As a prototypic negative emotion, anger would seem to have little in common with positive activation, as measured by the Positive and Negative Affect Schedule (PANAS; D. Watson, L. A. Clark, & A. Tellegen, 1988). However, growing evidence suggests that both anger and positive affect are associated with approach motivation. This suggests the counterintuitive hypotheses that positive affect should be increased by anger-evoking situations, and that positive affect and anger should be directly correlated in such situations. Four studies tested and supported these hypotheses. Discussion focuses on the implications of these results.Keywords: anger, approach motivation, positive affect, PANAS Some dimensional models of emotion postulate that positive affect is directly associated with approach motivation, whereas negative affect is directly associated with withdrawal motivation (e.g., Watson, 2000). In this article, we examine how anger, an approach-oriented, negatively valenced affect, relates to positive activation (PA), which has also been linked to approach motivation (Watson, 2000). PA is measured by the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988), one of the key measures used in tests of these dimensional models. More important, we tested the relationship during the experience of anger. If anger relates directly to PA, this would have implications for the structure of emotion and motivation. It would also suggest that the PANAS does not provide a pure measure of positivity, and this would have implications for the interpretation of research using the PANAS.The linking of affective valence with motivational direction may be appropriate for most affects. Joy, interest, and enthusiasm are positive affects that appear to be associated with approach motivation. Similarly, disgust, fear, and anxiety are negative affects that appear to be associated with withdrawal motivation. Most past research has examined these affects.Far less research has been conducted on anger, an affect that violates the direct linking of affective valence and approach motivation. Indeed, according to recent research, anger is a negative affect associated with approach motivational tendencies (for review, see Carver & Harmon-Jones, in press). Thus, in contrast to the theoretical idea that approach motivation is only associated with positive affect, we propose that approach motivation also involves certain negative affects, particularly anger. Consequently, if the PA measure of the PANAS taps approach motivation rather than purely positive activation, it should be related to anger even though PA and anger seem to have different valences. Watson (2000) suggested that positive activation reflects activation of the approach motivational system. He wrote, . . . a growing body of evidence has suggested that negative mood experience is part of a larger Behavioral Inhibition System (BIS), whereas positive mood experience is linked to what has variously been called the Behavioral Activation System, Behavioral Engagement...