This article reviews the problem of 'boundaries' in psychopathology and its conceptualization in the diagnostic classification of psychiatric disorders.A modest number of publications in 2004-2005 contribute concepts, methods and data relevant to several aspects of the problem: scientific 'facts' versus value attribution in the definition of mental disorders; the problem of comorbidity between diagnostic entities; the spurious dichotomy between categorical versus dimensional approaches to diagnosis and classification; and the distinction between validity and pragmatic utility.The relative paucity of research explicitly addressing these issues should be a reason for concern in the debate leading up to the next revisions of the International Classification of Diseases and the Diagnostic and Statistical Manual of Mental Disorders.