Graphs function plainly to summarize data. They hardly seem momentous. They are not like a famous discovery, whose significance is often marked by an eponymous name: Mendel’s laws, the Watson and Crick model of DNA, Darwinian theory. Who would name a mere graph? They seem mundane fragments of science, hardly worth celebrating. A notable exception, however, is the Keeling Curve (Figure 2.1). This simple graph depicts the steady rise in the concentration of carbon dioxide (CO2) in the Earth’s atmosphere over the last half century. It helps document how humans have transformed the atmosphere and, with it, the Earth’s temperature. The Keeling Curve is a linchpin in the evidence that humans have changed the planet’s climate. The Keeling Curve starts in 1958 and continues uninterrupted for over five decades. The scale of the data is extraordinary, an ideal rarely achieved in science. The hard data from real-time measurements show the steady accumulation of CO2 from burning fossil fuels. The results, presented in a simple yet striking visual format, serve to warn an energy-hungry culture of its environmental hubris. Although just a graph, it is monumental in scope and significance. The Keeling Curve, viewed in retrospect, raises an interesting question about how science works. How do such important long-term data sets emerge? Often we assume that scientific investigations find just what they intend to find. That is an implicit lesson of the tidy scientific method, as widely presented (see essay 5). But should we trust this sacred bovine? Could anyone have predicted this curve or its importance in advance? How did these important data originate? What happened before the graph was fully created? What happened, literally, ahead of the Curve? The Keeling Curve is named after its creator, Charles David Keeling. In the 1950s, as a handsome young man frequently enjoying the great outdoors (Figure 2.2), he hardly fit the stereotypical image of a scientist clad in a white coat, isolated in a lab. Indeed, with a fresh degree in chemistry, he turned down many job opportunities because he wanted to be closer to nature on the West Coast.
Stochastic Revealed Preferences with Measurement Error
