Corpus-based variationist linguistics
Lexical change in a century in Turkish
This paper aims to identify what archaic words/word groups were still known and used both among language speakers
and Turkish National Corpus (TNC) as an indication of lexical change in Turkish from 1900 to 2020. The present study explores the diachronic
variation of lexical change in Turkish by combining the corpus-based variationist sociolinguistic approach with the perspective of
historical sociolinguistics. The words/collocations thought to be outdated from the original version of “Eylül” novel, written in
1900, were selected and randomly subsampled using a computer-based randomization algorithm. A survey was formed using the outdated
words/collocations along with the context. The results indicated that demographical variables did not affect word knowledge and
that the archaic words were unfamiliar to all participants uniformly. The overall comparison of words/collocations tested in TNC
and survey indicated similar results as the most and the least frequently used words were also the most and least abundantly
present in TNC.
Article outline
- 1.Introduction
- 2.Methodology
- 2.1Corpus
- 2.2Data collection tool
- 2.3Statistical analysis
- 3.Results and discussion
- 4.Conclusion
-
References