The development of theory and conceptualisation of militias is surprisingly sparse in comparison to the study of other types of combatants. The quest for conceptual clarity in the wider literature is hindered by the vast number of groups that are labelled "militias". The extent and range of studies of militias may be driven by a preferred nomenclature for conflicts, where the term militia is used as a residual category applied to seemingly atypical rebel groups, or any group short of a statutory military.

This paper presents a new theory of militias, which examines the relationships between militias, the state and society, and finds that the key relationship dynamic lies in terms of the legitimacy relationship between militias and society. It is this legitimising relationship which influences the militias relationship with the state. The theory is developed from an examination of the prominent role of militias in American military and political history, Clausewitz's delineation of militias' key attributes in On War and other works, and a concept analysis of similar and related combatant-types. The example of the Kamajors of the Sierra Leone civil war demonstrate the analytical utility of this new conceptualisation.

Elections in Nigeria, particularly the oil rich Niger Delta, are often fraught with irregularities and pockets of violence. The resource allocation formula used by the federal government to offset the impact of oil exploration activities in the region boosts the income of the Niger Delta states in relation to other non- oil producing states in Nigeria. The relative 'richness' of the Niger Delta states increases the stakes for electoral representation but in a predatory manner. Electoral positions are seen as avenues to enrich the elite and their cronies at the expense of the state and this results in electoral malpractices and the use of violence.

Since the return of Nigeria to democratic government in 1999, there has been an increase in the use of militias to seize or maintain control of political institutions in most of the Niger Delta states. This paper explores the roles played by militia groups before, during and after elections and the relationship between the groups and their patrons. Contrary to the notion that more established militias can effectively serve the interest of their patrons, the paper argues that the lack of trust between politicians and militias result in politicians employing the services of different groups for future elections and this development leads to the formation of new groups during each electoral cycle. This analysis is based on an empirical research carried out in the three states in the Niger Delta between 2013-2016.

Nigeria's Fourth Republic has now survived longer than any prior attempt at democratic rule in the country, yet violence during electoral contests has continued to undermine the legitimacy of their results and of elected leaders. It seems a truism that in order to stop political violence, one most understand its causes. However, questions on what constitutes political violence, what its drivers are, and who are the responsible actors can be in themselves matters of perception and contestation. Especially in the conflict-affected country, public debate on political violence can become a power-struggle which, in turn, can affect the further development of conflict. In this paper we look in particular at how public debate on electoral violence is formed in three Nigerian newspapers. Based on these sources, we established a dataset on different types of conflict (protests, riots, deadly armed conflict, terrorist attacks) in Nigeria. The dataset covers the period April 2014-March 2015 —the one-year period before the presidential elections. In qualitative as well as quantitative analyses we demonstrate how newspapers' differing political affiliations can affect media representations of the dynamics of electoral violence in Nigeria. Most media attention was directed to the two major Nigerian parties: The People's Democratic Party (PDP) and the All Progressives Congress (APC). Results demonstrate the difficulty of capturing and understanding electoral violence in Nigeria due to political claim-making. It also emerges that parties and political leaders are seen as major drivers of violent conflict in public debate.

In Uganda's 2016 elections, international and national commentators questioned the role that the government's crime preventers—or community police—would play. Many claimed that they would be used "as tools" to rig the elections, intimidate voters, and vote en masse for the ruling NRM regime. In contrast, this paper shows that the government never intended the crime preventers to play an explicitly coercive role. Instead, the NRM leadership intentionally structured the crime preventer program as indefinite and fluid, allowing political authorities and citizens to understand the purpose of crime preventers alternately as dangerous tools of the regime, family men in search of work, or patriotic citizens of Uganda. Used interchangeably, these logics—which are described in this paper as ideal-typical categories of political, economic, and social—prevented Ugandans from accurately assessing the program. The resultant uncertainty fragmented organization of crime preventers, civil society, and members of the opposition; limited the government's responsibility for crime preventers; and helped ensure that crime preventers would bolster the strength of the NRM regime in the 2016 elections.