Skip to the main content

There is no shortage of opinions about what should be done about hate speech, but if there is one point of agreement, it is that the topic is ripe for rigorous study. But just what is hate speech, and how will we know it when we see it online? For all of the extensive literature about the causes, harms, and responses to hate speech, few scholars have endeavored to systematically define the term. Where other areas of content analysis have developed rich methodologies to account for influences like context or bias, the present scholarship around hate speech rarely extends beyond identification of particular words or phrases that are likely to cause harm targeted toward immutable characteristics. 

This essay seeks to review some of the various attempts to define hate speech, and pull from them a series of traits that can be used to frame hate speech with a higher degree of confidence. In so doing, it explores the tensions between hate speech and principles of freedom of expression, both in the abstract and as they are captured in existing definitions. It also analyzes historical attempts to define the term in the United States, from the brief period of time when the United States punished hate speech directly. From this analysis, eight traits are surfaced that can be used for the development of a confidence scoring system to help ascertain whether a particular expression should be considered one of hate speech or not.

You might also like


Projects & Tools 01

Harmful Speech Online

The Berkman Klein Center for Internet & Society is in the third year of a research, policy analysis, and network building effort devoted to the study of harmful speech, in close…