Just as we can be more or less certain that there is extraterrestrial life or that Goldbach’s conjecture is correct, we can be more or less certain about normative matters, such as whether euthanasia is permissible or whether utilitarianism is true. However, accommodating the phenomenon of degrees of normative certitude is a difficult challenge for non-cognitivist and expressivist views, according to which normative judgements are desire-like attitudes rather than beliefs (Smith, in: Evaluation, Uncertainty, and Motivation. Ethical Theory and Moral Practice 5: 305–320, 2002). Several attempts have been made on behalf of non-cognitivism and expressivism to meet the challenge (Lenman in: Non-cognitivism and the dimensions of evaluative judgment, Brown Electronic Article Review Service, 2003; Ridge in Synthese 2003. https://doi.org/10.1007/s11229-018-1884-7; Ridge in: Shafer-Landau (ed) Studies in metaethics, Oxford University Press, Oxford, 2007; Sepielli in Philos Stud 160: 191–207, 2012; Eriksson and Francén Olinder in Aust J Philos 94: 719–735, 2016). These attempts have all been found wanting (Bykvist and Olson in Philos Q 59:202–215, 2009, Aust J Philos 95:794–799, 2017; Bykvist and Olson 2012). Michael Ridge has recently offered a quasi-realist solution, according to which expressivists can say exactly what cognitivists say about certitude, including normative certitude. In this paper, we explain the basic problem and Ridge’s quasi-realist solution. We then argue that the quasi-realist account of normative certitude faces severe difficulties that do not arise for cognitivist accounts, according to which normative judgements are beliefs.