Month: March 2024

Before opening the article “Racism in the Machine: Visualization Ethics in Digital Humanities” I took a peak at the maps depicting lynching in the United States discussed in the article. Even without the background the article provided me, many of the ethical issues of the Equal Justice Initiative’s (EJI) map were readily apparent to me. I found the map quite limited in the information and context it could provide as a visualization technique.

As mentioned in the article, the EJI’s map focused almost solely on the South.[1] It did this to the point that I found it difficult and somewhat frustrating that the map could only really provide me information about that region. Also, the fact that states that did not register lynchings of African Americans were not represented with their political borders on the map, while those that did register lynchings of African Americans were highlighted in contrast with the dark background made understanding a relatively complete picture of lynching in the United States impossible via this map, and thus the map more frustrating to use.

All this felt misleading or limiting for this data visualization. As a reader, I want to understand these lynchings in context, context that the map’s focuses on political boundaries in the South and presenting solely lynchings of African Americans severely limits.[2]

When I opened the map from Monroe & Florence Work Today, I was absolutely stunned by its ability to provide me with much of the context the EJI’s map lacked. The notification and explanation screens displayed before I could access the map not only got me to think about what this map visualized, but also what it did not visualize.

The most effective way this project gets its readers to think about how lynchings in the United States are visualized is through the choice it gives readers between depicting lynchings according to a “narrow definition” and lynchings according to a “broad definition.”

In all honesty, no other digital mapping project I have seen has made me think as long and hard about data visualization as this seemingly simple, two-option question. As Hepworth and Church pointed out in their article discussing ethics in the digital humanities by comparing the EJI and Monroe & Florence Work Today maps, the Monroe & Florence Work map places it emphasis on humanizing the victims of lynchings and not simply depicting them as data points. The map does this by plotting each lynching victim as a single point on the map which, when clicked, gives information and references discussing who the person was and why they were lynched.

The choice the Monroe and Florence Work map provide readers of what lynching data to depict I discussed above also help to avoid depersonalizing the victims of lynchings as mere datapoints. It instead makes readers think about the context in which these victims were killed. It makes readers think about the social dynamics of these killings and what it must have been like for those subjected to such atrocities.

Analyzing these maps has reminded me of how data and date visualizations are not neutral in any way, shape, or form. Data and algorithms can never be fully divorced from humans. As Church and and Hepworth state in their article, “[H]umans are at the center of algorithms, not only as their creators, but, in the case of data-driven algorithms, as the producers of the content they shape and present.”[3]

Data and data visualizations occupy a unique ethical landscape. While the ethics of a written piece may become quite obvious with a thorough read, it may take much more to uncover whether data you observe have been ethically collected, produced, and represented. Raw data, with its cold, seemingly authoritative, numbers and figures give off the explicit impression of impartial authority. However, methods of data collection, who collected the data, who funded the data collection, and the purpose of the collection of the data can all result implicit which can often remain invisible until thoroughly inspecting or investigating a dataset.

Data visualization also has a similarly authoritative nature. When you look at a picture, map, infographic, or other visualization its graphics can often captivate you and provide you with what appears to be an authoritative and unquestionable narrative as sources and methods and put into the background while the story of the data is pushed to the front.

All this makes me recognize the importance of providing transparency in any mapping projects and data visualizations I create. By being upfront with my data sources and the processes by which I conglomerated, prioritized, and ultimately visualized data in my projects will allow readers to be more critical of my work, seeing it not as monolithically authoritative, but instead as a piece of a larger puzzle of the historical reality. It is also important that I acknowledge what my maps and data visualizations omit and why I chose to (or subconsciously) omitted those datapoints or features. All these steps promote the ethical use, display, and dissemination of digital humanities projects like the one I will create for my final project in this class.

Bibliography

[1] Katherine Hepworth and Christopher Church, “Racism in the Machine: Visualization Ethics in Digital Humanities Projects,” DHQ: Digital Humanities Quarterly 12, no. 4 (2018), 3.

[2] Hepworth and Church, “Racism in the Machine,” 3.

[3] Hepworth and Church, “Racism in the Machine,” 1-2.

Map Citations in Order of Appearance:

Equal Justice Initiative and Google, Lynching In America: Racial Terror Lynchings, Lynching In America, https://lynchinginamerica.eji.org/explore.

Plain Talk History, Map of White Supremacy’s History of Lynchings/Map of White Supremacy’s Mob Violence, Plain Talk History, https://plaintalkhistory.com/monroeandflorencework/explore/.

Growing up in what is colloquially called the “Dirty South,” there is an expectation of stupidity and racism associated with Southerners, especially those from the Dirty South. There is the misconception that we tote around shirts with “THE SOUTH WILL RISE AGAIN” in bold letters amidst the backdrop of the Confederate flag. Our accents are seen as hillbilly (especially in Appalachia) and there is no regard to the rich culture and there is no mention of black Southerners unless it is in regards to slavery or acts of racial violence. These stereotypes are not only rooted in racism, but also classism (often ignored in the United States). Only recently, with Georgia (my home state) becoming a swing state, are there more talks around political disenfranchisement in terms of voter suppression and the realization that a lot of Southern voices have been stifled by the (often) loud, far-right populations of those states. I credit Stacey Abrams for finally opening the floor up to conversations about modern-day voter suppression of black voices.

In digital mapping, there is a reliance on human-made algorithms and the whims of the mapmaker themself. The article we read used the example of the Twitter-made, AI, Taytweets, which used people’s Tweets to learn how to communicate less like AI and more like an actual person. Of course, humans are not infallible, and the computer coders that create AI are not neutral, so, as the articles states:

And that:

“Racism in the Machine: Visualization Ethics in Digital Humanities Projects”

It makes sense that misconceptions and stereotypes about differing geopolitical states would be prevalent in mapping and geographical history projects. The maps we used for this blog post show the same subjects, but in contrasting ways.

(Lynching in America, EJI and Google).

This map, made with a partnership with Google and EJI (the Equal Justice Initiative), focuses almost entirely on the state of racial violence in the Deep South. We’ve spoken in class about how maps tell stories and this one recirculates the stereotype of racial violence and racism being an act mostly committed in the South. While a majority of black lynchings did take place in the South, it is also important to note that that is because a higher concentration of Black people lived in the South than the North.

“Visualizing data that exclusively focuses on the African-American experience in the Southern United States provides an important argument about the nature of Jim Crow racism. However, purporting such data to be an inclusive representation constitutes a harm in the sense that it perpetuates common narratives of racial violence as a southern exception to an otherwise inclusive nation.“

“Racism in the Machine”

This shapes the systematic oppression and structure of white supremacy as being something linear- of power birthed from the top and trickled down to the bottom percent of people within white hegemony. White supremacy is not, pun intended, black and white. Racial violence, while extremely horrific is not the full story of white supremacists and their maltreatment and persecution of other racial and ethnic groups. Racial violence is usually the escalation of decades of whites being able to get-by with “smaller” crimes and abuses:

“The conflation of lynching with the full extent of racialized violence in United States history obscures the historic depth and breadth of the oppression of people of color. A black individual was far more likely to suffer public humiliation, assault, rape, and murder than a public lynching.“

“Racism in the Machine.”

I understand the reason why the mapmakers chose lynching, however, it is a physical and tangible manifestation of racial violence. The stories of black people being raped or abused are far too commonplace, so therefore there are few records documenting the crimes. However, white supremacy thrives not just on murderers and extremist groups like the Aryan Brotherhood, but on the complacency and the actions of everyday white people, who use “not me” statements and ignore the white supremacists tendencies that are conditioned into the white American populous through years of American historical propaganda and nationalism, regardless of geographic region or state. I lived in Southwest Missouri for a bit and I was surprised by how prevalent the Klan was there. I knew someone whose grandfather was a founding member of the Klan in Southwest Missouri. One day when I was cleaning out a closet in their home, I found their grandfather’s bloodstained Klans robes. It was terrifying- to see the ghostly manifestation of the United States racist terrorism in person.

The other map, made by Monroe Work, showcases more nuances in white supremacy, showing it is not burdened by state borders and that the white nationalists do not simply punch-down on African-Americans, but other racial groups and even some white ethnic groups (referencing Italians).

(White Supremacy Mob Violence).

This map depicts a number of lynchings within Westward Expansion, California, and the Midwest.

I wanted to use my county from home as an example for why Work’s map does a better job than the EJI map.

Here is Lee County, GA on Work’s Map:

Here is Lee Ct. on the EJI map:

There is no mention of the names of those who were murdered, nor is there a reference to sources in the EJI map. Work’s map not only provides the reasoning behind the hate crimes, but also their sources to add supplementary materials to their storytelling.

I think within digital humanities, one has to be careful how their story is portrayed to an audience. As an English major as well as a history major, I have learned a lot in creative writing classes that whilst a story I write means one thing to me, once it is published it takes on different meanings depending on the audience. If I wasn’t a Southerner or if I didn’t read the article, I cannot help but wonder if I would understand the focus on the American South as a hotspot for racism in the EJI map. I think this is something to be mindful of when mapping minorities or the trauma of others. Data should come from multiple sources and nuance should be added if possible and applicable to the subject matter. I understand that the EJI map is political- it is meant to shock and it is meant to be minimal. They are demanding collective action as fast as possible, but (and not to go all Foucauldian again) ignore the non-linear construction of power under white supremacy. This is dangerous, white supremacy has seeped into the minds of not just white people, but minoritized groups as well, who foster hatred for other racial groups and/or even their own. This allows racism to fly under the radar. Lynching is just a stark example of what happens when this racism goes unchecked and festers before it becomes infected and is only cured through egregious acts of violence.

“Monroe Work Today’s Map of White Supremacy Mob Violence is an exemplar of ethical visualization, not because it is free of two centuries of baggage and biases, but because it acknowledges them, while also acknowledging the potential pitfalls of the very endeavor of transforming human beings into visualized historical data.”

“Racism in the Machine.”

Stories, especially those that are not are own, are hard to tell. It is important to acknowledge biases and provide the resources for audiences to reach out and interact with stories from members of the community that the mapmaker is trying to tell. People are not unerring, but there is something to be learned from accountability and the avoidance of knee-jerk reactions of “not me!” or “I don’t see color!” Only then can conversations begin to take place and the fostering of learning can begin. Readers should also, then, question the sources of information and the potential argument presented from visualizing history. It would be impossible for a historian to show every facet of history in their map, but one should question if those silences are intentional.

The Jim Crow era is a dark chapter of American History. African Americans across the country experienced discriminatory laws. Among some of the challenged African Americans faced were lynchings. When mapping racial history, like lynchings, it is important to be mindful of the ethical ideas surrounding it.

According to Katherine Hepworth, ethical implications can vary. She writes, that these maps are “demonstrating how choices about representation, interaction, and annotation in their data visualizations either do harm in the sense described above, or challenge dominant narratives” ¹. In other words, when it comes to mapping race, representation and how we define our terms matter. These maps highlight the importance.

In both of these maps, we see visualizations of racial violence across the United States. In the first map, created by Equal Justice Initiative, we see it centered on the Southern United States. Granted, a majority of racial violence occurred in the South, however, the first map gives you the idea violence outside of the South was nonexistent. In the Map of White Supremacy mob violence, we see much more detail and inclusivity. This map highlights White Supremacy mob violence which includes every racial minority group. This includes, Chinese, Black, Italian, Latinx, and Native American. This map also shares details about each individual case.

Besides representation, how we define our term matters. According to Plain Talk History, there are two definitions of “Lynching”. A strict definition of lynching is when the lynchers “believed they were righteous, never worried they might face charges. To them, a dead victim meant justice: the matter was settled, and townspeople would move on” ². The broader definition of lynching is the belief when “perpetrators did not claim to be agents of justice. They knew it was murder—in their rage, they didn’t care. This was a homicidal spree to avenge the established order” ³. As you can see, these two defintions of lynching are different. Because of this, how you define “Lynching” means including or excluding a wide variety of victims.

In conclusion, when mapping racial violence, it is important to understand representation. This means that we need to think about who we want to include and how we want to include them. To do this, we must be able to define what we are researching. By doing this, we create the ability to expand or reduce our representation.

1. Hepworth, Katherine & Christopher Church. 2018. “Racism in the Machine: Visualization Ethics in Digital Humanities Projects.” Digital Humanities Quarterly, 7.  ↩︎
2. Monroe & Florence Work Today. 2016. “Map of White Supremacy Mob Violence.” PlainTalkHistory ↩︎
3. Monroe & Florence Work Today. 2016. “Map of White Supremacy Mob Violence.” ↩︎