Answers are grounded in this paper and cite their sources.
Over the past century, leadership research has accumulated numerous constructs and theories. We characterize this literature as a “nebula,” rich but clouded by construct redundancy that blurs conceptual boundaries. Drawing on 2,226 articles published between 2000 and 2023, we use a large language model (LLM) to identify 146 primary leadership constructs and 389 root behaviors and to code each construct’s descriptions for these behaviors. Redundancy concentrates in a positive and a negative cluster with different internal structures. The positive cluster collapses toward a common center whose most prominent behaviors form a shared core, and its constructs depart from that center along five dimensions, labeled CHAMP (Change-focused, Harmonization-centered, Authority-based, Morality-driven, and People-oriented). The negative cluster instead separates into two distinct prototypes, the control-obsessed Dictator and the disruption-driven Saboteur. This asymmetry is surprising: the less studied negative constructs form distinct groups, whereas the positive ones, which the field has worked hardest to differentiate, do not. It suggests that positive constructs are modeled on a culturally shared prototype of the ideal leader, whereas negative constructs are distinguished by how leaders do harm. This structure helps explain why positive constructs keep proliferating and indicates where constructs with incremental value may emerge
No takes yet. Share an insight, caveat, or question.
Lin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: