March 26, 2017

Slaves in America Would’ve Been Slaves in Africa Anyway


Based on the fact that ~41% of all blacks in West and Central Africa prior to European conquest were slaves, the extreme increase in price of slaves from the relatively small increase in demand that the Atlantic Slave Trade represented, and the anecdotal evidence that more slaves were put to work within Africa following Britain’s global ban on slavery, we can say that the Europeans likely did not cause a single black person to be a slave who wasn’t already going to be one.

1. The Population of West and Central Africa 1500 to 1800

  1. The Proportion of Slaves in West and Central Africa
  2. The Number of Slaves in West and Central Africa
  3. Atlantic Slave Disembarkations
  4. The Inelasticity of Supply
  5. Our Hands are Clean

1. The Population of West and Central Africa 1500 to 1800

Population Data on Central and West Africa

Region Year Population WAF proportion of CAF
West Africa 1907 33.385 million 2.606
Central Africa 1907 12.81 million
West Africa 1950 70.54 million 2.674
Central Africa 1950 26.38 million

Population Estimates for West Africa, Central Africa’s population inferred as a proportion of the West African population estimate based on the trend from 1907 to 1950:

Region Year Population WAF proportion of CAF (estimate)
West Africa 1500 20 million 2.361
Central Africa 1500 8.47 million
West Africa 1800 25 million 2.453
Central Africa 1800 10.19 million

Some other population estimates that will be used later based on the above extrapolations:

Region Year Population WAF proportion to CAF (estimate)
West Africa 1525 20.42 million 2.369
Central Africa 1525 8.62 million
West Africa 1866 30.17 million 2.547
Central Africa 1866 11.85 million
West Africa 1675 22.92 million 2.414
Central Africa 1675 9.49 million
West Africa 1775 24.58 million 2.445
Central Africa 1775 10.05 million
West Africa 1805 25.08 million 2.455
Central Africa 1805 10.22 million

And so the populations of West and Central Africa as regions average to 31.83 million from 1500 to 1800.

Central Africa is defined as Chad, Cameroon, Central African Republic, Congo-Kinshasa, Congo-Brazzaville, Angola, Equatorial Guinea and Gabon.

West Africa is defined as Nigeria, Niger, Benin, Togo, Ghana, Ivory Coast, Burkina Faso, Mali, Liberia, Sierra Leone, Guinea, Guinea-Bissau, Senegal, Gambia, Mauritania.

Population of West Africa in 1500 and 1800

1950 population numbers

1907 population numbers

2. The Proportion of Slaves in West and Central Africa

Encyclopedia Britannica’s Guide to Black History – Slavery

Encyclopedia Britannica’s claims for the proportion of slaves in various African states

State Approximate Modern Location Years Proportion slave
Sokoto Nigeria, Cameroon “19th century” (1800-1899) “One half”
Fulani Jihad States All West Africa North of Coast 1750-1900 “One half”
Ghana Ghana 1076-1600 “One third”
Mali Mali 1200-1500 “One third”
Segou Mali 1720-1861 “One third”
Songhai Mali, Niger 1464-1720 “One third”
Ouidah Benin “19th century” (1800-1899) “One half”
Kanem-Boru Chad, Niger, Nigeria, Libya, Cameroon, Sudan, Central African Republic, Algeria 1580-1890 “One third” , “40 percent”
Berber-Tuareg Sahel region “Until 1975” “15 percent… to perhaps 75 percent”
Senegambia Senegal, Gambia 1300-1900 “One third”
Sierra Leone Sierra Leone “19th century” (1800-1899) “One half”
Yoruba Benin, Togo “19th century” (1800-1899) “One third”
Ashanti Ghana “19th century” (1800-1899) “One third”
Duala Cameroon “19th century” (1800-1899) “One half”
Ibo Niger “19th century” (1800-1899) “One half”
Kongo Congo “19th century” (1800-1899) “One half”
Chokwe Angola “19th century” (1800-1899) “One half”

If the estimates for the proportion of the population that are slaves are taken as precise, it averages to 41.61%.

If these numbers seem extreme, remember that Athens was said to have been about one-third slave. And the whole of the Confederate States of America was approximately one-third slave; these are not outrageous numbers for slave societies.

3. The Number of Slaves in West and Central Africa

So we have a good idea of what the population of West and Central Africa was, and we have consistent subjective estimates of the proportion of people that were slaves. The next question is how many people were alive in West and Central Africa from 1500-1800.

The average age of childbirth in African countries is today around 27. So if we assume the average age of motherhood was 25 back in 1500-1800, then we have a generation time of 25 years.

If the average life expectancy was 50 years, then there were roughly 190.98 million people who lived in West and Central Africa from 1500-1800. If 41.61% were slaves, that would give us 79.45 million slaves within West and Central Africa between 1500 and 1800.

4. Atlantic Slave Disembarkations

Roughly 12.5 million slaves disembarked from Africa between 1525 and 1866 to the Americas. Over that same period, we would expect there to have been 242.32 million people to have lived in Africa, and of those 100.82 million would have been slaves between 1525 and 1866.

And so based on this, the Atlantic Slave Trade was about 5.84% of the total slave trade within West and Central Africa at the time.

5. The Inelasticity of Supply

Elasticity of supply is simply a fancy way of saying “does quantity supplied increase in response to an increase in demand?”. For slaves, if the supply is “elastic”, that means that the suppliers of slaves can easily supply more slaves if the demand goes up. If it is “inelastic”, that means the suppliers can’t easily get more slaves just because more people are willing to buy them.

If the supply is inelastic, then any increase in demand will simply result in the price of slaves increasing. For example, if the supply of slaves was totally fixed and could not be increased, then new buyers would simply bid out some of the previous buyers for whom slaves are now too expensive.

And here we can actually compare the price of slaves in Africa, the number of slaves shipped to the Americas in the prior 25 years (roughly the generation time), their price in Africa, and the Atlantic Slave Trade as a percentage of the overall slave trade in Central and West Africa.

Price of slaves in Africa, number of slaves shipped to the Americas, est. number of slaves in West and Central Africa, and the Atlantic Slave Trade as a proportion of the overall slave trade in West and Central Africa by year

Year Price of Slave IN AFRICA Number of slaves shipped to the Americas in prior 25 years Estimated number of slaves in West and Central Africa Atlantic Slave Trade as proportion of overall slave trade in West and Central Africa
1675 3.33 pounds 0.488 million 13.486 million 3.492%
1775 18.43 pounds 1.925 million 14.410 million 11.785%
1805 26.86 pounds 2.009 million 14.688 million 12.032%

Slaves shipped to Africa by year available here

Price of slaves in Africa by year available here

Now the extreme increase in price in response to the very small increase in quantity demanded that the Atlantic Slave Trade represented is evidence that Africans couldn’t readily just increase the supply of slaves. I.e. there wasn’t much or any “excess supply” with which to supply the Europeans. And so they would have to either sell some of their slaves they were already using, or pick up arms and go try to enslave some peoples who up to that point had managed to resist enslavement.

The price increases are not necessarily entirely caused by the increase in demand over those years. It’s impossible to say with any precision what caused the price increases. But the general pattern is of extreme PRICE inelasticity, with an 8.293% increase in relative quantity demanded coinciding with a 453.453% increase in price, and then a 0.247% increase in relative quantity demanded coinciding with a 45.741% increase in price.

Again, there are all sorts of factors that could be at play that I don’t know about. However, the limited data that exists points to extreme price inelasticity, which is evidence of inelasticity of quanitity supplied – which is a long way of saying “it looks like the Africans couldn’t increase the number of slaves, that the supply was fixed, and as a result when new buyers came along, the price just shot through the roof.”

According to the Encyclopedia Britannica article:

“After the limiting and then abolition of the transatlantic slave trade, a number of these African societies put slaves to work in activities such as mining gold and raising peanuts, coconuts (palm oil), sesame, and millet for the market.”

– Of course that is not data, but merely a subjective impression being that the end of the Atlantic Slave Trade did not result in fewer black slaves. They just stayed in Africa.

And when ~40% of the population is enslaved, and people resist with violence being enslaved, it is certainly not difficult to imagine that the supply of slaves was probably quite inelastic.

6. Our Hands are Clean

The life of a slave in the Americas was certainly better than the life of a slave in Africa. It’s possible that the average non-slave African had a better life than the average slave in the Americas, but that’s not the relevant comparison.

Now lets say that the Atlantic Slave Trade did not create a single slave, that it merely moved slaves from Africa to the Americas, where they ended up having a better life and a better chance of surviving (even taking into account the middle passage, which later generations would not have to endure anyway).

Would we then say that the Europeans engaged in a wicked act? Consider that if the Europeans did not engage in this act, those slaves would have remained in Africa, where they would have been more likely to die, where their lives would have been worse – still slaves. In this case, the Atlantic Slave Trade IMPROVED their lives relative to being a slave in Africa.

And so whether or not the Atlantic Slave Trade was immoral is totally dependent on the elasticity of supply. Did the supply of slaves increase in response to the Atlantic Slave Trade, or was a relatively fixed supply of slaves simply transferred out of Africa?

And everything points to no increase in the number of Africans who were slaves. And if there was a slight increase in blacks enslaved, it is certainly less than the 12.5 million that disembarked from Africa, and exchange those 12.5 million had a better chance of surviving, and certainly their progeny had a much better life than those who remained in Africa.

We did not cause them to be slaves, nor was it our duty to buy their freedom. Nobody would be cursing Europeans if they did not buy a single slave from Africa. Yet if Europeans did not buy a single slave, those Africans would have been worse off. But by doing something that made them better off compared to inaction, Europeans are condemned as if we caused their enslavement.

The Africans were slaves to themselves well before we came along. We caused none of it, only better conditions in the Americas and their eventual freedom.

Facebook Comments
  • Awakened Saxon

    How do you take the life expectancy, generation time, and average age of motherhood, and move to a population estimate from there?

    • Ryan Faulk

      Average age at birth = 25, average age of death = 50, then you have 2 full generations.

      On average, some die sooner, some die younger.

      These are used to extrapolate how many adults lived in Africa during that time. The total number of births will of course be higher, but that’s going to have infant mortality.

      • Awakened Saxon

        Both of your replies make sense, but it doesn’t tell me the actual math behind how you reached your conclusion. I intuitively understand that you could reach some sort of an estimate of the number of individuals that lived across a span of time using information like generation time, life expectancy, and population size, but I don’t know how you would actually do that mathematically. Basically, if I had a hypothetical example spanning 200 years where the average population was 10 million and the generation size was 30 years and the life expectancy was 50, how would I reach an estimate of the number of people that lived during those 200 years mathematically? I’m curious how the analysis was actually done. I linked this to a liberal friend of a friend and they accused the author of “dishonesty” because the analysis wasn’t included, I just assumed it wasn’t included because it would be excessively lengthy relative to its perceived importance.

    • Ryan Faulk

      The “real” life expectancy would probably be something like 30, but that factors in infant mortality. I’m not interested in that because I only care about how many “adults”, i.e. people who could have been enslaved, there were.