Data Visualization, Old School:
A Review of W. E. B. Du Bois’s Data Portraits
So it has been a hot minute. I’m not even going to get into how long it has been since Meg and I have written a book review. Here is what I will say on the topic. Life happens. My dad passed, Meg and I were both starting or ending different UX jobs, and things just came up. The good news is that we didn’t stop reading UX books or having informed opinions about them. We also didn’t stop sending each other gifs about the terrible people in our lives — see attached. We just lost bandwidth to keep writing these posts for a while; we’re sorry. We appreciate those who have joined us in the last few months and continue to show interest in the books we’ve reviewed. Thank you.
I figured upon my return to writing reviews I would pick up with perhaps my favorite UX related book that I’ve read of late, W. E. B. Du Bois’s Data Portraits: Visualizing Black America. It is published by Princeton Architectural Press, one of my favorite publishers for UX and UX-adjacent books. I say this because every book of theirs that I own is beautifully printed, well-constructed, and is a book that I return to multiple times. Goodness me, just pulling up the website now to put the link in this post has incited a buying spree. Just check out all of their books on design and I dare you to not buy one.
W. E. B. Du Bois’s Data Portraits: Visualizing Black America is not one to disappoint. Before I get into the contents of the book, I think it is important to state that I’m not going to get into the history of Du Bois, or anything controversial about his interesting and opinionated life. The editors of the book do not get into it, and I think that it is important to frame the visualizations in the book around the art, and not about the life of the man making the art.
I picked up this book because, as I said above, I clearly have an addiction to Princeton Architectural Press books. I also really love books that involve visualization before the time of computers. The examples that I have of non-computer-generated vis are few and far between (and mostly in Tufte books). The reason why there is a dearth of hand crafted pre-1970’s data visualizations is because visualizations were stupid consuming. To create something that visualizes data costs time, materials, and a level of obsession that few can muster without the everyday capabilities of modern Microsoft Excel. Du Bois, though, had a whole team of people that he co-opted to convey his very important message. In fact, the book shows pictures of people working on the visualizations with their noses to the desk cracking out work. I do not have a whole team of college students at my beck in call to spend weeks crafting visualizations. I have D3 and Excel. The difference between our situations is what interests me. What time, creativity, and a lack of abundant computer tools can create forces the designers to put message first rather than convenience.
In this book the editors Whitney Battle-Baptist and Britt Rusert contextualize the visualizations that Du Bois had presented at the Paris World Fair in 1900 in an exhibit titled The Exhibit of American Negros. The Du Bois and contributors frame the exhibit around the argument that African Americans were equal to their white counter parts and that slavery was the central cause of continued black oppression. To make this argument Du Bois, college students, and others worked, penned, and colored vivid visualizations that left little argument to be had. The visualizations said it all.
The book starts with a framing of how Du Bois made the visualizations and why the arguments were necessary to make. It then moves into a reproduction of each visualization that was presented along with useful descriptions. The visualizations are also split into roughly two sections with one focused nationally and one focused on a particular region.
The book took maybe 2 hours to read, but even 2 months later I enjoy picking it up and revisiting the visualizations because they are so well done. What I really should say is that the visualizations are masterful. Each one tells part of an overlapping story, like waves crashing, that repeatedly hammers in the theme of black oppression and a resurgence while also demonstrating growth and emergence. This book demonstrates that when done well, data visualization does more than make numbers understandable, it tells a story.
Beyond what I learned about the state of African Americans in the 1900s, I spent a considerable amount of time learning about how to be creative with visualizations. Du Bois didn’t just get creative, he invented many visualizations to properly tell his story and used colors to point to themes that he wanted to relay. Reading the book made me think about what stories I want to tell about the data I work with. It made me question how much more I should be doing; that I don’t need to turn everything into a bar chart. It made me question how transient my visualizations are (since many deal with streaming data) and what visualizations last the test of time.
Good books are like good visualizations. They do more than make you think about the subject; they make you introspective about your craft and your life. This book hits the mark.
