Diseases as Homeostatic Property Clusters

Several philosophers have recently drawn on property cluster accounts of natural kinds to argue that individual diseases form natural kinds. According to this view, diseases have a super-explanatory property (their pathophysiology), by virtue of which their other properties (their symptomatology, biological signature, response to treatments, and prognosis) tend to co-occur. I argue that we can say more: Disease pathophysiologies are mechanisms, so that the property clusters diseases form are homeostatic. I examine the two objections that natural kind theorists have raised against the homeostatic property cluster account, and I argue that they should not deter us from applying it to disease.