Where people have come back from the edge of death, and what they say they saw. Tap a bubble, then Walk this experience.
Where the experience happened, or as near as the source allows. Er in Pamphylia, not Athens where Plato wrote. Drythelm at Melrose, not the disputed Incuneningum. Tundale at Cork where he collapsed, not Cashel where he was from.
Basemap © OpenFreeMap, data © OpenStreetMap contributors.
26 pins. 472 accounts. Some pins are collections, so the two counts tell different stories, and the first two charts show both.
Among the published and historical pins, the tunnel and the light is the rarest category by pins: 2 of 26, against 8 each for guided tours and journeys. By accounts it is the commonest by a long way: 252 of 472.
Both numbers are right. Moody's 150 cases and Ring's 102 are both filed under the tunnel, and between them they outweigh every historical account on this map combined. That is a fact about who got collected, not about what happens.
The six categories, as the bubbles are coloured.
The same six, weighted by how many experiences each pin covers.
Named by the account, inferred from its setting, or never recorded.
As each source gives it, including the ones that are not a cause.
The deep past is thin and the gaps are long.
The third bar is the size of what is not known.
Russia. Not for want of people who survived. Vladimir Negovsky founded resuscitation medicine in Moscow and revived more people, earlier, than anyone, and he did collect what they reported: he published it in 1984 as a neurophysiological analysis of "hallucinations" experienced by post-resuscitation patients, the quotation marks his own. It went into a resuscitation journal rather than a near-death one, filed under neurophysiology, which is part of why nothing from the Soviet period reads as a near-death account today. His entry is in the sources; it is not on the map because a career is not one person's experience.
The Islamic world and Jewish tradition. Both traditions' teaching about what comes next is on the Beyond page. What is missing is a first-person account from either that this atlas can quote or properly cite.
Africa and the Americas. Nothing citable found yet. Kellehear's census of non-Western accounts reaches neither in a form that gives figures.
It shapes the content. Kellehear found the life review largely absent from hunter-gatherer accounts, and cites Bellah on conscience belonging to religions that put moral weight on the individual. Pasricha's north Indian series: 75% met a being of light. Her 1993 series: none.
It does not. Long argues from children under five, and from 500 non-English accounts, that content is "strikingly similar" regardless. Van Lommel's paper reports, citing earlier work, that experiencers do not differ from controls by religion or degree of religious belief.
Kellehear, JNDS 2008;26:249-65 · Pasricha, JNDS 2008;26:267-82 · Long & Perry 2010 · van Lommel, Lancet 2001;358:2039-45. Knoblauch's sharpest split, secular east against west Germany, is on Patterns.
Every bubble is one experience, or one collection of them, coloured by the shape it took. The label under it is the religion of the person reporting it. Walk any bubble to the end and you reach a fork: back, or on into what has been proposed.
Each account got the category that fits its own source, not the one tradition assigns it. Plato's Er is not summoned by anyone: he is killed in battle and travels with a crowd, so he sits under a journey.
Prevalence studies are not on this map. They measure many people at once and have no single shape to place or colour. They are in the chart below, on Patterns and in Sources.
Archive accounts coded here are on the map too. NDERF and IANDS accounts this atlas has read are placed in the same categories as everything else, at the centre of the country (or US state) the account gives, never closer. Accounts from one place with the same kind, cause, religion and decade share a bubble, with the count on it. Their walk plays each stage as strongly as the share of those accounts that mention it. Only the codes are kept here: what they report.
The timeline runs from 2100 BC to the 2020s, stretched so the crowded last century stays readable. Play the atlas taps through everything like a reel.
Read the denominators, not the heights. 64% is seven of eleven children; 4% is 82 of 2,044 Germans. Brackets are 95% confidence intervals.