There's something screwy with the MLA's numbers. The Inside Higher Ed and the NY Times articles (
http://www.nytimes.com/2009/12/18/education/18professor.html) both give the percentages for the foreign languages as "Spanish (35.5 percent), French (16.0 percent), Chinese (9.5 percent), German (4.0 percent), Arabic (3.0 percent) and Italian (2.0 percent)." Now, Chinese is having a very good year, but it's not
that good.
If you search on the ADFL job information list for these languages (combining "Spanish" with "Latin American" and "French" with "Francophone Studies," and adding "all dates" [=current] and "only expired listings"), you find, for all academic ranks combined, the following:
Spanish/LatAm 259
French/Franc 112
Gmc/Scand 60
Chinese 50
Italian 30
Arabic 28
According to MLA, these languages represent 70% of all their jobs (add up the percentages above). So if you divide the number of jobs by the total of just these six languages and multiply by .7, you should get more or less what the MLA suggests...but you don't. Instead, you get
Spanish/LatAm 33.6%
French/Franc 14.5%
Gmc/Scand 7.8%
Chinese 6.5%
Italian 3.9%
Arabic 3.6%
The numbers are close enough to the MLA's that I suspect I'm not completely off here, but it looks like MLA is reporting significantly lower percentages for Germanic/Scandinavian than the posted jobs indicate, and over-stating the numbers for Chinese.
Now there are some caveats, most notably that searching the ADFL list by "all ranks" picks up some oddball postdocs and weird international positions that get duplicated in many languages, so these numbers aren't perfect by any means. But if you look, for example, only at assistant professor-rank, which should eliminate some of that effect, you still find:
Spanish/LatAm 198
French/Franc 79
Gmc/Scand 32
Chinese 30
Arabic 19
Italian 14
In other words, I suspect that the Big 3 have not been rearranged quite yet.