The current phase of political conflict in Tibet began with pro‐independence protests in the late 1980s and saw a significant surge of unrest in 2008. But that unrest was not continuous and for much of the last 25 years was at a low level of intensity. Yet the Chinese authorities have categorised the situation in Tibet as a ‘life‐and‐death struggle’ against pro‐independence forces throughout this period. This paper notes earlier debates in Chinese history about political strategies for managing borderland peoples, including late imperial era attempts by Chinese officials to forcibly change Tibetan culture that provoked rather than assuaged conflict. It suggests that this happened again in the 1990s when a group of Chinese officials proposed policies that sought directly to change core cultural practices among Tibetans. These policies of selective cultural intervention, unprecedented in the post‐Mao era in Tibet, fuelled long‐term resentment, leading to the violence and unrest of 2008. The paper argues that these policies were inseparable from the institutional interests of the agency within the Chinese Communist Party, the United Front, which had promoted them, to the extent that its status and influence within the state bureaucracy depended on it preventing them from being challenged or reversed. It made cultural intervention in Tibet seem normative to the Chinese policy elite by invoking three interlocked imaginings about ways of managing borderland peoples – the perception of perpetual war, Han expertise at borderland management, and latent threat within borderland cultures. That these have led to the prolonging of conflict in Tibet for over a quarter‐century is a reminder of the importance of considering institutional dynamics in the analysis of ethnic conflict.