Yunhan Wen

I’m a Ph.D. candidate in Sociology at Princeton University. My research is driven by a  fundamental curiosity about the urban: how cities come into being, what they produce, and how late urbanization—especially after 1950 in the Global South—differs from earlier patterns observed in industrializing Europe and North America. This broad interest anchors both my empirical work and my theoretical orientation.

My dissertation examines how local governments in China built state capacity from a position of weakness to manage rapid urbanization without producing large-scale slums—an outcome that sets China apart from many late-urbanizing authoritarian regimes. Challenging the common view of the Chinese state as uniformly strong, I adopt a comparative approach that combines ethnography and archival research in two iconic cities at pivotal moments: Shanghai in the 1950s and Shenzhen in the 1980s. Despite starting with limited administrative resources, both municipal governments deployed similar strategies to govern urban growth, often relying on non-state actors to carry out state functions—a mechanism I call governing by proxy. While these strategies were broadly similar, variations in how they were implemented had lasting effects on local political economies. Through this comparison, I shed light on the micro-foundations of authoritarian state-building. I also contextualize the Chinese case in a broader trend of late urbanization in developing countries. 

While China has been an empirical focus for both personal and intellectual reasons, I believe the most powerful insights emerge through comparison and interdisciplinary conversations, and I strive to stay intellectually omnivorous. To better understand outcomes of development policies outside China, I’ve collaborated on research exploring sexual practices among older adults in rural South Africa during the rollout of antiretroviral treatment. To engage with methodological frontiers, I lead a collaborative project examining how large language models (LLMs) and computer-assisted text classification can benefit from the insight of traditional qualitative methods. You can find more about my work in my CV.