It's a hoax!

Incredible: the fact that Inuit languages have lots of different words for snow is a myth!

In his article, "The great Eskimo vocabulary hoax," Geoffrey K. Pullum argues that people cite this "fact" without checking.

Once the public has decided to accept something as an interesting fact, it becomes almost impossible to get the acceptance rescinded. The persistent interestingness and symbolic usefulness overrides any lack of factuality.

First of all, the statement "Inuit languages have lots of different words for snow" is ambiguous: which Inuit languages? How is a 'word' defined? And how do we determine that it is a word that defines 'snow'?

In addition to being a very vague and ambiguous statement, it is uninteresting: English has lots of ways to talk about snow too! Powder, sleet, blizzard, snowdrift, snowbank, slush, etc.

C. W. Schultz-Lorentzen's Dictionary of the West Greenlandic Eskimo Language (1927) gives just two possibly relevant roots: qanik, meaning 'snow in the air' or 'snowflake', and aput, meaning 'snow on the ground'.

[...]

Even in English, the distinction between internally unanalyzable roots (like snow and slush) on the one handand inflected word forms of nouns on the other is worth noting. Snow is one word, but it is easy to generate another dozen directly from it, simply by applying inflectional and derivational morphological rules to the root: snowball, snowbank, snowblower, snowcapped, snowdrift, snowfall, snowjuke, snowlike, snows, snowshoe, snowstorm, snowy... You get the picture.

Now, this may not seem like too wild a profusion of derived words. But in the Eskimo languages there is a great deal more inflection (grammatical endings) and vastly more fully productive derivational morphology (word formation). For each noun stem there are about
280 different inflectional forms. And then if you start adding in all the forms derivable by word formation processes that yield other parts of speech (illustrated in a rudimentary way by English to snow, snowed, srzowittg, snowier, srtowiest, etc.), you get an even bigger collection-indeed, an infinite collection, because there really is no such thing as the longest word in a language of the Eskimo type where words of arbitrary complexity can be derived.

So if you identify four snow-related noun stems in some Eskimo dialect, what do you report? Four? Or the number of actual inflected noun forms derivable therefrom, certainly over a thousand? Or the entire set, perhaps infinite, of relatable words of all parts of speech?

Linguistics is fascinating.