Much attention has been paid in the literature to the writing and discourse structure of research articles, but the focus is predominantly on evaluating English research articles from an Anglo-American perspective. The unstated assumption of such studies is that writers whose first language is not English tend to write differently. Despite a number of cross-cultural comparisons in the literature, only surface discursive features have been examined, such as explicit devices that signal clausal relations, or the genre investigated has been stabilized with strong normative traditions. This study investigates how third party ideas are evaluated in research articles written in English and Chinese. The data come from three domains - the humanities, the social sciences, and the hard sciences. The results indicate that there are no large differences between English and Chinese in the weightings of evaluation categories, even though they are realized by different grammatical resources in the two languages. Chinese research articles tend to use more explicit linguistic resources with less frequent use of mitigators, possibly because of the socio-pragmatic and typological differences between English and Chinese. The study also proposes a taxonomy of evaluative resources in English and Chinese research articles, which can be used in future research.