I've been taking advantage of the rabid interest in "Downton Abbey" lately to report on some verbal anachronisms that have cropped up in the show's second season (originally broadcast on ITV in the UK late last year and now wrapping on PBS in the US). Over the past few days I've written about it in columns for The Boston Globe and the Visual Thesaurus, and I was interviewed on the topic for NPR Morning Edition earlier today. I also put together a video compilation of questionable lines from the show, and it's been making the rounds in culture-y corners of the blogosphere:
There seems to be a real appetite for anachronism-spotting, given how widely the video has circulated since I posted it (along with the NPR website, it's been featured on Slate, The Awl, Lapham's Quarterly, WSJ Ideas Market, Flavorwire, Nerve, Metro, and BBC America). And in the UK, reports of too-modern slang in the second season of "Downton" made national news last year (in The Telegraph, Daily Mail, and elsewhere). But is this all much ado about nothing — a tempest in a vintage teapot? Are there any broader linguistic questions at stake here beyond the nitpicking?
I'm happy to report that Benjamin Schmidt — history grad student at Princeton, fellow at Harvard's Cultural Observatory, and co-creator of the Bookworm interface for Google Ngrams — has taken the ball and run with it. In a post on his Sapping Attention blog, he guides us through an experiment in historical collocations:
I found some copies of the Downton Abbey scripts online, and fed every single two-word phrase through the Google Ngram database to see how characteristic of the English Language, c. 1917, Downton Abbey really is.
His results chime with my more anecdotal inquiry, but he also raises some deeper questions about the (re)invention of tradition. It's exciting to see what can be done along these lines, in terms of judging linguistic verisimilitude of period-specific drama. This is something I've been looking forward to ever since I heard J.-B. Michel and Erez Liberman Aiden give a preview of their blockbuster "culturomics" paper in late 2010. I had mentioned to them that the writers and researchers for the '60's-era TV show "Mad Men" used Google Books, among other sources, to test the accuracy of words and phrases in their dialogue. The culturomicists were taken by this idea, imagining how Google Ngrams could be applied to the creation of historical fiction on the page and screen. And now we can begin to see the applications more clearly. It leads me to wonder: just as sabermetrics has created a place for statisticians in baseball, could we be seeing the birth of new job prospects for digital humanities scholars in the entertainment industry?