Wordhood in Ghana-Togo Mountain languages, with a focus on Akebu and Adele

• Applicant: Dr. Andrey Shluinsky
• Subject Area: General and Comparative Linguistics, Experimental Linguistics, Typology, Non-European Languages
• Term: since 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 532192624

The project aims at solving the problem of wordhood for Ghana-Togo Mountain languages (GTML; < Kwa; West Africa), with a focus on Akebu and Adele. Extensive discussion of wordhood in linguistics during the last decades has shown that defining words as language-specific concepts is promising. At the same time, evaluating wordhood in underdescribed languages where it has not yet been analyzed in this perspective is urgent. The study will be based on my existing field data, as well as additional data collected as part of the project. As both Akebu and Adele are understudied languages with no published grammars, the project has a documentation dimension and includes a significant fieldwork component. The languages represent two distinct groups of GTML, and thus the project will develop a preliminary picture of this unity. Working hypothesis (i) will be tested for the two languages: (i) In a given language, there exist phonological and morphosyntactic features that correlate with each other and allow to distinguish between a limited number of possible morphological units and a limited number of possible morpheme boundaries. Combining (i), formulated in universal terms, with a narrow focus on specific data, the project has two goals: to make the analysis of the grammar of GTML non-arbitrary and consistent both if (i) is correct or not and to contribute to cross-linguistic studies of wordhood. I will focus on three research questions. RQ1: Is there consistent evidence of phonological wordhood in GTML? Is there a correlation or an implicational hierarchy between features evidencing phonological wordhood in GTML? The following phenomena will be used as criteria for phonological wordhood: vowel harmony; tone and stress; fusion and phonological reduction; pauses. RQ2: Is there consistent evidence of grammatical wordhood in GTML? Is there an implicational hierarchy between features evidencing grammatical wordhood in GTML? Possible criteria for grammatical wordhood are functional grammaticalization; inflectional marking; projection of an entity’s own full phrase structure; non-specificity and anaphoric islandhood; restricted use; linear insertion of syntactically independent material. RQ3: Is there a correlation between phonological and grammatical units in GTML? Are there criteria for distinguishing between morphological units simultaneously in phonology and in grammar? If in a particular language there are wordhood-relevant phenomena that correlate with one another both at the phonological and grammatical level, one can use this correlation for distinguishing between morphological units as a language-specific criterion. If there is enough of such language-specific criteria to establish a limited number of types of morphological units, RQ3 gets an affirmative answer and hypothesis (i) holds true for the language in focus. The planned published outcome includes 3 research papers submitted to peer-reviewed journals.

• DFG Programme: Research Grants