Case S02E01 – The Iceland Confessions


See the Self-care page if you need support.

Can you make a person confess to a crime they haven’t committed? To a crime as serious as murder? What about a whole group of people? This episode tells the story of the disappearances of two men and the series of confessions spun out of thin air afterwards. en clair is a podcast about forensic linguistics, literary detection, and language mysteries from past to present. Below you will find data, audio credits, further reading, and a transcript of the podcast.

Audio credits

Scott Holmes – Edge of Nowhere
Kai Engel – November
Kai Engel – Twinkling Stars Won’t Answer Me
Lee Rosevere – Quizitive
Lee Rosevere – The Ambient Ukulele

Credits, sources, and more

The Supreme Court ruling of the acquittal can be found here in Icelandic. It seems to translate somewhat reasonably into English:

Ástvaldsson, J. P. (2018). Acquittal in Guðmundur and Geirfinnur Case. [Accessed 22nd June 2020]

Cocozza, P. (2017). ‘Deep down, I knew it didn’t happen’: The woman who imagined a murder. [Accessed 22nd June 2020]

Cox, S. (2014 [Updated 2018]). The Reykjavik Confessions. [Accessed 22nd June 2020]

Crossing Continents, The Reykjavik Confessions, 20:30, 19/05/2014, BBC Radio 4, 30 mins. (Accessed 22 Jun 2020)

Guðjónsson, G. (2017). Memory distrust syndrome, confabulation and false confession. Cortex, 87, 156-165.

Guðmundur and Geirfinnur case. (n.d.). [Accessed 22nd June 2020]

Helgason, E. (2011). The Tragic Story of Sævar Ciesielski. [Accessed 24th June 2020]

Hilmarsdóttir, S. K. (2018). All persons acquitted in the Gudmundur and Geirfinn cases [Automatically translated from Icelandic]. [Accessed 24th June 2020].

Howitt, D. (2017). Out of Thin Air [Film]. London: Mosaic Films.

Iceland Review. (2018). Acquittal Requested in Unsolved Murder Case. [Accessed 24th June 2020]

National News. (1998). Prime Minister David Oddsson in parliamentary debates: Not one judgment, but many [Automatically translated from Icelandic]. [Accessed 24th June 2020]

Olgeirsson, B. (2018). David sponsored Saevar for a “good” amount. [Accessed 24th June 2020]

Out of Thin Air: Characters. (n.d.). [Accessed 22nd June 2020]

Out of Thin Air: Story. (n.d.). [Accessed 22nd June 2020]

Tómas, R. (2020). Compensation Awarded in Guðmundur and Geirfinnur Case. [Accessed 22nd June 2020]


Case S02E01 – The Iceland confessions

It is the 26th of January, 1974. 18-year-old labourer Guðmundur Einarsson is out clubbing with his friends in Reykjavik (Out of thin air: Story, n.d.). As the night draws to a close, a bitter snowstorm blows in. Despite this, Einarsson decides to walk the 10km back to his house (Guðmundur and Geirfinnur case, n.d.). That’s about six miles, and, at an average walking pace of around 5kmh or about 3mph, and in reasonable conditions, such a distance would take someone around two hours. But of course, the conditions are far from reasonable.

Einarsson’s friends see him leaving the club and heading out into the snowstorm.

A little later, a motorist will see him walking at the side of the road, unsteadily, alongside another man.

And some time after that, another motorist will see him. This time, he is alone, and he almost falls in front of a vehicle (Out of thin air: Characters, n.d.; Guðmundur and Geirfinnur case).

After that final sighting, Guðmundur Einarsson will never be seen again.

For days, the police search the lava fields that they assume he will have walked through, but they find no sign of him. Almost inevitably, they assume that he has died of hypothermia – that is, of extended exposure to the cold leading to death as a direct result of spending hours outside during a snowstorm (Out of thin air: Story). Disappearances into thin air are not uncommon in Iceland, particularly in harsh weather conditions. Indeed, such occurrences feature often in Icelandic folklore, and so, as time passes by, his disappearance is gradually taken to be just another such occasion.

Ten months later, and it is now the 19th of November, 1974. 32-year-old construction worker Geirfinnur Einarsson (who is, I hasten to add, unrelated to Guðmundur – the names are similar due to Icelandic patronymic names) receives a mysterious phone call while at home (Guðmundur & Geirfinnur case, n.d.). He then drives to a nearby harbour café in Keflavik, parks up, and leaves his keys in his car. And then he, too, simply vanishes into thin air.


Welcome to en clair, an archive of forensic linguistics, literary detection, and language mysteries. You can find case notes about this episode, including credits, acknowledgements, and links to further reading at the blog. The web address is given at the end of this podcast.

Fairytales and nightmares

Due to the suspicious circumstances, and especially the unexplained phone call, the police carry out an extensive search, but they cannot find Einarsson, nor a body that could be reasonably identified as his. The circumstances are such that the police launch a murder investigation (Out of thin air: Story, n.d.), but public pressure quickly builds. Where is Einarsson? Has he been murdered? Is he simply missing? Is a killer on the loose? At this time, Iceland has a population of only 220,000 people so how is it possible that a witness, a lead, a potential murderer cannot be found? (Cocozza, 2017)

Despite the intense interest, a long year passes by in which the police are unable to make any progress (Ástvaldsson, 2018; Howitt, 2017). Finally, in December of 1975, there appears to be a small breakthrough. A young couple, Erla Bolladóttir and Sævar Ciesielski, are being questioned about an embezzlement case they have committed together. The police decide to take Erla, who happens to be the mother of a newborn, away for an individual interview (Out of thin air: Story, n.d.). Whilst questioning her, they show her a picture of Guðmundur Einarsson, the man who tried to walk home from the nightclub. When asked if she recognises him, surprisingly, she answers that yes, she does. She’d met him a few years ago at a party. She’d thought him handsome. She was flattered that he decided to talk to her. The police then asked her about the night of his disappearance. She recalls that she had a nightmare that night (Cox, 2014; Out of thin air: Story). Her nightmare involved her boyfriend Sævar and his friends whispering outside of her room.

And here is where I will pause briefly to mention something called the witchcraft analogy. I’ve mentioned it before but it bears repeating. The witchcraft analogy, or in its fancy academic dress, confirmation bias, is where you already believe, or want to believe something, and you then set yourself the task of finding proof. It comes from the idea of the witchcraft trials in which people set out to find witches, on the assumption that witchcraft, in its supernatural, demonic pact form, was real. So a door would slam – malevolent spirits. The cows would sicken and die – wicked curses. A whole harvest would fail – the devil. All one need do is start with a prior belief, and then, to quote Sherlock Holmes as he unravels the Scandal in Bohemia, “Insensibly one begins to twist facts to suit theories, instead of theories to suit facts”.

The police are desperately looking for a killer. And under so much public pressure to make some sort of progress, anything can start to look like a possible lead. Even a dream. So begins the first step down the road of confirmation bias as the police unwittingly begin to find ways to turn this dream into a concrete piece of damning evidence. This nightmare, they reasoned, could very well be a suppressed, traumatic memory. It could have something to do with the man’s disappearance…

The police then take further steps. They interrogate Erla for several hours. I deliberately call this an interrogation because of its coercive nature. During this time she is not allowed access to her newborn. That is, not until she gives what they deem to be useful information. This is another large stride down the road of confirmation bias. They have already decided that only a certain set of answers will be deemed acceptable, and of course, these will be answers that have begun to fit their pre-conceived theory. If Erla wishes to be reunited with her baby sooner, she must comply.

But there’s more than just a coercive element to this interrogation. There’s also an insidiously persuasive aspect. By the end of the questioning, Erla begins to suspect that she has indeed witnessed something traumatic. The next day she testifies that she had seen Sævar and three of his friends with, potentially, a body (Out of thin air: Story). Through their methods of questioning and interrogation, the police have potentially tampered so completely with a suspect’s memory that they have transformed a vague nightmare into a concrete statement of involvement in murder.

The dangerous freight train of confirmation bias, fuelled by the relentless public pressure has now left the station and the police rush out to arrest Sævar, and his two friends Kristján Viðar Viðarsson and Tryggvi Rúnar Leifsson.

All three are subjected to lengthy interrogations (Out of thin air: Story).

And all three testify to having killed Guðmundur.

Another friend, Albert Klahn Skaftason even confesses that he had driven them into the lava fields to hide the body (Guðmundur & Geirfinnur case, n.d.).

What was their motive? A dispute over the cost of a bottle of spirits.

The accounts of the four men seem to match Erla’s account exactly and the police believe they have cracked the case.

But maybe there’s more. They should question these suspects over the missing Geirfinnur. The two disappearances could well be linked. Furthermore, there had been rumours that Sævar had information of the second man’s disappearance.

Another round of interrogations.

Another round of confessions.

But the first cracks begin to show. Initially, Erla confesses to having killed the man with a shotgun. Sævar later says that he did it. He claims that Geirfinnur had died after falling from a boat, following a fight with him and his friends (Out of thin air: Story). But this story changes. Instead, now, Geirfinnur was killed in a boatyard filled with scrap and then he was buried in the lava fields (Cox, 2014).

So many confessions.

And yet, the momentum stalls.


Gold standards

In many jurisdictions, confessions are held as the gold standard of evidence. They have been sufficient to secure convictions for people on trial for the most serious crimes imaginable, even in the face of contradictory forensic evidence.

But, six months later, by the summer of 1976, there are, as yet, no convictions. Instead, the case has drifted into a strange purgatory. Four of the suspects are still being held in custody (Out of thin air: Story, n.d.) but in arguably the clearest moment of justice in this whole case, the police want something more solid than words alone. They want physical evidence. And as yet, they have found no bodies. There are no murder scenes. And without either of these, there is simply no way to get incontrovertible, hard forensic evidence.

At yet another impasse, the Icelandic police finally determine to bring in outside help, and they contact the German ambassador, Pétur Eggertz, and this leads to a delicate negotiation with Germany’s Secretary of the Interior Ministry, Siegfried Fröhlich. The outcome of these discussions is that then-64-year-old Karl Schütz takes over the investigation.

So who is this Schütz? It’s worth spending a moment getting to know him since he now looms large in this story. Schütz is newly-retired from German Federal Criminal Police Office, also known as the BKA. Whilst with the BKA, he had been in charge of their investigation and evaluation units, though the German newspaper, Spiegel, suggests that he was not especially successful in this role. As an expert in large criminal cases, he sets up a task force of ten officers with the sole aim of resolving the case.

At first, the team work hard on finding physical forensic evidence – from the car supposedly used to transport the bodies, from the clothing worn by the suspects when the murders supposedly happened, from a knife, from carpets, even from handwriting samples. But even with the might of the BKA crime labs behind him, Schütz can find no incriminating evidence. No blood. No faeces. No vomit. Nothing.

With this avenue firmly closed, Schütz chooses another way forward. He turns back to the confessions. These have been conflicting and contradictory. Too contradictory. A jury would not be convinced. What these stories need, Schütz decides, is to harmonise. Everyone needs to be saying the same things. Telling the same story. So how do you make six people all tell the same story without directly instructing them to do exactly this? How do you force matching confessions from people who cannot meet up to all agree on the same set of circumstances?

Putting the harm in harmony

For a start, remember that the police started from a rather tenuous position of interpreting a bad dream to be a supressed traumatic memory. From there they swiftly moved onto a general presumption that they had found the murderers, and after that, they dedicated themselves to finding evidence to fit this theory. In other words, they began to believe in witchcraft, and all that remained was for them to find their witches. Like so many things, confirmation bias doesn’t start out as a towering oak tree growing in the middle of a desert. It starts as a tiny seed that slips into a little opening – the ambiguity of a dream, the suspect’s self-doubt, the police force’s eagerness, and from there it thrives in the hot, pressured atmosphere of ongoing public displeasure. Another example of this exact same issue in another police investigation involving multiple murders: the Yorkshire Ripper case, from Series One.

Once the confirmation bias has put down such deep roots, once the majority of people, and especially those in power, have bought into this theory, this presumption of guilt, it becomes extremely difficult to start suggesting alternative narratives in which the suspects are, in fact, innocent. Why? Because your alternative story implies that a whole raft of police officers and their superiors are wrong. And not just a little bit wrong. Hugely, embarrassingly, excruciatingly wrong. That they have treated innocent people with egregious inhumanity. That they are bad at their jobs. And with that, comes even more public humiliation than they have already been experienced. If they had to let the suspects go without charges, the news would break instantly, and to mass uproar. Imagine the headlines. After a year in which no progress has been made, the Icelandic police have managed to follow up with an almighty, hideous blunder. The damage to the force’s reputation and competence would be severe. So by this stage, even if officers and superiors really do have doubts, it’s too late. Too much reputation and time and money have been invested. To admit any possible error now would be a PR disaster. And so, just like in the Yorkshire Ripper case, the confirmation bias becomes self-protecting. It isolates out those who would oppose it, and rewards those who protect it. At least, it does in the short-term. It is, you might say, almost bewitching in its deadly simplicity.

Of course, we will probably never know how extensive the doubts were in the force at that time. All we can do is reconstruct the story as it unfolds, and with the arrival of Schütz, things escalate quickly.

Collectively, the suspects had already been interrogated over a dozen times, but that number now spirals into the hundreds. The police also begin to spend lengthy amounts of time taking the suspects out to search the lava fields for bodies and to re-enact the crimes. So, where’s the linguistics here? Well, it’s woven into every interaction. Imagine that I ask you, “When did you kill him?” Firstly, this is guilt presumptive. I’ve assumed that you really did kill him. We call these presuppositional questions. But there’s more. Imagine that I finally break your spirit and force you to “confess” that you did, indeed, kill someone. Then I start saying, “Okay good. Now we’re making progress.” Nice little bit of positive reinforcement there for your supposedly cooperative behaviour, right? And then I say, “Look, I’ve already been talking to your co-conspirators and they’ve already told me everything. So all I need from you are some details.” Now I’ve framed my upcoming narrative as authoritative. It’s backed up by supposed evidence. I know stuff you don’t. You’re likely to assume that I’m telling the truth. After all, I’m working with law enforcement. Who are you to question me. So then I say, “Who buried the body?” Notice this is presuppositional again – I’ve assumed that there actually is a body. And, more importantly, I’ve just indirectly told you that in the current version of events, that body was buried. But what happens when we inevitably run into contradictions or inconsistencies between the stories? Maybe you say that you buried it and the other person says they did. Well, one of you is obviously being loyal and protecting the other. Or what if both of you each blame the other? Well obviously, one of you is a cowardly liar. And if it just so happens that your answers all agree, then this is just further proof that the police are onto the right track.

And so, like a parasite or a disease, the confirmation bias continues to protect itself against any counter-arguments and evidence that might come along to destroy it.

So now these suspects are being questioned over, and over, and over. With every interrogation, every re-enactment, every discussion, every question, every response, the police are, wittingly or otherwise, feeding the suspects the “correct” information, and in so doing, they are spinning up a whole narrative out of thin air.

In between interrogations, the suspects are kept in solitary confinement. The shortest is 87 days – that’s about three months. The longest is 655 days. That’s almost two years, and it’s the longest period of isolation reported outside of Guantanamo Bay (Guðmundur and Geirfinnur case, n.d.). You probably don’t need me to explain that solitary confinement is considered a form of torture and is a serious human rights violation. Several suspects will later say that they only confessed to end the confinement.

But there is more. The suspects and especially the supposed ringleader, Sævar, are hypnotized, subjected to water torture, deprived of sleep, and given regular doses of powerful drugs, allegedly including mogadon, diazepam, and chlorpromazine. These act as sedatives, and sedatives affect our ability to accurately recall (Howitt, 2017). As we already know, Erla is denied access to her newborn, the suspects are regularly denied legal counsel, and as I’ve mentioned before, throughout the whole, they are presumed guilty. And then there remain perhaps the two most problematic issues in the case overall: the first is the lack of hard forensic evidence. No bodies. No crime scenes. But the second is the lack of any reasonable motive. It is worth noting that some killers do indeed kill without any substantive motive beyond what might be called pleasure, for want of a better term. But statistically such individuals are extremely rare, and a group of them all working together is substantially rarer still. Wildly improbable, even.

In spite of these two huge issues, however, under Schütz’s direction, the case finally goes to court.

In justice, we trust

In the duration between the confessions being given and the trial starting, finally, the suspects have time to think, and inevitably, several of them start trying to revoke their confessions (Out of thin air: Story, n.d.). In fact, they begin to openly doubt that they were ever even involved in the crimes.

Yeah, comes the response. Sure. So why would you confess? Everyone knows that you don’t confess to a crime you didn’t commit. Least of all to murder.

It is easy to imagine the blunt scepticism that they face, and it is easy to understand that scepticism too. I mean, criminals often plead their innocence. Some will continue to do so in the face of overwhelming and incontrovertible evidence. And why would someone say that they committed murder? In this case at least, there are no obvious wins for doing so. They’re going to lose their jobs and families and any social standing they have. They’re going to become notorious criminals, and in a country with a small population this infamy is only exacerbated. Everyone will know their names and what they’ve done. Moreover, it’s not just one person confessing to a crime. This is several people, all apparently supporting each other’s versions of events. What are the odds of that? And so, one can readily imagine the public at large thinking, well of course you have regrets now that you’ve seen all the consequences that await you! How could a whole group of people end up fabricating a story that would only do themselves enormous amounts of harm? It just didn’t make sense. The only logical explanation from the outside, was that these people were obviously guilty, and of course now they were sorry. Unsurprisingly, then, the increasing protests that they only confessed due to police coercion go largely ignored, and so, the original confessions are used in court.

The outcome is painfully predictable, and ultimately, six people are convicted.

Sævar and Kristján are found guilty of killing both men.

Tryggvi is convicted for killing Guðmundur.

Albert is convicted for helping to hide Guðmundur’s body.

Sævar’s former teacher and friend Guðjón Skarphéðinsson is convicted for killing Geirfinnur.

And Erla, the one whose bad dream was the at the start of all of these events, the young mother who was kept from her newborn and told that she had obviously repressed some sort of trauma? In an awful twist of irony, she is convicted for perjury.

Yes, that’s right.

She’s convicted for lying during a criminal investigative procedure. According to the prosecution, she wrongly implicated her half-brother Einar and three of his friends (Guðmundur and Geirfinnur case, n.d.). Somehow her statements about him and his friends are deemed false, but her confessions around the others are deemed to be true. It is, I’m sure, irrelevant that the half-brother in question, Einar Bollason was then a three-time national champion basketball player…

Ultimately, these convictions lead Iceland’s Minister of Justice to joyfully report that, “the nation’s nightmare is over”. However, although these headlines are published in February of 1977, the suspects are not finally sentenced until 1980. Finally, Erla is given the shortest sentence of two years, and the supposed ring-leader, Sævar, gets the longest – seventeen years in prison (Out of thin air: Characters, n.d.).

Civil twilight

The months pass by. And then the years. And one by one, each of the convicts are released from prison. With time, and reflection, some of them fight to reopen the cases, that there has been an awful miscarriage of justice… but the Supreme Court rules that it will not re-hear the case (Helgason, 2011).

Nearly two decades pass by, during which time the murders and convictions saturate the Icelandic national consciousness at every level. And finally, in 1998, the then Prime Minister of Iceland, Davíð Oddsson criticises the investigation and prosecution procedures, and notes that he is disappointed that the case is not being reopened (National News, 1998). It will later transpire that he had actively been sponsoring Sævar’s attempts to appeal the case (Olgeirsson, 2018), though ultimately none of the efforts would succeed. But even with this impetus from the country’s own premier, there is no movement.

Yet another decade passes by. Thirty-five years after the first alleged murder, in 2009, Tryggvi passes away from cancer. Two years later, in 2011, Sævar also dies. To the very end of his life, he continues to protest his innocence, but it is his death, rather than anything he was able to do in life, that renews interest in the case (Out of thin air: Story, n.d.). Finally, someone starts to listen. Journalist, Helga Arnardottir starts researching the case from a different angle (Out of thin air: Story, n.d.), and discovers that Tryggvi’s daughter, Kristin, has secretly kept hold of her father’s prison diaries, in which Tryggvi detailed his innocence and the many doubts he had over his guilt (Cox, 2014). Recognising their importance, Helga Arnardottir immediately flies to London to show them to forensic psychologist Gísli Guðjónsson. Guðjónsson has previously worked in high-profile miscarriages of justice cases, such as the Birmingham Six, and is an expert in false confessions (Out of thin air: Characters, n.d.). Within three hours of reading the diaries, he goes on record to say that the case should be reopened, and two weeks later, a government inquiry is announced (Out of thin air: Story).

So what does Guðjónsson see in the diaries that is so compelling?

The mortal memory

When reviewing the diaries, Guðjónsson finds evidence that these do indeed look like false confessions. Moreover, they appear to be one particular subtype.

False confessions can fall into roughly three categories. There are the voluntary kind, as sometimes happen when a parent takes the blame for their child’s actions, and they do this without any external pressure from, say, the police. Then, as an obvious contrast, there are pressured-compliant, or coerced confessions. Essentially, someone simply can’t cope with whatever is being said or done to them, whether it’s threats, or psychological torture, or physical force, and they try to escape the situation by giving what they know to be a false confession – usually whatever they think their interrogator wants to hear. Whatever will make the awfulness stop (Guðjónsson 2017: 158).

That might sound like the kind of confession we have here, and that’s probably partly it, but there’s also a third type – the pressured-internalized confession, and this is caused by memory distrust syndrome (Guðjónsson, 2017). In Guðjónsson’s words, memory distrust syndrome is ‘a condition where people develop profound distrust of their memory and become susceptible to relying on external cues and suggestions from others’ (p. 156). Essentially, the investigator or officer or whoever, knowingly or otherwise, erodes someone’s confidence in their own recollections so completely that they end up having more faith in the investigator’s claims about the events, than they have in their own lived experiences of it. It is a grotesque reverse gaslighting. Rather than adamantly saying something never happened, the figures in power or with authority claim over and over that something did happen, and finally, again in Guðjónsson’s words, ‘due to external pressure, the person is persuaded that he or she has committed a crime of which he or she is innocent, but typically the person has no clear memory recollection of committing the crime but accepts that they have done it’ (Guðjónsson 2017: 158).

Fortunately, Tryggvi was not the only one of the suspects to have kept a diary through their time in prison or solitary confinement, and Guðjónsson’s study ultimately focused on the diary entries that Guðjón Skarphéðinsson kept throughout his 412 days spent in solitary confinement (p. 160). Guðjónsson saw Skarphéðinsson’s diary as a perfect example of memory distrust syndrome. Added to this, Skarphéðinsson was the last to retract his confession – he didn’t finally take that step until 1996 – and even today he still has no idea whether or not he was involved in the crime.

We all think, or believe, that we know what we have and have not done. What we are and are not capable of. So how do you so completely distort and destroy someone’s memory of an event as visceral and unique as a murder to the point that they can no longer be sure about even whether they were involved? But there’s another consideration to add to this. I think it’s reasonable to argue here that even if the Icelandic police force desperately wanted something that looked like a successful outcome to the case, they presumably did not want to actively manufacture a bunch of false confessions from people, and so I extend to them the courtesy of believing that, whatever the ultimate outcome, they did actually intend to get to the truth and perhaps they truly believed that they were finally getting there. And false confessions happen too often and too widely around the world for every possible case to involve malicious bad actors who know what they’re up to and simply don’t care. So how does a team of presumably well-meaning but misguided police officers ultimately induce such profound memory distrust syndrome in not just one person, but in a whole group of six or seven people?

Guðjónsson charts five steps along the road to creating a full false memory of committing a crime (p. 159). These are a trigger, plausibility, acceptance that the event may have happened, reconstruction of the event in their imagination, and the resolution of the true memory coming back. So let’s apply that to the case here.

For step one, there were three vital contextual risk factors that were key components in the development of a trigger (Guðjónsson 2017: 162). These were social confinement, the investigators’ guilt presumption and persuasive interrogation, and the high emotional intensity. These were all well-documented in the diary of Guðjón, which exhibits high emotions and confusion. Erla, too, stated that in her interrogations the officers befriended her and used leading questions that implanted information. These questions would use phrases like, ‘Do you recall…’ or ‘Is it possible that…’ or ‘Do you think that…’ and then they would provide some situation or other. Hypotheses. Guesses. Pure fabrications. Effectively, this created a whole series of potential narratives that could later be developed depending on which provided the best and most complete traction with all the other suspects (Crossing Continents).

What of step two, plausibility? Well, with so many stories to choose from now, the individuals and their whereabouts and activities are forced to fit into narratives wherever they can, conceivably, just barely, maybe, have been involved in the murders. Worse, Skarphéðinsson’s pre-existing vulnerabilities made him especially easy to coerce (Guðjónsson 2017: 162). His personality, as evidenced in the diary, was one of high compliance and low self-esteem. He had great trust and respect for authority, and he had a vivid imagination. Moreover, he had good faith in the investigators and felt a strong sense of importance that he could help save the case if he gave them the information they wanted. It was therefore all too easy for the officers that he greatly respected and wanted to please to encourage him to doubt his own memory when compared with their assertions about the case, and as a result. He was, essentially, a prime target for suggestibility.

From here, it’s but a moment’s work to reach step three – acceptance of the events, and that will only be increased as each hears that the others are now agreeing that yes, maybe they were involved after all. Finally, in the fourth step, we have reconstruction of the event in their memory, and remember, the police repeatedly took the suspects to the supposed crime scenes to re-enact their various stories until they achieved some sort of agreement. From physically visiting and revisiting the spots to working over a narrative time after time to hearing others in authority claim that a situation occurred, finally it is possible to implant in someone’s mind a memory of an event that simply never happened.

And at long last, the diaries take us to step five, when the true memory tries to resurface, along with the doubt, and confusion, and questions.

Guðjónsson’s conclusions describe how memory distrust syndrome appears to have been a prevalent factor in the false confessions of all the suspects, and that this was caused by the inhumane investigation procedures used at the time.

A pale dawn

Remember how, in September of 2011, a Government inquiry was instigated on the back of Guðjónsson’s report? Well, eighteen months later, in March of 2013, this inquiry will detail the problems, limitations, and abuses of the police investigation (Out of thin air: Story, n.d.). Within, it will discuss the amount of time a person can be kept in solitary confinement, and the use of lengthier periods as a means of breaking down resistance (Cox, 2014). It will also consider the rules on interrogation, how, even by the standards of the time, the methods used were extreme, and whether the approaches taken could be justified by the pressure the police found themselves under.

Despite this, it will still take four more years before the Interior Ministry’s Rehearing Committee will finally conclude, in February of 2017, that that the five men’s cases should be reheard by the Supreme Court.

That’s right.

The five men, yes, but not the woman. Not Erla (Guðmundur and Geirfinnur case, n.d.).

At long last, almost half a century after the first alleged murder, in February of 2018, the State Prosecutor will request that the Supreme Court acquit them. Six months later, on the 27th September 2018, the Supreme Court will indeed acquit all five of the men, but it will not declare any of them innocent (Hilmarsdóttir, 2018; Iceland Review, 2018). Despite this unwillingness to declare them innocent, two years later, in January 2020, just a little over a year ago as I record this, compensation has been paid out to the acquitted parties and to the families of the deceased (Tómas, 2020).

Finally, at the time of this recording, in 2021, there has been no update on the fate of the two missing men, Guðmundur Einarsson and Geirfinnur Einarsson – the men whose disappearances started the whole story almost half a century ago.


The story behind this episode of en clair was brought to my attention by Professor Tony McEnery, who emailed it to me almost two years ago now. It was then painstakingly researched by my intern, Debbi Tomkinson, and it was narrated and produced by me, Dr Claire Hardaker. However this work wouldn’t exist in its current form without the prior efforts of many others. You can find acknowledgements and references for those people at the blog. Also there you can find data, links, articles, pictures, older cases, and more besides.

The address for the blog is And you can follow the podcast on Twitter at _enclair. Or if you like, you can follow me on Twitter at DrClaireH.