Getting Primaried
The artificial way that “primary sources” are packaged for history classrooms doesn’t help students make sense of the information they’re bombarded with in the real world.
“One of them is lying.”
That’s how a student in my U.S. history survey class explained the discrepancy between two eyewitness accounts of the Boston Massacre given as testimony in the subsequent trial of the British soldiers. This student thought that the soldier had an agenda – he wanted to be found not guilty – and thus was “biased.” Because his words could not be trusted, the “true” account of events must have come from the colonist testifying for the prosecution.
Never mind that both accounts were offered months after the event being described, leaving plenty of time for memories to be muddled by trial preparation, or by hearing competing accounts of that day. Never mind about the fog of war, and the likelihood that perspectives were clouded by the adrenaline of dodging bullets and rocks in a melee. In the moment, I resorted to a sports metaphor. The quarterback, the defensive line coach, the radio announcer, and a fan in the nosebleed section all experience the same game. But they all experience it differently, and may recall divergent narratives of the game, without any intent to spread disinformation.
That, though, can be hard to remember when the popular discourse surrounding us frames politics and media in terms of lies, fake news, interest group agendas, and bias. Whether they are highly attuned to current events or casually picking up on the zeitgeist of the times, students can hardly be blamed for applying the “bias” framework to the people and events they read about for class. This media milieu, in fact, was one reason I decided to include primary source analysis in my class, even though it had not been an important part of my own history education. I imagined how difficult it might be for young people to build the skills of media literacy when they were the targets of disinformation that is often framed in ways intended to discourage critical analysis and engagement.
As an historian of media and culture, I see useful parallels between analysis of historical sources (both primary and secondary) and the skills involved in media literacy. Understanding a resource requires reflecting on the conditions of its production, distribution, and reception. It also requires considering the possibilities and limitations of the resource’s form and the conventions of its genre.
None of this can be done without also considering the ways people have contested its meanings or used it to shape their understandings of its contexts. The key idea, though, is context, not source. In the work historians do, understanding a primary source is neither the first step nor the ultimate goal. It is, instead, a process that is interwoven with other reading and analytical work. Yet most of the materials designed to help history educators integrate primary sources into their classrooms do not support this approach.
The textbook I had assigned for my course presented primary source excerpts that had been separated out from the narrative. The excerpts were paired in text boxes labeled “Dueling Documents,” and marked by a drawing of crossed swords. This presentation was rife with problems.
Framing the comments on the relationship between social equality and economic opportunity by feminist and social reformer Frances Wright and political philosopher Alexis de Toqueville as “he said, she said” detracted from the ideas of both writers. Pairing plantation mistress Mary Boykin Chesnut’s discussions of cruelty and sacrifice with Harriet Jacobs’ narrative of her enslavement risked devaluing complex experiences as mere difference of perspective. One of the excerpts did present an instance of actual disinformation – Jefferson Davis’s post-war claim that the Civil War was not about slavery, offered as a legitimate opinion alongside Alexander Stephens’s earlier assertion that secession was about slavery – undermining broad historical consensus and potentially strengthening the effectiveness of people who still spread Davis’s falsehood today.
Nor is the adversarial framework unique to primary source pedagogy. The same publisher has a long-running series of textbooks called “Taking Sides: Clashing Views,” that pit excerpts of scholarly arguments against one another on questions such as: “Were the first colonists in the Chesapeake region ignorant, lazy, and unambitious?”
Presenting resources as competing viewpoints may seem like a harmless simplification of the more complex reality that historical sources, and the scholarship that draws on them, are often in conversation with each other. But it misrepresents what historians do with sources. Rather than encouraging contextualization, the pairing of conflicting sources suggests that the goal is simply to judge their utility based on an evaluation of their accuracy. (Even if it were, two sources do not give readers much information with which to make such an assessment.)
Similar deficiencies lie under the surface of one of the other most common styles of presenting primary sources: the “primary source reader.” This model, used for the past 50 years or so, features a selection of excerpts accompanied by short explanatory introductions that answer all the who, what, when, where, and why questions indiscernible from the resource itself. It is tremendously difficult to analyze a single primary source excerpt in isolation, and the introductory paragraph gives just enough information to overshadow the source itself in classroom discussion.
Some such readers are designed to accompany a particular textbook, but the yawning gap between a 17th-century sermon and grand narrative synthesis does not lend itself to productive discussion either. The synthesis may help students evaluate the sermon, but the sermon probably does not help disrupt or complicate the synthesis. At best, a source excerpt may serve as a superfluous illustration; at worst, it may seem unconnected altogether. In the end, students and educators are still left with the task of trying to evaluate a source based on its internal clues, which neither represents how scholars approach their source material, nor gives students necessary practice contextualizing sources of information.
A more recent trend toward the creation of “primary source sets”– curated clusters of resources relating to a particular topic – seeks to avoid this limitation. The premise is that juxtaposing several sources allows them to provide context for one other, and to invite more nuanced observations about similarities and differences. Some educators facilitate inquiry-based learning in which students are asked to formulate research questions based on primary source sets.
While this approach may work in a research class, applying it to the scope of a survey class raises the specter of the dreaded “DBQs,” the document-based questions familiar to high school students who have faced the AP History exam. A handful of sources, unfortunately, does not contextualize itself. I learned this lesson early, when I took the AP exam and was confronted with a set of sources from the 1950s, inscrutable to me because my class had only gotten as far as the bombing of Pearl Harbor. Tackling primary sources as an introduction to an historical theme seems artificially difficult when contextual reading typically precedes and intermingles with an exploration of sources.
The collection that comes the closest, in my experience, to reproducing the ways that historians encounter primary sources in the wild is the Bedford/St. Martin’s reader Going to the Source. Each chapter is centered around an event or theme and a type of resource (advice literature for parents and children in the early republic, newspapers covering the Pullman strike, post office murals from the New Deal, etc.). Each chapter provides a cluster of similar resources, their historical context, and strategies for analyzing resources of that type. While this model holds a lot of potential, the time it takes to effectively use these sorts of source sets can make integrating them into a survey class quite difficult.
The desire to find short, self-contained, high-impact sources to analyze within the compressed schedule of a survey class must be one reason that many educators are drawn to the political cartoon. The very qualities that make it an effective form of political commentary also make it an effective genre of educational material. Political cartoons pack a lot of meaning into a small space, they are broadly accessible to readers without a lot of background knowledge, and they are attention-grabbing and maybe even humorous.
But how do we most frequently use them in the classroom? Do we look at them to gain a better sense of the political leanings of the newspapers in which they are published? Do we use them to catalog the ambient imagery that affects public opinion or normalizes stereotypes? Or do we mostly just break down the symbolism to reveal arguments and persuasive techniques? The latter approach is much more similar to the way historians analyze secondary sources. But maybe that’s where their greatest potential lies.
Our most common approaches to teaching history with primary sources are not only failing to give students the tools to understand the past. They’re also lulling educators into familiar patterns that make it more difficult to teach important concepts. Maybe, bombarded with thousands of resources and collections offered as “primary sources” for history classrooms, we’re missing the forest for the trees. The American Historical Association, as Ed Ayers has noted, describes historical practice not as analysis of primary sources, but as interpretation of evidence about the past, a task that requires contextualization, synthesis, and engagement with other historical arguments. Instead of thinking about materials for history classes in terms of primary and secondary sources, what would happen if we chose and analyzed resources based more on medium, genre, and the themes we might use to connect them with other resources?
In a recent blog post for AHA, Joseph D. Martin advocates for just such an approach: one that does not stop at analyzing how sources relate to the people and events they describe, but instead emphasizes how those sources can be used to interpret history. Reading his piece, I couldn’t help but think back to the class in which I truly began to cultivate history’s habits of mind. The professor who taught that class, Patrick Rael, did not assign a single primary source. Instead, he focused on an even more important, yet seemingly less frequently taught skill: reading and analyzing historical arguments. Because history is not the past, but rather our attempts to interpret that past, it necessarily exists at the intersection of the past and present.
The myriad ways history comes to us on the internet demonstrate the shortcomings of focusing on the distinction between primary and secondary. Several of the issues that have animated public discourse in the past few years, such as the calls to remove Confederate monuments from public spaces and the attempts to censor history education, are specifically about how we commemorate and learn about the past. Many compelling lessons come from people’s quests to uncover their family histories, while others result from digging into local histories. Video documentaries require audiences to grapple with original sources and interpretation simultaneously. Newspapers and magazines regularly publish historical analyses of current events, many of which quickly become valuable snapshots of history themselves. Historical writing from March 2020 on the history of supply shortages and hand washing, for example, now seems like a time capsule from the days when we did not know much about the virus that was beginning to sweep the nation. It’s tempting to think of this as a case of today’s secondary sources becoming tomorrow’s primary sources, but then how do we explain the reverse? Writings from decades ago sometimes have so much contemporary resonance that they can serve as de facto analysis of the present.
On the Bunk platform, we are less concerned with distinguishing primary from secondary sources than in organizing content to maximize its utility for its audiences. We categorize resources by form (e.g. digital humanities, visualizations, Q&As) and by sub-genre (e.g. biography, origin story, comparison, vignette). We link resources by theme, so users can identify resources with relevant ideas, no matter how those users will ultimately relate those resources to their own research questions and historical arguments. Perhaps most importantly, Bunk highlights the many connections between resources. There are countless ways of classifying sources, but whether it’s in the history classroom or on the opinion pages, it’s in those in-between spaces where meanings are made.
