The Historical Jesus

The question behind this lesson is narrow and should be stated narrowly. It is not whether Jesus is Lord. It is: what can be established about Jesus of Nazareth by the ordinary methods historians use on any figure from antiquity, using no assumption a non-believer would refuse? That is a deliberately limited question, and both its friends and its critics sometimes forget the limits.

It begins with Hermann Samuel Reimarus, whose manuscripts Lessing published anonymously between 1774 and 1778. Reimarus argued that Jesus was a failed Jewish political messiah and that the disciples invented the resurrection after his death. The argument is not much read now, but the move was decisive: for the first time the Jesus of history was set against the Christ of the church as a matter of method. David Friedrich Strauss followed in 1835 with Das Leben Jesu, introducing myth as a category that was neither fact nor fraud, and lost his academic career for it.

Albert Schweitzer closed the first phase in 1906 with the book known in English as The Quest of the Historical Jesus. His survey showed that the liberal lives of Jesus had each produced a Jesus resembling their authors, and his own reconstruction offered a figure nobody wanted: an apocalyptic prophet who expected the end of the age within his own generation and was wrong. Schweitzer's verdict, that the historical Jesus comes to us as one unknown, ended the enterprise for a generation. Rudolf Bultmann opened his 1926 book with the statement that we can now know almost nothing about the life and personality of Jesus, and turned his attention to the preaching of the early church instead.

Ernst Käsemann reopened it in a 1953 lecture, arguing that a faith wholly indifferent to the historical figure risked becoming myth in the bad sense. That second phase, the so-called New Quest, was pursued mostly by Bultmann's own students. The phase we are still in was named the Third Quest by N. T. Wright in 1988, and its distinguishing mark is the insistence on placing Jesus firmly inside first-century Judaism rather than in contrast to it. Its major figures disagree profoundly with one another. E. P. Sanders and Dale Allison and Paula Fredriksen and Wright and John Meier and Geza Vermes produce recognisably different portraits from broadly the same evidence.

Running alongside it was the Jesus Seminar, founded by Robert Funk in 1985, which voted on sayings with coloured beads and published The Five Gospels in 1993, concluding that a small minority of the sayings attributed to Jesus went back to him. Its methods were widely criticised, including by scholars with no stake in orthodoxy, chiefly for the composition of its voting membership and for treating a vote as a finding. Its media presence was out of all proportion to its standing in the field, and students should know both facts.

For most of the twentieth century the tools of the trade were criteria of authenticity, applied to individual sayings and episodes to sort what goes back to Jesus from what the church produced. John Meier's list in the first volume of A Marginal Jew is the standard statement: multiple attestation, discontinuity or dissimilarity, embarrassment, coherence, and rejection and execution.

Each has an obvious logic. Multiple attestation: a saying found in independent sources is less likely to be one source's invention. Embarrassment: material awkward for the later church, such as the baptism by John or the family calling him mad, is unlikely to have been made up by that church. Dissimilarity: material unlike both contemporary Judaism and the later church cannot be explained as borrowed from either. Rejection and execution: whatever reconstruction you produce has to explain why this man was crucified.

The criteria are now under serious and largely successful attack, and a student should know this because it changes the shape of every argument built on them. The heaviest charge falls on dissimilarity. By construction it yields a Jesus who is unlike his own Jewish setting and unlike the movement he founded, which is a historically absurd person; and a saying failing the criterion is not thereby shown to be inauthentic, only unproven. Multiple attestation depends on a settled answer to the synoptic problem in order to know which sources are independent, and that answer is itself disputed. Embarrassment requires us to know what embarrassed communities we cannot interview. The 2012 collection edited by Chris Keith and Anthony Le Donne, Jesus, Criteria, and the Demise of Authenticity, gathered the case, and the title is not a joke.

The deeper objection is about what the criteria assume. They assume that beneath the interpreted tradition lies a layer of uninterpreted fact that can be recovered by peeling. Work on social memory, much of it by the same scholars, argues that memory is interpretive from the first moment, that there is no unvarnished layer, and that the question is not what the church added but how it remembered. Dale Allison, whose scepticism about the criteria is as sharp as anyone's, argues in Constructing Jesus that we should stop adjudicating individual units and look instead for the recurrent patterns that run across the whole tradition, on the ground that a general impression repeated everywhere is likelier to be reliable than any one saying.

The consequence for your teaching is practical. Arguments that begin 'this passes the criterion of embarrassment, therefore it is historical' are weaker than they were twenty years ago, and students who use them in a graduate seminar will be corrected. Arguments of the form 'this pattern recurs across sources that do not depend on one another' remain strong.

Take them one at a time, and be exact about the date and the weight of each.

Tacitus, Annals 15.44, written around AD 115 to 120, describing Nero's scapegoating of Christians after the fire of 64: the founder of the name, Christus, had been executed in the reign of Tiberius by sentence of the procurator Pontius Pilatus. This is the strongest non-Christian attestation of the execution. Two honest qualifications. Tacitus calls Pilate procurator, while the inscription found at Caesarea Maritima in 1961 gives his actual title as prefect; this may show Tacitus using the title current in his own day, which was common practice, or may suggest he was not working from an official record. And we cannot show that Tacitus had a source independent of what Christians were saying, though his contempt for them makes borrowing from them without comment less likely. The passage is preserved in a single eleventh-century manuscript, in which the word for Christians appears to have been altered from an original spelling with an e.

Suetonius, Divus Claudius 25.4, written around AD 120: Claudius expelled the Jews from Rome because they were constantly making disturbances at the instigation of Chrestus. Chrestus was a common slave name, and whether this refers to Christ is genuinely uncertain. Acts 18:2 independently records an expulsion of Jews from Rome under Claudius, which is the reason many take the two notices together. Do not build anything heavy on it.

Pliny the Younger, Letters 10.96, around AD 112, reporting to Trajan that Christians met before dawn and sang a hymn to Christ as to a god. This is first-rate evidence for what Christians did and believed by that date, and no evidence at all for the life of Jesus. Keep the distinction.

Then the weaker items, which are often overstated. Lucian of Samosata, writing satire around AD 165, mentions the sophist the Christians worship, crucified in Palestine. Mara bar Serapion's Syriac letter refers to the wise king of the Jews, and its date is disputed across two centuries. The Babylonian Talmud at Sanhedrin 43a says Yeshu was hanged on the eve of Passover and had practised sorcery and led Israel astray; it is late and may be reacting to Christian claims rather than preserving independent memory. Thallus is sometimes cited for a note on darkness at the crucifixion, but we have him only through a ninth-century writer quoting a third-century writer quoting him, and the chain will not bear weight. A student who cites these as though they were early independent testimony will lose the room.

Josephus, a Jewish priest and historian writing in Rome, finished the Antiquities of the Jews around AD 93. Two passages mention Jesus, and they are in very different conditions. The shorter one, Antiquities 20.200, is incidental: Josephus reports that the high priest Ananus convened a council and brought before it the brother of Jesus who was called Christ, James by name, and had him stoned. The phrase is unadorned, the Greek is Josephus' own, the point of the passage is Ananus' misconduct rather than James, and no Christian copyist would have written such a flat description. This passage is accepted as substantially authentic by the large majority of specialists, and it is the more valuable of the two.

The longer one, Antiquities 18.63-64, is the Testimonium Flavianum, and here is the problem. As it stands in every surviving Greek manuscript, it calls Jesus a wise man, says that he was perhaps more than a man, that he was the Christ, that he did astonishing deeds, that Pilate condemned him at the suggestion of the leading men among us, and that he appeared to his followers alive again on the third day as the divine prophets had foretold. A Jewish priest who remained a Jew did not write that he was the Christ. Origen, writing in the 240s and using Josephus, states twice that Josephus did not believe in Jesus as the Christ, which is hard to square with Origen having read this paragraph in this form. The earliest writer to quote the passage as we have it is Eusebius, in the early fourth century, and the oldest Greek manuscripts of the Antiquities are medieval.

Three positions exist. A very small number still defend the whole paragraph as authentic. A real minority, including Ken Olson, argue that the whole thing is a Christian composition, possibly by Eusebius himself, and point to vocabulary the passage shares with Eusebius' other writings. The majority hold that Josephus wrote something about Jesus at this point, and that a Christian hand later sharpened it: remove the three Christian clauses and what remains reads like Josephus and fits its context. A tenth-century Arabic version preserved in the chronicle of Agapius of Hierapolis, published with fanfare by Shlomo Pines in 1971, appeared to give a less Christianised form and was widely cited as support for the partial-interpolation view; Alice Whealey has since argued that the Arabic derives from Syriac translations of Eusebius rather than from an independent recension, which substantially weakens that particular argument. This is a live dispute, not a settled one.

Now be exact about what the passage can carry, on the majority reconstruction, which is the most favourable reasonable reading. It supports the following: that a Jewish historian writing around AD 93 knew of Jesus as a teacher with a following among Jews and Greeks, that he was reputed a doer of astonishing deeds (the course's own rendering of the Greek), that he was condemned to the cross by Pilate at the instigation of leading Jewish men, and that the movement named after him had not died out by Josephus' day. That is worth having, and it corroborates the outline of the Gospel narrative at exactly the points a Roman-era historian would be in a position to know.

It does not support the resurrection. The clause reporting that he appeared alive on the third day as the prophets had foretold is one of the clauses removed on every standard reconstruction, along with the statement that he was the Christ and the suggestion that he was more than a man. Anyone who cites the Testimonium as extra-biblical evidence for the resurrection is either unaware of the textual situation or is relying on an audience being unaware of it. The same goes for citing it as evidence that a non-Christian confessed Jesus as Messiah. Students should hear this from you first, in a classroom, rather than later from someone who will use it to suggest that everything else they were taught was handled the same way.

Here is the sceptical case, stated as its best defenders state it and with nothing pulled. Every extended account of Jesus comes from inside the movement that worshipped him. The earliest of them, Mark, is usually dated some thirty-five years after the events, and the latest between sixty and eighty years after, all in Greek, all outside Judea so far as we can tell, all by authors whose names are attached by later tradition rather than by the documents themselves. Between the events and the writing lies a period of oral transmission in a community with strong reasons to shape what it passed on, and the study of memory shows that even honest transmission reshapes. The criteria that once claimed to reach behind that period have been largely dismantled. The non-Christian sources are late, brief, and possibly dependent on Christian informants; the best of them, Tacitus, tells us only that a man was executed. And the places where the sources are theologically loaded are exactly the places where believers want historical confirmation, which is the worst possible correlation for the historian. On this view, a modest core survives and the rest is confession.

Here is the confident case, also at full strength. The sources are early by the standards of ancient history: Paul's letters are twenty to twenty-five years from the events, and they embed material older than themselves, including a formula in 1 Corinthians 15:3-7 that most scholars of any persuasion date within a few years of the crucifixion. Paul is not a distant figure; he spent fifteen days with Cephas and met James the Lord's brother, as he says in Galatians 1:18-19, and he knew the Jerusalem leadership well enough to quarrel with them. Multiple streams that do not depend on one another, in whatever way one resolves the synoptic problem, converge on the same outline. The tradition preserves material that damages its own transmitters. There is an Aramaic substratum visible under the Greek, and place and custom details that fit pre-70 Palestine and would have been hard to reconstruct afterwards. And the existence, shape and speed of the movement itself is a historical datum requiring explanation. Compared to what we know about most ancient figures, this is an unusually rich body of evidence unusually close to the events.

Now what almost everyone grants, whatever their faith. That Jesus was a Jew from Galilee. That he was baptised by John. That he gathered disciples and taught in parables and aphorisms about the kingdom of God. That he had a reputation as a healer and exorcist. That he confined his activity largely to Jews. That he caused a disturbance in the temple. That he was crucified under Pontius Pilate in Jerusalem, at or near Passover, with a charge board naming him king of the Jews. That after his death his followers said they had seen him and that a movement continued, and that it was persecuted and rapidly spread. E. P. Sanders assembled a list of this kind and called the items almost indisputable. It is a short list, and it is not nothing.

The claim that Jesus never existed sits outside this range, and it is worth saying why plainly rather than dismissively. Richard Carrier has argued the case at book length with Bayesian machinery, and Robert Price has argued it too. The reason it has gained almost no traction is not that scholars of religion are all believers. Bart Ehrman, who is not a believer and who spends most of his career arguing against conservative conclusions, wrote a whole book in 2012 rejecting it. The sharpest opposition to mythicism comes from people with no confessional reason to oppose it, and that is the relevant fact about the position, rather than any count of how many people hold it.

What should you do with all this? Not reach a verdict for your students. Show them the shape of the argument: which claims rest on sources everyone accepts, which rest on a contested reconstruction, and which rest on a prior judgment about what sort of events are possible. Those are three different kinds of claim and they fail in three different ways. A student who can sort a statement into the right category is doing history. A student who has memorised a set of conclusions is doing something else, and it will not survive the first serious challenge.