The relationship between first language acquisition and dialect variation: Linking resources from distinct disciplines in a CLARIN-NL project

• A new measure taking four factors of lexical variation into account.
• TypeTokenRatio, Lexical complexity, Heterogeneity and Entropy.
• A new tool to define time and frequency of production in child language.
• Two combined measures are significant in a multiple linear regression analysis.

It is remarkable that first language acquisition and historical dialectology should have remained strange bedfellows for so long considering the common assumption in historical linguistics that language change is due to the process of non-target transmission of linguistic features, forms and structures between generations, and thus between parents or adults and children. Both disciplines have remained isolated from each other due to, among other things, different research questions, methods of data-collection and types of empirical resources. The aim of this paper is to demonstrate that the common assumption in historical linguistics mentioned above can be examined with the help of Digital Humanities projects like CLARIN. CLARIN infrastructure makes it possible to carry out e-Humanities type research by combining datasets from distinct disciplines through tools for data processing. The outcome of the CLARIN-NL COAVA-project (acronym of: Cognition, Acquisition and Variation tool) allows researchers to access two datasets from two different sub disciplines simultaneously, namely Dutch first child language acquisition files located in Childes (MacWhinney, 2000) and historical Dutch Dialect Dictionaries through the development of a tool for easy exploration of nouns.