About This Case Study

In 1900, W. E. B. Du Bois and a team of students and alumni from Atlanta University prepared a set of hand-drawn charts, graphs, and maps for "The Exhibit of American Negroes" at the Exposition Universelle in Paris — the world's fair of that year. The exhibit, organized by the newspaper editor Thomas Junius Calloway, set out to document the lives of Black Americans a generation after emancipation; Du Bois's contribution was a sociological study told largely in pictures. Working from census figures and his own Atlanta research, his team produced sixty-three data visualizations, divided into two series: "The Georgia Negro: A Social Study," which examined a single state, and a national and international series titled "A Series of Statistical Charts Illustrating the Condition of the Descendants of Former African Slaves Now in Residence in the United States of America." The charts used bright color, bold geometry, and an inventive visual style that still looks modern more than a century later.

These visualizations were not neutral displays of information. They were an argument. At a moment when "race science" and social Darwinism claimed to demonstrate Black inferiority through statistics and measurement, Du Bois turned the same tools — data, charts, comparison — to the opposite end, presenting evidence of education, landownership, business, and population growth in order to counter the prevailing narrative. This case study asks you to read the data portraits the way you would read any argument: to ask what each chart claims, to whom it speaks, and against what assumption it pushes. The analytical move is to treat a visualization not as a transparent window onto facts but as a designed act of persuasion — and to ask how Du Bois used the language of charts to contest "the color line," his term for the system of racial division he named the central problem of the twentieth century.

Before You Begin

Have ready:

• The Library of Congress's digitized Paris Exposition collection — the charts, photographs, and related materials. Begin with the collection overview; the hand-drawn statistical charts are gathered under LOT 11931.
• A note-taking surface — paper, a document, or a shared doc if you are working in a group.

A note on language: the chart titles are reproduced from the 1900 originals and use period racial terminology that is now considered offensive. The terms are part of the historical record you are examining; this guide preserves them in quotation marks where they appear on the original plates.

Before you open the collection, write down, in a sentence or two each:

• When you hear that statistics or charts are "objective," what do you take that to mean?
• What would it look like for a chart to make an argument rather than simply report a number?

Keep these answers nearby; the exercise will return to them.

The Exercise

Phase 1: Orientation (5 minutes)

Open the Library of Congress collection in a new tab. Scroll through it once to get a sense of its overall shape — the range of charts, the maps, and the photographs that accompanied them. Then go through a second time and note three visualizations that surprise you, puzzle you, or strike you as especially striking in their design. For each, jot down its title (as given on the plate), what it appears to measure, and one design feature that caught your eye — a color, a shape, a comparison.

For groups: each member does this individually. Do not share yet.

Phase 2: Close Reading of One Chart (10–12 minutes)

Choose one chart to examine in detail. If you are not sure which to pick, choose one from each of these kinds so that your selection is concrete:

• A chart that compares two groups or two moments in time — for example, one tracking population, literacy, or property across decades.
• A map that distributes data across space — several plates map Black population or landownership across Georgia or the United States.
• One of the radial or spiral forms — the bold, abstract designs for which the portraits are now best known.

For your chosen chart, write or discuss:

1. What it shows. Be specific. What is being measured, in what units, across what range? What is the single fact the chart most wants you to take away?
2. How it shows it. What design choices is Du Bois making — color, scale, orientation, the decision to bend a line into a spiral or to leave an axis unlabeled? What do those choices do to your reading of the data?
3. What it leaves out or assumes. What does the chart not tell you? What knowledge does it assume you already have? Where, if anywhere, does the design make one reading easier and another harder?

For groups: each member presents their chart in two minutes, then the group chooses one to develop further.

Phase 3: Reading the Visualization as Argument (8–10 minutes)

Return to the two answers you wrote in Before You Begin, and now consider the historical setting. In 1900, influential strands of science held that statistics could demonstrate the biological inferiority of Black people; measurements and displays at world's fairs often served that claim. Du Bois designed his portraits in direct response — using data to show development, capability, and progress achieved under conditions of oppression.

Holding your chosen chart against that context, ask:

• What claim does this chart make about Black American life? State it as a sentence Du Bois might have endorsed.
• To whom is the claim addressed? The exhibit's audience was largely white and European. How might that audience have shaped what Du Bois chose to show, and how he showed it?
• Against what belief does the chart push? Name the assumption it is built to unsettle, and identify the specific feature of the design that does that work.
• Pair your chart with one of the photographs in the collection. What does the photograph do that the chart cannot, and the reverse? How do image and data together build a fuller argument than either alone?

For groups: spend the first five minutes developing the chosen chart together; spend the last five pairing it with a photograph.

Phase 4: The Portraits Today (optional, 5–7 minutes)

Du Bois's data portraits have found a second life among contemporary data-visualization practitioners. Since 2021, an annual online challenge — the #DuBoisChallenge, organized by Allen Hillery, Anthony Starks, and Sekou Tyler — has invited people to recreate the 1900 charts using modern tools such as Tableau and R and to share their versions on social media; the original plates and their underlying data are gathered in a public repository. Browse a handful of the recreations (search the #DuBoisChallenge hashtag, or open the repository). Then consider: what is gained, and what is lost, when a hand-drawn 1900 chart is rebuilt with twenty-first-century software? Does the recreation clarify Du Bois's argument, aestheticize it, or change it?

Closing Reflection

In two or three sentences, complete this thought:

The chart I examined argues that __________. It makes that argument not only through its data but through __________. What I now understand about "objective" statistics that I did not before is __________.

Write the most precise version of this you can. The value of the sentence is in the specificity of what you name.

A Note on Modes

Solo mode. Work through the phases in order, writing brief notes at each step. The close reading in Phase 2 and the argument analysis in Phase 3 are the central artifacts; keep them for your own record.

Group mode (3–6 people). Designate a timekeeper. Members work individually in Phase 1, then choose one chart to develop collectively in Phases 2 and 3. If time is short, the closing reflection can be assigned as individual writing after the session.

After the Case Study

• For the fullest collection of the portraits with scholarly commentary, see Whitney Battle-Baptiste and Britt Rusert, eds., W. E. B. Du Bois's Data Portraits: Visualizing Black America (Princeton Architectural Press, 2018), which reproduces the plates in color and reconstructs how they were made.
• On the photographs Du Bois assembled alongside the charts, see Shawn Michelle Smith, Photography on the Color Line: W. E. B. Du Bois, Race, and Visual Culture (Duke University Press, 2004).
• To work with the data directly, the public data-portraits repository (github.com/ajstarks/dubois-data-portraits) provides the original plates and their underlying datasets, so you can attempt your own recreation of a chart and discover what choices Du Bois faced.