Motivation for Study. Worldwide, teachers are often blamed for failures in curriculum reforms. Yet, do teachers actually shape reforms, or are teachers instead shaped by reforms? How may teachers’ roles in shaping or being shaped by reform processes vary from one education system to the next? These questions motivate my comparison of teachers’ roles in cases of curriculum reforms in Botswana and South Africa, adjacent middle-income southern African countries with divergent histories and educational outcomes. Although Botswana and South Africa have initiated a number of curriculum reforms, exemplified by processes that began almost simultaneously in the early 2000s, students from Botswana typically have had higher test scores than their South African counterparts. Divergences in the educational outcomes of the two countries point to differences in their socio-political histories, and raise questions about differences in their respective curriculum reform processes. My study specifically focuses on the roles of teachers in the reforms of the respective countries. Approach. This study began during my participation in a Spencer Foundation funded study of teaching quality in Botswana and South Africa’s North-West Province (NWP) in 2009 (see Carnoy et al., 2012 forthcoming). I present an organization studies perspective that builds upon prior political and sociological perspectives. I conceptualize curriculum reforms in SSA as processes, or sequences of “individual and collective events, actions, and activities unfolding over time in context” (Pettigrew, 1997, p. 338), with teachers being among other groups of actors that play roles. Within the framework provided by institutional theory on institutional processes (Scott, 2001, p. 93), I conceptualize curriculum reforms as multi-level policy-practice processes, which emerge from specific histories and occur in socio-political contexts that may differ from one country to another (Thornton & Ocasio, 2008, p. 114). I situate my study among process studies, which are more concerned with “a series of occurrences of events rather than a set of relations among variables” (Mohr, 1982, p. 54), and do not attempt to locate “singular causes” for outcomes (Abell, 2004, p. 296).