Kinship terminology varies considerably across the world's languages, but this variation is constrained. It has been suggested that kinship terms evolve via linked sets of changes, such that a change in one part of the system will lead to a predictable change in another, related part of the system, thereby constraining the theoretical design space. If so, systems of kinship terminology should exhibit predictive structure: the terms in one part of the system should be informative of other terms. In this project, we tested whether predictive structure is universal for a sample of more than 1000 languages, and whether it emerges through biases for predictive structure that emerge during generalisation of known kin terms to new referents.
To find out more, you can read the abstract in the proceedings of the Evolang 2024 conference or have a look at a poster I presented at CogSci 2024.
One way that systems of kinship terminology vary is their number of terms. Even though the number of possible kin term referents is absolute, different languages carve up the space of referents into different categories of varying size. In this project, we are interested in the scope of this variation, the extent to which the number of kin terms can vary, and whether the size of the kinship lexicon is an adaptation to cognitive or cultural constraints.
To find out more, you can view the slides from my talk at the Cultural Evolution Society Conference 2024.
During my time at the HRAF Summer Institute for Cross-Cultural Anthropological Research, I conducted a short project testing whether societies who traditionally recognised more than two social genders were less likely to use gender as a distinctive feature in their kinship terminology. With a limited dataset of societies that definitively recognised multiple genders, the results were not conclusive. If you're interested in the data I collected for this project, feel free to contact me.