Risk and Quantification: A Linguistic Study

AbstractIn risk analysis and research, the concept of risk is often understood quantitatively. For example, risk is commonly defined as the probability of an unwanted event or as its probability multiplied by its consequences. This article addresses (1) to what extent and (2) how the nounriskis actually used quantitatively. Uses of the nounriskare analyzed in four linguistic corpora, both Swedish and English (mostly American English). In total, over 16,000 uses of the nounriskare studied in 14 random (n = 500) or complete samples (wherenranges from 173 to 5,144) of, for example, news and magazine articles, fiction, and websites of government agencies. In contrast to the widespread definition of risk as a quantity, a main finding is that the nounriskis mostly used nonquantitatively. Furthermore, when used quantitatively, the quantification is seldom numerical, instead relying on less precise expressions of quantification, such ashigh riskandincreased risk. The relatively low frequency of quantification in a wide range of language material suggests a quantification bias in many areas of risk theory, that is, overestimation of the importance of quantification in defining the concept of risk. The findings are also discussed in relation to fuzzy‐trace theory. Findings of this study confirm, as suggested by fuzzy‐trace theory, that vague representations are prominent in quantification of risk. The application of the terminology of fuzzy‐trace theory for explaining the patterns of language use are discussed.