Cities of Slavery
A Brief Walk Through the Antebellum Urban World
I was asked an interesting question on Twitter:
There’s a lot there, but the basic question is: what was urban life like before the Sunbelt migrations and, especially, during the period when slavery still dominated the south? It’s a fun question, and, as a loyal son of the south, of great personal interest to me. So I set out to do it.
But it turns out to be way harder than I expected. I’m genuinely shocked by how little information I was able to find on the historic populations of major Caribbean cities like San Juan, Havana, Port-au-Prince, Caracas, and others. I even did French- and Spanish-language searches, and came up dry in many cases. As such, I won’t be able to provide nearly the level of detail I wanted.
But to start out, let’s look at just raw population. How big were southern, American cities compared to the rest of the nation, and other slave-holding areas?
The largest slave-holding city in 1830 in the US or the Caribbean was almost certainly Havana, Cuba. Sadly, I found virtually no data whatsoever on the populations of some cities I would really have liked to include, like Maracaibo and Santo Domingo. It is possible that Santo Domingo was larger than Havana But I have no data to show that.
Havana had about 140,000 people in 1830. Were it in the United States, it would have been our #2 city after NYC. Now, I should say, modern day Philadelphia, would actually be about 160,000 people, but was at the time classified as several different cities. The world’s largest city was probably either Beijing or London at around 1.3–1.4 million people. Comparably-sized cities to Havana included Glasgow, Milan, Mexico City, or Venice at about 170,000 people, or Copenhagen at 140,000, or Berlin at around 200,000. In other words, Havana was a substantial city, only rivaled in north America by New York, Philadelphia, and Mexico City. For comparison, Montreal had probably under 70,000 people and Quebec City probably under 40,000. I have less clear information about the South American cities like Buenos Aires, Rio de Janeiro, Santiago, or Montevideo, but they could range from 30,000 to 170,000 in the case of Rio.
No surprise, then, that Cuba was seen as a key trading partner, and coveted by Southern expansionists. Havana could have been to the South what New York City was to the north, if only Cuba could be annexed (see: Ostend Manifesto).
Within the United States, the leading slaveholding city was Baltimore. This probably surprises many readers, who expected to hear Charleston, or New Orleans, or Atlanta. But no: Baltimore. And it’s not close, by the way. Baltimore had 81,000 residents, while the next biggest American slave-city was New Orleans, with 46,000 residents.
Change in Slave City Populations
The Rise of… Some.
Over time, many of these relationships would change. Below, I have the best I could come up with for a time series of major slave-holding cities, as well as the two Haitian cities. It involves tons of disparate sources and some guess-work on my part, so please take it as a best guess, not as too authoritative.
As you can see, by 1860, Baltimore had surpassed Havana, and New Orleans was getting close. But New Orleans was itself set to be surpassed by a newcomer. From just 1,600 people in 1810 and 5,000 in 1830, St. Louis had 161,000 residents by 1860.
You’ll also notice that many of the foreign cities have lost ground. Louisville, Kentucky passed up Port-au-Prince in Haiti. Charleston has made serious headway to catch up with Caracas, as has Richmond. Kingston is essentially stagnant, while Memphis, Mobile, Savannah, and Nashville rise to its level. Cape Haitian actually declines a bit before recovering to 1860, but is nonetheless surpassed by Wilmington, North Carolina. Lexington, Kentucky rises to or beyond La Guaira, Venezuela. Throughout the enslaved world, American populations outpaced others.
There are many reasons for this. Many countries began regulating the slave trade beginning in the early 1800s, and even interdicting some foreign slave-trading ships. The result was that some areas found it harder to maintain enslaved populations, as, everywhere outside of the United States, enslaved people tended to experience higher mortality than fertility. Forced immigration was necessary to maintain populations in many cases. As those flows were cut off, populations plateaued or fell.
Many cities were struck by fires or earthquakes, Spanish colonial holdings suffered from neglect and poor administration, and, crucially, virtually the entire Caribbean failed to attract high levels of European immigration.
But before we talk about immigration, let’s look at our city-size map in 1860.
Couple of neat things to notice here. First, wow New York grew. By 1860, it has 1.2 million people, and is almost certainly the largest city in the western hemisphere. It’s so big you’d almost miss the fact that Pennsylvania under it has another 565,000 people. Baltimore, meanwhile, remains the largest American slave city, at 212,000 residents. Then comes Havana at 203,000 residents. In other words, Baltimore added 131,000 people, Havana added 64,000. And Havana fell from being #2 on my list, and probably in the Western Hemisphere top-10, to being #4 on my list, and quite possibly no longer in the Western Top 10.
We also have some new arrivals in the north. I only show northern cities if they were a top-10 US city in 1830 or 1860. Chicago has shown up in a big way, with 112,000 residents, outpacing Louisville, Charleston, and others. Buffalo has shown up with 81,000 residents, while Albany has grown to a handy 62,000. Meanwhile, Venezuela and Jamaica have abolished slavery, making them free cities now.
But we also have new southern cities. Memphis has grown to over 20,000 people, fueled by a boom in Irish immigration. New Orleans has tripled to a population of 168,000. Louisville has grown by 6 times to 68,000. Within the south’s interior as well, new cities lice Macon, Nashville, Atlanta, and Montgomery are cropping up and growing.
But the big southern grower is, of course, St. Louis. From 1830, it has grown by a factor of thirty, from 5,000 to 161,000, especially pushed by German immigrants.
So now, let’s talk about immigration.
Immigration Versus Slavery
The Contest of Labor Supply Institutions
Looking at the Americas in the 1790s, it would not have been immediately obvious that New York City was destined to become the uncontested pre-eminent city of the western hemisphere for the next century or two at least. Philadelphia, Boston, Charleston, and Baltimore could all give it a run for its money, and in fact more people resided within the current boundaries of Philadelphia than resided within the current boundaries of New York City. Likewise, an astute observer would see that Havana was nearly twice the size of New York City. Mexico City, Lima, Rio de Janeiro, all were substantially larger than any American city, and in many cases appeared more prosperous. Back in 1770 or 1780, Cap-Francais in Haiti (which I have labelled Cape Haitien; long story there) would have looked like a major metropolis for the area.
But by 1860, everything had changed. Free cities like Philadelphia and New York were super-dominant. Whatever shot the American south may have had at urban parity with the north was gone. Of course, our hypothetical astute observer would probably have noted that the south’s economy was more agricultural than the north’s, and also that the north had received more free migration historically, and maybe also that the largest southern state, Virginia, didn’t have a single top-10 city. Its largest cities like Norfolk, Alexandria, and Richmond simply were not the equals of northern cities in terms of size and complexity.
These southern disadvantages need not necessarily have persisted. Although today we often conceptualize migration as a “chain,” where past predicts future migration, the truth is that early “settlement” migration is often much more sporadic. New England was settled practically in a single wave. The Scots-Irish came in a very distinct period of time. Before modern transportation and communication technology, migration tended to be somewhat more episodic than it is today. Likewise, a greater frequency of wars and other catastrophes caused migration flows to face more interruptions and jump-starts than we see in developed countries today. Probably the only contemporary migration flows that are structurally similar to the typical migrant flows of the 18th and early 19th centuries are crisis migrants, like those currently arriving in Europe in large numbers.
In other words, it’s not unreasonable to think an observer in 1800 would expect immigration in the south to be roughly similar in scale to immigration in the north. And since immigration is a major source of population growth, which drives urbanization, it’s reasonable to expect the south could have kept pace.
And, at first, they did. Immigration to the south was equal to or greater than immigration to the north from 1790–1807. How do we know? Easy: we can assume at least some free immigrants went south, while all enslaved immigrants went south. From 1790–1807, enslaved migration amounted to 120,000 people, while all non-enslaved migration was 156,000. We can assume that the north got 100% of all free immigrants and still see that the south roughly kept parity with the north.
But in 1807, the legal slave trade was banned. Until about 1860, another 250,000 or so enslaved people would be brought into the future United States, such as by sale into independent Texas, or illegal human trafficking into the US. However, non-slave migration would amount to 7.8 million migrants, which is, ya know, quite a bit more than 250,000. Most of those 7.8 million would go to free states.
While urban slavery did exist on a large scale, the most valuable deployment of unskilled labor in the American south was in cotton cultivation. Hot, sweaty, back-breaking work, most people believed that free-labor cultivation of cotton on an industrial scale was impossible. All they had to do for confirmation was to look at the economic collapse of Haiti, or the struggling plantations of Mexico. In the north, however, the most valuable deployment of unskilled labor was not agricultural, but rather tended to be industrial: mining, factory work, infrastructure building, etc.
That last is worth noting. Academic studies have shown that the Irish in particular were heavily influenced by the availability of infrastructure jobs, building railroads and canals, and that the geographic distribution of these jobs influenced their distribution around the country. Likewise, these projects opened up new lands that could attract agricultural immigrants, who tended to be skilled or at least semi-skilled, especially those coming from Germany. And how did infrastructure get built? With public support! In some sense, then, at least part of the north’s attractiveness to immigrants was policy-driven. The choice to offer public support to infrastructure projects created unskilled jobs for immigrants and made new lands viable for settlement and small-holder agriculture at a rapid pace. The south tended to make less aggressive investments in these projects.
Even where the south did industrialize, Southerners often deployed enslaved labor rather than recruiting immigrants. The institution and asset existed due to the high returns to cotton cultivation, but once the institution of slavery was widespread, this labor market institutional path-dependence made southern industrialists often lean on enslaved labor for projects free workers did in the north.
Why avoid free labor? Oh, so many reasons! Consider China today. Labor costs are rising, and indeed above many other countries, like India. So why keep production in China? Simple: because China’s state supports business’ efforts to keep workforce discipline, maintain tight production schedules, and avoid regulatory hurdles. So why use slave labor for factories in the south? Simple: because you already have skilled managers who can keep them working all day long in awful conditions, because the racist southern society has set up special protections for white workers, and because you can sell a slave if you need to get out of the business.
This is a long tangent to get at a key point: the north supplied its unskilled labor demands by encouraging unskilled labor immigration. The south supplied its unskilled labor demands by encouraging the institution of slavery, which allowed the mobilization of a larger share of the slave population for market production than free labor allowed for the northern immigrant population. I know some readers will object to this sanitized description, but it’s crucial to understanding this period. Slavery can be seen as the most extraordinary intensive labor supply growth, while free labor immigration represents extensive labor supply growth. Whereas northerners recruited more people, southerners exploited their workers more and more efficiently. And, of course, brutally.
And, to be clear, the southern method made the south quite rich. Many estimates suggest that the south’s output per capita was equal to or higher than the north’s. The point here is that while slavery is obviously immoral, we shouldn’t be confused about why it persisted: it successfully met the demand for labor, and did so in a fashion that made free southerners vastly richer than agriculturalists anywhere else on the globe, and indeed richer than many industrialists in their own country.
So when we compare immigration and slavery, we should think of these as two different means of meeting the labor supply problem of industrial capitalism: one method primarily focused on intensifying the exploitation of the worker, the second method primarily focused on extending the size of the workforce. Of course both north and south pursued both strategies to some extent.
Living In The City
Charting the Population
Putting together comprehensive tables for all of the southern cities is too darn much to ask. So instead, I’m going to look at just 7 cities: New York City, Baltimore, St. Louis, Louisville, New Orleans, Charleston, and Philadelphia. This gives us 2 deep southern cities, 2 free cities, and 3 border-state cities. For each city, I’ve collected all the data available from the decennial censuses on the enslaved population and the foreign population. While the U.S. Census is a key source, I also used various state Censuses to supplement and confirm these figures. I extrapolate between Censuses using a fairly simple method I can outline for anyone curious.
So let’s look at our 7 cities. First, population totals from 1790 to 1870.
For each city, I try to use as close to the current boundaries as I can. So, for example, for Philadelphia, I include the whole current Philadelphia city, not just the original Philadelphia, now known as Center City. For New York, I use Manhattan, Queens, Brooklyn, Staten Island, and a portion of then-Westchester County for the Bronx. This results in some quirks: while the U.S. Census lists New York City as the largest city in the U.S. for every decade from 1790 to 1870, my data suggests that Philadelphia was the largest city until the early 1800s, and remained neck-and-neck with New York until 1820.
You can also see some interesting changes in the smaller cities. As of 1840, New Orleans caught up with and nearly surpassed Baltimore in population, but then lost ground over the next decade. We can also see New Orleans exceed Charleston’s population sometime in the late 1810s as American control of the Mississippi encourages westward migration and economic development, and American victory in the War of 1812 eviscerates Native American claims and continued resistance.
Then by 1850, Louisville, Kentucky also passes Charleston. But, curiously enough, Louisville isn’t the first western American city to pass up the traditional leader of the south. That honor goes to St. Louis.
Although it had been an outpost of the French, then Spanish, then American governments for a long time, St. Louis was nonetheless extremely small. As of 1810, estimates place it at around 1,600 residents. By 1820, it was still just 3,600. But then again, that’s 3,600 out of 67,000 people total in Missouri. St. Louis’ peer city in many regards is Louisville, with similar geography, demographics, history, and even name; in 1820, Louisville had about 4,000 residents, but that’s out of over 560,000 Kentuckians. St. Louis benefited from an ideal location on the Mississippi, but also from being the capital of the Louisiana Territories north of today’s state of Louisiana: it was a political hub, and originally settled mostly by southerners.
By 1830, Louisville had 10,000 people, and St. Louis just 5,000. But by 1840, Louisville had risen to 21,000, while St. Louis had risen even faster, reaching 16,000 people. And by 1850, the tables had truly turned: St. Louis had 78,000 residents while Louisville had 43,000. The disparity grew from there. By 1860, St. Louis had reached parity with its downriver counterpart, New Orleans, and by 1870, it had surpassed Baltimore to become one of the nation’s largest cities.
Where Was Urban Slavery Prominent?
Charleston… But Mostly Rural Areas
The chart below shows the enslaved share of the population for these cities, from 1790 to 1870. Obviously, over the course of the civil war, slavery was abolished.
The above chart shows my best attempts at estimating the enslaved share of each city’s population from 1790 to 1870. As you can see, not all cities are available in all years. I have nothing for Louisville before 1800, and nothing for St. Louis before 1810. Frustratingly, I have the best data for New York City due to New York’s state-level decennial Censuses in 1814, 1825, 1835, 1845, 1855, and 1865; but New York City also happens to have a slave population that declines early in the period. I also have Charleston-specific Censuses for 1848 and 1861, but they don’t seem extremely reliable, as they don’t even come close to matching U.S. Census data. Missouri Census records exist, but do not appear to be available for free anywhere. Louisiana Census records may exist, but I can’t find solid confirmation of that. Broadly speaking, I defer to the U.S. Census and, where state Censuses produce results that seem compatible, I incorporate their point-estimates as well.
What I want you to notice is that even slave-state cities tended to have a lower enslavement ratio than the nation on the whole by 1860. In St. Louis, the enslaved population peaked at about the national average level, in 1820, when the city had just a few thousand residents. In Baltimore, the enslaved population never even rose close to the national level. In New Orleans and Louisville urban slavery was very prominent, but declined from 1830 onwards.
Although urban centers did have slavery, it tended to be less common than in surrounding rural areas, and by 1860, was almost universally below national rates of enslavement. The exception here is Charleston. Indeed, Charleston is genuinely weird as far as cities go, with 53% of its population enslaved in 1860.
Charleston was the only significant urban area in 1860 to have over half of its population enslaved.
The graph below shows the enslaved share of the population for every city in the south that was a top-100 city in 1860. I’m going to call these 21 cities “the urban south.” About 60% of these “urban southerners” lived in Baltimore, New Orleans, or St. Louis, so let’s be clear that those three cities really were the pre-eminent cities of the south.
As you can see, the “big 3” southern cities had lower enslaved share than most southern cities. The other low-share cities include several right across rivers from free states: Louisville, Covington, Newport, Wheeling, Wilmington, DE. The next group, Alexandria, Memphis, Washington, DC, and Nashville, are all upper-south cities as well. When we look at the deep south in the Carolinas, much of Georgia, Alabama, or deeper into Virginia, we find cities with much higher rates of enslavement, well above the national level of 13–14%.
But even for that group of slavery-friendly cities, urban areas tended to have lower enslavement rates than surrounding counties, and none of the cities crack 40%. At 53% enslaved, Charleston was an extreme outlier among even deep-southern cities. At the same time, New Orleans’ 8% enslavement rate is also an outlier: the highest for a large city, but the lowest for a deep-southern city. It is perhaps remarkable that the two oldest cities in my sample are also the two most statistically peculiar as far as enslavement rates go.
This might also be a good moment to give you a sense of national urban geography in 1860. Here are the 100 biggest urban places (well, more like 98, as I had to combine a few to match my definitions):
You’ll notice the red dot in the middle of Cincinnati: that’s Newport and Covinton, Kentucky. And the blue dot on the edge of Louisville: that’s New Albany, Indiana. And the red dot on the edge of Philadelphia: that’s Wilmington, Delaware. Note that Louisville, Newport, Covington, and Wilmington all have pretty low enslaved shares, though Louisville has the highest.
Smaller cities tended to emulate the labor practices of neighboring large cities where possible. So Louisville has a substantial enslaved population despite being a pretty large city because it established a slave-based labor system and thus did not immediately and easily transition out of it. Despite being literally on the border to freedom, Louisville was over 7% enslaved, and a major port for sale of enslaved people down the Mississippi. Meanwhile, just up the river in Covington and Newport, the enslaved share was virtually zero, despite very comparable local cultures and ethnic mixes. What made Covington and Newport primarily free-labor cities was not just immigration or local culture or laws, but the dominance of Cincinnati-based economic networks and labor practices.
In St. Louis we get a large slave-state city with a near-zero enslaved share: why? Well, it’s because the city was still very small with essentially no established urban or industrial labor system when large immigrant groups began to arrive. They implemented their familiar labor systems, which didn’t depend on slavery.
The next two maps show the same cities, but instead of raw population, they show the enslaved population.
As you can see, this is a different geography. Despite being far smaller than Baltimore or St. Louis or New Orleans, Charleston had the largest enslaved population of any city. Also, just to be clear about the scale: the enslaved population of Wilmington, Delaware, is 4. Not 40. Not 400. 4. So, yeah, that dot should be smaller, but I couldn’t make it smaller. The enslaved population of Wheeling, Virginia was 0. Zip. Zilch. Nada.
So that’s the urban slave economy. A few other items worth mentioning: cities tended to have significant free black populations, in many cases equal to or larger than enslaved populations. So keep in mind that a 30% enslavement ratio probably means a greater than 30% black or mixed-race population. In many cities, whites were, if not a minority, then not an overwhelming majority.
Within cities, enslaved populations were typically female-dominated, about 55–70% depending on the city. Enslaved women were often used as domestic help, while physically fit enslaved men often commanded a higher price in agricultural use. Many of these men may have been leased out to nearby plantations, or even on sailing vessels, and may have had wives in urban areas. At the same time, it’s likely that many of these women were never able to marry. In agricultural areas, enslaved populations tended to be more male-dominated, and we know many of these men were never able to marry or have families.
The key to keep in mind here, however, is that slavery was almost always less prominent in cities than in surrounding rural areas, especially by 1860. As cotton agriculture became more intense and as the price of an enslaved person rose with import restrictions, it became less and less economical to deploy slaves as non-cotton laborers, even as a growing number of immigrants did make their way to the south. Indeed, it’s worth noting that the states of the future confederacy hit peak enslavement rates in 1860, at 41% of the population. In other words, the urban south, even excluding the big 3, had about half the enslavement rate of the south on the whole.
In 1860, the urban areas of the future Confederacy were about 21% enslaved. The rural Confederacy, meanwhile, was about 42% enslaved.
But let’s move on to immigration in the antebellum urban south now.
Why Weren’t There More Southern Cities?
Because There Were Too Many British Cities
Modern readers have a warped view of how humans have typically lived. This becomes especially obvious if you read fantasy literature. It’s astonishing how much of fantasy literature revolves around cities when the reality is that urban centers made up under 10% of the human population until sometime in the mid-to-late 1800s. In 1800, just 3% of the world’s population lived in a city.
Most people in most times have lived rural lives. With the advent of modern industry and technology, urban living has become far more viable and useful than ever before: but that technological and economic leap did not arrive simultaneously in all places, and in some places rising urbanization in one area drove intensified ruralization elsewhere. This is the case in the American South.
We have to ask the question: Why did the south have less urbanization than the north? The answer has to go back a long way, because even the very earliest population records show less urban concentration in the south than the north. Even as of 1790, of the 24 identified urban places, just 6 were in the south.
I’ve heard different answers. One perplexing answer I’ve heard is that the south had few cities because it had numerous navigable rivers, allowing each farmer to have his own dock for shipment. This doesn’t make a lot of sense, however, given that most of the South’s major cities are on those very rivers, and given that navigable rivers should create clustering around those rivers. Are we supposed to think that the South would have rivaled northern urbanization rates if there had just been a few more major waterfalls? That seems silly. Rather, good natural transportation ought to make urbanization easier: you can support a larger population when it’s easier to move people, goods, and especially food in and out of the city at need. That’s why cities grow up along natural harbors and riverfronts; not for the view, but because before railroads, waterways were vastly more rapid conduits of travel, trade, and supply than roads were.
There’s another explanation out there that I don’t think makes sense: the south was less urban because of its climate. I don’t buy it. Yes, the climate can be unpleasant. But that didn’t prevent large cities like Charleston or New Orleans from forming: nor cities like Havana or Cap Haitien or Santo Domingo or Kingston in the Caribbean. Yes, the climate could facilitate spread of disease that probably slowed urban growth, but that effect hardly seems likely to have prevented otherwise viable urbanization.
No, there’s a way more elegant answer for why the south had fewer cities, any why the cities that did exist grew up around the periphery: the South existed as the global countryside for industrialization. You can’t have the industrial revolution in England without American cotton; at least, not to nearly the same extent. And before that, early industrialization was underwritten in some part by profits from American tobacco, timber, and other raw materials.
Now, being the raw materials supplier is not normally a great deal for countries: you don’t get lots of value-added, so you don’t get rich. The American north transitioned away from this model and made a successful bid to capture value-added sectors, which benefited from urban concentration in order to provide access to labor, transportation, finance, skills, etc. But the American south had developed plantation slavery early on, and in this way, climate indirectly impacts southern urbanization: the South’s climate was good for many large-scale cash crops like cotton, sugar, or tobacco. The revenues from these cash crops were not high enough to make Southerners wealthy, but they were high enough to make free Southerners wealthy. The American South is perhaps one of the only rural societies in history to be rich and to politically dominate its urban neighbors without waging war against them. Historically, rural societies have generally only been richer or more powerful than more urban societies when they pillaged and dominated them, as they often did.
Why did rural societies often militarily dominate urban societies? Partly, rural life was more fraught with everyday peril, less likely to provide state-sponsored defense against violence, and thereby created “natural warriors.” Partly rural areas were often political borderlands with frequent wars, so lots of experienced combat veterans. And finally, rural areas lacked the high-value trades and skills of cities, so talented, ambitious people could only get ahead by taking up arms and becoming leaders in war. As such, the more rural and chaotic a society, the more likely it was to be able to dominate nearby urban societies if it could organize under a leader. Think of Roman dependence on barbarian soldiers, or steppe invasions of China, if you want golden historical examples, or indeed the recruitment patterns of the modern U.S. army, or even the American South’s vaunted cavalier-soldier tradition.
In other words, slavery made it possible for free southerners to command returns equivalent to the high-value trades and skills found in northern cities. Meanwhile, northern willingness to treat enslaved people as assets for use in financial markets meant southerners had fantastic access to credit: New York City was, in many ways, a southern city, given that a huge share of cotton was exported via NYC, and cotton-related securities represented the largest single financial asset traded there alongside U.S. treasury bonds.
So why no urbanization in the south? Because vigorous urban-led industrial demand in Europe and the American north created super-normal returns for free southerners to agriculture, which meant that the high-value trades and skills of urban centers were less appealing, especially given the specter of urban diseases. And where did southern cities crop up? In border states where plantation agriculture was less viable due to higher rates of escaped slaves or less favorable climate, and in a few major export-oriented ports like Charleston, New Orleans, or Mobile.
Industrialization and slavery yielded super-optimal returns for masters of the whip and of the wage, driving excess urbanization in industrial areas, and suboptimal urbanization in raw materials-producing areas.
Where Was Immigration Most Prominent?
Immigration and Population Growth
Just like I charted out the enslaved share of the population, so I now want to turn to the foreign-born share of the population. For the foreign-born shar, I have firm figures for 1850, 1860, and 1870, as well as a pretty good figure for New York going back to 1825. I also have a figure for “aliens not naturalized” for all these cities in 1820. Using some figures from New York State censuses on the non-naturalized share versus the foreign-born share, as well as some educated guesstimates, I extrapolate to a foreign-born population from the non-naturalized population. This method is fraught with peril, but it’s the best we’ve got.
I want to point out one fact here that shocked me: St. Louis had a higher foreign-born share than New York every year until 1870. Heck, in 1860, New Orleans almost had a higher foreign-born share than New York. This was all incredibly surprising to me.
You’ll also note that the foreign-born share declined from 1860 to 1870. That’s because (1) the foreign-born made up a disproportionate share of soldiers and casualties in the Civil War, (2) far fewer immigrants came to the US during the war than before it, for obvious reasons, and (3) post-war immigration to the south plummeted, with many foreigners leaving.
But really, St. Louis! Wow!
It’s also worth noting that urban Charleston had just barely above, and in some cases below, the national average level of foreign-born people. How crazy is that? It’s quite rare for large cities to have very low foreign-born populations, especially prosperous cities deeply embedded in international markets like Charleston. The apt comparison to New Orleans shows just how large an effect such international exposure could have.
But many southerners were displeased with the urbanization they saw in New Orleans, Baltimore, and other cities. Many openly worried about the rise of free-labor, immigrant-led, mass-movement unionism in cities like New Orleans, Baltimore, and St. Louis. The large southern cities were seen as being, in a sense, bastions of the north.
It’s easy to see why from the above chart. From 1850 to 1860, over half the population of St. Louis wasn’t even American-born, let alone a native southerner. New Orleans approached 40% foreign-born. These cities were as or more foreigner-dominated than the great northern industrial cities, and these foreigners brought with them linguistic and cultural diversity, different ideas about slavery, different expectations about cities and labor rights, and different preferences regarding public policy. Native-born southerners, therefore, often regarded even their own cities with suspicion, as these cities were largely non-Southerner-dominated.
Of course, these immigrants weren’t necessarily opposed to slavery. Many were attracted by the prospect of becoming a plantation master, or else profiting from the cotton trade somehow. Immigrant families dominated much of the cotton trade in New Orleans, especially through family connections in Europe and New York City. Memphis was an immigrant-dominated city in many regards, and also a rapidly-growing cotton-shipping hub. Louisville and Newport, Kentucky, were key ports for the sale of enslaved people further south, large foreign populations.
And on that note, let’s look at the foreign populations in 1860 in the south. First, their share of the urban population.
As you can see, the big cities also happened to be immigrant-rich cities. And most of the cities with the most immigrants were around the periphery of the south, while cities in the heart of the south like Nashville, Richmond, or Atlanta had far smaller immigrant populations. I’ve also included New York City and Philadelphia for reference.
As you can see, the urban south actually did attract immigrants. The reason the south on the whole didn’t attract as many immigrants isn’t that their cities were less appealing, but that the countryside was less appealing, and there were fewer cities to begin with. And why was the countryside less appealing? Because to be able to make enough money to justify the price of land, you had to have slave labor, and that was expensive, meaning there were significant barriers to entry. And why were there fewer cities? Because the economic model that made the South distinctive gave slave-owners super-optimal returns as they benefited from unremunerated labor, meaning there was less incentive to settle in cities. The rural poor as well, being enslaved, were not free to move to seek their fortune in the city.
We can look, then, at the distribution of the foreign-born population in the south.
The above map is showing you two key pieces of data at one time. The size of the bubbles indicates the size of the foreign-born population for each of the 100 largest urban places in America in 1860. The color of the bubble shows the foreign-born percentage of each city’s population. The more blue-green bubbles are above the national urban average, the brown bubbles are below the national urban average. And for reference, the national urban average is 34.8%. And that, my friends, is remarkable.
In 1860, foreigners made up about 13% of the population of the United States. However, they made up about 35% of the urban population, versus just 9% of the rural population.
In other words, while only about 12% of native-born Americans lived in cities, about 40% of foreigners did. That’s a remarkable difference.
But look at the geographic distribution of colors too. Sure, foreigners were more likely to live in eastern cities all things considered, but foreigners made up the largest share of population in the newer, western cities in Iowa, Illinois, Missouri, Wisconsin, Ohio, California, and Minnesota. This tells us something interesting about foreigners: they were uniquely likely to head into the frontier. International migration seems to have often been tied to a subsequent domestic migration.
To Be Continued
I’m going to drop this story right there, on a very inconclusive note. Why? Because this is already an excessively long post. But rest assured, I am not done. I will pick up in the next post on this topic at 1870, and look at southern urbanization until 1940.
But for now, you have enough information to start to reach some interesting conclusions. For example, the term “southern city” was, for the antebellum period, close to a contradiction in terms. Southern elites viewed their own cities with skepticism and concern, and the South’s position in the global economy was emphatically as a global countryside. To the extent that cities were cities as we think of them today, they ceased to be “southern,” regardless of their geographic location. The prime examples of this are cities like St. Louis and New Orleans, where immigrants changed the city and its local politics.
It is telling that the Union was able to secure all the major cities of the south with minimal contest. The urban population of the south simply was not the core of confederate sympathy. New Orleans, St. Louis, Baltimore, Louisville… these areas either didn’t secede, or fell quickly and gave the Union relatively little problem in holding them. The cities we think of as offering protracted resistance are cities like Chattanooga, Vicksburg, Petersburg, or Atlanta: cities that in most cases didn’t even make it into the list of major urban places for the nation. These were just towns with railroad junctions, not cities, with the political and economic implications that urbanization entails.
We can show this in election results. In 1860, southern Democrats secured 19% of the vote in Missouri. They secured just 2% in St. Louis. The winning candidate in St. Louis was Abraham Lincoln, with 40% of the vote, versus 10% statewide. In Maryland, Constitutional Unionists carried Baltimore, despite the southern Democrats taking the state on the whole. In Louisville, the Constitutional Union party won out as a home-town favorite, but the 2nd-place contender was Stephen Douglas’ northern Democrats. In Louisiana, while the southern Democrats won the state handily, they were in a dead heat in New Orleans itself.
Southern cities were not like the south that would ultimately secede. They were a region-within-a-region. Pro-slavery to be sure, but quite unlike their fire-breathing secessionist neighbors. And when it came to the actual secession votes, the cities were likewise rather less enthusiastic than the plantation-dominated countrysides.
The dominant cities of the south will change rapidly in the next time period. I leave you until then!
Check out my Podcast about the history of American migration.
If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. Or, just as valuable for me, you can click the recommend button at the bottom of the page. Thanks!
Follow me on Twitter to keep up with what I’m writing and reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!
I’m a graduate of the George Washington University’s Elliott School with an MA in International Trade and Investment Policy, and an economist at USDA’s Foreign Agricultural Service. I like to learn about migration, the cotton industry, airplanes, trade policy, space, Africa, and faith. I’m married to a kickass Kentucky woman named Ruth.
My posts are not endorsed by and do not in any way represent the opinions of the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I did not receive any financial support or remuneration from any party for this research. More’s the pity.
