From the Merc:
The Google Books "Ngram Viewer" and the downloadable raw data set achieve what mere mortals can't: analysis of 500 billion words from 5 million books published over the past four centuries, part of Google's ambitious book-scanning project.
The search tool makes it possible for anyone to chart usage of a word or phrase over time, using computer technologies that scroll through a sequence of letters. If strung together, the letters would reach to the moon and back, 10 times over.
Then word frequencies, charted on a graph, can be compared and contrasted. And by clicking on the decades at the bottom of the graph, it's possible to view actual titles.
The tool's main attribute is that you can track the usage of words and phrases over time -- 1800 to 2008, to be exact. So, as cited in the Merc, you can see that the word "God" peaked in the 1820s, hit a nadir in the 1970s, and has since begun a comeback.
A joint Harvard-Google research team's paper on this data, called "Quantitative Analysis of Culture Using Millions of Digitized Books," has been published in Science. The Harvard Cultural Observatory has put up a site called Culturomics, which includes a user-guide to the N-gram Viewer.
Keeping in mind that the search is case sensitive, see how it works by clicking on some of the random names, words, and phrases I spent way too long typing into the thing, below.
Then get back to work.
Google Ngram Graphs