敬应奇
I am currently a visiting researcher at the Department of Linguistics and Philology, Uppsala University. Before that, I got my Ph.D. at the Department of Comparative Language Science, University of Zürich, with a background of evolutionary linguistics and linguistic typology. My research focuses on exploring the evolutionary dynamics of language syntax from two perspectives:
- comparing the distributional patterns in language usage against random baselines;
- modeling the evolutionary histories of typological features via phylogenetic inference.
A data-driven approach
We are living in a world that is defined by languages. We are communicating with the world via languages, and the languages we acquire may in turn affect the way we think and behave. To better understand our languages and ourselves, we need to a data-driven approach, especially in this digital age. I am open to all sorts of data (movie subtitles, Wikipedia databases, parallel corpora, dependency treebanks) in different languages, though they can sometimes be quite noisy. We need to preprocess these raw texts (tokenizing, tagging, and parsing) with pre-trained language models, and get a more structured format for further research. This corpus-based approach is important for us to get an idea of how our languages are developed and constrained.
An evolutionary approach
Languages are evolving! The languages we are using can tell us where we are from and where we are going. Like genes, languages also encode important information about our ancestral histories. Comparative approaches, like phylogenetic inference, allow us to estimate the language histories and detect the evolutionary trends of linguistic features in time and space. Like natural selection, those typological traits that are easy to process, produce or learn, will have a higher chance to be preserved and transmitted from generation to generation. Thus, it is necessary to model the dynamics of language systems and understand the driving forces in language change across the world’s languages.