This study examines how language operates as a tool of manipulation in academic grooming within Indonesian higher education. Using a qualitative forensic-linguistic approach integrating Critical Discourse Analysis, Speech Act Theory, and Thematic Analysis, it analyses approximately 850 utterances from authentic lecturer–student communications, supplemented by interviews with six victims and one expert. The findings show that grooming unfolds through three interrelated linguistic mechanisms, affective persuasion, instrumental exploitation, and symbolic control, that gradually transform professional discourse into emotional dependency and compliance. Affective and instrumental lexis dominates grooming communication (71%), embedding coercion within rhetoric of care, mentorship, and professionalism, and producing discursive entrapment in which victims internalise control as trust. The study proposes a Forensic Linguistic Indicator Model (FLIM) for the early detection of grooming language, conceptualising academic grooming as institutionalised linguistic coercion and offering policy-relevant insights for safeguarding and prevention in higher education.