This started off as an email to a friend, me taking an opportunity to lay out something I’d meant to try to think through for a long while, about slavery in the US south as a form of labor under capitalism.

Marx’s series-as-abbreviations are useful. I know you know all this as well as I do, I’m not trying to be pedantic, but laying this out helps me think. Waged labor is M-C(lp)+C(mp)…P…C’-M’. The difference between M and M’ is surplus value. C(mp) are purchased at value, as a general rule and systemic tendency, and so they may transfer value to C’ via P, but are not productive of new value. C(lp) however is the peculiar commodity with the use value of producing value greater than the value it took to produce. So C(lp) is purchased at cost and yet produces a surplus anyway. The whole serious M-C…P…C’-M’ has no fixed temporal scope. Workers in an high speed industrial workplace might turn out multiple products every day producing many individual commodities that fall under C’, or early carriage makers and producers of luxury goods today might spend a long time all producing one commodity which fall under C’. In the former case, C’-M’ happens frequently, in the latter it happens slowly. In both cases, the surplus is re-invested. This too has no fixed temporal scope, it can be reinvested quickly and an ongoing basis (continual purchase of small innovations in machinery etc as well as new C(mp)+C(lp)) or reinvested less frequently in major purchases to expand operations. The purchased time of C(lp) is also variable. The time worked per 24 hour period changes along a number of axes, as does the pay period (“working day” seems to me a misleading term, mixing together both sustained periods of labor and pay periods when really those rarely coincide – few people work by the day and also get paid per day, most people nowadays seem to have multiple labor periods per pay period). In the former type of example, high speed industry producing many goods, workers may produce a great many commodities over a single period of labor and even more over a single pay period. In the latter, slower rate of production (but not necessarily slower in terms of surplus value), the workers may work multiple periods and even go multiple pay periods before producing a single object or batch of objects C’.

Maybe this is obvious, but I want to see slavery as practiced in the US south generally as fitting into this description of capitalism a relatively unproblematic case in one sense, in that it fits into the highly abbreviated series above. (I know slavery is highly particular and so on, and it’s not explained by so much as described by this series. I’m not trying to minimize complexity, I just don’t think the “was southern slavery capitalist or not” issue is complex).
Slavery was a case of M-C…P…C’-M’ where the time of purchase for C(lp) was very, very long and the initial expenditure was very high.

I can’t recall where I got the numbers from, but in the 1850s US slave prices could range from $300 to $1300 or so. Let’s say a cotton planter bought 10 slaves for an average of $1000 including costs to house and feed them in that purchase price for simplicity’s sake, and put them all to work in cotton production. Let’s say they produced for 10 years.
Let’s say other production costs — the C(mp) — cost $50,000. That’s a total cost of $60,000. If the cotton produced, the total C’ produced, is sold for greater than $60,000 then there’s a surplus produced. Assuming that all of that surplus comes only from the sale of cotton produced (so no sale of slaves or other goods) then it only comes from the slaves’ labor. This seems to me an unproblematic example of M-C…P…C’-M’, and of capitalism, making the slaves exploited in the same sense Marx described, despite the many other incredibly important particularities of slave labor. Whether not there were actual surpluses is an empirical question but it seems implausible to me that there weren’t in general across the south.

Another way to think of the point – Marx polemicizes about women and children at work in the chapters on machinery and the working day in v1 of Capital, he accuses heads of household of getting into slavery by making their wives and children work. This is mostly polemic but there’s an element of truth to it in that children and women were legally disqualified labor, not formally free in the sense of rights-bearing and able to own their wages, and Marx implies I think a level of informal coercion here as well by heads of household. In those cases, workplaces full of women and children, the exchange is not (or at least not solely) a case of exchange between capitalist and someone who owns their own labor, rather it’s at least in part or is at least potentially an exchange between capitalist and heads of household for the labor of a third party (the women and children), and is in this sense really quite similar to slavery.The thing is, in workplaces full of women and children who are not really free laborers even in Marx’s sense of the term, it seems to me that those are still fully capitalist workplaces, fitting into Marx’s descriptions.