COMMENTARIES Beyond the Fundamental Attribution Era?

Since Jones and Harris (1967) first introduced it to the published world, the bias that Sabini, Siepmann, and Stein (this issue) refer to as the fundamental attribution error (FAE) has been the catalyst for a tremendous amount of research. Around the time of its 13th birthday (an age of particular significance in some cultures), Jones (1979), beaming like a proud father, characterized it as a "remarkably robust and easily replicated phenomenon" (p. 107). Almost a dozen years later, Jones (1990) called it "the most robust and ubiquitous finding in the domain of interpersonal perception" (p. 164). This important, influential concept is now in its mid-30s, and one might expect it to be cashing in its stock options and retiring to the south of France, taking the occasional sommelier class to keep itself entertained and making special guest appearances with cognitive dissonance theory on the celebrity lecture circuit. But not unlike so many other similarly aged "celebrities," I fear, the FAE is not without its demons. Its discovery, if not its conception, was rather by accident, and it took a great deal of testing before many were convinced of its ability to survive long beyond its birth in the Jones and Harris laboratory. There has been almost unprecedented inconsistency in what to name it, enough to create anxiety and identity confusion in even the most robust adolescent. It took Jones about 20 years to settle on a name—the correspondence bias (Gilbert & Jones, 1986)—and by now it has been called, among other things, the overattribution effect, the overattribution bias, the overattribution-to-persons tendency, the observer error, the observer bias, dispositionalism, and, most dramatically, the fundamental attribution error. More important, the cohesiveness of its identity has been rocked by the observation, made most articulately by Gilbert and Malone (1995), that there really are multiple correspondence biases, each with its own set