Schütze (2001) argues that Universal Grammar makes use of the so-called default case. These are “case forms used to spell out nominals that do not receive a case specification by assignment or other syntactic means” (Schütze 2001:205). This article proposes a theory of default case whose goal is to capture the phenomenon without adding anything to the theory beyond the defining property of the default (the lack of case features): after the DP exits syntax without case features, it is realised in the morphological component as any other syntactic object, namely by finding the best matching lexical item(s). I argue that this theory is not only the simplest one theoretically, but also the most empirically restrictive one. Specifically, once it is combined with the so-called cumulative case decomposition (Caha 2009), it restricts the range of possible values of the default: out of all cases, only the nominative (or absolutive) can be the default. This leads me to investigate languages with an apparent accusative default. This is a small set of languages with case on pronouns only. I argue that these languages, too, have a default nominative, but it is the nominative of a strong pronoun, which happens to be syncretic with the weak accusative pronoun.