TheIdea of explaining natural phenomena by appealing to laws of nature is one that is thoroughly familiar to the modern mind. This idea does not perhaps appear quite as natural as it did a century ago, when Engels proclaimed to the mourners at Marx's funeral that just as Darwin had discovered the law of development of organic nature so Marx had discovered the law of development of human history. Twentieth-century historians do not in general conceive their task as including the formulation of laws of history, and the discoveries of modern physics since Maxwell have for the most part been expressed in terms of principles and equations rather than laws. Nevertheless, despite these changes, we are still quite accustomed to thinking in terms of laws of nature; and just because it seems natural it is easy to assume that it is natural for human beings seeking to explain the phenomena of nature to do so by enquiring after the laws by which these phenomena are governed.

Despite Durkheim's powerful argument that suicide is caused by social forces only (Durkheim 1966), recent studies have viewed suicide as resulting from a combination of psychological and environmental factors, as springing from the suicide's personality conjoined with his immediate social setting. ‘Suicide is both a personal and a social act’ (Reynolds and Farberow 1976: 33). Social forces alone cannot explain why only a few people commit suicide nor can psychological factors by themselves provide the necessary set of conditions for individual suicides. Certain syndromes are widespread: illness, loneliness, bereavement, shame lead to suicide in many parts of the world. But each society's structure and institutions determine the environmental setting and constellations involved in suicide. How does the society regard illness? Who will be lonely? In what relationships is loss most sharply felt? What constitutes shame? Social forces define the various trouble spots and categories and groups of people most at risk in the society. Their interplay with psychological factors outlines who will commit suicide.

During the past few years there has been a rapid growth of interest in the sociological history of psychiatry. Prior to this, the history of psychiatry had been left largely to the psychiatric historian, who tended to proceed, as Thomas Szasz claimed, ‘as a socially neutral person, discovering the historical “facts” when in truth, he is a psychiatric propagandist, actively shaping the image of his discipline’ (1). Writers such as Michel Foucault, Vieda Skultans, Andrew Scull, David Rothman, Klaus Doerner, and Szasz himself have attempted to underline, as Skultans says, ‘the specific uses to which psychiatry has been put in the past, in order to make a more general claim about the nature of psychiatry as such’ (2). This aim, however, is not always made fully explicit (3). In this paper it will be argued that psychiatry, viewed as a historically constituted social activity, was characterised by a dualism. It was constituted by a medical or curative model of practice, in that psychiatry developed as a branch of medicine. Yet the ‘diseases’ which psychiatrists have historically come to regard as part of their field competence are distinguished by at least two criteria: first, their symptoms consist primarily of actions that are highly inappropriate to their social context; secondly, that their etiology is ambiguous. It will be argued here that an ambiguity regarding the etiology of mental disorder, which is often seen as both physically and psychologically caused, was central to psychiatric discourse.

Deaths of sovereigns or political leaders are generally accompanied by dramatic representation and celebration of the political order over which they have presided. The circumstances of death, funeral rites and destination of the corpse (cremation, burial or public display) proclaim the value and necessity of the ideas embodied in the ruler's office. However practically deficient or scandalous any particular ruler's interpretation of that office, the activities which surround his death reaffirm the invulnerability of the transcendent order to any local or temporary individual failings. Sometimes, however, the circumstances of a sovereign's death can be appropriated by his opponents not merely to decree that death but to destroy the ideological underpinnings of the political system itself: the trials and executions of Charles i and Louis xvi were not simply the punishment of individuals for specific crimes but rather symbolic destructions of monarchy itself staged by Cromwell and the conventionnels (Walzer 1974). Such occasions have been rare. A radical political opposition can expect at most to intervene in the timing of the ruler's death, by assassination, but draw no benefit from this, since it is likely to be even more effective than peaceful death in stimulating public affirmation of the existing order.