This thesis seeks to assess the relevance of existing theories about the origins of nationalism and investigate more specifically the claim that nationalism is rooted in ethnicity. It does so by examining the cases of Eritrea and Somaliland, which proclaimed their independence in May 1991 after seceding from the states to which they were formerly united. Having explained in the introduction why International Relations needs to take a closer look at the causes of nationalism, the second chapter proceeds to review some of the main theories about the origins of nationalism. It retraces the history of the primordialist-modernist debate, discusses the main contentions of the ethnonationalist approach and presents some of the factors singled-out by recent scholarship as propitious for the emergence of nationalism. Given that most of the theories about the origins of nationalism presented in chapter two centre on Europe, chapter three surveys the literature on the rise of nationalism in Africa i_n order to determine whether any additional factors need to be considered before analysing Eritrea and Somaliland. Chapter three also includes a discussion of the anthropological literature on ethnicity in Africa and questions the ethnonationalist claim that ethnic groups are pre-modern. Using as a framework the factors identified previously, chapter four offers a historical account of the emergence of nationalism in Eritrea. Chapter five does the same for the case of Somaliland. As the analysis provided in chapters four and five illustrate, the claim that nationalism and secession have ethnic roots is not empirically substantiated by the cases of Eritrea and Somaliland. The thesis concludes by discussing the practical implications of these fmdings with regard to the right of secession and proposals for boundary adjustment in Africa. It also highlights the ways International Relations may contribute to our understanding of the causes of nationalism.