Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamily
In this paper, we present the results of our analysis of the 110-item basic wordlists for four reconstructed and one ancient languages, the linguistic ancestors of five language families which are hypothesized to constitute the Altaic (a.k.a. Transeurasian) macrofamily: Proto-Turkic, Proto-Mongolic,...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Cambridge University Press
2021-01-01
|
Series: | Evolutionary Human Sciences |
Subjects: | |
Online Access: | https://www.cambridge.org/core/product/identifier/S2513843X21000281/type/journal_article |
_version_ | 1811156409894567936 |
---|---|
author | Alexei S. Kassian George Starostin Ilya M. Egorov Ekaterina S. Logunova Anna V. Dybo |
author_facet | Alexei S. Kassian George Starostin Ilya M. Egorov Ekaterina S. Logunova Anna V. Dybo |
author_sort | Alexei S. Kassian |
collection | DOAJ |
description | In this paper, we present the results of our analysis of the 110-item basic wordlists for four reconstructed and one ancient languages, the linguistic ancestors of five language families which are hypothesized to constitute the Altaic (a.k.a. Transeurasian) macrofamily: Proto-Turkic, Proto-Mongolic, Proto-Tungusic, Middle Korean and Proto-Japonic wordlists. Protolanguage wordlists were reconstructed according to strict criteria of semantic reconstruction, based on accurate semantic glossing of forms in daughter languages. Each involved form was encoded into a bi-consonantal CC-shaped sequence using the consonant class method, after which a recently developed weighted permutation test was applied. In a typical situation, our algorithm makes a small number of type 1 errors (false positive), but the number of type 2 errors (false negative) can be substantial. Our main finding is that pairs between the Nuclear Altaic taxa – Turkic, Mongolic and Tungusic – as well as the Turkic-Japonic and Tungusic-Japonic pairs demonstrate significant p-values. In some cases, this can be attributed to either ancient contacts or genealogical relationships, but at least for the Turkic–Japonic pair, a contact scenario is unlikely owing to geographical remoteness. |
first_indexed | 2024-04-10T04:50:01Z |
format | Article |
id | doaj.art-00b58b4401c84e20a55b0adfe67cd69e |
institution | Directory Open Access Journal |
issn | 2513-843X |
language | English |
last_indexed | 2024-04-10T04:50:01Z |
publishDate | 2021-01-01 |
publisher | Cambridge University Press |
record_format | Article |
series | Evolutionary Human Sciences |
spelling | doaj.art-00b58b4401c84e20a55b0adfe67cd69e2023-03-09T12:32:19ZengCambridge University PressEvolutionary Human Sciences2513-843X2021-01-01310.1017/ehs.2021.28Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamilyAlexei S. Kassian0https://orcid.org/0000-0003-1129-2611George Starostin1https://orcid.org/0000-0002-4024-6625Ilya M. Egorov2https://orcid.org/0000-0002-1443-4438Ekaterina S. Logunova3Anna V. Dybo4https://orcid.org/0000-0002-6077-7183School of Advanced Studies in the Humanities, The Russian Presidential Academy of National Economy and Public Administration, Moscow, RussiaInstitute for Oriental and Classical Studies, National Research University Higher School of Economics, Moscow, Russia Santa Fe Institute, New Mexico, USASchool of Advanced Studies in the Humanities, The Russian Presidential Academy of National Economy and Public Administration, Moscow, RussiaInstitute of Linguistics, Russian State University for the Humanities, Moscow, RussiaInstitute of Linguistics, Russian Academy of Sciences, Moscow, RussiaIn this paper, we present the results of our analysis of the 110-item basic wordlists for four reconstructed and one ancient languages, the linguistic ancestors of five language families which are hypothesized to constitute the Altaic (a.k.a. Transeurasian) macrofamily: Proto-Turkic, Proto-Mongolic, Proto-Tungusic, Middle Korean and Proto-Japonic wordlists. Protolanguage wordlists were reconstructed according to strict criteria of semantic reconstruction, based on accurate semantic glossing of forms in daughter languages. Each involved form was encoded into a bi-consonantal CC-shaped sequence using the consonant class method, after which a recently developed weighted permutation test was applied. In a typical situation, our algorithm makes a small number of type 1 errors (false positive), but the number of type 2 errors (false negative) can be substantial. Our main finding is that pairs between the Nuclear Altaic taxa – Turkic, Mongolic and Tungusic – as well as the Turkic-Japonic and Tungusic-Japonic pairs demonstrate significant p-values. In some cases, this can be attributed to either ancient contacts or genealogical relationships, but at least for the Turkic–Japonic pair, a contact scenario is unlikely owing to geographical remoteness.https://www.cambridge.org/core/product/identifier/S2513843X21000281/type/journal_articleAltaic languagesTranseurasian languageslanguage classificationpermutation testAltaic hypothesis |
spellingShingle | Alexei S. Kassian George Starostin Ilya M. Egorov Ekaterina S. Logunova Anna V. Dybo Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamily Evolutionary Human Sciences Altaic languages Transeurasian languages language classification permutation test Altaic hypothesis |
title | Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamily |
title_full | Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamily |
title_fullStr | Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamily |
title_full_unstemmed | Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamily |
title_short | Permutation test applied to lexical reconstructions partially supports the Altaic linguistic macrofamily |
title_sort | permutation test applied to lexical reconstructions partially supports the altaic linguistic macrofamily |
topic | Altaic languages Transeurasian languages language classification permutation test Altaic hypothesis |
url | https://www.cambridge.org/core/product/identifier/S2513843X21000281/type/journal_article |
work_keys_str_mv | AT alexeiskassian permutationtestappliedtolexicalreconstructionspartiallysupportsthealtaiclinguisticmacrofamily AT georgestarostin permutationtestappliedtolexicalreconstructionspartiallysupportsthealtaiclinguisticmacrofamily AT ilyamegorov permutationtestappliedtolexicalreconstructionspartiallysupportsthealtaiclinguisticmacrofamily AT ekaterinaslogunova permutationtestappliedtolexicalreconstructionspartiallysupportsthealtaiclinguisticmacrofamily AT annavdybo permutationtestappliedtolexicalreconstructionspartiallysupportsthealtaiclinguisticmacrofamily |