Google Releases Books Ngram Viewer, Massive Searchable Database of Words

(Dec. 17) -- Never one to be caught resting on its laurels, Google has released a massive, searchable database that will give linguists and historians a new tool for quantitatively understanding how language and culture have changed over time. It also makes for a good time-waster on languishing Fridays before the holidays.

The Books Ngram Viewer, which Google created with the Encyclopedia Britannica and scientists from Harvard and the Massachusetts Institute of Technology, takes 500 billion words from 5.2 million digitized books and allows you to track their usage over time. The result is a database that shows when certain phrases, people, ideas and trends faded in and out of fashion.

For example, if you wanted to see when certain words became outmoded in the English language, you might compare them to their more popular versions today. Below, a comparison of "trousers" and "pants" shows pants falling out of favor around 1830 as trousers took off. But, after trousers reached their apex in the 1940s, pants rebounded, and eventually reclaimed their rightful place at the top of the English lexicon around 1980:

Or, for a more controversial comparison, try looking at "dog" and "cat" and you will see that the former has appeared far more frequently since 1800 than the latter in books published in English in the United States:

The Harvard team has named the new analysis "culturomics," as a means of expressing the idea that culture can be studied quantitatively. They have come to some interesting conclusions already, that people become famous at a much earlier age today than before, for instance, but that they also fall from notoriety much faster too. Their study was published in the journal Science today.

"They've come up with something that is going to make an enormous difference in our understanding of history and literature," Robert Darnton, cultural historian and director of the Harvard University Library, told The Wall Street Journal.

Erez Lieberman Aiden, one of the lead researchers at Harvard, told Scientific American that they do not see the Ngram Viewer as an answer factory. Instead, many of the findings in the database will elicit a multitude of questions.

The current Ngram database took about four years to assemble, and the team intends to add more books, magazines, newspapers and blogs -- and also non-text-based work such as art -- in the future.