Lexicon Valley 2014 highlights: Did you miss any great writing on Slate's language blog this year?

Did You Miss Any Great Language Writing This Year? Top Lexicon Valley Posts of 2014.

Did You Miss Any Great Language Writing This Year? Top Lexicon Valley Posts of 2014.

Lexicon Valley
A Blog About Language
Dec. 30 2014 11:55 AM

Did You Miss Any Great Language Writing This Year? Top Lexicon Valley Posts of 2014.

lexicon_valley_2014_highlights
Here's to more fun with language in 2015!

Lexicon Valley logo modified by Gretchen McCulloch

Did you miss any great language-related writing on the Lexicon Valley blog this year?

Based on a combination of shares, views, and reader feedback, here are the 10 posts on Lexicon Valley that readers liked best in 2014. For the sake of variety, I've only included one post per author, but if you're curious, you can also click on the author names for more by the same person.

Advertisement
A few years ago, Top Gear host Jeremy Clarkson pointed out that the best way to expose a German spy would be to ask them to say the word squirrel, because "no German, no matter how well they speak English, can say 'squirrel.' " So naturally, someone tested the premise back in 2013 and made a video of it. (Read more.)
All too often, what happens is something like this story I heard from a math teacher in a first-grade classroom, "One of the kids, an African-American kid, was playing a game and he said, ‘I don’t got no dice.’ He didn’t have the materials he needed. And the teacher said, ‘You know, Joshua, we speak English in this class.’ Really harshly. And I just thought, oh gosh. There must be a better way to respond."
But what's a teacher to do? On the one hand, they need to help students prepare for a world that—like it or not—isn't particularly accepting of linguistic variation. But on the other, they want to do so in a way that lets students continue to be proud of who they are and where they come from, rather than pushing them into tongue-tied linguistic insecurity. (Read more.)
Advertisement
It's that post-sangria effect of increased Spanish fluency: In one study in the 1970s, college-age English speakers were asked to perform a pronunciation test in Thai, a language they had no experience with, after drinking different amounts of alcohol. What happened? Those who drank 1.5 ounces of alcohol performed better on the pronunciation test than those who had drunk an alcohol-free placebo. An ounce and a half of alcohol was apparently the perfect amount to relax the subjects' egos enough to not feel silly while trying to produce the sounds of a language very different from English. But those subjects who drank 1.5 ounces of alcohol also performed better than those who had consumed 2- and 3-ounce drinks. (Read more.)
Whether the action was feeding, walking, or fornicating, though, all of these early examples were used to mean "to loaf around" or "to waste time" (dogs have often been associated with laziness, as in the expression "dogging it"). Later on, possibly around World War II, "fucking the dog" and its euphemistic equivalents took on a secondary meaning of "blundering." (Read more.)
Advertisement
The funny thing is, uptalk isn't actually just used by the young and female. When you’re on the lookout for it, you’ll hear uptalk from people of many demographics. Yet I’ve never heard anyone condemn New Zealanders’ speech for not being authoritative or confident enough, despite their rampant use of uptalk at all ages and genders. I also hear many men, including former President George W. Bush, using uptalk, and have yet to hear any of them be chastised for not sounding authoritative enough. In fact, there's no conclusive evidence that women even use uptalk more than men.
But even if women did uptalk more than men, we've all heard enough uptalk to know that its rising intonation doesn't indicate a question. No one's actually confused. So why should anyone have a problem with it? (Read more.)
at_least_ten_ducks
The animated GIF above—along with the caption "Look at all those ducks there are at least ten"—is currently floating around the Internet. The thousands of people who have liked and shared it on various sites, including Tumblr, Reddit, and Imgur, presumably think it's entertaining, but why?
At the risk of spoiling the joke by overanalyzing it, a linguist named Paul Grice has an answer. (Read more.)
Advertisement
As the proportion of boys in the group increases, boys interrupt more and girls interrupt less. The data suggests that the reverse may also be true, but it takes more girls overall to bring the interruption rate to parity. And we'd need some even more girl-skewed playgroups to confirm.
When girls play together without boys, they interrupt more. A lot more.
Remember that the overall interruption rate is similar regardless of whether or not boys are part of the group. That means that when boys aren't around, girls pick up the slack in their own interruption rate. That's just basic math. (Read more.)
Advertisement
In the 17th century, English speakers fell under the spell of a peculiar linguistic fad. With some exceptions, they started to use a seemingly plural form of a field of study to refer to it in the singular. Enter physics, acoustics, economics, acrostics. The rule wasn’t applied uniformly: Disciplines that had been around for a while, such as arithmetic, had already rooted deeply enough in people’s minds to avoid the trend. But mathematic, the classical and somewhat arcane science of all things numerical, acquired an S. (Read more.)
What about this idea that non-native speakers produce a lower quality of input than native speakers? It makes intuitive sense—we know we make grammatical mistakes in a second language, so why wouldn't children learn them?—but it's not supported by the evidence. In fact, kids who are exposed to early language from non-native speakers usually grow up to be full speakers of that language. For example, deaf children of hearing parents benefit greatly from early exposure to Sign Language from non-natively signing parents, and in fact end up almost as fluent as deaf people who have Sign Language exposure from birth. Another striking example comes from Daryl Baldwin and David Costa’s work on revitalizing the Native American language Myaamia, where children fluently use sounds and grammar that their parents, who learned the language as adults, still struggle with. (Read more.)
It's the idea that somewhere out there, there's a perfect, unadulterated version of English, and what your everyday person speaks is a poor copy. I call it the kilogram model of language, because there is literally a physical object in France by which the unit kilogram is defined, and there are in fact multiple and worryingly imperfect copies of it around the world. But what linguists have discovered is that language is definitely not like the kilogram. The only place where English really exists is in the minds of its everyday speakers. To the extent that varies geographically and socially, so does English. There are no imperfect copies. (Read more.)
The index cards appear to just be a random series of letters, and had confounded the poster's family for years. But it only took Metafilter 15 minutes to at least partially decipher them. User harperpitt quickly realized she was using the first letters of words, and that she was, in fact, writing prayers. (Read more.)
The oblique, tongue-in-cheek invocation of "penis envy"—a theory of female sexual identity that only a man could conjure—imputed a you-can't-fight-your-nature air of inevitability to our use of pronouns, ignoring the many decades of struggle by women to reform language that undermines them. Ironically, as the late authors Casey Miller and Kate Swift point out in Words and Women, Watkins had previously written that language "is at once the expression of culture and a part of it," and here was failing to apply his own insight, as if grammar were handed down from on high with no choice but to maintain it. (Read more.)

To keep up with Lexicon Valley in 2015, follow @lexiconvalley on Twitter or on Facebook.