It’s the culturomics, stupid
The Internet Archive is pretty awesome. But the Google Ngrams tool, which allows anyone to search millions of books from 1800 to 2008, is pretty neat too. This tool is part of a new field called “culturomics” which is the study of historical trends by quantitatively analyzing words from a very large digital data set, like Google Books. (Ironically, if you do an Ngram search for culturomics you get an error message since it hasn’t entered the corpus of digitized books yet.) Culturomics has led to some cool discoveries, especially related to the development of the modern English lexicon. Here’s how it works, in graphic form:
Personally, I found most fascinating the ability of culturomics to predict, retroactively at least, the location of Osama Bin Laden to within 200 kilometers of his eventual hideout in Abbottabad. Using mentions of Bin Laden in Summary of World Broadcasts, which tracks global news outlets, researchers realized that the two cities “identified” closest with Bin Laden were Islamabad and Peshawar. Lo and behold, Abbottabad is within 200 kilometers of both cities.
But as Patricia Cohen wrote in her New York Times article when the Ngram too was first released, this groundbreaking historical research tool was not without its initial detractors. There are two main complaints from traditional historians: 1) only academics from the sciences, and not the humanities, were part of the initial research and study; 2) this tool doesn’t allow for close readings or interpretation (as Elena Razlogova describes it “today builders of digital tools and archives herald a non-interpretive, ‘post-theoretical age'”).
I think those complaints are missing the point. This tool, like so many other digital resources, is not about replacing traditional historical approaches, but is about giving historians and other “humanists” supplementary tools that are only now possible thanks to the power of the internet. As one of the creators of the Ngram project said “I don’t want humanists to accept any specific claims — we’re just throwing a lot of interesting pieces on the table.” So unlike quantitative history, which has more or less faded out as a historical method, I believe this tool is here to stay.
(Up next: creating an actual chart using Ngram!)