Gloves are like multiple personality

Historical ontology, we learn, is a term put forward - though not dwelt on - by Foucault. It teeters on the oxymoronic since one would think that matters ontological would be about as timeless as things could get. But Ian Hacking's interest, in most of the essays that make up this collection, concerns the ways in which novel entities can come into being. Without falling - as he would see it - into the trap of constructionism (and adopting the suggestion that all our claims about what exists are equally made up), he wishes to focus on created identities: "Historical ontology is about the ways in which the possibilities for choice, and for being, arise in history."

In his introductory essay and the central chapters exploring Foucault, he spells out just which kinds of things can have an ontological history. He takes four (apparently mismatched) examples: horses, planets, gloves and multiple personality. The first, he argues, is an easy case: the entity of horses is not made up; there is no historical ontology. Equine creatures have more in common than simply that we call them all horses, and they had that in common before we ever invented the category.

The planets seem to be trickier since, for centuries, we got by successfully thinking about the planets differently from the way we do now.

But still, there is a best way to think about this class of astronomical bodies. Moons differ from planets and planets from stars while planets resemble other planets. Gloves, however, are constructed. Without people being the way they are, there would be no gloves. For billions of years there were no gloves; they had a coming-into-being. Hacking's interest is in investigating which seemingly natural things are more like gloves than like horses and planets. Multiple personalities, he argues, are of the glove kind. Before the last quarter of the 19th century, this was not a candidate identity at all, though people might be possessed by spirits and manifest some similar behaviours. Subsequently, doctors, medical schools, other social institutions and the patients' personalities - and sometimes even their multiple identities - orientated themselves towards this possibility. It thus became possible for people with multiple personality to be.

The claim is not that medical diagnosis woke up to this possibility only in 1875; it is that the condition and the treatment of the condition arose together. This, for Hacking, is what gives his inquiry its Foucauldian character: it was not "that there was a kind of person who came increasingly to be recognised by bureaucrats or by students of human nature, but rather that a kind of person came into being at the same time as the kind itself was being invented". The invention is both an intellectual accomplishment and a matter of practices and institutions.

Through this complementary focus on intellectual categories and social practice, Hacking connects with Foucault's famous emphasis on the play of power/knowledge.

The key thing that seems to characterise Hacking's examples is that these identities are in a sense reflexive. Thus, another example he offers, also from the 19th century, is that of suicides. Suicide, along with such now-standard accoutrements as the suicide note and the coroner's willingness to cite suicidal impulse as a cause of death, is understood by the suicide themselves and constituted as a form of action in part through that understanding. But one could not have been a perpetrator of suicide in this sense, nor for that matter a multiple personality or possibly a sexual pervert, at just any point in human history. Such new ways of being could arise at particular times only out of particular resources and practices; in that sense Hacking's thinking about the development of these identities is rather Hegelian. And to demonstrate the ubiquity of such phenomena, Hacking concludes with an engaging essay on the historicity of dreaming.

The great strength of the essays in this book is that they are learned and lucid. Each chapter is quite narrow in focus, often having been written for a lecture. As a philosophically robust introduction to Foucault and to Hacking's thoughts about Foucault's work and related concerns, the book is very strong. But it is determinedly philosophical; the reader who wants to understand more about the construction of historical identities will have to turn to Hacking's other studies or to the many other sources cited.

Steven Yearley is professor of sociology, University of York.