Chapter 6

The Boundary Question

When Are You Actually Dead?

Her heart stopped at 9:14 on a Tuesday morning. The line on the heart monitor — the line that had jagged up and down for fifty-six years, beat by beat, second by second — fell flat. It just sat there. A single green stripe of nothing.

The cath-lab team did everything right. Compressions began within seconds. Within a minute, the crash cart was beside the bed and the first shock had been delivered. Within two minutes, the anesthesiologist had secured an airway and was pushing oxygen into her lungs by hand. Within five minutes, three rounds of epinephrine had been pushed through the IV. Within ten minutes, the team had cycled through chest compressions twice, with three nurses and two physicians taking turns so that no one's arms gave out before her heart restarted. Within fifteen minutes, the attending was beginning to say the things that attendings say when they are about to call it.¹

And then — at the sixteen-minute mark, on the fifth dose of epi — the line on the monitor moved. A small flutter at first. Then a recognizable rhythm. The team kept working. By twenty minutes, she had a steady pulse. By thirty minutes, she was stable enough to move to the cardiac ICU.

Four days later, when she was awake and talking, she told her daughter what had happened during those fifteen minutes. She had risen up and watched it from the corner of the room. She had seen a tall doctor with reddish hair who had not been part of her usual team — a man she would later identify, from his hospital photo, as the cardiologist who had been called in from another floor. She had heard her sister talking on the phone in the family waiting room — a conversation her sister did indeed remember having, and a conversation she had not relayed to anyone in the cath lab. She had felt herself moving toward something she described, with some hesitation, as a presence. She had been told, by something or someone in that presence, that she had to go back. And then she was waking up, four days later, sore in every rib from CPR, with a daughter weeping at the bedside.²

Now I want you to sit with this case for a moment, because there is a question hidden inside it that the rest of this chapter is going to be about. The question is not whether she had an experience — she did, and we will spend much of this book examining the evidence for the reality of such experiences. The question is not whether her perceptions were accurate — for the moment, take it on the documented evidence of cardiac-arrest NDE research that some such perceptions are.³ The question I want to put on the table is different. It is this:

During those fifteen minutes between 9:14 and 9:30, was she dead?

Most people, when they hear this question, give a quick answer. Some say yes — clearly she was dead, her heart had stopped, she had no spontaneous breathing, the line was flat. Others say no — clearly she was not dead, because she came back, and dead people don't come back. A few stop and think and say, well, what do you mean by dead?

That last answer is the right one. And the surprising thing — the thing this chapter is here to show you — is that the medical and philosophical communities have spent the last sixty years trying to answer that very question, and they have not finished yet. There is, at this moment, no settled definition of what it means to be dead.⁴

I know that sounds startling. It sounded startling to me when I first started reading the bioethics literature on this topic. Most of us assume that death is one of those words medicine has nailed down — a precise, technical, agreed-upon thing. We assume that when a doctor signs a death certificate, she is recording an objective biological fact, the way she might record a blood pressure or a temperature. The assumption is wrong. What she is recording is the satisfaction of a set of criteria, and those criteria have been argued over, revised, and fought about for decades. The fight is not over. And the answer to the question "when are you actually dead?" is, for the time being, an answer that depends on which committee, which jurisdiction, and which medical team you ask.

Stay with me here. This matters more than it might first appear. What we are about to find is that the very gray zone where the medical community is still arguing — the minutes after the heart stops, the period of measurable brain shutdown, the window before what we used to call biological death — is precisely the gray zone where near-death experiences happen. The cases the rest of this book examines occur in territory the medical community has not yet mapped. To understand the evidence, we have to understand the territory.

The Question Behind the Question

Let me try to say what is at stake. There are at least three different questions hiding inside the everyday question "is she dead?" and they are not the same question.

The first question is biological. Has the body, as a coordinated biological system, ceased to function in a way that cannot be reversed? This question is about cells, organs, and systems. It is about whether the heart can be restarted, whether neurons are still alive, whether the body is still doing the work of being a living organism. This is a real question, and modern medicine can sometimes answer it with reasonable precision — though, as we will see, the answer keeps changing as medicine becomes able to reverse more things.

The second question is medical-legal. By the criteria that this society has agreed to use for legal and practical purposes, has this person met the threshold of being declared dead? This question is about paperwork. It is about whether a death certificate can be signed, whether organs can be procured for transplant, whether the body can be released to the funeral home, whether the will can be read. The answer to this question is set by law and by professional convention. It tracks the biological question, but it is not identical to it.⁵

The third question is philosophical-theological. Has the person — the someone, the self, the soul — departed? This is the deepest question, and the one that families at the bedside actually want answered. When a daughter asks "is she still in there?" she is not asking a biological question or a medical-legal one. She is asking whether her mother is still present in the body, or whether her mother has gone. The Christian tradition has consistently said that a person dies when the soul separates from the body.⁶ The trouble is that no one — not the cardiologist, not the philosopher, not the priest at the bedside — has any direct way to observe the soul's departure. We have to infer it from physical signs. The medical criteria are, in the end, attempts to approximate the moment when the soul departs. They are not direct observations of it.

These three questions can come apart. A patient can be biologically dead by the older criteria (no breath, no heartbeat) but not yet legally declared dead, because no one has found her yet. A patient can be legally declared dead by current brain-death criteria but, on the most rigorous biological account, still doing some of the work of being a living organism (we will come to Alan Shewmon's striking findings shortly). A patient can have stopped breathing and had her heart stop — by the standards of every century before our own, dead — and then come back, with stories about what happened during those minutes that suggest the deepest question, the philosophical-theological one, may have a different answer than the biological question alone can give.

Here, then, is the framing this chapter will work out: death is not a single instant; it is a process, and the medical-legal markers we use to identify "the moment of death" are conventions — useful conventions, often necessary conventions, but conventions all the same. The contemporary debate over what counts as death is not a sign that medicine has failed; it is a sign that medicine has begun to ask the right question. And the gray zone where the question is most contested — the zone where a person is, in some real sense, neither clearly alive nor clearly dead — is the zone where NDE research has gathered its strongest evidence.

How We Used to Know You Were Dead

For most of human history, the question "when are you dead?" had a simple, intuitive answer. You were dead when you stopped breathing and your heart stopped. The standard test of life was breath. In the Old Testament, the word for the breath of life — nəšāmâ — was nearly the same as the word for spirit.⁷ When breath went, life had gone with it. The Greeks held a similar view. The Romans tested for death by holding a feather or a small mirror to the nostrils; if the feather did not stir or the mirror did not fog, the person was dead.⁸

For most practical purposes, the breath-and-heartbeat criterion was a good one. It picked out the right people most of the time. A person without breath or pulse for any extended period was, in pre-modern medicine, almost certainly dead. The signs of life — warmth, color, responsiveness, breath — went together. When they all departed at once, the patient had died.

The criterion was not flawless. There were always cases that troubled it. Plato told the story of Er, the soldier whose body lay on a pyre for many days before he revived and told what he had seen.⁹ Pliny the Elder catalogued cases of people thought dead who later sat up.¹⁰ The medieval Christian world had its tales of figures whose corpses moved or spoke. Most of these were probably either misperceptions or premature declarations — but they were enough, in the eighteenth and nineteenth centuries, to give rise to a real and widespread fear of being buried alive.¹¹ The Victorian era saw a small industry of safety coffins, mortuary alarms, and waiting mortuaries where the not-yet-decayed body would be observed for several days before burial. The fear was that the breath-and-pulse criterion was sometimes wrong; that a person could appear breathless and pulseless without being truly dead; that the difference between "no detectable signs of life" and "actually dead" was a difference that could be missed.

For most of the time those fears were live, the medical solution was patience. Wait long enough, and the question would settle itself. A truly dead body would begin, within hours, to show the signs of biological decomposition: the cooling (algor mortis), the stiffening (rigor mortis), the pooling of blood (livor mortis). A person who was not actually dead, but only deeply unconscious, would stir before any of these arrived. The boundary was crossed when the body, left alone, did not return.

The criterion held — more or less — until the middle of the twentieth century. And then, in a stretch of roughly fifteen years, several technologies arrived that broke it.

The first was mechanical ventilation. The iron lung had been around since the polio epidemics of the 1930s, but it was only one of many devices. By the 1950s, hospitals had positive-pressure ventilators that could push air into a patient's lungs whether the patient's brain was telling the diaphragm to move or not. A patient whose brain had stopped sending the signal to breathe could now be kept breathing — or, more precisely, kept oxygenating — for hours, days, weeks, sometimes years. The chest still rose and fell. The chest still made the sound of a chest. But the breathing itself had been outsourced to a machine.¹²

The second was cardiopulmonary resuscitation. Closed-chest cardiac massage was developed in 1960; the standard of CPR became routine in emergency medicine by the late 1960s.¹³ A heart that had stopped could now sometimes be restarted. The fact of the heart having stopped was no longer the end of the matter; it was the beginning of an intervention.

The third was the cardiac defibrillator and the rest of the modern resuscitative arsenal — emergency drug protocols, hypothermia therapy, extracorporeal membrane oxygenation (ECMO), and most recently, the protocols for prolonged resuscitation that have stretched the window of recoverable cardiac arrest from minutes to, in some carefully managed cases, hours.¹⁴

The fourth, and most consequential for our subject, was the development of solid-organ transplantation. The first human heart transplant was performed by Christiaan Barnard in December 1967. The donor was a young woman whose brain had been catastrophically damaged in a car accident; her heart was still beating when it was removed.¹⁵ Barnard had to make a judgment that would have been unthinkable a decade earlier: the donor was dead, even though her heart was still beating. He had to decide that, even though her heart was still functional, she was not.

The breath-and-heartbeat criterion could not absorb these developments. A patient on a ventilator with no brain function had breath of a sort. A patient whose heart had been restarted by CPR had a heartbeat — but had also, by the older criterion, just been dead and was now alive again. The transplant donor had a beating heart that no longer belonged, in any meaningful sense, to a living person. The criterion that had served for thousands of years was no longer doing the work the medical community needed it to do.

Something had to give. What gave, in 1968, was the criterion itself.

The Harvard Criteria, 1968

In August 1968, the Journal of the American Medical Association published a paper that quietly changed how medicine would understand death for the next half-century. The paper was titled "A Definition of Irreversible Coma." It was written by an ad-hoc committee at Harvard Medical School, chaired by Henry Beecher, an anesthesiologist whose previous fame had come from a landmark essay on the ethics of human research.¹⁶

The Harvard committee proposed that medicine should adopt a new criterion of death: the irreversible loss of all brain function, including the function of the brainstem. A person whose entire brain — cortex, brainstem, all of it — had irreversibly stopped working should be considered dead, even if a ventilator was keeping the chest moving and even if the heart was still beating. The criterion came to be called whole-brain death.¹⁷

The committee laid out four criteria for the diagnosis: (1) unreceptivity and unresponsivity — the patient does not respond to any stimulus, however painful; (2) no movements or breathing for at least an hour, including no respiratory effort when removed from the ventilator for three minutes; (3) no reflexes, including no pupillary response, no eye movement, no swallowing or yawning, no postural reflexes; and (4) a flat electroencephalogram, taken twice with a twenty-four-hour interval. They added, sensibly, that the diagnosis should not be made if the patient was hypothermic or had a central-nervous-system depressant in their bloodstream, since both of those conditions could mimic brain death without actually producing it.¹⁸

Now I want to be honest about something. The Harvard committee did not invent the concept of brain death out of pure scientific curiosity. The paper itself names two reasons for the new criterion. The first is the burden on patients and families of indefinite mechanical support of the irretrievably comatose. The second is, I quote — and this is in the published paper, not somewhere else — that "obsolete criteria for the definition of death can lead to controversy in obtaining organs for transplantation."¹⁹ The new criterion solved a real medical-ethical problem (when to discontinue support for the irretrievably comatose) and a real practical problem (how to get organs while they were still useable). The two motivations were tangled together from the start, and the tangling has never quite come undone.

I do not say this to discredit the Harvard committee. The committee was made up of serious people doing their best in the face of a genuine new challenge. But the criterion they proposed was not handed down from biology. It was a convention, agreed to in a particular institutional setting for particular medical-ethical purposes. It was a useful, defensible convention — but it was a convention.

The Harvard criteria spread quickly. By the early 1970s, most major American hospitals had adopted them in some form. Other countries developed their own versions. The United Kingdom's "Conference of Medical Royal Colleges" issued guidance in 1976 that focused, slightly differently, on irreversible loss of brainstem function — a criterion that came to be called brainstem death and that, in the UK, has been the standard ever since.²⁰

By 1981, the United States needed a single legal definition. President Carter's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research issued a report. The report's recommendations were turned, that same year, into the Uniform Determination of Death Act (UDDA), which has been adopted, with minor variations, by every American state. The UDDA reads, in its entirety:

An individual who has sustained either (1) irreversible cessation of circulatory and respiratory functions, or (2) irreversible cessation of all functions of the entire brain, including the brain stem, is dead. A determination of death must be made in accordance with accepted medical standards.— Uniform Determination of Death Act, 1981²¹

That is the law of the land. It looks tidy. The two criteria — either circulation-and-respiration or whole-brain — are presented as if they pick out the same underlying state from two different angles, like two different thermometers reading the same temperature. The framers of the UDDA thought they did. The trouble is that, as we are about to see, they do not always.

Shewmon's Challenge

Throughout the 1980s and into the 1990s, the consensus around whole-brain death held. Most physicians treated the criterion as settled science. Most philosophers and theologians who weighed in accepted the consensus. Most bioethics textbooks presented brain death as a real, biological fact that medicine had at last properly identified.

And then a Catholic neurologist at UCLA named Alan Shewmon began to raise some very uncomfortable questions.

Shewmon had originally been a defender of the brain-death criterion. He had taught it, written in support of it, helped develop the technical guidelines used to diagnose it. But he had also been, for years, gathering case after case in which patients diagnosed as brain dead by all the standard criteria did things that the standard model said brain-dead patients should not be able to do.²²

The argument that whole-brain death is real death had always rested on a particular biological claim. The claim was that the brain — and only the brain — is the integrator of the body. Take the brain out of the equation, the argument went, and the body cannot function as a coordinated whole. It will fall apart within hours or, at the most, days. The body parts may continue to be alive in some narrow sense (cells in a Petri dish are alive in some narrow sense), but the body as an organism — as a unified, integrated, living whole — has died. Brain death, on this account, is the death of the organism.²³

Shewmon began to publish papers showing that this claim is empirically false. He documented cases — many cases — of patients diagnosed as brain dead who continued to do many of the things that integrated organisms do. They maintained body temperature. They fought infections. They healed wounds. They digested nutrients delivered through feeding tubes. They responded to stress with appropriate hormonal cascades. Pregnant women diagnosed as brain dead carried fetuses to term. Children diagnosed as brain dead grew, sexually matured, and in some cases lived for years. Shewmon assembled a series of more than 175 cases of "chronic" brain death — cases of survival beyond a week, with one extraordinary case, "TK," surviving for more than twenty years.²⁴

Now I want to be careful about what Shewmon was claiming. He was not arguing that brain-dead patients are conscious. The brain-dead, by every measure we have, are not conscious. They cannot perceive, think, choose, love, or pray. Whatever ability the body has to do the work of being a living organism, that ability — when the brain is destroyed — does not include the work of being a person. Shewmon was not denying any of this.

What Shewmon was arguing is that the philosophical claim used to justify calling brain death "death" — the claim that the brain is the integrator without which the body cannot function as an organism — is empirically false. The body has more capacity for self-integration than the original argument allowed. Some of the integration the brain was supposed to do is, it turns out, done at the level of organs, the spinal cord, and various humoral feedback loops. The brain-dead body is not a heap of disconnected parts being held together by a ventilator. It is a damaged organism, badly damaged, but still capable of substantial coordinated function for a long time.²⁵

The implication is uncomfortable. If brain death is going to be defended as real death, it cannot be defended on the grounds that the brain-dead body is no longer an integrated organism. It must be defended on different grounds. Some defenders of the criterion have shifted to arguing that brain death is the loss of consciousness — that what makes the brain-dead "dead" is not biological disintegration but the permanent absence of the higher functions that make a person a person. But this is a different argument, and it has its own problems. (It would, for instance, make patients in persistent vegetative state — patients with no higher cortical function but with a working brainstem — also dead, which most defenders of the standard criterion are not willing to say.²⁶) Other defenders have shifted to a more pragmatic position: brain death may not be a perfect biological boundary, but it is a useful and stable medical-legal one. That is honest, but it is not the original claim. The original claim was that we had finally identified what death really is. The shifted claim is that we had agreed on a useful convention.

The Five-Minute Wait

If brain death is one part of the gray zone, circulatory death is the other. And here the medical-ethical questions are, if anything, sharper.

Most organ donors today are not whole-brain-dead patients on ventilators. Most are patients who die in a different, more familiar way: their hearts stop, breathing stops, and shortly afterward — minutes afterward — their organs are recovered for transplant. This is called donation after circulatory death, or DCD. It has become, in the last twenty years, the largest source of transplantable organs in many countries.²⁸

The DCD protocol works like this. A patient is in the hospital who is not going to recover but who does not, by current criteria, meet whole-brain death. Perhaps she has had a massive stroke. Perhaps she has end-stage heart failure. Perhaps she has suffered a catastrophic accident from which she will not return. The family, with the medical team, makes a decision to withdraw life support. Life support is withdrawn. The patient's heart stops.

And then the team waits.

How long? Most American protocols call for a wait of two to five minutes after the heart stops before organ procurement begins. Some protocols use ninety seconds. A few have used as little as seventy-five seconds.²⁹

The wait is there for a specific reason. It is supposed to ensure that the cessation of circulation is permanent. The protocols want to be confident that the heart is not going to start again on its own, which, rarely but documentedly, hearts do — a phenomenon called the Lazarus phenomenon or autoresuscitation.³⁰ If the heart might restart on its own, then the patient is not yet permanently dead, and procurement would be the act of killing her. If the heart will not restart on its own, then she is permanently dead, and procurement is taking organs from a dead body.

Now sit with the math here. The wait is two to five minutes. After two to five minutes, the team begins the procurement. They reopen the chest. They cool the organs. They take them out and put them in coolers and rush them to recipients. In some recent protocols, especially in heart-DCD cases, they actually restart the donor's heart in the donor's body, with extracorporeal pumps, in order to preserve the heart and other organs while procurement takes place. The heart that had been declared permanently stopped, two to five minutes earlier, is now beating again.³¹

How can it be true both that the heart was permanently stopped and that the heart can be restarted? The answer the bioethics literature has settled on, with some unease, is a distinction between irreversibility and permanence. James Bernat, one of the most careful defenders of the current protocols, argues that the heart must be permanently stopped — meaning, no one is going to try to restart it — but it does not need to be irreversibly stopped — meaning, restarting it is impossible. So long as the decision has been made not to attempt resuscitation, and so long as no spontaneous restart will occur in the wait window, the patient is "dead enough" for the purposes of procurement.³²

I am not going to argue here against DCD protocols. I think the practice is, on balance, defensible — the patients are not going to recover, the families have consented, and the organs save lives. But I think we should be honest about what is happening. The DCD protocols depend on a definition of death that the developers themselves describe as a convention. The convention is that we will treat a patient as dead after a sufficient waiting period in which no spontaneous return of circulation occurs and the medical team has decided not to attempt resuscitation. The convention picks out a useful pragmatic line. But the line is not a biological essence. It is a line we have drawn.

Robert Truog, a Harvard bioethicist, has been particularly direct about this. Truog argues that the bioethics community has been engaged, for decades, in a kind of "noble lie" — pretending that the lines we draw for organ procurement track real biological facts about death, when in fact they are pragmatic compromises designed to allow procurement without quite saying we are taking organs from people who are still alive. Truog thinks we should be more honest about this; some of his colleagues think his honesty is corrosive. The debate is unresolved and unresolvable on current terms.³⁴

The Reversibility Puzzle

There is one more piece of the puzzle to put on the table before we step back and ask what all this means for our subject. It is the strangest piece, and I want to spend a few paragraphs on it because the strangeness matters.

A standard moment in any cardiac-arrest case is the moment when CPR is stopped. The team has been working for fifteen, or twenty, or thirty minutes. The line is still flat. The drugs are not working. The compressions have not produced a pulse. The attending physician makes a judgment call: the patient is not coming back. Time of death, 9:30. The team stops working. The bed is wheeled to a quieter part of the unit. The family is told.

Now suppose, instead, that the team had not stopped. Suppose they had continued for another five minutes, and on minute thirty-two, the rhythm had returned. The patient would have lived. There are real cases like this — cases of late-resuscitation success, especially with the modern protocols of high-quality compressions, hypothermia, and ECMO. Some patients have been resuscitated after more than an hour of CPR.³⁵

If the team had stopped at minute thirty, the patient would have been declared dead at 9:30. If the team had continued and gotten a pulse at minute thirty-two, the patient would not have been declared dead. The biology of the patient's body is, at minute thirty, the same in both cases. What differs is the team's decision about what to do next.

This is a very strange feature of the current criteria. Whether a person was dead at 9:30 depends, in part, on what the team did after 9:30. If they kept going and succeeded, she was not dead. If they stopped, she was. The biological state at 9:30 was identical. The status — dead or not dead — depends on a downstream choice.

The bioethics literature has worked hard to absorb this. The careful position is that "irreversibility" in the criterion of death does not mean "could never have been reversed under any conceivable intervention" but "will not be reversed, given the actual decisions made about treatment." On this view, irreversibility is permanence-given-the-decisions-actually-made. The patient was dead at 9:30 because, given that the team stopped at 9:30, no further reversal occurred.

I think this is honestly the best one can do with the current concepts. But notice what it commits us to. It commits us to saying that whether someone was dead at a particular moment is, in part, a fact about what the medical team chose to do — not just about the biology of the patient. Dead turns out to be a more relational, more decision-dependent concept than common usage allows. This is not a result the framers of the UDDA quite intended, but it is the place the careful analysis ends up.³⁶

Now bring this back to NDEs. A patient who experienced a coherent, lucid, sometimes verifiably accurate experience during the period between 9:14 and 9:30 — between cardiac arrest and successful resuscitation — was, by the standards of every century before our own, dead during that window. By the standards of the current UDDA, she was probably not dead, because the cessation of her circulation was not permanent (the team kept going and succeeded). But she met every observable criterion of death during the window. Her heart was not beating. Her lungs were not moving. Her EEG, if monitored, would have been flat within thirty seconds of arrest.³⁷ Her brainstem reflexes were absent. There was no public, observable difference between her state at 9:20 and the state of a patient on whom CPR had been stopped at 9:20 and who was about to be declared dead.

The difference between dead and not yet dead, in such a case, is not a difference any onlooker can see. It is a difference that depends on what happens later. The window of NDE experience falls inside this gap.

The Christian Tradition and the Gap

I want to step back now from the medical literature and put alongside it the Christian tradition's account of death. The traditional Christian account is older than any of the medical criteria we have just walked through, and it is, in a certain way, more honest about what is going on.

The Christian tradition has held, with remarkable consistency across two millennia, that death is the separation of the soul from the body. The body, on this account, is the soul's instrument and home in the embodied life. At death, the soul departs. The body, without the soul, ceases to be a living person and becomes — in the language of the tradition — a corpse, awaiting resurrection.³⁸

This is the view assumed by the biblical texts. When Rachel dies in Genesis 35:18, the text says her nepheš — her soul, her life — was departing.³⁹ When Elijah revives the widow's son in 1 Kings 17:21–22, he prays that the child's nepheš would return to him, and the text says it did. Ecclesiastes 12:7 says that at death, the dust returns to the earth as it was, and the spirit returns to God who gave it. The Lord Jesus, on the cross, commits his spirit to the Father (Luke 23:46) and breathes out (Mark 15:37). James writes that the body without the spirit is dead (James 2:26). These are not metaphors imported from Greek philosophy; they are the consistent biblical idiom of death-as-departure of the inner self.⁴⁰

The patristic and medieval Christian traditions developed this account into a fairly precise theology. The soul separates from the body. The angels accompany the soul (Luke 16:22). The saved soul goes to be with Christ (Philippians 1:23, 2 Corinthians 5:8). The unsaved soul goes to a holding place — what the Old Testament calls Šəʾôl and what the New Testament calls Hadēs — to await the final judgment.⁴¹ The body remains on earth, awaiting the resurrection. The Christian tradition, on this view, is a tradition that takes the gap between death and resurrection seriously. The dying person is conscious throughout. The Christian tradition has, in fact, almost always assumed a conscious intermediate state.⁴²

Now notice what the traditional Christian account does not do. It does not pretend to give a publicly observable test for the moment of separation. The soul is not visible to the doctor at the bedside. The moment of departure cannot be timestamped on a chart. What the doctor can observe are the physical correlates of the soul's withdrawal — the cessation of breath, the failure of the heart, the loss of brain function — and from these correlates the doctor infers that separation has occurred or is occurring.

The Christian tradition is therefore willing to say something the contemporary medical-legal account is reluctant to say: that the moment of death is a real metaphysical event that has happened or has not happened, but that our inferential access to it is approximate. The medical criteria are heuristics. They are not direct observations. The actual event — the soul leaving the body — is not the kind of event the medical instruments can measure.

This is, I think, more honest than what the contemporary literature can manage. The contemporary literature is forced to stretch and twist the concept of death because it is unwilling to admit that the underlying event is not a publicly observable physical change. The medical-legal community has been trying for sixty years to find some physical fact such that, the moment that fact obtains, the patient is dead. The trying has not been successful. The trying has not been successful because there is no such physical fact. There is a process of bodily failure, more or less rapid, that begins at some point and ends at some point, and somewhere along that process — somewhere — the soul departs. The departure is not the same event as the cessation of cortical activity, or the cessation of circulation, or the cessation of integrated organ function. It is the underlying event that the cessations are all approximating.

Stay with me. I want to be clear that this Christian account is a metaphysical claim that goes beyond what the medical literature can establish on its own resources. I am not claiming that the medical literature, looking at the evidence, must conclude that there is a soul. I am claiming something different. I am claiming that the medical-legal account has run into the limits of what it can establish on physical evidence alone, and that the most natural account of the gap — the account that makes sense of why the boundary is so hard to draw — is an account in which the boundary is a metaphysical event that the physical signs approximate. The Christian tradition has been giving that account for two thousand years.

And here is where the NDE evidence enters.

NDEs Live in the Gap

The boundary question is not just a question for bioethics. It is a question for our subject. The cases this book examines occur, almost without exception, in the gray zone we have been mapping. Cardiac arrest patients whose hearts stopped, whose breathing stopped, whose EEGs went flat — and who, during that period, had experiences they later reported.⁴³ By the standards of every century before the twentieth, those patients were dead during the window. By the standards of the contemporary UDDA, they were not (because the cessation was reversed). By every observable physical criterion in the moment, they met the threshold of death. Their reports of consciousness during this window are reports of consciousness during precisely the window where the medical-legal community is unsure whether to call them living or dead.

The full evidential force of cases like these will be developed when we come to the case files (especially Chapter 10 on accurate distant observation and Chapter 12 on cardiac-arrest NDEs and the EEG problem). For our purposes here, the point is more limited but still important. It is this: the cases occur in territory the medical community has not finished mapping. The cases involve patients whose status — dead or not dead — depends on choices made after the window of experience and on philosophical commitments the bioethics community has not settled. Whether to call these patients "dead during the experience" turns out to depend on what we mean by "dead" — and that, as we have seen, is not a settled question.

This means a particular kind of objection to NDE evidence loses its bite. Some critics have argued that NDEs cannot be evidence about life after death because the patients were not really dead — by definition, they came back. The argument seems forceful at first glance, but it depends on the older breath-and-heartbeat criterion together with the new requirement of irreversibility, which is incoherent on its face.⁴⁵ If "dead" means "permanently dead in the sense that no return is possible," then the term has become synonymous with "biologically extinct" and the question of whether the patient was "dead" during the window becomes a tautology — they came back, so by definition they were not dead. But on this stipulation, no one we can interview can have been "dead" during their NDE, because we can only interview people who came back. The objection collapses into the claim that the dead cannot be interviewed, which is true but trivial.

The serious question is whether, during the window, the patient's body and brain were sufficiently shut down that, on the standard physicalist account, no consciousness should have been possible. The answer to that question is well-established in the medical literature. Within seconds of cardiac arrest, blood flow to the brain ceases. Within ten to twenty seconds, organized cortical activity disappears. Within thirty seconds, the EEG goes effectively flat. Within minutes, the metabolic substrate that consciousness — on the physicalist account — depends on is no longer there.⁴⁶ The strict physicalist prediction is unambiguous: during this window, there should be no consciousness, no perception, no coherent experience, and certainly no veridical perception of events distant from the body. The reports we will examine in later chapters say otherwise. The contradiction between the strict prediction and the actual reports is the heart of the evidential case the book develops.

Two Honest Objections, Honestly Handled

Before we close this chapter, I want to bring forward two objections that thoughtful readers will likely have. I want to handle them honestly because both deserve more than dismissal.

The first objection runs like this. Even if the boundary of death is contested in bioethics, surely there is something simple and definite about the patients we are talking about. Cardiac arrest is well-defined. Flat EEG is well-defined. The reports are coming from people whose medical status during the window is, in the relevant respects, not actually mysterious. The bioethics gray zone is a real but separate question. We do not need to settle bioethics to evaluate NDE evidence.

This is partly right. The neuroscience of brain shutdown during cardiac arrest is, as Chapter 5 developed, well-established. The EEG goes flat. The cortex is offline. The strict physicalist prediction is clear. We do not need to resolve every question in bioethics to know that consciousness, on the physicalist model, ought to be impossible during the window. So the evaluation of the NDE reports does not depend on getting bioethics right.

But the bioethics question is not separate. It is the question that frames what the NDE evidence is evidence of. If the medical-legal community had simply stipulated that no patient who comes back was ever dead, then NDE reports would be reports of unusual experiences during medical crises — interesting but not metaphysically loaded. The very fact that the bioethics community is unsettled about whether such patients are or were dead during the window opens up the metaphysical question. The NDEs are reports of consciousness during a period the medical community itself describes as the gray zone. That gray-zone framing is what makes the evidence probative.

The second objection is more philosophical. If the moment of death is a metaphysical event that physical signs only approximate, then how can we ever know when it happens? We are dependent on the signs. The signs are imperfect. The honest position would be a kind of agnosticism about whether anyone has, in any given case, "actually" died.

I think this objection is correct as far as it goes, and the right response is not to deny it but to live with it. We are dependent on the signs, and the signs are imperfect. There is no test that an attending physician can run to certify that the soul has departed. What the physician can do is observe the relevant correlates and make a reasoned inference. In ordinary cases — the elderly woman who has been failing for weeks, the cancer patient at the end of a long decline, the trauma victim with no realistic chance of recovery — the inference is reliable enough for all the practical purposes that matter. We declare death; we mourn; we bury; we proceed. The metaphysical event has, in those cases, almost certainly occurred when we say it did.

But there are also unusual cases in which the inference is less reliable. The cardiac-arrest patient who comes back. The DCD donor whose heart is restarted in another body. The brain-dead pregnant woman whose body carries a fetus to term. These cases are genuinely puzzling, and the right response to them is not to pretend our usual inferences work cleanly. The right response is to acknowledge that the boundary, in such cases, is harder to draw and that our certainty about what has happened is correspondingly lower.

NDE evidence enters this picture not as a way to settle the bioethics question but as a new kind of data point. If the central thesis of this book is right — if the soul is a real entity that survives the body and that NDE research has begun to give us empirical traction on — then the metaphysical event of soul-body separation is not entirely unobservable. It leaves traces. Some of those traces are the experiences NDErs report from the period of their separation. The experiences are not direct observations of the soul's departure; they are first-person reports of what the soul did during the gap. But they are reports we did not have access to before. They are, for the first time in human history, a body of accumulating, systematically studied first-person testimony from the boundary itself.

What We Can Say

So let me try to say, in plain terms, what this chapter has and has not established.

It has established that "death" is not a single instant but a process, and that the medical-legal criteria for "dead" are conventions that approximate the underlying reality without exactly capturing it. It has established that the criteria have changed three times in sixty years (from breath-and-heartbeat to whole-brain to a contested current set), that they are still being argued about, and that the gray zone where the argument is most active is the zone where NDE research has gathered its strongest evidence.

It has established that the philosophical-theological account of death — the soul's separation from the body — is a coherent and defensible account that has the advantage of being honest about the inferential gap between the underlying event and the physical signs. It has established that this account is the historic Christian account and that the historic account's inability to give a clean public test is not a defect but an admission of what is actually the case.

It has not established — and could not establish — exactly when, in any particular case, the soul departs. The question of whether the cardiac-arrest patient I described at the opening of this chapter was "actually dead" between 9:14 and 9:30 is a question I do not pretend to settle. By the older breath-and-pulse criterion: yes. By the contemporary UDDA reversibility-included criterion: probably no, because she came back. By a Christian metaphysical criterion: only God knows, and the answer turns on whether her soul was, in the relevant sense, separated during the window. What I can say is that her experience during the window was not a hallucination produced by her cortex (her cortex was offline), and it was not a confabulation produced afterward (some of what she reported was independently verified). What she reported is the kind of evidence we will examine in detail in the chapters to come.

The careful position, then, is this. The medical-legal community's lines around death are useful conventions. They are not lies, and we should keep using them — they serve real purposes, and they roughly track the underlying reality most of the time. But we should be clear-eyed about what they are. They are tools, not truths. The deepest question — when a person actually dies, in the sense of when a self ceases to be embodied — is a question the medical-legal community cannot answer on its own resources. It requires philosophical and theological work. It also, increasingly, requires careful attention to the empirical evidence that has begun to come in from the boundary itself.

Why This Matters

I want to close with three reasons this chapter's argument matters for the rest of the book — and for the readers who will be sitting at the bedsides where these questions are no longer abstract.

The first reason is evidential. The next chapters will present cases — many cases — of NDErs who report experiences during the window of cardiac arrest, flat EEG, and clinical death. Some readers will be tempted to dismiss these reports on the grounds that the patients were not "really" dead. This chapter has tried to show that the dismissal does not work. The patients met every observable criterion of death during the window. Whether to call the window "death" or "near-death" or "clinical death" or "the gray zone" is partly a matter of convention. What the patients reported during the window is not. The reports are what they are. They occurred during a period the medical community itself describes as the territory where, on the physicalist account, no consciousness should be possible. The chapters to follow examine, with care, what those reports show.

The second reason is theological. The historic Christian tradition has a clean and coherent account of death: the soul departs the body. This account does not depend on the medical-legal community settling its criteria. It does not require that medicine produce a perfect test for the moment of departure. It only requires that the underlying event — separation — is the kind of event that the physical signs approximate. The contemporary failure to settle the criteria, far from being a problem for the Christian account, is exactly what the Christian account predicts. If death is the separation of an immaterial soul from a material body, then no physical test can directly observe it. We will always be, in the end, working with signs and inferences. The Christian tradition has lived with this honestly for a long time. The medical-legal tradition is just beginning to admit that it must, too.

The third reason is pastoral. Families at the bedside ask "is she still in there?" The medical staff often have nothing helpful to say. Hospital chaplains sometimes have less than they should. Pastors trained in a generation that lost the historic Christian wisdom about dying may not know how to answer. The answer the historic tradition offers is sober and clear. Yes, she is still in there — until the moment she is not. That moment is real, but it is not the same as any of the public physical signs you can observe. It is the moment the soul departs. It may have departed already; it may be about to depart; it may not have happened yet, even when the EEG has been flat for several minutes. We are not in a position to say with certainty exactly when the moment occurs. We are in a position to commend her, in word and prayer, to the One in whose hands her departure rests. That commending is part of the historic Christian wisdom, and it is one of the things this book exists to help recover.

The chapter ahead — on the history of NDE research — turns to the empirical material itself. We will trace how, beginning in 1975 with Raymond Moody's Life After Life, a small number of physicians and psychologists started to do something that medicine had not previously done with the dying. They started to ask the survivors what they remembered. The answers they got — gathered now in over fifty years of accumulating research, and including the dataset of more than five thousand cases that grounds my own work — are what the rest of the book is about. Whether those answers are the kinds of answers that confirm, refute, or merely complicate the Christian theology of dying we have just outlined is the question the next several chapters will work to settle.

For now, hold this in mind. The question "when are you actually dead?" is not a question medicine alone can answer. It is a question that requires philosophy, theology, and — increasingly — the careful empirical study of those who have been to the boundary and come back to tell us what they found there.