The "meaning through syntax" framework proposes lexical, decompositional representations of verb meaning. For several classes of verbs, the proposed representations have successfully predicted 2 types of data that pattern differently: the syntactic structures of sentences that are naturally produced by speakers and writers and the comprehension processing times of verbs and sentences. In addition, the framework assumes that syntactic structures carry meaning, and a particular meaning has been proposed for the reduced relative clause construction. Combining this meaning with the proposed meanings of verbs that might be used in reduced relative clauses explains why reduced relatives with some classes of verbs occur in natural production with near-zero frequency. K. McRae, M. Hare, and M. Tanenhaus criticized the meaning through syntax framework but offered no explanation for many of the empirical findings that support the framework.