The body washed ashore in a hurricane: Middle-aged male, pristine Terra work boots, tattered Urban Heritage jeans, face lost to decomposition and surf.
The location provided few leads.
Sandy Cove Beach, N.S., is situated along the Bay of Fundy’s southern shore. Powerful tides there are known to suck in flotsam from as far away as Boston before belching it back out into the Atlantic. With Hurricane Dorian’s churning winds added to the mix, it was impossible to say where the man had come from.
When nobody from the tight-knit Digby area came forward last fall, investigators began looking more widely.
But as winter descended, the identity remained an enigma, and Digby Man was added, alongside 714 other entries, to the RCMP’s national database of unidentified remains.
The database is a storehouse for some of the country’s most vexing cases. Many show signs of violent ends -- bullet holes, broken limbs, cracked skulls. But the entries consist largely of assorted bones and clothing fragments, not enough to create the facial reconstructions necessary for issuing public appeals.
To get anywhere on Digby Man, investigators needed a face. But how?
Earlier this month, the RCMP turned to art students and a world-renowned forensic imaging specialist -- a conjurer of lost souls -- for an answer. The force sent an industrious Mountie and 15 skulls to the Manhattan-based New York Art Academy in a last-ditch effort to solve some of the country’s toughest cases.
Within two weeks, the identity of Digby Man would be solved by means that show both the possibilities and limitations of the RCMP’s efforts.
“A face is just so vitally important to these cases,” said Corporal Charity Sampson, the RCMP identification specialist who accompanied the skulls to New York. “Without good facial reconstruction, they may be lost forever.”
The idea emerged from a class Ms. Sampson took last summer on facial imaging. Joe Mullins, the instructor, had earned headlines in previous years for conducting a one-week workshop at the New York Academy of Art where students performed facial reconstructions on unidentified skulls held by the city’s chief medical examiner. In just four years, the students, who are classically trained in anatomy, had been so successful reconstructing faces that they’d nearly cleared the city medical examiner’s backlog of unidentified skulls. That left Mr. Mullins with a quandary. He told Ms. Sampson that he needed skulls.
“I thought to myself, I bet I can get Joe some skulls,” she said.
The RCMP normally employs three forensic artists, but this would be a rare opportunity to complete 15 faces in five days. Over the next four months, Ms. Sampson overcame a series of bureaucratic hurdles. She got buy-in from the RCMP, convinced the B.C. Coroners Service and the Nova Scotia Medical Examiner Service to put forward 15 well-preserved skulls and made nylon replicas of each one using a high-end 3-D printer at the National Research Council (handling and transporting real skulls has legal and ethical restrictions).
In the first weekend of January, Ms. Sampson hopped aboard an RCMP plane to New York City, more curious than ever about the backgrounds of the 15 skulls in her luggage.
“I have really thought of nothing else for five months,” she said shortly after landing. “I’m so excited to see faces on them. I have been looking at them a long time. A long time.”
Her thoughts kept wandering back to one skull in particular, Digby Man. “That’s the one that intrigues me most,” says Cpl. Sampson. “I’m from Nova Scotia. It’s the most recently recovered one. Memories are still fresh. It’s a good time to put a face on him and get him out to the public.”
The art students got their skulls on a Monday.
Kelly Robert, an MFA student with more than 20 years of experience in jewelry production, rubbed a tattooed forearm in nervous anticipation. When Mr. Mullins, the instructor, finally handed her a skull, she expelled a long “wooooooow” as she stared into the hollow eye sockets. “Oh wow.”
Ms. Robert had done the class before. Her art tends toward more abstract sculpture, but she returned for the sense of altruism in the workshop. Her skull this time around, that of a white or Indigenous man discovered in Vancouver in 1989, had no teeth. “It’s like the mouth is pulling in the rest of the face,” she said.
Other students pored over the scant details that came with their subjects’ back stories.
One skull had been recovered near a chairlift on Whistler Mountain in 1987 with a clear bullet exit wound in the cranium. Another four came out of B.C.’s Fraser River between 1972 and 2008. One skull was discovered in 40 feet of water accompanied by underwear labelled Edmonton Psychiatric Center.
“These are lost, lost souls,” Mr. Mullins told his students. “There are family members out there frozen in uncertainty. Hopefully you can help answer some questions.”
Mr. Mullins took on Digby Man.
Though computerized methods of facial reconstruction exist, Mr. Mullins prefers lower-tech tools: molding clay (200 pounds by week’s end), superglue, marbles, plastic straws, cheese cutters and assorted other sculpting tools.
“Even with a computer, it’s not like CSI,” he said. “There’s no instant add-face-to-skull button. It takes time.”
Despite the setting, this is not an art class, as Mr. Mullins continually reminds his students. “Leave your artistic license at the door, “ he warns. “You do not have it. There is no room for interpretation. You have to put the right face on.”
Most of the skulls come with detached jaw bones. The first order of business is attaching them using cotton balls, generous amounts of superglue and a dab of forensic humour. “Make sure you don’t get any cotton in your external auditory meatus," Mr. Mullins says. “Better known as your ear hole.”
He teaches according to the Manchester Method of forensic facial reconstruction, which puts an emphasis on facial muscles and soft tissue thickness to accurately gauge facial proportions. It’s a science, but an imperfect one. In studies where subjects have to match a reconstructed face with an original, they generally pick the right one 70 per cent of the time. A 2006 study that compared two skull reconstructions to their original faces using CT imaging found that 67 per cent of the reconstructions were accurate to within 2mm. The tip of the nose showed the highest degree of error.
Once the skulls are mounted on adjustable stands, the students layer 11 muscles on either side of the face. First is the temporalis, or temple. Last is the zygomaticus major, or the smile muscle. At this point, the skulls look somehow undignified, less human and more Terminator.
Next, students cut lengths of plastic straw coinciding with average soft tissue thickness at specific points on the skull. The pieces are depth markers guiding students on how thick or thin they should layer their clay skin. So adorned, the skulls acquire a distinctly spiky Hellraiser appearance.
As the week rolls by, Mr. Mullins gives demonstrations on eyes, ears, noses, lips and hair. The faces slowly come to life. The process has a profound effect on many of the students.
“It’s a little emotional,” says Anita Clipston, a Vancouver resident in the class working on a middle-aged Indigenous skull with dentures. She speaks in a respectful whisper, as if at a funeral. “When I went home last night I had two thoughts running through my head: Does this person have a family? And, if not, where do you have to be in life that you go missing and nobody is looking for you? In sculpting we’re used to working on generic skulls based on real ones, but this is a real life, a real person -- we don’t know who. I thought I’d feel more scientific about this, but I do feel this responsibility now.”
She opted into the workshop both because of a long-time fascination with shows like CSI and for the opportunity to restore a name to the nameless. She specifically sought out an Indigenous subject.
“This means a great deal to me,” she says. “I’m keenly aware from my First Nations friends of how many Indigenous people do go missing in Vancouver.”
Mr. Mullins supplies eyes in the form of clear white marbles. Students draw a circular iris on each one, 11.5 mm in diameter, with a dot in the middle for a pupil. They use a brown (the most common eye colour) Sharpie to make a wagon-wheel pattern around the iris. Suddenly, it’s like 15 new souls have just entered the room.
“At the beginning, all the skulls look similar,” said Ms. Sampson. “By the time the eyes and lips go on, there was true personality in the room. It was incredible.”
There’s a formula to placing most parts of the face. The tops of the ears, for instance, align with the eyebrow ridge and the lobes typically line up with the tip of the nose. The shape of the lobes is related to the shape of the mastoid process, that pointy part of the skull directly behind the ears.
The nose is more complicated. Mr. Mullins shows his students how to project its shape by following the paths of the nasal bone (which forms the bridge) and the nasal spine (the bony projection between the nostrils). The nasal spine is one of the most telling points on any skull. It acts like an arrow to identify a nose that points up, down or straight ahead, often one of the key defining features of any face.
This little information-rich nub also happens to be one of the most fragile parts of the skull. “You can flick it with your finger and snap it off,” says Mr. Mullins. “Of the hundreds and hundreds of skulls I’ve done, it’s very rare that I get a good nasal spine. It’s one of the most elusive pieces of the skull.”
The Digby skull has no nasal spine, so Mr. Mullins has no choice but to give it a straight-ahead, generic nose.
By Thursday morning, Digby Man is nearing completion. The remains discovered on a Nova Scotia beach last year now has a face. Digby Man is handsome, with a tall forehead and sad eyes. Still, what happens next is anyone’s guess.
Mr. Mullins has worked on hundreds of reconstructions that have led to dozens of identifications, though he doesn’t know exactly how many. He can never guess which ones will get solved.
“Remember, this is a last-ditch effort,” he said. “Nothing is happening with these cases otherwise.”
The Academy workshops have led to at least four positive identifications. In a previous class, Kathleen Gallo reconstructed the skull of an apparent migrant border crosser whose body was found in the desert of Pima County, Ariz. The man was identified shortly after Ms. Gallo’s finished sculpture went public and his remains have since been returned to his family. “After that, I was hooked,” says Ms. Gallo, who took the workshop again this year and is pursuing forensic reconstructions as a career. “Not only is it an artistic workout, but it’s a mental and ethical workout as well.”
The RCMP uploaded all the faces to canadasmissing.ca on Jan. 13. Solid tips began to trickle in. For a positive ID to be made, someone would have to come forward linking the face to a name. Dental and DNA work would then be conducted to confirm the match. “You just need the right person to see that face,” says Cpl. Sampson, after returning to Ottawa. “It may not be today or even this year. At some point, the right person will see someone they love in this database and the link will be made. The important thing is they now have a face.”
One week later came a bombshell in the Digby case. Nova Scotia RCMP announced they’d identified the remains. There was a caveat: The reconstruction played an indirect role, at best.
Digby Man was actually Brent McLellan, a 43-year-old Saint John man who’d leapt from Reversing Falls Bridge the previous summer. A tourist’s photograph had provided confirmation of the death, but the body wasn’t recovered at the time.
He’d been a star athlete and belonged to the Saint John Sports Hall of Fame through his membership on 2001-2002 Saint John Alpines, winners of the 2001 Canadian Senior Baseball Championship. More recently, however, he’d struggled with a bipolar diagnosis and addictions issues, his mother said. The family held a memorial mass in July. Eight hundred people showed up, but it did little to comfort his mother.
“The whole time I’ve been thinking about Brent in that cold water,” Marjorie McLellan told The Globe this week. “That’s not a good thing for a parent to be thinking about every single day.”
The day the workshop started, RCMP headquarters promoted the program on its social media feeds. Nova Scotia RCMP added a picture of the Terrra work boots from Digby Man. A friend of Mr. McLellan’s saw the photo and notified police, saying she’d been with Mr. McLellan when he bought the boots for a job at a graveyard. They had a DNA match within days.
“You will never know how good we felt that day when the DNA came back positive,” said Ms. McLellan. “Oh, it was just wonderful.”
She was less thrilled about her son’s reconstruction. It lacks perhaps the defining feature of Mr. McLellan’s face, an upturned nose. “I was disheartened when I first saw it because it didn’t look anything like Brent to me,” she said. “My girls looked at it and disagreed. They said if you look at the ears and the eyes it’s him.”
The McLellan case offered Mr. Mullins a rare opportunity to compare his handiwork to the source material. He understands Ms. McLellan’s reservations, but says the absence of a nasal spine limited what he could do with the nose. “Without that nasal spine, the only thing you can do is build a straight-out nose,” he said. “That’s the only choice you have.”
Upon review, though, he says the work stands up. “When I did a side-to-side comparison, I felt pretty good,” he said. “The proportions are there. The stature is on point. Everything lines up where it should be. Given the information I had, it was the best face I could do.”
Without the publicity surrounding the workshop, it’s difficult to say whether the match would’ve been made. But the details don’t matter to Ms. McLellan. Any method that provides even a remote shot at recovering lost souls and providing relief to grieving families is worthwhile, she said. “This method maybe didn’t work for Brent, but I know it will work for others,” she said. “It was euphoria knowing that Brent wasn’t in that cold water anymore. Other families deserve that feeling.”