Is this the true face of Anne Boleyn? | BBC News

Added: 2026-05-09

[00:00:02]When I started looking into this story, I thought it was going to be a simple one. Then this happened. The result has shocked us completely. Has facial recognition discovered a hidden portrait of Anne Berlin, wife of Henry VII?

[00:00:17]>> I'm afraid to say it, but I think it's a load of rubbish.

[00:00:19]>> Is this traditionally recognized portrait of the tuda queen actually her mom instead?

[00:00:25]>> I don't get walked off. Just results are results, isn't it?

[00:00:30]To really understand our fascination with this TUDA queen and what she looked like, I've come to London to visit the National Portrait Gallery.

[00:00:43]>> So, welcome to the Tudtor Gallery.

[00:00:46]>> Wow, look at this.

[00:00:47]>> So, I think it's fair to say that Amberlin is one of the most famous figures from English history. So, she is Henry VII's second wife. And I'm just looking because obviously her husband is there. You wouldn't want to mess with him, would you?

[00:01:01]>> He ultimately divorced Katherine of Araggon, breaking England from the church in Rome for Henry to then marry Anne Berlin. So she triumphed, became queen of England. There's their baby who became Queen Elizabeth the first. Right.

[00:01:14]>> Exactly. But then after only three years, court kind of conspired against her, her own mistakes, her own actions, and she ended up being executed on the scaffold at the Tower of London. We know this painting was made in the late 16th century, i.e. long after Anne's death.

[00:01:29]>> So, how would they know what she looked like?

[00:01:31]>> Well, this is the kind of conundrum of the many mysteries that surround Anne Berlin. One of them is the fact that we don't have a lifetime painted portrait of her that's kind of absolutely secure, a kind of um a wonderful painting that we can use as a reference point to tell all those stories about her. Her reign wasn't necessarily long enough for a kind of established iconography. And there is this kind of tantalizing suggestion that perhaps some of her images might have been deliberately destroyed.

[00:02:00]>> And we've been fascinated with what she looked like ever since.

[00:02:08]>> So could computer science reveal what an actually looked like during her lifetime?

[00:02:14]>> More recently, we've got a lot more powerful facial recognition systems.

[00:02:18]>> If evidence can be tested, then it should be tested. Facial recognition allows computers to identify or verify someone from their face. The technology isn't without controversy. It's used for things like unlocking your phone, passport control, and some police investigations. But what about using facial recognition on sketches from the past?

[00:02:42]>> An objective tool like this can be used to actually answer quite a lot of interesting and unknown questions. Even though there are no known paintings of Amberlin made during her lifetime, there are a few depictions, including this sketch by Hans Holine the Younger. It is considered by many modern historians to be a genuine preparatory sketch of her.

[00:03:07]So, she would have had to sit for this sketch, giving us a close likeness, except for the questions. The age of the lettering, it's not in a TUDA style, but likely an 18th century hand. The light colored hair when she was thought to be dark-haired. The full chin when accounts suggest a slender neck and her informal dress on a queen. And there are, of course, arguments to counter these. More on those later. A team led by the University of Bradford has used facial recognition on the digital copies of the whole collection of drawings that this one belongs to. to see if an algorithm can put to bed the debate.

[00:03:47]>> So the algorithm has been trained on photos of people. What we are looking at is a bunch of um drawings and then we are comparing these drawings through a machine learned algorithm where there's no human input. What we've done here is we've compared these drawings to Amberllin's first cousins and to her daughter Elizabeth to look for the family um similarity and geometry and they cluster and we've used drawings that we absolutely know are non relations and they don't cluster.

[00:04:23]They've interpreted the percentages and clustering where images grouped together based on facial resemblance to propose a new theory that this drawing is more likely to be Amberlin's mom and this previously unidentified woman could in fact be the doomed queen. I think it's very exciting we are opening new doors >> to generate leads in art history you know where the evidence is fragmentaryary and historians face a momentous task case solved then >> not so fast >> a lot of uh art historical colleagues are a bit nervous about this kind of sort of gloves off stuff but I think you know if you see it you've got to say it why not I think academically I would describe it as a load of old fooy I think this is trying to treat works of art as modern photographs and it's simply impossible to do that. That's before you get into any of the other problems about their methodology and how they have approached this whole question. First of all, the drawing that we have for a long time thought was Anne Berlin has her name on the top left hand corner. It says Anne Berlin Queen and we know that that identification came from someone who knew what Ambulin looked like. to attempt to demolish that with the aid of computer science 500 years after the event I think is suspicious.

[00:05:42]But I also think it's suspicious that they then say that another completely unrelated drawing which no one has ever said is ambulin before is in fact amblin. I think one of the things that convinces me that this drawing is in fact amblin is that she's shown in a very informal pose. And it sounds paradoxical to say it, but actually a queen would be more likely to be seen at court in this kind of private dress. For much of the period she was queen, Amberlin was pregnant. So I think it makes perfect sense that this is recording her in that moment.

[00:06:16]>> So where does this leave us? It's important that you know this work and the result of it isn't about replacing scholarship or this is definitively an it's not the end of a conversation it's the beginning of a conversation.

[00:06:34]>> So this is just another tool I think that will help the people who are working in this field especially art historians. The incredible emotional tragedy ultimately of her life is this story that people want to revisit. She's a subject in which people are so fascinated that there is this curiosity and and I think that kind of drives sort of ongoing research. People throwing different methodologies to try and answer a question that has been thought about for hundreds of years.