People are often willing to embrace rights-restricting policies if they are seen to provide security or restrain out-groups. Such policies are typically framed as security benefits. What happens when the public is prompted to consider the costs of limiting human rights? Research in this field has rarely tested public responsiveness to an explicit defense of human rights. To shed new light on this, we address two related questions: What arguments can strengthen support for human rights of others? How much does the answer to this question depend on people’s attitudes towards those whose rights are affected? We test the resilience of supporting the protection of basic rights, and the reasons behind it, in a low-threat context. In a novel survey experiment of over 6,000 adults in Germany, we investigate attitudes towards police violence against peaceful protesters. We vary the argument against amnesty for excessive police violence and identify protesters as out-groups with an innovative research design. We find that pro-human rights arguments do not generally sway people’s opinion. But they do lower support for a rights-restricting policy in the most unlikely circumstances, if it is targeted against an out-group and if people have strong priors against human rights. Exploring justifications for respondents’ opinions reveals the chasm in human rights attitudes when the right-holder does, or does not, belong to an out-group. Our study paints a cautiously optimistic picture. A stronger human rights narrative might reach those who are otherwise least committed to human rights, but negative views of others poses the greatest obstacle to popular support for universal human rights.