Introduction
The Journals of Lewis & Clark document the expedition of Meriwether Lewis and William Clark from 1804 to 1806. The journals detail their exploration of the American West, encounters with Indigenous peoples, natural observations, and daily travel challenges.
This project examines these journals to compare Lewis’s and Clark’s word usage and narrative focus. By analyzing their entries separately, the project highlights differences in emphasis and vocabulary. Using Voyant Tools, the results are visualized through interactive word clouds, trends, and word-link networks to make linguistic patterns accessible and interpretable.
Sources
The dataset comes from the Journals of Lewis & Clark, available via Project Gutenberg.The original text includes transcriber notes and introductory material that are not part of the journal content. To prepare the dataset for analysis, these sections were removed, leaving only the actual letters and daily entries. Using Python scripts, the text was split by the author—creating separate corpora for Lewis and Clark—and further cleaned by removing header lines and first lines within each letter (e.g., “Tuesday May 15th”).
Processes
Once cleaned, the texts were analyzed using Voyant Tools to explore patterns in word usage and textual relationships. Following tools were used for a deeper understanding of narrative focus, stylistic differences, and linguistic patterns.
- Summary: summarized the word counts and density.
- Word Cloud & Mandala: Identified the ten most frequent words in each corpus to highlight dominant vocabulary.
- Links: Visualized co-occurrences of these frequent words to explore contextual relationships within the text.
- Trends: Tracked the relative frequencies of these key words across the journals to show how topics appeared over time.
Presentation
The analysis is presented on a WordPress site hosted by my subdomain. The site layout uses multiple WordPress blocks to organize content clearly and make the visualizations interactive:
- Paragraph blocks provide context and descriptions for each section of the project.
- Group blocks display Lewis and Clark side by side, making it easy to compare corpus statistics and analysis results at a glance.
- Custom HTML blocks embed Voyant visualizations, using the HTML export code from Voyant.
The combination of these blocks allows the page to be both informative and visually engaging, presenting the results in a structured format.
Significance
This project demonstrates how digital humanities approaches can illuminate historical texts by finding linguistic patterns that may not be apparent through close reading alone. While word frequency and co-occurrence analyses offer quantitative insights, they require human interpretation to connect results to historical and cultural context. Unlike traditional data science, which emphasizes prediction and modeling, the DH approach emphasizes interpretation, context, and humanistic understanding. The project shows how digital tools can complement literary and historical scholarship, providing new perspectives on familiar texts while highlighting the potential and limitations of computational analysis.