essays from linguistics, some technical, some funny

Pullum is funny, cutting, and doesn't have time to fuck around. You wouldn't want to be a confused undergrad in his class, but luckily he's embraced public outreach as well, so you can read his posts on Language Log, or articles like this one, from your armchair with your coffee and come away amused and edified with no danger of drawing his attention. This tossed-off opinion piece has become the definitive refutation of the "Eskimoes* have x words for snow" canard, despite Laura Martin's earlier yeo-work in actually looking at the development of that idea and its inconsistencies and actually, like, proving it wrong, and Steven Pinker's later misappropriation of Martin's and Pullum's work to make ad hominem attacks on Benjamin Whorf, the popularizer of the idea, and ridicule the broader concept of linguistic relativism of which the Eskimo vocab thing is a trivial example.

I do not know what Geoffrey Pullum's stance on linguistic relativism is, which says good things about him: he's not trying to mischaracterize anything. (As opposed to anyone: he gets in some easy digs at Whorf for being a fire-safety inspector and not a real linguist, which is fucked because the guy did good, meaningful work, but I get the feeling Pullum's just trying to entertain.) He gets in and gets out: looks at how the thing was born from an offhand observation of Boas's (that English could have had a single root meaning "water–" for words like river, lake, rain, etc., and that some languages, like Inuktitut, do in fact have multiple productive roots for kinds of snow, where we have only one). Then he looks at how it was repeated and repeated until it was management consultants saying Eskimoes have 1000 words for snow and their brain is different and you need to maximize their productivity differently! Then he looks at how it's not meaningfully the case, that there are only two productive roots, qanik 'snow in the air' and aput 'snow on the ground', and that the multiplicity of word forms is an entirely unremarkable feature of the Inuktitut language, which agglutinates to show various different features of language, so that the number of possible snow-words is essentially infinite and their frequency essentially zero. Then he notes that even if there were multiple productive roots it would not mean any more than how printers have mutiple names for fonts and type-stuffs. He does it all without even mentioning linguistic relativity, and it's quite good. Too bad he couldn't resist being a dick to Whorf for no reason but.

*as long as we're debunking the popular ignorance, why don't we stop calling them "Eskimoes," guys? Inuit. INUIT. Is this an American thing, like "Amerind" (that rare and delicious fruit)? ( )