This article presents a case study of an instance in which the influence of linguistic theory on descriptive fieldwork has led to both the discovery and the remedy of missing gaps in the documentation record of a language. It focuses on the restriction of wh - in-situ induced by intervention effects in Krachi, an endangered Kwa language of Ghana. Investigating Krachi intervention effects both enriches the depth of description of wh - constructions in the language and reveals patterns of intervention effects that differ from what has been documented in other languages in the literature. The Krachi data therefore provide a new set of empirical challenges for current theoretical accounts of intervention effects and thus help to set the theoretical agenda for further work. This case study thus supports the position that the relationship between linguistic theory and language documentation and description is a symbiotic one in that each complements and drives progress in the other.