Purpose - Integration is policy, practice as well as object of systematic investigation. What we do not know is whether or not integration can be understood as a science. In his book The Structure of Scientific Revolutions, Thomas Kuhn formulated a notion of (natural) sciences based on the emergence of commitments amongst a community of scientists to a set of logics, model and exemplars. He called this a paradigm. The purpose of this paper is to assess the scientific nature of integration by perceiving it as a paradigm in Kuhn's sense. Design/methodology/approach - The paper proceeds by conceptual reflection through matching existing components, theories and exemplifications of integration to Kuhn's model of a scientific paradigm. Integration is understood broadly, either vertical or horizontal, and located within the practical domains of policy formulation, policy implementation and evaluation research. The nature, scope and depth of group commitments amongst students and practitioners of integration receive particular attention in line with Kuhn's social interactionist approach. Findings - Employing Kuhn's notion of paradigm in the context of integration highlights the fundamental tension between integration efforts and integration outcomes. Whilst integration defines itself in contradistinction to professional boundaries and fragmentation, the paper argues that it fails to develop a strong theoretical and empirical foundation for a robust and stable group commitment. The reason is that the key motivational force that may create a stable group commitment amongst those engaged in integration, the patient perspective, remains outside the integration paradigm. This leaves integration as a practice and policy model underdeveloped, mainly paradigmatically illustrated by singular exemplars and rooted in aspirational policy vocabulary, while clustered around a near dogmatic belief that working together between services must lead to improved quality of care. To become a scientific paradigm the group commitment in integration would have to coalesce around a clear ontology (symbolic generalisations), epistemology (models of knowledge) and manifestations in practice (exemplars). Research limitations/implications - At present both the ontology and epistemological foundations of integration practice and research are insufficiently clear. This hampers the development of integration practice as well as a better understanding of how to evaluate integration outcomes. Future studies should focus on the depth, nature and subject of group commitments to assess whether integration is a viable candidate for scientific paradigm or an assorted construct of policy aspirations. Originality/value - The paper questions the rigour and trajectory of integration practice, policy and research. It identifies a tension at the centre of the field between group commitments to scientific exemplars (case studies) and symbolic generalisations, encapsulated in the desire to improve patient care. The notion of a scientific paradigm thus helps to re-frame the discussion about research and practice in integration.