When Small first entered the cemetery in the summer of 2012, she burned sweetgrass – a plant with spiritual significance in native cultures. “The sweet grass brings the spirits in, wakes them up,” she said. She spent her early days walking through the rows, cross-referencing a list of cemeteries with the names carved into each tombstone. One day at dusk, as she reached the gate on one side, she stared at the horizon. The sun was setting and Small’s eyes followed the long shadows reaching back to the school. All the graves, she noted, were laid out according to Christian custom with their feet facing east—a blatant disregard for the multitude of burial practices and belief systems held by different tribes around death.
“I got super emotional,” Small recalls. ‘I couldn’t write anymore, couldn’t concentrate anymore – because there were so many of them. And a lot of them were babies. Many of them were sisters and brothers. I saw the Davis family name in there three, four times and I thought, ‘You wiped out an entire family! A generation.’ It was just breathtaking.” She walked to her car and sat silently behind the wheel.
After a while a train rumbled past the cemetery. She got out and headed for the railway—the same line that would have brought children to Chemawa 100 years earlier. “I was trying to focus on that moment,” Small explained. “The horror of it, the unfamiliarity. Maybe even, for some, the excitement of doing something new. She bent down and touched the cool steel of the rails with her cheek.
By the time Small had used the GPR machine at the cemetery for a few days, she felt transfigured with a sense of calling. As she stood among the graves of children who never returned home, she felt there was important work to be done, work she knew she could do if she continued to persevere. “I felt I had found my place in the whole spirit of things,” she said. “Not just the world, but in the universe.”
But she still had a lot to learn and there were few clear paths to professional enlightenment. Usually used as a tool to study groundwater, soil and bedrock, ground penetrating radar was first used by a researcher in 1929 to measure the depth of a glacier in the Austrian Alps. The technology is commonly used today to identify underground utility lines. Both utility lines and graves are dug in places with histories of different uses, each leaving their own traces underground, but because utility trenches differ so much from the surrounding soil and contain metal pipes, water-filled plastic, gravel or sand, they are easier to identify.
Any abnormality — an air pocket, a layer of soil that retains moisture differently from the surrounding environment — can appear as a visual aperture (as soft tissue can be nearly invisible on an X-ray) or as a solid. , a point of light, like a hard drive going through an airport baggage scanner. Modern data processing software can help, but underground exploration can still be a tedious, often ambiguous process.
When Small submitted a partial study of the Chemawa cemetery for her master’s thesis comparing the location of graves and grave markers, she also shared some of her GPR images with the company that supplied the machine. She hoped for confirmation. Instead, an anthropologist there who works on forensic applications of GPR politely explained that Small’s footage didn’t necessarily show graves where she said it did. She realized she had been badly misled when she conducted her survey and interpreted the data. She had done most of her fieldwork unsupervised, and no one in the state of Montana had direct experience using GPR in this way. “It was beat, really beat,” said Small. “At the time I thought you could see bones with that damn thing.”
But Small didn’t give up; even as she began her PhD program, the calling to get reliable data on Chemawa stayed with her. Realizing that she “needed someone to teach me nuclear-level GPR,” she found her way to Jarrod Burks, an archaeologist who lives in Columbus, Ohio, and conducts surveys for the Defense POW/MIA Accounting Agency on recovery missions for missing soldiers. He agreed to join her promotion committee. In 2017, Small invited Burks to help produce a new report on Chemawa. After five days of painstaking work at the cemetery, the new data Burks and Small gathered revealed where she had gone wrong. He confirmed the fundamental limitation of Small’s earlier analysis: tree roots and grave shafts can look alike in raw radar data, and Small had neither the experience nor a large enough data set to tell the difference. “Marsha, I don’t see any graves here,” Burks said, pointing to where she thought there might be a few.