Nominal classifications in languages are morphosyntactic systems that impose a classification on their nominal lexicon, as shown by the masculine/feminine/neuter grammatical gender distinction in German and other Indo-European languages, or the lexico-syntactic numeral classifiers widespread in East and South East Asian languages, such as Mandarin Chinese (Aikhenvald 2000; Seifart 2010). Linguists are interested in these systems due to their diverse lexical and pragmatic functions as well as their cognitive and cultural correlates. Most recent studies have shown that languages rely on different types of nominal classification to express related functions such as the expansion of the referential power of the lexicon and identifying participants in discourse (Contini-Morava and Kilarski 2013). The book under review here is an important contribution to these ongoing debates concerning the functions of nominal classification systems, since the use of grammatical gender in the Mawng (Iwaidjan) language of Northern Australia calls into question prevailing ideas about the functions of nominal classification. First, Singer demonstrates that unlike the gender systems observed in Indo-European languages, Mawng's gender has a strong semantic basis and plays an important role in the construction of meaning in discourse, rather than only being used in referent tracking. Second, in Mawng, gender agreement on verbs is frequently lexicalized, meaning that a verb tends to combine with arguments of a specific grammatical gender, creating a structure similar to noun-verb idioms. Since most research on nominal classification has focused on nouns rather than verbs, this rare phenomenon makes the book of great interest to linguists working on nominal classification, discourse analysis, and syntax, among other fields.