The Relative Marker and Long Distance Dependencies in the L2 Acquisition of Swahili Relative Clauses

Of the several African languages that have become regularly offered at U.S. colleges and universities, Swahili consistently garners the highest enrollments, and is the second most widely taught after Arabic, with the greatest variety of instructional materials compared to its Sub-Saharan African counterparts. According to the Modern Language Association, the number of Swahili-language learners in the U.S. jumped by nearly 36% between 2002 and 2006 (Furman, Goldberg, and Lusin 2007). These greater enrollments have led to increased opportunities for investigations of how Swahili is learned by adult learners. The present study represents an initial investigation of how linguistic factors interact with proficiency in the second language (L2) acquisition of relative clauses in Swahili by native speakers of English. An eastern African Bantu language spoken from northern Mozambique to the south of Sudan, and through much of the eastern Democratic Republic of the Congo, Swahili (G42) exhibits the agglutinative verb structure typical of Bantu languages (e.g., Guthrie 1967-71). This verb structure, as it is implicated in relative clause formation, is the subject of the present study, which seeks to extend the discussion of relative clause acquisition to include this understudied language that is so widely taught. Thus far, linguistic analysis of the relationship between a null element (gap) and its lexically specified antecedent (filler) has proven useful in predicting processing difficulty of relative clauses for second language (L2) learners of languages like English (e.g., Eckman, Bell, and Nelson 1988, Gass and Ard 1980, Izumi 2003), French (e.g., Hawkins 1989), and Korean (O’Grady, Lee, and Choo 2003). These studies consistently demonstrate that L2 learners comprehend and produce subject and direct object relative clauses in a manner that largely parallels the noun phrase accessibility hierarchy (NPAH) outlined by Keenan and Comrie (1977). This hierarchy builds upon evidence in a number of languages and shows subject relatives to be the most dominant relative clause type, followed by direct object relatives, and others. Though studies on the L2 acquisition of relative clauses have largely demonstrated subject relatives to be the best comprehended, this research also acknowledges that mechanisms for this hierarchy are not thoroughly understood.