One of the longest-running and most contentious debates about the historicity of the Bible involves the date of composition of the four New Testament gospels. An early date, while not necessarily ensuring the historical reliability of the gospels, would seem to allow less time for embellishment and distortion to creep in. Conversely, a late date would leave more room for at least the possibility of these things. Naturally, conservative and inerrantist scholars and apologists want to push that date back as far as possible, sometimes as early as a few years after they believe the death of Jesus occurred.
So, what conclusion is best supported by the facts? This essay will examine four categories of evidence: internal evidence from the text of the gospels themselves; documentary evidence from the early church record; physical evidence in the form of surviving manuscripts; and negative evidence in the form of missing references to the gospels where we would have every right to expect to find them.
- Internal Evidence
- Documentary Evidence
- Physical Evidence
- Negative Evidence
In general, internal evidence is not a reliable marker of when a text was composed. After all, there is always the possibility that a text which was composed late was written to seem early. This is a known problem in the biblical world, where many apocryphal documents were attributed to famous past figures to give them greater authority. Nevertheless, Christian apologists sometimes cite verses from within the gospels as a way of establishing their date, so we’ll consider several of the most common examples.
In “2000 Years Late“, I called attention to this passage from the so-called Olivet Discourse, regarding Jesus’ prophecies of the end of the world:
“And as he went out of the temple, one of his disciples saith unto him, Master, see what manner of stones and what buildings are here! And Jesus answering said unto him, Seest thou these great buildings? there shall not be left one stone upon another, that shall not be thrown down. And as he sat upon the mount of Olives over against the temple, Peter and James and John and Andrew asked him privately, tell us, when shall these things be? and what shall be the sign when all these things shall be fulfilled?”
—Mark 13:1-4 (KJV)
The historical fact is that the Temple was destroyed, along with much of Jerusalem, during the Jewish War that occurred around 70 CE. Conservative Christians naturally take this passage as proof of Jesus’ miraculous powers of foresight, and an indication that the gospels were written before the events they foretell. However, an equally possible interpretation is that this passage was composed after these events, and was written to “foretell” them in order to make it seem as if the author had access to a source of supernatural insight. (Textual scholars are in wide agreement about other places in the Bible where this strategy was used, particularly the Book of Daniel.) Since no internal evidence of this passage can decide between these alternatives, by itself it is of little help in dating the gospels.
Another point often appealed to by Christian apologists is that the Book of Acts ends with Paul under house arrest in Rome, while Christian tradition states that he was put to death in that city around 60 CE. Since Acts does not record Paul’s death, it’s a logical deduction that it was written before that happened. And since Acts identifies itself as having been written by the same person who wrote the Gospel of Luke, as a sequel to that book, we can conclude that Luke is even earlier. The Gospel of Mark, which Luke and Matthew are based on, must therefore be earlier still.
However, reasonable as this sounds, the premise is shaky. There are other possible reasons why Acts ends where it does. One, as previously mentioned, is that it’s a late document whose author sought to make it seem earlier by leaving out Paul’s death. Another, as skeptical historian Earl Doherty suggests in his book Challenging the Verdict: “[Some] scholars see in Acts’ plot line a symbolic progression of the faith’s early expansion from Jerusalem to Rome, from a Jewish beginning to a gentile culmination, so the author may well have wanted to avoid ending on a negative note. That symbolic progression would have been somewhat compromised by having Paul get his head chopped off.”
Another verse from the Gospel of Luke, which I haven’t often seen mentioned, points in a different direction:
“Many have undertaken to draw up an account of the things that have been fulfilled among us, just as they were handed down to us by those who from the first were eyewitnesses and servants of the word. Therefore, since I myself have carefully investigated everything from the beginning, it seemed good also to me to write an orderly account for you, most excellent Theophilus, so that you may know the certainty of the things you have been taught.”
—Luke 1:1-4 (NIV)
In Christian tradition Luke was the companion of Paul, so it’s not surprising that he doesn’t claim to be an eyewitness to Jesus’ ministry. But even so, his saying that the events he records are things “handed down to us” does not sound like the wording that would be chosen by someone writing within a few years of the event. Although it’s not conclusive, this verse implies that Luke’s gospel, at least, may have been written a generation or more after the events it describes.
Moving beyond the text of the gospels themselves, the next category of evidence is the written records left behind by the evangelists and bishops of the early church. In most cases, these records (unlike the gospels themselves) can be fit better into an established historical context, and thus dated more securely. Definite references to the gospels in the writings of these figures can help to fix the time period within which they must have been originally written – although as we’ll see, what isn’t there can be almost as illuminating as what is.
One of the early church figures commonly appealed to in dating the gospels is Papias, a bishop who lived in the early second century. His best-known work is The Sayings of the Lord Interpreted, though the book itself is lost and we know of it only in fragments. One surviving passage from Papias, quoted by the fourth-century bishop Eusebius, has some tantalizing details that Christian apologists often take as a reference to the Gospel of Mark:
“This, too, the elder used to say: Mark, who had been Peter’s interpreter, wrote down carefully, but not in order, all that he remembered of the Lord’s sayings and doings. For he had not heard the Lord or been one of his followers, but later, as I said, one of Peter’s. Peter used to adapt his teaching to the occasion without making a systematic arrangement of the Lord’s sayings, so that Mark was quite justified in writing down some things just as he remembered them. For he had one purpose only — to leave out nothing that he had heard, and to make no misstatement about it.”
Although Papias agrees with Christian orthodoxy in making Mark the follower of Peter, it’s clear from this passage that the document he’s discussing is not the Gospel of Mark as we have it. Papias says clearly that Mark wrote down all he could remember, “but not in order”, not as a “systematic arrangement”. This indicates that the document Papias knew of was a collection of sayings and parables, probably similar in form to the apocryphal Gospel of Thomas, rather than the coherent narrative with a storyline that we possess. When (or whether) this collection was edited and reworked into the Gospel of Mark as it exists today is the question we want to know the answer to, but it’s not a question that Papias helps answer. Note especially that he himself was not even in possession of this document, but is passing this claim along as second-hand information which he got from an unnamed “elder”.
When it comes to the Gospel of Matthew, Papias is even less helpful:
“Matthew compiled the Sayings in the Aramaic language and everyone translated them as well as he could.”
Again, Papias states that the document known to him was not a narrative with a storyline, but a collection of sayings. The reference to Aramaic is also puzzling, since as Earl Doherty points out, New Testament scholarship has long established that the Gospel of Matthew was originally written in Greek, based on the Greek Gospel of Mark.
Another first-century Christian, Justin Martyr, is the first writer to give recognizable quotes from the gospels and to refer to them as such. For instance, from his First Apology, dated to around 150 CE:
“For the apostles, in the memoirs composed by them, which are called Gospels, have thus delivered unto us what was enjoined upon them; that Jesus took bread, and when He had given thanks, said, ‘This do ye in remembrance of Me, this is My body;’ and that, after the same manner, having taken the cup and given thanks, He said, ‘This is My blood;’ and gave it to them alone.”
Notably, Justin gives no specific names for any gospel author, nor does he say how many there were. As in the above passage, he refers to them only collectively, as “memoirs” of the apostles. This may have been a careless omission on his part, or a stylistic choice. However, it may also have been because he was unfamiliar with them or had heard of them only dimly.
Writing around 180 CE, Irenaeus of Lyons is the first Christian commentator we know of who lists the four canonical gospels by name, describes them as containing eyewitness testimony to the life of Jesus, and presents them as authoritative and reliable (in his book Against Heresies). This reference sets a latest-possible date for the gospels’ composition, although it doesn’t tell us how much earlier they may have been composed.
One thing that Christians should find unsettling, though, is just how late this latest-possible date is. Not until a hundred and fifty years after the events they purport to describe do the gospels definitively make their entrance into the historical record. It’s as if the first detailed records of World War II weren’t written until the year 2100. One would think that documents written by the first followers of Jesus would have immediately made a large and memorable splash in the Christian community, but that doesn’t seem to be what happened. Instead, over a period of about sixty years, we begin to see vague and ambiguous references to gospels and gospel-like documents trickling into the consciousness of early Christianity, until finally clear references like Irenaeus’ emerge. It seems as if the gospels were circulating, in various forms and versions, for some time before their authenticity was finally accepted.
Although physical evidence could theoretically be used to set the exact date of the gospels’ composition, we’d have to be extremely lucky to find a first-generation copy of the text. Most of the earliest Christian writings have long disintegrated in their original autographs, and all that we have left are later copies and fragments. This section will briefly consider a few of the most significant archaeological findings.
Of all the ancient papyri known to biblical scholars, one of the most controversial is called 7Q5. This sliver of papyrus was found among the Dead Sea Scrolls, in a cave that was part of the ancient Essene Jewish settlement at Qumran, and was claimed by the Jesuit historian Jose O’Callaghan to be part of a verse from Mark.
The Dead Sea Scrolls are believed to be no older than about 70 CE, so if this identification were correct, it would make 7Q5 by far the earliest of all known remnants of the New Testament. However, that identification is highly contentious at best. 7Q5 is extremely small and fragmentary, literally no more than a handful of Greek letters – and some of those letters are simply wrong for the verse Callaghan claimed it to be, as even most Christian scholars argue:
It is like observing a sheet of paper that has been left out in the rain. Only a handful of letters can be made out clearly; all else is up for grabs. Now suppose I come along and say that one or two of the clear letters need to be changed. And of the unclear letters, I propose three or four nearly impossible suggestions. I do this because I have a certain text in mind that I want this sheet to be a copy of.
An identification this dubious, in a text this fragmentary, offers us no reliable evidence whatsoever that the gospels existed by this date. And Callaghan’s hypothesis has a further difficulty: even if it was in existence by this time, what would the staunch, reclusive Jews of Qumran be doing with a copy of the New Testament?
The earliest undisputed physical copy of any part of the gospels is P52, a scrap of papyrus containing fragments of a few verses from the Gospel of John. Unlike 7Q5, the textual identification is far more secure. However, the dating of P52 is based solely on paleographic arguments, and allowing for the imprecision that must come with such considerations, it could lie anywhere between 100 and 160 CE, or even later. Nor can we tell, from this small scrap, whether and to what extent the document it was originally part of agrees with the Gospel of John as we now have it.
After P52, the next oldest New Testament papyri are further out. Many, like P52, are little more than scraps. Of the ones that are more complete, one of the oldest is P75, also called Papyrus Bodmer XV, which dates to between 175 and 225 CE and contains substantial portions of the Gospels of Luke and John. Another one from this time period, P66, contains almost the full text of John, minus the Pericope of the Adulteress, which was a later addition. Although these documents are of little help in determining when the gospels were originally written, they do agree quite well with Irenaeus in setting a last-possible date for their composition.
The final category is documentary evidence, but of a different sort: negative evidence, which comes in the form of references to the gospels that are missing where we would have every right to expect them. Given the fact that many early documents are lost to us, relying on this kind of evidence has its potential pitfalls; we can never know what a Christian may have said in works that did not survive. On the other hand, the use of negative evidence is more appropriate when we have a surviving work in which the author would have had a very good reason to cite one of the gospels, but does not do so. This pattern of omission casts a detectable shadow in the historical record, helping us to infer the date range in which the gospels had not yet been written, or at the very least had not made a significant impact on the larger Christian community.
Of all the Christian figures who know nothing of the gospels, by far the most prominent is Paul. Although Paul obviously contains some similar creedal statements, he never explicitly quotes from the gospels, nor does he refer to their existence either directly or indirectly. Critical scholars are in wide agreement that Paul’s letters were early, probably around the middle of the first century CE. It is implausible that, if he knew of the gospels, he would never find occasion to cite them in any of his surviving writings. This is all the more true because the gospels contain discussion of several theological issues that Paul does discuss and even debate with his audience, such as the continued validity of purity laws (Romans 14:14; cp. Matthew 15:20) or divorce (1 Corinthians 7:10; cp. Mark 10:11), where citation of Jesus’ actual words would have been a great help to him.
This silence indicates either that the gospels had not been written by this date, or that they were written but were not in wide circulation, such that Paul had not heard of them. But if the gospels are what traditional Christianity says they are, that latter possibility is not plausible. If any of the handpicked followers of Jesus had written books about his life, how could Paul – or any Christian – not have heard of them?
The Didache (Greek for “teaching”) is an epistle of the early church, a sort of theological instruction manual, which contains teachings for Christians on how to pray, how to practice rituals, how to receive fellow believers and wandering evangelists, and so on. Its style and content hint that it originated in a very early branch of Christianity, one that had not yet become formalized into a hierarchy. It’s commonly judged to have been written sometime around or before 100 CE.
The interesting thing about the Didache is that, although it contains teachings clearly related to some that appear in the gospels – including praying for your enemies, loving God and loving your neighbor as yourself, turning the other cheek, and so on – it never attributes them to a historical, human Jesus. Indeed, Jesus is scarcely mentioned at all, save for fleeting mentions in the prayers for Eucharist and communion. He is never described as a teacher, never described as the source of the moral lessons quoted there, and his death and resurrection are never even mentioned. Rather than being derived from the gospels, it seems that both the gospels and the Didache drew on a common pool of material, which the gospels’ authors put into the mouth of their Jesus character while the Didache left them as anonymous proverbs.
Like the Didache, the First Epistle of Clement is one of the oldest authentic Christian works outside the New Testament canon. It was written in Rome, attributed to the bishop Clement, and probably dates to the end of Domitian’s reign, around 96 CE. Also like the Didache, 1 Clement’s author quotes material that sounds very similar to the teachings in the gospels, yet never attributes it to a specific written source:
“Let us especially remember the words of our Lord Jesus Christ which he spake when teaching gentleness and long-suffering, for he spake thus: ‘Show mercy, that ye may obtain mercy; forgive, that it may be forgiven unto you; as ye do, so shall it be done unto you; as ye give, so shall it be given unto you; as ye judge, so shall ye be judged; as ye are kindly affectioned, so shall kindness be showed unto you; with whatsover measure ye measure, with the same shall it be measured unto you.’” (13:1-2)
More strikingly, he tries to persuade his readers of the plausibility of resurrection by citing a mythological creature, the phoenix, and examples from nature. He never mentions what would seem to be the most obvious example of all: the gospel story of Jesus’ own resurrection from the dead.
“Let us look, beloved, at the resurrection that is ever taking place. Day and night show to us the resurrection; the night is lulled to rest, the day ariseth; the day departeth, the night cometh on. Let us consider the fruits, in what way a grain of corn is sown. The sower goeth forth and casteth it into the ground, and when the seeds are cast into the ground, they that fell into the ground dry and naked are dissolved; then after their dissolution, the mighty power of the providence of the Lord raiseth them up, and from one seed many grow up and bring forth fruits.” (24:2-5)
“There is a bird which is called the phoenix. This, being the only one of its kind, liveth for five hundred years. And when the time of its death draweth near, it maketh for itself a nest of frankincense and myrrh and the other perfumes, into which, when its time is fulfilled, it entereth, and then dieth. But as its flesh rotteth, a certain worm is produced, which being nourished by the moisture of the dead animal, putteth forth feathers… Shall we then think it great and wonderful, if the Maker of all things shall make a resurrection of those who, in the confidence of a good faith, have piously seized him, when even by means of a bird he showeth the greatness of his promises?” (25:2-26:1)
Clearly, the author of 1 Clement, though a Christian, has never heard of the gospel story. Why else would he use such strained examples of resurrection, if a much better one was staring him in the face?
Ignatius of Antioch was a bishop in the early church during the reign of Trajan, between 98 and 117 CE. He’s most famous for a series of letters he wrote to churches throughout the Ancient Near East as he was being taken to Rome, where he would ultimately be put to death, by the Roman authorities. Unlike many first-century church figures, Ignatius does declare his belief in Jesus’ life and career in no uncertain terms:
“Concerning those things, my beloved, I wish you to be warned beforehand (not because I knew that any of you were so disposed, but as being less than you), so that you fall not into the snares of vainglory, but may be fully persuaded of the birth, the passion, and the resurrection which happened in the time of the governorship of Pontius Pilate, which things were truly and surely done by Jesus Christ, our hope, from which hope may it happen to none of you to be turned away.” (source)
Although Ignatius clearly believes in a story much like that of the gospels (and in fact, is the first Christian writing outside the canon to say so), he never specifically cites a written gospel in support of his claims. Notably, he writes very much as if there were others who did not believe this story, and he was trying to persuade his readers not to be seduced by them.
Even when all the evidence is summed up, the question of when the gospels were written remains a murky one. The best we can do is make educated guesses. The most likely conclusion, I believe, is that the Gospel of Mark was written around the time of the Jewish War (c.70 CE), either soon before it or, more likely, considering the evidence of the Olivet Discourse, soon after it. This would mean that Paul wrote his epistles before the gospels ever existed, which explains why he shows no knowledge of their contents, and likewise shows why other early first-century Christian sources omit mention of them.
Matthew and Luke, both of which are based heavily on Mark, came later, sometime between 70 and 100 CE. This would account for Luke’s opening remark about “things handed down to us”; by then, enough time had passed between the date of composition and the presumed date of the events recorded for this to be true. John, the last of the gospels and the one that shows the most divergence from Mark, was probably written sometime after 100 CE.
These dates are at the outer limit of the range when eyewitnesses could conceivably still have been around, though the shorter life expectancies of the ancient world weigh against that. But the more important consideration is how the gospels were received by the early Christian community. The orthodox view holds that Jesus’ life and the identities of his followers were universally known and unquestioned facts. But this view fails to explain why the gospels, if they were written by the closest followers of Jesus, didn’t immediately make a large and noticeable impact on the Christian documentary record.
Instead, what we find is that they circulated, seemingly below the radar, for several decades. Gradually, the events they describe began to seep into Christian consciousness – which accounts for Ignatius’ seeming knowledge of the gospel story despite lacking any specific sources, or Papias’ mention of disconnected collections of sayings and miracle stories under the names of Mark and Matthew. (These may have been forerunners or earlier drafts of the gospels by those names. It may even be that later Christians, seeking to put authors’ names on these anonymous documents, misinterpreted Papias to have been referring to the gospels and used the names he inadvertently proposed.) The gospels may have been the sources of these ideas, or they may have been just one expression of things that Christians were gradually coming to believe. In either case, depending on the exact date of composition, they may have been around for between fifty and one hundred years before they were finally accepted by all Christians as canonical and orthodox.
This does not accord with the usual picture of early Christianity, promoted by pious fictions like the Book of Acts, which envisions the church as unified, organized, and possessing a clear hierarchy and line of succession from the beginning. On the contrary, the evidence indicates that early Christianity was a broader and more diverse movement, encompassing a wide variety of sects that had little in common with each other theologically: docetists, adoptionists, Marcionists, gnostics, and others. Their beliefs about Jesus were equally varied: some envisioned a Christ figure who, like the saviors of the mystery religions, had performed his saving act in a heavenly sphere and had never walked on Earth. Others believed in a Christ who was a kind of supernatural phantom, one who seemed to take the form of humanity but was not made of flesh and blood. Yet others believed in a fully mortal and genuinely human Jesus. It was this latter sect that slowly gained in prominence, no doubt aided by the gradually widening circulation of the gospels. In time, it absorbed the early and probably well-known epistles of Paul (who, ironically, believed something closer to the first view). Its final triumph came when it gained the ear of an emperor, at last giving it the power to declare what would be dogma and stamp out opposing views. There must have been pitched theological battles between these early Christian sects. Regrettably, little evidence has survived of them – although we do have hints, such as when Paul anathematized those who preach “another Jesus” (2 Corinthians 11:4) or when 1 John refers to “spirits of antichrist” who deny that Jesus had come in the flesh (4:3). Ignatius refers to heretics as “mad dogs biting by stealth” (Ephesians 7:1) and “wild beasts” (Trallians 7:1).
The lesson to take away from all this is that the gospels were neither the first nor the only strand of Christian belief. In contrast to the carefully cleaned-up and sanitized story presented by orthodoxy, the actual historical record testifies that the theology represented by the gospels was just one thread in a larger tapestry, one that took decades to rise to preeminence. Once it had achieved that status, its advocates engaged in a conscious process of historical revision, making it seem as if their favored belief was the predestined winner all along. In reality, if just a few historical events had turned out differently, the version of Christianity that dominates might have been very different from the one we now see.