C. S. Carver and E. Harmon-Jones have presented considerable evidence to support their argument that "anger relates to an appetitive or approach motivational system, whereas anxiety relates to an aversive or avoidance motivational system" (p. 183). However, they have failed to take sufficient account of the extensive psychometric data indicating that anger is strongly related to anxiety (and other negative affects) and more weakly associated with the positive affects. Considering all of the available evidence, the most accurate conclusion is that anger shows both approach and avoidance properties. Moreover, viewed in the context of the hierarchical structure of affect, some evidence suggests that the nonspecific component of anger (i.e., its shared variance with the other negative affects) is primarily related to the aversive or avoidance motivational system, whereas its specific component (i.e., its unique qualities that distinguish it from other negative affects) has a stronger link to the appetitive or approach system. The author concludes by considering the broader implications of these data for affective structure. (Contains 2 tables.)