Quantifying superstrate and substrate influence

Abstract

To quantify the degree to which the structure of superstrate and substrate languages influence that of a creole, this paper compares the nearly 100 grammatical features of Guiné-Bissau Creole Portuguese surveyed in Baptista, Mello, & Suzuki (2007) with the corresponding structures in Balanta (one of the creole’s substrate/adstrate languages) and Portuguese (its superstrate), proceeding from one area of syntax to another. However, tables summarizing the presence or absence of features in each of the three languages are not organized by area of syntax but rather by the patterns of the features’ presence or absence in the three languages, allowing a quantification of the various patterns of influence, e.g. the percentage of cases in which a feature is found in the creole only, in the creole and its superstrate only, in the creole and its substrate only, or in all three. These percentages are then discussed regarding what support they might lend to the various hypotheses purporting to explain the sources of creole language structures: the influence of superstrate and substrate languages, universals, creole-internal innovations, and the convergence of all or some of these. The issue of Balanta being both a substrate and an adstrate language with many speakers bilingual in the creole is also discussed, as is the dated bias in the very grammatical categories chosen for the survey, which assumes an Atlantic creole prototype based on ‘Kwa’ rather than West Atlantic languages, the Niger-Congo subfamily to which Balanta belongs.