This paper investigates the lexis associated with the church and the manor, two major institutions of medieval England, in order to assess the effects of language contact across lexical domains and levels of specificity. The data for this study, consisting of lexical items denoting locations and objects, was extracted from existing thesauri and analysed in terms of source languages, technicality, and survival, with the aim of finding out to what extent the influence of borrowing from French and Latin varied depending on the domain and level of specificity of meaning. The results reveal a consistent predominance of Romance-origin lexis in the domain relating to the church and of native English lexis in the manorial vocabulary across all levels of specificity, although some variation was found in the subsets relating to locations and objects. In addition, survival rates are relatively higher in the religious vocabulary; however, neither of the two domains displays a considerable difference in loss of lexical material between different languages during the Middle English period. The differences and commonalities identified between the two lexical domains reveal the ways in which language contact operated in two major multilingual contexts of post-Conquest England.