A Very Brief Demographic History of Slavery
Tracing the Enslaved Population from 1630 to 1866
Regular readers know that I’m working on a new Podcast that will cover the history of migration in America, to go live in late January or early February. We’ll have a website and Kickstarter hopefully by the new year. The first episode or two will offer the surprisingly migratory history of bourbon. Once we’re done with that topic, my co-host and I will look at the history of the forced migrations of enslaved people. In preparation for that, I’ve put together what I believe to be one of the more comprehensive databases on the enslaved population of the Americas.
This post will essentially lay out five types of data. First, the raw numbers of the enslaved population, by future state, and for the whole territory of the future United States. Second, the enslaved proportions for the same. Third, a history of the legal and illegal forced immigration of enslaved people. Fourth, an outline of the internal slave trade, especially after 1807. Finally, a history of emancipation. My main focus here is not on offering a detailed discussion of slavery, as that’s what the Podcast will be for, but rather to just provide what I believe to be a pretty unprecedented look at the data on slavery.
Regular readers will recall that I have an existing interest in the history of slavery as it relates to migration. However, I am not by a long shot an expert in the history of slavery. If you are, and I’m wrong on something, I’d actually really love it if you could tell me. For this post, I’ve tried to rely on good outside sources as much as possible, but the nature of my project requires me to fill in the blanks.
Growth of the Enslaved Population
Natural Growth and Forced Immigration
You may have noticed I’m being a bit circuitous in my terms. I refer to slaves being brought to the United States as “forced immigration,” rather than “importation” or “trafficking.” I avoid the first because, as a trade expert by training, I don’t believe it’s correct to include human beings as a commodity, regardless of how woefully backwards a country’s tariff schedule may be. I avoid the second because it is often imprecise. Forced or coerced migration is my preferred language. I also generally try to say “enslaved people” rather than “slaves,” because, well, slavery is not an innate condition. It’s something done to a person. So, “enslaved” correctly highlights that these people were the target of a violent process.
So with that, let’s get to the core data. The below chart offers the raw enslaved population and, as usual, is interactive.
The thing that should stand out to you is that slavery never decreased in the United States. Never. The enslaved population grew pretty consistently. This, by the way, is fairly remarkable. American slave owners made a big deal of the fact that the American enslaved population was one of the only such populations in the world that was experiencing natural growth. They claimed this proved that American slavery was humanitarian. And, by the way, slavery in the United States was vastly gentler than in, say, French Saint Domingue or some Spanish colonies or other British colonies. So, I mean, okay, American slavery was not quite as horrific. Small praise given that slavery in Saint Domingue was so hyperviolent, it eventually led to the only successful slave rebellion in history and the establishment of Haiti.
Until 1808, when the slave trade was made illegal, the forced immigration of enslaved people was a major component of growth. After 1808, illegal forced immigration of slaves continued, but did, gradually, decline. However, as of the 1850s, some southerners were calling for the legalization of the slave trade again to meet strong demand for slaves and for the products they produced.
You’ll notice if you explore by state that some states did see declining slave populations. Virginia’s fell from 1830 to 1840, before rising again after. Maryland’s enslaved population declined slowly but steadily after 1810. West Virginia’s fell after 1850 (referring here to the counties that eventually became West Virginia). Delaware’s enslaved population declined fairly steadily after 1790. But overall, such declines were small, marginal, and rare. The big slave states like the Carolinas, Georgia, and Virginia saw increases even as of the very end of the period. Newer players like Mississippi and Alabama had explosively growing enslaved populations. Frontier and settlement states like Missouri, Arkansas, Florida, and Texas likewise saw rapid growth in the slaved population. Overall, very small declines in the old border states were being more than offset by fairly huge increases in core slave states and the frontiers.
Now, let me take a moment to discuss the data here. Obviously, I use Census data where available. But I’m going back to 1630 here. Some limited data from colonial censuses is available, and I use that source where I can. There’s of course debate over how many slaves lived in French Louisiana or Illinois, the rate of increase in Texas before and after independence, etc. Many readers will also be surprised to find that slavery continued in some northern states up until the 1830s. That certainly surprised me, and I’m indebted to the work done at Slavery in the North for really good research on that topic. For anyone with quibbles over the pre-1790 numbers (or my estimates for Oklahoma, for example), I invite you to provide a better number. I’m more than happy to answer questions about any state for any year. I should also note that I include both African and Native American enslaved people.
So with that, let’s jump over to the enslaved share of the population.
Shrinking Enslaved Share of the Population
The Great But Limited Power of Immigration
So the raw number of enslaved people was rising exponentially. But then again, so was the general population. As it turns out, the enslaved share of the population peaked around 1750, then reached another low peak around 1790, then fell steadily until emancipation. This decline was mostly due to rising immigration of free people, not emancipation.
The national total chart, as I said rises to 1750. Some caveats here: the very early periods are disproportionately reflective of Dutch and then early British New York, Spanish colonial holdings (especially Florida), and even French Louisiana. Eventually South Carolina and Virginia, then North Carolina, really get it into gear and run up the score on the slave share. Though it bears remembering that, as of 1750 for example, New York was 10–15% enslaved. Only North Carolina, Maryland, French Louisiana, Virginia, and South Carolina topped that. In 1776, on the eve of the American Revolution, there were small enslaved populations in Connecticut, Massachusetts, New Hampshire, New Jersey, and Pennsylvania: Rhode Island was over 5% enslaved, compared to Georgia, which was about 6%. In other words, while slavery was far more prominent in the southern colonies, it was by no means exclusive to them. Allegedly, slaves made up almost a third of the Boston workforce at various times in the mid-1700s.
When emancipation did come, it was almost always gradual, freeing enslaved people not yet born, and only after they were full adults, and usually with compensation for the previous masters. Emancipation almost always came alongside new restrictions on free blacks. But more on that later. The point is, although partial or full emancipation spread throughout the northern states over time, the “clean” Free vs. Slave state lines you’re used to are pretty much fabrications, at least until the 1840s.
You can also go state-by-state and see that the enslaved share of the population was rising in states like South Carolina, Mississippi, Tennessee, Texas, and Arkansas. Now, it was falling in some states, like Kentucky, Virginia, Florida, Missouri, Delaware, and West Virginia. These areas, however, didn’t see a decline in the enslaved population. They were just beneficiaries of relatively high immigration or domestic migration. So what created these declining shares, and the nationally declining share overall, was not any commitment to ending slavery, but just the arrival of new free people. It’s not clear this process would ever have naturally ended slavery.
Sure, high immigration made it possible for free areas to politically overpower the south, but the decline of the enslaved share of the population didn’t make the natural or peaceful elimination of slavery more likely, it just made the violent and coerced end of slavery more feasible. This is a crucial lesson to be learned about immigration and politics, I think. I expand on it in extreme detail in my previous post about Star Wars, nationalist politics, and immigration, but the short version is: using immigration to shore up your political position might not up your odds of achieving a natural, orderly, civil end to a disliked policy. But it probably will up your chances of winning the bloodbath that arises from the natural conflict of two polities within one state.
The Execrable Commerce
Forced Immigration of Enslaved People, Legal and Illegal
With the overall population dynamics of the enslaved population covered, I’ll look at what Thomas Jefferson called “this execrable commerce” in the first draft of the Declaration of Independence: the trade in enslaved persons. We’ll start with legal slave trade, and then progress to illegal. But first, the graph:
You can explore the chart and see the numbers as either a raw inflow number, as I’ve shown, or a % of the enslaved population, or a % of the total population, or a % of total immigration. I recommend doing this (click the heart icon). You’ll find that in the 1730s, enslaved people made up 70% of all immigrants to the future United States, for example. From the 1660s to the 1820s, enslaved people rarely made up less than 1/5 of all immigrants. And what led to their eventual fall below that threshold was less a decline in their immigration as a sharp rise in other immigration.
Here, I have to make a shoutout to the invaluable work done by the Transatlantic Slave Trade Database. These folks have meticulously catalogued known information on the transatlantic slave trade and made it downloadable in a usable format. I use their data as a baseline and annual volatility and fluctuation estimator. However, they track exclusively transatlantic slave trading. A very large share, perhaps a majority, of slaves sold into the future United States were shipped from the Caribbean. So where the TSTD estimates that 389,000 enslaved people were brought into the future United States between 1655 and 1866, I estimate the true number at around 720,000, based on other sources tracking inflows into the U.S. All the same, I do adjust annual flows in order to mimic annual volatility before 1807.
After 1807, I estimate illegal force immigration of enslaved people. This is obviously tenuous. I’ve found estimates suggesting between 100,000 and 350,000 enslaved people were brought into the future territory of the United States after 1807. The most common number seems to be 200,000 or 250,000. That appears to include slaves ultimately sold into Texas before 1848, though it’s not entirely clear. I have annualized this as a declining number of illegal inflows over time, with a bump during Texan independence. Honestly though, the true profile could be quite different. The overall scale, however, should be about right.
And that overall scale makes one thing very clear: from the 1720s to the 1850s, despite huge changes in legal regimes and global trade patterns, the number of enslaved people sold into the United States remained fairly constant. Sure, there are disruptions for wars, and the threat of the ban on imports of enslaved people in 1807 drove a huge spike as buyers rushed to get the slaves they would need. But even after 1807, illegal forced immigration of enslaved people averaged around 4,000 to 5,000 people per year, essentially the same as it was before the ban.
One reason this policy may have failed is how interdicted slave ships after 1807 were treated. “Cargo” was confiscated, including the slaves aboard. Any confiscated cargo could then be sold.
Wait. WHAT!?
Yes. U.S. law stipulated that enslaved people should be dealt with according to state laws where the confiscated cargo was to be disposed of. So if, as was generally the case, a slave ship was interdicted off the coast of a state that permitted slavery, then the enslaved people who would have been illegally brought in by a slave trader were taken from that trader, depriving him of income of course, but then promptly sold by the government to local plantations. In other words, one of the biggest revenue-earners from the forced illegal immigration of slaves was in fact the U.S. government, and state governments. One can’t help wondering if, between the bribes and the confiscations, the officials in charge of suppressing the slave trade really had their heart in it.
The Domestic Slave Trade
Trade Barriers and the Economic Interests of Slave Breeders
For many readers, my approach to slavery will seem somewhat detached. And it is: my goal here is just to present the facts as best I can. There are moments when I can’t help not commentating, however, like the above note on the revenues from the sale of confiscated enslaved persons. Now we’re going to reach another such point.
I mentioned that the American enslaved population was, alone in the world, experiencing rapid natural growth. This was not accidental. Nor was it a product of some exceptional fecundity among enslaved people. High growth of the enslaved population was at least partly a result of what can only be described as systematic breeding programs in the old slavery states. Here I rely on the valuable work done by In Motion: The African American Migration Experience entirely, with no new data of my own.
The below map shows which states were “gainers” and “losers” in the domestic slave trade from 1820–1829, for example:
As you can see, the old slave states of Virginia, the Carolinas, Maryland, New York, New Jersey, and Delaware were net exporters of enslaved people. Meanwhile, the newer, frontier states were net importers. Its from this domestic slave trade that we get phrases like “sold him down the river,” a reference to the role of Louisville and Newport, Kentucky as key points for the sale of Kentucky-bred slaves “down the river” to the deep south.
But it gets even more perverse than that. Remember how I mentioned the “gradual” emancipation of slavery in the north? Well, it was gradual enough that those states often saw large sales of slaves after such laws passed. Essentially, future emancipation lowered the net present value to slave ownership, creating better conditions for southern buyers to swoop in and make a purchase on the cheap. As I’ll reiterate in the next section, the northern record on slavery and emancipation is far dirtier than I expected.
Moreover, there’s a nasty side to the ban on the international slave trade.
Consider the economics here. The intensive slave-based agriculture practiced in the south was broadly less profitable, in terms of reinvestable returns, than industrialization. The ban on the slave trade drove up input costs on slaves as well, so it made slavery even less economically viable. However, you don’t need to import enslaved people if you can breed them. Virginia, Maryland, and South Carolina slave owners in many relied on the sale of slaves to meet operating costs, that is, the breeding of slaves for export was itself a business. And no surprise: by the 1840s, a slave could cost $1,000 within the U.S., while a slave bought in an African port costed just $30. That’s a monumental price wedge. Some returns were captured by smugglers, some by governments selling smuggled slaves, but mostly the winners were slave breeders and slave-exporting states.
That’s not all the import ban did. These excess returns made slavery economically viable in some areas even if slave-labor agriculture was not itself economically viable there. Why do Virginia and South Carolina slave populations continue to rise even after there is far more productive land being brought into cotton cultivation out west, and even after there’s ample local capital and labor to fuel more urbanized industrialization? Well, partly because slave-breeding offered a kind of extra value-added to the plantation agriculture value chain.
Now, with no ban on the slave trade, slave-labor agriculture would have been viable in more areas too, but perhaps with less value-added, and with greater supply disruptions as more foreign countries banned the trade. It’s possible that a legalized trade would have made slavery extend even further. But it’s also possible it would have allowed slavery to decline faster in border states like Delaware, New Jersey, Maryland, and Kentucky, where urbanization and industrialization and capital accumulation were giving rise to new investment opportunities. Instead of being poured into a slave-breeding and exporting business, those planter resources could have fueled a less slave-dependent economy.
On the other hand, maybe cheaper slaves would have made slavery even more widespread in border states where it eventually became cost-prohibitive for medium-size farmers to own slaves, and thus slavery would have been even more entrenched. That’s possible too.
But the point is, the ban on the slave trade directly increased the incentives to engage in the part of the slave trade that has the best shot at being as horrific as the transatlantic trade: the domestic trade and breeding program.
Emancipation, Gradually and then Suddenly
The Empirical Story of Legislated Freedom
There was a fun infographic that made the rounds a while back called “This is How Fast America Changes Its Mind.” They tracked various social reforms over time, showing how quickly various movements gained federal traction and action.
But they didn’t include the greatest social movement in post-Revolutionary American history, the cause of the abolition of slavery. I’ve put together that data. In the chart below, I provide a count for how many states had abolished slavery and emancipated enslaved people. The trick is, there are lots of ways to do this. For states with “gradual emancipation,” do we count the data the law was passed? The date it went into force? The date when there were truly no more enslaved people? What about states that kinda-sorta abolished slavery, but not totally, and had to take a couple whacks at it?
What I’ve done is come up with a simple system. There is “first emancipation,” which is whenever the first law prohibiting enslavement came into effect, and then “full emancipation,” which means that these states freed essentially all slaves.
As you can see there was a flurry of full and incomplete emancipations after the American Revolution. Then, from the early 1800s on, the number of free and slave states rises steadily. The emancipated states rise faster, and by the late 1840s, there are no “partially free” states left. The country is well and truly divided, slave and free.
But holy cow, it takes until 1848 for that full division to occur. Before the late 1840s, it’s more appropriate to talk about some kind of spectrum of free- or slave-statehood.
Now, that chart only tracks states that are part of the union. We can also track the status of slavery in all the territories that would eventually become the United States. So this will show the impact of policy decisions made about non-state territories, as well as the policies of foreign governments holding territory that would eventually become the United States, most notably Mexico.
This chart shows a bit more clearly what the political stakes were. Looking at just the states it actually does kind of look like emancipation might have eventually won out. But we should recall that the most bitter fights about slavery, and in fact the political fight that started the civil war, was about whether or not slavery should be extended into the territories, or contained to its then-current limits.
As you can see, the number of future states in which slavery was legal fell fairly sharply during the late 1700s due the accession of the Northwest Territory states closing those lands to slavery. Then the closure of the northern Louisiana Territory states to slavery created a further drop. Then the closure of the territories taken in the Mexican-American War caused a further drop. But then in 1836 Texas secedes from Mexico and embraces slavery with gusto. In 1851, the Utah Territory permits slavery. In 1858, slavery was permited in the New Mexico Territory, including the territory that eventually became New Mexico, Arizona, and Nevada. While the accession of Oregon and Minnesota tipped the Senate balance against the slave states, as far as the overall territory of the United States was concerned, slavery was advancing.
So of course Lincoln’s platform was about keeping slavery contained: because the 1840s made it look like slavery was going to advance, claiming more territory, extending the reach of slavery through cases like Dred Scott, and the implementation of more and more invasive fugitive slave laws.
When emancipation came first for some states in 1863 and then for all in 1865 or 1866, it was both building on a pattern of gradual emancipations and the extension of emancipation that had been occurring for 80 years and reversing the revanchist gains of slavery over the previous decade. In other words, emancipation was far messier than our imagination makes it. The clear lines between “slave” and “free” states were actually fairly muddled well into the 1840s. Who was “winning” was far from clear, and depended on how you defined the playing field. And there’s essentially no evidence that the major slave states were ever going to emancipate their own slaves, gradually or otherwise.
Conclusion
Slavery and Migration
As I said at the start, I’ll be venturing much more detailed explorations of some of the stories behind the forced migrations of American slavery in the Podcast come February. For today, I hope that the information I’ve offered helps you see that the history of slavery in America has far fewer “good guys” than we might like and is far more muddled than the manichean story we tell ourselves. Policies that strike us as obviously humanitarian, like banning the slave trade, may indeed have been good, but also came with some perverse side effects. Even state-level emancipation often simply induced faster export of enslaved people instead of any actual freeing them.
If you want more information on this topic, check out the links I included. And look at these cool county-level maps. And also, stay tuned for more information about my upcoming Podcast!
See my previous two posts, offering a detailed, interactive history of immigration, and discussing the domestic geography of immigration.
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.
