Janneken Smucker

Digital platforms have created the infrastructure that enables exciting, transformative work centered on the lives of individuals who are too often missing from the archive and the classroom, merely because the imprints they left behind are not the ones we can easily access. This process-oriented chapter describes and evaluates the opportunities and challenges of working with students to interpret and disseminate oral histories with Southern newcomers to Philadelphia, focusing on interviews with women who struggled to find home as they made the transition from the Jim Crow South to the perceived promised land of the urban North. Unlike digital humanities projects that use computational tools to analyze a macro-level view of history by turning historical accounts into data, thereby elucidating patterns of large-scale movement or revealing complex networks of relationships, Goin’ North: Stories from the First Great Migration to Philadelphia does the opposite: revealing intimate moments within individual lives. Goin’ North is a collaborative initiative linking students and faculty from West Chester University (WCU) with the Louie B. Nunn Center for Oral History at the University of Kentucky Libraries.

The Goin’ North: Tales of the Great Migration oral history collection, housed at the Nunn Center, comprises more than seventy interviews conducted by my colleague Charles Hardy and a team of interviewers in the early 1980s with elderly African Americans in Philadelphia. At that time, accessing an oral history interview typically meant reading a transcript, an act in which the voices, silences, emotions, and nonverbal forms of communication are absent. Digitizing oral history interviews can enable the recovery of narrators’ voices. In the case of Goin’ North, this recovery had particular significance, as these voices were largely otherwise missing from the archival record that is typically dominated by literate, privileged, and largely white historical actors—those who left a paper trail. By the 2010s, digital audio technologies, digital media storage, and cloud-based interactive platforms created a vastly different landscape through which to encounter archival materials.

Before their transfer to the Nunn Center, the cassette tapes had sat untouched and untranscribed for more than thirty years, a fate common to many oral history initiatives undertaken in the decades before greater accessibility through digital audio. In essence, these were lost voices, resulting in the absence of crucial memories of a significant demographic, cultural, and social transformation within American life. Yet Hardy knew the accounts were powerful and important, and in 2014, before the centennial anniversary of the start of the Great Migration to Philadelphia, he became determined to rescue these narratives and do something with them that would make this archival record accessible, discoverable, and engaging.1 He wanted this project to involve students. And he wanted it to take advantage of digital technologies not conceived of when he conducted the interviews.

Some of the most compelling moments from these interviews with the female narrators have continued to rattle in my brain during the years since I first heard them. How did Minnie describe the impact of her grandparents’ experience in slavery? What was it that Idelle said about “passing”? Why was Fannie afraid to walk on sidewalk grates in Philadelphia? With a few clicks, I can now search for each of these moments in each interview. I am able to hear in clear fidelity Minnie’s voice:

You see, my father, his mother and father both were slaves. And my mother’s father and mother both were slaves. See, my mother’s mother came out . . . my mother came, her mother hadn’t been freed too long. She was still living kind of under the bondage of that slavery. You know there was a rule they say that whatever the white man would tell them, they believed him. And if he says, “Well you didn’t earn but $5, this year,” they believed him. So see they, some of them still was livin’ under their bondage of slavery. But they, I never heard ’em complain.2

I found this passage in the Oral History Metadata Synchronizer (OHMS) index for Minnie Whitney’s 1984 oral history interview. The index, created by WCU undergraduate student Stephanie Loeh in the course “The Great Migration and Digital Storytelling” that I team-taught with Hardy in fall of 2014, allows the user to search fields using a description, keywords from a controlled vocabulary, and image captions, as well as the transcript of the interview, to find a timestamped moment in the digital audio recording.3 I typed “slavery” into the search bar and instantly identified the extract just quoted in the synced transcript, along with another segment six minutes later describing the “slave rules” of the Jim Crow South.

I next remembered how Idelle Truitt Elsey described young African American women in Philadelphia who passed for white. Beginning in 1927, Elsey worked for the Armstrong Association, a Philadelphia organization with social service programs, including a job placement service, designed to help Southern newcomers acclimate to the city.4 Elsey used the term “passing” in her interview, because it was in common use at the time of her 1984 interview, although it was not used frequently during the time period she recalled.5 But she did not use the term “colorism” to describe the hierarchy among African Americans of various skin tones that existed in 1920s Philadelphia, which often pitted “Old Philadelphians (O.P.s)”—African Americans who had lived in the city for generations—against the newcomers from the South with their typically darker skin.6 Yet when WCU senior Ahleah Miles created Elsey’s OHMS index, she drew on her familiarity with the concept, knowing that a twenty-first-century researcher or curious listener might be interested in the power dynamics reflecting one’s level of skin pigment. She used this term not found in the interview transcript itself to describe a key theme of Elsey’s interview, thereby translating its meaning into language a researcher might use. I found this passage by searching for “colorism”:

I sent a very fair colored girl, and the lady called me up and said, uh, “We don’t—we can’t use her because she looks like white. We must have someone that looks like colored.” I said, “Well, you told me light-skinned colored girl, and that’s what we call a light-skinned colored girl.” I am a brown. I might be a light brown, but I’m not a light—but as far as white people went, uh, I’m light. There’s so many light-colored girls that went over and got mixed up in the crowd, see? So that the white people called someone like me light, but I know that we could not send—we had to send light girls on the job.7

The rush to digitize archival sources during recent years has privileged text over other types of media, in large part because of the lower cost of digitizing text and the ease through which text can be rendered machine readable compared to other media types, including audio and video.8 Consequently, those historical actors who left behind the written word have also been privileged. African American women who traveled from the rural South to Philadelphia as part of the first Great Migration fall into the category of those who did not leave behind extensive paper trails. Hardy’s pre-digital public history project included oral history interviews with women who worked as domestic servants in the homes of white families, in tobacco or garment factories, and occasionally in offices and banks.9

Digitizing the audio from their interviews was the first step in creating access to and enabling discovery of these firsthand accounts of internal migration within the United States. The next step was creating a classroom / archive partnership between WCU students and professors and the Nunn Center—the creator of OHMS and the archival home of the interviews—which enabled students to work closely with these interviews, making them searchable and discoverable. Using OHMS—an open-source tool designed to provide access to moments within an interview—while pairing interview segments with curated descriptions, hyperlinks, images, keywords, and GPS coordinates, students became intimately acquainted with the Jim Crow South, domestic servitude, the segregation of the urban North, and the sacrifices and decisions women made to empower their children and themselves. Integrating OHMS with the open-source content management system Omeka allowed students to create horizontal integration, as they identified disparate resources spanning multiple repositories, which they assimilated into a growing digital archive to use in the interpretation of the lives of these previously unknown historical actors.

Although digitization alone does not necessarily constitute digital humanities practice, it is an essential first step in many digital history projects by making primary sources machine readable, digitally accessible, or both. Our first step in the Goin’ North project indeed was the digitization of analog audio. Digital audio, however, is only slightly riper for interpretation and analysis than analog audio. Certainly, it is preserved, which is a vital objective of the digitization process. But how then to transform the ones and zeros of digital audio into sources that are more than just a different medium than the analog tapes from which they originated?

OHMS takes the best aspects of text—its variability, its nuance, its ability to say one thing in a variety of ways—by synchronizing it to the ones and zeros of digital audio. Earlier generations of oral historians considered the transcript of an interview as the primary source; however, when synchronized with the digital audio through the OHMS platform, the transcript becomes a descriptive form of metadata allowing the previously inaccessible audio to serve as the primary source used by researchers.10 Better yet, OHMS pairs additional metadata with the audio, essentially letting administrators create chapter marks within the audio file and populating each chapter with machine-readable metadata that can guide end users to the specific moments they seek.

A researcher studying cultural, rather than demographic or macro-level, aspects of the Great Migration, may want to access moments of oral history interviews that demonstrate how individuals remembered and understood their transition from South to North: the power structures, the processes of assimilation, the perceptions of difference between Southerners and Northerners. What sorts of keywords—natural-language terms—might this hypothetical researcher use to discover the hopes, fears, and superstitions of a young woman from the South who was newly arrived in Philadelphia?11 Fannie Hutchinson’s 1984 oral history interview recounts her experiences growing up in a sharecropping family in Virginia and arriving in Philadelphia in 1926 to live with an overprotective uncle. Sara Hasted, the WCU senior who indexed Hutchinson’s interview, wrote the following synopsis for a segment describing the warnings she received about living in the city:

Hardy asks Hutchinson about any stories that young men and women may have heard about what could happen to them in Philadelphia. She recalls stories of witch doctors taking girls away and student doctors experimenting on or killing them. She was also told not to walk on the grates in the sidewalks. She remembers knowing several girls who disappeared in Philadelphia; her uncle was very protective and didn’t want her to go to school or work.12

Hasted further crafted metadata for this segment in the form of keywords: Advice about moving north; Experiments; Stories; Urban legends; Witch doctors.13 Thus, end users could discover this segment of Hutchinson’s interview by searching for one of these terms, or they could skim the index, like a table of contents, and easily discern the segment’s key themes.

Mapping the moments of an interview into natural language is one of the powerful acts of translation that the students in the Goin’ North course embarked on.14 Doug Boyd, the head of the team that conceptualized and created OHMS at the Nunn Center, gives the example that an oral history narrator could talk for three hours about segregation without ever using the term.15 Using just a transcript of the interview, a researcher could search for clues of segregation—restrooms, swimming pools, drinking fountains, and so forth—perhaps yielding productive results. Or an index translating the verbatim text of the interview into terms a researcher might use could include the following keywords: Segregation; Segregation in Philadelphia; Segregation: Housing; Segregation: Motion Picture Houses; Segregation: Schools; Segregation: Theaters; Housing discrimination.

These terms are drawn from a thesaurus of more than one thousand terms developed by students over two semesters. Rather than using a text-mining computational tool, our students did it the old-fashioned way: they listened closely while consulting draft transcripts. Students each submitted a list of the proper names, places, and organizations, along with what they identified as key themes, from their assigned interviews. I culled the list, cleaning up and standardizing terms. Nunn Center archival staff took a final pass, formatting the list into a controlled vocabulary to load into the OHMS application. After this comma-separated values (CSV) data were integrated into the platform, OHMS then auto-suggested terms from the keyword list for students to use as they were indexing the interviews.

By imposing order on the interviews through this creation of text-based metadata, the students interpreted and made sense of them, while coming to know the narrators intimately. But is a group of twentieth-century, predominantly suburban and white undergraduate and graduate students best equipped to translate the spoken words of elderly African Americans from Philadelphia? Student Rachel Dixon noted of her contributions to a similar oral history indexing assignment, “What indexers deem important about these oral histories is a rhetorical act in and of itself.” She goes on to say that such translation “can be dangerous because it can feel as though the indexer is putting words into the interviewee’s mouth. . . . My desire to stay true to the person’s words and the identity claims they represented made the assignment both challenging and valuable.”16 The tensions Dixon describes are all the more prominent because of the large chasm between the lived experiences of the deceased African American narrators and the student indexers.

Yet all descriptive cataloging of cultural artifacts has inherent biases, and archivists and amateur laborers—like our students—who assist them can only strive to do their best to convey an object’s meaning, even from their limited perspectives. “Creating metadata is a process of deconstruction; that process will always be indelibly marked by the person who does it, and it will never be static across contexts and time,” observes Elinor Mazé, a veteran editor and archivist of oral history interviews.17 Ultimately, the project of imposing text-based metadata on the spoken words of these interviewees benefits both the student indexers and the indexes’ future users by creating a mechanism for accessing previously silenced voices, which for me outweighs potential biases derived from students’ rhetorical acts of translation. Significantly, our students now understand a sliver of the Great Migration from the firsthand perspective of an individual who lived it, something they would be hard-pressed to grasp in any other way.

In addition to using text-based metadata, OHMS enables indexers to curate the interview by pairing visual materials—digitized photographs, ephemera, and other primary sources—with individual audio segments. Here students exhibited creativity with these interviews conducted with African American women, who had not left behind significant paper trails. Student Kristen Waltz borrowed a camera from the university and located the house in the Chestnut Hill neighborhood of Philadelphia where Beulah Collins worked as a domestic servant. She paired the image with a segment of Collins’s interview in which she described living with this family.18 Sara Hasted scoured the Philadelphia City Archive’s PhillyHistory photography project to find a street scene along the bus route Fannie Hutchinson describes in her interview. Even more significantly, she found an archived photograph of the West Philadelphia luncheonette Hutchinson owned in the 1940s, a rare image of a Black woman–owned business; she paired the photo with the segment in which Hutchinson recounts with pride her ability to save up capital to purchase it.19

Often, students had to settle for stand-in images to illustrate the life stories of their interviewees. Erica Knorr found an image in the National Archives’ Department of Labor, Records of the Women’s Bureau, of an African American woman working in a tobacco factory, doing tasks similar to those described by Marie Mathis in her interview. Similarly, students drew on images of Black domestic servants taken as part of the Farm Security Administration’s photography project from the late 1930s and early 1940s. These images can only serve as illustrations, rather than as true primary sources, but they still help the end user imagine the world described in the interview. By developing a style guide as a class, we determined an appropriate and clear way to caption these images to be transparent that they were stand-ins for photographs of the narrators that we wished existed.

How have the processes of digitizing and indexing the words of African American women who traveled north with the Great Migration changed our understanding of this one aspect of the Black experience in the United States? First and foremost, these accounts are no longer missing from the archival record. Digital technologies allow us to hear the cadence and timbre of the actual voices; now we can listen to and analyze their riveting memories of oppression and opportunity, struggle and success. Rather than further obscuring these individual lives in order to draw macro-level conclusions, OHMS indexes and the accompanying primary sources reveal moments, intimate glimpses into distinct lived experiences. These moments humanize. And to our students working with these interviews, these historical figures—unknown to the textbooks too often used to teach key events in U.S. history—became companions as they learned about this pivotal shift in the cultural, social, political, and economic life of both Philadelphia and the nation at large. The course was a high-impact educational experience: the students left it not only with greater historical knowledge but also with the experience of having engaged in a collaborative assignment with real-world lasting impact, all while gaining transferrable technical skills. More beneficial than the specific skills and knowledge, however, was their empowerment: these students—many now college alumni—are pursuing aspirations and vocations in public history, having enrolled in library and public history graduate programs or embarking on internships and entry-level jobs, applying their newfound abilities to other projects. Several students have presented their work on Goin’ North at professional conferences, taking ownership of the final project and pride in knowing that their course contributions would extend far beyond the classroom.