This book explores the transformative power and inherent risks of text mining in historical research. It examines how the quantitative analysis of words over time, a technique known as text mining, enables new forms of historical inquiry, while simultaneously issuing a stark warning about the potential for distortion and error when data science lacks humanistic guidance. The author presents case studies of both flawed applications and groundbreaking collaborations between humanists and data scientists, illustrating how text mining can reveal shifts in historical understanding, such as the fading public memory of slavery or the congressional silencing of environmentalism.

The book is positioned as a critical examination of text mining's application in digital history, highlighting both its potential and its pitfalls. It aims to bridge the gap between quantitative data analysis and qualitative humanistic interpretation, suggesting that a collaborative approach is essential for accurate historical understanding. The work likely appeals to historians, digital humanities scholars, and data scientists interested in the ethical and methodological considerations of analyzing large textual datasets.