Amateur photographer Charles Cushman’s photo collection offers a unique portrait of America between 1938 and 1969. Cushman photographed landscapes, buildings, and points of interest using Kodachrome color slides. His work describes the evolving industries, priorities, and social changes of the pre and post World War II era. Cushman’s photos are compelling because they are taken from the perspective of a working American. Professional photos lack the intimacy and honesty found in Cushman’s album.
I chose to examine a dataset of Cushman’s photographs between 1940 and 1942. This set is collected and maintained by the Indiana University Bloomington Library. It was compiled after the Cushman collection gained notoriety in the early 2010s. The metadata was organized and shared to encourage digital history projects about place and space.
My goal was to create a tool that could showcase how Cushman’s photographic style and choice of subject evolved over time and location. I began by taking the raw data into OpenRefine. Although the data was neatly categorized, I wanted to eliminate unnecessary columns and combine location variables to create a set to serve my objectives. First, I removed columns like “Person”, “Roll Number”, and “Personal Names”. Next, I joined “City”, “State”, and “Country” columns to create a general location variable. I also refined the “Display Date” column to include year.
Since I hoped to create a visualization including the photos themselves, I needed to collect direct image links for each photo entry. To do this, I created a column based on the Indiana University Library link and selected the “Add column by fetching URLs” option. This option downloads the html code from each photo page which I could then use to retrieve each photo URL using a cell transformation.
Since I wanted to describe the Cushman collection’s progression over time, I chose to use the TimelineJS tool by Knight Lab. I felt like this tool could give users a chance to explore Cushman’s photos at whatever point between 1940 and 1942 they were most interested in. Also, the Timeline JS tool’s handy data template helped me structure my dataset in OpenRefine.
My final TimelineJS model had embedded photos, names, descriptions, and dates with a sleek design. I didn’t want the metadata to distract too much from Cushman’s photos themselves, so I only included essential context for interpretation. My final timeline is embedded below.
I was also curious as to where Cushman took his photos. I used Flourish to create a neat, interactive visualization to give users a better idea of where Cushman spent his film. To do this, I edited the dataset in OpenRefine to include only the state category. I believe this visualization helps to contextualize the overall set and to make the timeline more interesting. This graph is attached below.
I tweaked both the timeline and interactive plot several times before their final versions. I streamlined the design of the timelineJS interface and added a title slide to give it a more finished look. I also changed the size and color of the dots on the Flourish graph to make it easier to understand.
My approach to the data can shed light on how Cushman’s photo style and choice of subject developed over time. The Flourish visualization can also contextualize which areas of the United States Cushman is portraying. Instead of just looking at the images, this approach can place the photos in conversation with one another and situate them within their broader temporal context. My current approach is limited to exploring the photos over time. A more useful tool could include an embedded map or an option to sort the photos by location, environment, or category. My visualizations are fairly one-dimensional in this respect.
This project relates to Digital Arts and Humanities because it uses digital tools to tell a story. While Cushman’s photographic journey could be described in writing, these visualizations make his collection more meaningful and add an extra layer of context to these photos. Specifically, the timeline can help humanists place photos within their exact time periods in a way that a written contextualization cannot. DH tools like this one make the exploration of sets like the Cushman collection more accessible and informative.