President’s Corner: Obliteration by Incorporation
Seven decades ago, Robert Merton alerted social scientists that the application of their work in policymaking “has not been systematically reviewed and codified.” Consequently, he wrote, we don’t understand “its potentialities.” It was three decades later that an NAS committee was tasked to document who benefits from extensive public investments in social science. The committee concluded that there was little “systematic evidence as to whether these [investments] are having the results their sponsors hope for.” Another three decades passed and another NAS committee (that I chaired) picked up the thread: we were charged to explore “what is known about how social science knowledge is used in policymaking.” Its report, Using Science, delivered the same answer: we know very little. Meanwhile, our British colleagues launched a more ambitious study, only to conclude: “Yet the processes involved in social science research influencing wider decision-making have been relatively little studied in systematic ways, and are consistently under-appreciated by observers outside the academy.”
In a previous President’s Corner, I complained about this failure: “…[W]hy have we not yet gotten around to measuring trust in the social sciences? We know much more about how research on the natural and biological sciences contributes to society than how equally important social science does.” Then a colleague (and friend) said: “Ken, quit complaining, do some research.” I was shamed into action, and, fortunately, was rescued by an imaginative team at NORC. Together we have initiated pilot research projects that promise to mature into the kind of systematic study Merton urged. In fact, one of our projects is motivated by a Merton insight—obliteration by incorporation. Readers will recall that his term was designed to draw attention to a truth of science history: when they become generally accepted by fellow scientists, discoveries are no longer attributed to the original contributors. Discoveries live on, but discoverers fade from the record.
I take license with Merton’s phrase, stretching it to suggest that in social science, not only do contributors fade, but discoveries themselves are taken for granted. Consider, for instance: supply and demand, early childhood intervention, social capital, and peer pressure. These are social scientific discoveries, but they no longer belong to social science. They have become part of generally accepted common sense. Social science, including its influence on wider decision-making, is, as the UK study put it “consistently under-appreciated by observers outside the academy.” This distinguishes us from natural science. Dark Matter and DNA are science, not common sense; they are not at risk of being under-appreciated.
The NORC pilot project selected 50 concepts from the Oxford Dictionary of the Social Sciences, including the three mentioned above. Using Mechanical Turk, we asked respondents which of these terms they recognized, and, when recognized, to provide a working definition. We then used the dataset to examine concepts widely recognized by the public but not viewed as discoveries from social science, sometimes not even by advanced degree holders. The pilot project was sufficiently instructive to take the next step: we launched an NORC AmeriSpeak probability-based panel to more fully compare concepts that might be in different stages of obliteration. We are also interested to know if obliteration by incorporation varies across the disciplines—are psychology and economics more likely to be treated as science than sociology and anthropology?
A companion pilot project scanned various key data platforms, for use of the same 50 concepts. For example, Congress-Gov was scanned for those interested in policy; Wall Street Journal, Harvard Business Review, The Economist for business; Public Affairs Information Service for think tanks. As expected, most of the concepts appeared in high frequencies. More than 80 percent of the terms were noted 500 or more times (stretching back five decades). About 65 percent were referenced in federal bills that reached the desk of the president, with economic policy-based concepts, as we expected, frequently appearing in the media. The complexities of this dataset cannot be summarized in this brief note, though questions may already be coming to mind: under what circumstances do think tanks serve as the bridge linking basic social science discoveries to congressional debates? Can datasets with a long history tell us about the lag time between new concepts—early childhood intervention or deterrence theory—and their informed usage in public discourse?
These two pilot projects barely skim the surface, but the major conclusion is obvious: obliteration by incorporation is a real phenomenon. However, if its patterns are understood, it need not be an insurmountable barrier to social science reclaiming credit for significant achievements. I save my next (and last) President’s Corner to comment on how reclaiming credit can be furthered.
October 12, 2020