Many three-gendered languages have in common that some nouns are assigned conceptual gender – where the value of gender correlates with the interpretation of the noun – while other nouns are assigned arbitrary gender – where there is no such correlation. Strikingly, however, such languages do not always pattern together in how they resolve agreement with gender-mismatched coordinated nominals. If coordination resolution reflects feature representation, variation across languages with similar gender categories presents a puzzle. We hypothesize that resolution with gender-mismatched human and inanimate coordinated nominals is predictable from how properties like animacy and individuation are encoded within a language’s gender system. Focusing on Greek and contrasting patterns in Icelandic and Bosnian/Coratian/Serbian (BCS), we capture resolved agreement patterns through i) an interpretable vs. uninterpretable feature distinction, ii) a feature-geometric account à la Harley and Ritter 2002; and iii) universal coordination resolution mechanisms we refer to as percolation and conversion. Our system correlates resolution with other language-internal properties for gender agreement across the languages we investigate and captures complex patterns of resolution that have not been fully appreciated.
No takes yet. Share an insight, caveat, or question.
Adamson et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: