Whenever we want to research or discuss the consequences of any phenomenon, we need to have a clear idea of what that phenomenon is. So it is when we hypothesize that what we call "disaster" has negative consequences for mental health. While it may seem easy to conceptualize what constitutes a disaster, the task is far more complicated than appears at first glance. Even more complicated is the task of identifying the mental health effects of disasters. The phenomena are not self-evident. The disaster literature yields little consensus on the definition of the basic concept of disaster or any of its derivatives. Nor does the literature provide much of an empirically based position on any possible relationship between disaster occasions and mental health. An underlying theme of this paper is that we will never clearly understand the effects disasters may have on mental health unless we clarify what we want to consider as a disaster, including the most important dimensions of disaster. The lack of consensus on any such definition has impeded research and hindered our ability to draw valid and significant conclusions about the relationship between disasters and mental health. While enough work has been done to suggest relevant models, significant questions, and reasonable hypotheses, there is likely to be little progress in this research area unless we move toward conceptual clarification of the key concept involved, that is, disaster.