This text is intended as a reflection on the most important contradictions and limitations of evaluations performed on Graduate Programs submitted to the CAInter/CAPES Interdisciplinary Evaluation Committee. Some of the concerns have to do with the search for quality in the production of knowledge in those programs. This search is, on the one hand, associated with the organization of curricula, the composition of the faculty and student bodies, the dynamics of research groups, the conditions under which theses, dissertations, and research are prepared, where multi- and interdisciplinarity are viewed as goals. On the other hand, it related to the Evaluation System to which the Programs are subordinated and whose origin is based on disciplinarity. The conclusions mainly point to the need for discussing alternatives to the manner by which disciplines are evaluated, for without change there is a risk of jeopardizing the potential for the construction of knowledge aimed at establishing interchanges between distinct scientific areas.