By the final decade of the twentieth century, rates of suicide in Sri Lanka ranked amongst the highest in the
world. However, in 1996 the suicide rate began to fall, and was soon at its lowest level in almost thirty years.
Posing problems for classic sociological theories of suicide, the decline forces us to question some fundamental
assumptions underlying social scientific approaches to the problem. Drawing from sociological, medical
epidemiological, historical, and anthropological secondary sources as well as twenty-one months of original
ethnographic research into suicide in Sri Lanka, I argue that there are four possible readings of the country’s
suicide rate. While the first three readings provide windows onto parts of the story, the fourth, a composite view,
provides a new way of thinking about suicide not just in Sri Lanka but also cross-culturally. In so doing the
article poses questions for how the relationship between suicide and society might be imagined