Theoretical and practical issues concerning the justification of paternalism towards children are widely debated in a variety of philosophical contexts. The major focus of these debates lies either on questions concerning the general legitimacy of paternalism towards children or on justifications of paternalism in concrete situations involving children (e.g., in applied ethics). Despite the widespread consensus that the legitimacy of educational paternalism in important respects hinges on its principled, temporal and domain-specific limitation (e.g., via a soft paternalist strategy), surprisingly little has been said about conditions and criteria that determine what exactly (if anything) is morally wrong with paternalism towards children. This contribution aims to further the understanding of these normative issues by providing a critical analysis of the theoretical and methodological difficulties involved in developing context-invariant criteria for the identification of specific wrong-making features of paternalist rationales and paternalistically justified practices in cases involving children. I am going to show that the moral status of pro- and anti-paternalist reasons is much more context-sensitive than usually assumed by proponents of standard generalist justificatory strategies. In conclusion, my argument is that a moral particularist and casuistic framework may offer an adequate theoretical alternative to make sense of the context-dependent wrongs (and rights) of educational paternalism.