Two different event-related potential (ERP) components were used to investigate the temporal processing of semantic and phonological encoding during implicit picture naming. Participants were shown pictures and carried out a dual choice go/nogo decision based on semantic information (i.e., whether the picture was of an object or an animal) and phonological information (i.e., whether the picture's name starts with a vowel or a consonant). In addition to the already established lateralized readiness potential (LRP; related to response preparation), we introduce the N200 (presumably related to response inhibition) as a tool for measuring online language processing. Both, the LRP and the N200 data indicated that semantic processing began earlier than phonological processing. The data are discussed in the context of language production models. Therein, the LRP and N200 results, taken together, favor a serial or cascaded processing model of language production in contrast to a parallel processing account.