I was looking through the newest information from Medicare, which looked to my untrained eyes just like the previous version and the one before that and the one before…

What caught my eye won’t surprise you: some sentences (presumably equivalent ones) in 16 languages, none of which was English. Take a look, and then I’ll raise a few questions.

We notice immediately that the languages are alphabetized by their English names. Fair enough, since no other system would really be both fair and easy to use. But what questions do we have?

• One wonders about the choice of languages: presumably these are the 16 non-English languages most often spoken in the U.S. But are they? The latest data are unfortunately from 2009–2013, but that’s the best I could do. Here are the top 16 in this spreadsheet, listed in order of numbers of speakers:
  • Spanish (no surprise there)
  • Chinese (i.e., all Chinese languages put together)
  • Tagalog
  • Vietnamese
  • French
  • Korean
  • German (including Luxembourgian…hmmm…)
  • Arabic
  • Russian
  • French Creole (presumably Haitian)
  • Italian
  • Portuguese (including Portuguese Creole)
  • Hindi
  • Polish
  • Japanese
  • Urdu

• So the answer is no. Fourteen of those match the Medicare list, but what happened to Hindi and Urdu? Do they think everyone from South Asia speaks English?
• If two of the top 16 aren’t on the Medicare list, then two of the languages on the Medicare list must not be in the top 16. Somehow Armenian and Persian (Farsi) have replaced Hindi and Urdu. Why? Do they have a thing for Mideast-adjacent languages, where Armenian and Persian are (slightly) closer? Inquiring minds want to know.
• Finally, we wonder whether the sentences really are equivalent. Not knowing very many of the languages in question, it’s hard for me to tell. I don’t really want to rely on Google Translate. I could compare the French and German, which might tell me something, and I might make a stab at the Italian and Spanish, which are at least somewhat familiar to me. So what does that attempt get me?
• Those four do seem to be equivalent, so by induction we can conclude that all 16 are equivalent. Right?
• Well, no. Those four are all Indo-European. Worse yet, three of them are Romance languages, so we have a very narrow sample. I guess we have to resort to Google Translate after all. We’ll try two. Tagalog (which Google calls Filipino) checks out as fully equivalent. But Japanese doesn’t! All the other languages that I’ve checked are addressed to “you or someone helping you,” but the Japanese is only addressed to “you.” Do they think Japanese people don’t need help? Can we solve that mystery? Inquiring minds want to know.

Categories: Linguistics