Modeling Discrimination with Causal Abstraction

2025-01-14

AI Summary

This paper addresses a core tension in causal analyses of racial discrimination: race is both a socially constructed category and a potential causal variable in discriminatory processes. To resolve this, we propose a *causal abstraction framework* that models race as a higher-order abstract concept constituted by concrete social features—such as name, accent, and residential address—thereby enabling rigorous causal attribution of differential treatment while respecting its social construction. Crucially, we explicitly align the abstract concept with its constitutive features for the first time, rendering the “race-as-social-construction” hypothesis empirically testable and rigorously distinguishing constitutive relations from causal ones—clarifying longstanding conceptual disagreements. Integrating causal modeling, conceptual abstraction theory, and social ontology, we develop a falsifiable framework for discrimination assessment, supporting systematic evaluation of normative positions and empirical assumptions. This work establishes a formal foundation for causal fairness in AI.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

A person is directly racially discriminated against only if her race caused her worse treatment. This implies that race is an attribute sufficiently separable from other attributes to isolate its causal role. But race is embedded in a nexus of social factors that resist isolated treatment. If race is socially constructed, in what sense can it cause worse treatment? Some propose that the perception of race, rather than race itself, causes worse treatment. Others suggest that since causal models require modularity, i.e. the ability to isolate causal effects, attempts to causally model discrimination are misguided. This paper addresses the problem differently. We introduce a framework for reasoning about discrimination, in which race is a high-level abstraction of lower-level features. In this framework, race can be modeled as itself causing worse treatment. Modularity is ensured by allowing assumptions about social construction to be precisely and explicitly stated, via an alignment between race and its constituents. Such assumptions can then be subjected to normative and empirical challenges, which lead to different views of when discrimination occurs. By distinguishing constitutive and causal relations, the abstraction framework pinpoints disagreements in the current literature on modeling discrimination, while preserving a precise causal account of discrimination.

Problem

• Racial Discrimination
• Social Perception
• Causal Analysis

Innovation

• Racial Analysis Framework
• Causal Relationship of Discrimination
• Social Perception Influence