It always ends in statistics

farewell-to-arms

Thanks to a coffee break with Benjamin Mitra-Kahn, I had already heard of this 2007 famous book on the “long” history (millenniums) of economic growth. The book, “A Farewell to Alms” by Gregory Clark, is now discussed in a symposium published in the August issue of the European Review of Economic History. I thought it was a good occasion to know more about this book – the issue is available for free on CUP until the end of the month. I started reading the rejoinder by Clark (or, how to learn about something by the very end of its tail).

Clark answers to the criticisms of three economic historians, McCloskey being one of them. In essence, Clark’s critics say that he is either wrong (like: he contradicts some of Angus Maddison’s figures) , or that what he says is just rewrapped well-known facts. McCloskey adds something else: that Clark commits social Darwinism.

Economic history has a long relationship with biology. Robert Fogel and Stanley Egerman, in their Time on the Cross (1974), already used measurements of protein intake to estimate the well-being of slaves, to reach the conclusion that “biologically speaking”, being a slave was not that harsh after all. The book caused a lot of stir as one can imagine. Paul David and Peter Temin wrote a lengthy review entitled “Slavery: The Progressive Institution?”. Basically, the reviewers had taken the time and effort to study in detail the 2nd volume of Fogel and Egerman’s book, where all their statistical work and various assumptions were compiled. After a very dense study (the review takes 45 pages of the JEH), David and Telmin concluded that fundamental assumptions were biased, and the results flawed.

The “technical statistical discussion” is the direction the debate on Clark’s book is taking: to answer McCloskey’s criticisms, Clark proposes a measurement of the degree of heredity of wealth. His few equations are simple, but they are the promise of surely loooong debates on the correctness of such or such method: Clark also mentions twin studies, another sure sign that economic history is now far behind, with statisticians in control. And if I judge from past intricate heated debates on wealth and inheritance in statistical economics, the result is that after a few ping-pong exchange, the wider audience gets bored and moves on.

6 thoughts on “It always ends in statistics

  1. I am not sure about the rhetorical connection you made between social darwinism and biology in your post. The Fogel/Egerman measurings were about finding a ‘as solid as possible’ yardstick to provide a measurement of the standard of living of slave against that of free populations. The problem they face was that they could not rely on money measurement since an important of the slave economy was based on non-market relationships. Thus, there was no moral issue there, it was basically a statistical problem and the recourse of material measurement based on biological data was their way out of this problem. I see no darwinism or social darwinism issue here.
    On the other hand, Clark’s notion of heredity of wealth seems something quite different (The book is on my reading list but I did not find the time to read it yet, i have to check the debate you mention, though): no biology (at least this is what I figure from your post) but the idea that wealth can be analysed through a biological metaphor.
    However, thanks for the references.
    On the intellectual history of the Fogel/Egerman debate, see: The revolution that bit its own tail : how economic history changed our ideas on economic growth by Jan W. Drukker, Amsterdam : Aksant, 2006.
    For a more sociology of science perspective: see Christel de Rouvray’s Phd Thesis.
    Both are interesting in their own way.

  2. I think the connection Clement is drawing is one between biology and economic history, while the social darwinism is a point made by McCloskey in reference to Clarks work. The commonality – to my mind, and I stand ready to be corrected – is that Clark has used a biological theory to further his economic theory, while Fogel/Egerman used biological data to further their economic theory.

    The point that McCloskey is making (and it’s been a while since I read the book, but the comments are lying in front of me) is that Clark is arguing that the world was trapped in a Malthusian Trap up until the 18th century and as a result of the spreading bourgeoisie classes – through higher rates of reproduction than their poorer countrymen – England’s growth exploded, as opposed to other countries where this was not the case. McCloskey says that a lot of his arguments are not new, and the reproduction argument in particular is reminiscent of social Darwinism explanations from a hundred years ago. (Then there is some data quibbling, and a lot of un-cited work which Clark uses as well).

    What I did like about McCloskey’s answer is the point she makes about the transfer of values, and that it may be much more important for economic betterment that people are situated within a free market based system rather than who their parents were, She illustrates this by pointing to immigrant communities in the US and Europe who came from non-market countries and have prospered. Also she points to the rise of India and China in too short a time to have come from a genetic change in the ‘virtues’ of people, but rather she says, is based on the market creation in those countries. The point that Shakespeare’s father was a glover and Kant the sone of a saddler is raised to point out the problem with a biological link between father and son as being the key to creating virtues and progress. That is the link which Clark argues exists, which McCloskey takes a stance against.

    The title of your post reminded me of a line from Terry Pratchet’s ‘Mort’ :
    “They Listened.
    They Argued.
    They Resorted to Mathematics.”

    And now for a coffee.

  3. [Answer to Loic, before seeing Ben’s comment]

    That’s interesting, and maybe should I have been more precise.

    What I meant is that in Clark and Fogel, there is a will to make historical pronouncements more quantitative by referring to biology. In Fogel, this is the measurement of diets as a proxy for physiological well-being, while in Clark, this is evolutionary theory – but also body heights, diets, and racial differences (read his rejoinder). This is what I meant when I said that economic history has a long relationship with biology.

    No moral issue? Choosing a yardstick is not a neutral operation, all the more when it concerns the body. A few years ago, I had heard a beautiful talk on the rationalization of hunger through the definition of living standards in late XIX France. The talk showed clearly how simple an operation as defining a daily ration could be, by any standard, morally dubious (This is now a PhD dissertation: Dana Simmons, Minimal Frenchman: Science and Standards of Living, 1840-1960, Univ Chicago 2004). And when talking about the evaluation of the well-being of **slaves**, this is all the more morally controversial to me.

    The “social Darwinism” issue is difficult: first, I try myself not to use it to label a scientific work, as there is no good and stable definition for it. So this is why I don’t use it in my post. In my opinion McCloskey uses it because, as Clark replies, this is an old rhetorical trick. But social Darwinism or not, what is sure is that Clark does use Darwinism operationally, not metaphorically. In fact, a good result of McCloskey’s severe review of Clark is that in his rejoinder, Clark is forced to (or keen on) making things clear.

    Thx for the refs!! I wanted to read Christel’s thesis for a long time, would anybody know if it is available electronically?

  4. Moral issue? It was not so for Egerman and Fogel and loads of demographers and quantitative historians who used this kind of data in the 1970s and 1980s. This does not mean that there is no moral issue at stake – deciding at what level of daily calories diet you are starving can certainly be viewed as a moral question, but again I do think that whan Fogel and others were stating that they were not concerned with these undertones, they were genuines.
    To put it in a more general perspective, most of the quantitative measuring of human question can be viewed as moral issues – for example setting the ceiling of poverty of one-half of the median standard of living of a certain population -, but it should not mean that you cannot do or discuss these measurements solely in moral terms or as indicating moral statements. This does not mean that there are not moral issue at stake, but one ‘is suspending them’ at the moment one decides to approach a subject scientifically. Otherwise, I do not see how you can approach social questions without being emprisonned in the moral issue. This is one of the variant of “Memory” VS “History” debate still going on in France (there had been a few recent articles in Le Monde on this subject). If Memory or moral issue is all that counts then no history (as a rational and independant from moral valuation investigation of the past) for its own stake can exist, I can not accept it.

  5. I am curious to hear what Beatrice’s comment would be on your argument: “one is ‘suspending the moral issues’ at the moment one decides to approach a subject scientifically” . Let’s wait after the 20 Nov.

Leave a comment