SOAS Research Online

Abstract

The question of which numbers are most easily borrowed, and in which contexts, has implications for an understanding both of historical change and language contact and of the extent to which the linguistic behaviour of numbers can be related to independent cognitive factors. In the Berber languages of North Africa, numbers are commonly clear-cut loanwords from Arabic; some languages retain as few as one or two non-Arabic number words, while others preserve a complete inventory. Closer examination reveals differences in intensity of borrowing even within single languages, depending on the numbers' functional usage. The languages in question are closely related to one another and are all influenced by mutually comprehensible varieties of Maghrebi Arabic, allowing what amounts to a controlled experiment, with extremely similar contact situations in different areas yielding a wide spectrum of possible outcomes. Careful examination of this spectrum allows us to set up a typology of numeral borrowing in Arabic-Berber contact, showing how linguistic, social, and cognitive factors all affect the process of number borrowing and how synonymy may emerge as an unstable transitional stage in the adoption of a new system.