LEADERSHIP PROJECT PLAN: EXPANDING ACCESS TO PENSION FILES
Background
Ever since the American Revolution occurred, historians have contested its causes, meanings, and enduring consequences. Today, school board members and legislators have entered the debate over how we should remember the Revolution, who has the authority to dictate or recover historical memory, and what sources we should use to teach it. Several have tried to make a statement about United States history by passing bills censoring the narratives teachers share about the Revolution: they limit and often outright ban teaching about slavery and the Revolution. For their part, historians of memory and the American Revolution often put forward consensus-oriented narratives of the Revolution. Scholars like Sarah Purcell and Keith Beutler write of the “democratization of memory.” They contend that elites created a national memory with literature and public ceremonies that celebrated the “Founding Fathers.” In the mid-nineteenth century, when most of these elites died, Americans turned toward lower-class Revolutionary War veterans for commemoration. Marginalized actors including poor veterans, women, and African Americans took advantage of this shift and used elites’ celebratory rhetoric to include themselves in national memory. In doing so, they founded American history: “a shared history of common sacrifice.”
My own research and primary source body dispute this consensus-oriented approach to memory. In the National Archives’ 80,000 pension files, veterans from all across the United States, Canada, France, Ireland, and Africa use oral and material history to negotiate a memory of the American Revolution that often challenged the dominant narrative put forward by politicians, printers, and gentry historians themselves—one that still continues to influence textbooks and teaching across the United States. These pension files contain valuable information about early national debates over the meanings of disability, welfare, and immigration and naturalization. They can also make visible the complex social networks that veterans of the War for Independence forged within and between families, occupations, and religions. These files can reframe national conversations about the American Revolution; they provide fodder for new academic research, as well as teaching tools for secondary and post-secondary classrooms, community centers and historical societies, and genealogists.
Statement of the substantive problem
Several scholars and independent researchers have approached me with a desire to integrate pension files into their research, but they admit to being unable to access the pension records. For scholars to make use of the files by physically visiting the National Archives, they need the names of specific veterans to read about. Researchers who want to draw large data from the pensions as a whole cannot readily do so with the physical collections. Several pension files are available online through Ancestry.com’s service, Fold3. This database features digital images of microfilmed pension files; in fact, the website’s team has has digitized approximately 99 percent of Revolutionary War and Bounty-Land Warrant application files. But none of the files are transcribed. Optical Character Recognition (OCR) attempts to make available some of the names mentioned in each file, but it misses many names and mis-transcribes many others—not to mention, it entirely fails to make available battles and other key words (for example “uniform,” “smallpox,” “child,” etc.). To search the files, researchers must enter the pensioners’ name or home state. This system makes it difficult for researchers to approach the collections as a whole with key term searches or broad data (for instance, what percentage of applicants were from each state? What percentage had a given occupation or religion? When did applicants tend to marry—before or after the war? etc.). And, perhaps most challengingly, Fold3 costs $79.95 a year or $7.95 each month. Consequently, many historians, educators, and independent researchers struggle to afford and use these files.
Solutions: general project statement
My general solution is to launch a digital humanities project to transcribe the pensions, make them searchable, allow researchers to extract data from the collections at a macro-level, and make visible veterans’ complex social networks. This humanities project will also center community engagement by holding online workshops with independent researchers, local historical societies and archives, and community centers to discuss how the pensions might further their own local public history projects and educational programs. Project members will also reach out to schools by establishing example lesson plans to demonstrate how teachers in each state might incorporate the project into K-12 curricula, as well as examples of how post-secondary instructors might use the project to engage students in original research.
This is an example of one of the pension records I frequently work with: the application of Sarah Osborn Benjamin, who traveled with the Continental Army from 1780 to 1783.
Project plan
Below, I describe the project plan in three steps. It is important to note that this project is a large undertaking and will necessarily operate in several phases. Ideally, the project will eventually incorporate all pensioners from the American War for Independence and, later, the United States Civil War. To establish the project, researchers will first need to start with a much more narrow population. For example, a team might draw regional parameters by starting with all pensioners from a given state (Virginia, South Carolina, Massachusetts, etc.). Or they might create a demographic timeline, starting with all African American pensioners. After creating a stepwise plan for how to divide the pensions into manageable subsections, project members may repeat step two, transcribing and organizing sources, with each additional project phase. To be sure, as their pension sample expands, team members will also continue augmenting the skills learned in step one, applying for additional funding sources, and developing new community outreach workshops and lessons.
Plan Overview and Relationship to Key Insights
Step 1: Solidify necessary skills and funding
In order to create a digital humanities project, team members (my current team consists of university faculty and independent researchers) will need to establish a familiarity with their technological platforms. One free and accessible web-publishing platform for online archival projects is Omeka. This digital humanities web publishing platform, designed by George Mason University, is designed specifically for non-IT specialists. It has a Universal Viewer plug-in that will allow audiences to engage with several filetypes (including images of pensions, audio, etc.). It also integrates an inbuilt document search module and comment module to maximize audience engagement. Another useful platform for the project is Onodo, a free digital humanities tool that allows researchers to create interactive network visualizations. Onodo networks can be embedded into web platforms like Omeka, so viewers can engage with sources on both the microlevel (by studying individual pensioners’ language, memories, etc.) and macro-level (by studying broader community networks). Numerous universities and digital humanities institutes across the United States offer workshops in how to use Omeka and Onodo, which project members will use to facilitate conversations about how a technological platform can best make the documents accessible to viewers. The members of this project are currently in the process of applying for a summer digital humanities institute held by the University of Kansas and National Endowment for the Humanities, which offers training in Omeka as well as workshops on participatory and collaborative design, digital archive ethics, metadata, mapping and visualization, and copyright and intellectual property. In addition to developing the above technological expertise, project members will also need to secure project funding. University digital humanities institutes and local humanities centers offer smaller grants for projects that may provide foundation support. The National Archives also offers several grants for digital archival projects like this pension digitization project. Their National Historical Publications and Record Commission (NHPRC) grants include large funding opportunities for collaborative digital projects.
Step 2: Transcribe and organize sources
After securing adequate funding and training (ideally after year one of the project), we will hire additional team members to join us in transcribing and organizing the pension files. After narrowing our pension sample for a particular phase (for instance, Massachusetts pensions, African American pensions, etc.), we will consult with the National Archives to assess what sources are already digitally available to projects like ours and will travel to Washington DC to image additional files as needed. We will then individually transcribe the documents and verify them through a tandem-reading process. These transcriptions will serve as a foundation for our additional project components: historical context and social network visualization. After ensuring the accuracy of our transcriptions, project editors will tag key terms within each file: for instance, battles, places, people, books, organizations, etc. They will also tag each document according to broader categories (women’s history, African American history, material history) that viewers will be able to use to identify which documents are most relevant for their research. Finally, editors will broadly summarize the data offered within each pension application: when did the applicant marry, what is their occupation, what is their religion, etc. These summaries will allow us to eventually extract broader data about pensioners. After completing the transcription and editing process for each sample phase, editors will be able to use the people tags within each document to create community pension network webs with Onodo. They will build these webs regionally and then embed them in the site. Viewers will be able to see not only individual pensions, but also how pensioners fit into broader networks of veterans and witnesses. Team members will publish the files one phase at a time, repeating the step as they expand regionally, demographically, and chronologically.
Step 3: Develop community outreach
After publishing our first phase of the project, team members will assess state social studies standards in order to propose how educators might consider integrating the project into their lesson plans. We can post these example lessons and research project ideas/guiding questions to an education page on the project website. We will also promote the project by holding Zoom webinars for special research interests. For example, one webinar might engage historical societies and archives to demonstrate how the archives can further local social and demographic histories. We can also hold webinars on using the pensions to further studies of early American women’s history and African American history as well as genealogy. One goal of these webinars is to teach users how to read the pension files and navigate our web platform, to ensure our sources are as accessible as possible. Another is to equip community independent researchers and scholars to start their own projects based off of the source, in the process teaching students how to conduct research with the records and contributing original analysis to our studies of early America.
Evaluation
This is a long-term project, and because we aim to reach so many audiences (both academic and non-academic), we have many different options for evaluating success. Perhaps the simplest form of evaluating success is internal to the website itself. We can monitor website user metrics to learn more about who is accessing our site institutionally and regionally. We can establish target goals for audience growth based off of user numbers. For instance, perhaps we set a goal of having five-hundred viewers within the first three months after publishing documents within our first project phase. But meaningful interactions cannot be measured with purely quantitative metrics. To assess our website’s accessibility, we can include a feedback or survey section on the webpage. This survey might ask questions including: what are you using the pensions for (classroom engagement, personal independent research, academic research project, public history project, etc.), how can our web design or organization improve to better meet your research needs?, what additional resources can we provide?, etc. Finally, more personal dialogue with educators and researchers during our online workshops and professional conferences can allow us to better assess our audiences’ needs and adapt our project goals accordingly.
Current progress and applications
When I drafted the above steps, I was also working with a friend, Michelle Cook, to apply for the University of Kansas and National Endowment for the Humanities' Public Digital Humanities Institute. (For more on the Institute, see the end of Key Insight 3, "Reimagine the Archive.") Michelle studies African American Civil War soldiers. She, too, relies heavily on pension applications and has struggled to access them. When we realized that we share this archival problem, we decided to start a project to digitize the pensions of United States Colored Troops (USCT) servicemen from seven Missouri counties. Our project plan drew on the above steps, so our application (below) gave me a chance to apply my GLD reflection. Michelle and I will have the opportunity to acquire technological skills (step one) at the Institute's summer training session. Here, we will also receive help creating a financial plan and identifying grants to apply for (step one). This means we will be able to start implementing the plan very soon: we will take a week of classes with the Institute in early June 2022. During 2022 and 2023, we will continue to engage with Institute teachers and participants through online workshops before presenting on our project's progress in June 2023.