Because I study information networks, there are all kinds of opportunities to manipulate, visualize, and otherwise goof around with data. I have mostly worked with topic modeling, text mining, mapping, and social network visualization.
On this page, I’ll share a few of the digital, digital-adjacent, and quantitative history projects I’ve worked on.
Mapping eighteenth-century newspapers.
The Age of Revolutions inspired a widespread hunger for news. Demand for news created a growing market for newspapers in North America, Europe, the Caribbean, India, and elsewhere. I have sought to visualize the resulting expansion of newspapers through maps like the one below, which shows when newspapers were founded in the eighteenth-century Caribbean.
Visualizing Information Networks
Who cited whom? Visualizing newspaper citations can tell us a great deal about how information flowed through the revolutionary Atlantic. As one example, the two images below each show a network of citations from Boston, Philadelphia, and New York City (the three darker nodes). The above image shows this network from 1755 through 1779. The second image shows what this network looked like from 1780 through 1804. Both show the information networks emanating from three cities over the course of a 25-year period. But the second is, obviously, much more complex. My dissertation’s third chapter explains why.
Eighteenth-cenutry Newspaper mottoes
Check out this page for information on newspaper mottoes in eighteenth-century America. It includes links to pages containing a dataset of these newspaper mottoes, by decade.
Modeling Newspaper Carrier Addresses
Topic modeling doesn’t lend itself as well to visuals as mapping and network analysis do, so I offer this fella plucked from the masthead of a 1742 issue of the Boston Gazette. It’s a representation of a newspaper carrier. Carriers were men and boys (and in one documented case a young woman) who delivered newspapers in a city. They were usually poorly-paid, and often apprentices. In order to give them an opportunity to earn some money, printers gave them a poem to read (or present) to their customers on or around Christmas or New Year’s in order to solicit tips.
These poems are amazing documents. One of the interesting things about them are the sections that list the most important news of the previous year, in verse. I grabbed these, in the hundreds, and used an open-source package called MALLET to topic model them. This provides an extraordinary window onto what news was considered to be the most noteworthy and important each year.