Chapter 10

Cases of Accurate Distant Observation

When the dying see what their bodies cannot.

It is August 1991. Pam Reynolds, a thirty-five-year-old singer-songwriter from Atlanta, lies on an operating table in Phoenix, Arizona. She has a giant aneurysm — a swelling in a blood vessel — at the base of her brain. If it bursts, she dies. To reach it safely, her surgeon has chosen one of the most extreme procedures in modern medicine.

His name is Dr. Robert Spetzler. He runs the Barrow Neurological Institute. The procedure he is about to perform is called hypothermic cardiac arrest — sometimes nicknamed "standstill."¹ The team will cool Pam's body to roughly sixty degrees Fahrenheit. They will stop her heart. They will drain the blood from her head. For about an hour, by every measurable standard medicine has, Pam will be dead.

To make sure her brain itself stays offline, the team fits molded clickers — small earphones — into both of Pam's ears. The clickers fire pulses of sound at one hundred decibels, about as loud as a chainsaw, every second. A monitor watches her brainstem for any reaction. When that monitor goes flat, the team knows the deepest part of her brain has shut down. Tape covers her eyes. Her body cools. Her heart stops. The blood drains.

And then something happens that, on the standard physicalist account of consciousness, should be impossible. Pam will later say she "popped out the top of her head." She will describe the bone saw the surgeon used — a tool she had never seen, called a Midas Rex² — with surprising accuracy. She will hear what a female specialist said about the size of her femoral artery. She will note where, in the room, that conversation happened. When she wakes up, she remembers all of it. And the medical record agrees.

What "Distant Observation" Means

Pam Reynolds's case is famous for a reason. She did not just report a vague feeling of peace. She did not just say she saw a light. She reported specific, checkable facts about a room and a procedure she could not see, hear, smell, or touch — facts that turned out to be correct.

NDE researchers call this kind of perception veridical. The word comes from the Latin veridicus, which means "truth-telling." Veridical means checkable against the world. Not "I felt loved." Not "I saw the colors." But: I saw what was on the third-floor window ledge. I saw what the surgeon did with his hands. I heard what was said when my heart was stopped. Chapter 8 walks through the precise definitions and the three-part scoring system I used in the dissertation. Here we apply that framework to the strongest single body of evidence in the entire NDE literature.

This chapter is about one category: cases where a dying person reports accurate observations of locations, events, conversations, and people at a distance from their physical body. They report seeing the operating room from the ceiling. They report seeing what the family is doing two floors below. They report seeing a tennis shoe on a window ledge that no nurse has noticed.

I want to walk you through the strongest cases. Then I want to lay the larger pattern alongside them — the 1,114 distant-observation cases I cataloged in my own dissertation research, drawn from a database of 5,278 accounts.³ Then I want to be fair to the physicalist response, and explain why, in the end, that response does not rescue the strict view that consciousness is what the brain does.

The Cases

Pam Reynolds

Michael Sabom, a cardiologist who spent his career studying NDEs, devoted a long chapter of his book Light and Death to the Reynolds case.⁴ Sabom interviewed Pam. He interviewed members of the surgical team. He cross-checked her account against the medical record. The result is one of the most carefully documented veridical NDEs ever published.

Begin with what Pam should not have been able to describe. The Midas Rex bone saw is not a kitchen-tool kind of saw. It is a small, pneumatic, high-speed device that runs on compressed gas. Pam, in interviews months later, said it looked like an "electric toothbrush" with a case lying nearby holding interchangeable blades.⁵ She had never seen such a device. She had no medical background. Her description matched the actual instrument.

She also reported overhearing a conversation. A female cardiac specialist, working on her femoral artery to set up the bypass machine, remarked that Pam's vessels were too small. The team had to switch sides. Pam's account of who said what, and where in the room they were standing, lined up with the medical record.⁶

None of this should be possible if consciousness is identical to brain activity. Pam's brainstem responses had gone flat under the click stimulus. Her cortex had no measurable electrical signal. Her eyes were taped shut. Loud clicks were firing into her ears. The room she "saw" was the room she could not, by any normal route, perceive.

Maria's Tennis Shoe

In 1977, a migrant worker named Maria suffered a cardiac arrest at Harborview Medical Center in Seattle. She survived. After her resuscitation she told a hospital social worker, Kimberly Clark Sharp, that she had floated outside her body, drifted out a window, and seen a tennis shoe on a ledge of the third floor on the north side of the building.⁸

Sharp was skeptical. To prove the story wrong, she went to the third floor and began checking the windows. On one of the windows, on a narrow ledge that could not be seen from inside the building, she found a man's tennis shoe — exactly where Maria had said it would be. Maria had described details that could only be seen from outside, hovering at window-level: a worn place on the little toe, the laces tucked under the heel.⁹

Critics have rightly pressed on this case. Sharp wrote it up years after it happened. Maria herself was never located again for follow-up interviews. The shoe was not preserved. These are honest weaknesses, and I will not pretend otherwise. But the basic outline is clear, and the social worker's reputation is not in dispute. Maria's case is not the strongest single case in the literature — Pam Reynolds is — but it is the most widely retold for a reason. The shape of the claim is sharp: a specific physical object, in a specific place, that could not be seen by any normal route.

Al Sullivan and the Flapping Elbows

In 1988, a fifty-five-year-old Connecticut truck driver named Al Sullivan went in for a triple-bypass at Hartford Hospital. During the surgery he had an out-of-body experience. He saw himself on the table. He saw his surgeon, Dr. Hiroyoshi Takata, doing something he could not understand: pressing his gloved hands flat against his chest, then pointing at instruments by flapping his elbows like a chicken.¹⁰

When Sullivan told his story afterward, the surgical team confirmed it. Dr. Takata had a peculiar habit. To keep his sterile hands above the operative field while he gave instructions to the staff, he would press his hands flat to his chest and gesture using his elbows. It was an unusual personal mannerism. Sullivan, who had been under anesthesia and draped under sterile cloth before the surgery began, had no normal way of knowing it.

This case is reported in the peer-reviewed Journal of Near-Death Studies by Cook, Greyson, and Stevenson, three respected investigators at the University of Virginia.¹¹ The detail is the kind that resists confabulation. Sullivan did not describe what people imagine surgeries to look like. He described something so odd he asked his doctor to explain it.

Sabom's Atlanta Studies

Long before Pam Reynolds, Michael Sabom had run a quieter but more important study. In the late 1970s, working as a young cardiologist in Atlanta, he interviewed cardiac-arrest survivors who reported NDEs. Then he interviewed a control group: cardiac patients with similar medical histories who had not had NDEs. He asked the controls to describe, as best they could, what a resuscitation looks like.¹²

The results were striking. About eighty percent of the controls made characteristic medical errors — describing the wrong device, the wrong order of steps, the wrong placement of equipment. Their descriptions were the kind a person picks up from medical television. The NDErs, by contrast, were essentially error-free. They described what had actually happened in their resuscitations, including small details like which staff member had stood where and which procedures were tried first.

This is the kind of evidence that does not depend on a single dramatic case. It is a comparison study, with a control group, with a measurable difference in accuracy. And the more accurate group is the one whose hearts had stopped.

The AWARE Study

In 2008, the cardiologist Sam Parnia launched a multi-hospital prospective study called AWARE — short for "AWAreness during REsuscitation." Thirty-three hospitals across three countries. Four years of data collection. Over two thousand cardiac-arrest patients enrolled. Of the survivors well enough to be interviewed, nine reported full NDEs. Two reported veridical out-of-body content.¹³

One of those two is the case most often cited from AWARE I. A fifty-seven-year-old man at Southampton General Hospital in England suffered a cardiac arrest. During the resuscitation, he reported, he was watching from above. He described a senior nurse, a bald man in a blue cap, the placement of the defibrillator pads, and a sequence of events that lasted about three minutes — events that occurred during the time his heart was documented to be stopped.¹⁴ The hospital records matched.

AWARE is sometimes described, by skeptics, as a "failed" study. The reason: Parnia had placed images on high shelves in resuscitation rooms — images visible only to someone hovering at ceiling height. None of the AWARE patients reported seeing the targets. That is true. But it is also a narrow result. Of the patients with veridical content, none of their resuscitations occurred in rooms with the targets. The shelf experiment is one specific test. Its inconclusive result does not erase the veridical reports outside it.¹⁵

The 1,114-Case Pattern

These famous cases matter, but they are not the whole story. They are the visible peaks of a larger range. In my dissertation I worked through 5,278 NDE accounts — 832 from peer-reviewed scholarly sources and 4,446 from public databases at NDERF and IANDS. Of those, 1,114 contained at least one report of accurate observation made at a distance from the body during clinical death.¹⁶

The patterns are consistent across the dataset. The most commonly reported veridical observations are, in this order: accurate descriptions of resuscitation procedures and equipment; accurate identification of medical staff (their faces, their actions, sometimes their names); accurate observations of family members in waiting rooms or in halls; and accurate reports of conversations between staff or family that occurred while the patient was clinically unconscious. Cases in cardiac-arrest contexts — where brain shutdown is well documented — show up at higher rates than cases in other medical settings, and their veridical content tends to be sharper, not vaguer. Chapter 12 develops the cardiac-arrest sub-dataset in detail.

What the Physicalist Says

The strict physicalist position is that consciousness is identical to brain activity. On that view, no veridical perception during clinical death should be possible. The brain that is doing nothing measurable cannot perceive a tennis shoe on a ledge or a flapping elbow at the operating table. Pam Reynolds's account, on the strict view, is impossible.

So what do physicalists say? Three main responses, and each deserves a fair hearing.

First, "These cases are selection bias." The argument is that researchers cherry-pick the impressive cases and ignore the misses. There is something to this for the qualitative literature. Sabom and Ring did not publish hundreds of mundane NDEs alongside Pam Reynolds; they published the strong ones. But the response cuts only so far. The 4,446 quantitative cases in my dataset come from public submission databases — anyone with a story can submit, and most do not have to be talked into it. The pattern of veridical reports shows up in the unfiltered database too. Janice Holden, in her review for the standard Handbook of Near-Death Experiences, found that across all major published case collections, about ninety-two percent of veridical claims that could be checked turned out to be accurate.¹⁷ If the appearance of veridicality were just selection bias, the unfiltered samples should look different. They do not.

Second, "The reporters were unconscious for less time than they think." The idea is that perception happens just before or just after the period of medical crisis, while some brain activity is still on, and the patient later assigns it to the wrong moment. This response works for some cases. It does not work for the cases that matter most. Pam Reynolds had monitored brainstem responses; her perception window included the period of confirmed flat-line. The AWARE Southampton case is timestamped against the cardiac-arrest record. Sabom's Atlanta NDErs were describing details — like the placement of crash-cart drugs — that occurred during their documented arrests, not before or after. The "wrong-time" hypothesis works only when the timing is loose. The strongest cases have tight timing.

Third, "The accurate details are confabulation — the brain filling in plausible content after the fact." This is the response taken seriously by the most careful skeptics, like Christopher French and the late Susan Blackmore in her earlier work.¹⁸ The brain, recovering from anoxia (lack of oxygen), reconstructs a story. The patient, asked what happened, generates a narrative consistent with what they have learned about the procedure since waking. The narrative feels like memory but is not.

This is a real psychological process. The brain does fill in. But it does not fill in verifiable detail that the patient could not have learned through normal channels. Confabulation explains why someone might say "I saw the doctor working on me" when the doctor was, in fact, working on them. It does not explain why Pam Reynolds described an unusual surgical instrument she had never seen. It does not explain why Al Sullivan described the unique mannerism of a surgeon he had never watched. It does not explain why Maria reported a tennis shoe in a place no one inside the hospital had reason to mention. Confabulation is a real factor in many NDE reports. It is not a candidate explanation for the strongest veridical content.

What This Chapter Has and Has Not Shown

I want to be careful here. The case for substance dualism is not won by any single chapter, and certainly not by any single case. Chapter 23 brings the whole evidential pattern to its synthesis. What this chapter has shown is more modest, but it is real.

It has shown that distant veridical observation during clinical death is not a rare anomaly. It is a documented, reproduced, multi-source pattern. Pam Reynolds is not a single oddity. She is a vivid representative of a class of cases — a class of more than a thousand cases in a single research dataset, reported across decades, across medical settings, across continents. The cumulative weight of those cases does not depend on any single one being airtight. Some of the cases are ironclad. Some are merely strong. Some are weaker. The pattern survives the loss of the weaker cases.

It has shown, further, that the standard physicalist responses — selection bias, mistimed perception, confabulation — handle some cases but not the strongest ones. Each response is plausible somewhere. None of them is plausible everywhere the evidence runs.

It has not shown that physicalism is logically impossible. A clever non-reductive physicalism can always retreat to "some unknown brain mechanism that we have not yet discovered." That is a serious philosophical position. But notice what it concedes. It concedes that the strict identity claim — consciousness is brain activity — has failed empirically. The fallback is no longer "the brain produces consciousness through known mechanisms" but "the brain might produce consciousness through unknown mechanisms that violate every prediction we currently know how to make." That is not the same claim. Chapter 15 takes up the leading version of that fallback — the dying-brain hypothesis built on Borjigin's rat-EEG work — and shows where its mechanism does and does not reach.

And it has not shown which theological framework is correct. The evidence here supports the claim that a person — a self, a consciousness, a soul — can perceive accurately while the body is offline. It does not, by itself, settle whether that soul is metaphysically simple, whether it is created at conception or generated through procreation, whether it survives indefinitely or only for a window. The evidence does theological work, but it does not do all the theological work. It does not need to.

Why This Matters at the Bedside

Most of this book is about evidence. This chapter is the heaviest stretch of evidence in the book. So I want to close with something that is not evidence but pastoral observation.

Families at hospital bedsides ask one question more than any other: Is she still in there? The mother whose teenager has been in cardiac arrest for forty minutes. The husband whose wife has just been pronounced. The daughter sitting with her father after the ventilator is turned off. They do not phrase the question theologically. They phrase it as a parent or a spouse or a child phrases the question. Is she still there?

The evidence in this chapter does not give them a complete answer. But it gives them something. It gives them grounds — empirical, careful, peer-reviewed grounds — for believing that the person they love does not simply switch off when the brain switches off. Pam Reynolds was watching from the corner of the operating room while her body lay drained on the table. Maria was hovering outside a third-floor window while her heart was stopped. Al Sullivan was looking down at himself while his sternum was open and his surgeon was flapping his elbows. The simplest description of what was happening in those moments is that the person was still there. Their body had stopped. They had not.

The next chapter takes up a category of veridical evidence even harder to explain: the NDEs reported by people who have been blind from birth. If you have found this chapter persuasive, the next one will press the argument further. The brain that has never had a working visual cortex cannot, on the physicalist account, generate visual content at all. And yet, again and again, blind people in cardiac arrest see.