Cross-linguistic comparability and language-specificity in CorpAfroAs

One of the aims of CorpAfroAs is to allow queries within and across the language samples composing the corpus. Through the study of phenomena represented in several languages of the corpus (directional morphemes, case, and gender) we show that CorpAfroAs indeed allows the retrieval of a body of data amenable to cross-linguistic comparison, within the Afroasiatic phylum and beyond. However, given the annotation scheme of the corpus, the retrieval of relevant data has to rely on information given in the accompanying grammatical sketches.