Perfume smell aboard a Navy ship at the reported time of an aunt’s death
[name withheld], a caller from Manhattan Beach, California, describes a wartime experience aboard a Navy attack transport near Okinawa. The caller woke suddenly around 3 a.m. to a sweet floral powder smell that they recognized as their Aunt [name withheld]’s. They say they later learned by letter that she had died that day. They say that after a time conversion, her death matched the moment they woke. The story ends with the host’s comments on visits from the dying.
This clip comes from the source recording. AI chose where it starts and ends, so it may start or end a little early or late. The word-for-word transcript is not published. You can hear the source recording on the Internet Archive (file “Ghost to Ghost 1999.mp3”, starting at 2:20). Is this you, or someone you know? Ask for a correction or removal.
The story as told
[name withheld], calling from Manhattan Beach, California, tells Art Bell that they enlisted in the Navy at 17. They trained at a Navy radio school at Bedford Springs, Pennsylvania, then at Fort Pierce, Florida, and were assigned to an APA (attack transport) that went to the Pacific. The caller says the story begins on April 1, Easter Sunday, which they give as the day their unit landed troops on Okinawa. Their ship carried a second wave of troops ashore in landing craft. The crew hoisted the boats back aboard, and they went to bed close to midnight. They report waking suddenly about 3 a.m. with no noise or startle. They sat up in their bunk and smelled a sweet, floral powder. They say they could not normally tolerate the diesel-exhaust odor aboard the ship, but this time the diesel odor was gone and the sweet smell was strong. They say they first thought another sailor had spilled perfume. Then they recognized the smell as that of their favorite aunt, [name withheld], “back east.” They say they looked at their watch, it was around 3 a.m., and they fell back to sleep. They call that the end of the incident. About two weeks later, they say, they received a letter from their mother telling them Aunt [name withheld] had died on Easter Sunday, April 1, in the afternoon. They say they wrote back asking the exact time she died. Once the time conversion was made, they say, it was exactly the moment they woke and smelled the powder. The host then comments that such stories are common and that the dying may be near those they love.
Reported events, in order
- 0:32
[name withheld] says they enlisted in the Navy at 17 and, after boot camp and testing, were sent to radio school at Bedford Springs, Pennsylvania.
- 2:21
They say they were assigned to an amphibious group at Fort Pierce, Florida, and then to an APA that went to the Pacific.
- 3:10
They give April 1, Easter Sunday, as the day their unit landed troops on Okinawa.
- 3:39
Their ship carried a second wave ashore and hoisted the landing craft back aboard, and the crew went to bed close to midnight.
- 4:04
They report waking with no noise or startle around 3 a.m., sitting up in their bunk, and checking their watch.
- 4:33
They say they smelled a sweet, floral powder and that the diesel-exhaust odor was gone.
- 4:49
They first suspect a sailor spilled perfume, then recognize the smell as their favorite aunt’s.
- 5:36
They say they fell back to sleep and call that the end of the incident.
- 5:42
About two weeks later, they received a letter from their mother saying Aunt [name withheld] had died on Easter Sunday, April 1, in the afternoon.
- 6:22
They say they wrote asking the exact time of death and that, after time conversion, it was exactly the moment they woke and smelled the powder.
Claims
Each claim is labeled as something the person telling the story noticed (observation), what they made of it (interpretation), something others told them (hearsay), or something the host said (host statement).
- observation· 4:33
The caller says they woke about 3 a.m. aboard ship to a sweet floral smell and that the diesel-exhaust odor was absent.
- observation· 5:10
The caller says they recognized the smell as that of their Aunt [name withheld].
- interpretation· 5:31
The caller says they concluded the smell was connected to their aunt at the time of her death.
- hearsay· 5:50
The caller says a letter from their mother reported that Aunt [name withheld] died on Easter Sunday, April 1, in the afternoon.
- interpretation· 6:22
The caller says that after they got the exact time of death and converted time zones, it matched the moment they woke.
- host statement· 6:43
The host asserts that the dying may visit or be near those they love.
Scores, with reasons
An AI model gave these scores using a fixed set of scoring rules (scoring-v1). They are editorial judgments and rough guides, not the odds that anything paranormal happened. How scoring works.
A coherent firsthand story with concrete detail (unit history, date, ship routine, time on a watch, a named relative). The caller offers an ordinary explanation before rejecting it. The link to the time of death rests on a secondhand letter and a time conversion.
Evidence for and against
- Firsthand witness of the smell and of waking at about 3 a.m.
- Concrete sequence and sensory detail, including the absence of the diesel odor.
- The caller separates what they observed from what they concluded and says they wrote to their mother for the exact time.
- The key link depends on a time of death supplied secondhand by the caller’s mother.
- A single-sense event with no other witness and no backing detail beyond the smell.
- The caller’s own ordinary explanation (spilled perfume) is left unruled out rather than tested.
- The caller’s year and day (“1944,” “Easter Sunday”) cannot be checked from the transcript.
- Some lines carrying the caller’s words are labeled as the host, and the reverse, so who said a few sentences is uncertain.
- Only the smell is reported; nothing seen, heard, or felt.
- 4:08
- 4:43
- 6:22
What the model inferred (not said on air): Component judgment (21+15+15+10+10=71, rounded down conservatively to 68): I credited the caller for raising and then discarding an ordinary explanation and for seeking the time of death. I deducted points for the single-sense report and the reliance on a secondhand time.
The story rests on the caller’s own word. They give checkable specifics (a named aunt, a letter from their mother, dates), but none is produced or verified on air.
Evidence for and against
- A named relative and named places and dates that could in principle be checked.
- A claimed document, the mother’s letter.
- The letter is not read or examined on air.
- No independent witness speaks on air; the death time comes through the caller’s mother.
- The letter is not read out or examined on air.
- The caller reports the exact time of death and the conversion, but does not show them.
- No second witness speaks.
- 5:42
- 6:22
What the model inferred (not said on air): Placed in the 11–25 range because checkable specifics are given but nothing was checked. I did not raise it to the 26–40 range because the letter is not offered or examined as evidence on air.
A gentle, comforting story with no threat, intrusion, physical contact, or escalation. The only sad element is the death of a relative, reported afterward.
Evidence for and against
- The smell is described as sweet and familiar, tied to a beloved aunt.
- No fear, harm, threat, or menacing presence reported.
- The incident ends and the caller falls back to sleep.
- Listeners may find the death-timing coincidence unsettling rather than frightening.
- 4:33
- 5:50
What the model inferred (not said on air): My judgment of the effect on listeners; the caller expresses no fear at any point.
A familiar story of a sign at the time of a death, with one distinctive element: the sign is a remembered smell rather than a figure seen, and it happens on a naval vessel with a later time-zone comparison.
Evidence for and against
- A sign at the time of an unknown death is a standard motif.
- A smell-only sign with nothing seen, aboard an APA near Okinawa.
- No unusual being, place, or event beyond the smell.
- The structure (wake, smell, later death notice, time match) follows a common pattern.
- The comparison with the rest of the archive is provisional; no comparison set is provided here.
- Similar stories of a sign at a death are common in the archive, but the smell-only detail is not.
- 4:33
- 5:21
What the model inferred (not said on air): The first-pass originality score is a judgment that the pattern (a sign at the time of a relative’s death) is familiar, while the smell-only, at-sea detail is distinctive.
Scoring confidence 0.65 · analysis confidence 0.78 · checked against outside sources: no · evidence available: caller’s firsthand story about the smell and waking; claimed letter from the caller’s mother reporting the death (not produced or examined on air); claimed time of death and time conversion, reported by the caller. The originality score blends the model's score (52) with how rare the story's details are across the archive. This method is still provisional.
Ordinary explanations to consider
These are possibilities only. None is a finding about what happened.
- The smell was an ordinary shipboard event, such as another sailor’s spilled perfume or toiletries. The caller raises this and does not rule it out. · The caller says: [quote withheld]
- A smell experienced on waking from deep sleep, possibly a memory of a familiar smell, on a night of extreme tiredness after combat operations. · The caller says: [quote withheld] They also say: [quote withheld]
- The exact-time match depends on a time of death reported secondhand by the caller’s mother and on the caller’s own time conversion. An error or approximation in either could explain the apparent coincidence. · The caller says: [quote withheld]
What stands out
This crisis-type story involves only smell: no figure, sound, or object, just a remembered powder scent at the moment of a relative’s death, aboard a naval vessel at sea. The caller says they later got the exact time of death from their mother and converted time zones to compare it with the moment they woke.
Related stories
Matched by shared recurring details, categories, and wording.