This article examines the status of dynamical models of movement coordination qua phenomenological models. After a brief outline of the aims, methods and strategic assumptions of the dynamical systems approach, a survey is provided of the theoretical and empirical progress it has made in identifying general principles of coordination. Although dynamical models are constructed for phenomena at a particular level of analysis for which they provide descriptive explanations, their dynamics can sometimes be linked to or associated with the dynamics of processes at other levels of analysis. The article concludes with a tentative scheme to clarify the position of the dynamical approach relative to other extant approaches in movement science. © 1995.