Oddity Journal
← Archive

Scotland · sea and archives · Luoghi

Flannan Isles — three men missing

On 26 December 1900, on Eilean Mòr in the Flannan Isles, a relief keeper found an empty lighthouse. Three men had disappeared. The records do not tell of a curse: they tell of a closed door, a damaged landing and eleven days of silence.

A light that did not answer

On 26 December 1900, Joseph Moore arrived on Eilean Mòr, the largest of the Flannan Isles, to change the lighthouse shift. From the ship Hesperus he tried to call the station with a whistle and a rocket. No one answered. When he landed, the outer doors were closed. The fire was out, the beds were empty and the lamp was ready to work. James Ducat, Thomas Marshall and Donald McArthur were gone.

That empty house generated an almost perfect story: an abandoned meal, an overturned chair, diary pages in which three men supposedly wrote about a storm and a presence. Contemporary sources do not deliver that scene. Moore’s letter and inspector Robert Muirhead’s report instead describe an orderly station, clean pots and stopped clocks. A stopped clock is a real object; it is not a stopwatch for a disappearance.

The lighthouse had entered service only a year earlier. The three men did not live in an ancient ruin cut off from the world, but in a workplace with shifts, equipment and landings to check. That difference changes the question: not what imaginary force emptied the lighthouse, but what work took them outside.

Source note

The station description comes from 1900–1901 documents transcribed by the Northern Lighthouse Board; the meal and chair belong to later tradition.

The darkness of 15 December

A temporary board, used for the log, contained measurements and notes from the morning of 15 December. From those notes Muirhead inferred that the morning work was complete and suggested that the three men went out in the afternoon. That evening the steamer Archtor did not see the lighthouse light and reported it on reaching port. The news did not reach the Board in time to start a search.

Here too, the document must not be loaded with more than it can support. On earlier nights the light had not been visible from the coast even with keepers in place: fog and distance could hide it. The absence observed by the Archtor is a credible alarm, not direct proof of the hour or cause of the disappearance.

Eleven days passed between the night of the 15th and the Hesperus’s arrival. That interval is what leaves the file short of answers. The sea can move equipment, erase traces and change the meaning of an object. We cannot know whether an immediate rescue would have saved anyone; we can say that by the time Moore arrived, the island may have shown only the consequences.

Source note

Muirhead’s afternoon is an inference, not a moment seen by a witness.

The west landing

The first entry does not solve the case. The decisive part of the inspection is the west landing. Moore recorded a missing equipment crate, damaged railings and a lifebuoy torn from its lashings. At the east landing, by contrast, the ropes were still in place. Muirhead also found water marks and damage along that exposed coast.

Moore observed that the clothes left at the station suggested Ducat and Marshall had gone out in work gear; McArthur’s coat remained inside. That detail drives many accounts toward a sudden rush. But the coat does not say why McArthur went out, when he went, or whether he reached the others. It says only that the explanation must account for a man less prepared for bad weather.

A frequently repeated figure, 110 feet, created a false image: a thirty-metre wave swallowing the lighthouse. The report does not measure a free-standing wave; it records where objects and water marks were observed on a rocky coast. Water can run up and concentrate in a landing. The figure demonstrates high damage; it does not restore a film of the storm.

Source note

The crate, lifebuoy, railings and coat are inspection observations. The men’s route is an inference.

The inspector’s conclusion

In his report of 8 January 1901, Muirhead proposed the explanation he considered most likely: the keepers had gone down to the west landing to secure the equipment crate, and a large wave had swept them into the sea. It is an investigative conclusion built on damage, not a witness account of the final minutes.

Its strength is parsimony. It needs no unknown ship, unrecorded fight or unseen attacker. It explains why the event may have centred on the place where the damage remained. Its limit is just as clear: it does not explain the order in which the men went out or prove that all three were together at the water’s edge.

Former keeper Walter Aldebert later proposed a more dramatic sequence, with two waves and a return to the lighthouse to call McArthur. He knew the coast and had experience of the local sea, but his reconstruction is late; historian Mike Dash did not locate the photographs and report on which Aldebert said he based it. It may be consistent with the coat, but it is not a recovered chronicle.

Source note

Muirhead’s wave theory is the one recorded in the report; the two-wave sequence is a later, unverified reconstruction.

The mystery that remains

The station’s original log is not among the papers deposited at the National Records of Scotland. Its absence does not prove that someone removed it, and it does not authenticate sentences attributed to the keepers. It establishes a simple limit: no one can quote those pages as if they were still available for inspection.

Escape, violence, a clandestine ship and the supernatural remain abstract possibilities. In the selected documents they have no object, route, testimony or motive comparable to the damage at the west landing. Saying they are unproven does not declare them impossible; it means refusing to turn an archival gap into evidence.

The disappearance therefore retains an opaque area, but it is not a shapeless enigma. Moore found an empty station. Muirhead found damage and water marks at the west landing and interpreted them as the trace of a wave. Between those observations, the accident explanation remains the best supported. What is missing is the human sequence: who went down first, who saw the danger and who tried to help.

Source note

The legends have cultural value; they do not close the historical file.

LEADS

Explanations under review

A workplace accident

This is Muirhead’s hypothesis. Damage and missing equipment at the west landing make it the explanation best anchored in the records. It does not establish the order of events.

Working hypothesis

Two waves

Aldebert proposed this many years later to explain McArthur’s coat. The available sources preserve neither his supporting material nor a 1900 testimony.

Working hypothesis

Escape, conflict or the supernatural

These are possibilities only in the abstract. The sources consulted offer no equivalent trace: no evidence of a ship, violence or a mysterious force.

Working hypothesis

What remains

The Flannan Isles do not ask us to choose between an ordinary sea and a fantastic story. They ask us to look at the difference between observed damage and a scene invented to connect it to three lives.

The closed door, torn lifebuoy and missing crate do not tell us everything. They tell us enough, however, to return the story to where it began: a working landing, facing a sea that no one saw at the decisive moment.

SOURCES

References

Documents, studies and archives cited in this article:

Read the source list →