Although reputation analyses have been developed utilizing NLP technology, such studies have focused on web texts. Critiques with the humanities have not been regarded as primary targets for such analyses. The purpose of this study is to establish a basis for the deep semantic analysis of critiques. Movie and theater critiques, which are complete pieces of artistic output, were targeted, and all names, both for individuals and works, were extracted for analysis. As a result, the influence of certain individuals, the trends of certain factions and relationships to other genreswere revealed through frequency and network analyses. Moreover, the semantic and functional characteristics of pronouns within the critiques were extracted by using a database of personnel.