You will all have to pardon me for not posting for the last few days, because I have been busy developing a new and exciting scientific theory. Here it is, the fruit of my labors:
(Smart Science Person + Science Reporter / Anecdotal Hook)^# of Repetitions = Absolute Drivel
Before you dismiss my glorious equation as nothing more than a clever rhetorical device used to spice up some minor nitpicking, allow me to explain. It all begins with this piece over at–of all places, the BBC News website–titled ” ‘Oldest English Words’ Identified“.* According to the article, a group of researchers at Reading University have developed a computational model that analyzes the rate of change of words in the Indo-European language family, and from this model they have concluded that words change more slowly the more often they are used.** And using this model the researchers believe they can speculate fruitfully about what words might have looked like 10 and 20 and even 40,000 years ago.
Yet somehow, after an extended conversation with the BBC science editor and other affiliated reporters, the head researcher on the project, Professor Mark Pagel, ends up spouting utter nonsense like this:
“You type in a date in the past or in the future and it will give you a list of words that would have changed going back in time or will change going into the future,” Professor Pagel told BBC News.
“From that list you can derive a phrasebook of words you could use if you tried to show up and talk to, for example, William the Conqueror.”
You heard him right. When a wild-haired scientist pulls up in a Delorean and tells you that you’ve got to get back to the 1066 Enchantment Under the Sea dance*** to make sure that William the Conqueror and Matilda hook up, you should run a computer model that simulates average linguistic change over time and rewind through 1000 years of linguistic evolution in order to figure out how to talk to the locals. Rather than, say, telling Doc Brown to swing by Barnes and Noble on the way up to 88 MPH so that you can snag a copy of Clark-Hall’s Concise Anglo-Saxon Dictionary.
I mean, it’s not like the English language in 1066 is some great mystery that we poor scholars can only theorize about today. Ever hear of a little poem called Beowulf, BBC? You do realize that the weird scribbles on the left-hand pages of the Seamus Heaney edition are not just some sort of fanciful elven language cribbed from Tolkien, right? Wouldn’t using this model to make a medieval phrasebook be a lot like taking a model that analyzes average genetic mutation and using it to predict what the Mesozoic ancestor of humans might have looked like?
OK, I know you probably think I’m being too hard on the BBC, that I’m being purposefully obtuse. Surely, you say, all the reporter meant for you to take from that little quote above was that Professor Pagel’s team’s model is “kinda nifty”. They didn’t mean to give the impression that you could type “Now, we’ve all heard about how awesome those Danish leaders used to be back in the day” into their linguistic wayback machine and have it spit out “Hwæt! We Gardena in gear-dagum, þeodcyninga, þrym gefrunon, hu ða æþelingas ellen fremedon.” Instead, they meant that their model would have been able to predict ahead of time that words like “we,” “in,” “day” and “how” would more likely to be the same 1000 years ago than words “awesome” “leaders” and “heard”.
As much as I’d like to believe you, I don’t know how that explanation would jive with the pictures and captions they use to illustrate the article. Right by the nonsense about phrasebooks, we have this helpful sidebar to make it absolutely clear that they have no intention of talking about what the model actually does:
 |
| Time-travelers would find a few sounds familiar in William’s words. |
See, it really is all about the time-travelers and what’s best for them on their hypothetical visit to Hastings.
At this point, I suppose I should go ahead and mention that William was a Norman, and so would be far more likely to be shouting at you to move your inoperative time machine**** from his battlefield so that the archers can get in position in French, not English, so this whole line of discussion is probably moot.
Though if I may be permitted to beat this dead cheval just a few sentences longer, time-travelers trying to speak French to William would have had a whole lot more luck than those trying English, even if we ignore the question of William’s English fluency. While eleventh-century English looks a lot like those few lines I quoted from Beowulf above,***** eleventh-century French is much closer to modern. Take this snippet from the eleventh-century Life of St. Alexis:
‘E! Deus,’ dist il, ‘bels reis qui tot governes, se tei plöust, ici ne volsisse ester.”
You might have a little problem deciding on tenses here and there, especially in the second half, but I’m pretty sure you could get ” ‘Oh, God,’ he said, ‘beautiful king who rules all’ ” with only second-semester French. (For the record, this seems to me like a problem with the whole premise of the computational linguistics model the folks at Reading put together. If there really is a law-like constant rate of linguistic change, why is French so conservative and English so promiscuous?)
The other sidebar picture and caption they use for this story is also a treat, if only for manuscript snobs:
 |
| Medieval manuscripts give linguists clues about more recent changes |
I suppose it’s technically true that linguists use medieval manuscripts to track linguistic changes more recent than 1066. But that’s only because they use them to track linguistic changes before 1066 as well. Indeed, I’m pretty sure all the linguistic inputs concerning the middle ages for the fancy computer model are derived, at one remove or another, from evidence culled from manuscripts. This is like running a story about how the Republicans are going to try to stop Obama’s health care plan and illustrating it with a picture of Garrett Augustus Morgan (the inventor of the stoplight)***** labeled “Other famous African-Americans have had challenges involving stops.” Technically true, yes, but not exactly relevant.
And, as if to add insult to injury, the picture they chose is a stock image of the Macclesfield Psalter. I’m almost 100% certain than nobody’s studying the Macclesfield Psalter for clues about English’s recent changes, because it, like every other lavishly illuminated fourteenth-century psalter, is written in Latin.
So let this be a lesson to you. If you’re a smart person with a clever new theory or process, stay as far away from the BBC’s science reporters as you can. You might think you’re going to get the chance to tell the world about your new method of detecting dark matter, but they’re just as likely to try to get you talking about whether, on Bizarro World, Bizarro Superman is just plain Superman or if he’s some sort of extra-Bizarro Bizarro Superman. And if they put you on the front page, it’s only going to be as an excuse to show a picture of Bizarro with the caption, “Does dark matter explain the Bizarro universe? Bizarro says, ‘This am good question.’ “
—
*Or perhaps I should say that it begins with this post over at the Language Log, which first directed my attention to this story.
**Apparently, these researchers first published their results in a reasonably well received paper in Nature back in 2007. That paper was called”Frequency of Word-Use Predicts Rates of Lexical Evolution Throughout Indo-European History“, but there is no way to determine that from the article. Again, I must thank the good people at the Language Log for said information.
***Though I would imagine, given the way that dashboard display works on that Delorean, we’ll probably have to settle for going back to 1055. And listening to Huey Lewis while we do.
****You should have asked the Libyans for more plutonium before you left. It’s 1066 1055. You can’t just pop over to the corner store for some more.
*****That’s right, I’m a late dater for Beowulf. Kaluza’s law my ass, that’s all I have to say.
******Thank you, Black History Month. [Music cue] The more I know….