Chapter 4

The Strongest Cases: Veridical Perception During NDEs

A man arrived at a Dutch hospital on a cold night in 1979, comatose and turning blue. Paramedics had pulled him from a meadow. In the emergency room, a nurse took the upper dentures from his mouth and dropped them onto the rolling crash cart so the doctors could intubate him. The patient’s heart was not pumping. His brain was without oxygen. His pupils were fixed. By every measure available at the bedside, he was as close to death as a person can come and still be brought back. CPR went on for about ninety minutes.¹

A week later — back on the cardiac ward, awake, and stable — the same man saw the same nurse and said, “Oh, that nurse knows where my dentures are.” Then he described, accurately, the exact sliding drawer of the exact crash cart where his teeth had been placed during the resuscitation. He described the small room. He described who was there. He had watched it all from above, he said, and he had been afraid the team would stop CPR and let him die.²

The nurse was stunned. So am I.

This chapter is about cases like that one. They are the heart of the case for taking near-death experiences seriously — not as comforting stories, but as evidence. A veridical NDE is one in which the experiencer accurately reports something specific they should not have been able to know by ordinary means — the layout of a room, the placement of an object, the actions of staff. (“Veridical” just means “truth-telling”: the perception matches reality and can be checked.) These are the cases the critics must explain. The question of this chapter is whether they can.

A. The Critic’s Argument

Michael Marsh has read these cases. He is not impressed. He devotes a careful section of his book to specific case studies, and his verdict is steady: corroboration is missing, hearsay piles on hearsay, and the alleged perceptions can almost always be traced to ordinary sources. He surveys roughly eight popular NDE authors — Moody, Sabom, Ring, Grey, the Fenwicks, Atwater, and a few more — and concludes that the field’s “exemplary” cases simply do not pass the kind of scrutiny he would expect of a clinical trial.³ The reports come months or years after the event. Witnesses, when they can be tracked down, often cannot remember what the experiencer claims to have seen. Descriptions are loose enough to fit many actual states of affairs. By the time Marsh reaches his summary on page 27 of his book, his judgment is firm: the case for true paranormal perception is weak, the documentation thin, and the published cases inadequate as evidence for a soul that leaves the body.⁴

John Martin Fischer and Benjamin Mitchell-Yellin make the philosopher’s version of the same move. They are open about their strategy. They take two famous cases — the man with the dentures and Pam Reynolds — and argue, point by point, that each apparent veridical perception can be explained by some ordinary process: subliminal hearing, lucky guessing, memory reconstruction after the fact, or unconscious learning during the days the patient spent on the ward.⁵ They are not trying to prove that those alternative scenarios actually happened. Their burden, they say, is much lighter. They only need to show that the alternatives are possible. If a story can be explained without invoking a soul that floats free of the body, the more ordinary explanation should win on grounds of simplicity. They call this their “Single Explanation” strategy, and it sets the bar exactly where they want it set.⁶

Both critics agree on the bottom line. Experiencers are sincere; something striking happened. But none of these cases, taken one at a time, forces us to conclude that consciousness left the body. The brain is clever. Memory is reconstructive. Hospital staff talk loudly when they think the patient cannot hear. Coincidence, confabulation, and unconscious cueing can do a great deal of work between them. Marsh will grant that the experiences are real and meaningful to the people who have them. He will not grant that they are evidence for an immaterial mind.

Take that argument seriously. It is well within the bounds of mainstream science, and if the dentures man were the only case on offer, I would not dismiss the skeptical reading lightly. The question is whether the skeptical reading survives the rest of the evidence — and whether the procedure used to generate it is itself sound.

B. Where the Argument Breaks Down

Two weaknesses stand out. They both have to do with how the critics decide what to engage.

The first is the narrowness of the corpus. Marsh writes a careful book, but he engages roughly eight popular NDE authors, with most of his attention on Moody, Ring, Grey, Sabom, and the Fenwicks.⁷ That is a sensible place to start a survey. It is not the same as engaging the evidence base. The authors he reviews wrote in the 1970s, ’80s, and ’90s. Since then, the field has produced prospective hospital studies, blind-NDE research, peer-reviewed analyses, and large case databases. Marsh’s book appeared in 2010. By that point van Lommel’s landmark Lancet paper was nine years in print; Janice Holden’s comprehensive review chapter in the Handbook of Near-Death Experiences had just been published in 2009; and the team behind The Self Does Not Die had begun gathering more than a hundred verified veridical cases. To read Marsh, you would not know how much had changed. He spends his energy on the popular books while passing lightly over the hospital studies and the case archives that ought to have been the heart of the conversation.⁸

Fischer and Mitchell-Yellin show a related pattern in tighter form. They are philosophers, not clinicians. Their book is about how to argue, and so they pick two famous cases and analyze them in great detail.⁹ That is a fine philosopher’s move. It is also a way to keep the conversation small. The hundreds of other documented cases are not addressed. They are gestured at and dismissed as “more of the same.” But “more of the same” is exactly the issue. A single case can be explained by an unlikely coincidence. A hundred independent cases, each explained by its own unlikely coincidence, start asking the skeptic to believe in a great many unlikely coincidences.

The second weakness follows from the first. Both critics treat each veridical case as a stand-alone puzzle, asking, “Could this one perception have been picked up by ordinary means?” — and almost always answering yes, somewhere on the edge of plausibility. But evidence is not a series of stand-alone puzzles. It is a collection. The right question is whether the same kind of ordinary explanation accounts for the cases together. If hypoxia explains case one, post-hospital confabulation explains case two, lucky guessing explains case three, and overheard staff talk explains case four, the skeptic is no longer offering an explanation. He is offering a different excuse for each piece of evidence. That is not parsimony. That is special pleading.¹⁰

Marsh himself seems to sense the problem. More than once, he concedes the field needs “a far larger corpus of credible events studied prospectively” — and then, having said so, declines to engage the larger corpus that already exists.¹¹ Fischer and Mitchell-Yellin are more candid: they admit they only need to show that a non-supernatural explanation is possible. But “possible” is a very low bar. With enough creativity, almost any story can be made to fit a physicalist mold.¹² The question is whether such fits remain plausible, taken together, and whether they predict the data better than the alternative.

C. What the Evidence Actually Shows

I want to do three things in this section. First, I want to walk through the two cases the critics handle worst — the dentures man and Maria’s tennis shoe — in enough detail that you can judge for yourself. Second, I want to add a third case from Sam Parnia’s AWARE Study, because Marsh wrote his book before AWARE’s results were public, and they matter. Third, I want to step back and look at the body of evidence as a whole, through a quiet but devastating analysis by Janice Holden.

The Dentures Man

I told the opening of this story already. Here is the rest. The night the patient came in, he was deeply comatose; his pupils did not respond to light; he had no measurable cardiac output; CPR continued for roughly ninety minutes. The nurse, identified in the literature as “TG,” removed the upper dentures and placed them on the crash cart so an airway could be inserted.¹³ TG worked his shift and went home. A full week passed before he saw the patient again on the cardiac ward. When he walked in to deliver medication, the patient looked up and said, “Oh, that nurse knows where my dentures are” — and went on to describe, in detail, the small resuscitation room, the people present, what they wore, the appearance of the nurse himself, and the fact that his teeth had been placed in a sliding drawer of the cart with bottles on top. He told TG that during the resuscitation he had watched from above. He had been afraid the team would give up.¹⁴

Notice what an alternative story has to do here. It must explain how a patient — comatose, with fixed pupils, pulseless, undergoing CPR — picked up a precise visual layout of a small room and the position of his own dentures in a drawer he could not have seen, and then preserved that information through ninety minutes of resuscitation, several days of intensive care, and a week on the ward, before producing it spontaneously when the right nurse walked through the door. Fischer and Mitchell-Yellin offer the suggestion that he might have “subconsciously pieced together” the sound of a drawer opening and the feeling of dentures being removed.¹⁵ Set aside that this would require an unconscious, pulseless man to be hearing and feeling. Even granting that, the story does not explain his accurate description of the room and the people. None of that can be reconstructed from a single drawer-sound and a single sensation. A person well enough to register and integrate that much sensory data is not a person undergoing CPR for a heart that has stopped pumping.

Maria’s Tennis Shoe

The case is from 1977. A migrant worker named Maria, a Spanish speaker with limited English, suffered a heart attack and was admitted to Harborview Medical Center in Seattle. Three days later, she had a second arrest. During that arrest she experienced an out-of-body episode. After resuscitation, she described to a hospital social worker named Kimberly Clark Sharp the printouts strewn across the floor of her room and one particular detail: a man’s left tennis shoe sitting on a third-floor window ledge on the outside of the hospital. The shoe, she said, was dark blue. The fabric was worn over the little toe. One lace was tucked under the heel.¹⁶

Sharp was skeptical. So she walked the third floor. Window by window, she pressed her face against the glass and looked down at the ledges. Eventually she found it. A blue tennis shoe — worn over the little toe, lace tucked under the heel — sat on the ledge outside a north-end window. From inside, even with her face on the glass, Sharp could not see the worn toe or the tucked lace. Maria had described details visible only from a position outside the building.¹⁷

The skeptical literature has worked this case over hard. Hayden Ebbern, Sean Mulligan, and Barry Beyerstein — later seconded by Keith Augustine — argued that the shoe could have been seen from the ground or from inside the room, and that Maria might have overheard staff talking about it. Sharp answered point by point, and her answers held: the defining details (worn toe, tucked lace) were not visible from the ground, were not visible from inside without pressing one’s face to the glass, and could hardly have been picked up by a Spanish-speaking patient with little English overhearing English-speaking staff in the brief window before her arrest.¹⁸ Augustine’s “she probably overheard it” hypothesis is exactly the kind of stand-alone excuse I warned about in section B. Maybe — just possibly — for this case. But run the strategy across a hundred cases, and the cumulative cost becomes painful.

The AWARE Study and the Talking Defibrillator

Marsh wrote his book before Sam Parnia’s AWARE Study results were published. In 2014, Parnia and his team reported on the first phase of a multi-hospital study designed precisely to test claims of conscious perception during cardiac arrest. The most striking case involved a 57-year-old social worker at Southampton General Hospital in England who went into cardiac arrest while being prepped for a catheter procedure.¹⁹ An automated external defibrillator — an AED, the same kind of machine you may have seen mounted in airports — was used twice. After resuscitation, the patient described looking down at his own body, watching the team work, and hearing a voice — coming from the AED — saying “Shock the patient, shock the patient.”

Parnia’s investigators did something most NDE researchers cannot. They pulled the medical record. The AED used in the case did, in fact, issue voiced commands during the resuscitation. According to Professor Enrico Facco of the University of Padua, who reviewed the timing of the AED’s automated prompts, the patient’s reported awareness implied conscious perception persisting for at least three minutes during cardiac arrest — well past the 10-to-20-second window in which a few residual neurons can still misfire.²⁰ No one needed to take the case on faith. The hardware itself produced the corroborating data.

Holden’s Ninety-Two Percent

In 2009, Janice Miner Holden — a longtime NDE researcher and past president of the International Association for Near-Death Studies — did the book-length version of what every skeptical critic should have done. She gathered every published case of “apparently nonphysical veridical perception” reported between 1975 and 2007. Every case where an experiencer reported perceiving something specific that, by ordinary means, they should not have been able to perceive. She found 107 such cases, drawn from 39 different publications by 37 different authors or author teams.²¹ Then she scored them.

Using her most stringent criterion — counting any case as inaccurate if even one detail did not check out — Holden found that only about 8 percent of the cases contained any inaccuracy at all. That leaves roughly 92 percent of cases in which every reported detail matched reality. Of the full set, a remarkable 37 percent had been verified, in the original published reports, by independent objective sources — researchers who tracked down the witnesses, the medical records, and the staff.²² The authors of The Self Does Not Die, applying their own rigorous criteria, confirmed Holden’s general result: discrepancies of any kind appear in about 10 percent of cases, and major discrepancies in only about 2 percent.²³

Think about what those numbers mean. If NDE perceptions are hallucinations — manufactured by a confused, oxygen-starved brain — the accuracy rate should be near zero. Hallucinations would not predict, with 92 percent accuracy across a hundred unrelated cases, the actual contents of a hospital room. Random guesses about lost dentures, hidden shoes, AED voices, and surgical instruments would not produce a 37 percent independently corroborated rate. The numbers are exactly the wrong shape for a dying-brain hypothesis. They are the right shape for the hypothesis that the experiencer was, in some way we still do not fully understand, actually perceiving.²⁴

Pam Reynolds is the most famous veridical case in the literature, and her story deserves the chapter it gets. We treat her case fully in Chapter 5; here, I only note that her story is one more entry in the same growing column.

D. Counter-Objections

Two objections deserve a direct answer here, even though both are owned more fully later in the book.

The first is the slogan: “Anecdotes are not evidence.” It is a bumper sticker, and like most bumper stickers, it is half-true. Anecdotes — uncollected, unchecked, and unverified — are weak evidence. But the cases I have just described are not bare anecdotes. The dentures man’s report was made spontaneously to a working nurse who had been on the resuscitation; the case was investigated, the witnesses interviewed, and the medical record consulted by van Lommel’s team for Consciousness Beyond Life and by Smit for the Journal of Near-Death Studies.²⁵ Maria’s tennis shoe was investigated almost immediately by Kimberly Clark Sharp, who walked the floor, located the shoe, and recovered it. The AWARE case was logged in a peer-reviewed journal, with the medical record cross-checked. When skeptics complain about “anecdotes,” they usually mean they would prefer laboratory conditions. So would I. So would NDE researchers — that is precisely why van Lommel ran his prospective Dutch study and Parnia ran AWARE. Prospective studies show that bedside cases are not random folktales; they are the same cases, gathered with witnesses present.

The second objection is memory. Could the patients have reconstructed their reports after the fact, building a “memory” from staff comments, hospital sounds, and the natural human pull toward a coherent story? It is a serious objection, and Chapter 18 develops it in full. Two short answers will do here. First, many veridical reports are made within minutes or hours of the event. The dentures man spoke to TG immediately upon recognizing him. Maria spoke to Sharp the same day. There is little time for confabulation to do its work. Second, the content of the reports is testable. If the memory is reconstructed, the details should be wrong. They are not. At the rates Holden has documented, they are right roughly 92 percent of the time. A confabulating brain does not score 92 percent on a test of independent fact.²⁶

What I have tried to show is that the strongest veridical NDE cases are not just colorful stories. They are data — collected at the bedside, checked against the medical record, gathered into a 107-case archive that the leading skeptical critics have never engaged on its own terms. Marsh, Fischer, and Mitchell-Yellin have tested the door at certain hinges and found it loose. They have not tried the lock. The next chapter takes the case the critics most want to dismantle — Pam Reynolds — and lets her story answer for itself.