Mõttus alerts us to the widespread predictive heterogeneity of different indicators of the same trait. This heterogeneity violates the assumption that traits have causal unity in their developmental antecedents and effects on outcomes. I would go a step further: broader traits are useful units for description and prediction but not for explaining personality development and personality effects. In most cases, the measured trait indicators are closer to relevant causal mechanisms, and within a network perspective on personality, broader traits as entities with causal potential can be dismissed completely. Copyright © 2016 European Association of Personality Psychology