This is a book about ethnic conflict, but it is not about ethnic conflict in the usual sense. Most treatments of the subject focus on explaining variation in the occurrence or intensity of ethnic violence across time and space. They ask: why did tensions between Hutus and Tutsis in Rwanda erupt into full-blown genocide in 1994 (Prunier 1995; Gourevitch 1998)? Why have Hindu–Muslim riots taken place in some Indian towns but not in others (Varshney 2002; Wilkinson 2004)? Why, since the mid-1970s, has there been so much more ethnic violence in Sri Lanka than in Malaysia (Horowitz 1989)? This book, by contrast, seeks to explain not when or where communal conflict breaks out but why it breaks out along one line of ethnic division instead of another. It seeks to explain why politics comes to revolve around the particular axis of ethnic cleavage that it does.

I fell into this research topic by accident. When I first went to Zambia in 1993, I thought I was going to be studying how Zambia's recent transition to multi-party politics had affected the relations among the country's ethnic groups. I thought I was going to be studying whether the introduction of competitive multi-party elections had deepened ethnic divisions and made ethnicity a more salient part of the country's political affairs. Given this motivating question, I began my research by conducting a series of open-ended interviews in one of Lusaka's poor residential compounds to probe how people from different ethnic communities were getting along in the new multi-party era.