By R. Alexander Bentley and Alberto Acerbi, University of Bristol, with Alan Hall, Socionomics Institute | Excerpted from the January 2015 Socionomist
Here we excerpt an article by University of Bristol researchers R. Alexander Bentley and Alberto Acerbi, which reviewed their research on the sentiments reflected in authors’ word choices in 20th century literature. They found that changes in emotions expressed in US, UK, and German literature reflect changes in social mood.
During the 20th century, changes in the sentiments of authors’ word choices provide “evidence for distinct historical periods of positive and negative moods.” That was the key finding of the 2013 paper, “The Expression of Emotions in 20th Century Books,” that the two of us—Bentley and Acerbi—co-authored with Vasileios Lampos and Philip Garnett.
We used Google Books’ data to assess the language. Google’s n-gram tool draws on a database of millions of books in multiple languages to show the annual popularity of any published word or phrase over the past several centuries. We extracted raw data from Google’s corpus of English literature from 1900-2000. We limited our study to the 20th century because the composition of Google’s corpus changed after 2004. To determine whether and how the expression of emotions had changed over this period, we calculated the frequencies of six lists of terms that characterize anger, disgust, fear, joy, sadness and surprise. A previous study of UK Twitter content used these same emotion-word lists and categories and showed that “changes in these emotion word frequencies identified real-world events such as the unexpected deaths of popular personas, public unrest, or natural disasters.”
We were able to distinguish “happy” and “sad” periods in the data, and we developed what you might call a “Literary Happiness Index.” We found that,[Emotions] tracked broad historical trends, including a “sad” peak corresponding to the Second World War, and two “happy” peaks, one in the 1920s and the other in the 1960s. In more recent years we can see a “sad” period starting from the 1970s, with an increase in “happiness” in the last years of the data set.
We showed the data to Alan Hall at the Socionomics Institute, who graphed Figure 1. The top line plots our Literary Happiness Index, while the bottom line plots trends in social mood as indicated by the Dow/PPI. As you can see, the graph of literary happiness roughly corresponds with major trends and turning points in social mood as indicated by the stock market. This is especially true from 1929 through the 2000 peak of the dotcom bubble.…
In the remainder of this five-page article, Bentley and Acerbi compare their ‘Literary Misery Index’ – an inversion of their Literary Happiness Index – to economic misery and negative social mood in the US, Germany and Russia.
Want more content like this?
The Socionomist is the only monthly publication that offers you practical insights on the relationship between social mood, financial markets and cultural trends. Each issue warns you about big societal changes before they can harm you and reveals breakthrough opportunities emerging from trends in society. Become a socionomics member today and get instant access to The Socionomist.
(Socionomics Members: Log in for the full article and your complete, exclusive archive.)