Background/Context: Increasingly, instructional reforms in US schools place teachers in differentiated roles, such as literacy coach or data analyst. Not only do these roles hold promise for reforming instruction, but they also may make the teaching career more rewarding by offering teachers new challenges over time. In the past, teachers who held roles that distinguished them encountered resistance from colleagues who questioned their distinction and rebuffed their instructional advice (Lortie, 1975; Little, 1988). In doing so, colleagues defended their classroom autonomy and appealed to teaching's traditional egalitarian culture (Little, 1988; Johnson, 1990; Mangin, 2005). As veteran teachers retire, the teachers that take differentiated roles today may be relatively inexperienced and young. Thus, these teachers may violate teaching's allocation of privileges based on seniority (Lortie, 1975) in addition to autonomy and egalitarianism. Purpose: In this study, we set out to understand how second-stage (third-tenth year) teachers experienced differentiated roles in the current context of accountability, where instructional change is the order of the day. Participants: We selected 20 second-stage teachers who held differentiated roles that were formal, compensated by time or money, and promised to be ongoing rather than temporary. Our sample included teachers working in several metropolitan areas and in a range of school settings. Research Design: This is an interview-based, qualitative study. Data Collection and Analysis: Data were collected through hour-long, semi-structured interviews. Constructing matrices and writing analytic memos, we conducted cross-case analysis. Findings: Among the second-stage teachers we interviewed, only those whose roles sought to change colleagues' practice provoked resistance from their colleagues. Teachers who held these roles, which we designated "reform roles," reported that colleagues resisted their efforts to provide feedback and resented their special recognition, especially given their inexperience. Moreover, teachers in reform roles performed these roles strategically, in order to reduce colleagues' opposition. In some cases, the teachers in our sample reduced the scope of their role in an effort to avoid provoking their colleagues. We concluded that the norms of autonomy, egalitarianism, and seniority continue to exert great influence among teachers, whether veteran or second-stage. Conclusions/Recommendations: These findings suggest that instructional reform built on such roles should be designed and implemented with more consideration of the power and persistence of teaching's traditional norms. In this way, these roles might better promote whole-school instructional change and provide appealing career opportunities that retain teachers.