Ontological assumptions of psychiatric taxonomy: main rival positions and their critical assessment.

The scientific classification of human psychopathology raises a host of important philosophical issues ranging over all major areas of philosophical inquiry; logic, semantics, ontology, epistemology and ethics. The paper focuses on the major views of the ontological status of our psychodiagnostic constructs; idealism, nominalism and finally, neoaristotelian realism. Idealism holds that our psychodiagnostic constructs represent ideal essences, instantiated imperfectly by individual mental patients of the same psychodiagnostic class. These ideal essences are construed either as abstract forms dwelling in an autonomous realm of eternal abstract objects--platonistic idealism--or as ideal objects emerging in and grasped directly by a pure or absolute consciousness--husserlian idealism. By contrast, nominalism claims that our psychodiagnostic constructs are nothing but conventional common names of sets of individual mental patients, invented and adopted on purely pragmatic grounds such as their practical utility, especially in predicting their clinical course and outcome under various types of treatment. Finally, neoaristotelian realism holds that, though our psychodiagnostic constructs are concepts, and thus of our own making, they are not arbitrary ones but grounded in objective features shared by individual mental patients of a given psychodiagnostic class. It is argued that both idealism and nominalism are flawed and thus untenable--though for different reasons--and a refined version of neoaristotelian realism defended.