Hafsteinn Björnsson (medium)

Hafsteinn Björnsson (1914–1977) was one of the greatest trance mediums of the twentieth century and the most important one from Iceland. Hafsteinn’s mediumship, which included drop-in communicators and xenoglossy, was investigated in the 1970s by Icelandic parapsychologist Erlendur Haraldsson in Rekjavik and at the American Society for Psychical Research in New York.

The Mediumship of Hafsteinn Björnsson

Psychical researchers recognize two forms of mediumship, the primarily physical and the primarily mental. Indridi Indridason was the most significant physical medium Iceland has produced. Hafsteinn Björnsson was the most important mental medium from that country. He was a trance medium, in contrast to the psychic mediums with whom most contemporary mental mediumship research has been conducted.

In Iceland there are no proper surnames, only patronymics. Hafsteinn was the son of Björn and hence Björnsson. He was born 30 October 1914 in Skagafjordur, northern Iceland. His psychic powers began to manifest in early childhood and remained strong throughout his life. As an adult, he was employed as a bill collector for the National Radio in Reykjavik. The radio’s general manager encouraged the development of his mediumship, although his principal mentor was Einar H Kvaran, who previously had trained Indridi Indridason.1

Icelandic parapsychologist Erlendur Haraldsson learned about Hafsteinn from a book by Elinborg Larusdottir, published in 1970.2 Erlendur contacted Hafsteinn and found that he was receptive to having his mediumship studied. Ian Stevenson, who was interested in drop-in communicators, xenoglossy (unlearned language), and other facets of Hafsteinn’s mediumship, participated in some of Erlendur’s investigations.

Hafsteinn was polite, punctual and orderly, making him very popular. After work at the radio station, he would walk to the nearby Iceland Society for Psychical Research, where he held two sittings each evening, and most nights did not return home until almost 10 pm. The séance room was illuminated by a soft red light. A prayer was said and hymns were sung at the start of each sitting, then there was a sharp blow as Hafsteinn expelled air from his lungs and went into trance. The voices of his controls and communicators sounded different from the way he normally spoke. A steady stream of communicators appeared evening after evening, usually addressing particular sitters and giving their full names.

Hafsteinn used his holidays to go to places far from Reykjavik, where he would conduct up to five sittings a day. He would also lead large events, which he called clairvoyant meetings. These meetings were held in community halls in the countryside and large cinema houses in Reykjavik. They differed from his séances in that Hafsteinn would be in half-trance, as he called it; he knew what was going on round him, yet at the same time could sense discarnate spirits in his presence. Typically about 150 communicators appeared at the meetings. Hafsteinn would start by describing certain persons and their close relatives, acquaintances or co-workers. He would then call out into the hall, usually to a specific area, and ask whether anybody recognized the description; usually someone did. In March and April 1972, Erlendur recorded 53 séances and several clairvoyant meetings with Hafsteinn.3

Erlendur was with Hafsteinn one Easter when he travelled to the east of Iceland. Their flight was delayed due to inclement weather. As they awaited its departure, Hafsteinn told Erlendur that spirits from the east coast were already gathered round him in anticipation of the meetings. At their destination, Hafsteinn hosted five meetings a day. He did not have anything to do with the bookings, and did not know who would come to each meeting. Erlendur noticed that Hafsteinn avoided contact with attendees before and after the events. In the short interval between meetings, he only wanted to drink coffee and relax. Some critics maintained that Hafsteinn gathered information about sitters prior to his séances and clairvoyant meetings, but Erlendur saw no evidence of this.4

Hafsteinn was married and lived for many years with his wife Thordis, but they divorced when both were middle-aged. Hafsteinn remarried and lived with his new wife until he died suddenly of a heart attack in 1977. When Thordis died in December 1985, Erlendur and his wife Margret attended her funeral, as they had come to know her well. There something extraordinary happened. The church suddenly filled with a peculiar atmosphere that always existed at Hafsteinn’s séances. It was if Hafsteinn and his spiritual entourage had appeared with an energy that filled the church. Their arrival was accompanied by a sound as if from afar, but this faded out after a while. When Erlendur and Margret left the church, he asked her whether she had sensed anything unusual. She described the same thing he had experienced. This event was Erlendur’s last connection with Hafsteinn and his circle.5

The following are the more significant séances, meetings and tests with Hafsteinn documented by Erlendur Haraldsson.

Runki Runólfsson

Runólfur (Runki) Runólfsson was a drop-in communicator (a communicator not known to the medium or sitters) at Hafsteinn’s séances beginning in the late 1930s. Elinborg Larusdottir told his story in her 1970 book. Erlendur and Ian Stevenson made additional enquiries and wrote a journal paper about it.6 It is also the subject of a documentary available for viewing on YouTube. Because this case is treated elsewhere in the Psi Encyclopedia, it will only be summarized here.

In the autumn of 1937 a strange communicator began to manifest at the regular séances Hafsteinn held in the home of Einar Kvaran in Reykjavik. The communicator refused to give his name; when asked why he was there, he explained cryptically, ‘I am looking for my leg. I want my leg’. Pressed on where he thought his leg was, he said, ‘It is in the sea’. This made no sense to any of the circle until, in January 1939, a new sitter joined the group. The mystery communicator accused this new sitter of having his leg in his house. The man did not know what the communicator was talking about and he was told firmly that the circle could be of no use to him unless he revealed who he was.

The communicator then vanished for several weeks. When he reappeared, he identified himself as Runolfur Runolfsson, whose nickname was Runki, and told his story. After a night of drinking, he had been making his way home alone along the beach. He had sat down on a rock to rest and passed out. When the tide came in, it carried his body out to sea. Most of his body later washed ashore, was found and buried, but without his leg. When his leg later washed up, it was not associated with him; it was placed in the wall of the new sitter’s house, which was then under construction. The sitters was asked how his identity might be confirmed and he directed them to a certain parish, where indeed records of the late Runolfur Runolfsson were found. The new sitter had walls of his house opened and discovered a femur behind them. The bone was ceremoniously buried, to Runki’s pleasure. His gruff demeanor disappeared and he went on to become Hafsteinn’s chief control (discarnate master of ceremonies).

Gudni Magnusson

Gudni Magnusson was another drop-in communicator described by Elinborg Larusdottir and followed up by Erlendur Haraldsson and Ian Stevenson.7 This case has its own entry in the Psi Encyclopedia and will only be summarized here.

Unlike Runki, Gudni appeared at a single séance, in January 1941. Most of the information about him was conveyed by Finna, Hafsteinn’s control on this occasion. Finna said that there was someone by the name Gudmundur or Gudni Magnusson, who wanted to contact Hjalmar Gudjonsson, one of the sitters, because he was an acquaintance of his relatives. Finna said she saw Gudni as a young man of between twenty and thirty years of age, blond, with hair thinning on the top of his head. He had been a car or lorry driver who had died in a place associated with Eskifjordur and Reydarfjordur. His vehicle had broken down and he had crawled beneath it in an attempt to repair it. In so doing, he had ruptured something inside his body and although he had managed to reach his home, he had to be transported to hospital by boat. Unfortunately, he had died in the boat before reaching the hospital.

None of this meant anything to Hjalmar Gudjonsson, but enquiries in Eskifjordur determined that indeed there had been a young man named Gudmundur (Gudni) Magnussen, a blond with hair thinning on the top of his head, who had died at 24 after his lorry had run out of petrol on a mountain pass, necessitating a four-mile hike to and from a filling station. It was thought that this strenuous activity had caused an old surgical repair to his intestines to split. Although he reached his home, Gudni began to complain of severe pain in his stomach. He had to be transported by boat to hospital (it being by that time dark, so that flights were impossible), but did not make it there.

Erlendur tracked down all witnesses still living and obtained some additional contemporary documents which confirmed the course of events but added no new information to the story. The connection between Gudni and Hjalmar’s relatives was never ascertained, although inasmuch they lived in the same part of Iceland, they might well have met. Moreover, it is not known whether Gudni had crawled under his truck before he began his hike or whether he injured himself in walking. However, other details of Finna’s account were confirmed.

Greenlandic Inuit Xenoglossy

A Danish man, Svend Fredriksen, sat with Hafsteinn on a visit to Iceland in 1966. Fredriksen had grown up in Sisimiut, the second largest town on the west coast of Greenland. Among his playmates were Inuit (Eskimo) children, whose language he learned to speak. Frediksen later moved to Denmark and worked in Copenhagen, where he became acquainted with the Icelandic writer Jon Björnsson, who lived there from 1932 to 1945. Still later Svend Fredriksen joined the faculty of the Catholic University in Washington, DC, as a professor of Inuit languages and anthropology. On his way back to Copenhagen from Washington in September 1966, Fredriksen stopped in Reykjavik to visit Jon Björnsson and it was on this occasion that he attended the sitting with Hafsteinn. He went without Jon but returned home  – Jon told Erlendur – very excited and happy. Among the many people from the other side that came to him was a shaman he had known in Greenland as a boy. The shaman spoke to him in Inuit and they conversed briefly in that language.

Erlendur tried to meet Svend Fredriksen in Washington in the 1970s, but by then he had passed away. Erlendur was able to interview Jon Björnsson and two sitters who had been present at the séance, however, and their testimony confirmed the account given by Elinborg Larusdottir in her book. Elinborg reproduced remarks written out immediately after the séance. The sitter said that because Fredriksen did not speak Icelandic, another sitter, who spoke Danish, interpreted for him. Soon a large group of communicators began to address him directly in Danish and then Runki, acting as control, began to describe a group of Greenlanders who showed themselves in a strange sort of clothing, along with dogsleds and kayaks. They mentioned names ‘that are not understandable to us and make Fredriksen almost elevate from his seat in astonishment. He knew these people and lived daily with them in his youth before they moved from our realm of existence’. Then Runki gave way to a direct-voice communicator. ‘A joyous reunion takes place, and we become witnesses to that from the lips of the medium sounds the unique language of the Greenlandic Eskimos that no Icelander understands. Prof. Fredriksen responds in the same language and they exchange a few sentences’.8

Crew of the Trawler July

Erlendur was able to arrange two sittings for which he chose the sitters. The first of these was in October 1972. He selected five people he knew well, including Sveinn Freyr Rognvaldsson and his wife Alma Andresdottir. At the séance, a group of communicators addressed Sveinn. The four identified themselves as men who had died on the Newfoundland fishing banks, giving their full names. Indeed, these men were among the crew of the trawler July, which had sunk off Newfoundland on 12 February 1959 with thirty men aboard. Sveinn knew them from earlier tours on another boat from the same firm. In fact, he had been scheduled to accompany them, but had backed out after strong opposition from Alma, who had had a premonition that he would not return alive if he sailed that day.

Sveinn had never met Hafsteinn, who had no reason to know who he was. Hafsteinn could have heard about the tragedy of the July, because that was well-known in Iceland. Hafsteinn might have read the names of the crew members in newspapers of the time. But why, Erlendur asks, did these men from the trawler crew address Sveinn and not another sitter that day, or another day? Sveinn was the only sitter with connections to the crew of the July. After writing a draft of his account, Erlendur rang Sveinn, who still remembered the incident and described it to him again. Sveinn read Erlendur’s final text and approved it. Erlendur deposited a sound recording of this séance with the Icelandic National Library.9

Experiments in the 1970s

In the early 1970s, Erlendur conducted several experiments in which he had sitters try to identify the readings intended for them. These experiments were done at the American Society for Psychical Research in New York and in Reykjavik. Reports were published in the Journal of the American Society for Psychical Research.10

The first experiment was the most successful. Erlendur invited Hafsteinn to come to New York and with the help with the Icelandic consulate recruited ten Icelandic citizens residing in that city to act as participants in a clairvoyance meeting format. In the experimental room, thick curtains separated Hafsteinn from the participants, who entered one at a time and sat in a chair wearing earplugs and headphones with music playing loudly while Hafsteinn gave his readings. Erlendur chose the participants, but their order of entry was random, determined by Stevenson, who did not know their identities. Hafsteinn’s readings were recorded and transcribed and all ten readings were given to each of the participants, who tried to identify the one intended for them. Four of the ten were successful, not a great number, but nevertheless more than would be expected by chance.

After the experiment was completed, it occurred to Erlendur that all of the participants were young, so he asked them to give the readings to their parents to score. It turned out that the parents performed much better than their children, indicating that Hafsteinn had been more successful than the formal test made it seem.

Erlendur conducted another ten experiments with Hafsteinn using a similar methodology, some in Reykjavik and in some in New York, but in none of them were the participants consistently able to select the readings Hafsteinn did for them. Erlendur began to think there might be another factor at play. He noticed that participants often recognized persons mentioned in readings meant for other participants. He wondered if Hafsteinn might be thrown off by having too many participants in one room; communicators for various participants in the experiment might be present at the same time, confusing him. To investigate this possibility, Erlendur set up an experiment in which each participant scored twenty readings, ten of them from their experiment and ten from an earlier experiment. The results showed that even when participants did not identify the reading intended for them, they did select readings from their experiment rather than from the other one. The statistical odds of this being due to chance was one in a thousand.11

Memorable Communicators

In Hafsteinn’s séances, there was usually a control, such as Finna or Runki, who conveyed messages to the sitters, but there were also communicators who spoke directly through Hafsteinn. This occurred several times at each séance. When communicators spoke through him directly, Hafsteinn would grasp the hand of the sitter that the communicator wished to contact. Once he did this with Erlendur, although Erlendur was present as an observer, recording the sittings, and did not otherwise take part in them. Erlendur did not immediately understand who the communicator was, but then he heard a firm voice say, ‘What is this, man, don’t you know who I am? This is Stefan. Don’t you remember me in the swimming pool?’ At this, Erlendur recalled a patient who had stayed for a long time at a sanatorium at which he had been employed as a masseuse.

This patient’s name was Stefan. He had a stiff hip but moved quite swiftly, was very strong, and had a lively personality. From where he worked, Erlendur could see into the swimming pool, and Stefan was always there; typically he was the first in the pool in the morning. When he greeted people, Stefan shook them firmly by the hand, and now Hafsteinn, in trance, was shaking Erlendur’s hand in the same way. Then he added, leaving Erlendur in no doubt about his identity: ‘They liked you pinching them, the ladies.’ This was a characteristic way of talking for Stefan. The incident was memorable for Erlendur, because Hafsteinn would have known nothing about Stefan or that Erlendur knew him.

On another occasion, a communicator was addressing a sitter from Gardur, in the south-west of Iceland. Suddenly Hafsteinn turned to Erlendur and the communicator announced, ‘I knew this man when he was young.’ Indeed, Erlendur knew the communicator, a man who had been married to his father’s sister. As a boy, Erlendur had sometimes spent the summer with this family. He could not imagine, however, how Hafsteinn could have had the faintest idea of this.12

James G Matlock


Haraldsson, E. (2021). Towards the Unknown: Memoir of a Psychical Researcher. Hove, UK: White Crow Books.

Haraldsson, E. (n.d.). The medium Hafsteinn Björnsson. [Unpublished manuscript.]

Haraldsson, E., Pratt, J.G., & Kristjansson, M. (1978). Further experiments with the Icelandic medium Hafsteinn Bjornssön. Journal of the American Society for Psychical Research 72, 339-47.

Haraldsson, E., & Stevenson, I. (1974). An experiment with the Icelandic medium Hafsteinn Bjornssön. Journal of the American Society for Psychical Research 68, 192-202.

Haraldsson, E., & Stevenson, I. (1975a). A communicator of the “drop-in” type in Iceland: The case of Gudni Magnusson. Journal of the American Society for Psychical Research 69, 245-61.

Haraldsson, E., & Stevenson, I. (1975b). A communicator of the “drop-in” type in Iceland: The case of Runolfur Runolfsson. Journal of the American Society for Psychical Research 69, 33-59.

Larusdottir, E. (1970). Hvert liggur leidin? (Where Does the Road Lead?). Skuggsja.


  • 1. Haraldsson (n.d.) and Haraldsson & Stevenson (1975b, 35). Although technically incorrect, Haraldsson is treated as a surname in the citation of Erlendur’s work in languages other than Icelandic.
  • 2. Larusdottir (1970).
  • 3. Haraldsson (2021, n.d.). These recordings are now housed in the sound department of the Icelandic National Library.
  • 4. Haraldsson (n.d.).
  • 5. Haraldsson (2021, n.d.).
  • 6. Larusdottir (1970); Haraldsson & Stevenson (1975b).
  • 7. Larusdottir (1970); Haraldsson & Stevenson (1975a).
  • 8. Haraldsson (2021, n.d.).
  • 9. Haraldsson (2021, n.d.).
  • 10. Haraldsson & Stevenson (1974); Haraldsson, Pratt, & Magnus Kristjansson (1978). Accounts of the experiments are also given in Haraldsson (2021) and Haraldsson (n.d.).
  • 11. Haraldsson (2021, n.d.).
  • 12. Haraldsson (2021, n.d.)