In the Digital Humanities, it is common to weigh the research potential of collections as data by evaluating their representativeness. That is to say, we ask to what extent the data have the capacity to characterize a person, an event, a period, or an experience. Where the data exhibit significant informational paucity, indeterminate values, inordinate biasing, or limited scope it is common to cast them aside in pursuit of something held to be more representative. Alternatively, a move is made to systematically qualify data absence as a means of shoring up grounds for a redefined notion of representativeness to stand upon. Both responses generally fail to engage with data absence as a feature rather than a bug to be quashed. How might data driven scholarship be conducted in a manner that centers data absence?
I turned to Twitter with a question in this vein and did my best to document the generous response below.
Scott Weingart referred to absence as, “more a creative wellspring than a lacuna”, and shared a concise presentation on Fidelity at Scale. Scott raised the notion of a workshop or conference focused on productive explorations of archival absence at scale. I am all in for that. I’d guess that others would be to.
As Matthew Davis explains, a database is useful as a methodological tool because it does not permit ambiguity. This means that all decisions must be documented and justified. To create the schema for my database, I have already made important decisions about what data to extract from my primary documents – the Slave Narrative Collection, the first-person testimony culled from the Ku Klux Klan hearings, and the records of the Freedmen’s Bureau. Extracting some data does not require significant forethought, such as bibliographic information, dates, or geographic locations. Other data, however, require clearly defined keywords and a rigid workflow. When inputting data on incidents of racialized violence, for example, I must decide how to code types of violence. This issue will be the subject of my upcoming talk at the Graduate Colloquium on 26 April 2018. It is also necessary, however, to define resistance. Recently, there has been a proliferation of scholarship on resistance. But scholars have often failed to define resistance in any systematic way. This poses a challenge for creating a database that requires a concrete definition to ensure consistency. Where resistance is loosely defined, it is possible to see it almost everywhere and nowhere. This blog post, then, will outline how I define resistance.
According to Jocelyn Hollander and Rachel Einwohner, resistance has been variously defined as questioning and objecting, engaging in behaviour despite opposition, and opposing abusive behaviour and control. For my purposes, resistance can be understood as an action that results from a conscious decision to thwart attempts at subjugation. Resistance can be further defined as either formal or informal. Formal resistance refers to actions that are organized and conspicuous, whereas informal resistance refers to actions that are unorganized and clandestine. For most of the nineteenth century and early twentieth century, resistance to racialized violence was informal; African Americans relied on clandestine actions with limited risk of reprisal. My dissertation, therefore, is primarily concerned with informal resistance.
While the event focused mainly on how far things have come within the digital humanities world in the last ten years, the ODH’s ten year milestone is also an opportunity to reflect on the relationship between the digital humanities and the humanities-at-large (see this EDUCAUSE report for a great definition of the digital humanities). How are the digital humanities positioned within the academy after about a decade of increasingly institutionally recognized and grant funded activity?
“You good?” a man asked two narcotics detectives late in the summer of 2015.
The detectives had just finished an undercover drug deal in Brentwood, a predominately black neighborhood in Jacksonville, Florida, that is among the poorest in the country, when the man unexpectedly approached them. One of the detectives responded that he was looking for $50 worth of “hard”– slang for crack cocaine. The man disappeared into a nearby apartment and came back out to fulfill the detective’s request, swapping the drugs for money.
“You see me around, my name is Midnight,” the dealer said as he left.
Before Midnight departed, one of the detectives was able to take several photos of him, discreetly snapping pictures with his phone held to his ear as though he were taking a call.
Two weeks later, police wanted to make the arrest. The only information they had about the dealer were the smartphone pictures, the address where the exchange had taken place, and the nickname Midnight. Stumped, the Jacksonville sheriff’s office turned to a new tool to help them track down the dealer: facial recognition software.
It’s about time to infuse feminism into data science and visualization. At least, that’s what Emerson data visualization and civic tech professor Catherine D’Ignazio says based on her research into what an intersectional feminist perspective on data could look like.
“We’re in this moment when big data and visualization are being heralded as powerful new ways of producing knowledge about the world,” D’Ignazio said at a recent talk hosted by the Northeastern University Visualization Consortium. “So whenever anything has lots of power and is valued very widely by society, we just want to interrogate that a little more and say ‘Is it being valued equally?’ and ‘Is it benefitting all people equally?’”
She and her research partner found that the field has major problems with inequality, inclusion and quantification. Those who have the resources to collect, store, maintain, analyze and derive insight from large amounts of data are generally corporations, governments and universities. This creates an imbalance between who data is about and who has access to that data.
There is an imbalance between who data is about and who has access to that data.
D’Ignazio says this issue is compounded by the fact that women and people of color are underrepresented in data science and technical fields in general, a trend that is worsening. She also highlights skewed quantity and quality of data that is collected about various groups of people. For instance, there are very detailed datasets on gross domestic product and prostate function, but very poor datasets on hate crimes and the composition breast milk.
“Even when there is institutional and political will to collect data, data on sensitive topics — such as domestic violence, war crimes, sexual assault — is often highly flawed because there is powerful incentives for institutions and individuals not to report, not to collect, not to come forward,” she said.
So how do we take a feminist perspective on the design of visualizations? D’Ignazio cited six points that might bring us there.
At DHNow we try to use our Editor’s Choice pieces as an opportunity to highlight debates and important scholarship related to the Digital Humanities. Below is a round-up of commentary on two controversial twitter debates related to Safiya Umoja Noble’s (@safiyanoble) forthcoming book Algorithms of Oppression: How Search Engines Reinforce Racism. The controversy began when the manager of the IEEE History Center’s Twitter account (@IEEEhistory) sent several tweets denying the argument and evidence of Noble’s book without reading the work. In a second separate exchange, twitter users questioned and dismissed the existence of racist data structures — a component of Noble’s argument about digital redlining in Algorithms of Oppression. Noble’s book is set to be released on February 20 and contributes to a growing body of scholarship on the issue of bias in computing and digital ethics. Below are links to two responses prompted by the twitter exchanges as well as a link to Safiya Umoja Noble’s talk at the Personal Democracy Forum in 2016 which serves as a preview of her forthcoming book.
Everyone knows what you post online is never truly gone, but rarely are attempts to scrub something from the web quite this ironic—or infuriating.
Last week, the Institute of Electrical and Electronics Engineers History Center tweeted out an apology to author Safiya Umoja Noble after one of its historians shared a glaringly insulting criticism of her work from the organization’s Twitter account. But it appears the IEEE History Center—which was established to help preserve information on electrical technologies—has since deleted the apology.
After admitting and accepting that I was wrong, I discovered a few ways to move forward.
I began reading the work of Black feminist scholars who Safiya and other women of color were citing and discussing in their scholarship. Through this process, which continues today, I began learning that there are multiple ways of knowing about the world. By understanding the perspectives of women of color, I could begin to see why men continue to question women of color and why it needs to stop.
I also began to recognize that I, as a White cisgender heterosexual man from a privileged background in a position of power in higher education, can use the powerful platform that I have to work toward changing these stereotypes and addressing oppression in academia and beyond.
More recently, I have begun to assign Black feminist scholarship in my classes at Simmons. This is because, as Patricia Hill Collins (2000) explained, “It is more likely for Black women, as members of an oppressed group, to have critical insights into the condition of our oppression than it is for those who live outside this structures” (p. 39). Collins’s writing, as well as the writing of many other Black feminist scholars, can be an incredibly important starting point for library and information science students in learning to develop the tools, skills, and knowledge needed to challenge the oppressive systems and structures that continue to impact our profession and the communities we serve in harmful ways.
My hope in writing this post is that it will ultimately serve as a call to other men, like myself, to begin questioning not only what we know but how we know. More importantly, I hope this post will cause other men to stop questioning women of color and to start asking ourselves critical questions such as, “maybe what I know is wrong.”
“He was always on the lookout for what the next big thing would be, and made sure I knew about it.” In an email interview with DML Central, Parham explained that her grandfather was also an enthusiastic booster for tennis as a sport, although “pre-Williams sisters” women of color like Parham might have felt hesitant to follow his lead. “The tennis didn’t take, and we never got the Kaypro, which was, of course, too expensive and too useless for an 8 year old. But, the imagination of that computer did take hold, and soon after, we went to Sears to buy me a Commodore 64.”Professor Marisa Parham of Amherst College, who has led the Five College Digital Humanities initiative has a long history with digital media. “My earliest experiences with computers and devices mainly stemmed from my grandfather’s obsession with Kaypros in the 1980s. I was 8 or 9 years old. He would take me downtown to ogle what must have been some iteration of the Kaypro II, which for some reason, we found more intriguing than the Compaq II, though I remember thinking that the Compaq was hideous to behold.
As a future digital humanist, access to home computing proved critical to her literacy story. “The Commodore 64 was transformative for me because I could do so many different kinds of things with it. I still remember making my way through Turtle and then Basic. Over the years, though, I spent most of my time playing interactive text narratives, which is an interest I still have today!”
Since 2009 we have been contributing to the development of Europeana, the European platform that provides access to the digitised collections of cultural heritage institutions (CHIs) across Europe. One of our main contributions to Europeana is the Europeana Licensing Framework which ensures that data published on Europeana can be freely reused, and that all digital objects available via Europeana come with easy-to-understand information about their copyright status and under which conditions they can be reused…
We have constructed a methodology in which we individually assessed the accuracy of the rights statements+Kennisland has been working with Europeana since 2009 to make cultural heritage available for reuse. Read here more. of a representative sample of the digital objects made available via Europeana. The results from the sample give an indication of the accuracy of the rights statements of the entire database of Europeana. The results show that at least 61.8% of the rights statements were accurately applied and that at least 9.1% were inaccurate based on the available information. The accuracy of 17.4% of the rights statements is questionable, while for 8.8% it was not possible to determine the accuracy.
As a digital humanities librarian, E. Leigh Bonds (The Ohio State University) undertook an institutional environmental scan as the basis for assessment, identifying gaps, and developing recommendations. In this post, Bonds details her approach and framework, which prompted conversations and coordination across campus.
In August 2016, I became The Ohio State University’s first Digital Humanities Librarian. I’d already been “the first” at another institution, so I was acutely aware that distinction is both a gift and a curse: on one hand, I have the opportunity to define the role; on the other, the responsibility of defining that role. More importantly, I knew “the first” typically has the task of mapping previously uncharted (or partially charted) territory—the scope of digital humanities on campus—and exactly one week into my new position, I received that first charge: conduct an environmental scan of DH at OSU.
Having never conducted a formal environmental scan before (or even witnessed someone else doing one), I turned to the literature: no one charged with such an undertaking—regardless of campus size—does so without consulting those who have already charted their own environments. From recent publications (see Works Consulted), I gleaned that the scan should determine the nature of DH work underway, researchers’ interests, researchers’ needs, existing resources, and gaps in resources. All of the information gathered would then be complied into a report—in my case, an internal report for the Libraries’ administration and the head of the research services department—that included recommendations based on the findings.
What follows is my strategy for making these determinations and framing my report. Rather than provide a one-size-fits-all template (which would most likely work for no one), I explain the process I followed—and the thinking behind that process—to guide other “firsts.”
Electronic health records, quantified health, and diagnostic tools are all ‘digital technologies’ that co-create meaning and knowledge throughout the medical industrial complex. The initial connection between digital humanities (DH) and medicine is an easy association to make: DH works with data, with structures of data, with big data, with various forms of tech. Medicine and health are already ‘digital,’ and create and use data and data structures in relationship to various technologies and bodies. Easily, digital humanists can investigate these formations.
We can also untangle the underlying structures of the U.S. medical industrial complex in order to create new formations founded in justice and care. This is where #transformDH is foundational to the kinds of work that can be done in these intersecting fields. As an academic guerrilla movement invested in transformative scholarship that works for social justice, accessibility, and inclusion, #transformDH’s ideals are exactly what is needed to investigate and change not only how we practice and study medicine and health, but to change the structures of power within the larger medical industrial complex.
“What counts?” Fiona M Barnett asks, “…what is the effect when the conversation is not about recognizing similarity across differences or disparity in order to build a common ground, but rather, about declaring something to be unrecognizable within the confines of a field?” The U.S. medical industrial complex is founded on preventing difference, on creating normative categories of health, illness, and wellness as well as normative bodies and minds. These structures create invisibility, an inability to recognize “similarity across difference or disparity.”