Are the Gospels Historically Reliable?
Why this question is so difficult to answer...
For the student of the gospels, there are few topics more tantalising than the gospels’ historicity. How accurately do the gospels convey Jesus and his times?
Online, a content machine of apologetics and counter-apologetics dispenses easy answers to these questions. The crudest apologists assure us that the gospels are four sources close up to the facts, capturing the past as it actually played out. Meanwhile, their sceptical counterparts inform audiences that the gospels bear almost no connection to the historical Jesus – if there was a historical Jesus at all.
In the view of mainstream critical scholarship, however, neither party is to be fully trusted. It is not that the question of historical reliability is of no interest. The study of the historical Jesus is – in the words of one Oxford historian – “perhaps the most thorough and sophisticated analysis of any set of texts in the history of human thought.”1 Rather, it is that the sources do not afford us an easy ‘yes’ or ‘no’.
In this piece, I hope to sketch why this is. Why is the gospels’ historicity so difficult to determine? And why is it unlikely that this debate is soon to be settled?
#1 The Gospels’ uncertain provenance
When ascertaining the reliability of an ancient source, it is generally helpful to know something about their provenance: who wrote the texts, where and when. Which brings us to our first difficulty in assessing the gospels’ reliability: the gospels’ provenance is highly contested.
Traditionally, the titles of the gospels are attributed to Matthew and John (eyewitnesses and disciples of Jesus) and Mark and Luke (followers of eyewitnesses). Yet as I have unpacked elsewhere, there are good reasons to think that the gospels were originally anonymous writings, like many Jewish texts.
In the view of most scholars today, these titles were attached later, when the gospels came together as a four-fold collection. This would explain why they each have the same formulaic title – ‘The Gospel according to X’ – and why they are not referred to as the work of a particular disciple until Irenaeus in the late second century.
What then are we to make of the names we now find them? One early tradition, found in the writings of Papias, is that Mark knew Peter, and that Matthew composed something in Hebrew or Aramaic. Yet Matthew probably did not write our present Greek Gospel, and the Petrine association with Mark is similarly contested.
The situation with Luke is similarly muddied. The idea that Luke wrote his gospel finds no earlier attestation than Irenaeus, and we know that Marcion knew a version of the text without its traditional title. Many scholars now think that Luke the physician was inferred as the author of the text, since Acts has a number of ‘we’ passages in the second half of the book, and Luke was one of Paul’s companions.
Meanwhile, the author of John hides behind the moniker, ‘the disciple whom Jesus loved’. Yet the quest for the beloved disciple has not yielded clear results. Though some think that an eyewitness stands behind the gospel, the fact that the sons of Zebedee are referred to in the narrative (21:2) makes its traditional attribution unlikely. Some scholars now think that the work’s authorship is deliberately obscure, or that the ‘beloved disciple’ functions as a literary device, not a historical figure.2
Other aspects of the gospels’ provenance are enigmatic. For example, the jury is still out on when and where the gospels emerged. Some common dates (70-110 CE) and locations recur in the literature. Mark, for example, is often said to have been written in Rome shortly after Jerusalem’s destruction. Yet it is difficult to pin the gospels to any specific place, and a minority of scholars prefer earlier dates.
I always keep in mind what a Professor once said to me regarding the gospels’ dating: that these really all are just guesses – our best attempt to work out what went on. Without a clearer image of the gospels’ provenance, we are not afforded some of the context about an author which we would like to gain insight into their reliability.
#2 The Gospels are ancient biographies
A second problem is that the gospels are ancient ‘lives’ (bíoi). For some scholars, this fact is somewhat consoling. Many ancient biographers saw their life-writing as something adjacent to history. This meant that they would not generally invent scenes, but relied upon source material to compose their biographies.
The difficulty is that biography was a fluid genre, perhaps especially prone to blur the lines between fact and fiction. Some biographies, such as the Life of Aesop, written around the time of the Gospels in a similar style, were almost entirely fictional. Though drawing upon earlier material and believed to reflect a historical figure, the narrative itself was fictional and bore little connection to the actual past.
The gospels are not about a character of the mythical past, and claim a more direct connection to history (see Lk. 1:1-4). Yet knowing this about their genre tells us little about their historicity, for the value of ancient biographies, like modern ones, often varied. A classicist today might recognise the general value of Plutarch’s Cicero, while also cataloguing a host of errors –3 biography does not equate to historicity.
Perhaps more problematically, historically-minded biographers often worked with different methods of composition to their modern counterparts. Consider this summary of Tomas Hägg, a leading classical scholar of ancient biography:
“Ancient life-writers did not encounter among their contemporaries the same demands for documentary truth as their modern colleagues do, nor did for that matter ancient historiographers… Conversations are allowed to be fictitious, and insight is readily granted into the acting characters’ feelings, thoughts, and motives, as long as some kind of verisimilitude is maintained. The establishment of any form of higher truth – be it poetic, psychological, philosophical, or religious – overrules demands for the truth of facts.’4
As studies by classicist Christopher Pelling and New Testament scholar Michael Licona have demonstrated at length, biographers like Plutarch were trained to rework their sources, using a range of literary techniques.5 This should give us pause for thinking that what we have in the gospels is straightforward reportage of events as they actually occurred. Rather, what we often encounter is a curated representation of the past – a literary attempt to bring to life the character or éthos of an individual.
#3. The Gospels lack independent attestation
A further difficulty for assessing the gospels’ reliability is that they contain many events that are not independently corroborated. The historian is therefore left to work out whether any particular event is plausible – but doing so is not always clear cut.
Consider the custom of Pilate to release a prisoner at Passover. There is no independent attestation of this custom, and it seems highly unlikely for a number of reasons. Not only does it run up against Pilate’s cruel character in independent sources, it makes little sense to release an insurrectionist at one of the most politically fraught festivals of the year. That the rebel is called Barabbas (‘Son of a Father’) – even Jesus Barabbas in some manuscripts – seems to be the icing on the literary cake.
On the other hand, the event does not stretch the bounds of belief. For all his cruelty, we know that Pilate sometimes did make concessions to crowds. He was not stupid. If he knew that the crowds had turned against Jesus, he might have made a concession to them (whatever the reliability of the details.) Might this event be historical?
Or take another example: Jesus’ burial by Joseph of Arimathea. For some scholars, it seems quite clear that the story of burial is a fiction. It is more likely that Jesus was buried in a common grave than that Joseph – a member of the very group that condemned him to death – used his private wealth to honour a convicted criminal. That Paul mentions only that Jesus was ‘buried’ offers little confirmation of the tomb.
I understand this. Given that many crucifixion victims were not buried, it makes sense that Jesus could have met the same fate. But of course, the data we have suggest that he was an exception – and he was not the only exception. The heel-bone of a crucifixion victim named Jonathan has been preserved, precisely because he was buried in a tomb. There is also that obscurest of name, Joseph of Arimathea. Was it preserved precisely because he had done something exceptional for Jesus?
Perhaps one might think I am being too generous to the gospels. There are certainly events that are singularly attested and stretch credulity. A classical example is Matthew’s mass resurrection (27:52-53), where many saints rise from their tombs and come into Jerusalem. The problem is that this most marvellous of miracles is not once found in any other source. For this reason, quite apart from its miraculous nature, even many conservative-minded scholars explain it away as a literary device.
It should be noted, however, that historical biographies also contain events which lack independent attestation and are judged unreliable. This does not necessarily invalidate the overall portrait of the biographer. A wider range of factors have to be taken into account to determine a biographer’s overall reliability. Yet for the reasons I have outlined above – and continue to do so below – this task is a complex one.
#4 The Gospels are intertextual and theological works
Another hurdle in assessing the gospels’ reliability is that they are highly intertextual works. More so than other biographies, they allude to other stories in the Hebrew Bible and the pagan world. For example, Mark draws upon the Elijah-Elisha cycle in the presentation of Jesus’ miracles, while Matthew portrays Jesus as a New Moses.6 Some scholars also see echoes of Homer and pagan mythology in the gospel texts.
What are we to do with this? The answer is not at all obvious. On the one hand, some scholars would see biblical allusions in the gospels as ‘prophecy historicised’.7 The early Christians looked to the Jewish Scriptures to make sense of Jesus’ life, and then they created narratives from those scriptures. For example, just as Elisha fed one hundred people with loaves and bread, Jesus went further and fed five thousand.
I don’t find this account implausible. Presumably, the reading of the scriptures at least influenced the memory of Jesus. For instance, if Jesus had done some Elisha-like things, such as being engaged in a ministry of healing, is it that unlikely that this typology generated other Elisha-like events in the absence of historical data? We have examples of oral tradition moving towards this kind of idealisation.8 There is also evidence that Jews in Jesus’ time crafted characters and stories from scripture.9
Yet there is an alternative to the idea that the gospels are prophecy historicised: that they are ‘tradition scripturalised’.10 That is, the gospel writers drew upon scriptural language – the language most familiar to them – in describing the past. On this reading, scriptural language may have distorted the past, but it does not mean that the evangelists were simply weaving their material whole-cloth from scriptural scripts.
How are we to assess these competing models? We can look at examples one-by-one, but there is no straightforward method to determine which is at play. To take just one example, some will point to Jesus’ birth in Bethlehem as a contrivance designed to fulfil the prophecy of Micah 5:2. The Messiah needed to be born in Bethlehem, and Luke’s awkward census places him there. For others, Micah 5:2 was not read messianically, and Matthew reads Micah 5 messianically because Jesus was born there.
This intertextual problem is just one part of a wider theological one: the gospels are not disinterested texts, trying to chronicle the life of their teacher. They are theological works, written to believers, and devoted to their master, who they believe is alive. As a result, the selection and presentation of their materials are coloured not only by Jesus as a figure of the past, but by the Spirit at work in the Church.
This is seen most of all in John, whose teachings diverge quite sharply from the Synoptics, and whose Jesus serves as a mouthpiece for the evangelist. This is not unusual in ancient biography. Yet it does mean that parsing the teaching of Jesus from his later interpreters is made all the more difficult (if not entirely without hope.)
#5 The Gospels develop and contradict one another
Another difficulty in assessing the gospels concerns the way they are re-written. In our first and shortest gospel, Mark, we find a bare-bones story of Jesus. It begins with Jesus’ baptism by John and ends with an anticipated resurrection appearance. Yet the later evangelists, clearly not quite content with this story, develop it in various ways: they add birth stories, prolegomena and resurrection appearances, and much else.
In the case of John, the overall portrait of Jesus which emerges is quite unlike the earlier Synoptics. In John, Jesus does not speak in parables, does not perform exorcisms, and does not preach the Kingdom of God. The expectation of an imminent in-breaking of God in history is substituted with eternal life known here and now. John also contains a vast amount of material which is absent from the earlier sources. This raises the question: if John was simply narrating history, how was all of it missed?
Yet the developments within the synoptic gospels also raise questions. Consider the story of Jesus walking on water. In Mark, this story ends with the disciples’ hardness of heart, because they don’t understand him. This is a thread running through the gospel. In Matthew, however, the scene ends on a more positive note. Peter initially walks out on the water to Jesus, and the disciples worship Jesus as the Son of God.
How are to assess the development of this story – and others like it? Has Matthew drawn upon his own superior sources to narrate this fuller episode? Or is it more likely that he has freely adapted Mark’s version, finding it unsatisfactory that Mark’s episode ends with the disciples’ lack of faith? One can easily conjecture reasons for why Matthew would want to change Mark’s ending, but it is difficult to say for sure.
#6 The Gospels contain fantastical material
Here we come to the elephant in the room: the gospels contain much material which, to audiences ancient and modern, has a mythic or supernatural character. By some estimates, Mark alone narrates eighteen miracles, more than any comparable narrative in the Graeco-Roman world. What are we to do with this?
It is widely believed that there is good evidence that Jesus was known in his own time as a healer and exorcist. These conclusions are broadly accepted, for healings and exorcisms are attested cross-culturally, and our evidence is diverse. Some gospel material even points to Jesus cultivating a reputation as healer among his enemies.
The difficulty is that there are other events which are less widely attested: a virgin birth, transfiguration, miracles of extraordinary power over nature, and of course the resurrection. According to the historical principle of analogy – that our judgement of the past is based to some extent on our experience of the present – it is less clear to know what to do with these events, which are almost by definition rarities.
I have written on how the gospels’ infancy stories reflect the literary conventions of their time and face a number of historical challenges. I have also written on how much of the common apologetics for the resurrection is not wholly convincing.
Yet it must also be said that considerations of metaphysics come into play with all historiography – and differences of worldview inevitably affect the way one sees an event like the resurrection. Arguably, the Resurrection is simply not the sort of event that could be verified by secular history. And perhaps that is not a problem.11
#7 The Gospel materials are not easily probed
Our seventh difficulty in assessing the Gospels’ historicity is that we do not have good tools to recover their sources. In the past, it was often thought that a meticulous analysis of the ‘forms’ (sub-genres) and sources which comprise the gospels would allow us to peek behind the gospels to recover some of the earlier materials.
Yet today, these recovery projects are often seen as hopeful at best. It is no longer believed that by working back from putative laws governing the oral tradition that we can recover the earliest ‘form’ of a text. Meanwhile, hypothetical written sources such as a pre-Markan passion narrative or detailed constructions of the ‘Q’ source behind Matthew and Luke are themselves facing an increasing number of detractors.
There is also a problem with the material we do have. The building blocks of the gospels are chreiai, the ‘anecdotes’ we find in many ancient lives. Yet anecdotes are often wandering and their origins indeterminate. Where did these anecdotes come from and what were they doing prior to their inclusion in the gospels?
To help respond to this problem, scholars have formulated a number of ‘criteria of authenticity’, through which the gospel materials are sifted. Yet as I have outlined in a previous post, these criteria have not made Jesus research more objective, nor produced anything like a consensus. The oral period remains mysterious, and the ‘criteria’ devised to mine it are increasingly abandoned by scholars.
Are the Gospels Historically Reliable?
In asking whether the gospels are historically reliable, it is important to recognise that we are asking a particular question. We have not asked whether the gospels are true. They can be true in ways that do not satisfy our contemporary historical interests – for example, in expressing profound theological or mythological truths.
Insofar as historical truth is concerned, however, I have suggested that modern historians are faced with a set of challenges. These can be distilled into a single point: the gospels are so difficult to study because evidence about them is underdetermined. That is, one can formulate several theories which plausibly account for their data.
In the philosophy of science, the ‘underdetermination of the theory by the evidence’ is a familiar notion. It may be wise to take it to heart when studying the New Testament, too.12 Like putting together a jigsaw with many of the pieces no longer in the box, there will never be a completely satisfying solution to many questions we pose about the gospels. But if we are to make progress, we must recognise how little we know.
Thank you for reading!
This piece was an updated version of a piece I wrote last year. If you enjoyed it, you might like some of my other pieces which relate to the gospels’ historicity:
I would love to know what topics you would like me to address on Behind the Gospels. If you have any ideas, drop them in the comments below!
Diarmaid MacCulloch, A History of Christianity: The First Three Thousand Years (London: Allen Lane, 2009).
For the argument that John is a work of disguised authorship, see Hugo Mendez, The Gospel of John: A New History (Oxford: Oxford University Press, 2025).
See Plutarch, Life of Cicero, ed and trans. J.I. Moles (London: Oxbrow, 1988), 46-53.
Tomas Hägg, The Art of Biography in Antiquity (Cambridge: Cambridge University Press, 2012), 3-4.
See Christopher Pelling, Plutarch: Eighteen Studies (Swansea: Classical Press of Wales, 2002); Michael R. Licona, Jesus, Contradicted: Why the Gospels Tell The Same Story Differently (Grand Rapids, MI: Zondervan, 2024).
Adam Winn, Mark and the Elijah-Elisha Narrative: Considering the Practice of Greco-Roman Imitation in the Search for Markan Source Material (Eugene, OR: Pickwick Publications, 2010); See Dale C. Allison Jr., The New Moses: A Matthean Typology (Eugene, OR: Wipf & Stock, 2013).
See, for example, John Dominic Crossan, The Cross that Spoke: The Origins of the Passion Narrative (San Francisco: Harper & Row, 1988).
See Eric Eve, Behind the Gospels: Understanding the Oral Tradition (Minneapolis: Fortress Press, 2014), passim.
See Nathanael Vette, Writing with Scripture: Scripturalized Narrative in the Gospel of Mark, LNTS 670 (London: T&T Clark, 2022).
Mark Goodacre, ‘Prophecy Historicized or Tradition Scripturalized? Reflections on the Origins of the Passion Narrative’ in The New Testament and the Church: Essays in Honour of John Muddiman, eds. John Barton & Peter Groves, LNTS 532 (London: T&T Clark, 2016), 37-51; idem, ‘Scripturalization in Mark’s Crucifixion Narrative’ in The Trial and Death of Jesus: Essays on the Passion Narrative in Mark, ed. Geert van Oyen and Thomas Shepherd (Leuven: Peeters, 2006), 33-47.
Jonathan Rowlands, The Metaphysics of Historical Jesus ResearchA Prolegomenon to a Future Quest for the Historical Jesus (London: Routledge, 2023).
See Dale C. Allison Jr., The Historical Christ and the Theological Jesus (Grand Rapids, MI: Eerdmans, 2009).
The gospels, like much other narrative in the Bible, are history with a figurative slant for a spiritual purpose. They will not yield up their deepest secrets to straightforward analysis (Prov 1:6). After all, we are told that the pure light in which God dwells is swathed in darkness (Ps 97:2). I like to characterize each gospel pithily with a single word: Matthew, “teaching” (5:2); Mark, “mystery” (4:11); Luke, “history” (3:1); John, “testimony” (19:35). And these four elements are contained in various mixtures in the rest of the NT. But to characterize Luke, for example, as “history” is to say it is quasi-historical, not history in the pedestrian sense.
A deep dive into the geographical designations, descriptions, puzzles, etc. in the gospels would be interesting, unless you have already done it. I know it has been touched on.
Hi John. This is really an excellent article. I would like to have your opinion on "Doctor Luke". Isn't ture that according to later church tradition, the author of Acts is a companion of Paul named “Luke” (a doctor moonlighting as both a missionary and a historian) but he never makes such claims. This is the Orthodox tradition and there is no multiple attestation to Paul's letters.
The author cleverly switches to the first-person plural (“we”) to indicate that he was some kind of fellow traveler with Paul. But what can we securely conclude from this? It could be a purely literary way to elevate his credibility. He never explains why exactly he switches back and forth from the third to the first person. I think it's a big problem. If he was a companion of Paul, he wouldn't leave it as an open debate. He would have declared his finest quality (being Paul's companion) to the world with a loudspeaker.