More Linguistic Slush From the Arctic

It seems impossible to stop people babbling on about how the Eskimoan languages have huge numbers of distinct word roots denoting different types of snow, and trying to squeeze conclusions about cognition and culture out of this tired turnip of a factoid. Recently I ran into it again on the Web site of Miller McCune, a magazine based in Santa Barbara. “SMART JOURNALISM. REAL SOLUTIONS.”, says the masthead. (Authoritative periods. After each phrase. Might try that.)

And what sort of smart journalism do we actually get? A limp article about endangered languages and “endangered ideas” that once again hands us the old myth about Yup’ik and Inuit vocabulary:

As the famous example says, Eskimo have numerous words to describe what Americans would just call “snow” and “ice.” This suggests language systems don’t merely translate universal ideas into different spellings; they encode different concepts. And when we lose a language, we risk losing those concepts.

Well, this is Lingua Franca. Smart blogging. About language. And we urge you to ask yourself three questions: (1) Do they? (2) Does it? (3) Do we?

The writer of the piece, Emily Badger, gets a morsel of credit for one thing: She links the word “snow” to a short Web article by a genuine Eskimologist, Anthony C. Woodbury of the University of Texas (see a copy here). So she is trying to connect with factual material on the topic. But she hasn’t fully understood the importance or purpose of Woodbury’s piece.

Woodbury, without making too much of a didactic meal of it, is showing us how few snow words there are in Eskimoan languages—specifically in Alaskan Yup’ik, which he has studied for decades. He invites you to reflect on four challenging questions, which you might like to consider applying to English first:

  • (a) What exactly are snow words? (Is slush a snow word in English? What about sleet? Or blizzard?)
  • (b) Do we count synonyms as separate words, given that by definition they denote the same concept?
  • (c) Do we count noun/verb pairs? (Is the underlined verb in It snows a lot here a different word from the underlined noun in You should have seen the snows we had last winter?)
  • (d) Do we count dialect variants as different words, or concentrate on a single dialect? (That is, does some rural word you don’t know count as a word in your language?)

Woodbury’s list includes at least some nonsnow words (e.g., the second word on his list means “frost”), and it has synonyms, and noun/verb pairs, and dialect-limited variants. And crucially, following his list of just 15 Yup’ik words for wintry phenomena he gives a list of 22 such terms in English.

His message is that the Eskimoan languages (Siberian and Alaskan Yup’ik, Canadian Inuktitut, Greenlandic Inuit, etc.) do not appear have large inventories of snow-referring words relative to English or other languages.

But what is perhaps more important is that even if they had scads of such words, it wouldn’t mean they were encoding concepts English speakers didn’t have or can’t get. There’s nothing wrong with using multiword phrases to encode some concepts. There’s no single word for “way to stamp out a stupid but popular belief,” or “spreading of ignorant twaddle by lazy journalists,” but we can form those concepts.

It was Laura Martin, an anthropologist at Cleveland State University, who first pointed out the size of the Eskimoan snow-word stock was an Arctic myth, in a paper read to the American Anthropological Association in 1982, and published in American Anthropologist (after several years of arm-wrestling with defensive anthropologist referees) in 1986. My humorous attempt to publicize her paper first appeared in 1989, and has been reprinted in various places and used as the title essay of The Great Eskimo Vocabulary Hoax in 1991. The trope is now quite well known to be something to poke fun at (Kate Bush jokes about it in the title track of her new album 50 Words for Snow). But journalists ignore the both of us and prattle on regardless, as if it were factual.

I have even known people who know about Laura’s essay and mine, yet tell me that even if the story is not true, we have to go on using it in our introductory lectures on language, because it’s such a good example.

And when someone says that, you need to ask yourself two questions: (1) Do we? (2) Is it?

Return to Top