Untangling Rhetoric, Pathos, and Aesthetics in Data Visualization

TL;DR

This paper investigates how rhetoric, pathos, and aesthetics influence data visualization, arguing that pathos has been underexplored and that a historical, etymological, and epistemological lens can guide more conscious design. It defines working terms, situates them within the Aristotelian and Campbell frameworks, and adapts Campbell's seven circumstances to visualize design strategies with concrete examples. Through a Greenhalgh-inspired meta-narrative review, it synthesizes interdisciplinary literature from computer science, social sciences, psychology, philosophy, and humanities to highlight how emotion, credibility, and beauty shape sense-making and persuasion in visuals. The contributions include working definitions, a historical-contextual framework, seven-circumstance adaptations with design implications, and actionable recommendations for balancing clarity, credibility, and emotional engagement in data visualization practice. This work aims to foster ethical and effective communication by making designers aware of the affective and aesthetic dimensions embedded in data representations, ultimately improving audience engagement and understanding across diverse contexts.

Abstract

In contemporary discourse, logos (reason) and, more recently, ethos (credibility) in data communication have been discussed extensively. While the concept of Pathos has enjoyed great interest in the VIS community over the past few years, its connection to similar but relevant concepts like aesthetics and rhetoric remains unexplored. In this paper, we provide definitions of these terms and explore their overlaps and differences in light of their historical development. Examining the historical perspective offers a deeper understanding of how these approaches in science and philosophy have evolved over time, offering a more comprehensive embedding into the design process and its role within it. Drawing from Campbell's seven circumstances, we illustrate how pathos is being used as a rhetorical device in data visualizations today, at times inadvertently.

Figures (2)

• Figure 1: This chart visualizes the progression of cited works in the paper, providing an overview of the historical context and distribution of references to clarify the paper's temporal citation landscape. Visualization examples were excluded, as this chart focuses on the theories and literature used for argumentation and analysis. *References prior to 2000 are widely spread, the exact years being: 360 BC, 1776, 1777, 1937, 1948, 1972, 1976, 1985, 1986, 1987, 1988, 1990, 1991, 1997, 1998, 1999 (3x).
• Figure 2: These six visualizations demonstrate how pathos techniques are implemented in data visualizations. They cover a range of topics, from global issues like urban greenery (A) MIT, and gender-based pricing disparities (B) pink, to a thematic analysis of the 2008 U.S. presidential debates (C) martinkr, and the human toll of the Iraq war (D) Scarr. Additionally, they include visualizations of historical events like Napoleon's march to Moscow (E) minard1869, as well as a contemporary personalized population calculator that connects global statistics to individual lives (F) wordldatalab2021. For full-scale images, please refer to the supplemental material. Visualizations A, B, C, D and F are reproduced with permission, visualization E is in the public domain .