Nouwen (2010a,b) presents a puzzle involving the interaction of degree expressions and modals: a class of apparently unremarkable sentences expressing minimum and maximum requirements, obligations, desires, etc. are predicted to be trivially false or undefined, or receive otherwise incorrect truth-conditions. I suggest that the puzzle can be resolved if we treat the affected modals not as universal quantifiers over possible worlds but instead as scalar expressions which map propositions to points on a scale of expected utility. Independent arguments indicate that these modals are scalar, non-monotonic, and information-sensitive -- facts which are highly problematic for quantificational theories, but predicted immediately by the proposed sca...