It sounds like the plot of a particularly gripping thriller but, according to German police, it really happened. In August last year, a Bavarian woman identified as Sharaban K attempted to fake her own death by killing a doppelganger she contacted through Instagram.
Her victim was a beauty blogger lured into a meeting by the promise of free cosmetics. The disfigured body of Khadidja O, stabbed 50 times, was initially identified as Sharaban K by members of her own family.
A few months later, on a London-bound train, a passenger saw the doppelganger of a friend he hadn’t seen in some years, “although this person was a lot rougher round the edges and about 20kg heavier”, the passenger told Reddit. They were intrigued enough to sneak a photo of the person and send it to the old friend, noting the close resemblance, “if you’d packed on a few pounds”. The terse reply was not long in coming: “That was me.”
Doppelgangers have been part of folklore since at least ancient Egyptian times. But now, science is demonstrating that they could serve more serious purposes.
Coincidentally last August, the medical journal Cell Reports described research undertaken by a team led by Dr Manel Esteller, of Barcelona’s Josep Carreras Leukaemia Research Institute. It’s a densely written report that makes the average reader yearn for a sentence beginning “To put it simply …” or “In other words …”, but it purports that close lookalikes who are not related are likely to share a considerable amount of DNA. Which, Esteller acknowledges, “would seem like common sense, but never had been shown”.
The research used a sample of 16 pairs, sourced not by the lure of some free lippy but from a fascinating project by Canadian photographer François Brunelle. The pairs were matched according to three different methods of facial recognition, and nine were close enough according to all three measures to warrant further study.
Reports tended to obscure the fact that the project’s conclusions were relatively modest and the study itself had caveats, including a small sample size, the use of 2D black-and-white images, and the predominance of European participants. However, it did produce some surprising results, noted Science Daily. “Examples of independent questionnaire variables (such as height, weight, smoking habit, or level of education) further demonstrate that lookalike pairs are closer than non-lookalike pairs. Thus, humans with a similar face might also share a more comprehensive physical, and probably behavioural, phenotype that relates to their shared genetic variants.”
To put it simply, people who share DNA are also likely to share non-genetic qualities. This has consequences in several areas.
“I think the research is useful and can help improve precision medicine, since anyone who looks like you also might have similar HLA alleles [alternative forms of a gene], which we also need for matching for organ donors,” says Professor Christopher Mason of Cornell University.
The study goes on to say that its “results are germane to the ongoing efforts [in] diagnosis of genetic disorders using facial analysis technologies”. In other words, people who resemble each other closely may also tend to share certain diseases. Avenues for further research include the use of facial features to predict the possible presence of genetic mutations, such as those associated with diabetes and Alzheimer’s disease.
And, Esteller told Science Daily, “These results will have future implications in forensic medicine – reconstructing the criminal’s face from DNA – and in genetic diagnosis – the photo of the patient’s face will already give you clues as to which genome he or she has.”
A DNA sample from a crime scene could potentially be used to construct an image of the face of the person it came from. Not everyone thinks that is a great idea, however. It also raises the grisly spectre of prenatal facial-reveal parties to add to the twee terrors of the gender-reveal party.
Comparative anatomy
Despite its limitations, the study attracted widespread international attention, reflecting our deep and abiding interest in the notion of the doppelganger. But how many of us have one, really?
It depends on what you mean by a doppelganger. And that depends on how many similarities you require to qualify someone as a lookalike. Quite a lot of people have the same hair colour. A smaller number have the same hair colour and eye colour. An even smaller number have the same shaped ears. And so on.
In fact, according to Maciej Henneberg, emeritus professor of anthropological and comparative anatomy at the University of Adelaide, we need just five facial features in common to find a resemblance between two people. This explains the phenomenon of thinking we recognise someone we know before finding, on close inspection, it’s not them.
Henneberg points out that for each of those five features, there is a big range of possible differences – the overall number of gene combinations is huge, so the doppelganger effect as such is not really meaningful.
“If we take the entire DNA into account, the number of possible combinations goes into quintillions. And therefore, yet again, there are no two individuals who are absolutely identical.”
Another factor, which explains the frequently observed fact that children of the same parents are not identical, “is that there are about 100 mostly neutral mutations influencing the sequence of nucleotides in DNA when a child is formed”.
Doppelganger hunters may also be led astray by delusional misidentification syndromes. Relatively rare, these include the likes of Capgras delusion. This is the belief that someone in your family has been replaced by a lookalike, which can have serious consequences for relationships. Even more bewildering is Fregoli delusion, a rare mental disorder that makes someone believe different individuals are in fact the same person who constantly changes their appearance.
Mistaken identity
Confirmation bias can make doppelgangers appear where they don’t exist. Our brains tend to connect new information with what we already know – in this case, to assign physical features to people of our acquaintance or to see more similarities than there really are.
As far as potential medical benefits suggested by the Barcelona research go, Henneberg is bracingly realistic.
“There are already physical characteristics that indicate genetic conditions. I’ll just mention Down syndrome. It’s a chromosomal aberration. People [with it] are recognisable and have various other little conditions as well.”
He’s not bullish on the potential use for organ matching, either. Live donors can provide bits of liver and lung, but for procedures such as heart or face transplants, the donor has to be dead, which is severely limiting in cases where murder is ruled out for moral or practical reasons. That may seem far-fetched, until you look again at the first paragraph of this story.
Henneberg has worked extensively in forensics, helping to identify people in criminal cases, especially where CCTV footage is being used to identify someone when there is no DNA sample – a practice made more likely by the Barcelona findings.
“Cosmetic surgery can reshape a person quite a bit,” says Henneberg. “You don’t even have to do plastic surgery. If you have a good makeup artist, you may change a person’s appearance very significantly.
“It’s nothing unusual that similar-looking people are genetically similar. But when it comes to identification in forensic cases, we have, let’s say, CCTV images of somebody committing a crime, and then we have a suspect. And we have to make sure that the person on the CCTV images is actually identical with the suspect.”
On one of his cases, it was only when he noticed the person in the CCTV imagery had a flat left foot that he was able to rule out a suspect.
Mirror image
The current interest in doppelgangers is heightened by the epidemic of narcissism powered by social media.
Philip Kronk is a retired psychologist and neuropsychologist from Knoxville, Tennessee. In a prescient column for the Knoxville News Sentinel back in 2016, he wrote, “The idea of having a double has become a positive thing … It is, however, still related to self-esteem, self-identity and healthy and pathological narcissism. The idea of having a double has been changed by our everyday use of technology: the internet, Twitter and, of course, the selfie.”
Kronk tells the Listener this reflects that “people want to think they can live forever – it’s a way of enhancing yourself”. The corollary of this is that the obsessive recording of every moment of every day is a death-defying way for individuals to create their own doppelgangers, who will continue to exist in cyberspace no matter what happens to their real-life selves.
Dr Naomi Murphy, a clinical and forensic psychologist and honorary professor of psychology at Nottingham Trent University, expands on this idea: “I think it’s a slight sense of narcissism that we value people who are like us, who look like us and feel like us.”
She extends the connection to the often-observed phenomenon of people choosing partners who look like a version of themselves when hooking up. “People are attracted to people who look like them. From an evolutionary perspective, I guess you want your DNA to survive, so you choose people who also represent your DNA to maximise the chance of passing that on to the next generation.”
A doppelganger also provides the chance to see ourselves in a more objective way. A reflection in a mirror is not how other people see you; a separate person is. “If somebody says to us, ‘Oh, that person looks so like you,’ it allows us to get a sense of how people see us from a distance that we don’t ordinarily get.”
Ominous sign
Celebrity lookalikes have been making a fortune from their famous faces for years, but if doppelgangers are indeed having a moment, it’s only the most recent in a long line that stretches back centuries.
“The double of a person in spirit form was a very big deal in ancient Egypt,” says UK folklore researcher Mark Norman. “They’re often about conveying people to the other side or linking to the other world, which was a very important part of the Egyptian culture. What seems to be the case in the majority of doppelganger folklore is that it’s portentous. It serves as an omen of either your own death or the death of somebody else.” Again, think of the Bavarian incident.
For everyone who thinks a doppelganger means double the fun, there is another person who finds the idea unsettling to the point of being existentially threatening. We are supposed to be unique, so the existence of another “us” has the potential to trigger an identity crisis. I am here, but there I am again over there. This is why so many doppelganger stories are down the spooky end of the library.
Norman finds these connotations less prevalent today. “The most intriguing probability, in terms of folklore, associated with modern lookalikes is something like the stories that surrounded Saddam Hussein and his numerous doubles. How much of that is apocryphal?”
We may never know. But we do know that similar stories surrounded Jacinda Ardern, spread by conspiracy theorists who knew for a fact that the woman in the Beehive was a lookalike taking the then-prime minister’s place while she was in an overseas court being charged with crimes against humanity associated with the pandemic.
Norman also sees that social media and technology have altered the doppelganger model. “With developments in deep fake technology, you can create a very convincing doppelganger.”
This is related to the old idea of the evil twin. “You can create anybody’s double and portray them as doing something evil that wasn’t them and wouldn’t be them.”
Doppelgangers have many storytelling uses. In comedies such as the movie Dave, or Mark Twain’s novel The Prince and the Pauper, lookalikes are played for laughs. In thrillers, finding someone who looks just like you to take the blame or provide an alibi is an everyday occurrence.
Attention-grabbing
But why is something so unlikely such a constant source of speculation?
David Aldous is a retired mathematician formerly at the University of California, Berkeley, who specialises in matters of chance and probability. He believes our need for stories is at the heart of our fascination with the doppelganger. Such stories help us appreciate and deal with uncertainty in life in general.
“It’s out of the ordinary because we expect different people to look different,” says Aldous. “The doppelganger is unusual, and we instinctively pay attention to the unusual. If I asked you what you saw the last time you drove your car, you wouldn’t remember. But if you saw an elephant walking down the street, you would remember. The unconscious takes for granted almost everything we see. But the unusual things come to attention.”
As for the criminological corollaries, “When you get into criminals and line-ups, and identifying people, it’s often wrong. In that sense, humans are not actually as good as they think they are. We think we’re very good at facial recognition. The reality is, we’re not.”
So, perhaps more amazing than the fact that lookalikes exist is the fact that, although there are eight billion of us on the planet, there is for every one of us, no matter how closely our faces might resemble each other, something at some level that makes us different from the other 7,999,999,999.
Reinforcing bias?
The possibility raised by the Barcelona research that DNA samples could one day be used to create facial images of potential criminals has raised ethical concerns in the United States.
“As with any form of scientific research, there are downstream implications,” says Dr Daphne Martschenko, an assistant professor at the Stanford Centre for Biomedical Ethics, California. “And in this particular study, some of the downstream implications were raised by the authors themselves, when they spoke about the potential for their research findings to aid in criminal profiling.
“In the United States context, we see that communities of colour – black, African American, Hispanic communities – are over-represented in criminal DNA databases. Using this technology to aid in criminal profiling may work to exacerbate those pre-existing systemic biases within the criminal justice system.”
Martschenko has a warning that applies to the use of research in general. It’s a feature of the US system that “the scientific research enterprise broadly is not incentivised to think about the broad social and policy implications of the work, and social harms in particular,” she says.
Review boards in US institutions monitor federally funded research and, under a concept known as the Common Rule, are expressly prohibited from considering downstream social and policy implications. “Researchers kind of outsource the ethical responsibility to review boards,” says Martschenko.
At its worst extreme, this could lead to scenarios where “globally, we see the use of genomic research on things like cognitive ability and intelligence being used by members of white supremacist alt-right groups”. Already, such groups have quoted mainstream research in their manifestos. “The Buffalo shooting in May 2022 [in which 10 Black people were killed]is one example of a white supremacist using genomic studies that have been published in top journals like Nature Genetics.”
The situation in New Zealand is complicated but different in one important respect. The National Ethics Advisory Committee’s standards are specifically grounded in Māori ethical principles. One of those is tika, which “relates to the design of a study, and whether the research achieves proposed outcomes, benefits participants and communities and brings about positive change”.
‘I am not Mr Bean!’
Without the art of François Brunelle we might never have found out that lookalikes share DNA to the extent they do. He has been patiently engaged in photographing unrelated doubles for more than 20 years in an ongoing project called “I’m Not a Look-Alike!”
Sixteen of them became the subjects in a scientific study. “I tend to be a collector of things,” explains the affable Canadian, who is based in Montreal. “I collect children – I have six. I speak five languages. I collect cameras, I collect photos, videos, food, plates. So, I guess this lookalike project is just another of my crazy collections.”
He even has his own lookalike story. “Some English acquaintances told me that I looked like Mr Bean, but I didn’t know what Mr Bean was. One day, I was watching television and I saw a guy who looked like me. I told myself, ‘Oh my god, this is me.’ But then the programme ended and I saw the words ‘Mr Bean’.”
He took his youngest son to see a Mr Bean movie, but it was years before he found out the boy had just assumed that was his dad on the screen.
He does not look like Mr Bean now. But the experience did lead him to start seeking out lookalikes. “I reached out to some of the lookalikes I had met over the years. Out of maybe 12 or 15 pairs that I knew, eight or nine came to my studio and they did photographs.”
How did they react when they were brought together? “They say, ‘Hello, John,’ ‘Bonjour, Pierre,’ then there’s a surprise and then it vanishes right away.”
He found others through the media, mutual acquaintances, friends of friends of friends. “And it took quite a while. Because nobody could understand what I was doing. I was not looking for famous people.”
Questions of identity and what makes a person that person are at the heart of the work.
“It’s about who I am. When I look at myself in the mirror, I see kind of myself, but when I look at the Mr Bean character, I see myself.
“We’re not the same, obviously. So, what am I? Who am I? I don’t talk too much about it, but that’s the core of my project.”
In an earlier, equally extraordinary work, he put together lookalikes across time, not space, pairing contemporary people with subjects of artworks from ancient times. The museum where they were shown invited people to submit pictures of themselves or acquaintances that resembled classical sculptures in its collection.
Again, the project emphasised what humans have in common rather than what separates us from each other. Brunelle saw a sculpture he thought looked like a friend of his.
“He is a photographer and sometimes they are a little bit arrogant, you know, especially in advertising. They know everything; they’ve seen everything. And this sculpture looked the same. The Roman looked like a person who had been around and would tell you what to do.”
Maybe it’s proof, concludes Brunelle, that we are all basically the same. “Maybe we didn’t change. We just have more stuff around us. We get born, we live and then we die. In between, we eat a lot.”