Alper Yagci is an Assistant Professor of Political Science at Bogazici University. He got his PhD in political science from the University of Massachusetts Amherst in 2016. His research interests are comparative politics, international political economy, public policy, and public opinion. He has published in numerous journals such as Comparative Political Studies, Party Politics, and Electoral Politics, among others. He also received the TUBA Outstanding Young Scientist Award in 2022.
Rival or Villain? A Machine-Learning Study of Media Content about Opposition during Turkey’s Authoritarianisation (joint work with Cemalettin Yılmaz and Uzay Çetin)
Abstract: It is increasingly recognized that democratic backsliding and electoral authoritarianization rides on affective polarization. Negative language towards the opposition may play a role in increasing affective polarization. We study this question in the context of Erdoğan’s Turkey. Inspired by Karl Schmitt's discussion of what counts as a valid opponent in politics versus an illegitimate enemy that needs to be destroyed, we distinguish between two types of negative language: rival versus villain. We track the evolution of the political language employed to describe the opposition in terms of the two types. To do so, we utilize machine-learning methods to examine at the sentence level the entire content of 2 newspapers from 2009 to 2023: Sabah, which has been a pro-government outlet all along, and Hurriyet, which started as closer to opposition and was later acquired by a pro-government business group. Our data allows us to compare the levels of depictions of the main opposition party as an villain and rival between media outlets, before and after ownership change, during election times and beyond. We also test how these depictions respond to currency shocks and military casualties, as the incumbents could try to contain the fallout from these adverse events by discursively attacking the opposition.