Chapter 8
Defining Terms with Care—and Why the Definition Carries the Whole Argument
Two stories, side by side. Both are presented to me, in the course of a year of doing this work, as evidence that something survives the death of the body. Both are told by sincere people. Both involve the words “I was on the ceiling, looking down.” And yet, by the time I am done with them, one of them is among the strongest single pieces of empirical evidence we have for the survival of consciousness, and the other—through no fault of the experiencer—cannot bear any evidential weight at all. The difference between them is what this chapter is about.
The first story is one I will keep brief here, because it belongs in full to Chapter 12. In August 1991, a thirty-five-year-old singer-songwriter named Pam Reynolds underwent a surgery so radical that, in plain medical terms, it required her to be killed and then brought back. She had a giant aneurysm at the base of her brain—a balloon-like bulge in an artery so deep and so dangerous that no normal surgery could reach it without rupturing the wall and causing a fatal bleed. Her surgeon, Robert Spetzler at the Barrow Neurological Institute in Phoenix, performed a procedure called hypothermic cardiac arrest, also called “standstill.” They cooled her body to 60 degrees Fahrenheit. They stopped her heart. They drained the blood out of her head and let her brain go silent. Her electroencephalogram, the device that measures electrical activity at the surface of the brain, went flat. The clicking earphones in her ears—monitoring the deepest brainstem signal, the auditory evoked potential—produced no detectable response from her brainstem. Her eyes were taped shut. By every clinical standard medicine has, she was dead, and not just for a moment but for the better part of an hour.1
When they warmed her, restarted her heart, and brought her back, she remembered being out of her body during the operation. She remembered watching from above the table. She described the bone saw the surgeon used to open her skull—a specific tool, a pneumatic drill called the Midas Rex, with a particular shape and a particular sound—and she described it accurately, including details about the case it came in. She remembered a comment one of the cardiac surgeons made about her femoral artery being too small. She remembered the music being played in the operating room. All of these details were verified, after the fact, by people in the room who had no reason to invent corroboration and several reasons to be skeptical of her report.2
That is one story. Now the second.
A woman writes to me. She is in her late seventies. Forty years ago, she says, she was very sick—she does not remember exactly what was wrong, but she was in the hospital, and they told her family she might not make it. One night, she remembers, she was on the ceiling looking down at her own body. There was a light. She felt peace. She came back. Ever since then, she has known there is a God and that her late husband is waiting for her.
I believe her. I have no reason not to. I think something happened to her that night in the hospital that changed her life and that she has held onto for forty years and that has, by every appearance, made her a kinder and braver person than she might otherwise have been. None of that is in question.
What is in question is whether her account, just as she has told it to me, can be used to argue with a hostile philosopher of mind that consciousness is not identical to brain activity. And the honest answer is no. We have no medical record. We do not know what was wrong with her. We do not know what her EEG was doing, because no one took an EEG. We do not know whether she was unconscious for thirty seconds or thirty minutes. We do not know whether the doctor or nurse said something within her hearing—before or after she lost full awareness—that her mind later wove into a remembered image. We have no witnesses to corroborate any specific detail, because no specific verifiable detail was reported. There is nothing to falsify and therefore nothing to confirm. The story is precious. It is not, in the technical sense I am about to develop, veridical.
An NDE becomes evidence for the survival of consciousness only when the experiencer reports something he or she could not have known by ordinary means, when that something turns out to be true, and when there are people or records outside the experiencer’s own memory that can confirm both halves of that claim.
The first case can do work. The second cannot. Not because the first is sincere and the second is not—both are sincere—but because the first carries within itself the kind of detail that lets a careful researcher check it, and the second does not. The job of this chapter is to spell out what makes that difference and how to measure it. Without that spelling-out, every chapter that follows is built on sand.
The word is veridical. It comes from the Latin veridicus, which combines verus, “true,” with dicere, “to speak.” A veridical report is, literally, a truth-telling report. In the technical literature on near-death experiences, the term has come to mean something more specific: an NDE in which the experiencer reports knowledge of facts in the external world—events, objects, people, conversations, places—that he or she could not have obtained by normal sensory means during the period of the medical crisis, and that turn out, on independent investigation, to be accurate.3
That is a mouthful. Let me say it again, slower. A veridical NDE is one where the experiencer—during the time he was unconscious, in cardiac arrest, on the operating table, drowning, or otherwise incapacitated—says afterward, “I saw X.” And then we check, and X is true. And the only way he could have known X was from where he says he was watching it—which is to say, from outside his body, while his body was in no condition to be doing the watching. That is the phenomenon. It is, on the face of it, the kind of phenomenon that, if real, breaks the standard physicalist account of consciousness. Which is why people fight about the cases.
I want to slow down here because the careful definition is the whole ballgame. If you ask a sympathetic NDE researcher what counts as a veridical case, you will get one answer. If you ask a hostile skeptic, you will get a much narrower one. If you ask a neutral methodologist—someone who is not trying to win a fight but who is asking how we should evaluate this kind of testimony—you will get something in between. My view, developed in the dissertation that lies behind this book and refined through analysis of more than five thousand individual cases, is that the neutral methodologist’s answer is the right one. The criteria should be strict, but not so strict that they exclude any case that does not happen on a videotape with a hidden numerical marker. They should be strict enough to exclude what could be explained by ordinary perception, ordinary memory, ordinary chance, and ordinary suggestion. They should be loose enough to admit what the medical setting and witness corroboration actually allow us to admit.4
Why does this matter so much? Because the whole evidential argument of this book—the case for substance dualism that Chapter 23 draws together—depends on it. If “veridical” is loose, the cases prove too much. They become a credulous heap, and any serious philosopher of mind can dismantle them by pointing out the cases that do not meet basic standards. If “veridical” is too strict, the cases prove too little. They become a tiny handful, easily explained as statistical flukes, and any committed physicalist can wave them away by saying that with enough hospital deaths, eventually one of them will get a few details right by chance. The right definition is the one that lets the strong cases stand and the weak cases be set aside, and that lets the reader see, in a transparent way, why one is which.
The chapter ahead has six tasks. First, I will set down the four criteria that, taken together, make a report veridical. Second, I will describe the three-dimensional scoring system I developed in my dissertation work—a way of giving each case a number that reflects its evidential weight. Third, I will note where this system stands in the longer history of NDE measurement, particularly its relationship to the Greyson Scale and the Ring Index, and why I needed something different from those. Fourth, I will walk through the verification methodologies—the practical work of checking a case—so that the reader can see what serious evaluation actually looks like. Fifth, I will distinguish three classes of NDE evidence, only one of which can carry the metaphysical argument. And sixth, I will face the major objections honestly: coincidence, cryptomnesia, confabulation, and researcher bias. By the end of the chapter, the reader will have what he or she needs to read Chapter 10 and the chapters after it with informed judgment—not as a sympathetic believer or a hostile skeptic, but as a fair evaluator of the strongest cases the literature has produced.
A case is veridical, in the sense that matters for the empirical argument, when it meets four criteria together. Any one criterion alone is interesting. Any two are suggestive. All four together constitute strong evidence. Let me take them one at a time.
The first criterion is the most basic. The experiencer must report something specific and checkable—a particular conversation, the placement of a particular object, the specific actions of a particular person, a piece of equipment with a particular shape, the layout of a particular room. “I saw a light” is not verifiable. “I saw the surgical resident drop his pen” is. “I felt loved” is not verifiable. “I heard my surgeon tell the anesthesiologist about his daughter’s soccer game” is.5
This is harder than it sounds. Many otherwise interesting NDE accounts are short on the kind of detail that lets a researcher check anything. Some of this is the nature of the experience—experiencers often describe themselves as overwhelmed, stretched, “more real than real,” struggling to put what they saw into ordinary words. Some of it is interview practice—early NDE researchers did not ask the kinds of follow-up questions that elicit the verifiable details, because they did not yet know how important those details would be.32 Some of it is the time gap between event and interview—the longer the gap, the harder it is to recover the kind of fine-grained detail that could be checked. The result is that even sincere experiencers often produce reports too thin to be falsified and therefore too thin to be confirmed.
What we are looking for is a report that contains specific factual claims about the world during the medical crisis. The more specific the claim, the higher the evidential weight if the claim turns out to be accurate. A claim such as “the doctor wore green scrubs” is verifiable but weak—most surgical doctors wear green or blue scrubs, and a guess gets it right roughly half the time. A claim such as “the doctor was using a Midas Rex pneumatic bone saw, and there was a smaller blade in the case beside it that he did not use” is verifiable and strong—there is no half-the-time guess that makes this come out right.
The second criterion is the one that does most of the heavy lifting against physicalist alternatives. The accurate detail must have been inaccessible to the experiencer through ordinary sensory means at the time of the experience. There are three ways inaccessibility can hold.
It can hold by physical barrier—the relevant event happened in another room, or behind the experiencer’s head while he was lying on his back, or in a hallway he could not have seen, or above the surgical drape that covered his face. It can hold by sensory incapacitation—the experiencer’s eyes were taped shut, or her ears were blocked by molded foam plugs, or she was under general anesthesia, or she was hypothermic and ventilated and unable to receive any external input through normal channels. It can hold by temporal factors—the relevant event occurred during a window of medically documented unconsciousness, with EEG showing no organized cortical activity (more on this baseline in Chapter 5), or during cardiac arrest with no detectable cerebral perfusion.6
Each form of inaccessibility shuts off a different kind of physicalist alternative. The physical-barrier form rules out simple visual perception. The sensory-incapacitation form rules out unrecognized incoming signals. The temporal-medical form rules out perception by a still-functioning brain. When two or more forms of inaccessibility hold at once—for example, the experiencer was in a hallway he could not see, while under anesthesia, while EEG-monitored—the case becomes very hard to explain by any normal route.33
The third criterion is corroboration. The accurate report must be confirmed by a source that does not depend on the experiencer’s own memory: a witness, a medical record, a contemporaneous note, a recording, an interview with someone present at the event. The reason for this criterion is straightforward. Memory is reconstructive. Even sincere reporters can, over time, fill in gaps with what they have learned later, and they will not always know they are doing so. Without a source outside the experiencer’s own recollection, we cannot tell what was reported at the time and what has been added since.7
The strongest corroboration is contemporaneous and multiple. A medical chart entry made by a nurse at the time, a recorded conversation, a statement from a surgeon noted before the patient described her experience, an interview with a witness conducted before the witness had heard what the patient said—these are the gold standards. Weaker corroboration is later and singular. A family member who heard the story years afterward and who was not present at the event provides only secondary support; he can confirm that the experiencer remembers reporting it that way, but not what actually happened in the room during the crisis.
The corroboration criterion is also the one that most reliably distinguishes the literature’s strong cases from its weak ones. In Janice Holden’s 2009 review of more than a hundred published cases of out-of-body perception during NDEs, she found that ninety-two percent of the cases involved at least some accurate veridical detail. But the strength of the cases varied enormously, and the variable that did the most to separate them was the quality of corroboration. Cases with multiple independent witnesses, contemporaneous records, and pre-disclosure verification were robust. Cases relying solely on the experiencer’s later report—however sincere—could not be evaluated as evidence in the same way.8
The fourth criterion is the most demanding intellectually. Even when the first three criteria are met, a case still has to be examined for the possibility that the experiencer obtained the information through some normal route that has not been considered. The standard alternatives are these. The patient could have overheard a conversation while still conscious or while under light sedation, before losing full awareness, and the heard content could have been incorporated into the later memory of the experience. The patient could have learned the relevant detail after the event—from a passing comment, from medical records, from a family member—and could have unknowingly incorporated it into his memory of the experience. The patient could have prior knowledge of the kinds of equipment and procedures used in his treatment—from television, from a relative who works in medicine, from a prior surgery—and could be retroactively projecting that knowledge onto a remembered scene. The patient could have made a vague claim that retrospectively fits the actual event because nearly any guess would have fit. None of these alternatives is impossible in any individual case. The veridical criterion requires that the case as documented give us reason to think the alternatives do not in fact apply.9
I will return to each of these alternatives in detail later in this chapter, when I discuss the major objections. For now, the point is methodological. A case that has been documented in a way that lets us examine these alternatives, and that survives the examination, has earned the label veridical. A case that has not been documented well enough to examine the alternatives has not—not because it is false, but because we cannot tell.
Some readers will wonder whether these criteria are too strict. Doesn’t requiring all four standards exclude many sincere reports? It does. That is the point. The job of the criteria is not to validate every NDE—it is to identify the cases that can carry weight in an argument with a determined physicalist. Many NDEs that fail the veridical test still matter for other reasons: pastorally, theologically, phenomenologically. They simply cannot be used as evidence for the metaphysical claim that consciousness survives brain shutdown. The whole next several chapters work with the cases that can carry that weight. The careful definition is what lets them do so.
The four criteria above tell us what to look for. The scoring system tells us how to weigh what we find. In the dissertation work behind this book, I found that the standard tools for measuring NDEs—the Greyson Scale, the Ring Index—were not designed to measure the thing I needed to measure. They were designed to measure the richness of an experience, not the strength of a case. So I built something different: a three-dimensional system that scores each case along three axes, each on a zero-to-ten scale, with the dimensions designed to track the four criteria above. The full system, with all the operational details, lives in Appendix A. What follows here is the working sketch.10
The first dimension scores the seriousness and verifiability of the medical event during which the NDE occurred. A score of zero means the experiencer says he “almost died” but there is no medical record, no clinical detail, no way to know what actually happened to his body. A score of ten means full medical documentation of cardiac arrest with continuous EEG monitoring showing no cortical activity for a measurable period, with anesthetic and brainstem-monitoring protocols in place, all noted in real time by clinical staff with no interest in the patient’s subjective report.
The middle of the scale tracks the increasing weight that documentation gives to a case. A score of three or four might apply to a hospital stay where the patient was very ill and unconscious but where no specific documentation shows the depth of unconsciousness. A score of five or six might apply to a documented surgical procedure under general anesthesia, where we know the patient was unconscious but we do not have continuous EEG. A score of seven or eight applies when the documentation includes evidence of cardiac arrest, defibrillation, or other clearly catastrophic medical events. A nine or ten requires the highest standard—full electrophysiological documentation that the brain was not, during the relevant window, doing the kind of work physicalist accounts of consciousness say it would have to be doing.11
The medical-context score is doing a particular kind of work in the overall framework. It does not, by itself, tell us anything about the experiencer’s report. It tells us how high a barrier the report must clear if it is to be explained by ordinary brain activity. A high medical-context score raises the bar for any physicalist explanation of the case’s veridical content. A low medical-context score leaves room for the possibility that the relevant brain activity could have supported the relevant perception, and the case’s evidential weight is correspondingly reduced.
The second dimension scores the strength of the report itself—how specific the accurate details are, how unlikely they are to have been obtained by normal means, how unambiguously they match the events as independently established. A score of zero means the report contains nothing externally checkable—only subjective imagery, feeling tones, encounters with figures who cannot be independently confirmed. A score of ten means the report contains multiple specific accurate details—at least one of which involves equipment, conversation, or behavior that the experiencer could not have known about by any reasonable normal route.
To score this dimension well, we must ask: how much of the reported content is checkable, how much of the checkable content is accurate, and how unlikely the accurate content is to have been obtained without the perceptual access the experiencer claims? A vague accurate detail (the surgeon was tall) is worth less than a specific accurate detail (the surgeon was using a Midas Rex pneumatic bone saw with a five-eighths-inch blade and a backup smaller blade in the open case beside him). A detail likely to be guessed correctly (the patient’s body was on a hospital bed) is worth less than a detail unlikely to be guessed correctly (the cardiac surgeon was complaining that the patient’s femoral artery was too small for the cannula).12
The veridical-quality dimension is the place where the four criteria of the previous section get translated into a number. Specificity, inaccessibility through normal means, and the ruling out of plausible alternatives all show up here. The dimension is, in effect, asking: how much work would a physicalist explanation have to do to handle this report by ordinary cognitive means?
The third dimension scores the strength of the independent confirmation. A score of zero means there is no source outside the experiencer’s own memory—no witness, no record, no document. A score of ten means there is contemporaneous documentation by multiple independent professional witnesses, with verification conducted by a researcher independent of the experiencer and the family, ideally before the experiencer had a chance to revise her memory.
Corroboration scores depend on three things: the number of independent confirming sources, the quality of those sources (a written record at the time is stronger than an interview years later), and the timing (verification conducted before the experiencer disclosed her account is stronger than verification conducted after, because pre-disclosure verification cannot be biased by what the experiencer has already said). The very strongest cases involve professional witnesses—surgeons, nurses, anesthesiologists—whose own credibility is staked on the accuracy of their professional records, and who often have no investment in confirming a patient’s spiritual claims.13
Add the three dimensions together and a case has a score from zero to thirty. In the dissertation, I found that 1,618 of the 5,278 cases analyzed—30.7 percent—reached what I called the “Strong” or “Exceptional” thresholds. Strong cases scored at least seven on each dimension and at least twenty-three in total. Exceptional cases scored at least eight on each dimension and at least twenty-seven in total. The thresholds are arbitrary in the sense that any thresholds are arbitrary—there is no fact of nature that says twenty-three is a different number from twenty-two. But they correspond to recognizable real-world differences. A case that clears twenty-three points has documentation, specific accurate detail, and independent corroboration of a kind that is very hard to dismiss. A case below seventeen points lacks at least one of those, and probably more than one. Chapter 14 walks through what the full distribution looks like.1434
To make the scoring concrete, consider the Pam Reynolds case from the chapter opening. Medical Context: 10. Hypothermic cardiac arrest with body temperature at 60 degrees Fahrenheit, blood drained from the head, EEG flat for an extended period, brainstem auditory evoked potentials absent. There is no stronger medical context in the literature. Veridical Quality: 9. Multiple specific accurate details, including the particular bone-saw model and shape, the conversation about the femoral artery, and the music playing in the operating room. The details are low-prior-probability and not plausibly available through normal sensory routes given her clinical state. Corroboration: 8 to 9. Verified by surgical staff after the fact, with the operative record matching key details. Pre-disclosure verification was not fully achieved (the surgical team had heard her account before all interviews were complete), which holds the corroboration score below ten. Total: roughly 27 to 28—an Exceptional case by the dissertation thresholds.
The 1,618 cases meeting Strong or Exceptional thresholds are not a curated list of the best stories. They are every case in the database that reached the threshold, including cases the author would otherwise have set aside, and excluding cases the author found personally moving but that did not meet the criteria. This is what it looks like to take the methodology seriously: it sometimes excludes what you wish it included, and sometimes includes what you wish it did not. That is the discipline of empirical work. It is the discipline this book asks the reader to bring along.
Anyone who has read in the NDE literature will know that there are several scales already in use. The two most important are the Greyson Near-Death Experience Scale and Kenneth Ring’s Weighted Core Experience Index. Neither, for the work this book is trying to do, is sufficient. Let me say briefly why, and what they are good for.
Bruce Greyson’s scale, published in 1983 in the Journal of Nervous and Mental Disease, is a sixteen-item instrument that scores an NDE on its phenomenological richness across four domains: cognitive (such as time distortion, life review), affective (such as feelings of peace, joy, encounter with light), paranormal (such as out-of-body sensation, vivid sensations), and transcendental (such as encounter with a mystical being or place, or with a border or point of no return). The scale produces a score from zero to thirty-two; a score of seven or higher is taken as the threshold for classifying an experience as an NDE rather than a generic medical episode.15
The Greyson Scale is the most widely used instrument in the field, and it has earned its place. It allows researchers to compare experiences across studies, to identify which medical events produce experiences with NDE-typical content, and to test for the presence or absence of core features. What it does not do, and was never meant to do, is measure the evidential weight of a case. It is not interested in whether what the experiencer reported was true—only in whether the experiencer reported the kinds of things NDErs typically report. A case can score very high on the Greyson Scale by being phenomenologically rich and very low on a veridical scoring system by lacking any externally checkable detail. The two scales answer different questions.
Kenneth Ring’s Weighted Core Experience Index, developed for his 1980 study Life at Death, has a similar character. It measures the depth and completeness of an NDE—how many of the typical elements were present, how vivid each was, how the experience unfolded. It is a phenomenological tool. It tells us about the shape of the experience, not the truth of any factual claim within it.16
What I needed—and what the dissertation system is—is something different in kind: a tool for sorting cases by their evidential weight, not by their experiential richness. The richest experience in the database might score zero on my system if its richness is all subjective. The thinnest experience in the database might score thirty if its single accurate report concerns specific equipment used during a documented cardiac arrest, witnessed by multiple medical professionals. The two ways of scoring are answering two different questions: what was the experience like, and what does the experience tell us about the world?
I want to underline that the dissertation scale is not in competition with the Greyson Scale or the Ring Index. It is an evidential tool that complements the phenomenological tools already in the field. A complete evaluation of an NDE case asks both kinds of question and uses both kinds of instrument. Chapter 9 takes up the phenomenological side of the picture in some detail, including the cross-cultural-consistency findings that depend on phenomenological measurement.
How does the practical work of evaluating a case actually unfold? The literature gives several good examples, and I will sketch the workflow now in plain terms, because it helps the reader to see what serious case evaluation looks like from the inside.
The first step is medical-record correlation. The researcher gathers the patient’s medical record from the relevant admission—the operative report, the anesthesia record, the nursing notes, the EEG strip if there was one, the cardiac monitoring data if there was one, the resuscitation log if there was one. These records establish what was happening to the body during the relevant window. They establish the medical-context score. They also produce a timeline against which the patient’s remembered observations can be compared.17
The second step is structured witness interview. The researcher identifies the medical staff and family members who were present during the relevant window and interviews them with a structured protocol—asking specific questions about what the staff did, said, or saw, and asking those questions in a way that does not lead the witness toward what the patient has reported. Where possible, the witness is interviewed before being told what the patient said. This pre-disclosure interview is the gold standard for corroboration; it cannot be subtly shaped by the witness’s desire to confirm the patient’s account, because the witness has not yet heard the account.18
The third step is timeline reconstruction. The researcher matches the patient’s remembered observations to the documented events of the medical procedure, asking: did the patient report seeing or hearing things that happened during the period of medical unconsciousness? Or did the patient report things that happened either before unconsciousness was complete or after consciousness had been restored? The temporal-impossibility criterion lives in this step. A report of an event that occurred during EEG-flat cardiac arrest is on a different evidential plane from a report of an event that occurred while the patient was being put under or coming back up—periods during which some sensory input may still have been registering at some level, even if the patient was not behaviorally responsive.
The fourth step is documentation review. The researcher writes up the case in a way that makes the evidence transparent: what the patient reported, what each witness independently confirmed, what the medical record shows, what alternative explanations have been considered and how they fare. The documentation is, finally, what allows a reader who was not present to make her own evaluation. The cases that do well at this step are the cases that survive scrutiny outside the original researcher’s study; the cases that do poorly tend to be the cases that other researchers cannot replicate or that turn out, on later investigation, to have weaker corroboration than initially claimed.19
The whole workflow is unglamorous. It is medical-records work and patient interviews and witness interviews and timeline-building and write-ups. It looks more like investigative journalism than like the popular image of paranormal research. That is, in fact, the right frame. The strong NDE cases are not stories that prove themselves by their drama. They are claims that were tested by the same kinds of methods used to test any other claim about events in the world.
The careful methodology this chapter has developed lets us see something that gets blurred in the popular literature: there are at least three different classes of evidence that come out of NDE research, and they support different kinds of conclusion.
The first class is veridical. These are the cases I have been describing. They involve specific, accurate, externally checkable claims about the world during a period of medical unconsciousness, with corroboration sufficient to rule out the major ordinary explanations. The veridical class is what the empirical case for substance dualism rests on. A small number of strong cases, properly documented, would be enough to make the case difficult for physicalism. A large number of strong cases, as the database actually contains, makes the case very difficult indeed.
The second class is phenomenological. This class consists of the consistency-across-cases evidence: the fact that NDErs from different cultures, different decades, different religious backgrounds, and different medical contexts report experiences with strikingly similar core elements—an out-of-body phase, a tunnel-like transition, an encounter with a luminous presence, a life review, an awareness of deceased loved ones, a return. The phenomenological evidence does not, on its own, tell us that any individual report is veridical. It tells us that something consistent and structured is happening to the dying mind, in a pattern that is not easily explained by random cultural construction. The phenomenological evidence is powerful in its own right, but it is a different kind of argument from the veridical one. It is taken up in Chapter 9 and developed across Chapters 18 through 22.20
The third class is transformational. NDErs, as a population, show measurable changes in their lives after the experience: reduced fear of death, increased valuation of love and relationships, increased meaning, increased spiritual orientation (often without specific denominational direction), and durable behavioral changes including occupational changes, marriage changes, and changes in ethical priorities. The transformational evidence is real and worth attention; it speaks to the existential weight of the experience for those who have it. But it is evidentially distinct from the veridical and phenomenological classes. It tells us that NDEs change people. It does not, by itself, tell us why the experience that produced the change is structured the way it is or whether the content of the experience is reliable. The transformational evidence is taken up in Chapter 22.21
The book’s central evidential argument—the part that, more than any other, claims to bear on the metaphysics of consciousness—rests on the veridical class. The phenomenological and transformational classes do real work in the surrounding chapters and in the pastoral sections at the end of the book. But the work this book asks the reader to take seriously, and the work the dissertation behind it tried to do at the highest empirical standard, is work in the veridical class. That is why this chapter has spent its time defining the term so carefully. The whole structure ahead is built on the foundation set here.
I have said several times that the veridical criteria are designed to rule out the standard physicalist alternatives. The honest thing to do, before moving into the case files, is to walk through those alternatives one at a time. They have to be presented at their strongest before they can be answered. If they are not—if we strawman them—the answer is worth nothing.
The first objection is the simplest. Hospitals are large places. Many people die or nearly die in them every year. Some patients, by chance alone, will produce reports that happen to match what was actually going on around them at the time. With enough trials, even highly unlikely matches will eventually occur. So the existence of any individual veridical-looking case proves only that statistics work as expected. It does not prove that consciousness survives brain shutdown.22
The objection is reasonable as far as it goes. It applies, however, to a different kind of case than the strong cases actually are. A coincidence explanation can handle a case where a patient remembers a single vague impression that turns out to match what was happening in some general way: she remembers seeing “people in green standing around me,” and this turns out to be roughly true of the surgical team. The match is real; the chance of guessing it is high; coincidence is a serious live possibility.
Coincidence cannot, however, scale to a case in which a patient reports a half-dozen specific, accurate, low-prior-probability details: the precise tool the surgeon used (a Midas Rex pneumatic saw rather than the older oscillating saw), the specific shape of the bit, a passing comment about an artery being too small, the music being played, a specific hand gesture by an anesthesiologist who briefly stepped away. Each individual detail might have, say, a five-percent chance of being guessed correctly. Six independent details, each at five percent, have a combined chance of about one in sixty-four million of matching by coincidence. Even if we are generous about the probabilities, the math gets very unfriendly to chance very quickly. And once we look at the cumulative case across hundreds of strong-scoring cases in the database, the coincidence explanation is no longer doing real work. It cannot account for the rate at which strong cases occur. It cannot account for the consistency of what is reported when reports are detailed.23
The second objection is more sophisticated. Cryptomnesia is the technical name for the phenomenon in which a person learns something, forgets that he learned it, and later remembers it as if it were original to himself. The objection runs: the patient may have learned the relevant details before the medical crisis, perhaps long before, and may be reproducing them now without consciously knowing where they came from. The detail looks like a veridical report, but it is in fact a recovered memory of ordinary information.24
Cryptomnesia is real. It happens. It is, however, easier to invoke than to make stick in the strong cases. The strongest cases involve details that the patient could not plausibly have learned beforehand by any normal route. A patient who has never been in an operating room, has never seen a Midas Rex bone saw, has no medical training, no family in medicine, and no reasonable prior exposure to the equipment cannot have a forgotten memory of the equipment to recover; the memory was never deposited.
The cryptomnesia objection also runs into trouble with the temporal-impossibility cases. A patient cannot have a forgotten memory of a conversation that occurred during her cardiac arrest, because the conversation occurred while she was incapable of receiving sensory input. Cryptomnesia operates on memory traces deposited at some earlier time; it does not generate accurate reports about events that took place during a window of medical unconsciousness.35 Where the strong cases involve such windows—documented cardiac arrest, hypothermic standstill, post-anesthetic burst suppression, and so on—cryptomnesia is inert as an explanation. It explains only what is not in dispute: that some claimed veridical content might, in some weak cases, be recovered prior knowledge.
The third objection is the one most commonly invoked and the one most worth taking seriously. Confabulation is the brain’s well-documented tendency to fill in gaps in memory with plausible-seeming detail, generally without the awareness of the person doing the filling-in. Patients recovering from cardiac arrest, severe trauma, or general anesthesia are in exactly the kind of cognitive state in which confabulation is most likely. So the objection runs: the patient, in trying to make sense of a fragmentary remembered experience, is unconsciously constructing detail to fit what she has since learned about her medical event. The detail is sincere and feels authentically remembered. It is, in fact, post-hoc confabulation.25
This objection applies, as it should, to the weak cases. A patient interviewed weeks or months after the event, with full access to family members, medical records, and her own subsequent reflection on what must have happened, has had ample opportunity to fill in plausible detail. Such an interview cannot, by itself, distinguish accurate veridical perception from sincere confabulation.
It is precisely against this objection that the methodological criteria of the strong cases earn their keep. Real-time verification is the answer. When the patient’s report is recorded shortly after the event—sometimes immediately on regaining consciousness, sometimes within hours—and when the medical staff is interviewed before they have heard what the patient said, the confabulation explanation has no foothold. The patient could not have learned the verifying detail from the staff if the staff had not yet been told what the patient was going to say. The patient could not have filled in plausible detail from medical records if she had not yet seen them. Where the documentation captures the report fast enough and the witnesses are sealed off from contamination, confabulation is ruled out by the structure of the investigation, not by argument.26
“Patients confabulate—they fill in detail after the fact without realizing they are doing so.” True, and that is exactly why the strong cases require pre-disclosure verification. When the medical team is interviewed about what happened during the operation before they have been told what the patient remembers, the verification cannot have come from confabulation; the staff have nothing to contaminate the patient with. Where the workflow is honest at this point, the confabulation explanation runs out of room to work.
The fourth objection is the broadest. It says: the NDE literature is curated by researchers who are sympathetic to it. Cases that did not pan out have been quietly discarded. Reports that did not fit have been edited or omitted. The cases we see are the survivors of a hidden filter, not a representative sample. Even if the cases we see look strong, we cannot trust the population they were drawn from.27
This objection has more force than NDE researchers sometimes allow. Curation does happen. Sympathetic researchers do prioritize cases that confirm their working hypotheses. Cases that do not confirm them are sometimes neither published nor preserved. Anyone who has read deeply in the literature has seen the shape of this bias.
The dissertation work behind this book tried to address the bias by pulling cases from two different sources with different curation patterns. The 832 cases drawn from peer-reviewed scholarly publications and academic books are, indeed, the curated cases—they have been through editorial filters, they have been written up by researchers with their own commitments, and they survive into the literature partly because they are interesting. The 4,446 cases drawn from the NDERF (Near-Death Experience Research Foundation) and IANDS (International Association for Near-Death Studies) databases are different. These are first-person reports submitted by NDErs themselves. The submission filter is much weaker; the database accepts what experiencers send in, with light editorial review for completeness and clarity but without the tight quality filtering that scholarly publication requires. The two source classes give us complementary windows. The scholarly cases are richer in documentation and weaker in representativeness. The database cases are weaker in documentation and stronger in representativeness.28
What did the analysis show? The same broad patterns appeared in both source classes. Strong veridical cases occurred in both. The proportion of cases meeting the Strong or Exceptional thresholds was somewhat lower in the database cases (because of weaker documentation) than in the scholarly cases, but the kind of veridical content reported, the kind of medical contexts in which it occurred, and the kind of corroboration that could be found were structurally similar. If the scholarly literature were merely a curated artifact, with no underlying real phenomenon, we would expect the database’s less-filtered population to look qualitatively different. It does not. The agreement between the two source classes is, in itself, an answer to the curation objection. The scholarly literature does filter; the unfiltered population shows the same shapes.29
I do not want to pretend the answer to the bias objection is complete. Some bias remains. NDErs who submit reports to public databases are themselves a self-selected population—more inclined toward reporting, more willing to talk about their experience, perhaps more inclined to interpret what happened in NDE-shaped ways. There are real methodological worries here, and the dissertation works through them with more care than this chapter can. The point I want to make is narrower: the strong cases are not artifacts of a single research community’s editorial choices. The phenomenon is robust across collection methods, even when the collection method is light enough to admit cases the sympathetic researcher would have screened out.
What does the careful methodology developed here let us say, and what does it not let us say?
It lets us say that there are well-documented cases in which a person, during a period of medically established unconsciousness, reported specific accurate information about events in his or her surroundings, and in which the standard alternative explanations—coincidence, cryptomnesia, confabulation, researcher bias—have been examined and found inadequate. It lets us say that these cases are not isolated curiosities. They occur with sufficient frequency, across enough investigators and enough patient populations, that the phenomenon is no longer a curiosity but a body of evidence. It lets us say that any account of consciousness must, sooner or later, address what is happening in these cases. It lets us say that the standard physicalist account, which holds that consciousness is identical to or wholly constituted by ongoing brain activity, has a hard time accommodating reports of accurate perception during periods of documented brain shutdown.30
It does not let us say that we have a complete causal story for what is happening in the strong cases. We have evidence that something is happening. We do not have a verified account of how the perception occurs, whether it is mediated by some unrecognized physical process, whether it is fully non-physical, what the relationship is between the experiencing subject and the body. The evidence supports certain conclusions and excludes others, but it does not generate, all by itself, a finished metaphysics of mind.
It does not let us say that any individual case is beyond critique. Even the strongest cases have weak points; even the cases that survive scrutiny survive imperfectly, with details that could in principle be reinterpreted. The empirical case is cumulative. It does not stand or fall on any single case. It stands on the pattern across many cases, scored consistently by criteria that were set out in advance and applied honestly.
It does not let us say that the NDE evidence settles the eschatological questions of conditional immortality, universal restoration, or eternal conscious torment.36 The evidence supports substance dualism and a conscious initial postmortem state. It does not, by itself, tell us what happens next, whether everyone who dies eventually finds Christ, whether the unrepentant are eventually destroyed, or what the final judgment looks like. Those are theological questions, settled (insofar as they can be settled this side of glory) by Scripture and reasoning over Scripture, with NDE evidence sitting alongside as supporting data on certain points but not as authoritative revelation. Chapter 28, on distressing NDEs, takes up the most contested theological question that arises from the NDE data.31
It does not let us treat each NDEr’s personal interpretation of his or her experience as authoritative. NDErs interpret their experiences through the cultural and religious frameworks they bring with them. A Christian NDEr is likely to identify the being of light as Christ; a non-religious NDEr is likely to identify it as “the universe” or “pure love”; a Hindu NDEr may identify it differently still. The veridical evidence speaks to the structure of the experience and to the existence of some who is encountered; it does not, by itself, resolve all the disputes about who that someone is. The chapter on Christian discernment of NDE content—Chapter 30—takes up the practical question of how a serious Christian reader should sift the content of NDE accounts.
What this chapter has tried to do is set the methodological framework that makes the rest of the empirical case possible. It has not made the case. The case lies in the chapters ahead: Chapter 10, on accurate distant observation; Chapter 11, on the cases of NDEs in those born blind; Chapter 12, on the cardiac-arrest cases with documented EEG flat readings; Chapter 13, on the “Peak in Darien” cases; and Chapter 14, on the full cumulative case. The methodology developed here is what the reader brings to those chapters. It is what lets the reader weigh each case carefully, ask the right questions, and avoid both credulous acceptance and dismissive rejection.
I have tried in this chapter to be honest in both directions. The veridical phenomenon is real and well-documented, and any serious account of consciousness has to address it. The veridical literature is also imperfect, often loosely documented, and prone to the same biases that affect any literature collected by sympathetic researchers. The strong cases survive both kinds of honesty. They are robust under careful examination. They earn the weight that the empirical argument places on them. The weak cases earn less weight, and a careful reader will know how to tell them apart.
If you are reading this chapter as someone who has had an NDE yourself, or who knows someone who did, and you find yourself worried that your story or your loved one’s story does not score high on the methodological scale, please hear me. The score is for evidential argument, not for spiritual significance. An experience that did not produce a verifiable detail observed by a hospital staff member can still have changed your life, drawn you to Christ, healed your fear of death, and been precisely what God intended for you. The methodological scale is for arguing with hostile philosophers of mind. Its silence about your particular experience does not mean your experience was unreal. It means only that your experience is one of the many that cannot be used to win the empirical argument—and the empirical argument is not, thank God, the foundation of your life with Christ.
I have spent a long chapter on definitions, and the reader who has stayed with me may be wondering whether all this care is necessary. The concern is fair. Definitions can become a place to hide—an opportunity to look rigorous without saying anything that costs anything. So let me say plainly what the careful definition costs and what it pays for.
What it costs is the cheap apologetic move. There is a kind of Christian writing about NDEs that takes any moving testimony, however thinly documented, as proof that the soul survives death and that heaven is real. I understand the impulse. The testimonies are moving. They confirm the hope. The problem is that the cheap move does not work outside the room of those already inclined to believe it. It does not persuade the philosopher of mind who already thinks NDEs are dying-brain artifacts. It does not engage the Christian physicalist who needs to be shown what the evidence actually requires. It does not honor the dying patient who deserves to know what we can and cannot say with empirical confidence. The careful definition gives up the right to use weak cases as evidence, and that is a real loss for anyone who wants every story to do every kind of work.
What it pays for is the stronger argument. When I tell the story of Pam Reynolds in Chapter 12, or Vicki Umipeg and the cases of those born blind in Chapter 11, or the “Peak in Darien” cases in Chapter 13, the methodology of this chapter is what lets the reader say yes—this case is doing real work. The case is documented. The corroboration is independent. The medical context is verified. The alternatives have been examined. The reader is not being asked to take the case on the strength of the storyteller’s sincerity; the reader is being shown the case in a way that lets it stand on its own. That is the difference between a chapter the dying can rest on and a chapter that only confirms what the believer already wants to believe. The first kind is what this book is trying to be. The second kind would not, in the end, comfort anyone who needed real comfort.
For pastors, the careful definition matters because pastors are asked, sooner or later, by parishioners who have lost loved ones or who are themselves dying, what we know about what happens at death. The pastor who has a careful framework can answer honestly: Here is what the strong evidence shows; here is what it does not show; here is where Scripture says what evidence cannot say. The pastor without a framework will either overclaim or underclaim, and both are pastoral failures. The dying do not need cheap reassurance. They need honest hope.
For the dying themselves, the careful definition matters because the dying are not stupid. A person facing his own death has, in my experience, an unusually sharp ear for what someone is and is not actually saying. The dying patient who is told a smooth story about all the wonderful NDE testimonies will often see through it, will often suspect that he is being protected from harder questions, and will often want the harder questions taken seriously. The dying are owed that honesty. The methodology of this chapter is part of how to deliver it.
For the bereaved, the careful definition matters because the bereaved have to live for years with what they were told at the deathbed. A bereaved mother who was told, in the days after her child’s death, that the NDE evidence proves her child is in heaven and at peace, and who then later learns that the proof was overstated, has been twice wounded—once by the loss and once by the false comfort. Better to give the careful comfort, the one that says the strong evidence is consistent with what Scripture has always taught: that your child is held by Christ, that his body lies in the dust but his life is held in love, that what you most fear—extinction, abandonment—is not in fact what is happening. That careful comfort, anchored in evidence that earns its weight, will hold over years. The cheap comfort will crack.
And for the curious skeptic, the careful definition matters because skeptics are not generally moved by enthusiasm or testimony. They are moved—or at least slowed—by methodology that they cannot fault. The skeptic who comes to Chapter 10 with the legitimate suspicion that NDE proponents are too credulous, and finds in this chapter a methodology stricter than her suspicion required, has been given a way to take the cases seriously that does not ask her to abandon her critical instincts. The strict methodology is, in the end, an act of respect for her reasoning. It says, I am not asking you to believe this on my say-so. I am asking you to follow the criteria and look at the cases.
The chapter ahead—Chapter 9, on the core NDE pattern—turns from the evidential criterion to the phenomenological one. It asks not what makes a case veridical but what is the structure of the experience that NDErs across cultures and decades report. The two chapters together set up the case files that begin with Chapter 10. The framework is in place. We are now ready to see what it can do.
↑ 1. The Pam Reynolds case is described in detail in Michael B. Sabom, Light and Death: One Doctor’s Fascinating Account of Near-Death Experiences (Grand Rapids, MI: Zondervan, 1998), 37–52. The clinical details of hypothermic cardiac arrest as a surgical procedure are well-established in the neurosurgical literature; see Robert F. Spetzler et al., “Aneurysms of the Basilar Artery Treated with Circulatory Arrest, Hypothermia, and Barbiturate Cerebral Protection,” Journal of Neurosurgery 68, no. 6 (1988): 868–79. The case is treated in full in Chapter 12 of this book.
↑ 2. Sabom, Light and Death, 41–47. Sabom obtained access to the operative record and interviewed members of the surgical team. The details of the Midas Rex bone saw and the conversation about the femoral artery were among those independently confirmed.
↑ 3. The contemporary technical use of “veridical” in NDE research is given its most careful treatment in Janice Miner Holden, “Veridical Perception in Near-Death Experiences,” in The Handbook of Near-Death Experiences: Thirty Years of Investigation, ed. Janice Miner Holden, Bruce Greyson, and Debbie James (Santa Barbara, CA: Praeger, 2009), 185–211, esp. 185–90. Holden’s definition, which has become standard, restricts “veridical” to NDE reports of apparent perception that are subsequently confirmed against external evidence to have been accurate.
↑ 4. The methodological discussion that follows draws on the dissertation: Matthew Friend, “Near-Death Experiences as Evidence for Substance Dualism within the Conditional Immortality Debate” (Th.D. dissertation, Trinity College of the Bible and Trinity Theological Seminary, 2025), Chapter 3 (“Methodology”) and Appendix A (“Three-Dimensional Scoring System”).
↑ 5. The verifiability criterion is treated extensively in Holden, “Veridical Perception,” 187–88, and in Pim van Lommel, Consciousness Beyond Life: The Science of the Near-Death Experience (New York: HarperOne, 2010), 19–25, where van Lommel walks through the difference between subjective NDE content and externally checkable claims.
↑ 6. The medical baseline for cerebral inactivity during cardiac arrest is well-established; see Michael J. Aminoff et al., “Electrocerebral Accompaniments of Syncope Associated with Malignant Ventricular Arrhythmias,” Annals of Internal Medicine 108, no. 6 (1988): 791–96. The implications for NDE research are developed in detail in Chapter 5 of this book.
↑ 7. The reconstructive nature of memory is well-established in the cognitive psychology literature; for an accessible overview see Daniel L. Schacter, The Seven Sins of Memory: How the Mind Forgets and Remembers (Boston: Houghton Mifflin, 2001). Its specific implications for NDE research are discussed in Bruce Greyson, After: A Doctor Explores What Near-Death Experiences Reveal about Life and Beyond (New York: St. Martin’s Essentials, 2021), 151–65.
↑ 8. Holden, “Veridical Perception,” 185–211. Holden reviewed 107 published cases of out-of-body perception during NDEs and found that 92 percent contained at least some accurate veridical detail. Her analysis of the methodological strength of the cases varied substantially with the quality of corroboration available, which is the point in the text.
↑ 9. The classic enumeration of alternative explanations comes from the skeptical literature; see, e.g., Susan Blackmore, Dying to Live: Near-Death Experiences (Buffalo, NY: Prometheus, 1993), and Keith Augustine, “Hallucinatory Near-Death Experiences,” in The Myth of an Afterlife: The Case Against Life After Death, ed. Michael Martin and Keith Augustine (Lanham, MD: Rowman & Littlefield, 2015), 565–94. Each of these objections is engaged at length later in the chapter.
↑ 10. Friend, dissertation, Appendix A. The full operational definitions, with worked examples at each score level, appear there and in Appendix A of this book.
↑ 11. The reasoning here parallels the “medical context” criterion developed in the prospective NDE studies; see Pim van Lommel et al., “Near-Death Experience in Survivors of Cardiac Arrest: A Prospective Study in the Netherlands,” The Lancet 358, no. 9298 (2001): 2039–45, and Sam Parnia et al., “AWARE—AWAreness during REsuscitation—A Prospective Study,” Resuscitation 85, no. 12 (2014): 1799–1805.
↑ 12. The principle that low-prior-probability accurate detail carries more evidential weight than high-prior-probability accurate detail is a standard application of Bayesian reasoning to testimony; see Richard Swinburne, The Existence of God, 2nd ed. (Oxford: Oxford University Press, 2004), 110–34, on the structure of inductive evidence.
↑ 13. The corroboration criterion as applied to NDE research is developed in Penny Sartori, The Near-Death Experiences of Hospitalized Intensive Care Patients: A Five-Year Clinical Study (Lewiston, NY: Edwin Mellen, 2008), 215–42; and in Holden, “Veridical Perception,” 195–205, where Holden distinguishes “independently corroborated,” “corroborated but with caveats,” and “uncorroborated” classes.
↑ 14. Friend, dissertation, Chapter 4 (“Data Analysis”), reports the full distribution of scores across the 5,278 cases. The Strong/Exceptional threshold count is 1,618 (30.7 percent). The threshold definitions and full distribution are reproduced in Appendix A of this book.
↑ 15. Bruce Greyson, “The Near-Death Experience Scale: Construction, Reliability, and Validity,” Journal of Nervous and Mental Disease 171, no. 6 (1983): 369–75. The scale has been validated and used in dozens of subsequent studies; for an overview, see Greyson, After, 30–38.
↑ 16. Kenneth Ring, Life at Death: A Scientific Investigation of the Near-Death Experience (New York: Coward, McCann, & Geoghegan, 1980), 32–39, presents the Weighted Core Experience Index. Like the Greyson Scale, it measures phenomenological richness rather than veridical strength.
↑ 17. The medical-record correlation step is described in detail in Sabom, Recollections of Death: A Medical Investigation (New York: Harper & Row, 1982), 23–54, where Sabom describes how he gathered records and matched them to patient reports as part of his original prospective study.
↑ 18. The pre-disclosure interview standard is articulated in the AWARE study protocol; see Parnia et al., “AWARE,” 1801–2. AWARE went further than most studies in introducing concealed visual targets in resuscitation areas, so that any patient reporting accurate observation of those targets could not have learned about them through any normal route.
↑ 19. The documentation standards described here parallel those in the qualitative research methodology literature; see John W. Creswell and Cheryl N. Poth, Qualitative Inquiry and Research Design, 4th ed. (Thousand Oaks, CA: SAGE, 2018), Chapter 10. The application to NDE case investigation is straightforward; what differs is only the specific kind of phenomenon being investigated.
↑ 20. The phenomenological case is developed especially in Allan Kellehear, Experiences Near Death: Beyond Medicine and Religion (New York: Oxford University Press, 1996), and in Greyson, After, 39–82, where Greyson summarizes the cross-cultural and cross-temporal consistency findings. The full treatment in this book begins in Chapter 9.
↑ 21. The transformational evidence is summarized in Kenneth Ring, Heading Toward Omega: In Search of the Meaning of the Near-Death Experience (New York: William Morrow, 1984), and updated in Pim van Lommel, Consciousness Beyond Life, 50–68. NDErs as a group show distinct durable changes; Ring’s work documents these in detail.
↑ 22. The coincidence objection is most fully developed in Augustine, “Hallucinatory Near-Death Experiences,” 581–84, who argues that NDE reports should be evaluated against the base rate of accurate-by-chance reports in the general medical population. The counter-argument—that the joint probability of multiple specific accurate details is vanishingly low—is developed in van Lommel, Consciousness Beyond Life, 19–25, and in Holden, “Veridical Perception,” 195–201.
↑ 23. The cumulative-probability argument is given quantitative form in J. Steve Miller, Near-Death Experiences as Evidence for the Existence of God and Heaven (Acworth, GA: Wisdom Creek Press, 2012), 47–72, who walks through the math of low-prior-probability accurate detail across multiple cases.
↑ 24. Cryptomnesia as a memory phenomenon is described in Daniel L. Schacter, Searching for Memory: The Brain, the Mind, and the Past (New York: Basic Books, 1996), 105–9. Its application to NDE skepticism is found in Blackmore, Dying to Live, 113–28.
↑ 25. The clinical literature on confabulation is large; for an overview see William Hirstein, Brain Fiction: Self-Deception and the Riddle of Confabulation (Cambridge, MA: MIT Press, 2005). The application to NDE skepticism appears in Christopher C. French, “Near-Death Experiences in Cardiac Arrest Survivors,” Progress in Brain Research 150 (2005): 351–67.
↑ 26. The use of pre-disclosure interviewing as a confabulation control is articulated in Sartori, Near-Death Experiences of Hospitalized Intensive Care Patients, 220–28, and in the AWARE protocol; see Parnia et al., “AWARE,” 1801–2.
↑ 27. The researcher-bias objection is most vigorously pressed in Keith Augustine, “Does Paranormal Perception Occur in Near-Death Experiences?” Journal of Near-Death Studies 25, no. 4 (2007): 203–36, and in the responses and counter-responses in subsequent issues of the same journal. The exchange is worth reading at length for any reader who wants to see the strongest skeptical version of the argument.
↑ 28. Friend, dissertation, Chapter 3 (“Methodology”) and Chapter 4 (“Data Analysis”), describes the two source classes and the rationale for combining them. The NDERF database is hosted at nderf.org and is described in detail in Jeffrey Long, Evidence of the Afterlife: The Science of Near-Death Experiences (New York: HarperOne, 2010), 7–15.
↑ 29. The structural similarity of patterns across the two source classes is one of the dissertation’s key methodological findings; see Friend, dissertation, Chapter 4. It functions as an internal robustness check on the curation objection.
↑ 30. The standard physicalist account—identity theory, functionalism, or supervenience-based variants—requires ongoing brain activity to support conscious experience. For the standard statement, see Paul M. Churchland, Matter and Consciousness, rev. ed. (Cambridge, MA: MIT Press, 1988), and the responses in J. P. Moreland, The Soul: How We Know It’s Real and Why It Matters (Chicago: Moody, 2014), 51–90. The empirical implications for NDE research are precisely what is at stake in this book.
↑ 31. The author’s position on these eschatological questions is conditional immortality with postmortem opportunity, with genuine openness to conservative biblical universalism. The questions are addressed at length in companion volumes; the present book is methodologically careful not to adjudicate them on NDE grounds alone. See Stephen Jonathan, Grace Beyond the Grave: Is Salvation Possible in the Afterlife? (Eugene, OR: Wipf and Stock, 2014); James Beilby, Postmortem Opportunity: A Biblical and Theological Assessment of Salvation After Death (Downers Grove, IL: IVP Academic, 2021); and the discussion in Chapter 28 of this book.
↑ 32. The progressive refinement of NDE interview methodology is described in Bruce Greyson, “Getting Comfortable with Near Death Experiences. An Overview of Near-Death Experiences,” Missouri Medicine 110, no. 6 (2013): 475–81; and discussed historically in Holden, Greyson, and James, Handbook of Near-Death Experiences, 1–16. Early Moody-era interviews focused on phenomenological richness; later prospective protocols, especially those developed by Sabom, van Lommel, Sartori, and Parnia, were redesigned around the kinds of questions that surface verifiable detail.
↑ 33. The compounding of inaccessibility forms is what makes the strongest cases evidentially decisive; see the analyses in Holden, “Veridical Perception,” 195–205, and Sabom, Light and Death, 184–91. Where physical, sensory, and temporal-medical inaccessibility all hold simultaneously, the explanatory burden on any normal-perception account becomes very heavy indeed.
↑ 34. The threshold-setting and the percentage findings are not the central evidential argument by themselves; they function as a summary description of where the dataset ends up. The substantive evidential work occurs at the case level, where individual strong cases survive close scrutiny. The thresholds matter because they let the reader see how many cases reach a level of documentation strong enough to make case-level scrutiny worthwhile. See Friend, dissertation, Chapter 4, for full distributional data.
↑ 35. The temporal-impossibility argument against cryptomnesia is articulated in Edward F. Kelly et al., Irreducible Mind: Toward a Psychology for the 21st Century (Lanham, MD: Rowman & Littlefield, 2007), 367–421, esp. 380–88. The Kelly volume gives the most extensive philosophical treatment of the methodological objections to NDE evidence and the structural answers to them.
↑ 36. On the methodological boundary between empirical evidence (which can support certain conclusions about consciousness and the soul) and theological doctrine (which is settled in this book’s view by Scripture and the historic creeds, with empirical evidence sitting alongside as supporting data on certain points), see the discussion in § 2.7 and § 2.10 of the master prompt for this project; the position is also articulated in Beilby, Postmortem Opportunity, 47–67.