Subtypes, specifiers, epicycles, and eccentrics: Toward a more parsimonious taxonomy of psychopathology.

Berenbaum, Raghavan, Le, Vernon, and Gomez (2003; this issue) have performed a valuable service by articulating a taxonomy of emotional disturbance. While acknowledging that this structure has many attractive features (e.g., its inclusion of specific negative emotions such as disgust and anger), I discuss three potential problems. First, in comparison to “predominantly unpleasant” dysfunctions, little evidence supports the existence of “predominantly pleasant” disturbances in psychopathol-ogy. Second, emotional reactivity is confounded with valence, and it is unclear whether these two proposed types of disturbances actually provide independent information. Third, Berenbaum et al.'s taxonomy fails to model the behavioral substrate of emotions. I conclude by arguing that emotion-based taxonomies should be used to supplant—rather than supplement—the existing DSM framework.