Abstract

Scientists must sometimes choose between competing definitions of key terms. The degree to which different definitions facilitate important dis- coveries should ultimately guide decisions about which terms to accept. In the short run, rules of thumb can help. One such rule is to regard with suspicion any definition that turns a seemingly important empiri- cal matter into an a priori exercise. Several prominent definitions of eco- logical “stability” are suspect, according to this rule. After evaluating alternatives, I suggest that the faulty definitions resulted from an over- emphasis on population dynamics in community ecology. Machine met- aphors of nature may have given rise to a related problem of experimen- tal design.