Explorations in creole research with phylogenetic tools

Proceedings of the EACL 2012 Joint Workshop of LINGVIS & UNCLH, pages 89–97, Avignon, France, April 23 - 24 2012. c⃝2012 Association for Computational Linguistics Explorations in creole research with phylogenetic tools Aymeric Daval-Markussen Peter Bakker Research Centre for Grammar and Language Use, Aarhus University Research Centre for Grammar and Language Use, Aarhus University aymemeric@gmail.com linpb@hum.au.dk Abstract The principal goal of this paper is to illustrate various ways in which phylogenetic tools can advantageously be put to use in investigating and visualizing the relationships of creole languages to other languages, both creoles and non-creoles. After introducing a test study on the English-based creoles, the major theories seeking to explain the emergence and development of creoles will be reviewed and assessed. The final part of the paper is concerned with the typological status of creoles, where various samples will be used to show that creoles form a typologically coherent group among the world's languages. 1. Introduction Although the linguistic processes underlying creolization remain far from being fully understood, computational methods offer today the opportunity of uncovering complex mechanisms of language evolution on the basis of quantitative investigations. More sophisticated and powerful algorithms now available enable the visualization of patterns in a straightforward manner that was not possible before. Creole languages emerged in situations of intense contact between several languages, more often than not in the context of massive forced population displacements as were typical of European slave-trading ventures. A diglossic situation with a high-prestige variety (the superstrate) and several low-prestige languages (the substrates) characterized the settings in which creoles developed. Therefore, creole languages can be said to have several parents, and possess as well many often recurring features that appear ex nihilo. Thus, the problem of determining relationships between creole languages and other creoles or unrelated languages has long haunted creolists and been recognized as one of the challenges in the field. Following recent developments in creolistics, where phylogenetic networks were used to investigate questions inherent to the field (Bakker et al. 2011, Daval-Markussen 2011, Daval-Markussen and Bakker 2011), the aim of this paper is to argue that creole languages offer an unparalleled venue for exploratory research in language evolution, and that available computational tools now permit to graphically represent the relationships between the languages considered. In our demonstration, we will exemplify various ways in which phylogenetic networks may advantageously be used to visualize the results. Following the argumentation in Daval- Markussen (2011: 6-13), only structural features will be taken into account in the present study, since the lexical stock of a creole is mainly derived from a single source language, and this would likely be reflected in the resulting graphs. In the first part of the paper, phylogenetic tools are used to represent the relationships between 33 English-based creoles, for which 62 typological features were selected and encoded binarily1. The second part examines the various scenarios proposed to account for the emergence of creole languages in the light of phylogenetic networks. To this end, samples of various sizes and including …

Content truncated to 500 words