Hsiao Su-ying (Taibei, Taiwan)
A corpus-based diachronic study of Mongolian ‘say’ verbs
(57th Annual Meeting of the PIAC Vladivostok, 2014)
A corpus is a large and structural language database. Observations of a corpus can find facts and patterns of a language.
As research tools, I have compiled the following databases of Mongolian texts and lexicon: Databases of Manchu Shilu ‘Manchu Veritable Records’ (842 book images, with references to volume and page numbers), Databases of Mongolian monuments in Phagspa script (70 documents), Databases of Mongolian monuments in Uighur-Mongolian script (XIII-XVI Centuries) (3,156 sentences), Databases of Mongolian-English Dictionary (1960) by F. D. Lessing
et. al., Correlation databases of Mongolian Beijing Geser (50,131 words, 8,599 lines, 346 images), Correlation databases of Mongolian Longfusi Temple Geser (86,141 words, 14,021 lines, 550 images), Correlation databases of Mongolian Laoqida. (linked to corresponding texts in Manchu and 4 versions of Chinese Laoqida), Correlation databases of Erdeni-yin Tobčiya (5,788 lines, with links to corresponding sentences and Chinese translation), Correlation databases of Secret History of the Mongols (4,801 lines, 1,156 sentences, 282 sections, 1,220 images), Database of QadunUndusunQuriyangguiAltanTobci (4,413 lines), and Databases of KökeSüder (about 34,000 lines).
This paper introduces corpora of Mongolian texts and Manchu-Mongolian Parallel Corpora (http://altaic.sinica.edu.tw) compiled by the author and their applications in linguistic research on Mongolian ‘say’ verbs from a diachronic perspective.Data are drawn from the Secret History of the Mongols (1228), Mongolian monuments in Phagspa script and Uighur-Mongolian script (XIII-XVI Centuries), Manchu Veritable Records (1635), Erdeni-yin Tobčiya (1662), Mongolian Beijing Geser (1716), Mongolian Laoqida (1790), KökeSudur(1830-1891)and some modern Mongolian texts such as textbooks and published field reports.