This dissertation aims to understand how Islamist ideology differs across Islamist political parties in majority-Muslim countries. It asks why, despite drawing from the same religious source, the ideology of Islamist parties differs significantly and ranges from a wide spectrum, reaching from rigid conservative ideologies to flexible and even liberal. In order to address this question, the first step I pursue is to create a classification of different types of Islamism based on a spectrum of Islamist thought and behavior. I call the most literalist approach which aims at top-down Islamization Traditionalist Islamism, while the most flexible variety which argues for a bottom-up approach and relies on secular civil law is is labelled Neo-Islamism. Political parties that do not clearly fall under any of these categories and mix characteristics of both are grouped under the Hybrid Islamism category. I argue that if they could, all Islamist parties would moderate their ideology in order to achieve electoral success. However, both the position of the party vis-a-vis other parties in the system, and a high degree of dependence on an internal clique or an external movement create constraints and limitations which prevent parties from moderating their ideological profile. I find support for my argument by comparing five different Islamist parties in Indonesia. I further strengthen my case by introducing several hadow cases from Turkey, Malaysia and Tajikistan (under the authoritarian regime).