Visualization

Visualizations are powerful, and I had never understood how powerful until this week. They are not just a nice picture of the Columbian Exchange, or a family tree. Visualizations are arguments themselves, as Johanna Drucker asserts in Graphesis, and they do not sit passively on a printed or web page. Everything from their layout to the color of text matters and impacts the argument in some way. Someone looking at and and using the visualization should approach it the same way as a scholarly work- critically. What is the argument? What are the methods? What is the data? Why present the information in this way and not another? Visualizations encompass the history of knowledge organization and the concepts of graphic design and cognizance of information. It is not a simple task to create useful and meaningful visualizations, let alone effectively employ them in history.

It’s important for a scholar to consider why a visualization might be useful for the argument at hand. Visualizations, when done right, can be very useful to building and presenting historical arguments. In “Visualizing San Francisco Bay’s Forgotten Past,” Matthew Booker argues that visualization “helps us recapture the forgotten past.” He uses maps of the bay to uncover its environmental history and to argue that these visualizations of how humans and animals have interacted with the bay can impact the future. In “The Image of Absence: Archival Silence, Data Visualization, and James Hemings,” Lauren Klein uses visualization in the form of a network to argue for James Hemings supposed absence in the archive of The Papers of Thomas Jefferson. These visualizations are explanatory, rather than exploratory, but they were integral to the respective arguments. The scholars furthered their points through the visualizations.

Both Drucker and Scott Murray in “Designing Kindred Britain” assert that visualizations can be explanatory, exploratory, or both. Visualizations in digital history can use an interface that allows for multiple points of entry. This user engagement with the material, as Murray notes, depends on the good use of graphic design and that exploratory interface. Kindred Britain, for example, not only provides the choice to investigate people, the connections between them, or the known stories, but it also provides multiple ways to then visualize that data. The authors used network analysis, a timeline, and a map to contextualize the information. Visualizations used in this way are more generative of research questions and open to interpretation.

It’s also useful to know what constitutes a bad visualization. Edward Tufte in Beautiful Evidence claims cherry-picking and chartjunk can make ineffective visualizations. John Theibault argues in “Visualizations and Historical Arguments” that “the challenge for visualization is to be transparent, accurate, and information rich” (paragraph 5). Cherry-picking involves using data that supports a point and doesn’t necessarily relate the context. Tufte states that there is an “evidence reduction, construction, and representation process” that comes between the data gathering and presentation stage and it’s important to be honest and transparent about that process (147). Chartjunk refers to charts that are void of content, or useful information. Just as historians should consider different points of view in their sources and provide a balanced analysis, so, too, should the visualization.