Journal of Historical Linguistics - Current Issue

This paper deals with lexico-semantic stability, specifically in the anatomical domain. The main goal is to develop a method for measuring semantic polysemy and shift, in order to address: (1) the validity of standardized vocabulary lists (e.g., Swadesh 1950, 1952, 1955; Holman et al. 2008; Haspelmath & Tadmor 2009a, 2009b) for investigating cross-linguistic stability; and (2) the difference between basic and stable vocabulary (Ratliff 2006; Matisoff 2009), and its implications for studying remote relationships between language families, on the one hand, and subgroup differentiation within language families, on the other. To study these problems, a total of 50 etyma from the anatomical domain were selected from the Preliminary Etymological Mayan Database (Kaufman with Justeson 2003), and these were then classified employing the novel metric, and further analyzed by means of statistical methods. The results point to: (1) no specific correlation with the stability rankings of the Swadesh and Leipzig-Jakarta lists; (2) support for the “basicness” of etyma from the anatomical domain; (3) several significant relationships between stability and polysemy scores and independent variables relevant to the anatomical domain; (4) evidence of lexico-semantic stability score affinities between Mayan subgroups; and (5) evidence supporting the utility of polysemies to investigate subgrouping and language contact. The paper also offers conclusions and areas for further research.

The traditional view within Kurdish linguistics is that the split between Central Kurdish (CK) and Northern Kurdish (NK) is mainly the result of a Gorani substrate within the former group. More recent studies refute this hypothesis, arguing instead that Kurdish was initially composed of two distinct but closely related subgroups and that the differences between CK and NK are partly due to distinct source languages and partly due to ensuing contact with neighbouring languages. This study aims to shed new light on the Gorani-substrate hypothesis within CK by examining a corpus-based study of the southernmost CK dialects located within the historical Gorani heartland. Combining recent historical accounts of language shift from Gorani to CK in the region with linguistic data, the paper claims that (i) Gorani borrowings and substrate features reflect different layers of historical contact in Southern CK dialects and (ii) the Gorani substrate has caused a split in the morphosyntactic features across vernaculars of CK, showcasing second-language learning in shaping the historically recent development of Southern CK dialects.

Relationships between universally recognized language families represent a hotly debated topic in historical linguistics, and the same is true for correlation between signals of genetic and linguistic relatedness. We developed a weighted permutation test which represents the classical permutation tests with weights introduced for individual Swadesh concepts according to their typological stability. Further, the obtained values were calibrated on a negative control group to override non-uniform distribution of phonemes within the Swadesh wordlist. We applied the calibrated permutation test to the basic vocabularies of nine languages and reconstructed proto-languages to show that three groups of circumpolar language families in the Northern Hemisphere show evidence of relationship through common descent or borrowing in the basic vocabulary: [Chukotko-Kamchatkan, Nivkh]; [Yukaghir, Samoyedic]; and [Yeniseian, Na-Dene, Burushaski]. The former two pairs showed the most significant signals of language relationship. Our findings further support some hypotheses on long-distance language relationships previously put forward based on linguistic methods but lacking universal acceptance.

Kassian et al. (2023) propose a method to detect ancient linguistic relationships by means of a weighted permutation test applied to word lists. The method provides evidence for an ancient historical connection between Chukotko-Kamchatkan languages and Nivkh languages, and between Samoyedic languages and Yukaghir. In this paper, I argue that while the word lists collected by the authors represent a valid dataset for language comparison and are supported by extensive etymological work, their statistical tests constitute a departure from the significance testing framework that has propelled contemporary literature on long-range comparison. I then advance some proposals on how their tests can be integrated and compared with those traditionally accepted in the literature.