Pages Navigation Menu

History: A science of thought experiments

“How do historians know when they’ve established, once and for all, the causes of any past event? The answer is, of course, that they don’t.”  John Lewis Gaddis


I regularly read Old West discussion forums. The most active such website may be the “Tombstone History Discussion Forum,” filled with lively arguments over the Gunfight at the O.K. Corral, the iconic Wyatt Earp, Tombstone (“the town too tough to die”), and all things related. The opinions of forum posters do appear to shift on occasion, but mostly participants seem to talk past one another, their historical and evidentiary arguments on any side of the controversies du jour making little apparent headway in the face of strong, preconceived ideas.

I thought a lot about the concrete stances taken on a fluid river of historical knowledge as I recently read an engaging book on what it is historians actually do and whether or not writing history is more of a scientific or artistic endeavor.

In The Landscape of History: How Historians Map the Past (Oxford Univ. Press, 2002), John Lewis Gaddis presents a multitude of useful metaphors to explain what historians do and the age-old question of whether history has more in common with science or art. As the title and subtitle suggest, history is a lot like cartography. Historians, like map makers, “represent realities they can’t replicate and wouldn’t want to: a truly accurate map of Oxford would be an exact clone of Oxford…. Maps vary scale and content according to need.” Historians, too, must choose what to include, and at what levels—varying the scales of time and space according to the historian’s and the audience’s needs. Any account of Pearl Harbor (or the O.K. Corral, for that matter), probably does not need to inform the reader what Yamamoto (or Wyatt Earp, etc.) ate for breakfast a week earlier, or even that morning.

While sometimes a mountainous stockpile of evidence exists, more often than not there are gaps, big gaps. Some historical information does not survive or was never the subject of recordation. The historian fills in these gaps with “imagination, even dramatization.” Ultimately, Gaddis writes, the historian’s product “must go before an audience.” Readers may approve or disapprove, according to their preconceptions, or they may “revise their own views so that a new basis for critical judgment emerges, perhaps even a new view of reality itself.”

What is the proper methodology to achieve the desired product, to produce a new view, a better representation of reality? Does the historian rely on scientific methods? Is history a science? No and yes, says Gaddis. It is not like mathematics, physics, or chemistry. The historian cannot replicate the same results time after time, or even once, in the lab. But history is a lot like astronomy, geology, paleontology, and evolutionary biology, scientific disciplines in which the subject studied often cannot even fit into a lab, and “the time required to see results can exceed the life spans of those who seek them.” Results have occurred in the past and, though subject to some prediction, are not reproducible. The only way these scientists “can rerun history is to imagine it… within the limits of logic.” Like historians, they do so through thought experiments. Geologists, biologists, and astrophysicists, like historians, rely on surviving artifacts. Like these scientists, historians deduce the processes that produced surviving archives, artifacts, and memories. Historians allow for the fact that “most sources from the past don’t survive, and most daily events don’t even generate a survivable record in the first place.”

Historians combine logic and imagination to create a representation of the past. So do artists. How is the historian different from the historical novelist or painter of historical subjects? “Artists, can, if they wish, conjure up their subjects out of thin air. Historians,” Gaddis explains, “can’t do this: their subjects must really have existed. Artists can coexist in time with their subjects, altering them as they please. Historians can never do this: they can alter their representations of a subject, but not the subject itself…. Imagination in history then, as in science must be tethered to and disciplined by sources; that’s what distinguishes it from the arts and all other methods of representing reality (italics added).”

Historians, Gaddis writes, look for “the point at which [historical] processes took a distinctive, or abnormal, or unforeseen course.” Or, as Aristotle put it, “when things become contrary to expectation.” But how, Gaddis asks, do we know what those expectations were prior to the event?  Counterfactuals (what has not happened or is not the case) are one way. What would have happened if….? The historian does have to follow certain rules in applying counterfactuals. “You can’t throw multiple counterfactuals into the pot, because this makes it impossible to pinpoint the effect of any one of them.” To stir them in all at once forces speculation on the results of a “historiographical witches’ brew where anything goes and no particular outcome is any more probable than any other.” You can’t experiment with single variables that weren’t within the range of technology or the culture of the times.” (Recall the first season Saturday Night Live routine postulating Napoleon’s air force at Waterloo.)

Used correctly, counterfactuals are useful in establishing causation. But “how do historians know when they’ve established, once and for all, the causes of any past event.”

“The answer is, of course, that they don’t. Because not all sources survive, because not everything gets recorded in the sources in the first place, because the memories of participants can be unreliable, and because even if they were reliable no participant could have witnessed all of an event from all possible angles, we can never expect to get the full story of what actually happened.” We can get closer to the truth, but the historian’s basis for determining causes is “always provisional…. Historians do the best they can, therefore, but our findings are subject to revision, just as they would be in any other field of human inquiry.” Historians are left with a representation they hope closely fits the reality they seek to explain. This representation is usually in the form of a narrative which simulates in thought what happened in the past.

Gaddis has much more to say about history and biography in this breezily written and easily understood book. Recommended for historians and readers of history alike.