating crosses ideological lines. The German New Testament scholar Gerd Lüdemann—who emphatically does not believe Jesus rose bodily and explains the appearances as visions—nonetheless concluded that “the elements in the tradition are to be dated to the first two years after the crucifixion… not later than three years.”⁶ Gary Habermas, who has surveyed the literature most exhaustively, reports that a wide range of scholars across the spectrum who address the question place the formula within the first few years—though this figure should be cited with care, since it is a count of published positions assembled by one researcher rather than a randomised poll.⁷

The force of this is hard to overstate, and it reframes the whole debate. Whatever one decides about the date of Mark, the proclamation of Jesus’s death, burial, and resurrection appearances is not a late accretion that grew over generations. It sits at the very foundation, traceable to the believing community in Jerusalem within a handful of years of the events—and naming individuals, some still alive, who could be questioned.

Pauline Chronology and the Silence of Acts

Paul’s undisputed letters—Romans, the Corinthian correspondence, Galatians, Philippians, 1 Thessalonians, Philemon—are dated by near-universal agreement to roughly AD 48–60. They are themselves first-generation documents, written by a man who knew the original apostles, and they already presuppose the central narrative of cross and resurrection. That alone places the load-bearing claims well inside living memory.

A second clue concerns the book of Acts. Luke–Acts is a two-volume work, and Acts ends with remarkable abruptness: Paul under house arrest in Rome, awaiting trial, the gospel preached “with all boldness and without hindrance” (Acts 28:31). What it never narrates is striking. It does not report the martyrdom of Paul (c. AD 64–67), nor of Peter, nor the death of James the brother of Jesus (AD 62, recorded by Josephus⁸), nor the Neronian persecution, nor—most conspicuous of all—the destruction of Jerusalem and its Temple in AD 70. Luke, who narrates the martyrdoms of Stephen and of James son of Zebedee in detail, and who returns again and again to the theme of the Temple, would have had every reason to mention these catastrophes had they already happened. On this reading, Luke wrote Acts in the early 60s, which places his Gospel (and Mark, one of its sources) earlier still. This is the heart of Robinson’s case, and it was pressed in great detail by the classicist Colin Hemer.⁹

Patristic Testimony

Finally, the early church remembered who wrote the Gospels and traced them to the apostolic circle. The earliest such testimony comes from Papias of Hierapolis (writing c. AD 110–130), preserved by the church historian Eusebius. Papias reports, on the authority of “the Elder,” that Mark became Peter’s “interpreter” and accurately wrote down what Peter recounted of the Lord’s words and deeds, though “not in order,” and that Matthew compiled the sayings “in the Hebrew dialect.”¹⁰ Later writers—Irenaeus around AD 180—name all four Evangelists. This testimony is external corroboration that the church tied its Gospels to eyewitnesses and their associates, not to anonymous late authors.

The Strongest Objection, Stated at Full Strength

The mainstream of critical scholarship does not accept the early-60s dating of the Gospels, and its case must be stated fully. The standard consensus, articulated by Bart Ehrman among many others, dates Mark to around AD 70, Matthew and Luke to roughly 80–85, and John to 90–95.¹¹ Ehrman regards these dates as “not really particularly controversial.” The reasoning is layered: Matthew and Luke used Mark as a source, which requires Mark to have circulated first; and—the decisive point—Mark 13 (the “Olivet Discourse”) and its parallels have Jesus predicting the encirclement and destruction of Jerusalem. The simplest historical explanation, on this view, is that the Evangelist wrote after AD 70 and shaped the prophecy in light of what had occurred—a vaticinium ex eventu, a “prophecy after the fact.”

This is not special pleading, and it is important not to caricature it. Historians as a rule do not assume miracles or genuine foreknowledge when an ordinary explanation is available, and the correspondence between Jesus’s words and the events of 70 is real. The great historian of Judaism E. P. Sanders, while granting that nothing in Mark must be dated after 70, still inclined to a date around 70 on broader grounds.¹² Luke 21:20, with its talk of Jerusalem “surrounded by armies,” reads to many scholars as more specific than Mark—suggesting an author writing with the siege in view. To a critic, the early-dater looks like someone willing to multiply improbabilities (genuine prediction, a string of coincidental silences) to avoid the obvious.

A Reply—and What It Can and Cannot Carry

The reply begins with a concession. The vaticinium ex eventu reading is a legitimate historical inference, and it cannot be refuted simply by insisting that Jesus could predict the future—that would smuggle the conclusion into the premise. The most one can say is that the argument is not as decisive as it sounds, for three reasons.

First, the prediction in Mark 13 is conspicuously unspecific about the manner of Jerusalem’s fall. It speaks of the Temple’s stones thrown down, of wars and rumors of wars, of flight to the mountains—but says nothing of the fire that gutted the Temple, the months-long siege, or the legions of Titus, the very details a writer narrating after the fact would find hard to omit. Some scholars—notably N. T. Wright and R. T. France—have noted that the discourse draws its imagery from the Old Testament and from earlier threats to the Temple rather than from the specifics of 70. Second, the silences in Acts remain hard to explain on a late date; a convergence of expected-yet-absent events (Paul’s death, James’s death, the fall of Jerusalem) is more telling than any single omission. Third, and most importantly, the early-dating case for the central claims does not stand or fall with Mark’s date at all. Even on Ehrman’s own chronology, the creed of 1 Corinthians 15 and Paul’s letters carry the death, burial, and resurrection proclamation back into the 30s and 50s. Whether Mark is 65 or 72 changes little here.

And here the limits must be stated plainly, in the critic’s favor. The argument from silence in Acts is genuinely an argument from silence, and such arguments are only as strong as the certainty that the author would have mentioned the missing event—a judgment about which reasonable scholars differ. Luke may have ended Acts where his narrative purpose was complete, with the gospel reaching Rome. The dissenting expert here is the consensus itself: most New Testament specialists are not persuaded that the silences outweigh the source-critical and Olivet-Discourse considerations, and they place the Gospels a decade or two later than Robinson did. The early-dating of the narrative Gospels to the pre-70 period is therefore a serious, defensible, but genuinely contested position—not a settled result.

So let the critic have the last objection before the reply. Ehrman would say: you have rescued the early creed, granted, but you have not rescued the early Gospels, and it is the Gospels that contain the miracles, the empty tomb, the detailed narratives—so the gap you most needed to close, you have left open. That is fair. The considered answer is that the argument never promised to close it. What the creed establishes is that the explanatory core—a crucified, buried, and (it was claimed) risen Jesus, attested by named witnesses—is too early to be slow legend. The narrative details are a further question, addressed by other lines of evidence, including the case that the Gospels themselves rest on named eyewitness testimony.

The Honest State of the Scholarship

Two things are true at once, and a responsible presentation holds them together. On the dating of the four Gospels, the mainstream consensus—Mark c. 70, the others in the 80s and 90s—remains the majority view, and the pre-70 dating championed by Robinson and Hemer is a respected minority position, not the field’s settled judgment. But on the dating of the earliest tradition—the 1 Corinthians 15 creed and the undisputed Pauline letters—there is broad agreement across the ideological spectrum, sceptics included, that the material is remarkably early. The live debate is not about whether the central proclamation is early (it is) but about how much follows from that, and about where exactly the narrative Gospels sit within the eyewitness generation.

Conclusion: What Early Dating Does and Does Not Establish

What the argument establishes is real and worth holding onto. The foundational Christian claims about Jesus’s death and resurrection appearances are not the residue of generations of mythmaking; they are demonstrably early, fixed in creedal form within a few years of the events, named to specific witnesses, and circulated while hostile parties in Jerusalem could have objected. Even on the mainstream (later) dating, the Gospels fall within living memory. The “legend had time to grow” hypothesis, in its strong form, simply does not fit the evidence.

What it does not establish is equally important, and pretending otherwise would forfeit the reader’s trust. Early does not mean true. An early date rules out one explanation—slow, undetected legendary development—but it does not rule out sincere error or deliberate fabrication; people can and do invent things quickly. Nor does it prove the pre-70 dating of the narrative Gospels, which remains contested. The early-dating argument is best understood as removing a whole class of objection and clearing the ground—handing the central claims over to be assessed as first-generation testimony, where the harder questions of honesty, accuracy, and the resurrection itself can be taken up on their own terms. It does not win the war. It just makes sure the right battle gets fought.