Exploring the Emergence of Social Order through Computational Social Science and SNS Data

The emergence of social order has long been a central issue in social science, tackled by thinkers like Thomas Hobbes, Adam Smith, and Talcott Parsons. In this presentation, we focus on Parsons' theory, which describes the emergence of social order through the interaction between factual and normative behaviors.

1) Factual order: Regular behaviors emerge in society. For example, during the COVID-19 pandemic, many people regularly wore masks.

2) Normative order: People begin to believe that these regular behaviors are desirable and expect others to hold the same belief. For instance, seeing others wear masks, individuals may come to think that mask-wearing is socially desirable and that others agree.

3) Social order: A process of mutual reinforcement between factual and normative orders leads to the emergence of social order.

While Parsons' theory provides a clear conceptual framework, testing it empirically poses challenges. Cross-sectional surveys fail to capture dynamic processes, and even longitudinal surveys struggle to detail the emergence of social order. Case studies offer insight into the process but may lack generalizability.

By contrast, digital trace data offers a solution. This type of data is "always-on" (Salganik 2019), allowing us to track the emergence of social order over time. We propose using X (formerly Twitter) data in two steps: (1) identifying clusters of users who interact around specific topics, and (2) analyzing their posts and replies. If these interactions align with the three stages outlined above, we conclude that social order has emerged.

We will present detailed findings at the session.