This paper investigates benevolent dehumanization in Canadian parliamentary discourse—a rhetorical mode in which compassion serves as a mechanism of moral hierarchy. Drawing on thirty years of House of Commons transcripts (1994–2024), it analyzes how appeals to empathy, inspiration, and protection recurrently frame disabled citizens as moral patients rather than moral agents. Using a mixed-method approach integrating computational text analysis with qualitative discourse coding, the study maps recurring patterns across debates on accessibility, employment, and medical assistance in dying. The evidence shows that benevolent framings reproduce existing institutional hierarchies within a broader adversarial environment where empathy coexists with competition, partisanship, and strategic positioning. The paper contributes an empirical framework for examining affective and paternalistic reasoning alongside adversarial performance, offering a way to quantify how moral language organizes political hierarchy and linking it to measurable patterns of institutional behaviour.