,
26 tweets,
9 min read
Read on Twitter
Economic history is returning to (particularly bad) form. Reading the economic history of slavery and thinking about contemporary economic history research has convinced me that this generation is making the same mistakes of previous generations, and hurting the field overall.
Context: I’m reading and teaching the economics of slavery. So going back through this debate is like a time machine full of famous economic historians and intense debate. It was the best and worst of economic history. We have not learned our lesson.
My take is that the main problems of the economic history of slavery were a lack of serious engagement with the historical literature and an overly zealous use of data to answer questions of historical importance, but which could not be answered with the data at hand.
Let’s unpack this. The historiography of slavery as told by economic historians (on BOTH sides of the debate) was narrow and incomplete. Even for its time it neglected the burgeoning work of African Americans scholars and could not/did not incorporate new Marxist readings.
Such an analysis would have opened up new questions about culture, interpersonal relations, and plantation operation and management that economists have yet to say much about. The lack of a cultural perspective is a glaring hole in the literature. Plantations were not monolithic
Culture feeds back to norms and behavior, so without understanding the culture our models are gross simplifications at best and caricatures at worst. With an urge to keep things simple we also make them ahistorical.
Example: the economic history of slavery is devoid of time. There is very little change in the institution and its operation. That’s inherently false. Yet the static history is the most widely known from the field. Temporal and geographic variation is not emphasized.
Example: the concentration on a profit motive is reasonable, but denies the fact that enslaves could see profit very differently depending on how and when they were looking at the institution. I think new work in history is shedding light on this (see @CC_Rosenthal)
Suffice to say that not being deeply involved in the work of historians led us to overlook at lot. The fact the John Hope Franklin (the doyen of the time) dismissed the enterprise is very telling about how the contribution of economic historians was received.
The second problem was the misuse of data. More specifically, the belief that data itself was an answer to a question, or that the answer from data was the better or best answer. This is the economic historians road to Hell.
Why? Data is itself the function of a historical process. For a variety of reasons we do not interrogate it the same way traditional historians do. We are better, but the questions we ask rarely touch on more probing questions. Why is what we find there instead of something else?
Example: Fogel and Engerman use the lack of black prostitutes as an indicator of lack of demand for miscegenation. This is a hot mess, rightly corrected by Gutman. The data here is never contextualized to the institution. A simple count of mulattos by age in 1880 would help.
Example: Using the age at leaving home is not the same as marital breakup. This was a mess too! Jon Pritchett and I have used data to show that the slave trade did indeed lead to high rates of marital breakup. It’s not a difficult calculation to make. sciencedirect.com/science/articl…
When the data and models are put together you get some scandalous conclusions. It may seem reasonable under the assumption of full ownership of human capital, but the “finding” that slaves were well fed does not hold up to serious scrutiny at all. cambridge.org/core/journals/…
But what does this have to do with today’s economic history? Well, in my opinion, a whole lot. First, we’re still doing a bad job of actually reading and understanding history and historical context. Let me be blunt— We. Are. Absolutely. Terrible.
When was the last time you read an economic history paper that included historiography? That considered multiple readings of history (feminist, critical race, Marxist, etc.)? These are important ways in which the narrative is formed and reformed. They should influence our work.
Even more, these are not even questions that we ask. There is no supply or good historical knowledge because there is no demand for it from the field. We can (and some do) get away with trivial historical knowledge. It’s a bad look for us. But our incentives are clear to be Econs
Similarly, we are still convinced that data replaces historical import. We are more prone to argue over identification (a second or third order concern) as opposed to historical significance. Not economic significance, historical significance. There is a huge difference.
I’ve seen debates about what factors may be omitted that are absolutely laughable because they lack any historical insight. But in the shallow pool we swim in its enough to tread water, I suppose. I’ve seen work where people are making things up, out of thin air and hope.
Let me be clear— machine learning and linking are not historical knowledge nor do they by themselves answer any question, even a trivial one, that a historian has asked. This is hyperbole, but goes to my point. The data and methods must be in the service of history, not Econ.
The economic questions we ask are, by definition, ahistorical. We must discipline them through our knowledge and understanding of history. We are doing less and less of that in an effort to impress our non-historian colleagues in Econ. But we shouldn’t play that game. Not at all
What we are (perhaps) gaining in identification we are losing in our comparative advantage as economic historians. Of all the folks in an Econ department, we are the ones who are supposed to know some things. I see less and less evidence of that.
If we look more and more like every other applied economist then what, exactly, are we bringing to the table? Old data (is it any good)? Methods (that everyone else knows)?
So, the same problems that plagued the analysis of slavery show up in work that is totally unrelated to it. We can do better and we should, and even though I’m personally optimistic, I’m realistic enough to know that the headwinds are quite strong.
Very eager to hear what @MarthaOlney @delong @drlisadcook @SandyDarity @martha_j_bailey @snaidunl and lots of others think about this. /END
Please weigh in @WorldProfessor!
