On Saturday, I presented at the twentieth annual conference of the American Copy Editors Society, held in Portland, Oregon. My topic was “Copyediting and Corpus Linguistics”, and my aim was to give editors a crash course in using corpora to research usage questions.
I was floored by the turnout—there were probably close to two hundred people in attendance. And I was surprised at how quickly the hour went by. I wish I’d skipped most of the background stuff and spent more time demonstrating how to do different kinds of queries and answering questions, but live and learn, I guess. I’ll have to remember that if I teach the topic again at future conferences.
I’m posting a slightly expanded version of my presentation below. And if you attended and you still have questions, or if you weren’t able to make it, feel free to leave a comment here. I’ll do my best to answer.
And, of course, a big thanks to ACES for hosting and to all those who attended. ACES folks are the best.
Last month, at the yearly conference of the American Copy Editors Society, the editors of the AP Stylebook announced that over in the sense of more than was now acceptable. For decades, newspaper copy editors had been changing constructions like over three hundred people to more than three hundred people; now, with a word from AP’s top editors, that rule was being abandoned.
According to Merriam-Webster editor Peter Sokolowski, who was in attendance, the announcement was met with gasps. Editors quickly took to Twitter and to blogs to express their approval or dismay. Some saw it as part of the dumbing-down of the language or as a tacit admission that newspapers no longer have the resources to maintain their standards. Others saw it as the banishment of a baseless superstition that has wasted copy editors’ time without improving the text.
The argument had been that over must refer to spatial relationships and that numerical relationships must use more than. But nobody objects to other figurative uses of over, such as over the weekend or get over it or in over your head or what’s come over you? The rule forbidding the use of over to mean more than was first codified in the 1800s, but over can be found in this sense going back a thousand years or more, in some of the earliest documents written in English.
Not only that, but parallel uses can be found in other Germanic languages, including German, Dutch, and Swedish. (Despite all its borrowings from French, Latin, and elsewhere, English is considered a Germanic language.) There’s nothing wrong with the German Kinder über 14 Jahre (children over 14 years) (to borrow an example from the Collins German-English Dictionary) or the Swedish Över femhundra kom (over five hundred came). This means that this use of over actually predates English and must have been inherited from the common ancestor of all the Germanic languages, Proto-Germanic, some two thousand years ago.
Mignon Fogarty, aka Grammar Girl, wrote that “no rationale exists for the ‘over can’t mean more than’ rule.” And in a post on the Merriam-Webster Unabridged blog, Sokolowski gave his own debunking, concluding that “we just don’t need artificial rules that do not promote the goal of clarity.” But none of this was good enough for some people. AP’s announcement caused a rift in the editing staff at Mashable, who debated the rule on the lifestyle blog.
Alex Hazlett argued that the rule “was an arbitrary style decision that had nothing to do with grammar, defensible only by that rationale of last resort: tradition.” Megan Hess, though, took an emotional and hyperbolic tack, claiming that following rules like this prevents the world from slipping into “a Lord of the Flies-esque dystopia.” From there her argument quickly becomes circular: “The distinction is one that distinguishes clean, precise language and attention to detail — and serves as a hallmark of a proper journalism training.” In other words, editors should follow the rule because they’ve been trained to follow the rule, and the rule is simply a mark of clean copy. And how do you know the copy is clean? Because it follows rules like this. As Sokolowski says, this is nothing more than a shibboleth—the distinction serves no purpose other than to distinguish those in the know from everyone else.
It’s also a perfect example of a mumpsimus. The story goes that an illiterate priest in the Middle Ages had learned to recite the Latin Eucharist wrong: instead of sumpsimus (Latin for “we have taken”), he said mumpsimus, which is not a Latin word at all. When someone finally told him that he’d been saying it wrong and that it should be sumpsimus, he responded that he would not trade his old mumpsimus for this person’s new sumpsimus. He didn’t just refuse to change—he refused to recognize that he was wrong and had always been wrong.
But so what if everyone’s been using over this way for longer than the English language has existed? Just because everyone does it doesn’t mean it’s right, right? Well, technically, yes, but let’s flip the question around: what makes it wrong to use over to mean more than? The fact that the over-haters have had such an emotional reaction is telling. It’s surprisingly easy to talk yourself into hating a particular word or phrase and to start judging everyone who allegedly misuses it. And once you’ve developed a visceral reaction to a perceived misuse, it’s hard to be persuaded that your feelings aren’t justified.
We editors take a lot of pride in our attention to language—which usually means our attention to the usage and grammar rules that we’ve been taught—so it can seem like a personal affront to be told that we were wrong and have always been wrong. Not only that, but it can shake our faith in other rules. If we were wrong about this, what else might we have been wrong about? But perhaps rather than priding ourselves on following the rules, we should pride ourselves on mastering them, which means learning how to tell the good rules from the bad.
Learning that you were wrong simply means that now you’re right, and that can only be a good thing.
Update: Parallel uses can also be found in cognates of over in other Indo-European languages. For instance, the Latin super could mean both “above” and “more than”, and so could the Ancient Greek ὑπέρ, or hyper. It’s possible that the development of sense from “above” to “more than” happened independently in Latin, Ancient Greek, and Proto-Germanic, but at the very least we can say that this sort of metaphorical extension of sense is very common and very old. There are no logical grounds for objecting to it.
I also created this helpful (and slightly snarky) timeline of the usage of over and its etymons in English and its ancestor languages.