Cliodynamics: A Future for History?

June 15, 2013 by Forrest Barnum 10 Comments

What is history? And what, if any, are its practical uses? These are the questions I’ve been pondering since being introduced to Cliodynamics – which claims to make history into “an analytical, predictive science.” To that end, I wish to address two questions: is it possible to make history into “an analytical, predictive science?” And is it desirable, for the purposes of attaining greater knowledge or understanding, to do this?

Firstly, I should say that I dispute some of the premises of Peter Turchin’s (the intellectual founding father of this movement) arguments as presented. I once engaged one of my colleagues about what the essential nature of history is, and she expressed it in a single word I find quite apt: counting. If you boil the historical method down to its most essential features, historians engage in the act of counting things though time, who lived where, where people worked, what they earned, even the kinds of beliefs they expressed or the mores they practices come down to numbers changing over time. So I find his claim to have ‘mathmatized history’ to be risible. To be fair, I suspect he means to say that he means to use mathematics in a more rigorous social science kind of way, he mentions things like data sets etc. that support this interpretation.

To me, this suggests that he wishes to practice a form of history which positions itself squarely as a social science. Social sciences (economics, most political science, parts of various other fields) tend to be more explicitly math-based and attempt to carry out the kinds of empirical analysis with practical implications he extols. The multiplicity of methods and practices encompassed within ‘academic history’ one of the strengths of the field in my opinion. Though there have been times when this was contested, there was an honest to god period called ‘the history wars’ in the 1980s when academia was riven by largely silly debates about proper historical methodology, this diversity seems to most modern practitioners to be more of a feature than a bug.

I personally would never therefore say that Cliodynamics is somehow a ‘wrong’ or ‘bad’ way to do history, since the field can encompass a broad variety of methods and goals, and of you buy my argument that history is at a fairly fundamental level underpinned by numbers, it certainly can’t be seen as somehow foreign to the field.

So the answer to the first question, is a decisive yes. The answer to the second is more ambiguous: I would say that making history more of a ‘science’ does not necessarily make it more sophisticated or right. Karl Popper provided an analysis which might be of some use here. In his thinking, there were two essential schools of looking at the world which could be typified by the ‘cloud-clock analogy.’ Essentially sciences examine phenomena which are akin to clocks, they are mechanistic and predictable. If you drop an apple from a tower, it will behave in a matter which previous apple drops and future apple drops are consistent with (at least if we stick to classical physics). Social sciences and humanities examine events and people which are cloud like. Clouds are broadly speaking identical, but specifically entirely distinct from one another. Think of human societies: it would be pretty easy to come up with a number of attributes which features in all human societies (contain people, feature use of languages, rituals etc) but when you drill down the distinctions would become clear. Basically all societies and peoples and persons are akin to clouds insofar as they are the same thing but different. This presents a problem for ‘social sciences’ since these distinctions mean its virtually impossible to develop some set of axioms for human behavior that remain true in a ‘scientifically’ explicable way.

For example, let’s take the Trolley Problem. Here a person is presented with problem for which there is no right answer or where the answer is exceptionally subjective depending upon the context in which the person is situated. Different societies might claim that one answer or another is superior, but there is no means by which to determine ‘an answer’ that is always true. Human behavior is simply not subject to ‘laws’ which hold true across cases.

Now Turchin (2008) says that societies have various similarities or relations that can be explored and explicated for analytic leverage. As I am familiar with it, this sounds like the version of political science I practice, comparative politics. In this sub-field (or as wags would have it, methodological ghetto) various aspects of societies are compared using tools such as John Stewart Mill’s method of similarity and difference. In a highly regarded recent example, Theda Skocpol (1979) used a version of Mill’s method to compare three revolutions over time, demonstrating how in spite of their wide divergence of time and place, they were in essence the same phenomenon. This is obviously somewhat different from hard science, given that her theory though highly regarded has been contested, debated, and elaborated on since it was published some 30 years ago.

History has always been in a somewhat ambiguous position when it comes to being used in a ‘practical’ sense. History has always been classed as somewhere between a humanity (i.e. more of an ‘art’ than an empirical field) and a social science, indeed many universities stick it in one section of the facility or another with little justification. Modern history as it is practiced today is generally most interested in discovering how people and societies lived in the past, viewing this information to be on intrinsic interest to us as humans. So painstaking archival research to construct the story of a particular Italian cheesemonger might not in any major sense be ‘practical’ but it is rated worthwhile as a means of understanding our past. History was often in previous generations seen as something of a moral art, since a person who knew lots of history would be wise enough to make inferences and avoid the mistakes of the past when similar situations cropped up in the present.

However, it seems a bit dubious to me to suppose one could literally predict the future with history. While I suspect it’s a tempting idea, it might better be said that history is best used in a more impressionistic and individual way as a kind of collective memory. Much how we as children learn to avoid burning our hands on hot objects, so to can properly read history teach the avoidance of harmful actions. Telling the future, as nice as that would be, remains outside the grasp of historians, who are after all, constantly grappling with our very imperfect understanding of the past.