Railroading history into Big Data
Last year I attended a lecture by Bruno Latour, here in Utrecht. At the time, J. started reading The History Manifesto, so I joined that too. Those two meshed into the usual incoherent mess of thoughts craving for structure. I was also invited to write a commentary for a pretty cool academic journal on the Manifesto, and much of the thoughts here arose or were discussed with my co-writers for that.
Right. Latour.
Latour talked about the anthropocene. I knew nothing about the anthropocene before the lecture. If you're in the same boat, here's a one sentence definition from Wikipedia: "The Anthropocene is a proposed geologic chronological term for an epoch that begins when human activities have had a significant global impact on the Earth's ecosystems." The take-home message that I got from Latour was about sticky concepts - such as anthropocene - that mobilize small armies of actors, other concepts, organizations, resources, words, political institutions and decisions, etc. etc. The details of the whole debate were beyond my comprehension, but the nature of anthropocene as a concept was something I could think about (and along) Latour.
The History Manifesto attempts to do such a thing, too, on a different scale. It tried to mobilize historians into action. Unlike the anthropocene, which is a playground (or maybe a battleground?) of agencies; the Manifesto is is a conscious effort at agency. The name, the tone, the introduction ("A spectre is haunting our time: the spectre of the short term") and the conclusion ("Historians of the world, unite! There is a world to win – before it’s too late") almost comically frames the text as the probably most famous call for action in our history - the one by Marx. Now, Armitage and Guldi's aimed audience isn't the proletariat, but the historiat. The three main lines of argumentation that I could discern, and consequently the three conclusions, are:
We as a society, and history as an academic discipline functioning within that society, are maligned by the lack of long-term perspective. We navel gaze in the short-term.
Historians have lost face, reputation, and position as the interpreters of the future. Social scientists, economists chief among them, have taken primacy and the most important societal role.
Data drives the world. Big data. Historians need to reclaim their birthright (academic-right) as interpreters of that data. They are uniquely skilled to do so, to augur the future from trend charts and network analyses.
Their manifesto, in its zest, reminded me of Latour's Reassembling the Social. I read that as an academic manifesto too, much less comically and explicitly framed (which is quite a statement for somebody of Latour's flamboyant writing style, that it's less ostentatious than the Manifesto). Latour also had quite a radical goal with his book - to uproot and destabilize what social means as an object of research for the social scientists - exposing it as a universal black box that can serve both as an explanation and explanandum. Latour wants the social scientists to be less hasty with tagging something as social without earning that label with meticulous ethnographic research. In a sense, the social hides the networks of actors and the meanings they assign to what surrounds them. It's a lazy way out, taking 'social' as something out there in the world, without investing effort to describe it precisely and by doing so derail it. The social needs to be destabilized and examined into the tiniest detail until we are allowed to wield it as any kind of an explanation. Even then we should take it with a grain of salt.
With the risk of oversimplification and misrepresentation of Latour's main argument in the book, I believed him. I became highly critical of social explanations that blackbox instead of explaining. I think that Guldi and Armitage tried to do something similar with what is history for historians.
I believed Guldi and Armitage much less than I believed Latour.
I am comparing the two to show that that has nothing to do with politically charged language, but with the consistency and nuance in argumentation. Both Latour and the authors of the manifesto use charged language - Latour because that's how he writes, and Guldi and Armitage (I am guessing) because they were explicitly interested in setting a particular tone for this particular text. I agree in that with Peter Mandler and Deborah Cohen's review of the manifesto, that the layers of the story were lost and disfigured to fit the argument the authors were trying to make. The context out of which it was supposed to arise - just to name two components of the said context: the short-termedness of historians and historical monographs; and the quintessential conflict between economists and historians as two distinct dichotomized groups - were just not persuasive. They were paper-thin. According to Mandler and Cohen, they were also factually wrong. This says nothing about the charge of the text - I have no problem with the ranty tone, on the contrary, I quite enjoy it - but the argument belies under railroaded assumptions that are there to fit the argument, and those assumptions go hand-in-hand with the fact that the text was written as a manifesto.
Then there is the point of big data.
The historians, described by Guldi and Armitage, are the best interprets of data out there. They almost have a divine right to tell us what charts and Google ngrams mean. In comparison to the other disciplinary priests - the social scientists and economists chief among them - historians know how to do this precisely because they work with contextualizing really difficult things. The stuff of their trade is making something insanely complex and possibly very remote from our daily experience, and tell a plausible causal chain of events out of it. This account seems quite similar to what big data does, or digital humanities.
Well, yes. I completely agree that historians need to embrace the digital. Like the archive, the digital is a technology and a tool that allows historians to work. Now, that being said, embracing the digital involves all kinds of skills and sensibilities historians do not posses by the virtue of their training. I think it's a big oversimplification just to say historians need to do this, and they will be the best at it out of all the possible experts in the big market of knowledge producers.
Embracing the digital will entail a debate among experts of how to do it, and a way of making it a building block that is part and parcel to how historians usually conduct research - of all historians, if they find the need to use digital tools. But thinking of computational tools as an easy way out, a revolutionary new thing that will destructively take the place of the old, as the Scientific Method that will streamline history and writing history, I think that is completely misguided and bound to fail. And if that happens, historians stand to lose much more than their hopes for the future.
Cover image credit: https://upload.wikimedia.org/wikipedia/commons/f/fa/07_cars_traffic_experimental_digital_photography_by_Rick_Doble.jpg