“Direct slavery is as much the pivot upon which our present-day industrialism turns as are machinery, credit, etc. Without slavery there would be no cotton, without cotton there would be no modern industry. It is slavery which has given value to the colonies, it is the colonies which have created world trade, and world trade is the necessary condition for large-scale machine industry.”
Marx’s claim reappears in a body of scholarship loosely called the “New History of Capitalism,” which is a stream of books and articles that have appeared over the last decade or so. They are written mostly by historians, and they put slave-produced cotton at the center of the story of American economic growth. Books like Sven Beckert’s Empire of Cotton, Walter Johnson’s River of Dark Dreams, and Edward Baptist’s The Half Has Never Been Told tell vivid and moving stories about the obscene brutality of American chattel slavery, and they use words like “indispensable” to describe slave-grown cotton’s role in American industrialization and modern prosperity. We are rich today, according to this line of reasoning, because of slavery. The cruel tyrant King Cotton, it seems, has returned to take his throne.
Economic historians have interrupted the coronation. The New History of Capitalism has come under withering criticism from economic historians like Stanley Engerman, Eric Hilt, Alan Olmstead and Paul Rhode, Robert Margo, and now Gavin Wright in a new paper in the May 2020 issue of the Economic History Review. American economic progress, they argue, did not depend on slave-produced cotton. In his new paper, Wright responds to the central thesis of the New History of Capitalism by arguing that American slavery was a drag on the American economy, not a source of economic growth.
This might seem puzzling. The New Historians of Capitalism are keen to remind us that slave-produced cotton represented about half of American exports and was the leading commodity in Atlantic trade. In addition, they show in gruesome detail that there was blood on everyone’s hands, from whip-wielding overseers to bankers approving loans collateralized with slaves. The leap from “slave-produced cotton was a big part of American exports” to “slave-produced cotton was ‘indispensable’ to American prosperity” runs into a lot of problems, though.
First, slave-produced cotton was a big part of a small sector. Compared to American exports, slave-produced cotton is huge. Compared to American output in its entirety, it’s a lot less impressive. Make no mistake: the 5 or 6 percent of GDP cotton is a lot when you compare it to other sectors. It is not the difference between national prosperity and national poverty, however.
In The Half Has Never Been Told, Edward Baptist claims that “more than $600 million, or almost half of the economic activity in the United States in 1836, derived directly from cotton produced by the million odd slaves” (Baptist, p. 322). He gets this number by double- and triple-counting the same economic activity, by confusing stocks and flows, and as the economic historian Bradley Hansen argues in a devastating critique, “simply pulling numbers out of thin air, or a hat, or wherever it is that he gets them.” Stanley Engerman’s description is apt:
“To go from a value of the Southern cotton crop in 1836 of ‘about 5 percent of that entire gross domestic product,’ to ‘almost half of the economic activity in the United States in 1836’ (pp. 321-22) requires his calculation to resemble the great effects claimed by an NFL club when trying to convince city taxpayers that they should provide the money to build a new stadium because of all the stadium’s presumed primary and secondary effects.” (Engerman, p. 641)
That’s a degree of credibility we should all work diligently to avoid. Baptist, like the subsidy-seeking owner of the local sports team, is essentially arguing that 2+2 actually equals 8 because of “secondary effects.” 2=1+1, of course; therefore, 2+2=2+2+1+1+1+1=8. QED.
Second, just because something happened as it did doesn’t mean it literally could not have happened any other way. That as a matter of historical fact a lot of capital was tied up in slave-produced cotton does not mean that the capital could not have been deployed elsewhere. Slavery, as Wright argues, was not essential for cotton cultivation the same way it might have been essential for Caribbean sugar cultivation. He explains:
“Early mainland cotton growers deployed slave labour, not because of its productivity or aptness for the new crop, but because they were already slave owners, searching for profitable alternatives to tobacco, indigo, and other declining crops. Slavery was, in effect, a ‘pre-existing condition’ for the nineteenth-century American South.” (Wright, p. 354)
Slavery was not essential or “indispensable” for cotton cultivation, as the widespread cultivation of cotton around the world and as the recovery of the southern cotton economy in the late nineteenth century demonstrate. Slavery and cotton, according to Wright, were connected “through historical legacies rather than technological or economic imperatives” (Wright, p. 355).
It’s also a mistake to think that just because textiles played a big role (compared to everything else) in industrialization that textiles had to play a big role in industrialization. Textiles can be made out of a lot of things, though Beckert notes that producing an equivalent amount of wool would have needed “about 1.6 times the surface area of today’s European Union” for sheep grazing (Beckert, loc 206 of Kindle edition). That might have been impractical, but people can do a lot of things besides graze sheep, spin yarn, and make cloth. The historical pattern might have been different and the Cotton Bowl in Dallas might not be one of the “New Year’s Six” college football bowl games, but a change in the composition of economic activity does not need to mean a change in the level of economic activity.
Think about the characters in Star Wars and the actors who almost played them. What if Luke Skywalker had been played by Kurt Russell? What if Al Pacino and Burt Reynolds hadn’t turned down the role of Han Solo? What if Jodie Foster and not Carrie Fisher had been cast as Princess Leia? What if Darth Vader had been voiced by Orson Welles? The cultural patterns of the last few decades would have been different, obviously, but it’s hardly clear that Star Wars wouldn’t have become an iconic film franchise without Harrison Ford as Han Solo or James Earl Jones as Darth Vader’s voice. It’s also not like Star Wars was the only thing in theaters or on TV in 1977. Maybe Smokey and the Bandit–which came in second at the box office in 1977–would have become an iconic franchise had Star Wars been canned halfway through production. Or maybe Starship Invasions. Or Spider-Man.
In other words, Star Wars still would have happened had Han Solo been played by Al Pacino or Burt Reynolds. The American economy would have grown had cotton–or something else–been cultivated by free labor rather than slave labor. The specific pattern of economic activity would be different, but it’s hardly clear the level would be lower. If anything, it almost certainly would have been higher: as Gavin Wright and AIER’s Robert E. Wright explain, slavery is a millstone around the neck of an otherwise-free economy. It enriches some people, but more generally it spreads poverty rather than prosperity.
As with many history books, the contributions to the New History of Capitalism assemble an impressive catalog of facts, and the books are written in such a way as to really bring the human tragedy of chattel slavery to the forefront. The New History of Capitalism misfires badly, however, in its interpretation of these facts. Slavery was not necessary for cotton, and cotton was not necessary for industrialization. Had chattel slavery never taken hold in the United States, we would very likely be richer than we are today. The “slavery=>cotton=>industrialization=>modern prosperity” argument seems to condemn American capitalism, but it is wrong.
Phil Magness has also written on this for AIER, e.g. here and here. In the wake of the New York Times 1619 project, the Economics Detective podcast did a series of episodes on the relationship between slavery and capitalism. The episode with Alan Olmstead and Paul Rhode discusses the empirical problems with the literature. The episode with Jeffrey Rogers Hummel takes a look at the Civil War. The episode with Robert Wright explains how slavery ancient, antebellum, and modern impoverishes. The episode with Phil Magness offers a broad overview of the debate. I thank Phil Magness and Robert E. Wright for useful comments on this piece.
Made available by the American Institute for Economic Research.