Bone scan databases offer scientists new ways to study human remains. But some worry they could be misused.
Ten years ago, it wasn’t possible for most people to use 3D technology to print authentic copies of human bones. Today, using a 3D printer and digital scans of actual bones, it is possible to create unlimited numbers of replica bones — each curve and break and tiny imperfection intact — relatively inexpensively. The technology is increasingly allowing researchers to build repositories of bone data, which they can use to improve medical procedures, map how humans have evolved, and even help show a courtroom how someone died.
But the proliferation of faux bones also poses an ethical dilemma — and one that, prior to the advent of accessible 3D printing, was mostly limited to museum collections containing skeletons of dubious provenance. Laws governing how real human remains of any kind may be obtained and used for research, after all — as well as whether individuals can buy and sell such remains — are already uneven worldwide. Add to that the new ability to traffic in digital data representing these remains, and the ethical minefield becomes infinitely more fraught. “When someone downloads these skulls and reconstructs them,” says Ericka L’Abbé, a forensic anthropologist at the University of Pretoria in South Africa, “it becomes their data, their property.”
Digital bone repositories already exist around the world, and while viewing those bones in a computer environment is often an option, most such repositories keep the underlying data — which could be used to print new, physical bone replicas — private. The repositories that do make the data open access typically only include human remains that are older than 100 years because of the legal issues surrounding the potential to identify a person from their remains, as well as the value of the data their remains might yield.
Such are the vagaries now facing L’Abbé and her colleagues, who launched a project called Bakeng se Afrika, or For Africa, in 2018. It is a collaboration with local and international partners to digitize about 700 recent, identified skeletons from university collections around South Africa. The scans will go in a single repository, and it is hoped that research arising from the new database will have immediate social applications. For instance, L’Abbé works with police to help identify people whose bodies are discovered in Gauteng province in South Africa, which is home to the country’s administrative and economic capitals, Pretoria and Johannesburg, respectively. More than 20,000 people are murdered in the country each year, with a disproportionate number in Gauteng.
The team hopes to share their data with researchers from other institutions in different countries, but they will need to resolve the ethical and legal concerns before their digital repository goes live. Sorting out what’s right and what’s wrong, L’Abbé concedes, won’t be easy: “What are the ethics,” she asks, “behind printing a bone and using it?”
Skeletons are fundamental tools for teaching anatomy, as well as researching human diversity. Understanding human variation can also help a variety of researchers and clinicians. In addition to forensic science, for instance, dentists can use data about cranial variation to improve the fit of dental implants; plastic surgeons can review studies of the average ear or nose for specific populations to help reconstruct a face after an injury; researchers can create databases to study bone abnormalities; and prosecutors have used prints of human bones in courts to illustrate how a person died.
Digital databases — which may have just a small number of bones from a private institution or as many as hundreds in larger collections — advance the relevant fields even more. And with the evolution of new scanning and viewing technologies, researchers are able to look inside bone images and manipulate them in ways that would not be possible with the real thing. Bone-scanning technology may also preserve remains for future generations, allowing researchers to avoid physical wear and tear on the originals. And for researchers in many developing countries, digital scans may be the only opportunity to research skeletal collections, as working on specimens in person is often limited by funding constraints.
But these repositories come with loaded ethical questions that have often kept access to human-skeletal data under lock and key. L’Abbé’s South African project isn’t the first to grapple with these issues, but the creators of other repositories haven’t managed to devise universal solutions.
Digitized Diseases, a collaboration between the University of Bradford, the Museum of London Archaeology, and the Royal College of Surgeons of England is one such database that launched in 2013 and showcases the variety of bone pathologies that can manifest in the human skeleton. But the collaboration had to suspend downloads from its repository after curators discovered that someone had downloaded a skeletal model and uploaded it into a publicly-available online 3D software platform — divorced from its original context.
“It is hard to police,” says Andrew Wilson, principal investigator on Digitized Diseases and a senior lecturer in forensic and archaeological sciences at the University of Bradford. “We have a real dilemma there. We want it to be open and freely available for education and research. That was the wish of our partners and funders.”
When the project launched, 3D imaging software and 3D printers weren’t as common as they are now, Wilson adds. These technological developments have made the team wary of posting additional downloadable data. Moving forward, Wilson says the company plans host their data directly on that same publicly-available online 3D software platform themselves and link directly back to Digitized Diseases, in order to “retain a degree of visibility of the resource that people can link back to.”
Duke University’s MorphoSource, an online library of skeletons where researchers can share data, avoids some ethical dilemmas by only containing human skeletons when all of the consent documentation is in place. And MorphoSource curators don’t upload data themselves: They oversee the infrastructure that allows any researcher or museum curator to upload and share data, or to simply use the site as an archive for private data. Bioarchaeological remains, which are older than 100 years, are “explicitly not made publicly available,” says Julie Winchester, co-director the project. A member of the public will not be able to find them in a search of the site. “We would additionally not allow the publishing of any human remains (regardless of age) that could be identified to an individual or whose relatives could be identified.”
Meanwhile, John Kappelman, an anthropologist at the University of Texas at Austin, has set up numerous sites where it is possible to view human remains, such as eanthro.org and eskeletons.org. But none of it is available for download. Kappelman says his team has “no idea or any way” to find out the identities of the original skeletons. “Most of what we have here, it’s been here for pushing on 100 years,” he says, “so I don’t have any good provenance on that.”
Like Wilson, Kappelman says he is unsure how one would manage those who abuse online repositories: “I don’t think there is any perfect way to police it, same as any kind of hacking.”
Keeping data safe is something the South African team will have to navigate. L’Abbé says that the new South African digital repository, for which her team received $1.1 million at the beginning of 2019, will serve as research infrastructure. The collaboration includes Sefako Makgatho Health Sciences University and Stellenbosch University, both in South Africa, the South African Nuclear Energy Corporation, the University of Bordeaux in France, the University of Coimbra in Portugal, and K.U. Leuven in Belgium.
Initially, the new digital collection will include three types of bones most commonly used in research and forensics — bones from the forearm (radii), the thigh (femurs), and skulls from identified cadavers of colored, black, and white South Africans. (“Colored” is one of four officially recognized racial groups in the country — put in place during apartheid — along with black, white, and Asian/Indian.) The team plans to scan more people and bone types as the project progresses. But creating the database will be slow going. It can take a full weekend to capture a single adult cranium using the group’s scanning equipment, called a micro-CT.
While the database will widen the access to the bone collection, it may also erode restrictions on how the bones are used. In order to access real human bones, researchers in South Africa require ethical clearance from their institution, but scientists from other countries do not.
Researchers will have to follow the same procedures to access the digital repository, which will be housed and backed up on servers at the University of Pretoria. But once the data are downloaded, the original curators cede control. Researchers who work in countries with less stringent ethical reviews could then use the skull images in research that is considered unethical in South Africa.
One example is race science, which is premised upon the scientifically-invalidated idea that humans can be broken into distinct racial categories, and that these differences can be measured. Race science is a sore topic in South Africa, where it underpinned the policy of apartheid, in which a person’s race determined where they lived, their employment, and who they could marry. In general, race science is deemed a morally and scientifically unacceptable area of study in South Africa.
Still, while plenty of evidence counters the core claims of race science, the field has resurged in recent years, which worries L’Abbé: “What if someone prints 20 skulls of people from South Africa — 20 skulls of black people?”
She also worries that someone could take data from the repository, print copies of a person’s skull, and sell it. “In South Africa, we need to have an ethics constitution, so they can’t download it and sell it,” she says. “The National Health Act” — which lays out the rules for the country’s public health system — “doesn’t touch on what you do with an image of something.”
The issue extends beyond South Africa. There is no way to enforce any country’s legal and ethical framework on researchers in other countries. And the laws on human remains are uneven. In the United States, for example, the only federal law regulating the sale and ownership of human remains is the Native American Graves Protection and Repatriation Act, which has criminal penalties for trafficking in Native American remains. A first-time offender can face a fine of up to $100,000 and a one-year jail sentence. But many U.S. states allow commerce in other human remains, thanks in part to a major shortfall in American legislation: There is no law articulating whether human remains are people or objects.
Meanwhile, in South Africa and the United Kingdom, researchers require special dispensation to work on recently deceased human remains, according to the National Health Act and Human Tissue Act, respectively. There is also a distinction between recent human remains, which fall under laws governing forensics, and older remains, which are regulated by archaeological heritage laws.
Despite the legal variations, countries have a long history of transporting human biological material across borders, whether it is blood for research or bodies for repatriation. Digital copies, on the other hand, are difficult to regulate because they don’t have to go through official channels and could be emailed to anyone anywhere in the world.
For a digital image, “even if you ask conditions to be met, it is out there,” says Francisca Alves Cardoso, a biological anthropologist at the Center for Research Network in Anthropology in Lisbon, Portugal, who is investigating the ethics surrounding printed human bones. “You are not able to control it.”
Digitized databases of human remains, or artifacts made from those remains, have potential ethical problems beyond data access: The databases’ potential source materials may have complicated pasts. Scientists have been using and replicating human skeletons, faces, and bodies for hundreds of years, including resin castings, which were popular in the 19th century.
But throughout this history, the question of consent has plagued the fields that most commonly use such replications of human remains, including anthropology, archaeology, and medical science. Consent is one of the cornerstones of biomedical ethics, and affects any field that works with human remains.
The issue is particularly prominent in South Africa, where science — and anthropology in particular — has a history of unethically representing people and their remains. And today, South African curators are unsure of what to do with several types of relevant artifacts, including life and death masks, which are plaster casts made of people’s faces that were once used to show physical differences between races. Some of these masks are on display in the Wits University Hunterian Museum in Johannesburg, although parts of the collection, including the death masks, are in a restricted area accessible only to medical students and museum staff.
Some of the casts came from unidentified bodies in the Johannesburg mortuary, without the necessary ethical clearance. It is possible that family members could, in fact, recognize lost loved ones in these casts.
Other examples include the “Bushman” lifecasts, once thought to typify the country’s first inhabitants, which have been in storage at Cape Town’s South African Museum since the early 2000s, after an outcry that the casts reinforced racist stereotypes and impugned the dignity of this group of people. The term “Bushmen” has a fraught history and is sometimes considered a pejorative, but there is no other collective name for the diverse tribes that make up the region’s first people. Other remains from these early tribes have similarly difficult pasts. It took years for South Africa to reclaim and inter the remains of Sarah Baartman, a young Bushman woman whose body was displayed as a curiosity in Europe in the early 19th century — both in life and death — because it allegedly typified a Bushman’s physique. Baartman, who became known as the “Hottentot Venus,” traveled Europe as a stage curiosity in part because of her large buttocks. When she died in 1815, her manager, without her prior consent, sold her body to scientists at the National Museum of Natural History in Paris. Until the 1970s, the museum displayed a cast of her body, skeleton, and preserved brain and genitals.
There is also great sensitivity in South Africa, given its history of race science, around the provenance of skeletal collections. While the 700 skeletons earmarked for digitization in L’Abbé’s project are from identified people, the country’s museums have many skeletons of unclear or dubious origins. Today, some institutions are working to correct these past errors. The University of Cape Town, for example, is identifying pieces from its skeletal collections that were not collected ethically and negotiating with the respective communities to repatriate the remains.
But repatriation takes time, and even if some researchers are trying to remedy past mistakes, it isn’t clear what will happen if the remains are scanned in the meantime. “What if someone wants to 3D scan Bushman remains from the McGregor Museum?” L’Abbé asks. The museum has acknowledged that some of the skeletons in its collection may have been unethically obtained. “What does it mean if you print a skull of someone who was classified as Bushman?” L’Abbé adds. “What will someone do with that print? What will the community think?”
And the issue of data possession leads to questions of ownership and power. “Who owns the right to print which bones?” asks Brenna Hassett, an archaeologist at University College London in the U.K. “Is it even possible to own the rights to part of a person?”
What happens, for example, if people build a business around selling 3D models of a leg from a skeleton that was freely available online. And then later, those people “find that skeleton should not have been published in the first place, because it was stolen from a cultural group with deep sensitivities to the representation of the dead?” Hassett wrote by email. “Where would they stand?”
At the heart of this issue is whether the underlying data and the resulting images are the same as human bone. According to research from Cardoso and Vanessa Campanacho, a bioanthropologist at the University of Coimbra in Portugal, many researchers, often from Western countries, argue that the data and source material are distinct — that a print of a bone is not human biological material, although the latter should still be treated with respect. But other groups, including aboriginal communities and indigenous groups, strongly disagree, and consider data and replicas to be part of the deceased individual’s personhood.
Meanwhile, researchers have been digging into the abilities and limitations of 3D scanning technology, as well as discussing the possibility of a digital database of skeletons that would be open-access and available to anyone. But while “there is much focus on the potential” of the technology, says Priscilla Ulguim, a bioarchaeologist at Teesside University in the U.K., most of the research has bypassed ethical questions.
“How we share these 3D models and how we communicate is just as important as the technology we’re using to create them,” Ulguim adds.
For example, in a 2018 paper published in Archaeologies, the authors point to a 2015 study that showed it was possible to create high-resolution models of Native American remains prior to repatriation, and also “ignored arguably the more pertinent question of whether these digital data [should] be created and stored.”
Other researchers are exploring how society feels about 3D printed human remains.
“I’m often told the general public thinks this, says this, but then there is no communication,” says Campanacho. “We don’t know what the general public thinks.”
Campanacho is also working with Cardoso to investigate attitudes in Portugal and the U.S. via online surveys and social media. In Portugal, where the team has so far received about 310 responses, “the majority would be okay having their own skeletons and those of their family members digitized after death, but there may be some conditions,” Campanacho says. Respondents, she says, also “seem positive” about the creation and sharing of 3D digital models.
In a study of attitudes in the U.S. toward 3D bone replicas, the team says their preliminary data — which includes about 230 participants — suggests that many people would be happy to have their remains digitized and printed, with one exception. “Regarding Native American human remains, the response was negative,” Campanacho says. Respondents felt these remains “should never, under any condition, be 3D digitized.”
As is often the case with fast-moving technology, the law, ethical standards, and research practices haven’t had time to catch up with 3D printed human remains. While L’Abbé fears that her team’s new digital repository could open the door for illicit or unsavory commercial endeavors, how her team addresses the ethical dilemmas in South Africa may show one possible path forward. Still, the answer in South Africa won’t necessarily translate to other skeletal collections across the world.
“The circumstances of each skeletal collection are very different,” says Gwen Robbins Schug, a professor of anthropology at Appalachian State University. “People create and share models for such a variety of reasons that it would be difficult to make a blanket statement about what is and what is not ethical.” The ethical considerations will also differ between communities and countries.
But the advantages of sharing information — including bone data — between countries are significant. Many of the researchers involved in the work so far are acutely aware that dealing with human remains, or even the data derived from human remains, comes with a heavy responsibility. “We always have to treat human remains with respect,” says Ulguim. “They are individuals, people. They are someone’s mother, someone’s father.”
One idea for ensuring consistency for dealing with digitized remains is forming a consensus set of ethical guidelines. Such guidelines are still some way off, but researchers at the British Association for Biological Anthropology and Osteoarchaeology have made a start by drafting rules they consider essential. And a resolution on digital bioarchaeological ethics, adopted by the Eighth World Archaeological Congress in 2016, lays out suggestions for how to deal with digital collections.
For new remains, one important step would be to include digitization on consent forms for people who want to donate their bodies to science. But that isn’t possible if the human remains are older. For Ulguim, context is vital. Sometimes it is not possible to engage with relatives or communities because researchers don’t know who they are, she says. In this case, the best thing to do is to contextualize the research and try to account for the lack of consent. “Talk about your site, how you find it, the associated materials, give context about what you’ve found in your research.”
Whether the data is of the recently deceased or a body that is centuries old, researchers require a strong rationale for wanting access to digital repositories of human remains or being able to print the bones, something which would be determined on a case-by-case basis. “Why would you want a skeleton of a person, a replica of a real person in your house to hang hats or coats on?” asks David Errickson, a forensic archaeologist and anthropologist at Cranfield University in the U.K. “There has to be a real scientific reason. I think it comes down to justification. Why are you doing what you’re doing? Do you have permission to do this?”
Curators of repositories should require more control over the datasets, Errickson suggests, “not like Big Brother, but some kind of data string, a chain of custody like we would have in forensics.”
This is MorphoSource’s solution. The library allows researchers to track who views and downloads their data. Researchers have complete control over the data, and the onus is on them to ensure that they have the ethical permissions to publish it. “We encourage the contributor to set access levels such that they feel comfortable with the risks incurred,” says founder Doug Boyer, an associate professor of evolutionary anthropology at Duke. If they’re publishing datasets, he adds, “it means they’re taking on the risks that someone may use the data for something they were not intending it be used for.”
For L’Abbé, her work wrangling with the ethics of the digital repository is just beginning. In South Africa, this is a nascent field with little written on the subject, so she is building the repository’s constitution from scratch. Her own research will have to wait, she says with a sigh, since the constitution will take time to complete: “I’ll spend the next year writing it.”