In this language production experiment, working adults (N ¼ 226) were asked to respond to unethical business requests. Our objective was to advance a communicative understanding of unethical organizational behaviors by analyzing the linguistic adjustments workers employ to deny unethical requests. Specifically, we measured responses to unethical requests on a continuous coding scheme, which captured degrees of denial directness. We hypothesized that command structures produce a hierarchical mum effect in which subordinates are more indirect in denying an unethical request than supervisors and coworkers. Results confirmed the presence of a hierarchical mum effect; data also indicated that females, younger workers, and those with the least work experience are most indirect in denying an unethical request.