Appraisal Theory Overview, Assumptions, Varieties, Controversies IRA J. ROSEMAN AND CRAIG A. SMITH What is appraisal theory? In simplest form, its essence is the claim that emotions are elicited by evaluations (appraisals) of events and situations. For example, sadness felt when a romantic relationship ends may be elicited by the appraisals that something desired has been lost, with certainty, and cannot be recovered (Roseman, 1984; see, e.g., Frijda, 1986; Oatley & Johnson‐Laird, 1987; Scherer, 1993b; Smith & Lazarus, 1993; Stein & Levine, 1987). Appraisal theories may be contrasted with other theories of the causes of emotions. For example, it has been claimed that emotions can be elicited, without an intervening process of evaluation, by: (1) events themselves, as in stimulus‐response theories (e.g., Watson, 1919); (2) physiological processes, such as patterns of neural activity in the brain (e.g., Cannon, 1927) or peripheral autonomic activity (e.g., James, 1894); (3) facial or other expressions (e.g., Tomkins, 1962) or behaviors such as attack and flight (James, 1890); and (4) motivational processes, as in hunger eliciting an infant's distress (Tomkins, 1962) or the desire to intimidate an opponent leading an individual to get angry (Parkinson, 1997b). Questions That Appraisal Theory Was Developed to Address Appraisal theories were proposed to solve particular problems and explain particular phenomena that seemed to cause difficulties for alternative models (such as those just listed). Stated in the form of questions, seven of these problems or phenomena‐to‐be‐ explained are as follows. 1. How can we account for the differentiated nature of emotional response? Behavioral theories that dominated academic psychology from the 1930s through the 1950s tended to conceptualize emotion as undifferentiated, a dimension of behavior (emotionality) corresponding to its degree of energy or activity, which might reflect an underlying dimension of physiological arousal (e.g., Lindsley, 1951). More recently, unidimensional theories have been supplanted by two‐dimensional models, which either add a pleasantness (or "valence") dimension to the arousal dimension (e.g., Russell, 1980), or conceptualize positive affect and negative affect as the two fundamental dimensions of emotional experience (e.g., Watson & Tellegen, 1985). However, starting in the 1960s, evidence has increasingly supported the conception (see, e.g., Aristotle, 1966; Descartes, 1989; Darwin, 1965) that there are several distinct emotions (such as joy, sadness, fear, and anger), as manifest in different facial expressions observable across cultures (e.g., Ekman, Sorenson, & Friesen, 1969; Izard, 1971; Tomkins & McCarter, 1964) and in characteristic action tendencies such as approach, inaction, avoidance, and attack (see, e.g., Frijda, 1987; Roseman, Wiest, & Swartz, 1994; Shaver, Schwartz, Kirson, & O'Connor, 1987). These findings cannot readily Scherer, K.R., Schorr, A., Johnstone, T. (Eds.) (2001) Appraisal Processes in Emotion: Theories, methods, research. Oxford: Oxford University Press.