Google Ngram, the tool that lets you chart the relative frequency of words as they appear in English and foreign literature over time, has now added support for wildcard searches. This means you can plug in a search for "bar *" and conceivably get results not just for "bar stool" but also phrases like "bar exam" and "bar none."
In a moment of meta-ness this weekend, I searched for "search *_NOUN" — which will return only phrases with the word "search" followed by a noun.
I expected the phrase "search engine" to appear somewhere in the results, but not nearly as prominently as it actually did. The phrase "search engine" — and its plural derivative, "search engines" — has experienced tremendous growth in the book world, one that more or less mirrors the technology's own development over the last two decades. You had the slow, initial rise in the mid-1990s attributable to first-gen companies like AltaVista, Excite and Lycos. Then it started to eclipse more traditional ideas like "search party."
By the time "search engine" surpassed "search warrant" in 1996, a couple of bright-eyed Stanford students had come up with something called BackRub, which crawled the Web to find out who had linked to a given page. BackRub became the basis for a little start-up known as Google.
Here's what a wildcard search for "google *" looks like on Ngram today:
Whether Google's rise has something to do with the more recent decline in the term "search engine" is anyone's guess. It's been seven years, after all, since Google was added as a verb to Merriam-Webster and the Oxford English Dictionary — and arguably even longer since people began using the company's name as a stand-in for "search" in common parlance.
The dip might also be partly explained by the industry's move away from search-engine optimization. As the Web got more complex, Google adjusted its algorithms so publishers couldn't simply load a page with keywords to boost its profile. A decline in the pace of new books mentioning SEO could have a small but measurable impact. In fact — bearing in mind that this is an unscientific study — that may be approximately what we see here; while SEO continues to appear in books, the rate of growth seems to have slowed beginning in 2005.
Another possible explanation could be our collective shift toward social media, which has become an increasing focus in literature since the 2000s.
Although "social media" appears in just a tiny fraction of English books — it doesn't even show up when you search Ngram for "social *_NOUN" — its rate of growth in literature has been even faster than it was for "search engines." That said, the track record for social media is a lot shorter; it'd be unwise to presume anything about its trajectory from just a decade of prominence.
It's also worth noting that 2008 is the most recent year for which Ngram provides data; we don't really know what's happened to "search engines," "social media," "google" or other terms in the years since. Still, what we are able to see paints a fascinating picture of where we've already been.