Print exposure leads to individual differences in the Turkish aorist

Abstract

Several studies have established that not all native speakers extract the same generalization for a given construction due to speaker internal or external reasons, challenging a widely held assumption in linguistics. While there is a considerable number of studies investigating individual differences in grammatical knowledge in other languages, very little is known about how L1 Turkish speakers might manifest such differences in their linguistic knowledge. This is the first study to examine individual differences in the constructional representation of the Turkish aorist in adult L1 Turkish speakers. The aorist is known to be irregular and pose acquisition problems, especially when combined with monosyllabic sonorant ending verbs. The variants of the Turkish aorist have different corpus frequencies across spoken and written modalities. The study investigates to what extent differences in print exposure would lead to differences in how L1 Turkish speakers would apply the construction to monosyllabic-sonorant ending nonce-verbs. Based on the results, people with more written language experience extracted a more sensitive rule that applies to monosyllabic-sonorant ending nonce-verbs, such that they produced more -Ir than -Ar. Contrastingly, people who read less used more -Ar (r = –0.35), and print exposure accounted for roughly 12% of the variance. Our findings are compatible with usage-based approaches and suggest that print exposure-borne differences are pervasive in linguistic knowledge, adding to the growing body of evidence that challenges the convergence hypothesis.