Abstract The nature of word etymology, long a topic of interest in linguistics, has been considered to a much lesser extent in the word recognition literature. The present study created a database of overlapping words from the English Lexicon Project (ELP) and a database with the age of acquisition (AoA) norms which were categorized as either Germanic or Latin-based. Results revealed that Germanic words were learned earlier than Latin-based words. Germanic words also showed slower reaction times and higher accuracy relative to Latin-based words even when controlling for AoA, word frequency, and length. Additionally, analyses were conducted using a publicly available database that used the English Crowdsourcing Project (ECP) data with native and second language (L2) English speakers. The results with native speakers were similar to those collected with the ELP. However, nonnative speakers showed better accuracy and faster reaction times for Latin-based words compared to Germanic words. The findings support a bidialectal view of English in that Germanic words serve as the base of lexical processing during childhood, whereas Latin-based words fill in the lexical space across adolescence and into early adulthood. Furthermore, L2 speakers appear to acquire English via more advanced Latin-based vocabulary relative to native speakers. These results carry implications for theories of word recognition and the processing of lexical items in populations that come from linguistically diverse backgrounds.