Electronic Thesis and Dissertation Repository

Thesis Format



Doctor of Philosophy




Sullivan, Jacqueline

2nd Supervisor

Klimchuk, Dennis


In recent years both philosophers and scientists have asked whether or not our current kinds of mental disorder—e.g., schizophrenia, depression, bipolar disorder—are natural kinds; and, moreover, whether or not the search for natural kinds of mental disorder is a realistic desideratum for psychiatry. In this dissertation I clarify the sense in which a kind can be said to be “natural” or “real” and argue that, despite a few notable exceptions, kinds of mental disorder cannot be considered natural kinds. Furthermore, I contend that psychopathological phenomena do not cluster together into kinds in the way that paradigmatic natural kinds (e.g., chemical kinds and species) do; and, in light of this fact, I conclude that the normative ideal of natural classification—i.e. classifying real or natural kinds—is not appropriate as a general strategy for psychiatry. In the conclusion to this dissertation I propose an alternative way forward. Rather than assuming a priori that psychopathological phenomena cluster together into kinds, a more tractable and theoretically promising approach would be to first explain how particular kinds of experiences and behaviours—e.g, hallucinations, delusions, disorganized thinking, mania, low mood, etc.—are produced.

Summary for Lay Audience

A problem that scientists, psychiatrists, and philosophers have been wrestling with in recent years is how to best classify mental disorders. The primary motivation for this discussion is the belief that current psychiatric classifications have done a poor job of promoting the development of effective therapeutic interventions for mental illness. Although there is widespread agreement amongst scientists and philosophers with respect to what psychiatric classification has failed to do, it is not yet clear precisely why it has failed and how to go about fixing it. One suggestion that has gained popularity in recent years is that the psychiatric classification has failed because its categories do not really reflect the way the world is. For example, two people diagnosed with schizophrenia can be so different (e.g., one might be suffering from delusions and hallucinations, while another might be suffering from speech and motor problems) that it is more likely than not that their problems are caused by different things, which strongly suggests that providing them with the same treatment will be unhelpful. Nevertheless, there is still a lot of disagreement about whether or not categories like schizophrenia do reflect the way the world is. Some argue that schizophrenia is real, while others argue that it is a made up. In this dissertation I clarify the sense in which mental disorders can be said to be “real” and argue that, despite a few notable exceptions, mental disorders cannot be said to be real, and are better understood as hypothetical constructs. Furthermore, I argue that hypothetical constructs like schizophrenia and depression are not useful for psychiatry, and a better way forward is to focus on trying to explain and treat particular experiences and behaviours—e.g, hallucinations, delusions, low mood, etc.