“When Emilia and I look at our elders in population genetics, there are very, very few women,” says Rohlfs. “But there were women and they were doing this work. To even know that they existed is a big deal to me.”
The project started with Hidden Figures, the 2016 movie about three black female mathematicians who helped NASA win the space race in the 1960s. After seeing the film, Huerta-Sánchez and Rohlfs felt surprised that they had never heard of its three protagonists. How many other historical female scientists were they similarly unaware of, they wondered?
One name sprang readily to mind: Jennifer Smith. Huerta-Sánchez remembered reading a classic, decades-old paper in which Smith was thanked in the acknowledgments “for ably programming and executing all the computations.” That seemed odd. Today, programming is recognized as crucial work, and if a scientist did all the programming for a study, she would expect to be listed as an author. “It was weird to me that Smith was not an author on that paper,” Huerta-Sánchez says. “[Rori and I] wanted to see if there were more women like her.”
The duo recruited five undergraduate students, who looked at every issue of a single journal—Theoretical Population Biology—published between 1970 and 1990. They pored through hard copies of almost 900 papers, pulled out every name in the acknowledgments, worked out whether they did any programming, and deduced their genders where possible. Rochelle Reyes, one of the students, says that she was “extremely motivated” to do this work, having grown up on stories of under-recognized pioneers like Rosalind Franklin, who was pivotal in deciphering the structure of DNA, and Henrietta Lacks, whose cells revolutionized medical research. “I was fortunate to grow up in a diverse environment with a passion for science as well as social justice,” Reyes says.
She and her colleagues found that in the 1970s, women accounted for 59 percent of acknowledged programmers, but just 7 percent of actual authors. That decade was a pivotal time for the field of population genetics, when the foundations of much modern research were laid. “Based on authorship at the time, it seems that this research was conducted by a relatively small number of independent individual scientists, nearly all of whom were men,” the team writes. But that wasn’t the case.
“It’s hard to know what sort of contributions people in the past have made behind the scenes,” says Jessica Abbott, a geneticist at Lund University. But this study “shows that it’s possible to get the right kind of data if you think creatively.”
Ezequiel Lopez Barragan, Rochelle-Jan Reyes, Samantha Kristin Dung, Andrea Lo?pez, and Ricky Thu present their work. Credit: Mayra Banuelos
Margaret Wu, for example, was thanked in a 1975 paper for “help with the numerical work, and in particular for computing table I.” She helped to create a statistical tool that scientists like Huerta-Sánchez still regularly use to estimate how much genetic diversity there should be in a population of a given size. That tool is called the Watterson estimator, after the 1975 paper’s one and only author—G. A. Watterson. The paper has since been cited 3,400 times.
Skeptics might argue that the programmers listed in these old papers were just doing menial work that wasn’t actually worthy of authorship. Rohlfs says that’s unlikely, especially in the cases of Wu, Jennifer Smith, and Barbara McCann, who were repeatedly name-checked in many papers. “They were doing work that was good enough that they were being called back again and again,” she says. The team even talked with William Hill, Smith’s former supervisor at the University of Edinburgh, who described her work as both technical and creative. (He didn’t, unfortunately, know where Smith ended up, and the team never managed to track her down.)
Afterwards, Wu didn’t consider trying for a Ph.D., although she told the team that “had someone suggested that I do it, I possibly would have found that an attractive idea.” She only got her doctorate in her 40s, after two decades working as a statistician and a math teacher. Now, she’s a faculty member at the University of Melbourne, where she develops statistical methods to analyze educational data. Wu didn’t return my request for an interview, but apparently has no regrets about the 1975 paper, Huerta-Sánchez tells me. She didn’t even know how heavily cited it had become. “She smiled,” says Huerta-Sánchez. “There was a little laugh. I felt like I was more upset than she was.”
“This is an opportunity for us to think about the norms we use in authorship and other metrics of academic success,” says Rohlfs. Even today, there are no clear rules about how much work someone must do to become an author. A professor could email some data to a colleague and become an author. A lab technician could do enormous amounts of labor, without which experiments could never be done, and be ignored. “There’s no standard, and surely the way we deal with authorship will be exclusive to some groups of people,” says Rohlfs. “If I look around at lab technicians, I’ll see a lot of women and people of color who aren’t being given authorship for creative work.”