The importance of scale in map-based history

I am a big fan of Donald Worster’s Dust Bowl: The Southern Plains in the 1930s, so I felt more than a little trepidation when I was ready to begin reading Geoff Cunfer’s essay in Placing History. I knew from an earlier discussion in class that Cunfer had disproven the long-established theory that reckless overplowing of the fragile Plains environment had directly caused the Dust Bowl of the 1930s. 

I needn’t have worried. Cunfer does, in fact, make a compelling case that the conventional wisdom is wrong, but he does so with plenty of respect for Worster. He acknowledges that Worster worked at a time when the sort of systematic, sweeping analysis that Cunfer was able to perform in GIS was not feasible, meaning that the case study approach of Worster’s book was the most logical choice. Furthermore, Cunfer acknowledges that within the limited (but deeply nuanced and highly detailed) parameters that approach imposes, Worster’s interpretation is the most logical one to make.

However, by using GIS Cunfer was able to take a broader view of the entire High Plains region in question, and by doing so was able to cross reference statistical data on dust storms, drought* conditions, and percentage of land plowed versus left as grassland–all on a county-by-county basis. Doing so, he was able to show that while “over-plowing” could be a mitigating factor, the most direct correlation was between the prevalence of dust storms and below-average rainfall, not with the percentage of grassland being plowed up.

With this GIS data, Cunfer was then able to use standard historical primary sources–written reports from the mid- and late-nineteenth century–to argue that dust storms are a regularly reoccurring phenomena on the High Plains–it was just that previous dust bowls hadn’t been widely observed or reported, and the drought of the 1930s was extreme in severity.

 

*Or “drouth”.

Advertisements

6 thoughts on “The importance of scale in map-based history

  1. It really is amazing the things GIS can reveal. We’re priviledged to be in an era where we can examine history in ways previous historians could only have dreamed of.

  2. Great post, Kirk. I like how you bring in scholarship to provide some historiographical background to the Cunfer essay. I responded to that piece as well — it seems like the perfect argument for GIS. “Here’s a well known historical story that turns out to be completely wrong when we look at the data.” Most of the time, it doesn’t work that way! It seems like digital methods, perhaps like all new methodologies, have the double burden of proving the argument and the validity of the method. It’s a tricky balance to pull off. Getting back to the Cunfer piece, if I were a historian skeptical of digital methods and GIS being used for historical research, I might say, “Sure this GIS stuff works when you’re looking at rainfall, but it has no broader application to history.” How would you respond to skepticism like this?

    • That’s a good question, Nate. I would respond by asking “what do you mean by ‘broader application’?” I would then point out that in this case, Worster had the narrower (if deeper and richer) approach, while Cunfer had the “broader” approach. Which allowed him to look at the Dust Bowl of the 1930s in a–that word again–broader context. From a statistical point of view, just looking at hard-hit counties in the 1930s without looking at less-affected (or less-plowed) counties, OR at other times periods is essentially a sample size of ‘1’. Which is about as not-broad as you can get.

  3. Pingback: GIS and Framing the Question | Nathan Sleeter's Blog

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s