Publicity is the top word
After the Global Language Monitor, following numerous publicity-maximising delays, finally announced they’d spotted the ‘millionth word’ in English (which was 2.0, lest we forget), I naively expected that they would go away for good.
Alas, they have not done so, and are now back with their latest piece of lexicographical tomfoolery, a list of ‘top words of 2009’. Apparently by means of a magical algorithm that takes into account “frequency, contextual usage and appearance in global media outlets, factoring in long-term trends, short-term changes, momentum and velocity”, they’ve come up with a list that includes Twitter (at no. 1), Obama, H1N1 (the “politically correct” name for swine flu), and vampire.
As far as I’m concerned, this is obvious nonsense. But I’m apparently alone. Not only did it get picked up in the media, it did the rounds on Twitter too. Because I’m sad like that, I looked through all 600+ Twitter posts that linked to the Global Language Monitor list, and not a single one of them was even mildly critical of it in any way.
I’m not sure that there’s a point to this in any way, other than this drives me mad.
2 Comments
leave a commentLeave a Comment
Additional comments powered by BackType
I’m almost prepared to give them the benefit of the doubt on this one (as opposed to the million word nonsense) because it’s quite conceivable that you could design quite a good system to search lots of raw corpora for spikes in frequency. It’s less conceivable that you could design a good automated system for sense distinction (‘transparency’, for example, has a lot more senses than ‘Obama’), and there’s no obvious coherence in their approach to compounds. (They track some ‘phrases’, but presumably these need to be coded as lexical items in order to be returned, so it’s not properly automated.)
However, since their website talks such absolute voodoo about their search algorithm, I’ll remain a bit sceptical. Hopefully one day they’ll open up their methodology to proper scrutiny.
Yes, conceivable ≠ likely. I'd love to see their methodology opened: I suspect it involves picking newspaper clippings out of a big hat, and would love to see it in action.