Implosives in Bantu A80? The Case of Gyeli

Implosive consonants in Bantu A80 languages are widely attested in the literature. The status that specific authors assign to them, however, differ significantly, ranging from mere phonetic contrasts to phonemic status or even absence in certain languages. Given this variety of language analyses, along with a controversy about necessary and sufficient features of implosive sounds, this paper aims at reassessing the range of implosives and non-implosives within A80 and especially Gyeli (A801). I show that though implosives are expected in Gyeli from previous literature, these sounds are better described as pre-glottalized stops with a relatively long prevoicing time. That raises the question whether this analysis might be more appropriate for other A80 languages as well. While this paper cannot provide any conclusive answer on the latter question, it hopes to raise awareness of the methodological problems associated with the present description of A80 implosives, encouraging a systematic re-evaluation of the data. It also encourages a discussion on how the general fieldworker should go about describing implosive(like) sounds.