Untangling Hate Speech Definitions: A Semantic Componential Analysis Across Cultures and Domains

TL;DR

This paper introduces a Semantic Componential Analysis framework to study how hate speech definitions vary across cultures and domains, revealing significant differences and the influence of definitions on detection models.

Contribution

It presents the first cross-cultural, cross-domain dataset of hate speech definitions and demonstrates how semantic components affect large language model responses in hate speech detection.

Findings

Definitions vary significantly across cultures and domains.

Many domains borrow definitions without cultural adaptation.

LLMs' hate speech detection responses are sensitive to the complexity of definitions.

Abstract

Hate speech relies heavily on cultural influences, leading to varying individual interpretations. For that reason, we propose a Semantic Componential Analysis (SCA) framework for a cross-cultural and cross-domain analysis of hate speech definitions. We create the first dataset of hate speech definitions encompassing 493 definitions from more than 100 cultures, drawn from five key domains: online dictionaries, academic research, Wikipedia, legal texts, and online platforms. By decomposing these definitions into semantic components, our analysis reveals significant variation across definitions, yet many domains borrow definitions from one another without taking into account the target culture. We conduct zero-shot model experiments using our proposed dataset, employing three popular open-sourced LLMs to understand the impact of different definitions on hate speech detection. Our findings indicate that LLMs are sensitive to definitions: responses for hate speech detection change according to the complexity of definitions used in the prompt.