Reading children’s moral dramas in anthropological fieldnotes: A human–AI hybrid approach
Cambridge Forum on AI: Culture and Society · 2025
DOI: 10.1017/cfc.2025.10008 ↗Kurz gefasst
Fachartikel in der Zeitschrift Cambridge Forum on AI: Culture and Society, 2025. Verfasst von Jing Xu und 1 weiteren Autor:innen. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Themen: Kinderkrankenpflege
Abstract (Originalsprache)
Language AI has become a popular tool across the humanities and social sciences, but it has yet to gain traction in socio-cultural anthropology. Fieldnotes, the core data for anthropologists, present a unique opportunity and challenge for applying language AI to understand diverse human behavior and experience. Anthropological fieldnotes are communicative products in cultural contexts through immersive, extensive and idiosyncratic fieldwork. To read fieldnotes, anthropologists typically engage in qualitative, reflexive interpretations, attuned to local meaning systems and intersubjective encounters. This paper demonstrates a novel synergy, combining anthropological expertise and various AI technologies to analyze natural observation texts about children’s peer-interactions, especially their moral dramas, in the historical context of rural Taiwan during the Cold War. These fieldnotes were collected by the late anthropologists Arthur Wolf and Margery Wolf in the world’s first anthropological study focused on Han Chinese children. Engagement with AI in this project began as methodological cross-fertilization, transforming raw fieldnotes into a text-as-data pipeline and discovering how ethnographic close-reading, machine-learning techniques (e.g., unsupervised topic modeling), transformer models (e.g., S-BERT) and generative models (e.g., GPT) can complement and augment each other’s value. Capitalizing on the systematic nature of Arthur Wolf’s fieldnotes, as well as the special protagonists of these fieldnotes – playful children, the most voracious learners – this paper compares how children, the anthropologist and AI make sense of pretend-fight moral dramas. Such a human–AI hybrid experiment embodies layered-interdisciplinarity at methodological, epistemological and, to some extent, ontological levels, anchored at children’s social cognition. Situated at the intersection of anthropology, digital humanities, developmental science and data science, this work sheds light on the similarities and differences in how machines and humans learn and make sense of morality, and by doing so, critically reflect on the nature of socio-moral intelligence.
Zitieren
APA
Xu, J., & Hernandez, J. M. (2025). Reading children’s moral dramas in anthropological fieldnotes: A human–AI hybrid approach. Cambridge Forum on AI: Culture and Society. https://doi.org/10.1017/cfc.2025.10008
Vancouver
Xu J, Hernandez JM. Reading children’s moral dramas in anthropological fieldnotes: A human–AI hybrid approach. Cambridge Forum on AI: Culture and Society. 2025. doi:10.1017/cfc.2025.10008
BibTeX
@article{Xu2025,
author = {Xu, Jing and Hernandez, Jose Manuel},
title = {Reading children’s moral dramas in anthropological fieldnotes: A human–AI hybrid approach},
journal = {Cambridge Forum on AI: Culture and Society},
year = {2025},
doi = {10.1017/cfc.2025.10008}
}
EUVIMED