Rapid automatized naming (RAN) is a strong predictor of reading across languages. However, it remains unclear if the effects of RAN in first language (L1) transfer to reading in second language (L2) and if the results vary as a function of the orthographic proximity of L1–L2. To fill this gap in the literature, we examined the role of RAN in reading accuracy and fluency in eight languages representing different writing systems. Seven hundred and thirty-five university students (85 Chinese-, 84 Japanese-, 100 Kannada-, 40 Oriya-, 115 English-, 115 Arabic-, 105 Portuguese-, and 91 Spanish-speaking) participated in our study. They were assessed on RAN (Digits and Objects) and reading (accuracy and fluency) in both L1 and L2 (English). Results of hierarchical regression analyses showed significant effects of L1 RAN on L2 reading accuracy in the Chinese-, Portuguese-, and Spanish-speaking groups. In addition, L2 RAN was a significant predictor of reading fluency in L1 in the same language groups. No cross-language transfer was observed in the other languages. These findings suggest first that L1 and L2 RAN capture similar processes and controlling for one does not leave unique variance for the other to explain. Second, to the extent there is cross-language transfer of RAN skills, this appears to be independent of the orthographic proximity of the languages.