Over and over again

“I agree with your objections to the unhelpful comparisons using the stack of money and distance to the moon,” wrote T. Roger Thomas in a comment on my criticism of a New York Times op-ed. Then he went on:

I would also add an objection to the use of "over" in the original piece. To my way of thinking, one can climb over a fence. I would prefer to see the term "more than" used to denote a greater amount of things, which, in this instance, happen to be dollars.

He’s talking (I assume) about Ezekiel Emanuel’s “$2.6 trillion on health care, over $8,000 per American.” But this is a nit that I didn’t pick even when I was a professional nitpicker, despite the temptation to go along with my fellow journalists. "Disapproval of over ‘more than’ is a hoary American newspaper tradition,” says MWDEU, but the usage was standard for many centuries before some 19th-century crank decided he didn't like it.

We haven't identified that crank; as I said in an April 2010 post on "over" and "more than," the earliest mention of the  issue I’ve found is in an 1856 usage book by Walton Burgess, son of a New York City printer/publisher, grandly titled “Five Hundred Mistakes in Speaking, Pronouncing, and Writing the English Language, Corrected.” At No. 130 we get: "'There were not over twenty persons present:' say, more than. Such a use of this word is not frequent among writers of reputation."

William Cullen Bryant followed Burgess's lead in 1870, and Ambrose Bierce piled on in 1909. But even in its heyday, the “rule” was far from universal. Scott and Denney's "Elementary English Composition" (1900) said that “over a million dollars” was correct usage. And in 1856 -- the very same year that Walton Burgess declared war on this "over" -- a rival usage book, with an anonymous author, defied his ruling in the boast above its title: “Over 1000 Mistakes Corrected.”

For more on “over/more than,” see John McIntyre (at his non-paywalled former blog) on the AP Stylebook as a "repository of extinct rules"; Mark Liberman at Language Log; and Paul Brians's Non-Errors Page: "This absurd distinction ignores the role metaphor plays in language. If I write 1 on the blackboard and 10 beside it, 10 is still the 'higher' number."