Mason researcher uses machine learning to examine use of Bible quotations in American newspapers

Lincoln Mullen

Lincoln Mullen is taking a divine approach to computational research with a grant from the Library of Congress. The award will advance the work of America’s Public Bible, which explores how American newspapers from the 19th and early 20th centuries used the Bible.

The website identifies biblical quotations and allusions in millions of newspaper papers from the Library of Congress’s Chronicling America digitized collection and other newspaper archives.

The initial version of the site won first place in the 2016 National Endowment for the Humanities’ (NEH) Chronicling America Data Challenge. A much-expanded version of the site is in progress, and it will be published as a digital monograph by Stanford University Press.

“America’s Public Bible has always been an experimental project,” said Lincoln A. Mullen, an associate professor in the Department of History and Art History at George Mason University, who is one of the faculty directors of the Roy Rosenzweig Center for History and New Media (RRCHNM). “For the NEH Data Challenge, it was a chance to see if a machine learning prediction model could be useful for religious history.”

Until now, the project, which stems from the RRCHNM, has searched for quotations from the King James Version (or Authorized Version) of the English Bible, by far the most commonly used Bible among American Protestants during the 19th century.

The version in preparation for Stanford University Press will include a larger newspaper dataset and factor in more versions of the Bible. The peer-reviewed digital monograph will interpret the role of the Bible in American public life through prose and visualizations.

“With the support of Stanford University Press, the forthcoming interactive scholarly work is an experiment in the form of scholarship,” said Mullen. “And now with the Library of Congress Labs, I have a chance to experiment with cloud computing with the library’s amazing digital collections.”  

America’s Public Bible is one of three inaugural “Computing Cultural Heritage in the Cloud” projects at the Library of Congress, supported by a $1 million grant from the Andrew W. Mellon Foundation. The projects offer a chance to leverage the library’s collection to support digital humanities research.

“We know there’s so much more treasure to be unlocked here at the Library and in institutions across the country and around the world, and technology can help us understand these collections even more,” said Kate Zwaard, director of digital strategy at the Library of Congress. 

The superiority and engagement of America’s Public Bible is enhanced by its ability to explore the trend of how frequently a verse was used, Mullen said.

America’s Public Bible offers two ways of exploring how the Bible was used in 19th- and early 20th-century newspapers.

One option is an interactive chart, which allows a user to put in the reference to any of nearly 1,700 most quoted Bible verses and see the changing patterns in their usage over time.

Findings include that “Righteousness exalteth a nation: but sin is a reproach to any people” (Proverbs 14:34) peaked in 1865, while “Greater love hath no man than this, that a man lay down his life for his friends” (John 15:13) grew in popularity during World War I. “Suffer little children to come unto me” (and its variations) was the most popular verse in this collection of newspapers.

Visitors to the site are also encouraged to view the trends for collections of verses arranged in topics chosen by Mullen.

The second way to use site is to get links back to the images of the roughly 11 million newspaper pages at Chronicling America, from 1836 to 1922, with the quotations highlighted, showing how often and when a verse was used, as well as by whom, in what contexts, and for what purposes.

“Professor Mullen’s project, America’s Public Bible, demonstrates the power of computational methods for historical analysis and is one of the most exciting projects to come out of RRCHNM in several years,” said Mills Kelly, RRCHNM’s executive director and a professor of history at Mason. “In addition to being an important tool for scholars, this project creates an important resource for members of the public who are interested in the role that religion has played in American history.”