Situational nationalism: National framings among Chinese travelers using data mining of online travelogues

Despite many predictions of its inevitable decline, nationalism has become a recurring phenomenon of the contemporary world. Recent studies of nationalism have treated the nation as commonsensical, performed, and embodied—as a powerful, flexible frame of action that persists most forcefully outside of full-throated pronouncements and dramatic moments. The background assumption is that people do not think through the nation all the time; rather, national thinking and feelings are triggered in variable places and times.

While qualitative work has associated nationalism with objects, events, holidays, spaces, landmarks, and consumer practices, there is an absence of comparative perspectives, using control groups of triggering phenomena to understand when nationalist framings arise and when they do not. There is also a lack of systematic tracing of independent and dependent variables, and larger datasets are rare. One solution to the problems of reliance on surveys and major events to interpret the nation as construct is to use naturally occurring data, which is better able to capture the behavioral and discursive expressions of the nation in actual contexts.

My proposed project will use quantitative analyses of a large dataset taken from naturally occurring data—online travelogues—to map out the types of spaces, places, and situations that elicit political reactions and emotions from Chinese citizens and compare them to non-nationalistic situations. By examining across the various situations expressed in travelogue form, this study can more precisely map the appearance of national frames and trace how national identities are invoked across social and geographic space.

To do this, the PI will use first use text mining tools to create a random sample of travelogues from the Chinese website Mafengwo. Next, these travelogues will be coded for various situations. Because of the sheer volume of situations, this project will look at keywords and phrases representing “peak experiences”, emotional moments where the travel diarists express surprise, shock, disappointment, pride, etc. Finally, once these situations are collected, they will be coded for relevant dependent variables (whether the nation is invoked, the valence [positive/negative/neutral] of the reactions, etc.) and independent variables (the type of situation, specific location, foreign/domestic encounter, etc.). Other modeled independent variables will include destination, travel
companions, gender of author, age, and length of trip. These variables will be placed in statistical models to determine the factors that are more likely to produce nationalistic responses than non-nationalistic responses, understanding how geography and individual factors produce nation-inflected moments.