The Half Life of Facts

Resumen del libro

Absorbing and approachable treatise on the nature of facts: what they are, how and why they change and how they sometimes don't (despite being wrong). Facts matter. But when they change--as they seem today to do with alarming frequency, we begin to lose that control. In his debut, Arbesman, a research fellow at the Institute for Quantitative Social Science at Harvard, advises us not to worry: While we can't stop facts from changing, we can recognize that what we know "changes in understandable and systematic ways." Since it is often surprisingly predictable, we can get a handle on change. "Facts, in the aggregate," he writes, "have half-lives: We can measure the amount of time for half of a subject's knowledge to be overturned." With this, he introduces "scientometrics," the science of science. With scientometrics, we can measure the exponential growth of facts, how long it will take, exponentially, for knowledge in any field to be disproved--say, 45 years for medical knowledge. We can understand predictably how the spread of knowledge (even incorrect knowledge) occurs, and we can understand that those abrupt disconcerting changes that seem to stand the world on its head aren't really all that surprising. Some readers may lose interest as Arbesman discusses such esoteric topics as logistic curves, linked S-curve theory, semantic and associative data processing and actuarial escape velocity. But like a good college professor, Arbesman's enthusiasm and humor maintains our interest in subjects many readers may not have encountered before. Does what popular science should do--both engages and entertains.