In Justice and Reconciliation in World Politics Catherine Lu argues that structural reconciliation, rather than interactional reconciliation, ought to be the primary normative goal for political reconciliation efforts. I suggest that we might have good reason to want to retain relational approaches – such as that of Linda Radzik – as the primary focus of reconciliatory efforts, but that Lu’s approach is invaluable for identifying the parties who ought to bear responsibility for those efforts in cases of structural injustice. First, I outline Lu’s analysis of reconciliation, where she argues for the normative priority of structural approaches within the global political sphere, and propose that it will be useful to identify whether or not a relational account could instead identify underlying structural injustices. Second, I examine one particular relational account of reconciliation (based on Radzik’s account of atonement) and argue that this type of account brings to light underlying structural injustices of the kind Lu is concerned with. Finally, I identify an issue for relational accounts in identifying relevant responsible parties for reconciliation before returning to Lu’s structural account to address this gap.