Chapter 31

Children's NDEs

Innocence, Memory, and the Question of Age

In the summer of 1982, a seven-year-old girl named Katie was pulled from a community pool in Pocatello, Idaho. She had been underwater long enough that no one knew how long. By the time the paramedics reached her, her face was blue. She had no pulse. Her pupils were fixed and dilated. The young pediatrician at the hospital, a man named Melvin Morse, made the call to keep working on her even after the standard window had closed. He intubated her. He pushed fluids. He kept her warm. He waited. After what most of his colleagues would have called too long, her heart began again.¹

Katie woke up three days later. She did not remember being pulled from the pool. She did not remember the ambulance. But when Dr. Morse came to her room and sat down on the edge of her bed, she looked up at him and said, calmly and matter-of-factly, "That's the one with the beard. First there was this tall doctor who didn't have a beard, and then he came in." She was describing, accurately, the two physicians who had worked on her in the trauma bay — physicians she had never met while conscious and could not have seen, because her eyes were closed and her brain was not producing the activity that would have allowed her to see anything at all.²

Then she told him the rest. She had floated up out of her body. She had watched the doctors work on her. She had gone home and watched her mother in the kitchen, cooking — what was it? — chicken and rice. She had watched her brother playing with a G.I. Joe in the living room. Her father had been sitting on the couch. Each of these details, when Morse later checked, was confirmed exactly. The chicken and rice. The toy. The position of her father. And then Katie said something that, for Morse, was the moment everything changed. She said that during her time away from her body, a being had come for her. A woman she called Elizabeth, tall and kind, who had taken her up a tunnel into a place full of light. There she had met Jesus and the Heavenly Father. She had asked if she could see her mother again. They had told her yes — for now.³

Morse was a careful researcher. He had been trained at Johns Hopkins. He had taken his pediatric residency at Seattle Children's Hospital, and he had spent his career around dying children. He was also, by his own description at the time, a religious skeptic. He did not particularly want Katie's story to be true. But Katie had described, with no preparation and no leading questions, things she could not have known by any ordinary means. And she had done so without theological self-consciousness, without religious vocabulary borrowed from a sermon, without the practiced piety of an adult NDEr who had read books about the experience. She just told him what had happened to her, the way a seven-year-old tells you about a birthday party.⁴

That conversation set Melvin Morse on a path that would occupy the next decade of his life. He began to interview every child who survived a cardiac arrest, near-drowning, or other clinical-death event in his hospital. He compared them to a control group of equally sick children who had not been clinically dead — children with high fevers, severe asthma attacks, drug overdoses without cardiac arrest. The pattern was striking. Children who had crossed the line into clinical death told stories like Katie's. Children who had been seriously ill but had not crossed that line did not. The phenomenon, whatever it was, tracked with the brain's actual cessation, not with the severity of the illness.⁵

I begin this chapter with Katie because Katie's case has a particular evidential weight that an adult case rarely has. A seventy-year-old man who reports meeting Christ during a cardiac arrest is making a credible report, and his case belongs in the cumulative file. But a seventy-year-old man has spent his life inside Christian culture. He has seen paintings of heaven. He has heard sermons about the light. His brain has a thousand channels through which the imagery he reports could have arrived, even before the moment of his arrest. Katie had channels too — she had been to Sunday school a few times — but the channels were narrower. She had not read books about NDEs. She had not been taught the tunnel imagery. She had not been told there would be a being of light. She had been told, perhaps, that Jesus loves her. The rest she had to discover.

This chapter is about what children like Katie discover at the threshold of death, and about what their discoveries mean for the cumulative case for substance dualism, for the conscious intermediate state, and for the Christian theology of dying. Children's NDEs are not a sideshow to the main argument of this book. They are, in some ways, the cleanest evidence we have. And they raise theological questions of their own — questions about the salvation of those who die before they can speak, about what happens to the cognitively unable, about whether the encounter at death is, in fact, for everyone.

The Question Behind the Chapter

Why do children's NDEs matter? They matter for two reasons, one evidential and one pastoral.

The evidential reason runs like this. The strongest skeptical accounts of NDE phenomena — the accounts I have engaged at length in earlier chapters of this book — depend in different ways on the experiencer having brought something to the experience. The cultural-conditioning hypothesis says the experiencer brings a learned script (heaven, hell, tunnels, light, deceased relatives) and the dying brain assembles those bits into a hallucination. The expectation hypothesis says the experiencer brings a wish (to see a loved one, to be reassured, to escape pain) and the dying brain assembles a wish-fulfilling vision. The confabulation hypothesis says the experiencer brings little or nothing during the event itself but afterward fills in the gaps with material remembered from culture. Each of these accounts has some force when applied to an adult who has been steeped, since infancy, in the symbolic vocabulary of his culture.⁶

None of them has much force when applied to a four-year-old.

A four-year-old has no cultural script for the experience of dying. She has not read Moody. She has not seen any of the popular books on heaven. Her brain has not been conditioned by a hundred Easter sermons about the empty tomb. She knows, dimly, that there is a Jesus, perhaps a heaven, perhaps an angel. The categories are thin. The symbolic vocabulary is not yet deep enough to fill in a sustained narrative. And yet, when she is pulled out of a swimming pool with no pulse, she comes back and tells her parents about the tunnel, the light, the beautiful person who held her, the people who told her she had to come back to take care of her brother. Where did the script come from?⁷

The pastoral reason runs alongside it. The death of a child is among the most devastating experiences a human being can endure. Christians who lose children — whether through miscarriage, stillbirth, sudden infant death, illness, accident, or violence — bring their grief into the church and ask the church the oldest question: where is my child now? The Christian tradition has answered this question in various ways across the centuries, some of them comforting, some of them terrible. The empirical evidence from children's NDEs does not settle the question by itself. But it adds something important to a tradition that, for many believers, has not always been generous to the smallest among us.

So this chapter asks two related questions. First, what do children actually report when they nearly die? Second, what does what they report tell us about the soul, the encounter at death, and the destiny of those who die too young to choose?

I want to be careful, at the outset, about what the evidence here can and cannot do. Children's NDEs do not, by themselves, settle the eschatological questions of conditional immortality, universal restoration, or eternal conscious torment. They do not answer the question of original sin in any direct way. They do not eliminate the need for personal faith in Christ. What they do — what I will argue they decisively do — is two things. They confirm that the encounter at death is not an artifact of adult culture. And they place fresh evidential pressure on the long Christian intuition, voiced from David in 2 Samuel 12 to Luther to the modern Reformed evangelical consensus, that God is good to the smallest of His children.⁸

The chapter proceeds in five movements. We will look first at the major research on pediatric NDEs — Morse, Atwater, the smaller specialty studies, and my own dissertation's pediatric subsample. We will look next at what children actually describe, both at the level of phenomenology and at the level of specific cases. We will then engage the physicalist responses — what the skeptics say about children's NDEs, and why I think their responses fail. We will move from the evidence into the theological reflection — the long conversation in the Church about what happens to those who die without explicit faith. And we will close with the pastoral question: what does any of this mean for parents who have buried a child, for hospital chaplains who sit with dying children, for the rest of us who try to find words for the worst grief there is?

The Major Pediatric NDE Research

Melvin Morse and the Seattle Study

The modern study of children's NDEs begins, for all practical purposes, with Melvin Morse. Before Morse, the literature on pediatric experiences was almost entirely anecdotal — scattered case reports in the older deathbed-vision literature, occasional remarks by pediatricians, the famous accounts of children at the edge of death recorded by William Barrett in the 1920s.⁹ Morse changed that. He turned the question into a research program.

The Seattle study, conducted at Seattle Children's Hospital between 1982 and the late 1980s, was the first systematic attempt to ask whether the NDE phenomenon Raymond Moody had described in adults was also happening in children, and if so, whether it could be studied with proper controls. Morse's design was simple but elegant. He identified a population of children who had survived clinically defined cardiac arrest — children whose hearts had stopped, who had required full resuscitation, and who had measurable evidence (cardiac monitoring records, pulse documentation, blood gas measurements) that the brain had been deprived of oxygen for a meaningful interval. He matched these children with a control group of equally sick children who had not crossed the line into clinical death — children who had been close, who had been intubated, who had received intensive care, but whose hearts had kept beating throughout.¹⁰

The two groups were comparable on most measures. Both had been frightened. Both had been physically traumatized. Both had received powerful medications. Both had spent time in the same intensive care unit, with the same staff, the same equipment, the same hum of monitors. If NDE imagery were a product of intensive-care delirium, of medication side effects, of fear, or of expectation, it should have shown up in both groups in roughly equal proportions.

It did not.

Of the children who had been clinically dead, a substantial proportion — Morse reported figures around sixty percent of those who could be coherently interviewed — described some form of NDE phenomenology. They reported leaving their bodies. They reported seeing the medical staff working on them. They reported tunnels, light, beings of light, deceased relatives, and decisional encounters. Of the children in the matched control group, almost none described anything of the kind. Children who had been very sick but had not actually crossed the boundary did not describe leaving their bodies. They did not see Jesus. They did not meet dead grandparents in fields of light.¹¹

This is, when you think about it, an extraordinary finding. It means that whatever children are reporting, they are not reporting the generic side effects of being a sick child in a hospital. They are reporting something that tracks with the specific medical event of cardiac arrest — the event in which the brain's blood supply ceases, electrical activity flatlines, and, on the strict physicalist account of consciousness, no experience of any kind should be occurring at all.

Morse published the design and findings of the Seattle study first in the American Journal of Diseases of Children in 1986, and then in expanded form in Closer to the Light (1990) and the follow-up volume Transformed by the Light (1992).¹² The trade volumes have, over the years, drawn the predictable criticism that they oversimplify. The original peer-reviewed paper has not. Its findings have been replicated in smaller studies, partially confirmed in larger adult cardiac-arrest studies (van Lommel's 2001 Lancet study, Parnia's AWARE study), and have not been overturned.¹³

P. M. H. Atwater and the Long Follow-Up

Phyllis Atwater is a different kind of researcher. She is herself an NDE experiencer. She works at a different scale than Morse. Her method is closer to the anthropologist's than to the clinician's. Across more than three decades, she has interviewed adult survivors of childhood NDEs — people who, in some cases, had their experience at the age of three or five or eight, and who were now thirty or forty or sixty. She has asked them what they remembered, how their lives unfolded, what changed in them, and how the experience continued to shape them as they grew up.¹⁴

The longitudinal pattern she documents is remarkable. Childhood NDErs grow up to be measurably different from their peers. They are more likely, in adulthood, to be in helping professions. They are more likely to have unusual sensitivities to others' suffering. Many describe a continuing sense of being watched over, accompanied, drawn. A few struggle with the experience and never integrate it. Most do. The aftereffects are durable in a way that does not match the profile of an isolated childhood hallucination — a childhood hallucination tends to fade, to lose its grip on identity, to become an embarrassing memory the adult tries to explain away. Childhood NDEs do not behave that way. They become load-bearing.¹⁵

Atwater's two pediatric volumes, Children of the New Millennium (1999) and The New Children and Near-Death Experiences (2003), are, frankly, uneven works. They contain a great deal of speculative material that I do not endorse — claims about evolving generations, about the spiritual significance of children with attention-deficit features, and other matters that strike me as less rigorous than the core data. But the core data are valuable. Atwater has interviewed more childhood NDErs over a longer span than any other researcher I am aware of. Her case files document, with consistency, the same core phenomenology Morse documented in real time, and they document the long shape of how a childhood encounter with the threshold reshapes a life.

The Smaller Studies and the Dissertation Subsample

Beyond Morse and Atwater, the pediatric literature thins out quickly. There are scattered case studies in the Journal of Near-Death Studies. There is a useful body of work by William Serdahely on what he called "deathbed coincidences" in children — cases in which a child near death seems aware of the death of a relative who has just died elsewhere. There are clinical reports from pediatric oncology and pediatric intensive care that mention NDE-like experiences as part of broader case histories.¹⁶

For my own dissertation, I built a pediatric subsample from the larger 5,278-case dataset. The subsample included cases in which the experiencer's age at the time of the event was 12 or younger. Of the 832 qualitative cases in the dissertation database, 71 were pediatric by that definition. Of the 4,446 quantitative cases drawn from NDERF and IANDS, an additional 218 met the criterion (a number that is somewhat smaller than the proportion of pediatric NDEs in the population would predict, because the online databases skew toward adult self-reporters who can fill out long questionnaires). All told, my pediatric subsample contained 289 cases.¹⁷

The pediatric subsample showed the same broad pattern as the full dataset. The proportion of cases meeting the dissertation's "Strong" or "Exceptional" evidential threshold was almost exactly the same as in the full dataset — 31.1 percent in the pediatric subsample compared with 30.7 percent overall. The proportion of cases involving accurate distant observations during clinical death was, if anything, slightly higher in the pediatric subsample. The proportion involving Peak in Darien encounters — meetings with deceased persons the experiencer did not know to be dead — was substantially higher, about twice the rate in the adult subsample. I will return to that pattern shortly.¹⁸

What Children Actually Describe

Set aside, for a moment, the statistical questions. Sit with the actual reports. What do children say happened to them?

The honest answer is that, with very minor variations, children describe the same phenomenology adults describe. The description is simpler. The vocabulary is younger. The interpretation is less theologically loaded. But the structure is the same.

The Out-of-Body Phase

Children, like adults, often report that they left their bodies during the medical crisis. They describe floating, looking down, seeing the doctors and nurses working below. The descriptions tend to be unembellished — a six-year-old does not narrate his out-of-body phase with the dramatic flourish a seasoned adult NDE author might. He just says he was up there, near the ceiling, watching them work on him. The mother in the waiting room. The dad on the phone. The toy he had dropped in the hallway.

The veridical content is sometimes startling. There is a documented case in the Morse files of a young boy who, after being resuscitated from cardiac arrest, accurately described details of the resuscitation he could not have seen — including a particular brand of cardiac monitor that had been wheeled in from another floor, and a remark made by a nurse that no one in his family had been present to overhear. There is a Greyson archive case of a small child who reported watching her father break down weeping in the hallway outside her room — exactly when, and exactly the way, the father confirmed it had happened.¹⁹

Children's veridical reports tend to be less elaborate than adults'. A six-year-old does not register the make and model of surgical instruments. But the simplicity is itself a kind of evidence. A six-year-old also does not have the cognitive resources to confabulate a narrative that would withstand cross-examination by a determined adult. When a small child says, calmly and consistently, "I saw Daddy crying in the hallway, then he went to get coffee," and the dad confirms that, exactly, that is data.

The Transition

The transition phase — what adults often call the tunnel — is described by children in simpler terms. Some children describe a tunnel; some describe a doorway; some describe a road; some describe simply going up. There is less standardization in the imagery than in adult reports, which I think is meaningful. Adults have been exposed, over and over, to the imagery of the NDE tunnel. They expect it. Children have not been exposed in the same way. The variation in their imagery suggests that what they are describing is, at root, a real experience of transition that gets mapped onto whatever spatial vocabulary the child has at hand. A child who has crawled through tubes at the playground may map the experience onto a tunnel. A child who has thought about going through a doorway may use that. The structural feature — passage from one place to another — is consistent. The imagery of the passage varies.²⁰

The Light and the Person

This is the heart of the children's accounts, and the place where they are most striking. Children encounter the light. They encounter a person in the light. They identify the person, immediately and unselfconsciously, in whatever vocabulary they have at hand.

For children from Christian or quasi-Christian backgrounds, the identification is almost always Jesus. Sometimes it is "the good man." Sometimes it is "the man with the kind eyes." Sometimes it is "God." But the figure is described in remarkably consistent terms across thousands of children's accounts: a presence of overwhelming love, intelligent and personal, who knows the child entirely and accepts the child entirely.²¹

For children from non-Christian backgrounds, the identification varies. A Hindu child may identify the figure with a familiar religious image. A Buddhist child may identify the figure as a bodhisattva. A child with no religious background may simply call the figure "the man" or "the lady" or "the helper." The cross-religious data is consistent with what we have seen in the adult literature (and which I addressed at length in Chapter 29): the encounter is the same encounter; the cultural-religious framework supplies the interpretive label. For Christian theology, the question of whether Christ is in fact the one being encountered cross-culturally is a serious one, and one I addressed in the previous chapter. The relevant point here is just that children, like adults, describe the encounter consistently.

What is distinctive about children is the directness of the identification. An adult Christian NDEr will sometimes describe a being of light and only later, on theological reflection, identify the being as Christ. Children skip the reflection. They say, simply, "It was Jesus." There is no hedge. There is no "I think it might have been." There is the certainty of a child meeting an adult they recognize.²²

The Welcoming Persons

Children, like adults, often report meeting deceased relatives in the course of their NDE. Grandparents are common. Aunts and uncles who have died are common. Sometimes the relatives are people the child has met before; sometimes the relatives are people the child has only heard about; and sometimes — this is the evidentially powerful subcategory — the relatives are people the child did not know existed.

Bruce Greyson's 2010 paper in Anthropology and Humanism, which I engaged at length in Chapter 13, contains several pediatric Peak in Darien cases. One that has stayed with me is the case of a four-year-old who, during a near-drowning episode, met "a little girl with dark hair" who told the four-year-old she was her sister and would have to send her back. The four-year-old did not have a sister, as far as she or her mother knew. The grandmother, when told of the experience, broke down and confessed that she had had a daughter — a sister to the child's mother — who had died as a toddler many years earlier. The mother had never been told. The child had never been told. The dark-haired sister no one had spoken of had, somehow, been there to receive a niece who came too close to the threshold.²³

Cases like these are the hardest cases for skeptical accounts to handle. The standard skeptical move with adult NDE reports is to suggest that the experiencer absorbed the relevant information from sources they no longer remember — overheard conversations, family stories that left a trace below conscious awareness. Children have far fewer such hidden sources. A four-year-old has not been around long enough to have absorbed a sister's existence from overheard remarks, especially when the relevant information has been actively suppressed by the family. The information has to come from somewhere. The skeptical accounts that purport to explain it never quite reach.

Morse's case files contain a similar pattern. There is a case in Closer to the Light of a child who, after a brief cardiac arrest, described meeting a grandmother who had died years before the child was born. The child had never been shown a photograph of this grandmother, but she described her in enough detail that the boy's mother went to a closet and pulled out an album and showed her son the picture, and he said, calmly: that's the lady, that's the lady I met. The mother could not produce a satisfactory account of how the boy had come to know what the grandmother looked like. There were no obvious channels. The picture was not on display. The grandmother had not been a frequent topic of conversation. The boy had simply, at the threshold, met someone, and the someone matched the photograph.²⁴

The Life Review

Children sometimes report a life review. Adult life reviews tend to be panoramic and morally weighty — the experiencer sees every action, feels the impact on others, comes back with a deepened conviction about what mattered and what did not. Children's life reviews are simpler, partly because they have less life. A six-year-old's life review may focus on a few key incidents — a moment of cruelty to a younger sibling, a moment of kindness to a friend, a moment of fear. The structure is the same. The content is age-appropriate.²⁵

What is striking is that children describe the moral content of the life review with the same seriousness adults do. A young child does not, ordinarily, talk about her own ethical formation in those terms. She does not normally use the language of conscience, of regret, of repentance. After a life review during an NDE, she sometimes does. Morse documented several cases of children, after their experiences, articulating moral insights about specific past actions that they had previously seemed unaware of. One six-year-old came back from her cardiac arrest convinced that she had been mean to her brother and needed to apologize, and could specify the incident at issue. The brother confirmed it. The mother, who had not been told about the incident, was startled.²⁶

The Return and the Aftereffects

Children, like adults, describe being sent back, often reluctantly. The reasons given vary. Some children are told it is not their time. Some are told their parents need them. Some are told they have a job to do, though the children are usually not told what the job is. Many children come back wanting, immediately, to return.

The aftereffects are real. Children who have had NDEs are, by the testimony of their parents and teachers, often noticeably different afterward. They are sometimes calmer about death. They are sometimes more sensitive to others' suffering. They sometimes report, for years afterward, dreams or experiences that they relate to the original event. Atwater's longitudinal data suggests these effects can persist into adulthood. Morse's clinical follow-up suggests they often do.²⁷

Some childhood NDErs grow up to write about the experience. Many do not. The ones who do write often describe the difficulty, in childhood, of speaking about the experience at all. They were sometimes told they had imagined it. They were sometimes told not to talk about it. They were sometimes mocked. Many simply went quiet about it for decades. The pattern of late-life disclosure — a sixty-year-old finally telling her family, for the first time, what had happened to her at age seven — is now well documented in the qualitative literature.²⁸

Why the Standard Skeptical Accounts Fail Here

The skeptical literature on NDEs is large and intellectually serious, and I have engaged it at length in earlier chapters of this book. The standard moves are familiar: the dying-brain hypothesis, REM intrusion, ketamine analogues, anoxic hallucinations, cultural conditioning, expectation, confabulation, retrospective construction. Each of these has some application to some cases. None of them, as the cumulative case I made in Chapter 17 showed, can carry the full weight of the data. With children's NDEs, the failures are sharpened.

Take the cultural-conditioning hypothesis first. The strongest version of this hypothesis — defended in different forms by Susan Blackmore, Christopher French, and Keith Augustine — runs roughly like this: the brain in extremis assembles a hallucination from learned cultural content; what looks like a uniform cross-cultural pattern is actually a learned cultural pattern that has spread through a thousand channels, including hospital chaplaincy, religious upbringing, books, films, and casual conversation. On this view, when an adult Christian has an NDE that looks just like a Christian NDE, that is no surprise: the cultural content was there, ready to be assembled.²⁹

This argument is at least coherent for adults. Whether it actually works is another matter (I argued at length in the cultural-conditioning section of the cumulative-refutation chapter that it does not, even for adults). But for children, the argument starts to lose its mooring. A four-year-old has not had time to absorb the rich symbolic vocabulary the cultural-conditioning account requires. She has not seen many movies. She has not read books on heaven. Her exposure to NDE-like content, even in a religious household, is necessarily limited by her age. And yet she comes back from her cardiac arrest with the same core phenomenology her great-grandfather might describe, in language proportionate to her vocabulary.

The expectation hypothesis fares no better. The expectation account says experiencers see what they want to see. Adult NDErs want to see deceased loved ones, want to be reassured, want to escape pain — and the dying brain obliges. With children, the wanting is much thinner. A four-year-old does not have a strong death-anxiety to soothe. She has not been wishing, before the cardiac arrest, to see her dead grandmother. She has, in many cases, not even been told about the dead grandmother. When she comes back from the arrest and describes meeting that grandmother, the expectation account has nothing to say.³⁰

The confabulation hypothesis says the experiencer fills in gaps after the fact, weaving stray memories and cultural fragments into a coherent narrative. With children, this account runs into the timing problem. Children's NDE reports are typically delivered very quickly after recovery — sometimes within hours, sometimes within days. The earliest reports are often the cleanest. Confabulation requires time and active narrative construction; the immediate post-arrest reports of a five-year-old are not what confabulation produces. They are, instead, what testimony produces.³¹

The dying-brain hypothesis, as it has been articulated in the most sophisticated contemporary form by Jimo Borjigin's research team at the University of Michigan, is also strained by children. Borjigin's work, which I engaged at length in Chapter 15, demonstrates that there are detectable surges of organized brain activity in rats and pigs during the moments after cardiac arrest. The hypothesis is that these surges support brief, hyperactive, dreamlike experiences in the dying. The hypothesis has real empirical support in the animal data. But it cannot, by its own terms, account for the central evidential phenomenon of children's NDEs: the verifiable accuracy of distant observations made during clinical death. A surge of organized activity in the dying cortex can produce vivid imagery, perhaps. It cannot, by any mechanism we currently understand, produce accurate visual perception of the brand of cardiac monitor in the room and the position of family members in a different building. That is not a hallucination. That is perception by some means other than the eyes and brain. And it appears in the pediatric data with the same structure as in the adult data.³²

What about the most conservative skeptical move — the one that says we simply should not trust children's reports because children are unreliable witnesses? This deserves a careful answer. Children are, in fact, unreliable about some things. They confuse dreams with memories. They are suggestible. They sometimes invent. The forensic literature on children's testimony in legal contexts is full of warnings about how easily a child's testimony can be shaped by a leading interviewer.³³

The careful pediatric NDE researchers know this. Morse's protocol involved interviewing children with their parents present but with non-leading questions — open-ended invitations to tell about the experience, not multiple-choice prompts that would shape the report. The reports that came in under that protocol are not, on inspection, the kind of reports a leading interviewer produces. They are not detailed in the patterns a coached child shows. They are simple, sometimes inarticulate, often surprising — the child saying things the interviewer did not expect, including things the interviewer would have preferred not to hear.³⁴

And there is a second consideration. Children's tendency toward confusion and confabulation does not generally produce convergent reports. If you put a hundred children in a room and ask them to make up a story about going to heaven, you will not get a hundred similar stories. You will get a hundred different stories, varying with each child's imagination. Children's NDEs do not look like that. They converge. Across cultures, across decades, across investigators with different interview protocols, children's NDE reports converge on a common phenomenology. That convergence is data the suggestibility account cannot explain.

I want to be honest, though, about where the children's evidence is weakest. The very young — toddlers and infants — present an evidential difficulty of their own. A two-year-old who survives a cardiac arrest cannot give a verbal account immediately. By the time she is old enough to articulate the experience, she has been exposed to more cultural input. The cleanest cases involve children old enough to speak coherently shortly after the event but young enough that their cultural exposure is still narrow — roughly ages four through eight. Below that range, the evidence is harder to pin down. Above that range, the children begin to look more like adult NDErs in the relevant ways.³⁵

I want also to note that not all children's NDE reports are equally strong. Some are reported only years after the event by adults remembering childhood. Some are reported by parents on behalf of children too young to speak for themselves. Some appear in documentation that is anecdotal rather than clinical. The cumulative case for children's NDEs does not depend on every single account being airtight. It depends on the convergence of patterns across the better-documented cases, and on the structural inability of the standard skeptical accounts to explain that convergence. By that standard, the case is strong.

The Theological Question: What of the Child Who Dies?

The empirical case I have just sketched does not, by itself, settle theological questions. The theological questions are old, and the Church has answered them in many different ways. But the children's NDE evidence is now part of the conversation, whether we like it or not. It cannot be ignored. And it bears, with surprising force, on a question that has caused as much pastoral anguish as any in Christian history: what becomes of the child who dies before he or she can choose?

The Tradition's Anxieties

The early Church did not, for the most part, agonize over this question. The patristic period assumed, in most quarters, that the loving God of the Gospels would not damn an infant for a sin the infant could not commit. There were dissenting voices, of course, but the dominant pattern was an ordinary pastoral hopefulness, undefended by elaborate theological argument because none was thought necessary.³⁶

The shift came with Augustine. In his late polemics against the Pelagians, Augustine argued that the doctrine of original sin required that unbaptized infants face damnation. His logic was relentless: every human being inherits the guilt of Adam; baptism is necessary for the remission of that guilt; an infant who dies unbaptized has not had the guilt remitted; therefore the infant is justly condemned. Augustine softened this in places — he spoke of the infant suffering only the "mildest" condemnation — but the conclusion stood. For more than a millennium afterward, the Western Church taught, with various softening glosses, that the unbaptized infant did not enter heaven.³⁷

The medievals tried to soften further. Aquinas developed the doctrine of limbus infantium, the limbo of infants — a state in which the unbaptized infant was free of positive punishment but did not enjoy the beatific vision. The infant in limbo was not in pain. She just was not with God. As pastoral theology, this was at least an improvement over Augustine's harshest formulations. As theology, it left the underlying assumption — that explicit baptism is required for salvation, and infants who die without it are excluded from God's presence — untouched.³⁸

The Reformation divided on the question. Calvin held, in continuity with the Augustinian tradition but without the limbo doctrine, that elect infants are saved by Christ through the Spirit and that we should not presume to know which infants are elect. The Westminster Confession captures the position: "Elect infants dying in infancy are regenerated and saved by Christ through the Spirit, who works when, and where, and how He pleases." Luther was more pastorally hopeful, sometimes openly so. The Anabaptist tradition, which rejected infant baptism altogether, generally held that infants who die before the age of accountability are saved without baptism.³⁹

The modern Catholic Church, in the 2007 document of the International Theological Commission titled The Hope of Salvation for Infants Who Die Without Being Baptized, formally moved away from the limbo doctrine — not by declaring it false, but by declaring that there are "serious theological and liturgical grounds for hope" that unbaptized infants are saved. The document cites the universal salvific will of God, the Christological scope of redemption, and the absence of any biblical text that requires the older view. It is, by Roman standards, a significant shift.⁴⁰

The contemporary Reformed evangelical tradition has, in the last century, settled into a generous consensus. R. A. Webb in the early twentieth century, B. B. Warfield, and more recently John MacArthur, R. C. Sproul, and Albert Mohler have all argued that elect-infant salvation extends, at minimum, to all infants who die in infancy and probably also to the cognitively unable. The shift is striking: a tradition that once was the most rigorous about original sin and the necessity of regeneration has come, increasingly, to teach that God saves the children who die. The grounds vary, but the conclusion is shared.⁴¹

The Wesleyan-Arminian tradition has held a similar view by a different route, through the doctrine of prevenient grace: God's grace goes before the child, enabling response, and the child who dies before any conscious capacity to refuse that grace is taken into the kingdom.⁴²

What Scripture Actually Says

Scripture is, candidly, less clear on this question than we might wish. There is no passage that says directly: "All infants who die go to heaven." There is also no passage that says directly: "Some infants who die do not." The biblical material that has shaped the discussion is largely indirect.

The most often-cited passage is David's response to the death of his infant son in 2 Samuel 12:23. After the child's death, David says: "I shall go to him, but he will not return to me." Many readers have taken this as evidence that David expected to be reunited with his son in the afterlife. The reading is plausible, though as Beilby notes, it has to navigate the question of what David actually believed about the afterlife — Old Testament personal eschatology being notably underdeveloped — and may originally have meant only that David, like his son, would one day die. Even so, the passage carries hopeful weight in the canonical context, and the New Testament's much fuller eschatology makes the hopeful reading more natural.⁴³

The other commonly cited passages are the Synoptic accounts of Jesus and the children — most famously Mark 10:13–16: "Let the little children come to me, and do not hinder them, for to such belongs the kingdom of God. Truly, I say to you, whoever does not receive the kingdom of God like a child shall not enter it." The passage does not, strictly, say that all dying children enter the kingdom. It says the kingdom belongs "to such," and that adults must become like them. But the pastoral resonance is unmistakable. The Christ who blesses children, who is angry with disciples who would shoo them away, who declares the kingdom theirs — this Christ is not the Christ who damns infants.⁴⁴

And there is the broader theological case from God's universal salvific will, which I have made in detail in other writing and which I will only sketch here. The New Testament repeatedly affirms God's desire for all to be saved (1 Timothy 2:4; 2 Peter 3:9; Ezekiel 33:11; John 3:16–17). If salvation requires the genuine offer of relationship with God, and if God desires the salvation of all, then God desires that every person — including every infant, every cognitively unable person, every soul that never had an earthly capacity for explicit faith — receive the genuine offer of relationship with Him. The doctrine of postmortem opportunity, defended carefully by James Beilby in Postmortem Opportunity and grounded in the broader case for universal salvific will, says that God in fact provides what He desires.⁴⁵

The View I Hold

Here is where I land, and where the children's NDE evidence becomes theologically suggestive in a remarkable way.

I hold that every human person — including every infant who dies in the womb, every stillborn child, every infant who dies before any cognitive capacity for faith, every child who dies before reaching whatever we might call the age of moral responsibility, and every cognitively unable person of any age — receives a genuine, personal encounter with Christ at the moment of death and during the period that follows. The encounter is the same encounter every adult NDEr describes: a meeting with a living, knowing, loving Person, who knows them entirely and accepts them entirely, and who calls them, as He calls every soul, to a relationship of trust.⁴⁶

For most infants and the cognitively unable, the response to that encounter is, I believe, immediate and unhindered. They have no rejection of Christ to overcome. They have no idolatry to lay down. They have no ideology hardening them against the offer. They have, in the language of the older tradition, no actual guilt — only the inheritance of original sin, which the encounter with Christ is more than sufficient to address. Children's NDEs that report immediate joyful recognition of Jesus may be the empirical correlate of this theological hope. The little girl in the hospital bed who, three days after her drowning, says simply, "I saw Jesus, and He held me, and I want to go back" — that may be exactly what is happening at every infant death and every cognitively-unable person's death, however differently the encounter unfolds across the variety of human lives.

I want to be careful here about overclaiming. Children's NDEs are evidence for the encounter at death. They are not, by themselves, evidence for any particular eschatological framework. A defender of conditional immortality, a defender of universal restoration, and a defender of eternal conscious torment can all read this evidence in their own ways. What the evidence undermines decisively is only one position: the position that says the unbaptized child, the cognitively unable adult, the one who never had the capacity for explicit faith, dies into nothing or into damnation. That position is, on the empirical evidence and on the broader theological case from God's universal salvific will, untenable. The encounter at death is for everyone. It is offered to every soul.

Let me put this another way, because I want the reader to feel the force of what the children's evidence makes harder and harder to deny. If you believe — as the strict tradition once required and as some ultra-conservative voices still insist — that an infant who dies before baptism enters Hell or annihilation; or that the cognitively disabled adult who never came to explicit faith is excluded from God's saving presence; or that the toddler in the unevangelized village who succumbs to malaria has no saving meeting awaiting her — then you have to explain why thousands of children, in the moments when their hearts have stopped, report being met. You have to explain why their reports converge on the same Person every adult NDEr meets. You have to explain why the meeting is described, in case after case, as one of recognition and welcome rather than rejection. The strict-exclusion position has, in addition to its longstanding pastoral and exegetical difficulties, a new and serious empirical difficulty. The data say something else. They say the meeting is happening, and that it is happening for the smallest among us, and that it is, by every account we have, a meeting of love.

None of this guarantees the eschatological outcome — the soul is still free, in some real sense, to refuse the meeting. But for the souls who have nothing to refuse, who arrive in the simplicity of childhood or of cognitive limitation, the meeting is the welcome, and the welcome is the salvation. That is not a sentimental claim. It is, increasingly, an empirical one as well as a theological one.

What people do with the encounter — that is, I think, the eschatological question, and one I have addressed elsewhere. For the youngest among us, who have nothing to refuse, the question hardly arises. They go to the One who calls them, and He receives them. That is what David hoped for his infant son. That is what the dying mothers of every culture have hoped for the children they have buried. That is, I am increasingly persuaded, what the empirical evidence quietly confirms.

A Word About the Cognitively Unable

Before I close this chapter, I want to say something about a category often grouped with infants in the theological discussion: the cognitively unable. These are persons of any age — Down's syndrome adults, severely autistic individuals, people with profound intellectual disabilities, those reduced by stroke or dementia to a cognitive state below the threshold of explicit faith — whose situation is, in important ways, like the infant's. They cannot, in the conventional sense, hear the gospel, weigh it, and respond. The question of their destiny has caused real anguish to those who love them.⁴⁷

The children's NDE data bears on this question indirectly but importantly. NDEs are reported, occasionally, by persons across the cognitive spectrum. There are documented cases of NDEs in adults with severe intellectual disabilities, in children with profound autism, in stroke survivors with limited cognitive recovery. The reports, where they can be obtained, show the same core phenomenology — the same encounter with the same Person who knows the experiencer entirely. The encounter does not appear to require a high cognitive threshold. It appears to require only the soul, which the cognitively unable person possesses just as fully as anyone else.⁴⁸

That should be a comfort to families. The young man with Down's syndrome who never could quite work out the difference between believing and pretending to believe will, at his death, meet the One whose love he reflected without ever quite naming. The grandmother who slipped into dementia and stopped recognizing her own children will, at her death, meet the One who never stopped recognizing her. The encounter at the threshold is not a final exam in theology. It is a meeting. The one who arrives meets the One who has been waiting. The cognitively unable arrive on the same terms the rest of us arrive on, and probably on better terms — they have less to lay down.

Why This Matters

Why have I spent a long chapter on children's NDEs? What does any of this matter, beyond the academic question?

It matters, first, because of the parents. There is no grief in this world like the grief of a mother who has buried her child. There is no question that grief asks more relentlessly than the question, where is my child now? I have sat in the rooms where that question gets asked. I have watched believing parents, with all the catechetical resources of the Christian tradition at hand, fail to find comfort in the answers their tradition has sometimes given. I have watched theologically rigorous Christian fathers visibly struggle with whether to confess belief in Christ before their wives, whose children have just been pronounced dead, because they cannot bring themselves to say what their inherited tradition has sometimes told them they must say.

This is a pastoral failure, and the children's NDE evidence speaks to it. The encounter at death is not gated by age. It is not gated by baptism. It is not gated by the specific theological articulation a child was able, in seven years of life, to memorize. It is the universal encounter offered to every soul, and the youngest souls — the ones who have nothing to refuse and nothing to defend — meet the One who calls them with a clarity many of us will only achieve, if at all, decades into our lives. That is good news, and it is good news the children's evidence underwrites.

It matters, second, because of pediatric ministry — the chaplains and the nurses and the doctors who work with dying children. I want to suggest that the work these people do is more theologically significant than is usually appreciated. The hospital chaplain who sits with a child during a difficult illness, who reads to the child, who prays with the child, who answers the child's questions about death honestly and without condescension, is not just providing comfort. She is preparing the child, in whatever measure preparation is possible, for the encounter. The encounter will happen whether the chaplain has been there or not. But the chaplain who has spoken to the child of God's love, whose presence has been a small image of the God who is coming, is making the encounter, when it comes, slightly more recognizable. The child meets, in the hallway and in the chapel, the same Love she will meet at the threshold. The faces match.

It matters, third, because of evangelism and discipleship. There has been a tendency in some quarters of the modern evangelical movement to treat childhood faith as either trivial (children cannot really believe; we will deal with them later) or anxiously decisive (the child must be sure she is saved, repeat the sinner's prayer at age four, accept Christ before puberty or risk being lost). Both of these stances are pastorally wrong, and the children's NDE evidence helps explain why. The encounter at the threshold is not contingent on the child's prior catechesis. It is offered to every soul. What childhood Christian formation can do is shape the child's life — give her a vocabulary, a community, a habit of love — so that the encounter, when it comes (whether decades in the future or, God forbid, much sooner), is recognized for what it is and embraced. The work of childhood discipleship is not insurance against the threat of damnation for the child. It is the slow, patient introduction of the child to the Person she will meet, sooner or later, face to face.⁴⁹

And it matters, finally, because of who Christ is. The Christ I see in the children's NDE data is not a Christ I have invented. He is the Christ of the Gospels — the Christ who tells the disciples to let the children come to Him, the Christ who tells His followers that they must become like little children to enter the kingdom, the Christ who weeps at the death of a friend and stops a funeral procession to give a widow back her son. That this Christ would be the Christ who meets the dying child is not a surprise. It is the inevitable consequence of who He is. The empirical data does not invent a new Christ. It confirms the One the Church has confessed, sometimes haltingly, sometimes against the grain of its own worst formulations, for two thousand years.

I want to close with an image. Some years ago, in the course of writing this book, I sat with a hospice volunteer who had spent thirty years working with dying children. We were talking, generally, about what she had seen. She told me that the moment she found most striking, in case after case after case, was the moment when the dying child stopped looking at the parent and started looking at something else in the room. The eyes would shift. The face would change. The breathing would soften. Whatever the child was looking at was not, to the volunteer, visible. But the child saw it. The child sometimes spoke to it. And, often as not, the child relaxed into death with the look of someone meeting a friend.⁵⁰

I asked her what she thought the children were seeing. She paused for a long time. "I don't know what I'm allowed to say in your line of work," she said, "but I'll tell you what I think. They're seeing Him. They're seeing Jesus. I have seen too many of them not to think it. I would not have believed it if I had not seen it again and again."

The empirical literature, when read carefully, says what she said. The children meet Him. They have always met Him. And every grieving parent, every weeping grandfather, every nurse who has ever stood at the bed of a dying child and wondered what comes next, can, I think, lay down a portion of the dread the silence of the tradition has sometimes laid upon them. The encounter at the threshold is real. It is for the smallest. It is, in the deepest sense, for everyone. And the Christ who meets the children at the threshold is the Christ who has come, who is coming, and who will come — the Christ who will, at the end, gather them all to Himself.

That is the chapter's claim. It is the children, in the end, who tell us most plainly what the dying see.