Did Jesus rise from the dead?

The short version

1. The core data do not depend on trusting the Gospels.

   Paul’s own letters, written about 20 years after the crucifixion, pass on a list of appearance witnesses that critical scholars date to within about 2-5 years of it. Paul had met two of those witnesses in person.

   The evidence ↓
2. Two more facts are probable but genuinely disputed.

   The burial in a known tomb and the women finding it empty. We include them at a modest weight and show what happens without them.

   The evidence ↓
3. Every explanation, scored against every fact.

   Legend, theft, the wrong tomb, swoon and the rest each fail several well-established facts. The real contest is the resurrection against Dale Allison’s careful combination: grief visions, an end-of-the-age reading of them, and social spread.

   The matrix ↓
4. How strong: about 500 to 1, with a wide range.

   The data are about 500 times more probable if Jesus rose than on the best natural mixture, if our judgement calls are independent; if they move together, as one reader’s usually do, about 70-450. At the sceptical end of our own ranges the factor falls below 1; at the apologist end it reaches 10⁷. Not 10⁴⁴.

   The sensitivity ↓
5. Your prior decides it, and the prior comes from Tier I.

   If there is no God, no evidence can make this probable. If a God of the theist’s kind is reasonably likely, a factor of 500 makes the resurrection more probable than not. The reverse step is weak: even granted in full, the history does far less for the existence of God than for the event.

   Why the prior decides ↓

The question

Here is what is tested. Call the data about Jesus’ death and its aftermath E. Is E better explained by R, that God raised Jesus bodily from the dead and he appeared to his followers, or by the best natural alternative, not-R? Not-R is not one hypothesis. It is a mixture of every natural explanation anyone has proposed, each with the weight its own plausibility earns.

We assume nothing about inspiration. The Gospels and Acts are treated as ancient sources of varying reliability (III.1 weighs them). The primary evidence is Paul’s undisputed letters, 1 Corinthians and Galatians, written about AD 48-55.

Two terms. A likelihood ratio (LR) is how many times more probable the data are on one hypothesis than on the other: 1 means the data do not discriminate, above 1 favours R, below 1 favours not-R. A prior is how probable a hypothesis was before looking at these data.

Rules of evidence

These were fixed before any scoring.

1. Three tiers of data. Tier A, established: affirmed by the large majority of critical scholars, including non-Christians, and supported by Paul’s own letters or by non-Christian sources. Tier B, probable: majority support, with serious dissent by named scholars. Tier C, Gospel-only detail (touching, eating, the guard): never used in the core case, because R would then be partly assumed through the reliability of the very stories in question.
2. Predicting is not accommodating. A hypothesis predicts a fact if it would make it likely before seeing it. If it fits only after an extra assumption, that assumption is scored for its own improbability. This applies to R too: the probability that a persecutor converts, given R, is not 1.
3. Dependence. Facts that flow from one underlying event (the disciples’ belief, their preaching, their suffering, their worship of Jesus) share one factor, not one each. Thirteen disciples who talked to each other are one group report with some internal confirmation, not thirteen independent witnesses.
4. The null space counts. What a cherry-picker would leave out is scored as evidence, with a factor below 1 where it fits not-R better: no appearance to opponents or the public, visionary features in the stories, divergent narratives, visible legendary growth, and parallel claims in other movements.
5. Consensus is evidence about the evidence, not evidence. Head-counts of scholars are reported, but they do not enter the calculation.
6. The prior is kept separate. The probability of R before these data is P(God) × P(God raises this man | God). Hume’s maxim is honoured as a statement about the prior, not used as a veto.
7. Excluded: the Shroud of Turin (see the appendix), later martyr legends after about AD 150, and personal religious experience.

The evidence

What you can trust without trusting the Gospels

Jesus was crucified under Pontius Pilate. All four Gospels and Paul say so, and so does the Roman historian Tacitus, writing about AD 116, who calls the movement a “superstition”.¹ Josephus mentions Jesus’ execution too, in a passage whose core is probably authentic but which a Christian copyist expanded.² No serious historian doubts the crucifixion.

The key text is a tradition Paul passed on to the church in Corinth around AD 53-55:

“For I delivered to you first of all that which I also 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 he appeared to over five hundred brothers at once, most of whom remain until now, but some have also fallen asleep. Then he appeared to James, then to all the apostles, and last of all, as to the child born at the wrong time, he appeared to me also.”1 Corinthians 15:3-8, WEB³

“Delivered” and “received” are the technical terms for handing on a fixed tradition. Paul received it before he founded the Corinthian church, and he tells us where he could check it:

“Then after three years I went up to Jerusalem to visit Peter, and stayed with him fifteen days. But of the other apostles I saw no one except James, the Lord’s brother.”Galatians 1:18-19, WEB

Peter (Cephas) and James are both named in the list. And Paul says his message is theirs: “Whether then it is I or they, so we preach, and so you believed” (1 Corinthians 15:11, WEB). Critical scholars, including the atheist New Testament scholar Gerd Lüdemann, date the tradition to within a few years of the crucifixion.⁴ Bart Ehrman treats it as early pre-Pauline tradition and accepts that some followers had visions and believed Jesus raised soon after his death.⁵

Burial and the empty tomb: probable, and disputed

The doubters. John Dominic Crossan argues that crucified bodies were usually left to scavengers or put in common graves, so the burial story is a later hope.⁷ Ehrman argues that Jesus probably did not receive a decent burial from Joseph, an otherwise unknown figure.⁵

The replies. Craig Evans points to Deuteronomy 21:23 (“you shall surely bury him the same day”), to Josephus on Jews burying the crucified before sunset, and to the ossuary of a crucified man found in Jerusalem in 1968, which shows that a crucified Jew could receive family burial.^{8, 2} Jodi Magness holds that the Gospel burial accounts are largely consistent with the archaeology and with Jewish law.⁹ Allison, no apologist, argues for a proper burial and accepts Joseph’s involvement.⁶ In his 2005 book he called the empty tomb historically likely only with great hesitation.¹⁰

Against the empty tomb. Peter Kirby argues that every empty-tomb report depends on Mark, that Mark’s account has fictional traits, that Paul is silent, and that Roman practice points away from a discovered tomb.¹¹ Jeffery Jay Lowder proposes that Joseph stored the body temporarily and then moved it.¹²

For it. The claim was proclaimed in the city where the tomb was. All four Gospels make women the first witnesses, when women’s testimony carried less weight. And the earliest counter-claim, “Say that his disciples came by night and stole him away while we slept” (Matthew 28:13, WEB), presupposes an empty tomb; Justin repeats it about AD 160.¹³ That counter-claim reaches us through Matthew, not through a hostile document.

The appearances, and the grief-vision data

Paul’s list has individual appearances (Cephas, James, Paul) and group appearances (the Twelve, more than five hundred “at once”, all the apostles). The Gospels add Mary Magdalene and the women, Emmaus, the Eleven, and the lakeside.

The strongest sceptical point is that seeing the dead is common. Among the widowed in mid-Wales, “almost half the people interviewed had hallucinations or illusions of the dead spouse”.¹⁵ In two nursing homes, 28 of 46 widows (61%) reported such experiences.¹⁶ A 2015 review puts the prevalence among the widowed “ranging from 30% to 60%”.¹⁷ A sceptic needs only one or two such experiences among Jesus’ close followers to start the process.

What these studies do not show: they are mostly experiences of presence or brief perceptions, by individuals, of a spouse, read as the dead person “still with me”. None reports bereaved people concluding that the dead had been raised bodily, in the end-of-the-age sense, or groups converging on one shared appearance. Allison, who has collected reports of collective apparitions, treats them as real data, which is why we do not claim that hallucinations are never collective. He also writes: “I know of no close phenomenological parallel to the series of likely events as a whole.”⁶

Paul and James: the two outsiders

Paul writes against his own interest: “beyond measure I persecuted the assembly of God and ravaged it” (Galatians 1:13, WEB), and “Haven’t I seen Jesus Christ, our Lord?” (1 Corinthians 9:1, WEB). But he describes his experience as God choosing “to reveal his Son in me” (Galatians 1:16, WEB), and in Acts there is light and a voice that his companions do not fully share. So Paul’s experience was visionary in form. That is why Lüdemann takes it as the model for all the appearances, and why apologists stress that a hostile persecutor is the least likely candidate for a grief vision. Both points are fair.

James. “For even his brothers didn’t believe in him” (John 7:5, WEB). Yet Jesus appeared to James (1 Corinthians 15:7), he became a leader of the Jerusalem church (Galatians 1:19; 2:9), and he was stoned about AD 62.² The weak link: Paul says only that Jesus appeared to James, not that the appearance converted him. Family members joining a movement led by a dead relative is not unusual, and Acts 1:14 already places Jesus’ mother and brothers with the disciples.

Early worship and the cost of witness

Prayer to Jesus, hymns and the invocation of his name appear in Paul’s earliest letters. An Aramaic prayer, marana tha (“Come, Lord!”, 1 Corinthians 16:22), survives untranslated in a Greek letter. Larry Hurtado describes this devotion as a mutation inside Jewish monotheism, not a gradual pagan development, and Richard Bauckham argues Jesus was included in the identity of Israel’s God early.^{18, 19} This confirms that the Easter belief was extraordinary and immediate. But worship follows from the belief whether the belief was true or not, so it adds little: a factor of 1 to 2.

The witnesses suffered for their claim. Sean McDowell’s study of the apostles’ deaths grades each tradition rather than assuming them all. He concludes that their willingness to die “establishes their unmistakable sincerity”, and that the facts “prove neither the truth of the Resurrection in particular nor Christianity as a whole”.²⁰ Suffering shows sincerity, not truth. It defeats the conspiracy hypothesis; it does nothing against the vision hypotheses.

The evidence against: what a cherry-picker leaves out

• No public appearance. No appearance is reported to Pilate, Caiaphas or the crowds; Acts says plainly the risen Jesus was shown “not to all the people, but to witnesses who were chosen before by God” (Acts 10:41, WEB). A God who wished to be known might have done otherwise; visions come to the predisposed. Factor 0.3-0.8.
• Visionary features. Jesus appears and vanishes suddenly, is not recognised at first, and “When they saw him, they bowed down to him; but some doubted” (Matthew 28:17, WEB). Allison: the appearances “had uncanny features that suit visionary experiences better than everyday seeing”.⁶ Factor 0.4-0.9.
• Divergent stories. Galilee or Jerusalem; how many women and angels. Mark’s original ending has no appearance at all: “They said nothing to anyone; for they were afraid” (Mark 16:8, WEB). Factor 0.6-0.95.
• Legend at the edges. The saints who rose and “entered into the holy city and appeared to many” (Matthew 27:53, WEB); the guard; the Gospel of Peter’s talking cross. Legends grow around true events too, so the factor is mild: 0.7-1.0.

Together these four multiply to a median of 0.19: the counter-evidence alone divides the case by about 5.¹⁴

The analysis

Every explanation against every fact

Each hypothesis is scored on how well it predicts each fact, from ++ predicts it naturally, through 0 neutral, to −− contradicts it or needs a seriously ad hoc assumption. The result is simple to state:

• The single-cause explanations (legend, conspiracy, removal, wrong tomb, swoon, substitution, cognitive dissonance) each fail several Tier A facts. Legend fails on the creed’s date and Paul’s first-hand claims; conspiracy fails on sincerity and on Paul; the tomb theories explain only the tomb.
• The live contest is R against Allison’s combination (H3, or H3 plus a moved body or legend). R fits the positive facts better. H3 fits the counter-evidence better: private appearances, visionary features, divergent stories. The question is how much H3’s chain of steps costs, and that is what the likelihood ratio computes.

How strong: the likelihood ratio

We build the probability of the core datum on the best natural account from its steps, using ranges rather than single numbers, and then multiply in the other facts as separate clusters, each conditional on the ones before it. The core datum (A) is the early, sincere, group-attested proclamation that Jesus was raised and appeared. On not-R it needs four things to happen:

1. at least one vivid visual experience among 5-15 close followers (median 0.95, from the bereavement data);
2. that experience proclaimed as bodily “raised”, not as a ghost, an angel or a soul exalted to heaven (0.02-0.5);
3. spread into group claims fixed in a creed within about 2-5 years (0.1-0.8);
4. persistence under hostility, given sincerity (0.5-0.95).

On R the same datum has probability 0.6-0.95 (R predicts appearances, but not necessarily to groups or in this form). The result for A alone: a median likelihood ratio of 45 (5-95%: 7 to 300).¹⁴

The figure we carry forward is about 500. The median of the draws, 1,100, is not itself a likelihood ratio. The likelihood ratio implied by uncertain judgements is an average of how probable the data are on each side, and it comes out at about 300 if the uncertainty in James, Paul, the tomb, worship and suffering concerns the rivals, and about 600 if it is split evenly. 500 lies between. The range is about 1 to 20,000.

What drives the spread. The single input that matters most is step 2: would grief visions be read as “resurrection”? (Its correlation with the result is −0.53.) Next comes Paul (+0.49), then the tomb, the spread to groups, and James (about ±0.35 each). So the scholarly arguments that decide this are N. T. Wright against Allison on what “resurrection” meant to a first-century Jew, and the explanation of Paul. Not the head-count.¹⁴

0 of 6:  1/(n+2) = 0.125   Jeffreys 0.5/(n+1) = 0.071
0 of 8:  1/(n+2) = 0.100   Jeffreys 0.5/(n+1) = 0.056
0 of 10: 1/(n+2) = 0.083   Jeffreys 0.5/(n+1) = 0.045

Sensitivity: the sceptic, the middle and the apologist

The same facts, scored three ways. Each stance below grants all the Tier A data; they differ only in the judgment calls. The first sets every input at the sceptical end of our own ranges, leaning the way Richard Carrier and Ehrman argue: visions would quite naturally be read as “raised”, and the counter-evidence weighs heavily. The middle inputs are Allison-type; the last sets every input at the apologist end, leaning Wright- and Habermas-type. The two ends are corners of our ranges, not other scholars’ own numbers.

Three things follow. First, the sceptical result is a defensible position, not an error: a well-informed critic who grants every Tier A fact can end near or below 1. And because none of our ranges lets James, Paul or the tomb count against the resurrection, a sceptic who thinks a vision explains Paul’s or James’s turn better than a real appearance would go lower still; our ranges cannot represent him. Second, the apologist’s 10⁷ needs every judgment to fall one way at once; we report it, but we do not use it. Third, the hinge is step a2. If a vision would almost certainly be read as “raised” (a2 = 0.5), the core factor falls to about 2-10; if almost never (a2 = 0.02), it rises to about 100-1,000. Read Wright and Allison on that question and you will know where you stand.

Why not 10⁴⁴, and why not Swinburne’s 97%

Two well-known Bayesian treatments give far stronger results. We do not use either, and the reasons are instructive.

Timothy and Lydia McGrew (2009) assign a Bayes factor of 10² to the women’s testimony, 10³ to each of thirteen male witnesses (the Twelve and James) and 10³ to Paul. In their words: “Sheer multiplication through gives a Bayes factor of 10⁴⁴, a weight of evidence that would be sufficient to overcome a prior probability (or rather improbability) of 10^-40 for R”.²⁸

• Independence. Treat the thirteen as one group report, as our rule 3 requires, and the product is 10² × 10³ × 10³ = 10⁸. So 36 of the 44 orders of magnitude come from multiplying thirteen people who lived, preached and suffered together as if they were strangers. The McGrews say so themselves: “in the case of the calculation for D the independence assumption almost certainly breaks down.” They argue that this need not weaken the case, depending on how dependence works under each hypothesis. We think clustering is the honest default, and the script lets you see the difference.
• The Gospels are assumed. Their calculation “is predicated on the assumption that in matters other than the explicit claims of miracles, the gospels and the book of Acts are generally reliable”. That is the question of III.1, and the details it lets in (touching, eating) are exactly our Tier C.
• Hindsight. For Paul they write that “on the assumption of R there is no difficulty whatsoever accounting for” his conversion. But R does not predict that a particular persecutor would convert; our rule 2 makes R pay for that too.
• It makes the prior irrelevant. With a factor of 10⁴⁴, the posterior comes out at 1 for every prior in our script’s grid, even P(God) = 0.001 with P(God raises this man | God) = 10^-11. Any number that makes the reader’s prior beliefs irrelevant should make a careful reader suspicious.¹⁴

Richard Swinburne (2003) reaches a posterior that his reviewer Otte summarizes as “somewhere around .97” for the resurrection of God incarnate.^{29, 30} His inputs, as Richard Otte’s review reports them: the probability that God exists is 1/2; the probability that God becomes incarnate, if he exists, is 1/2, so the prior is 1/4; the probability of the evidence if he did is put “provisionally at 1/10”, and if he did not, at 1/1000.

• The evidence term is modest. 1/10 against 1/1000 is a factor of 100, the same order as ours. The 97% comes from pairing that factor with a starting probability of 1/4. Start at 1/1000 instead and the same factor gives about 9%.
• His evidence already names Jesus. Swinburne’s evidence is “that Jesus and only Jesus satisfies the prior and posterior requirements for being God incarnate”. So his 1/10 is not the chance that God would raise this man; it is the chance of evidence that already singles him out. What compares with our numbers is his starting probability times his factor: 1/2 × 100 = 50, which corresponds here to P(God raises this man | God) × the likelihood ratio.³⁰
• The prior is where the dispute lies. P(God) = 1/2 is imported from Swinburne’s own natural theology; on this site that is the job of Tier I. Otte notes that Swinburne’s own principle of indifference would make the incarnation about 1/n, where n is the number of equally best acts God could do: “it is difficult to see why Swinburne concludes the probability of God becoming incarnate, if he exists, is 1/2 instead of 1/3, and many readers will assign an even lower probability”. Of the 1/1000: “one would have expected a stronger defense of the claim that this is about 1/1000.” And the argument “is based on an appeal to God not deceiving us, which will be unconvincing to many philosophers.”³⁰ Michael Martin puts that figure at 1/500 and calls Swinburne’s unrealistic.³¹
• So we never quote the 97% without its inputs. With them it says something true and useful: a factor of about 100, on a theist’s prior, gives a high posterior. That is our conclusion too, stated conditionally.

Why the answer depends on Tier I

Here is the step most treatments hide. The probability that God raised Jesus, before looking at the Easter evidence, is

P(R) = P(God) × P(God raises this man | God)
P(R | no God) = 0

If there is no God, the probability that God raised Jesus is zero, whatever the evidence. No likelihood ratio, however large, can move a zero. So this page cannot settle the question on its own. It can only tell you how far the history should move you from wherever Tier I leaves you.

P(God) is the question of Tier I: is there a Mind behind nature? (the beginning of the universe, fine-tuning, the origin of life, consciousness), weighed against the case against. We do not count Tier I again inside this page’s likelihood ratio.

P(God raises this man | God) is not the chance that a random corpse revives. The background includes what we know of Jesus before Easter: an apocalyptic prophet who spoke and acted with claimed divine authority and was executed as “king of the Jews”. Two anchors: if God raised one person at random from the roughly 117 billion humans ever born,³² the figure would be about 10^-11; that is a floor, because Jesus is not a random person. There is no agreed upper anchor: Swinburne’s inputs are not a value of this number (see above), and the columns below are values to try, not anyone’s estimate.

The table shows the probability that Jesus rose, given the evidence, for a likelihood ratio of 100 and of 1,000 (our central 500 lies between). Bold cells are above one half at 1,000.

Run the whole Monte Carlo range through instead of a single number, and with P(God) = 0.5 and P(God raises this man | God) = 10^-2 the posterior has a median of 0.85 (5-95%: 0.24 to 0.99). With P(God) = 0.1 and 10^-4 it has a median of 0.011.¹⁴

What the event can, and cannot, tell you about God

The likelihood ratio on this page is for the event: did Jesus rise, or is there a natural explanation? Running the argument the other way, from the event to God, is much weaker, and there is a simple reason. To count the resurrection as evidence that a God who acts exists, you need to know how likely such a God was to do this particular thing.

1 + P(God raises this man | God) × (L − 1)

To count as strong evidence for a God who acts (a factor of 20), the advance probability that God would vindicate this man, times the strength of the evidence, would have to reach about 40 against a God who never acts, or about 80 against no God. Cautious inputs give about 3 to 5. Skeptical theism, the view that God’s reasons are beyond us, does not help: applied evenly, it removes this support together with the cost of evil, and leaves the question at even odds.¹⁴

That is why this page’s conclusion runs one way. If you already have reason to think God exists, the history makes the resurrection probable. The history alone does little to give you that reason.

Objections

What this shows, and what it doesn’t

What we don’t claim

• “The resurrection has a Bayes factor of 10⁴⁴.” 36 of the 44 orders of magnitude come from treating thirteen testimonies as independent, which the McGrews themselves say “almost certainly breaks down”; the figure also assumes the Gospels and Acts are generally reliable.
• “Swinburne calculated a 97% probability.” He did, given P(God) = 1/2 and three other contestable inputs. Never without the inputs. Nor “Swinburne puts the chance that God would raise Jesus at 1 in 10”: his 1/10 is the chance of evidence that already singles Jesus out.
• “The resurrection is evidence that God exists.” Not without the condition. It is evidence for the event; its support for a God who acts is capped by how likely God was to do this, which is a judgement about God’s purposes (the cap).
• “Habermas polled thousands of scholars; 75% accept the empty tomb.” He catalogued publications (1,400 to 2,000+ by 2005), coded by himself. The 75% is of the publications that address the question, with no denominator and no published dataset. Say: in Habermas’s catalogue, about three in four publications that address the question favour the empty tomb.
• “All the apostles died as martyrs for their testimony.” McDowell rates four at the highest probability, Thomas at 7, Andrew at 6, seven at 5 (“as plausible as not”) and John as improbable.
• “People don’t die for a lie, so it’s true.” It rules out conscious fraud only. Sincere people die for false beliefs.
• “The creed dates to within months of the crucifixion.” The defensible claim is within a few years, about 2-5.
• “Over 500 witnesses who could be checked.” Paul says most were alive, but gives no names; Allison suggests the Corinthians “were almost as much in the dark as are we, unable to name or quiz any of those involved.”⁶
• “Group hallucinations are impossible.” Collective-apparition reports exist. Argue from the whole pattern instead.
• “Women’s testimony was inadmissible.” As a blanket rule, that overstates the sources. Josephus does write “let not the testimony of women be admitted” (Ant. 4.219, Whiston), but the rule was not absolute in practice. Say it carried less weight.
• “JAMA proved Jesus died.” Edwards, Gabel and Hosmer (1986) is an interpretive review built on the Gospel accounts: “Modern medical interpretation of the historical evidence indicates that Jesus was dead when taken down from the cross.”³⁷ Its asphyxia mechanism is disputed.³⁸ The swoon theory fails for historical reasons.
• “Josephus says Jesus appeared alive on the third day.” That clause sits in the part of the Testimonium widely judged to be a Christian addition.
• “Tacitus is enemy attestation of the resurrection.” He attests the execution and the movement’s spread, and calls it a superstition.
• “The Jewish authorities admitted the tomb was empty.” We have Matthew’s report, decades later, of a counter-claim. It shows the dispute presupposed an empty tomb; it is not a hostile document.
• “Only the resurrection explains the facts; every alternative is fatally flawed.” Allison’s combination is a serious rival. Say “best explains”, with the likelihood ratio.
• “No first-century Jew expected a Messiah who would die.” Say: a dying and rising Messiah was not a common expectation.
• Wright’s widely circulated line, “I cannot explain the rise of early Christianity unless Jesus rose again…”. We could not verify it; quote Wright only from his books.
• Chuck Colson’s “Watergate conspiracy” argument. It is an anecdote, not data.
• Any Shroud claim in the main argument: “carbon dating refuted”, “new tests date it to Christ’s time”, “a burst of radiation made the image”, “type AB blood proves it”, “no known process”. See below.
• “The image took a burst of several gigawatts, about 120 joules, more than all the world’s nuclear plants.” These figures misquote the Italian laboratory work and contradict each other. That work coloured linen with laser pulses; it measured nothing from a body, and its authors did not say the image was made that way.
• “The image is only 20-40 nanometres deep.” The published figure is coloration “very thin, down to 200 nm”, about ten times deeper (see the appendix).
• “Coherence radiation”, an “ontological phase transition”, an isotopic “ash” of exotic particles on the cloth, or a constant of nature “derived from the Resurrection”. None has any published basis. These are inventions, some of them by AI systems, and the constant was defined by the very equation used to “derive” it.
• “A resurrection would leave testable isotopic traces in the tomb.” No mechanism gives an expected signal, and carbon-14 is not made from carbon-12.
• “NASA found a 3D property no other image has”; “the Sudarium of Oviedo is documented since the 7th century with the same AB blood”. The 3D analysis was done by two researchers of the 1970s Shroud project, not by NASA as such, and “no other image” overstates it; blood typing of degraded samples and the Sudarium’s early history are contested. We use none of it.

Appendix: the Shroud of Turin

The Shroud is Contested and not load-bearing: nothing on this page uses it, so its status cannot be used for or against the argument above.

Sources

1. Tacitus, Annals 15.44 (c. AD 116). Loeb Classical Library, vol. V (1937), via LacusCurtius, penelope.uchicago.edu. Checked.
2. Josephus, Jewish Antiquities 18.63-64 (the Testimonium; core probably authentic, partly interpolated), 20.200 (the death of James), 4.219; Jewish War 4.317. Tr. W. Whiston, Project Gutenberg #2848. 20.200 and 4.219 checked in Whiston; the others are standard references, not re-checked by us.
3. World English Bible (2020 stable text, public domain), eBible.org eng-web_vpl.zip: 1 Cor 9:1; 15:3-8, 11, 14, 20, 42, 44; 16:22; Gal 1:13-19, 23; 2:9; Mark 3:21; 6:14-16; 15:43; 16:8; Matt 27:52-53; 28:13-17; Luke 24:16, 31, 34; John 7:5; Acts 1:14; 10:40-41; Phil 2:9-11; Deut 21:23. ebible.org
4. Lüdemann, G. (1994). The Resurrection of Jesus: History, Experience, Theology. Fortress. (Dating as cited by several secondary sources; not checked against the book.)
5. Ehrman, B. D. (2014). How Jesus Became God. HarperOne. (Via secondary summaries.)
6. Allison, D. C., Jr. (2021). The Resurrection of Jesus: Apologetics, Polemics, History. T&T Clark. Quotations and page numbers as given in K. R. Harriman's review, krharriman.substack.com (checked against the review, not the book).
7. Crossan, J. D. (1994). Jesus: A Revolutionary Biography. HarperSanFrancisco; and Who Killed Jesus? (1995). (Via secondary summaries.)
8. Evans, C. A. (2005). "Jewish Burial Traditions and the Resurrection of Jesus." Journal for the Study of the Historical Jesus 3(2): 233-248; and "Getting the Burial Traditions and Evidences Right," in M. F. Bird et al., How God Became Jesus (Zondervan, 2014). (Via secondary summaries.)
9. Magness, J. (2005). "Ossuaries and the Burials of Jesus and James." Journal of Biblical Literature 124(1): 121-154. (Via secondary reports.)
10. Craig, W. L. "Dale Allison on Jesus' Empty Tomb, His Post-Mortem Appearances, and the Origin of the Disciples' Belief in His Resurrection" (review of Allison, Resurrecting Jesus, T&T Clark, 2005). reasonablefaith.org
11. Kirby, P. (2002). "The Case Against the Empty Tomb." Journal of Higher Criticism 9(2): 175-202; reprinted in R. M. Price & J. J. Lowder (eds.), The Empty Tomb: Jesus Beyond the Grave (Prometheus, 2005).
12. Lowder, J. J. (2005). "Historical Evidence and the Empty Tomb Story." In Price & Lowder (eds.), The Empty Tomb. (As summarised and answered in McGrew & McGrew 2009.)
13. Justin Martyr, Dialogue with Trypho 108 (c. AD 160). Standard reference; not re-checked by us.
14. Our calculation: truth/research/calc/12-resurrection_bayes.py (200,000 Monte Carlo draws, seed 12; the correlated-judgement tables and the cap in its sections 7-8) and its output, in the site's source repository. Every input is a sourced number or a labelled judgment range you can change. The figures for the cap against each rival (hiddenness, the silence of other candidate acts and the evidence from suffering included) come from our separate research record on divine action, omega/research/ceiling/c3-agency/RECORD.md, and its scripts, in the same repository.
15. Rees, W. D. (1971). "The Hallucinations of Widowhood." BMJ 4: 37-41. doi:10.1136/bmj.4.5778.37
16. Olson, P. R., Suddeth, J. A., Peterson, P. J. & Egelhoff, C. (1985). "Hallucinations of Widowhood." Journal of the American Geriatrics Society 33: 543-547. doi:10.1111/j.1532-5415.1985.tb04619.x
17. Castelnovo, A., Cavallotti, S., Gambini, O. & D'Agostino, A. (2015). "Post-bereavement hallucinatory experiences: A critical overview of population and clinical studies." Journal of Affective Disorders 186: 266-274. doi:10.1016/j.jad.2015.07.032
18. Hurtado, L. W. (2003). Lord Jesus Christ: Devotion to Jesus in Earliest Christianity. Eerdmans. (Via reviews.)
19. Bauckham, R. (2008). Jesus and the God of Israel. Eerdmans.
20. McDowell, S. (2015). The Fate of the Apostles. Ashgate (Routledge 2016); grades as summarised in McDowell, "Did the Apostles Really Die as Martyrs for Their Faith?", Christian Research Journal. equip.org
21. Habermas, G. R. (2005). "Resurrection Research from 1975 to the Present: What Are Critical Scholars Saying?" Journal for the Study of the Historical Jesus 3(2): 135-153. Author's online version
22. Strauss, D. F. (1864). Das Leben Jesu für das deutsche Volk (the critique of the swoon theory). (Via secondary summaries.)
23. Cavin, R. G. (1995). "Is There Sufficient Historical Evidence to Establish the Resurrection of Jesus?" Faith and Philosophy 12(3): 361-379.
24. Festinger, L., Riecken, H. W. & Schachter, S. (1956). When Prophecy Fails. University of Minnesota Press.
25. Klausner, J. (1925). Jesus of Nazareth. (As discussed in McGrew & McGrew 2009.)
26. Lake, K. (1907). The Historical Evidence for the Resurrection of Jesus Christ. (As discussed in McGrew & McGrew 2009.)
27. Wright, N. T. (2003). The Resurrection of the Son of God. Fortress. (Via secondary summaries.)
28. McGrew, T. & McGrew, L. (2009). "The Argument from Miracles: A Cumulative Case for the Resurrection of Jesus of Nazareth." In W. L. Craig & J. P. Moreland (eds.), The Blackwell Companion to Natural Theology, 593-662. Wiley-Blackwell. Quoted from the authors' draft (pp. 30-42), PDF mirror.
29. Swinburne, R. (2003). The Resurrection of God Incarnate. Oxford University Press. (Figures as reported by Otte.)
30. Otte, R. (2003). Review of Swinburne, The Resurrection of God Incarnate. Notre Dame Philosophical Reviews, 2003.09.04. ndpr.nd.edu
31. Martin, M. (2005). "Swinburne on the Resurrection." In Price & Lowder (eds.), The Empty Tomb, 453-468. (As cited in McGrew & McGrew 2009, draft p. 40 n. 31.)
32. Kaneda, T. & Haub, C. (2022). "How Many People Have Ever Lived on Earth?" Population Reference Bureau. prb.org
33. Hume, D. (1748). An Enquiry concerning Human Understanding, section 10, "Of Miracles", paragraphs 10.4 and 10.13. davidhume.org
34. Carrier, R. (2012). Proving History: Bayes's Theorem and the Quest for the Historical Jesus. Prometheus.
35. The Book of Mormon witnesses: FAIR, "Eight witnesses" (fairlatterdaysaints.org); Vogel, D. (ed.), Early Mormon Documents, vol. 5 (Signature, 2003). (Via secondary summaries; the primary documents were not checked by us.)
36. Haraldsson, E. (1987). Miracles Are My Visiting Cards; (2013) Modern Miracles (White Crow); "Sai Baba," Psi Encyclopedia (Society for Psychical Research). psi-encyclopedia.spr.ac.uk (Via secondary summaries.)
37. Edwards, W. D., Gabel, W. J. & Hosmer, F. E. (1986). "On the Physical Death of Jesus Christ." JAMA 255: 1455-1463. PMID 3512867 (abstract).
38. Maslen, M. W. & Mitchell, P. D. (2006). "Medical theories on the cause of death in crucifixion." Journal of the Royal Society of Medicine 99: 185-188.
39. Damon, P. E., et al. (1989). "Radiocarbon dating of the Shroud of Turin." Nature 337: 611-615. Reprint at shroud.com
40. Casabianca, T., Marinelli, E., Pernagallo, G. & Torrisi, B. (2019). "Radiocarbon Dating of the Turin Shroud: New Evidence from Raw Data." Archaeometry 61(5): 1223-1231. doi:10.1111/arcm.12467 (Checked: abstract.)
41. Rogers, R. N. (2005). "Studies on the radiocarbon sample from the shroud of Turin." Thermochimica Acta 425: 189-194. (Via secondary summaries.)
42. Freer-Waters, R. A. & Jull, A. J. T. (2010). "Investigating a Dated Piece of the Shroud of Turin." Radiocarbon 52(4): 1521-1527. doi:10.1017/S0033822200056277 (Checked: abstract.)
43. De Caro, L., et al. (2022). "X-ray Dating of a Turin Shroud's Linen Sample." Heritage 5(2): 860-870. doi:10.3390/heritage5020047 (Checked: abstract; the replication point via secondary summaries.)
44. Di Lazzaro, P., Murra, D., Santoni, A. & Baldacchini, G. (2010). "Sub-micrometer coloration depth of linens by vacuum ultraviolet radiation." Proceedings of the International Workshop on the Scientific Approach to the Acheiropoietos Images, ENEA Frascati, 4-6 May 2010. PDF (Checked. The 200 nm figure is theirs, citing Fanti et al., J. Imaging Sci. Technol. 54 (2010) 040201.)
45. Schwortz, B. (2011). "The Shroud in the News for Christmas 2011: An Editorial Response." shroud.com (Checked: “They did NOT say this was how the Shroud’s image was formed.”)