There has been a lot of discussion both here at AllTwitter and in the Twitter-sphere in general about how Twitter is affecting language.
The latest round of the debate was sparked by some rather incendiary comments made by actor Ralph Fiennes, who claims that Twitter is dumbing down the English language (even though he doesn’t actually have a Twitter account himself to verify this claim).
But those on the other side have a more experienced, if less famous, champion to back their protests up: a linguistic professor has stepped into the fray to pour water on the theory that Twitter is destroying language.
It’s all well and good for Fiennes to cry about the erosion of language thanks to “a world of truncated sentences, sound-bits and Twitter,” but his generalizations can’t stand up to number-crunching.
Mark Lieberman, a linguistics professor at the University of Pennsylvania who blogs at Language Log, was intrigued by Fiennes’ comments and wanted to test his claim.
He conducted an experiment to see just whether words are really becoming shorter, as Fiennes complained about to the media, or whether these were unsubstantiated worries.
Lieberman took the text of Hamlet and a number of P.G. Wodehouse stories and compared their word length against the latest 100 tweets from his university’s independent student newspaper (you can read a detailed explanation about how he did this here).
And here are the results from the comparison:
- Mean word length in Hamlet (using modern spelling): 3.99 characters
- Mean word length in P.G. Wodehouse’s stories: 4.05 characters
- Mean word length in the tweets: 4.80 characters
The tweets that Lieberman examined had a longer mean word length than the words in Shakespeare and Wodehouse – contrary to Fiennes’ complaint that Twitter is making society use shorter words.
Of course, this was just a small study of a handful of texts and a single Twitter account, so it can’t be extrapolated to the entire Twitter-verse. However, it is interesting that a linguistics professor was so quickly able to use just a smidgen of the scientific process to show that Twitter might not be the scourge of language as its detractors would like to claim.
Do you think that a larger study of some classic English texts against a wider variety of Twitter accounts would come up with similar results? Let us know your thoughts in the comments below.