“Futurology is almost always wrong,” the historian Judith Flanders suggested to me, “because it rarely takes into account behavioral changes.” And, she says, we look at the wrong things: “Transport to work, rather than the shape of work; technology itself, rather than how our behavior is changed by the very changes that technology brings.” It turns out that predicting who we will be is harder than predicting what we will be able to do.

Take away the object from the historical view, and you lose sight of the historical behavior. Projecting the future often presents a similar problem: The object is foregrounded, while the behavioral impact is occluded. The “Jetsons idea” of jetpacking and meals in a pill missed what actually has changed: The notion of a stable career, or the social ritual of lunch.