In this project we use comments collected on an online forum to describe the effect of a series of terrorist attacks in Germany during the summer 2016 in the levels of online hate speech. Using the terrorist attacks as a natural experiment, we examine the effect of terrorist events on the levels of online hate speech against immigrants, and other minorities. We also elaborate a theoretical argument that explains the increase on the level of online hate speech as a relaxation of social norms that regulate hate expression, and develop observable implications.
In the first part of the study we use pre and post attack data to identify a breakdown of social norms on hate speech expression. Hateful comments towards immigrants increased after the attack in the baseline treatment. We interpret the findings as a breakdown of norms against the public expression of prejudice towards immigrants.
In the second part of the study we compare the effectiveness of censoring prior hate content as an intervention to reduce hostile content before and after the attacks. We argue that censoring hate content biases the individual’s perception of the prevalence of hate speech, therefore highlight a descriptive norm against it. We find that following the attacks the effect of contextual normative cues is larger, suggesting that saliency of the topic, e.g. refugees, interacts with how people look for normative information in their context and how they ultimately use this information. In general, we find that people are highly responsive to changes in the perceived social acceptability of prejudice expression, and vary behaviour accordingly.
Implications of the findings for public opinion dynamics after terror attacks, online hate, and the literature on social norms and communication are discussed.
Alvarez, A., & Winter, F. The Breakdown of Anti-Racist Norms: A Natural Experiment on Normative Uncertainty after Terrorist Attacks