The premise of most Western thinking on counterinsurgency is that success depends on establishing a perception of legitimacy among local populations. The path to legitimacy is often seen as the improvement of governance in the form of effective and efficient administration of government and public services. However, good governance is not the only possible basis for claims to legitimacy. This monograph considers whether, in insurgencies where ethno-religious identities are politically salient, claims to legitimacy may rest more on the identity of who governs, rather than on how those people govern. Building on a synthesis of scholarship and policy regarding insurgencies and counterinsurgencies, the politics of ethnic identity, governance, and legitimacy, the author presents an analytic framework for examining these issues and then applies that framework to two detailed local case studies of American counterinsurgency operations in Iraq: Ramadi from 2004-05; and Tal Afar from 2005- 06. These case studies are based on primary research, including dozens of interviews with participants and eyewitnesses. In Ramadi, identity politics clearly trumped quality of governance in shaping the course of events. The grievances that fueled the insurgency had far more to do with a deep sense of disenfranchisement within Iraq's Sunni community and the related fear of sectarian persecution than it did with any failure in the government's performance. As a result, the evidence from this case points toward major limits to how much popular loyalty and legitimacy could be won through the improvement of governance. Other factors-namely security itself and identity-based concepts of legitimate rule (both tribal and sectarian)-appeared more decisive during the time of the case study. Moreover, the tribal "Awakening" movement that took hold in Ramadi the following year strongly supports this interpretation of events. The Awakening seems to have stemmed from two key changes in Ramadi and its surrounding Anbar province. First was the exhaustion of the population with violence and terror at the hands of Islamic extremists in their midst. Second was a new willingness of the Coalition to recognize the legitimacy of local tribal rule in spite of the sectarian tension this rule introduced between local and national sovereignty. Tal Afar's story is quite different, but suggests a similar conclusion. While the quality of governance mattered in the way both the population and the counterinsurgents perceived legitimacy, improvements in governance in Tal Afar were more a consequence than a cause of successful counterinsurgency. Without both the U.S. Army's dense presence in the city and its intensive focus on brokering compromises among the city's largely sectarian tribal conflicts, improvements in governance likely would never have taken root. Governance and political compromise between sectarian groups clearly reinforced each other there, but interviews with participants in the counterinsurgency in Tal Afar suggest that improvements in governance were of secondary importance in reducing violence in the city. The cases examined here yield ample evidence that ethno-religious identity politics do shape counterinsurgency outcomes in important ways, and also offer qualified support for the argument that addressing identity politics may be more critical than good governance to counterinsurgent success. However, the cases do not discredit the utility to counterinsurgents of providing good governance, and they corroborate the traditional view that population security is the most important element of successful counterinsurgency strategy. Key policy implications include the importance of making strategy development as sensitive as possible to the dynamics of identity politics, and to local variations and the complexity in causal relationships among popular loyalties, grievances, and political violence.