This thesis investigates the role of conceptual schemes in shaping how we make sense of experience in human affairs. Conceptual schemes involve values and beliefs, assumptions and presuppositions that originate in past experience and, projected onto new observations, guide what is seen, what is considered salient, and what is overlooked or dismissed. Paradigms (in the sense of exemplars) and loaded concepts play a decisive role in these dynamics, which I explore through cases, each showing how an opposing pair of conceptual schemes leads to contrasting descriptions of what is apparently the same problem. Disagreement in the study and measurement of intelligence rests on contrasting conceptions of success, science, and education. Different characterisations of identity, power, and religion lead to seeing aspects of international relations and terrorism that involve different explanations. Finally, divergent approaches to criminal justice rest on contrasting conceptions of the human being that involve different views of responsibility, motivation, and rehabilitation. All approaches present their positions as matters of fact: intelligence does predict success; identity is defined by antagonism, or not; criminals do lack self-control. However, descriptive statements are often loaded with value. I propose disentangling descriptive and evaluative elements to reveal the specific features that are endorsed, or deprecated, when we use certain expressions. Disentangling increases awareness of what we do when we talk about a phenomenon in a certain manner; it shows what is presupposed and allows the critical examination of elements that have become tacit. Contrast with other conceptual schemes, and focus on difference, shows us where to look for disentangling and opens a venue to unusual ways of seeing. Uncomfortable as it may seem, engaging across different schemes, rather than seeking to transcend them, is thus proposed as a way to expand possibilities for understanding, and action.