Purpose -The purpose of this paper is to develop a short and easy to use yet psychometrically sound instrument designed to measure destructive leadership behaviours in a military context. Design/methodology/approach -First, examples of destructive leadership behaviours in a military context were collected using a qualitative approach. Second, these examples were operationalised and pilot tested, which resulted in a 20-item questionnaire called Destrudo-L. Third, data were collected from three Swedish military groups (n ¼ 428). Dimensionality of the instrument was analysed using structural equation modelling. Conventional psychometric assessments of reliability and validity were performed. Findings -A nested hierarchical model with a general factor and the following specific factors emerged: arrogant, unfair; threats, punishments, overdemands; ego-oriented, false; passive, cowardly; and uncertain, unclear, messy. Meaningful subgroup differences and relationships with a criterion variable (lack of motivation/propensity to leave) were found. More use of active forms was reported by subordinates of younger military commanders and more use of passive forms was marked by subordinates of senior military managers. Practical implications -The instrument is easy to administer and interpret (norm values are provided) and can be used in leader evaluation, as well as leader development, contexts. Originality/value -The main contribution is methodological -the development of a new scale. Additional findings are a strong positive correlation between active and passive forms of destructive leadership behaviours in a military context, as well as significant differences between groups with different ranks.