Hallucination Theories of the Resurrection

"If the data are too meager for the apologist's needs, they equally do not suffice for the rationalistic antagonists of the church." — Dale Allison, 2021¹

"I know of no close phenomenological parallel to the series of likely events as a whole." — Dale Allison, 2021¹

Summary

Hallucination theories are not one theory.  They are a family of proposals that explain the earliest reports that Jesus "appeared" after his death as experiences produced by grief, guilt, conversion, group excitement or belief that came first, followed by the growth of the stories in the retelling.  Each part of that family has to be tested on its own, because the evidence for each part is different.  The short answer is that visionary explanations account well for some of the evidence, badly for some, and that much of the decisive evidence no longer exists on either side.

1. Experiences of the dead are common and normal. In the first systematic study, 137 of 293 widowed people in mid-Wales, 46.7 percent, reported experiencing their dead spouse.² Most of those were a felt presence. Only 14.0 percent of the whole sample had seen the dead spouse and 13.3 percent had heard the spouse.² Later studies and reviews find similar or higher rates. They find the experiences in people with no mental illness, though in some studies they go with harder grief and more distress.³ These studies document what people report and what goes with it. They do not show what caused any one experience. So an individual visionary experience of Jesus by Peter, or Mary in John's account, after a sudden and violent loss, needs no special pleading. Verdict: high confidence that such an experience could have happened; no way to show that it did.
2. Guilt as a trigger is unproven, and the portraits of Peter and Paul are speculation. Bart Ehrman says such visions come more often to people who feel guilt; the source he follows lists guilt as a common part of grief, not as a cause of visions. The studies that link these experiences to happy marriages measured the marriage, not guilt. One Harvard study found that sudden loss brought both self-reproach and, later, a sense of the dead person's presence.⁴ So guilt is neither shown nor ruled out as a cause. Gerd Lüdemann's guilt-ridden Peter and conflicted Paul, and Michael Goulder's "conversion visions", go beyond any evidence about what those men felt. Dale Allison says Lüdemann's story of Paul "fits the facts" but that "the facts hardly require it."¹ Verdict: low confidence in the specific reconstructions; the guilt claim is not demonstrated, and not refuted.
3. Group experiences are documented, but none examined matches the Gospel group scenes. Nobody can inspect two people's private perceptions to see whether they were identical, even at an ordinary event. The testable question is whether several people independently report the same figure doing the same things. The Society for Psychical Research's 1894 census collected 95 first-hand reports of a figure seen by two or more people at once, and could confirm a second witness in only 43.⁵ Several of the cases it prints are pairs who describe the same brief, silent figure. At a 1953 crowd of 100,000 or more in Puerto Rico, waiting for a promised appearance of the Virgin Mary, people reported different things, and one woman who saw the Virgin said her friend beside her "did not see anything."⁶ The apologists' claim that "the concept of collective-hallucination is not found in peer reviewed medical and psychological literature" rests on an unreported database search and a psychologist's letter, and its authors concede in the next footnote that group members can have "similar but not identical" experiences.⁷ Verdict: both slogans fail. Group reports of seeing something are documented; no case examined shows a group independently reporting the same familiar person talking with them at length.
4. The earliest group reports are too thin to test. Paul's list says Jesus appeared "to the twelve" and "to over five hundred brothers at once", but describes nothing they saw or heard.⁸ The touching, eating and conversation come from Gospels written decades later. If those scenes are accurate in their details, nothing in the collective cases examined resembles them. That is a limit of the comparison, not a proof. It leaves several possibilities: the details grew in the retelling, a real memory was retold in a simpler and more uniform shape, people present had different experiences, or something happened whose cause the modern cases cannot identify. Whether the details are early is the subject of Legendary Development. Verdict: a model built on contagion and story growth fits the bare list; the Gospel scenes, taken as accurate, are unlike any case examined.
5. Paul is the hardest case for a grief theory and the easiest for a conversion theory. Paul's own letters say he persecuted the church and then changed.⁸ No source records a personal bond with Jesus for him to mourn. A sudden vision of Jesus in a hostile person is a reported type: Sadhu Sundar Singh had burned a Bible three days before his. But Sundar Singh was also in a religious crisis and praying for a sign, and nothing in Paul's letters shows the inner conflict that Lüdemann and Goulder supply. Verdict: a grief model has nothing to work with in Paul; a non-grief vision remains possible and unproven.
6. James cannot be assessed with any confidence. Paul says only that Jesus "appeared to James."⁸ Mark and John suggest that Jesus' brothers did not follow him during his ministry, and Acts places them among the believers soon after his death. The order of James's belief and his experience is not recorded. Verdict: undetermined. The evidence on each side is thin, and neither the apologetic "skeptic converted" argument nor any skeptical model can rest on it.
7. Visions explain seeing, not concluding "he was raised." Modern mourners do not conclude that the dead have risen, and ancient texts offered other readings of such a sight: a spirit, an angel, or a righteous man alive with God. When Athenian soldiers were said to have seen the long-dead Theseus at Marathon, the result was hero worship and a search for his bones. Skeptics answer with the apocalyptic beliefs of Jesus' followers, an answer that Allison, who once pressed the objection, now accepts, and with belief formed first under failed expectation. The nearest modern case, the Lubavitch movement after its Rebbe's death in 1994, shows many believers expecting his resurrection and some denying he had died, while only "a few" said he had already been raised. Kris Komarnitsky answers that Lubavitch doctrine itself required the Rebbe to remain in this world, so the case cannot show how often such groups would claim a resurrection already accomplished. Verdict: a genuine gap, partly filled.
8. The tomb is a separate question. An account of experiences says nothing about where the body was. If the tomb was found empty, as the sibling article The Empty Tomb tentatively concludes, a visionary account needs a second explanation for that. That is a cost, not a refutation.

Overall verdict.  A combined visionary and social account is compatible with sincere, early claims to have seen Jesus.  That compatibility does not establish its proposed historical steps.  Each of its parts has modern analogues.  Its specific steps for Peter, Paul, James and the groups have no direct evidence, and it is weakest where the evidence is most specific:  a hostile outsider in Paul, the group reports, and the claim that Jesus had already been raised.  The popular apologetic replies, that hallucinations cannot be shared, need expectation, never speak and never change lives, are mostly overstated.  Allison knows of no case that combines everything the earliest sources report.  That is a fair observation about the record, not a measured probability.  The surviving evidence does not settle which account is true, and the reason is the thinness of the record, not a rule that history may not ask.

Part 1: What the words mean, and what the sources record

Six kinds of experience

The debate goes wrong most often when different experiences are given one name.  Six need to be kept apart.

1. Hallucination. A perception with no outside object. Bart Ehrman quotes the American Psychiatric Association's manual: "a sensory perception that has the compelling sense of reality of a true perception but that occurs without external stimulation of the relevant sensory organ."⁹ It can involve any sense: sight, hearing, touch, smell or taste.
2. Illusion or misperception. A real object seen as something else. A shadow taken for a person, a cloud taken for a face, or an old woman in black taken for the Virgin Mary.⁶
3. Sensed presence. A strong feeling that a person is nearby, often located in space, without seeing or hearing anything. W. Dewi Rees called it "the illusion of feeling the presence of the dead spouse", and it was the commonest experience in his study.²
4. Dream. An experience during sleep, which the dreamer may or may not count as real contact.
5. Trance or vision. An altered state that the person describes as seeing something revealed. Paul himself speaks of "visions and revelations of the Lord" and of a man "caught up into the third heaven."⁸ Acts has Paul fall "into a trance" in the temple and see Jesus.⁸ In Acts, Paul calls the Damascus event "the heavenly vision."⁸
6. A claim passed on. A report that someone else saw something. In the Puerto Rican crowd studied in 1953, 33 people gave 62 answers about what had happened. Twenty-six of the answers passed on what others had said about cures or about the Virgin being seen, while only three reported a vision of the Virgin or a sign in the sky.⁶

Researchers on bereavement now avoid the word "hallucination" for many of these experiences.  A 2020 review by Karina Kamp and eight colleagues, drawn from psychiatry, psychology and anthropology, notes that the phenomena cover "a wider range of phenomena than most uses of the term 'hallucination'", and adopts the neutral term "sensory and quasi-sensory experiences of the deceased."³  The term "hallucination" in this debate therefore often means "an experience the speaker thinks had no outside cause."  That is a judgment about the cause, not a description of what was experienced.  The studies below record what people report, how often, and what goes with it.  None of them tests whether any one experience had an outside cause, so this article keeps four things apart:  what was experienced, how often such reports occur, how the person interpreted it, and what caused it.

Normal experience is not mental illness

A bereavement experience is not a diagnosis.  Rees concluded that the experiences of widowed people were "normal and helpful accompaniments of widowhood."²  Allison, surveying the later research, calls them "a normal, non-pathological part of the mourning process."¹  Normal does not mean calm.  In a Danish study of 175 widowed people, those who had such experiences "scored significantly higher than those without BHs on prolonged grief, post-traumatic stress, depression symptoms, and emotional loneliness."¹⁰  Kamp's 2020 review, which calls the experiences SED, reads findings like this as a link with distress, not a mark of illness:  "it seems unlikely that SED, in general, serve to indicate psychopathology."³  For this debate the link cuts slightly toward the skeptic, since the disciples' grief was presumably hard.  So a hallucination theory does not need to call Peter or Paul mentally ill.  That a kind of experience is usually normal does not show that every instance is;  it shows that illness is not needed to explain one.  Allison writes that "We know enough to dismiss conscious deceit or illness as the cause of Easter faith."¹

The surviving accounts are also too sparse to support a confident medical diagnosis of Peter or Paul.  They lack the clinical history and examination needed to distinguish among possible causes.  Proposals that Paul had epilepsy or a "conversion disorder" therefore remain speculative, whether a skeptic or an apologist states them.  This article does not use them.

What each source actually says

The appearance reports come from sources of very different kinds and dates.  The details of dating are worked out in Paul's Letters as Historical Evidence and The Resurrection Belief: Event or Process?.  Four levels matter here.  Paul wrote 1 Corinthians in the mid-50s.  He had taught its formula at Corinth about AD 50.  The core wording of 1 Corinthians 15:3b-5 is older than that.  The claim that Jesus had appeared to Cephas was probably being made before Paul joined the movement, one to six years after the crucifixion.  None of those dates is the date of the experiences themselves, which are earlier still.

Three things follow.  First, Paul is the only person whose own description survives, and it is brief.  Second, every report of what a group saw and heard comes from the Gospels or Acts, written decades later.  A later source can preserve earlier material, and it can also add to it;  its date alone settles neither.  So a description that survives only in a Gospel is weaker evidence than Paul's bare list for what was first claimed, but it is not thereby worthless.  Third, Paul uses one verb, ὤφθη, "he appeared" or "he was seen", for every item on his list, including his own experience.⁸¹¹  That shows Paul put his experience in the same class as the others.  It does not show that everyone's experience felt the same.

Each claim below is therefore marked as one of three kinds:  what a source records, what a modern writer infers from it, and what is merely possible.

Part 2: The models, in their proponents' words

Five proposals and one critic are examined here.  The proposals overlap, but they are not the same, and none of them claims that five hundred people had one identical hallucination.

Lüdemann: mourning, guilt and a chain reaction

Gerd Lüdemann, a German New Testament scholar, gives the most detailed psychological account among the works examined here.  He grants that "It may be taken as historically certain that Peter and the disciples had experiences after Jesus' death in which Jesus appeared to them as the risen Christ."¹²  His explanation starts with Peter.  "With the dramatic events of Good Friday following close upon his denial of Jesus, Peter's world had collapsed."¹³  He applies a Harvard study of bereavement, which found three factors that hinder recovery from grief:  sudden death, "an ambivalent attitude toward the deceased, involving feelings of guilt", and dependence.¹³  All three, he argues, fit the disciples.  Peter's vision is "an example of unsuccessful mourning, because it abruptly cuts off the very process of mourning, substituting fantasy for unromantic reality."¹³

The vision then spread.  "Not long after Good Friday, Peter experienced a vision of Jesus that included auditory features, and this event led to an extraordinary chain reaction."¹³  Lüdemann reads the five hundred as "a mass ecstasy that took place in the early period of the community. One or more individuals provided the stimulus for it."¹³  He adds that "any nonecstatic interpretation comes to grief" at this point.¹³  For Paul, he proposes a "Christ complex":  Paul was drawn to what he attacked, and "When Paul approached Damascus, there was a catastrophic breakthrough. Paul fled from his painful situation into the world of hallucination".¹³  James, he says, was "caught up in the excitement", and his vision came under conditions "different from Peter's", since "James must have known of Peter's vision before he in turn 'saw' his brother".¹³

Goulder: conversion visions and collective delusion

Michael Goulder, a British New Testament scholar, treats Peter and Paul as cases of "conversion vision."  His model case is Susan Atkins, a member of the Manson family who reported seeing and hearing Jesus in her prison cell.  "Psychologists would say she had hallucinations, both in hearing voices and in seeing things."  Goulder writes, "That is what was experienced by Peter on Easter Day, when he saw the Lord, or by Paul outside Damascus."¹⁴  Atkins's own words, as Goulder quotes them, describe something seen "in my thoughts" as well as a voice, and a release from guilt:  "There was no more guilt!"¹⁴  Goulder holds that "it is characteristic of such experiences that they occur at deep crises of life and seem intensely real."¹⁴  This model does not need grief, so it can cover a hostile Paul.  It does need a crisis.

For the groups Goulder uses a different mechanism.  Peter's and Paul's experiences were "primary", while "the appearances to the apostles or the 500 brethren which Paul mentions are secondary in the sense that these groups have already heard, and perhaps credited, the reports of Jesus being alive."¹⁴  His parallel is a wave of Bigfoot sightings in South Dakota in 1977, which sociologists explained as "collective delusion" using Neil Smelser's theory of collective behavior.¹⁴  He grants that his account of Paul is speculative:  "The fact that we cannot provide a full account of a psychic event two millennia ago does nothing to commend theological dogmatism; even if speculative, a natural explanation is to be preferred."¹⁴

Ehrman: a few visions, then storytelling

Bart Ehrman, an agnostic New Testament scholar, proposes the smallest model.  "My tentative suggestion is that three or four people—though possibly more—had visions of Jesus sometime after he died."  He names Peter, Paul and Mary Magdalene.⁹  The rest is storytelling:  "as the stories of Jesus's 'appearances' were told and retold, of course, they were embellished, magnified, and even made up; so soon, probably within a few years, it was said that all of the disciples had seen Jesus, along with other people."⁹  On James, Ehrman says only that he possibly had a vision, "but I think it is difficult to say."⁹  Ehrman does not need group hallucinations at all.  He says of the five hundred that "Paul is the only one who mentions this event".⁹

Komarnitsky: belief first, then visions

Kris Komarnitsky, who works in the aviation industry and has written on Christian origins for many years, argues that the bare grief model has a hole.  "We know that many other Jews before the time of Jesus must have experienced post-mortem bereavement hallucinations of lost loved ones, and yet there is no record of such people concluding that their lost loved ones had been resurrected from the dead."¹⁵  His answer is cognitive dissonance:  followers who had staked everything on Jesus rationalized his death, and "cognitive dissonance may have given rise to the resurrection belief and then the appearance experiences followed only after the resurrection belief was in place."¹⁵  His modern parallel is the Lubavitch Hasidim after the death of their Rebbe in 1994.

Carrier: a culture primed for visions

Richard Carrier, an atheist writer trained in ancient history, argued in a 2005 essay that "the first Christians experienced hallucinations of the risen Christ, of one form or another."  His case leans on culture more than on grief.  In the ancient world, he writes, "hallucination occurred more often and more openly".¹⁶  He cites surveys in which "between 7 and 14 percent of people surveyed, who did not exhibit any mental illness, reported having experienced hallucinations", and adds:  "Of these identified experiences, over 8 percent were multisensory hallucinations, and 5 percent involved entire conversations."¹⁶  The disciples, he argues, "were primed for hallucination by their bereavement, their anxiety-filled circumstances, their cultural predisposition to see and believe things that confirmed their deepest desires in religious terms, and other factors, including social influence and suggestion".¹⁶  For the groups he proposes ecstatic gatherings "objectively comparable to the visitation at the Pentecost", after which "Everyone could be persuaded to agree they were seeing the very same thing".¹⁶  His answer to the question of why visions led to the word "raised" is that some Jews could already imagine a resurrection in a new body:  from the soul surviving death, "It is a very small step to go from that to an idea of the departed soul becoming or being clothed in an entirely new body."¹⁶  On this view the old body could stay in its grave.

Allison: the critic of both sides

Dale Allison, a New Testament scholar who believes that Jesus appeared to his followers, is not a proponent of a hallucination theory.  He is included because his 2021 book makes the most detailed comparison between the reports and modern cases among the works examined here, and because he sets limits on both sides.  He rejects most apologetic arguments against visions.  He also writes that "If one comes to the texts without such a predisposition, there is nothing in them that determines the nature of the experiences of Peter and the twelve and the others shortly after Good Friday. Historical knowledge just does not reach that far."¹  And he rejects the assumption, shared by apologists and skeptics, that all the appearances need one explanation:  "That Jesus showed himself after death to several people hardly entails that he truly appeared to everyone who claimed to have seen him."¹

What the models share and where they differ

All five proposals, and Allison, agree that real people had real experiences that they took to be Jesus.  None of the proposals is a theory of fraud.  They differ on three questions:  what caused the first experience (grief, guilt, conversion, a culture primed for visions, or prior belief), how the group reports arose (contagion, collective delusion, ecstatic gatherings, or storytelling), and how much of the later Gospel detail they accept as early.  So a test that refutes one model may leave another untouched.  A rebuttal of "group hallucination" does not touch Ehrman, who does not use one.  A rebuttal of grief does not touch Goulder's Paul or Komarnitsky's dissonance.

Part 3: What bereavement research actually measures

Every hallucination theory leans on modern studies of the bereaved.  Those studies are good evidence for some claims and no evidence for others, so their details matter.

The main studies

Rees's study is still the anchor because it tried to reach every widowed person in a defined area and interviewed each one.²  The figures that matter most are how many saw the dead spouse, 14.0 percent, and how many felt a presence without seeing, which was most of the rest.²  Grimby in Sweden found "Half of the subjects felt the presence of the deceased (illusions); about one third reported seeing, hearing and talking to the deceased".¹⁸  Kamp's Danish study reached a population register sample:  of 175 older widowed people, 92 reported some experience, and 48 of those 92 had seen the dead spouse.¹⁰  That is 27 percent of the whole sample.  Rates for seeing vary from about one in seven to about one in four in the studies that report them against everyone asked.

Reviews give wider ranges.  Kamp's 2020 review says "Prevalence estimates of having at least one of these experiences range from 47% to 82% across several studies."³  A 2015 review by Anna Castelnovo and colleagues gives "30% to 60%" among widowed people and warns that "all studies in the bereaved population had a very small number of subjects."²¹  A 2022 systematic review by Jenny Streit-Horn, Janice Holden and Julia Smith averages the strongest studies to estimate that "70-80% of bereaved people report one or more ADC experiences within months" of a death.²²  These ranges include felt presence, voices, dreams and signs, not only seeing.

A denominator trap

Olson's 1985 abstract is often quoted as if all its figures shared one base.  They do not.  It says "Twenty-eight (61%) of the widows reported hallucinatory experiences", "Twenty-four (86%) of the widows described the experiences as good or helpful", and "Nineteen (54%) of the widows had never discussed the experiences with anyone before this study."¹⁷  The 61 percent is 28 of 46.  The 86 percent is 24 of the 28 who had experiences.  The 54 percent fits neither:  19 of 46 is 41 percent and 19 of 28 is 68 percent.  Only a base of 35 gives 54 percent, and the abstract names no group of 35.²³  The full paper could not be obtained, so this article does not use the 54 percent.  The same caution applies to the percentages for each sense that Kamp's 2020 review prints for Olson's study:  they are fractions of the 28 with experiences, not of all 46.³

Popular summaries repeat other figures that do not survive a check.  Michael Goulder wrote that Rees found "as many as forty-five per cent of his patients recently widowed have seen or heard their dead partners."¹⁴  Rees's 46.7 percent includes felt presence.  At most 27.3 percent saw or heard, and fewer if the two groups overlap, which some did.  Rees's sample also included people widowed for decades.

What the studies show

1. The experiences are common. Every study finds them in a large minority or a majority of the bereaved. Most are a sense of presence, not a sight.
2. They are normal and mostly welcome. In Rees's sample, 94 of the 137 who had experiences, 68.6 percent, were helped by them.² Kamp's 2020 review reports that in Western studies most are "considered positive", with "a significant minority" ambivalent or negative.³
3. They are usually kept private. In Rees's study "only 38 (27.7%) had done so", that is, told anyone, and "No doctor had been informed and only one person had confided in a clergyman."² The most common reason given was fear of ridicule.
4. They last. Rees found that the experiences "often lasted many years", and that 106 people, 36.1 percent of all interviewed, were still having them when interviewed.²
5. In the studies that measured it, they go with happier and longer marriages. Rees: "Eleven widows stated that their marriage was unhappy and all these reported a complete absence of post-bereavement hallucinations."² Grimby: "Former marital harmony was found to make a person more prone to" the experiences.¹⁸ Kamp's 2019 study found that length of marriage predicted them.¹⁰ Kamp's 2020 review treats these links as tentative, because studies differ in method and several findings "have not been replicated."³ These studies measured how happy or how long the marriage was. None of them measured guilt.
6. Culture shapes them. Among the Achuar of Ecuador, any re-experiencing of a dead person is "a threat to the soul of the experiencer."²⁴ Kamp's review notes that such experiences are feared among the Navajo and sought among Catholic Mexican-American families.³
7. Shared experiences are rarely reported in the bereavement studies, and none of them studied one directly. Surveys of apparitions in general report more. Rees reports none. In Elsaesser's self-selected online sample, "36.4% of our respondents reported that they were not alone at the time of their ADC, and of these, 21.0% asserted that the ADC was witnessed by their companions."²⁰ That is about 7.6 percent of a sample already selected for interest in the subject, by the respondent's own account, with no companion interviewed.²⁵ A Los Angeles survey of 434 people found "slightly over 2 percent" reporting an encounter with the dead that another person present shared. Surveys of apparitions of every kind give much higher shares once they count only cases where someone else was there. Allison collects them: in the 1894 census, "when the percipient is not alone, about one third of the cases are collective", by G. N. M. Tyrrell's count; Hornell Hart, counting only cases where the others were awake and able to see, "came up with a much higher figure", 56 percent; and in Erlendur Haraldsson's Icelandic study, "of 89 cases in which two or more people were in a position to share a supposed encounter with a dead individual, they did so in 41 cases."¹ These are reports, mostly from one percipient: the census could confirm a second witness in only 43 of its 95 collective visual cases (Part 5). Allison also warns that "these assessments fail to distinguish between visions of figures known and unknown and between figures recently deceased or long deceased." Even so, they are the skeptic's best numbers on this point. When other people are present and able to perceive, a shared report is common, not rare.

Three claims the studies do not support

Guilt as a trigger.  Ehrman writes that bereavement visions "are more commonly experienced when a person has a sense of guilt over some aspect of his or her relationship with the one who has died", and applies this to disciples who "had all betrayed, denied, or fled."⁹  He gives this as Allison's summary of the research.  The features Ehrman lists, a sensed presence, guilt, anger, idealizing the dead and gathering to tell stories, follow the five points of Allison's 2005 excursus on the disciples and bereavement.  There guilt is a separate common feature of grief:  "Surveys show that, at least in the modern world, up to half of the grieving wrestle with serious guilt in one way or another".²⁶  The excursus does not say that guilt makes visions more likely.  Allison even notes that Peter's encounter "could have been" what "created guilt or intensified it rather than the other way around."²⁶  So Ehrman's sentence says more than the source he follows.

The primary evidence on guilt is thin and indirect.  Rees, Grimby and Kamp measured how happy or long the marriage was, and a widow can feel guilt about one day of a happy marriage.  Rees's finding shows that an openly unhappy marriage did not produce the experiences in his sample;  it does not show that guilt prevents them.  The nearest primary evidence is the Harvard study that Lüdemann uses, by Colin Murray Parkes and Robert Weiss, of Boston widows and widowers aged 45 or younger.⁴  After a death that came with little warning, they describe "feelings of self-reproach and despair" and, "As time passes", "a sense of the continued presence of the dead person".⁴  That shows guilt and a sensed presence occurring together after a sudden loss, in one small sample.  It does not show that the guilt caused the sense of presence;  the shock of the death could have produced both.  Elsaesser and colleagues, introducing their survey, write that such contact is typically taken to convey one of four kinds of message, one of them "settling old conflicts, allowing space for apologies and providing closure".²⁰  That fits experiences that deal with guilt, but it measures nothing about cause.

Lüdemann's own guilt factor comes from the same Harvard work, where it is one of three factors that hinder recovery from grief,¹³ which is a different question from what produces visions.  The result is that guilt as a trigger is neither demonstrated nor refuted.  Guilt may have mattered to Peter.  No study read for this article shows that it makes visions more likely, and no source records Peter's state of mind at the appearance.

Prevalence as probability.  A study that finds 14 percent of Welsh widowed people saw their spouse does not give the chance that Peter saw Jesus.  The samples are modern, mostly Western, mostly widows and widowers, mostly elderly, and asked years after the death.  Peter lost a teacher, not a spouse, by public execution, in a culture whose beliefs about the dead differed from those of modern Wales.  The studies show that such experiences are a normal human response to loss.  They cannot supply a number for one man in AD 30.

A rate for conversations.  Carrier's figure that "5 percent involved entire conversations"¹⁶ does not come from the surveys he names.  His source, a 1988 review by the psychologists Peter Slade and Richard Bentall, reports the general-population surveys of 1894 and 1948, and says that hallucinations in more than one sense were "comparatively rare (11.6 per cent in 1894; 8.4 per cent in 1948)".²⁷  The 5 percent comes from a different study, of 375 college students asked about hearing voices:  "Five per cent of the subjects reported holding conversations with their hallucinations."²⁷  That is 5 percent of all the students, not of the hallucinations, and it concerns voices, not a figure seen and heard.  Neither study is of the bereaved.  The corrected figures still matter to the debate.  They show that roughly one hallucination in nine to twelve involves more than one sense, and that some healthy people report talking with voices.  They give no rate at all for a familiar person who is seen, heard and answered.

Part 4: Individual experiences: Peter and Mary Magdalene

What is recorded

The formula Paul passed on names Cephas first.⁸  Luke also has the disciples say, "The Lord is risen indeed, and has appeared to Simon!"⁸  Allison notes that readers who think Luke knew Paul's letters can take this to depend on 1 Corinthians, while others can ask whether Luke knew the tradition behind Paul's list.¹  So it may be a second witness to the claim, or an echo of the first.  Neither source says what Peter saw, heard or felt, where he was, or how long it lasted.  Mary Magdalene is not on Paul's list.  Matthew names her with "the other Mary", then has Jesus meet the women: "They came and took hold of his feet, and worshiped him."⁸  John's version describes Mary alone at the encounter, initially taking Jesus for the gardener.⁸  An individual grief experience is a closer comparison to John's scene than to Matthew's shared, tactile encounter; the two descriptions should not be treated as one bare report.  Lüdemann nonetheless judges that "It seems to be historically certain that Mary Magdalene experienced an appearance of the risen Jesus."¹²

How well a bereavement experience fits

On the evidence of Part 3, a bereavement experience is a good fit for a single person after a sudden loss, and several apologetic objections to it fail.

1. Loss of a teacher, not a spouse. Most studies are of widows and widowers, but Kamp's review reports such experiences "in all types of relationship loss" and after violent deaths as well as natural ones.³
2. Expectation. Gary Habermas and Michael Licona argue that Paul "did not appear to have been in the frame of mind to experience a hallucination."²⁸ The same argument is often made for the grieving disciples: they were in despair, not hopeful. But bereavement experiences come to people who expect nothing. Allison, after reviewing the literature, writes that it "offers no support for supposing that seeing a dead individual is typically the product of enthusiasm or excitement, or the upshot of conscious expectation. Visions blow where they will."¹
3. Speech. The appearance stories have Jesus speak. Bereavement experiences sometimes include hearing the dead person, and sometimes sight and sound together. In Rees's study 13.3 percent of all the widowed people had heard the dead spouse, and 11.6 percent had spoken to the spouse, which records the living person speaking, not a reply.² In Kamp's Danish study, 45 percent of those with experiences had heard the spouse and 32 percent had "talked to/with" the spouse.¹⁰ Kamp's 2020 review says that hearing voices in bereavement "usually involves meaningful language": one woman it quotes heard her grandmother say "it's at the back", and then found the thing that needed turning.³ Allison cites two Icelandic studies in which "20% of the reports involved more than one of the senses" and "the most usual combination was hearing and seeing the deceased, amounting to 10% of the reports."¹ None of this documents a long two-way conversation.
4. Glory rather than a ghost. Larry Hurtado argued that the disciples met Jesus "in a new and more powerful mode of existence", unlike grieving relatives. Allison replies: "Many apparitions, however, appear in a 'new and more powerful mode of existence.'" He gives modern reports of the dead seen in brilliant light.¹
5. Conviction. Ehrman notes that people who have such experiences "tend not to doubt what they have seen."⁹ So a sincere report by Peter is what the model predicts, not evidence against it.

What the fit does not show

The fit shows that a grief experience is a possible cause of Peter's report.  It does not show that it was the cause.  The objection that "Peter must have been hallucinating" and the objection that "Peter could not have been hallucinating" both outrun the evidence.  Allison's summary applies here directly:  "there is nothing in them that determines the nature of the experiences of Peter and the twelve and the others".¹

Lüdemann's specific story, in which guilt over the denial drove Peter "by a bold if unconscious leap" into "the world of his wishes",¹³ adds detail that no source records.  The denial itself is in the Gospels, where Mark has Peter remember Jesus' prediction and weep.⁸  What Peter felt weeks later is not recorded anywhere.  William Lane Craig's criticism is fair on this point:  "there's not enough information to do psychoanalyses of Peter or Paul."²⁹  And as Part 3 showed, the bereavement studies do not single out guilt as a cause of such experiences.

Verdict on individual experiences.  The experience, a vivid and convincing sense of a dead loved one, is well documented in modern reports:  high confidence that such experiences occur.  Its historical fit to a single early report such as Peter's is good:  moderate to high confidence that it could explain such a report.  Whether it did cannot be decided from the evidence.  The specific psychological reconstruction of Peter's guilt is unsupported:  low confidence.

Part 5: Group reports

What Paul's list claims, and what it leaves out

Paul's list contains three group appearances:  "to the twelve", "to over five hundred brothers at once, most of whom remain until now", and "to all the apostles."⁸  The Greek ἐφάπαξ, "at once", makes the five hundred a single occasion.¹¹  That is all the list says.  Allison writes of the five hundred that "our knowledge is near nil. Who exactly were these people? Paul supplies neither names nor addresses."¹  Paul does not say what any group saw or heard, where they were, how long it lasted, or whether everyone present saw the same thing.  He does not claim to have been present at any of these occasions, and he places his own experience "last of all", so for these items his letter reports what others claimed.

The group scenes with detail come later.  Luke has Jesus show his hands and feet, invite the eleven to touch him, and eat broiled fish.⁸  John has Jesus invite Thomas to put his hand in his side.⁸  Matthew's mountain scene adds, "When they saw him, they bowed down to him; but some doubted."⁸  These are later narratives.  They may preserve what the groups first reported, add to it, or both, and their date alone does not decide which.  That question is examined in Legendary Development.  This article tests the visionary models twice:  against the bare claim in Paul's list, and against the later scenes taken at face value.

What "the same experience" can mean

Nobody can look inside two people's minds and confirm that their perceptions were identical, even when both watch an ordinary event.  So the question "did they all share one hallucination?" cannot be tested as asked.  What can be tested is whether separate reports agree:  the same figure, doing the same things, saying the same words, at the same time.  Four situations need to be kept apart.

1. Concordant reports. Everyone present independently reports seeing and hearing the same figure doing the same things at the same moment, with no outside object to account for it.
2. Simultaneous but different experiences in a group. Some people present see something, others see nothing, and those who see something describe it differently. Often an outside object is misperceived: the sun, a cloud, a statue, a person.
3. Sequential and social spread. One or two people report an experience; others later have their own experiences, or come to believe and repeat the report.
4. Convergence afterwards. Accounts that differed at first come to agree as people talk. Leonard Zusne and Warren Jones, in a 1982 textbook that both sides of this debate cite, wrote of collective hallucinations: "Once the general type of hallucination is established, it is easy to harmonize individual differences in the accounts. This may take place during the hallucination or in subsequent conversations."³⁰

Agreement between reports does not by itself show a real outside object, since agreement can come from the fourth process.  Disagreement does not by itself show a hallucination, since witnesses of real events also disagree.  Each kind of evidence below has to be read with both cautions.

The apologetic objection is aimed at the first kind.  None of the skeptical models in Part 2 depends on it.  Lüdemann's "mass ecstasy" is contagion, which is the second kind.  Goulder's secondary sightings and Ehrman's storytelling are the third kind, and both allow the fourth.

The evidence for collective experiences, by quality

The census figures come from the committee's chapter on collective cases.⁵  The examples it prints, as far as they were read for this article, mostly involve two people.  In one, a woman and her mother twice saw the same white cat, which "looked at us with the same melancholy gaze" before it "dissolved into nothing".⁵  In another, a man and his wife woke together and "both saw the same white unrecognisable figure, in the same place."⁵  The committee also set apart a class it called "Collectivity with Partial Dissimilarity", in which, for example, one woman "saw an angel fly through the room" while the friend sharing her bed "heard an angel singing."⁵  The committee notes:  "There is no case in which two percipients both see an apparition and at the same time hear a voice".⁵  Its 34 shared auditory cases are of a voice heard, not of a figure that talks with the group.  The committee's own conclusion was cautious:  "if the case for collective hallucinations rested on the Census alone, it would perhaps be rash to regard it as conclusively proved that hallucinations may be shared."⁵  Where two or more people were together and awake, the shared cases were "about a third", and the committee warned that "in view of the probability that in collective cases many of the supposed hallucinatory objects were real, it would be unsafe to draw any inferences from this."⁵

The Sabana Grande study is the best-documented crowd case found for this article, because social scientists were present before, during and after the event.  Schoolchildren had reported visions of the Virgin at a well, and an estimated 100,000 to 150,000 people gathered for a promised reappearance.⁶  In the last fifteen minutes before the appointed hour, "various people reported seeing different miracles":  rain falling "in many colors off the garments of the children", the Virgin "silhouetted in the clouds", and "wondrous rings of color" around the sun.⁶  Just after the hour, "a cry went up" that the Virgin was walking down the hillside, and "It took considerable time for the crowd to be persuaded that in fact this was not the Virgin, but only an old woman dressed in black."⁶  A second cry followed for a figure in white, who turned out to be an old man in a white shirt.  The children's own accounts had shown "basic inconsistencies":  "the color and costume and height of the Virgin varied from one child's report to another".⁶

The interviews have limits that matter.  The small team spent most of its time on "mass observation" and interviewed only thirty-three members of the crowd, and the authors write that there was "little opportunity to do any systematic sampling".⁶  Asked what had happened, the 33 gave 62 answers.  Sixteen mentioned seeing others cured and twelve described a cure of their own.  Two "saw visions of the Virgin" and one "saw an untoward celestial phenomenon".  Twenty-two had heard of cures from others and four had heard that the Virgin had been seen.  Two felt "a great spiritual peace", and three said nothing had happened.⁶  So hearsay outnumbered visions by 26 to 3, while first-hand claims of cures were the largest group of all.  These counts describe 33 people, not the crowd.  The authors themselves only ask whether "the bulk of the crowd were like minded".⁶  One pilgrim said, "Maybe the children saw the Virgin and we haven't."⁶  One woman who saw the Virgin said, "I told my friend who was at my side but she did not see anything."⁶  That is a first-hand report of the second kind of situation:  two people side by side, one with a vision, one without.

Jake O'Connell, writing in defence of the resurrection, surveyed six modern religious group visions in 2009.  He found that "Those Who Do See the Vision See It Differently", and concluded that "religiously inspired collective hallucinations can and do occur."³¹  In his first case "approximately two-thirds (67%) of those present saw the vision", and in his sixth, "fifty of two thousand (2.5%)."³¹  His sources are compilations and devotional accounts, not interviews of each witness, so the quality is mixed.

Two famous crowd events show the same pattern.  At Fatima in Portugal, on 13 October 1917, the journalist Avelino de Almeida of the Lisbon paper O Século recorded what the crowd said afterwards:  "O maior numero confessa que viu a tremura, o bailado do sol; outros, porém, declaram ter visto o rosto risonho da propria Virgem".³²  Most admitted seeing the sun tremble and dance;  others declared they had seen the smiling face of the Virgin herself.  The sun was in the sky the whole time.  At Zeitoun in Cairo, where lights appeared over a Coptic church from 1968, Allison, who cannot explain the lights and does not call them hallucinations, reports that "some claimed to see the clear outline of a haloed woman in white whereas others, at the same time and place, saw something less distinct."¹  He draws the lesson for Paul's list:  "we cannot know that everyone in those groups saw exactly the same thing."

The shared psychotic disorder literature is a different thing.  A 2006 review of 42 recent cases found that "The majority of the relationships were within the nuclear family (97.6%)" and that shared content was mostly delusion.³³  It describes a sick person passing beliefs to someone close over months or years, not a crowd seeing a figure.

What the apologists actually claim

Gary Habermas and Michael Licona wrote in 2004 that hallucinations "are not collective experiences", and that "a hallucination cannot be shared."²⁸  Their authority is a conversation with the psychologist Gary Collins:  "Hallucinations are individual occurrences. By their very nature only one person can see any given hallucination at a time."²⁸  In 2015 Joseph Bergeron and Habermas wrote that "the concept of collective-hallucination is not found in peer reviewed medical and psychological literature."  Their footnote says this "is based on Bergeron's comprehensive data base search of the Pubmed and American Psychological Association websites" and quotes a 2009 statement by the psychologist Gary Sibcy, who had "yet to find a single documented case of a group hallucination".⁷  No search terms, dates or results are reported, so the search cannot be checked.  The searches as described also would not reach the 1894 census, a textbook such as Zusne and Jones's, or O'Connell's theological journal.

Read closely, their claim is narrower than it sounds.  Their own next footnote says:  "Individuals within a group simply do not 'collectively' experience identical simultaneous hallucinations. Rather, when collective visionary experiences occur, some individuals in the group may experience similar but not identical personal hallucinations."⁷  Licona, in 2010, likewise grants that group hallucinations of O'Connell's kind are "rare but not at all impossible".³⁴  That narrower claim fits most of the sources above:  Sabana Grande, Fatima, Zeitoun and O'Connell's cases all show people in groups reporting experiences at the same time, with different content.  The census pairs who describe the same silent figure are the main exception found, and they are small, brief, retrospective and open to the committee's own doubt that the object was real.  No case examined shows a larger group independently reporting the same familiar person talking with them.

Allison's reply is that the apologists assume what they cannot know:  "These remarks appear to assume that the groups who saw Jesus beheld exactly the same thing, that they simultaneously had identical perceptions. Perhaps they did. Yet one fails to understand how anyone can ascertain this."¹  He adds that "None of those involved, however, have left us their first-hand accounts, so we cannot put their stories side by side and compare them."¹

The "everyone at once" argument

Licona aims his version of this argument at a specific reading of Goulder.  "In order to avoid the implausibility of a group hallucination, Goulder would have to suggest that the appearance to the Twelve ... involved each disciple experiencing an individual hallucination at the same time."³⁴  Against that, he sets background rates:  "50 percent of recently bereaved senior adults (only 14 percent of which are visual in nature)", against which "an incredible 100 percent of the Twelve would have experienced a hallucination, of the risen Jesus (rather than something else such as guards), simultaneously, in the same mode (visual) and perhaps in multiple modes."³⁴  Licona's rates do not match the primary study they resemble.  He takes the 14 percent as a share of those who have any experience, "or seven percent of all bereaved senior adults surveyed".³⁴  In Rees's study, 14.0 percent is the share of all the widowed people who saw the dead spouse, and about 30 percent of those who had any experience.²  Licona's source for the figures, a 2008 book by Aleman and Larøi, was not read here, so the slip may be his or theirs.

An arithmetic illustration shows why the number decides nothing.  If eleven men each independently had Rees's 14 percent chance of ever seeing a dead loved one, the chance that all eleven would do so is 0.14 to the eleventh power, about 4 in 10 billion.³⁵  Every input to that figure is borrowed or assumed: a retrospective rate from modern Welsh widowed people, applied to first-century men grieving a teacher; a number of participants that Paul's list does not fix; and independence between the men.  It also contains no model of timing, so it cannot say how likely experiences at the same moment would be.  The independence assumption matters because no proponent examined here makes it.  Contagion, suggestion and storytelling are all ways for one person's report to change what others experience or say.  Sabana Grande documents reports spreading and differing, but one case does not show the whole range of what dependence can produce.  The figure illustrates an assumed model only.  It supplies no historical odds and cannot establish that an actual proposed explanation is absurd.

The underlying objection still stands in a weaker form.  Licona's point is that the Gospels present all the disciples as having one experience together, and that differing experiences would be less likely to produce the belief Paul held.  He also grants that "The early oral traditions do not tell us anything related to the nature of the appearances."³⁴  So the objection depends, again, on how much of the Gospel scenes is early.

O'Connell's two arguments

O'Connell, having granted that collective hallucinations occur, offers two tests that he thinks the resurrection appearances fail.

The first concerns speech.  In his cases, "in no case does the vision ever carry on a conversation with the recipients", and he argues that "a group conversation would be impossible if the vision was a hallucination", because a crowd cannot expect a whole conversation in advance.³¹  This is a real difference between those cases and the Gospel scenes, in which Jesus talks with the disciples at length.  The inference to impossibility needs more: it assumes that a group's prior expectation must supply every perceived word.  Allison's finding that individual bereavement experiences need not arise from conscious expectation does not by itself refute a claim about groups.¹  But neither does the examined group evidence establish O'Connell's exclusive causal rule.  Zusne and Jones describe suggestion during an event and harmonizing afterwards.  These could help explain an agreed report, though they do not document a group having a lengthy conversation at the time.  O'Connell also excludes the cases nearest to a counterexample.  At Fatima, La Salette and Medjugorje the apparition is said to converse with a small group of seers, which he sets aside because he suspects fraud.³¹  Those disputed reports establish neither hallucination nor a real apparition.  The absence of a securely documented group conversation remains a limit of the comparison, not a proof of psychological impossibility.  Paul's list records no conversation with any group.

The second, and his main argument, concerns glory.  He argues that the disciples would have expected a glorified Jesus, so that "if the appearances were hallucinations, glorious appearances of Jesus would have occurred", and that "the absence of glorious appearances from the Gospel narratives is therefore indicative of the absence of any stories of glorious appearances at any point between AD 30 and 70."³¹  This rests on two premises: that expectation fixes the content of the experiences, and that early glorious stories would have been preserved.  Modern reports of glorious apparitions are consistent with his first premise; they do not refute it.  What is missing is evidence that every experience must follow that expectation, particularly when no description survives from the groups themselves.

O'Connell gives reasons for the second premise.  Jewish traditions honoured earlier heroes with heavenly splendour, which would also favour preserving stories of Jesus' glory.  He explicitly cites Luke's glorious appearance to Paul as evidence that Luke would have included other such stories if he knew them.³¹  Paul's body "raised in glory"⁸ and Acts' light from the sky⁸ therefore do not answer O'Connell by showing something he overlooked.  They show that glory was part of the tradition, but cannot establish that every earlier glorious story would survive, or that the later scenes preserve every feature of the experiences.  Lüdemann's proposal that the tactile emphasis "developed only later"¹² requires its own evidence.  Neither proposal can settle transmission simply by assuming the version of it that the argument needs.

Verdict on group reports

1. Concordant reports of the same figure: the 1894 census prints first-hand reports, from pairs in the examples read, of the same brief and silent figure, with a second witness confirmed in under half of the collective cases and real objects not ruled out. No case examined shows a larger group independently reporting the same familiar person speaking with them. No early source describes what any group saw, and no skeptical model examined needs concordant group perception.
2. Simultaneous, differing experiences in a group: documented, with evidence of mixed quality. Expectation, outside objects and suggestion are present in the best-documented cases. Moderate confidence that this kind of event happens.
3. Social spread of a report: documented at Sabana Grande, where among 33 interviewees hearsay of cures and sightings far outnumbered visions, and in Goulder's Bigfoot parallel. It requires no unusual psychology at all.
4. Fit to the earliest Christian reports: the bare list, with no description of what any group saw, is compatible with the second, third and fourth kinds. The detailed Gospel scenes of conversation, touching and eating, taken as accurate, are unlike any collective case examined. That leaves the visionary account with three options: the details grew in the retelling, the groups' experiences were more varied than the narratives say, or the comparison simply has no match. A defender of the traditional view reads the same gap as evidence that the groups met something real. The modern cases cannot decide between those readings. So the verdict on the groups depends heavily on the verdict about the later narratives, which belongs to the sibling article.

Part 6: Paul and James

Paul and James are the two cases that apologists treat as decisive, because neither is recorded as a follower grieving Jesus when his appearance occurred.  The evidence for each is different, so each is assessed alone.

Paul: what his letters record

Paul is the only person on the list whose own words about his experience survive.  He says he persecuted the church:  "beyond measure I persecuted the assembly of God and ravaged it."⁸  He describes his earlier life as a Pharisee, "concerning zeal, persecuting the assembly; concerning the righteousness which is in the law, found blameless."⁸  He says he "advanced in the Jews' religion beyond many of my own age", "being more exceedingly zealous for the traditions of my fathers."⁸  Later he calls himself "the least of the apostles", "because I persecuted the assembly of God."⁸  He reports what the Judean churches heard about him:  "He who once persecuted us now preaches the faith that he once tried to destroy."⁸  He describes the experience itself in a few words:  "Haven't I seen Jesus Christ, our Lord?"⁸  "he appeared to me also";⁸ and God was pleased "to reveal his Son in me."⁸  He also reports other "visions and revelations of the Lord."⁸

The three accounts in Acts add a light from the sky, a fall to the ground and a voice.⁸  They are later, and they come from one author, so they are not three independent witnesses.  They also differ on what Paul's companions experienced.  Acts 9:7 says the men "stood speechless, hearing the sound, but seeing no one."  Acts 22:9 says they "saw the light" but "didn't understand the voice."⁸  On any view, Acts reports an experience in which Paul perceived more than the people with him.

Paul: what fits and what does not

Grief has nothing to work with.  No source records that Paul knew Jesus or had any bond with him to mourn.  Hostility to a movement does not by itself rule out strong feelings about its leader, so this is a statement about the record, not a certainty about Paul's inner life.  But a grief model needs a bereavement relationship, and for Paul none is recorded.  Habermas and Licona are right that Paul "did not appear to have been in the frame of mind" for a grief vision.²⁸  Licona's book makes the same point.³⁴  The bereavement studies of Part 3 are not evidence about Paul.

A non-grief vision is a separate question.  Goulder's "conversion vision" needs no grief.  People who opposed Christianity have reported sudden visions of Jesus.  Allison cites Sadhu Sundar Singh, who "vigorously opposed the Christian message, stoning preachers and burning Bibles, until one day a dramatic vision of Jesus flipped his life."¹  Sundar Singh's own account, given years later and printed in 1921, adds what the summary leaves out.  He had found no peace in his own religion and decided to kill himself under the morning train unless God answered.  He prayed, "O God, if there is a God, wilt thou show me the right way or I will kill myself", then saw a light and "the form of the Lord Jesus Christ", and heard a voice say, "How long will you persecute me?"³⁶  Those words echo the voice in Acts.  His biographers note that he thought he did not then know the story of Paul's conversion, "though, of course, on a point of that kind the human memory cannot be implicitly relied on", and that he stressed the part the Bible had played.³⁶  Such cases are discussed in Visions of Jesus Outside Islam.  They show that a hostile person can report a convincing vision of Jesus.  They are self-reports, and they do not show what caused any of them.  Sundar Singh's case also came in acute crisis and after a prayer for a sign, so it is not a case of a vision with no inner turmoil.  Paul had also heard what the Christians claimed, since he persecuted them for it.  Allison:  "Even the pre-Christian Paul must have heard claims of people seeing the risen Jesus."¹  So Paul did not start from nothing.

The inner conflict is supplied, not recorded.  Lüdemann reads Romans 7 as Paul's hidden conflict before his conversion and speaks of Paul's "radical sense of guilt and unworthiness".¹³  He concedes that the biographical reading of Romans 7 "is now generally repudiated".¹³  Paul's direct statements about his earlier life speak of zeal and of standing under the law:  "found blameless" and "more exceedingly zealous".  They do not describe inner turmoil, but they are not a full inner history either.  Allison judges that Lüdemann's story "fits the facts, the facts hardly require it", and that "long-distance diagnosis of Paul's psychological state during a time for which we have only minute residues of evidence is more than tricky."¹  He adds that, in modern interviews, "as often than not, there was seemingly nothing distinctive about their emotional state at the time" of an apparition.¹  That cuts both ways:  a vision does not need a conflicted Paul, and the absence of a recorded conflict does not rule a vision out.

Diagnoses are unsupported.  Some writers have proposed epilepsy for Paul;¹  Habermas and Licona argue against a "conversion disorder."²⁸  No ancient source records the medical history that either would need.  Bergeron and Habermas themselves note that "hallucinations are not part of the diagnostic criteria" of conversion disorder.⁷  This article sets both labels aside.

Verdict on Paul.  A grief model has no evidential foothold in Paul:  high confidence in that statement about the record, though not certainty about his inner life.  A non-grief vision in a person who opposed Christianity has reported analogues, whose causes remain disputed.  That makes a visionary interpretation of Paul's report possible, without demonstrating the proposed psychological cause.  The specific reconstructions of Lüdemann and Goulder are speculation:  low confidence.  Paul's case is therefore the strongest single objection to a grief theory and a weaker objection to a visionary theory.

James: what is recorded

Paul says only "Then he appeared to James."⁸  Paul met James on his first visit to Jerusalem and calls him "James, the Lord's brother."⁸  The evidence that James did not follow Jesus during his ministry comes from the Gospels.  John says, "For even his brothers didn't believe in him."⁸  Mark has people close to Jesus go out "to seize him; for they said, 'He is insane.'"  Then his mother and brothers arrive.⁸  Acts places Jesus' brothers with the disciples soon after his death.⁸  A later story in the Gospel of the Hebrews, preserved by Jerome, has Jesus appear to James first;  Lüdemann concludes that "the rest of the text gives us no reliable information of any kind."¹³  The same story makes James a follower before Easter:  "James had sworn that he would not eat bread from that hour in which he drank the cup of the Lord until he should see him rising again from among those that sleep."³⁷  So the one ancient narrative of the appearance to James found for this article puts him at the Last Supper, already expecting the resurrection.  If it preserves anything early, it counts against the apologetic picture of a hostile skeptic converted by a vision.  If, as Lüdemann judges, it preserves nothing reliable, it counts for neither side.

Taken together, these sources give some evidence of a change:  not a follower during the ministry in Mark and John, among the believers soon after in Acts, and, by Paul's second visit to Jerusalem, one of those "reputed to be pillars".⁸  What none of them says is when James came to believe.  John's statement concerns a time during the ministry.  It is not an account of James's conversion.  Allison lists three possible orders:  James believed before Easter and then saw Jesus;  James saw Jesus and then believed;  or James believed after Easter, perhaps on others' reports, and then saw Jesus.  He writes:  "The sad truth is that we do not know the circumstances of the appearance to James. We know not where it occurred nor when it occurred. We cannot characterize James' state of mind at the time nor determine whether he had already thrown in with the Christian cause."¹

James: what each side claims beyond the evidence

Habermas and Licona argue that "there is no indication that James was stricken by grief over his brother's death."²⁸  That is true, and it is equally true that there is no indication that he was not.  The sources say nothing about his feelings.  Lüdemann, for his part, has James "caught up in the excitement" and perhaps among the five hundred.¹³  That is also supplied, not recorded.  Allison's comment fits both:  apologists favour the order in which a vision converts a skeptic "for the apologetical payoff", while the evidence does not exclude the other orders.¹  What can be said is that, if Paul's order is right, James's experience came after Peter's, and "so almost certainly after James had learned of others seeing Jesus."¹

Verdict on James.  Undetermined.  The evidence that James changed from outsider to leader is modest but real.  The evidence for the order of his belief and his experience is missing, so the case cannot be used as strong evidence by either side.  A skeptic cannot show that grief or contagion produced James's experience, and an apologist cannot show that it converted a hostile unbeliever.

Part 7: From seeing to "he was raised", and the tomb

The interpretation gap

Even if every appearance were a vision, something more is needed to get from "we saw him" to "God raised him from the dead."  Modern mourners who see a dead husband conclude, if anything, that he lives on in some other world, not that his body has left its grave.  In Elsaesser's survey, belief in life after death rose sharply after such experiences.²⁰  William Lane Craig puts the apologetic case:  "the bereaved do not as a result of such visions—however real and tangible they may seem—conclude that their deceased loved one has come physically back to life."²⁹  N. T. Wright puts it more sharply:  "The more 'normal' these 'visions' were, the less chance there is that anyone, no matter how cognitively dissonant they may have been feeling, would have said what nobody had ever said about such a dead person before, that they had been raised from the dead."³⁸

Ancient sources show what other readings were at hand.  Allison, who does not think the gap decisive, still calls the question "central to the discussion" and says Lüdemann "glides over the question far too quickly."  He notes that "Half of the Jewish texts from 200 BCE–100 CE that speak of an afterlife do so without mentioning resurrection", so that "It would have been easy enough for Peter, after seeing Jesus, to declare that God had vindicated his lord without using the concept of eschatological resurrection."¹  One Jewish text shows that other reading at work.  Second Maccabees teaches a bodily resurrection to come:  a dying martyr says that "the King of the world will raise us who have died for his laws up to an everlasting renewal of life."⁸  Yet the same book tells of a "dream worthy to be believed" in which Judas Maccabeus saw Onias, a high priest who had been murdered, "invoking blessings on the whole body of the Jews", and then the prophet Jeremiah, who spoke to Judas and handed him a gold sword.⁸  The book presents the two dead men as active after death and praying for Israel.  It does not say they were raised.  Luke names two more first readings of a sudden appearance.  When the risen Jesus stands among the disciples, they "supposed that they had seen a spirit";  when Peter, thought to be in prison, is reported at the gate, the believers say, "It is his angel."⁸  So the period offered at least three ways to read the sight of someone who should not be there:  a spirit or angel, a righteous man alive with God, or a raised body.  The first two needed no empty grave.

This evidence has limits.  Judas's experience is called a dream, and it is told in a history written to encourage its readers, not by a witness.  Onias did not lead a movement that expected the end of the age.  Luke wrote decades later, and his first scene argues against the very reading it reports.  None of these texts shows what a group expecting the resurrection soon would make of a sight of its dead teacher.

Two skeptics agree that the gap is real.  Komarnitsky says the needed reinterpretation is so large that "the hallucination seems secondary to the point of not even being necessary."¹⁵  Ehrman offers a bridge:  followers of an apocalyptic prophet who expected God to raise the dead at the end of the age would read any sight of him alive in those terms.  "In that view, a person who was alive after having died would have been bodily raised from the dead, by God himself, so as to enter into the coming kingdom."  He points to Mark's report that Herod Antipas thought Jesus was John the Baptist "raised from the dead", so "some such belief was not implausible."⁹

Allison now accepts Ehrman's reply.  In 2005 he had defended a version of Wright's argument.  In 2021 he wrote:  "I now think I was wrong and judge the criticism of Ehrman, How Jesus Became God, 185–6, to be cogent."  He gives two reasons.  The sightings came "to people whose religious convictions had been thoroughly molded by Jesus", and "Jesus himself had spoken of the new age, with its prefatory resurrection, as near".  And "many apparitions are not vaporous or ethereal but realistic, convincingly lifelike", so that people who met such a figure "would not have thought they were meeting a 'soul' or 'ghost.'"¹  That is the strongest form of the skeptical answer, and it comes from a scholar who believes that Jesus appeared.  Ehrman's Herod example cuts both ways.  In Mark, Herod says "John the Baptizer has risen from the dead, and therefore these powers are at work in him" about a living man working miracles in public, not about a vision.⁸  It shows that "raised" was a word for someone back in a body.  That fits Allison's lifelike apparitions, and it fits just as well the traditional claim that the disciples met Jesus in his body.

Carrier answers the gap another way, with a resurrection in a new body that left the old one in its grave (Part 2).  By his own account, though, the Pharisees already believed that souls of the dead could "hold conversations with the living, all before the general Resurrection even happens", which is a reading of such an experience that needs no resurrection at all.¹⁶  His pagan example also points the other way.  He writes of "Theseus, who rose from the dead to fight at Marathon".¹⁶  Plutarch, the source for that story, writes that "many of those who fought at Marathon against the Medes thought they saw an apparition of Theseus in arms rushing on in front of them against the Barbarians."  His word is φάσμα, an apparition.  The Athenians responded by honouring Theseus "as a demigod" and later, on an oracle's order, by finding his bones on the island of Scyros and bringing them home.³⁹  So an ancient report of many people seeing a dead man at once was read as the apparition of a hero whose bones still lay in a grave, not as a resurrection.  Plutarch wrote about six centuries after the battle, so the story shows how such a sighting was understood in his day, not what the soldiers saw.

The nearest modern test

The closest modern comparison found is the Lubavitch movement after the death of its leader, Rabbi Menachem Mendel Schneerson, in 1994.  Many of his followers believed he was the Messiah.  Allison summarizes what happened next:  "many Lubavitchers dropped this belief", others "said that he would soon rise from the dead or return as the Messiah", others sensed him as a spiritual presence, and "A few went even further, claiming that he had already been resurrected."¹  David Berger, a historian who opposes the messianists, judged in 2001 that "almost certainly a substantial majority" of the movement held that the Rebbe "will soon return to complete the redemption in his capacity as the messiah", and that "a growing number of messianists began to assert that the Rebbe never died".⁴⁰  Neither writer gives counts, and they differ in emphasis.  Both agree that expecting a future return was common and that claiming a resurrection already accomplished was rare.  Apparitions followed.  The anthropologist Yoram Bilu collected "a total sum of 76 testimonies of apparitions" from 1994 to 2010, among followers who "maintain that the Rabbi, the designated Messiah, has not died."⁴¹  Most were seen by one person, and "A few simultaneous apparitions, usually involving a dyad, were reported too."⁴¹

This case cuts both ways.  It shows that a messianic movement can survive its leader's death, deny or reinterpret the death, and report sightings of him, all without a missing body.  That supports Komarnitsky's order of events:  the conviction that the Rebbe was the Messiah and had not really died came first, and the reported sightings grew more frequent over the following years.⁴¹  It also shows that the step to "already raised" was rare there.

Komarnitsky answers that the difference has a cause in Lubavitch doctrine.  "In Lubavitcher theology, a 'prince,' in this case Rebbe Schneerson (who did not designate a successor), must be present in this world in some physical capacity in order to mediate the world's divine force or the world would cease to exist. Since the world still existed, Rebbe Schneerson had to still be physically present in the grave; he could not have been resurrected from the dead and transported up to heaven as Jesus' followers believed about him."¹⁵  If he is right, the Lubavitch case cannot show how often a group without that doctrine would claim a resurrection already accomplished, because the movement's own theology ruled that answer out.  His authority for the doctrine is the work of the anthropologist Simon Dein, which was not read for this article, so the doctrinal claim is reported here as his.  The answer has limits of its own.  It removes a counterexample;  it does not supply a case in which visions led a group to say "already raised".  And one movement, chosen because it is the closest parallel, cannot measure what comparable groups normally conclude.  Allison's list of disappointed movements does include one partial parallel:  after Sabbatai Sevi became a Muslim in 1666, some followers said that only "his shadow" remained on earth while "his body and soul are taken into Heaven".¹  That was a claim of an exaltation already accomplished, made against the plain facts, though it followed an apostasy rather than a death.

Verdict on the interpretation gap.  A real gap, partly filled.  Ancient texts show that the sight of a dead person had readings that needed no resurrection and no empty grave:  a spirit, an angel, a soul alive with God, a hero's apparition.  So "raised" needed something beyond the sight itself.  Apocalyptic expectation and dissonance can supply it, and Allison's version of that answer, from Jesus' own teaching that the resurrection was near, is the strongest reply and is not refuted.  They can explain how "raised" became the word for the experiences.  What they leave open is why this group said the resurrection had already happened, rather than that it would happen soon.  The Lubavitch case shows that the second answer was the common one there, but if Komarnitsky is right it sets no base rate, and the Sabbatean case shows that a group can assert an exaltation already accomplished against the evidence.  Moderate confidence that this is a genuine difficulty for visionary accounts, low confidence in any claim that it is decisive.

The tomb

An account of experiences says nothing about the body.  Wright's argument depends on it:  "if the tomb was still occupied by the dead body", he writes, the disciples would have concluded "he is still dead and buried."³⁸  The sibling article The Empty Tomb grades the links one by one and tentatively favours a located tomb found empty, while holding several links as uncertain.  That verdict is not treated here as settled fact.

If the tomb was not found empty, or the story of its discovery grew later, a visionary account needs nothing more.  If it was found empty, a visionary account needs a second explanation for that, such as reburial or removal, and the combined account has to be judged as a whole.  Combining two explanations is not cheating.  Each part must have its own evidence, and the combination must not require events that the sources contradict.  Habermas and Licona's objection that "hallucinations do not account for the empty tomb"²⁸ is true and beside the point:  none of the proponents examined here claims they do.

Part 8: The mechanisms compared

The table separates three different questions for each proposed mechanism.  Is the kind of experience documented in modern reports?  Does it fit what the earliest sources report?  Is there evidence for the specific story told about Peter, Paul or James?  A "yes" to the first does not carry over to the second or third.  And a modern report documents that the experience happened to someone, not what caused it.

Are combined explanations ad hoc?

Craig argues that a skeptic "cobbling together different, unrelated cases of hallucinatory experiences" shows "there's nothing like the resurrection appearances in the psychological casebooks."²⁹  There are two different claims in that sentence.

The first is that combining mechanisms is illegitimate.  It is not.  Real events often have more than one cause, and the earliest sources describe different kinds of experience for different people:  a follower who denied Jesus, groups, a family member, a persecutor.  Allison makes the same point against both sides:  "Apologists and skeptics typically write as though the appearances are a single phenomenon requiring a single explanation."¹

But legitimacy is not adequacy.  That each kind of experience has occurred to someone, somewhere, does not show that each occurred to these people in this order.  A combined account has to show evidence for its own steps, and the same test applies to each model examined.

1. Lüdemann's account needs a guilt-driven Peter, a mass ecstasy among five hundred, and a Paul with a hidden conflict. None of the three explanatory steps is recorded. The report of the five hundred supplies no description that could establish ecstasy rather than another kind of experience.
2. Goulder's needs a conversion crisis for Peter as well as for Paul, and a collective delusion for each group. Goulder draws Peter's crisis from Mark's scenes of his boasting, sleeping and denial,¹⁴ which is an inference from the Gospels, not a record of the setting of Peter's experience.
3. Ehrman's dispenses with collective visions and uses a few individual visions plus story growth. Its cost is that it must explain how the group claims arose. The Twelve belong to the older formula Paul had taught by about AD 50; the five hundred and "all the apostles" are securely attested in his mid-50s letter. Ehrman proposes storytelling within a few years, but no surviving source records that development. Fewer collective visions do not by themselves mean fewer unsupported assumptions.
4. Komarnitsky's needs belief to come before visions. No ancient source records the order either way.

Diversity therefore cuts both ways.  It is a fair challenge to any single-cause theory, and it is also a challenge to a combined account when each kind of report needs its own unrecorded step.  The traditional account faces the same record from the other side:  it covers all the reports with one cause, but what the groups experienced is just as undescribed for it as for its rivals.

The second claim is that the combination as a whole has no close parallel.  That is true, as far as this article's sources go.  Allison, whose 2021 book collects more modern parallels than any other work examined here, says so plainly:  "I know of no close phenomenological parallel to the series of likely events as a whole."¹  He lists what makes the series unusual:  "a missing body plus visions to several individuals plus collective apparitions plus the sense of a dead man's presence plus the conversion vision of at least one hostile outsider."  Each item has modern parallels.  The set does not.

What the lack of a parallel can and cannot show

Any historical series described in enough detail has no exact parallel.  The modern cases are also not a fair sample:  striking cases get reported and collected, ordinary ones do not, and the items in Allison's list are not independent of each other.  So "no parallel to the whole" is not, by itself, evidence for or against either account, and this article does not count it as a separate weight.  Allison's statement is his own judgment from the cases he knows, and it is reported here as that.

What matters is which particular observations each account struggles to predict, given how secure each observation is.

1. Paul's change from persecutor to apostle, after an experience he calls seeing Jesus. Secure: it rests on his own letters. A visionary account needs a non-grief vision in a hostile man. Reported cases exist, but those examined came in a crisis, and no crisis is recorded for Paul.
2. Claims of appearances to groups. The Twelve are in the formula taught by about AD 50; all three group claims are secure as claims in Paul's mid-50s letter. What the groups experienced is not secure. A visionary account is compatible with the bare claims. The Gospel scenes, if accurate, have no match among the cases examined.
3. The claim that Jesus had already been raised. Secure as the early claim. A visionary account needs an interpretive step whose nearest modern test is ambiguous.
4. A tomb found empty. Tentatively favoured by the sibling article, not established here. If granted, a visionary account needs a second explanation.

The traditional Christian account includes both a bodily resurrection and Jesus' subsequent appearances to followers and Paul.  Granted that whole account, all four observations are expected.  A resurrection by itself does not require Jesus to appear to these particular people.  The account's fit to the reports must therefore be distinguished from independent confirmation of it: the account and these observations come from the same sources.

What historical method can and cannot do

It is sometimes said that historical method cannot test a divine cause.  That needs care.  Historians routinely compare unrepeatable causes by the traces each would leave.  A resurrection account makes claims about traces:  a vacated tomb, experiences of a figure who spoke and could be touched, experiences shared by groups.  Those traces can be assessed, and this article and its siblings assess them.  What the evidence cannot supply is a frequency for such a cause, the way bereavement studies supply a frequency for grief experiences.  So the comparison cannot be reduced to a rate, and the weight a reader gives each account depends partly on what the reader already thinks possible.

That is the force of Allison's remark that, without a prior commitment, "there is nothing in them that determines the nature of the experiences of Peter and the twelve and the others shortly after Good Friday. Historical knowledge just does not reach that far."¹  It is a judgment about the surviving evidence, not a rule that closes the question.  A new early source, or a well-documented modern case of a group independently reporting the same figure in conversation, could change it.

Objections and responses

Objections from defenders of the resurrection

1.  "Hallucinations are private.  A group cannot share one."  Read with its authors' own qualification, the claim is narrower than it sounds.  Bergeron and Habermas grant that "when collective visionary experiences occur, some individuals in the group may experience similar but not identical personal hallucinations."⁷  Groups reporting simultaneous, differing experiences are documented, and the 1894 census also prints small cases, pairs in the examples read, who describe the same brief, silent figure (Part 5).  No case examined shows a larger group independently reporting the same familiar person talking with them.  No skeptical model examined needs that, and Paul's list does not say the twelve or the five hundred saw the same thing.

2.  "The disciples were not expecting anything, and hallucinations need expectation."  Bereavement experiences come to people who expect nothing (Part 4).  Expectation is prominent in crowd visions, as at Sabana Grande, and Zusne and Jones make prior information a condition of collective hallucination.³⁰  But once Peter had reported an experience, as Luke says he did,⁸ the others were no longer without expectation.  Allison asks, of the five hundred, whether "the large assembly was in an enthused or expectant state" if those who had already seen Jesus gathered them.¹

3.  "Visions do not speak.  The appearances include words."  Bereavement experiences sometimes include hearing the dead person speak, with words that carry a meaning:  13.3 percent of Rees's whole sample heard the dead spouse.²  What the documented cases lack is a long two-way conversation, and above all a conversation with a group.  O'Connell argues that such a conversation would be impossible for a hallucination.³¹  Part 5 sets out why that is a limit of his sample and his premise rather than a proof.  The difference is real, but it applies to the Gospel scenes, not to Paul's list.

4.  "Too many different people, places and circumstances."  Habermas and Licona argue that "It pushes credulity beyond reason to regard every last one of these appearances as hallucinations."²⁸  Each kind of report on the list has modern parallels:  a grieving individual, a group, a family member, a hostile outsider.  The census found 44 percent of collective experiences outdoors.⁵  The diversity is a fair challenge to a single-cause theory.  It is also a fair challenge to a combined account, wherever each kind of report needs its own unrecorded step, and Part 8 applies that test to each model.

5.  "Hallucinations do not transform lives."  Bereavement experiences change beliefs and decisions.  In Rees's study, "The four who rejected subsequent remarriage were all hallucinated" and gave as their reason "the feeling that the dead spouse was opposed to it."²  In Elsaesser's survey, 60.1 percent said their fear of death had decreased or disappeared, and belief in life after death rose "from 68.9% to 93.0%."²⁰  Goulder's answer is that conversion visions change lives by definition.¹⁴  None of the modern cases examined documents the whole sequence of an experience founding a movement and sending its recipients out on a mission.  Allison replies that the disciples were already missionaries before Easter, so a resumed mission is what one would expect.¹

6.  "Ancient people took visions of the dead as proof the person was dead."  This is the strongest objection.  Ancient texts support its premise:  2 Maccabees presents a vision of the murdered Onias and of Jeremiah as the righteous dead alive with God, Luke's characters reach first for "a spirit" and "his angel", and Plutarch's Marathon soldiers saw an apparition of Theseus whose bones were still in the ground.  But Allison, who once pressed this objection, now judges Ehrman's answer from apocalyptic expectation "cogent".  Part 7 assesses it as a real gap, partly filled by apocalyptic belief and dissonance.  The Lubavitch case, where only "a few" claimed a resurrection already accomplished, is the nearest test, but Komarnitsky argues that its doctrine barred that claim, so it cannot fix how rare the claim would be elsewhere.

7.  "Paul hated Jesus and James did not believe in him."  For Paul, no bond with Jesus is recorded, so a grief model has nothing to work with.  A non-grief vision remains possible.  For James, the evidence suggests a change from outsider to leader, but his state of mind and the order of his belief and his experience are not recorded (Part 6).

8.  "Hallucinations do not explain the empty tomb."  True.  None of the proponents examined here claims they do.  The tomb needs its own explanation, and its own evidence is weighed in The Empty Tomb.

9.  "The disciples touched Jesus and watched him eat."  Those details are in Luke and John, written decades later.  If they are accurate reports of what groups experienced, no collective case examined resembles them.  That leaves a visionary account three options:  the details grew, the experiences were more varied than the scenes say, or the comparison simply has no match.  A defender of the traditional view reads the same gap as evidence that the disciples met something real.  The modern cases cannot settle which reading is right;  the age and reliability of the scenes can, and that question belongs to Legendary Development.  Lüdemann holds that "the emphasis on a reality of the risen Jesus which could be perceived with the senses developed only later".¹²

10.  "They were willing to suffer and die for what they saw."  Sincerity does not tell between the two accounts.  Every model examined here holds that the witnesses sincerely believed what they reported.  Ehrman:  a vision "does not have to be veridical. It has to be believed."⁹  The evidence for the apostles' fates is weighed in Did the Apostles Die for Their Claims?.

11.  "The appearances lasted too long and went on for weeks."  Acts says Jesus appeared "over a period of forty days and speaking about God's Kingdom."⁸  Bereavement experiences can recur for years, so a series of experiences over weeks is not unusual.²  Single experiences are another matter.  In Bilu's Lubavitch sample, "Most apparition sightings are brief, ranging from a few seconds to a minute or two".⁴¹  Long teaching sessions with a group are unlike the documented visions.  But they are reported in Acts, decades later, not in Paul's list, so this objection, like objection 9, depends on the age of the later narratives.  The timing of the appearances as a whole is examined in The Resurrection Belief: Event or Process?.

12.  "Hallucinations would have shown a glorified Jesus, and the Gospels do not."  This is O'Connell's main argument: "if the appearances were hallucinations, glorious appearances of Jesus would have occurred."³¹  It requires both that expectation fixes the experiences' content and that early glorious stories would have survived.  O'Connell uses Luke's luminous appearance to Paul as a reason to expect preservation; citing that scene does not refute him.  The open questions are whether every experience must follow the proposed expectation and whether Luke's acceptance of one glorious appearance guarantees preservation of all earlier ones (Part 5).

Objections from skeptics

13.  "Allison's own family produced a list just like Paul's."  Allison says that after his father's death relatives and friends reported encounters, and that he "could compose a little list like Paul's."¹  The parallel is real for individual experiences told in sequence.  But the same author says no known case matches the whole early Christian series, which adds group experiences and a hostile outsider.¹  His family list contains neither.

14.  "Mass sightings happen.  The Virgin Mary has appeared to thousands."  Ehrman writes that there is "modern, verified eyewitness testimony" that Mary "has appeared to hundreds or thousands of people at once."⁹  The best-documented cases show crowds reporting different things in the presence of a real object:  the sun at Fatima, lights at Zeitoun, clouds and an old woman at Sabana Grande (Part 5).  They support group experience of the second kind.  They do not show thousands independently reporting one and the same figure.

15.  "Bigfoot waves show how a group comes to see what it expects."  Goulder's parallel fits a report of a vague sighting spreading through an excited community.¹⁴  It fits the bare report of the five hundred.  It fits a known individual speaking to a small group less well, since the Bigfoot reports were of a creature in the landscape, with footprints, not of a familiar person in conversation.

16.  "Only Paul mentions the five hundred, so the event may never have happened."  Ehrman makes this point.⁹  It is fair.  Paul adds that "most of whom remain until now",⁸ which suggests he expected no contradiction, but he gives no names or place, and no record shows anyone checking.  The report can be neither confirmed nor dismissed.

17.  "Grief and guilt explain Peter completely."  Grief can explain an experience like Peter's.  The guilt part is not shown by the bereavement evidence, though one study pairs self-reproach with a later sense of presence after sudden loss (Part 3).  The specific story of Peter's inner life is not recorded anywhere.

18.  "Apologists hide the fact that most bereaved people have these experiences."  The large figures, 47 to 82 percent in Kamp's review, include felt presence, voices and dreams.  Seeing the dead person is reported by a smaller share, about one in seven to one in four of all those asked in the studies that report it.  Both numbers should be given.

Objections to this article's verdict

19.  "The article is too kind to skeptics.  It concedes that it cannot decide."  The concession follows from the sources.  The earliest reports describe almost nothing of what was experienced, and the psychological claims about Peter, Paul and James are not recorded.  Allison, who believes Jesus appeared, reaches the same conclusion.

20.  "The article is too kind to apologists.  It counts the lack of a parallel as evidence."  It does not count it as a separate weight.  Any series described in enough detail lacks an exact parallel, and the modern sample is selective.  What the article weighs is the specific observations a visionary account struggles to predict:  Paul's conversion, the group claims and the claim "already raised", each graded by how secure it is (Part 8).

21.  "The article is too kind to skeptics.  It says history cannot test a resurrection."  It does not say that.  The traces a resurrection account predicts can be assessed, and are.  What the evidence cannot supply is a frequency for such a cause, so the final comparison cannot be reduced to a rate (Part 8).

Issues and unanswered questions

1. Olson 1985 in full. Only the abstract was read. Its disclosure figure, "Nineteen (54%)", fits no denominator the abstract names, and the sense-by-sense figures exist only as a later review's percentages of the 28 widows who had experiences.
2. The apologists' unpublished evidence. Gary Collins's statement that hallucinations cannot be shared is a personal communication. Bergeron's database search is described but not reported, and Gary Sibcy's survey is a letter. None can be checked.
3. Guilt has not been studied as a cause. No study read for this article measures guilt and then counts later experiences of the dead. The Harvard study's table that pairs sudden loss with sensed presence could not be read cell by cell. A study that measured guilt directly would settle the question better than any argument.
4. No study examined interviews every member of a group separately during a collective experience. Sabana Grande comes closest, with 33 interviews in a crowd of over 100,000, not chosen as a sample. The census cases are retrospective; O'Connell's are compilations; the Medjugorje instrument studies by Henri Joyeux and René Laurentin, which concern the seers' bodies rather than what they saw, were not obtained. A properly documented collective case, for or against concordant reports, would change the group verdict more than any argument.
5. Sources read only in abstract or through others. Grimby 1993, Castelnovo 2015, Kalish and Reynolds 1973, Yamamoto 1969 and Datson and Marwit 1997 were read as abstracts. The Icelandic studies of Erlendur Haraldsson are known here only through Allison. Komarnitsky's book was not read; his own essay was. Simon Dein's studies of the Lubavitch movement, on which Komarnitsky's doctrinal answer rests, were not read. The 2008 book by Aleman and Larøi that Licona cites for his rates was not read.
6. The later narratives. The group verdict depends on whether the Gospel scenes of touching, eating and conversation preserve early reports. That is the subject of Legendary Development, which is not yet written.
7. When James believed. No source settles the order of his belief and his experience.
8. Peter's and Paul's inner lives. No new evidence is likely. Claims about them on either side should be marked as inference.
9. A circulating quotation. A sentence attributed to Bilu, that "in one case no fewer than seven women saw the Rebbe", is not in Bilu's 2013 article, and no other source for it has been found. It is set aside in quarantine until one is.

Verdict

Visionary explanations are a family, and each member earns a different verdict.

1. A bereavement experience could have produced an individual report such as Peter's or Mary's encounter in John. Such experiences are commonly reported, normal, convincing to the person and sometimes include speech. High confidence that such experiences occur, moderate confidence in their fit to a single early report, no way to show that one happened here.
2. The specific psychological stories are unsupported. Guilt is neither shown nor ruled out as a cause of such experiences, and the feelings of Peter, Paul and James are not recorded. Low confidence in Lüdemann's and Goulder's reconstructions.
3. Group reports are neither impossible nor explained. Groups reporting simultaneous but different experiences, often with a real object in view, are documented, and so are small concordant cases of a brief, silent figure. No case examined shows a larger group independently reporting the same familiar person in conversation. Paul's list does not say what the early group experiences were. The detailed Gospel scenes, taken as accurate, are unlike any collective case examined, which is a limit of the comparison, not a proof either way.
4. Paul is a problem for grief, not for visions as such. High confidence that a grief model has no foothold in Paul's record; moderate confidence that a non-grief vision could produce a report like his.
5. James is undetermined.
6. The step from visions to "already raised" is a real difficulty, partly answered. In the ancient texts examined, such sights were read as a spirit, an angel, a soul alive with God or a hero's apparition, not as a resurrection. Jesus' own teaching that the resurrection was near is the strongest bridge across that gap. In the nearest modern case most believers expected a future return, but that movement's doctrine may have barred the other answer.
7. The tomb is a separate question, and a visionary account that faces an empty tomb needs a second explanation.

Taken together, a combined visionary and social account is compatible with sincere early appearance claims, and the popular apologetic replies to it are mostly overstated.  Its parts have reported modern analogues, but its specific steps for these people are unrecorded, and it struggles most with Paul's conversion, with the group claims if the Gospel scenes are early, and with the claim of a resurrection already accomplished.  The traditional account of a resurrection followed by appearances fits those observations, though it was framed from the same sources.  The absence of a measured frequency for a divine cause limits a numerical comparison; it is not itself evidence against that cause.  The surviving evidence does not settle which account is true.  That is a judgment about a thin record, not a rule that history cannot ask the question.

Sources