Mapping the Spread of American Slavery


Lincoln Mullen is an assistant professor in the Department of History and Art History at George Mason University, working on the history of American religions as a digital historian. He writes regularly on his own blog and for the Religion in American History group blog.  He also teaches a course on “Data and Visualization in Digital History” where he is working on nineteenth-century U.S. religious statistics.

It seems that Professor Mullen and I have a few things in common.  Both of us:

  • work with and enjoy using statistics to back up or generate our work;
  • acknowledge the value of data visualizations in pictorial or graphical formats to help others better grasp concepts or identify patterns in data;
  • enjoy sharing our discoveries with others through our blog sites;
  • love history and want to help others appreciate it, too.

The following blog was written by:  Lincoln Mullen and published on May 12, 2014.  It gives a fuller perspective on the distribution, growth, and demise of slavery that expands on the history of slavery in an earlier blog post of mine from May 21, 2013 Tobacco, Slavery, Earthworms, Honey Bees; Grains, Livestock, Disease, Oh My!

These Maps Reveal How Slavery Expanded Across the United States

As the hunger for more farmland stretched west, so too did the demand for enslaved labor

In September of 1861, the U.S. Coast Survey published a large map, just under three feet square, titled a “Map showing the distribution of the slave population of the southern states of the United States.” Based on the population statistics gathered in the 1860 census, and certified by the superintendent of the Census Office, the map depicted the percentage of the population enslaved in each county.

hergesheimer-map

Figure 1: U.S. Coast Survey, Map showing the distribution of the slave population of the southern states of the United States (Washington, DC: Henry S. Graham, 1861). Image from the Library of Congress. [PNG]

The map showed at a glance the large-scale patterns of slavery in the American South: the concentrations of slavery in eastern Virginia, in South Carolina, and most of all along the Mississippi. It also repaid closer examination, since each county was labeled with the exact percentage enslaved. The map of slavery was one of many thematic maps produced in the nineteenth century United States. As Susan Schulten has shown, this particular map was used by the federal government during the Civil War, and it was a favorite of Abraham Lincoln’s.1

hergesheimer-inset

Figure 2: A detail from the U.S. Coast Survey map of slavery, showing the Mississippi River and delta. [PNG]

Though such thematic maps, in particular of slavery, have their origins in the nineteenth century, the technique is useful for historians. As I see it, one of the main problems for the historians’ method today is the problem of scale. How can we understand the past at different chronological and geographical scales? How can we move intelligibly between looking at individuals and looking at the Atlantic World, between studying a moment and studying several centuries?2 Maps can help, especially interactive web maps that make it possible to zoom in and out, to represent more than one subject of interest, and to set representations of the past in motion in order to show change over time.

I have created an interactive map of the spread of slavery in the United States from 1790 to 1860.3 Using Census data available from the NHGIS, the visualization shows the population of slaves, of free African Americans, of all free people, and of the entire United States. It also shows those subjects as population densities and percentages of the population.4 For any given variable, the scales are held constant from year to year so that the user can see change over time. You can use the map for yourself, and I’ve also written briefly about what the map shows below.5 Historians have of course often made use of maps of slavery, in particular maps based on the Census, in support of their arguments. What I’ve tried to do in this interactive map is make it possible for users (including me) to explore the census data in support of making historical arguments.6

The Spread of U.S. Slavery, 1790–1860

How to use this map

I have written an introduction to this visualization. Zoom to any county by clicking on it. Clicking on the same county will zoom out. The scales preserve intensity for change over time: in other words, a color represents the same thing for each year on the map. However, the color scales do not necessarily preserve intensity from data field to data field: the darkest color for the total population does not represent the same values as for the enslaved population. The scales for population are logarithmic (with intermediate values) so every second step in the color ramp represents a ten-fold (not a two-fold) increase.

animation-slave-density

Enter a caption

Figure 3: An animation of the density of slave population from 1790 to 1860. Notice that slavery spreads more than it grows. Taken from the interactive map. [GIF]

animation-population-density

Fig: An animation of the density of the total population from 1790 to 1860. Notice that population in the north both grows in place and spreads westward. Taken from the interactive map. [GIF]

Another observation to make about slavery in the United States is what an extraordinarily high percentage of the population was enslaved. The majority slave populations of the Chesapeake, the South Carolina and Georgia coast were soon duplicated in the majority slave populations of the Mississippi River valley.

animation-slave-percentage

Figure 5: An animation of the percentage of the population enslaved from 1790 to 1860. Taken from the interactive map. [GIF]

A striking way to see the importance of slavery is to look at a map of the total free population: a photo negative, if you will, of slavery. When looking at the density of the population that was free, large swathes of the South appear virtually depopulated. (Perhaps this is what a history book looks like that fails to take includes slaves?)

total-free-1860

Figure 6: The population density of all free persons in 1860. Taken from the interactive map. [PNG]

Finally, the dynamics of the free African American population looked more like the free white population than the slave population. The Free African American population seems to have primarily settled along the Eastern seaboard and especially in the cities of the northern United States. Free African Americans were almost entirely excluded from most of the deep South, except the cities.animation-free

Figure 7: An animation of the free African American population from 1790 to 1860. Taken from the interactive map. [GIF]

Historians have long used maps of slavery to advance their arguments.9 I hope this map finds some use in making more arguments about the history of slavery, and especially for helping students to grasp the big picture of the “peculiar institution” which made the nation “half slave and half free.”10

Interactive Maps

The first thing to observe is that slavery spread more than it grew. The population of slaves in 1790 or 1800 was already very high compared the maximum population levels.7 In fact, in Charleston County, South Carolina (one of the counties with the highest populations of slaves) the number of enslaved people in 1860 was only 63% of what it had been in 1840. This is not to say that the total number of slaves in the eastern seaboard states did not go up over time. But the number of enslaved people in a particular place did not grow at anything like the rate of free people in the north. The free population in the north both grew in the same place and spread to the west. The slave population had a different dynamic. It grew in intensity in places around the Chesapeake bay, even as slavery was gradually abolished in the North. But primarily the slave population spread to the fertile crescent of lands in Georgia, Alabama, and Mississippi, and most of all to the Mississippi River valley. Below you can see two animations of the density of the slave population and the density of the total population (keep in mind that the scales are different). What you see in these maps is the spread of slavery through the domestic slave trade.8 You also see the origins of the sectional crisis in the continual expansion of slavery.

And, these data visualization maps from Cartographer, Bill Rankin.  Bill Rankin tries to find a balance between accuracy and readability in a set of maps that show slavery in a grid layout from 1790 to 1870.

Rankin made a map for each decade, but the most interesting one that shows all the data at once. Size of each circle represents the peak number of slaves per 250 square miles. Color represents the year this peak occurred.

YEARS:  mouse over, or click to download:

1790  (PNG, 551 KB)
1800  (PNG, 686 KB)
1810  (PNG, 821 KB)
1820  (PNG, 980 KB)
1830  (PNG, 1.10 MB)
1840  (PNG, 1.25 MB)
1850  (PNG, 1.48 MB)
1860  (PNG, 1.65 MB)
1870  (PNG, 1.56 MB)

or download:

all nine decades (ZIP, 9.88 MB)
big poster (PNG, 3.87 MB)
peak slavery (PNG, 1.05 MB)

The gradual decline of slavery in the north was matched by its explosive expansion in the south, especially with the transition from the longstanding slave areas along the Atlantic coast to the new cotton plantations of the Lower South. Although the Civil War by no means ended the struggle for racial equality, it marked a dramatic turning point; antebellum slavery was a robust institution that showed no signs of decline.

Mapping slavery presents a number of difficult problems. The vast majority of maps — both old (from Census 1860)  and new  (from Census 1790) — use the county as the unit of analysis. But visually, it is tough to compare small and large counties; the constant reorganization of boundaries in the west means that comparisons across decades are tricky, too. And like all maps that shade large areas using a single color, typical maps of slavery make it impossible to see population density and demographic breakdown at the same time. (Should a county with 10,000 people and 1,000 slaves appear the same as one that has 100 people and 10 slaves?)

My maps confront these problems in two ways. First, I smash the visual tyranny of county boundaries by using a uniform grid of dots. The size of each dot shows the total population in each 250-sq mi cell, and the color shows the percent that was slaves. But just as important, I’ve also combined the usual county data with historical data for more than 150 cities and towns. Cities usually had fewer slaves, proportionally, than their surrounding counties, but this is invisible on standard maps. Adding this data shows the overwhelming predominance of slaves along the South Carolina coast, in contrast to Charleston; it also shows how distinctive New Orleans was from other southern cities. These techniques don’t solve all problems (especially in sparsely populated areas), but they substantially refocus the visual argument of the maps — away from arbitrary jurisdictions and toward human beings.

(For a graphic explanation of this technique, see here.)

The bottom map shows the peak number of slaves in each area, along with the year when slavery peaked. Except in Delaware, Maryland, and eastern Virginia, slavery in the south was only headed in one direction: up. Cartographically, this map offers a temporal analysis without relying on a series of snapshots (either a slide show or an animation), and it makes it clear that a static map is perfectly capable of representing a dynamic historical process.

Resources:

Lincoln Mullen, “The Spread of U.S. Slavery, 1790–1860,” interactive map, http://lincolnmullen.com/projects/slavery/, doi: 10.5281/zenodo.9825.

Minnesota Population Center, *National Historical Geographic Information System: Version 2.0* (Minneapolis, MN: University of Minnesota, 2011), http://www.nhgis.org.


  1. See Susan Schulten, Mapping the Nation: History and Cartography in Nineteenth-Century America(Chicago: University of Chicago Press, 2012), especially chapter 4 on slavery and statistical cartography. Also see the book’s companion website, which includes many images of maps of slavery.
  2. For one discussion of the problem of scale, see David Armitage and Jo Guldi. “Le Retour de la longue durée: Une perspective anglo-saxonne,” Annales, in the press. Whatever the reason for the blockbuster success of Thomas Piketty’s Capital in the Twenty-First Century, it’s worth noting that the book is primarily a longue durée history of the structure of capital.
  3. I am grateful for suggestions from Yoni Appelbaum, John Hannigan, and Caleb McDaniel, who each looked at the map in development, though they will each find more things they wished were different.
  4. You might think of the visualization as 88 maps = 8 decades ✕ 11 variables.
  5. The map represents a lot of data, and I have not been able to make it snappy enough for my satisfaction, particularly for mobile devices. Hence the animated GIFs below.
  6. Of course, there is far more to the history of slavery than just the Census data, which alone cannot answer any of the interpretative questions that historians have asked.
  7. This is remarkable given that in the Revolution many slaves escaped to or with the British army.
  8. Steven Deyle writes, “I believe it is safe to conclude that between 1820 and 1860 at least 875,000 American slaves were forcibly removed from the Upper South to the Lower South, and that between 60 and 70 percent of these individuals were transported via the interregional slave trade.” Steven Deyle, Carry Me Back: The Domestic Slave Trade in American Life (New York: Oxford University Press, 2005), 289.
  9. Perhaps I will provide a few examples in a future post.
  10. From Abraham Lincoln’s “House Divided” speech: “Either the opponents of slavery, will arrest the further spread of it, and place it where the public mind shall rest in the belief that it is in the course of ultimate extinction; or its advocates will push it forward, till it shall become alike lawful in all the States, old as well as new—North as well as South.”

Explanation of Census Data:

The U.S. Census data and shapefiles for these maps comes from Minnesota Population Center, National Historical Geographic Information System, version 2.0 (Minneapolis, MN: University of Minnesota, 2011). For a description of the questions asked on the 1790 to 1860 censuses, see Measuring America: The Decennial Censuses From 1790 to 2000 (U.S Census Bureau, 2002). Bear in mind the reason the Census kept statistics on slavery. Slaves were counted in the Census because of the three-fifths compromise in the federal constitution, by which an enslaved person counted as three-fifths of a person when apportioning representation in Congress and direct taxes. I have tried to represent unavailable data on the map, but sometimes in the Census a value of zero actually means that the data has been lost or was never gathered. Treat the Census numbers skeptically: even in the best of circumstances, the Census undercounts the population. For example, Harvey Amani Whitfield has shown that Vermont did have slavery, even though no slaves were enumerated in the Census. The numbers are used chiefly for showing degrees of magnitude. Below are the fields in the NHGIS data that I have used. The total free population was always calculated by subtracting the slave population from the total population.

1790 Census

  • Slave population: “Race/Slave Status: Persons: Non-White: Slave” (AAQ002)
  • Total population: (A00AA1790)
  • Free African American population: “Race/Slave Status: Persons: Non-White: Free” (AAQ001)

1800 Census

  • Slave population: “Nonwhite Population, Except Indians Not Taxed by Slave Status: Slave” (AAY002)
  • Total population: (A00AA1800)
  • Free African American population: “Nonwhite Population, Except Indians, Not Taxed by Slave Status: Free” (AAY001)

1810 Census

  • Slave population: “Nonwhite Population, Except Indians Not Taxed by Slave Status: Slave” (AA7002)
  • Total population: (A00AA1810)
  • Free African American population: “Nonwhite Population, Except Indians, Not Taxed by Slave Status: Free” (AA7001)

1820 Census

  • Slave population: sum of “Nonwhite: slave” male and female columns for “Race/Slave Status by Sex” (ABB003 and ABB004)
  • Total population: (A00AA1820)
  • Free African American population: sum of “Nonwhite: free” male and female columns for “Race/Slave Status by Sex” (ABB005 and ABB006)

1830 Census

  • Slave population: sum of “Nonwhite: slave” male and female columns for “Race/Slave Status by Sex” (ABO003 and ABO004)
  • Total population: (A00AA1830)
  • Free African American population: sum of “Nonwhite: free” male and female columns for “Race/Slave Status by Sex” (ABO005 and ABO006)

1840 Census

  • Slave population: “Nonwhite: slave” column for “Race/Slave Status” (ACS003)
  • Total population: (A00AA1840)
  • Free African American population: “Nonwhite: free” column for “Race/Slave Status” (ACS002)

1850 Census

  • Slave population: “Nonwhite: slave” column for “Race/Slave Status” (AE6003)
  • Total population: (A00AA1840)
  • Free African American population: “Nonwhite: free” column for “Race/Slave Status” (AE6002)

1860 Census

  • Slave population: sum of “Slave” male and female columns for “Race by Sex” (AH2005 and AH2006)
  • Total population: (A00AA1860)
  • Free African American population: sum of “Free Colored” male and female columns for “Race by Sex” (AH2003 and AH2004)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s