In recent years a rapidly growing literature from history (Beinin, 2016), political science (Buehler, 2015; Hartshorn, 2018; Zemni, 2013) and sociology (Allinson, 2015; Feltrin, 2018.; Yousfi, 2018), has argued that unique success of Tunisian democracy after the 2011 Arab revolutions is best explained by the strength of organized labor. This supports the long tradition of comparative historical work arguing for the working class origins of democracy (Bellin, 2004; Rueschemeyer, Huber, & Stephens, 1992; Therborn, 1977; Usmani, 2018). The relative autonomy of the national labor federation in Tunisia (UGTT) meant that workers in Tunisia were able to participate in revolutionary and post-revolutionary politics en bloc, rather than by neighborhood, soccer club or workplace (Beinin, 2016). However existing scholarship has been less successful in explaining why it is that workers were able to obtain this autonomy in some places but not others. In the Tunisian context, it has been common to explain the coexistence of labor autonomy and dictatorship by pointing either to the fact that the UGTT was founded prior to independence, or to its nationalist legitimacy obtained by its role in the independence struggle (Beinin, 2016; Bellin, 2002). However these explanations fail on two levels. First, while it is true that the UGTT obtained its autonomy under French rule, this avoids rather than answers the question because it is no less of a puzzle to ask how labor achieved such a high level of strength under French authoritarianism. Second, they ignore the UGTT’s capture by the state and decline to near irrelevance between 1956 and 1969 (Toumi, 1989); obtaining autonomy from one despotism is no guarantee of retaining it under another. To address the question of labor autonomy, this proposal will offer a comparative historical study using a mix of archival sources and interviews to study three periods: the French protectorate and the Bourguiba and Ben Ali regimes. I will try to argue that this phenomenon requires understanding the interaction of elite political conflict and structural changes in the economy. This allows us to build towards a model capable of explaining why Tunisian labor failed so poorly in the 1960s, but relatively well since the 1990s, which in both cases goes strongly against broader global trends in both rich and poor countries.