For as long as there have been digital computers, American universities and federally funded agencies have helped to create a digital territory in which state-supported capitalists reap new fortunes. This is especially the case with medical research. Because tax dollars help fund research, researchers are legally required to make their data available to the public, including corporations. In effect, laws designed to encourage openness among federally funded researchers have made both researchers and private citizens vulnerable to the power of corporations who are accountable not to citizens, but to shareholders.
In short, our data crisis is not one of privacy, it’s one of piracy where governments and corporations aren’t in opposition. They are in a partnership.
The very agency we have charged with responsibility for our health — the National Institutes of Health — helped pioneer the practices that would make personal information economically powerful. In 1965 members of an indigenous group in the American Southwest became subjects in a NIH study aimed at understanding diabetes. These residents of the Gila River Reservation suffer from some of the highest rates of diabetes in the territorial United States. For more than 50 years, nearly all residents (who refer to themselves as Akimel O’Odham) agreed to have their personal medical information collected in the hope of helping researchers learn about diabetes. In fact, much of what biomedicine knows about diabetes is a result of the cooperation of those who the U.S. government referred to in the 1990s as “pathfinders for health.”
However, Akimel O’Odham neither consented nor imagined that their information would be used to fuel the emerging data market. By the 1980s, however, the research that NIH collected — while anonymized — was being used to develop a new kind of computer algorithm, the kind that would make Facebook possible.
How did this happen? Because, like the relationship of Stanford to Silicon Valley, NIH — headquartered in Bethesda, Md. — forged a close relationship with the nearby Johns Hopkins University and its privately funded Applied Physics Lab — the mission of which is to “provide solutions to national security and scientific challenges with systems engineering and integration, research and development, and analysis.” So it was easy for biomedical researchers to share data with nonmedical computer scientists.
Indeed, it seemed important to do so. Such collaborations offered the potential of using new computing technology to help predict the onset of debilitating and expensive conditions such as diabetes. Maybe more importantly, NIH was legally required to do so free since the data was collected by the federal government, with taxpayer dollars.
Public-health officials saw the Gila River Indian Community as a natural laboratory to experiment not just with solutions for diabetes but also with personal information. The medical information collected from their bodies — information about number of pregnancies, blood pressure, weight — became a valuable resource for improving and even defining general knowledge about the disease. In the process, however, it was translated into digital form that would enable it to be reused for knowledge projects unrelated to diabetes or even biomedicine.
As a result, before it was made unavailable to the public last fall for “permissions restrictions,” the “Pima Indian Diabetes Data Set” was cited nearly 100,000 times for uses such as predicting when fires might erupt in New York City’s underground power grid and teaching students in introductory computer science classes how algorithms work. The problem: Akimel O’Odham had no idea that information about their intimate details were helping others make money and fuel entire new industries. To make matters worse, they saw no improvements to their health and continue to struggle to find the resources to heal themselves.
The case of the Pima Indians diabetes data set is just one among hundreds, if not thousands, of examples of how citizens who put their trust in state-sponsored institutions were also unknowingly supporting the growth of industries they have come to fear. The experience of Akimel O’Ohdam is especially egregious because this community was first dispossessed of their land and then of their personal information. They are pathfinders for the health, and perhaps wealth, of others.
Facebook users have been horrified to find out that they have been cast as pathfinders for Cambridge Analytica and other companies that reuse their data without their knowledge. This formula, however unsavory it may seem, is not new. On the contrary, it’s what happens when the mantra “information wants to be free” leads common resources to be claimed for capitalism.
Although using biomedical data to help predict when fires might erupt or to teach students how algorithms work may seem relatively benign, compared with the Cambridge Analytica data breach, the point is that the same circumstances that led indigenous peoples to be dispossessed first from their land and then from their data structure our data economy more broadly. To move beyond the false notion of a battle between the state and the corporation, we must recognize their shared history of theft. This is, I’m sorry to say, the American way.
Debates over the neutrality of Facebook as a platform are a distraction from the fact that big data is anything but. We must go far beyond creating better user agreements that will be broken as easily as treaties with indigenous people have been, and grapple with the broken culture that fosters data reuse. The only information that Facebook, and for that matter, Congress, deserves is a lesson on the history of how they made data reuse not only possible but also legal. I’m available to provide that history. And I expect to be fairly compensated for the data I share.