Clefts in Naija, a Nigerian pidgincreole

This paper is a corpus-based study of the various forms and uses of clefts in Naija, the largest West-African English lexifier pidgincreole, spoken in Nigeria and its diaspora as a second language by close to 100 million speakers. The data on which this paper is based is taken from the 500,000 word ANR-NaijaSynCor corpus, consisting of 300 samples of spontaneous speech, recorded in 2017 in 13 different locations in Nigeria, from 330 different speakers of both sexes, of various ages, education levels, and geographic origins. The quantitative data is taken from a sub-section of 9,621 sentences (almost 150,000 tokens) that constitute a syntactic treebank mirroring the social and geographic sampling of the full corpus. Clefts, pseudo-clefts and reverse pseudo- clefts are examined. Four types of clefts are described: wey-clefts, bare clefts, double clefts and zero-copula clefts. The properties of those clefting patterns are represented using a UD-type annotation scheme named SUD for Surface-Syntactic Universal Dependencies. The quantitative analysis of the data and comparison with former descriptions of the language underline the massive domination of bare clefts, and the emergence, among these various patterns, of a relative pronoun nãĩ “who/which” used only with cleft constructions, while the relativiser wey is being abandoned and specialises as relative clause operator.