This thesis presents a quantitative study that aims to investigate whether Brian Martin is right in his theory about how more powerful actors have a greater capacity to prevent outrage and anger after opressions and thus suffer less from political jiu-jitsu, a process in which oppression becomes counterproductive. This is done by looking at whether more powerful regimes getaway more easily with repressing nonviolent campaigns. By designing a measuring scale for the scope of political jiu-jitsu, the connection between the scope and three different aspects of power - national capacity, wealth and state oppression - is investigated. The results shows that the more powerful the oppressive states are in terms of national capacity and wealth, the less extensive political jiu-jitsu. On the other hand, a higher degree of state oppression results in more extensive political jiu-jitsu. The results linked to the degree of staterepression are statistically significant and it can thus be said that the differences in the extent of political jiu-jitsu are not due to chance. The results indicate that more powerful states getaway with repressing nonviolent campaigns more easily, if power is measured in terms ofnational capacity or wealth. If, on the other hand, power is measured in the amount of noppression, it is more costly for the states that exercise more oppression.

Previous research has found that nonviolent campaigns are conducive for democratization, but variation in democratic outcomes still remains a puzzle. I address this research gap by analyzing whether democracy assistance that promotes political participation, civic political culture, and enabling environment for civil society before and during nonviolent campaigns can help explain why some countries democratize after regime changes initiated by nonviolent campaigns while other countries do not. I argue that sustained democracy assistance help maintain mass mobilization and build democratic institutions after the old regime has been removed. By using the method of structured focused comparison, I investigate based on data from USAID and OECD what kind of democracy assistance Tunisia and Egypt received before and during their nonviolent campaigns. I find that neither Tunisia nor Egypt to any great extent received the kind of sustained democracy assistance I hypothesized could impact democratization. I therefore conclude that it seems unlikely that democracy assistance had the kind of significant impact that could explain the different outcomes in Tunisia and Egypt. Lastly, I discuss how research on nonviolent campaigns could inform policymaking and contribute to designing more strategic democracy assistance in the future.

This study aims to investigate how two forms of non-violent campaigns, one run by unionsand one without, have arisen. It is investigated by analyzing the mobilization processes of the two campaigns. The materials used in this study is the theoretical framework of Resource Mobilization Theory’s five resource mobilization categories. It was used as an alternative way to examine how and understand why the two movements turned out to be so different, where one was a labor movement and the other were not. The method used was a qualitative comparative case study. Through the comparison the study could identify some differences between the two movements campaigns. Through the comparison, the study was able to identify certain differences between the two movement campaigns. On the one hand, the scope and network of the campaigns were important for their emergence, and on the otherhand, the temporal context was important for the mobilizations of the cases.

There is an ample body of literature which seek to investigate the role of religion in armed conflict. Yet, the role of religion in unarmed civil resistance has not received an equal amount of scholarly attention. Apart from some single and comparative case studies showcasing the pertinent role of religion on nonviolent campaign success, little has been done to investigate its effect across broader cases. By conducting a large-N analysis of 143 nonviolent campaigns from 1975 to 2013 globally, this study aims to fill in the research gap and answer the following question: How does religious support affect the success of nonviolent campaigns? Furthermore, this study seeks to investigate whether unique forms of religious support, i.e. traditional support and religiopolitical support, have a different effect on nonviolent campaign success. The results indicate that religious support does not have a statistically significant effect on the success of nonviolent campaigns. The study also reveals that religious support type does not have a statistically different effect on the chance of nonviolent campaign success. Thus, this study challenges the generalizability of case studies demonstrating the effectiveness of religiously-supported nonviolent campaigns. Avenues for future research include conducting a large-N study on the effect of religious support on the level of mass mobilisation and the resilience capacity of nonviolent campaigns.