I am about to take the French language exam as part of the requirements for the PhD program at my university. The exam will consist of having to translate (with the help of a dictionary) a subset of a mathematics paper written in French. As an Indian international student in the US, I speak a number of languages, but I've grown up knowing them (i.e. never went through the process of having to learn it) and none of them (except perhaps English) give me any insight into the French language, which makes me nervous!

I would appreciate any tips as to how one might prepare for this exam, and materials that might be handy. For example:

  1. Did you have to take a language exam? How did you prepare for it?
  2. Has knowing said language helped you out in your mathematical career?
  3. Do there exist collections of mathematics papers in French that one can practice on?

Thanks you in advance for your answers!

  • Have you asked how many people don't pass the language exam? For 3., a simple google search will be helpful.
    – John Engbers
    Jun 4, 2012 at 13:32
  • For (3), you could go to the website of the IHÉS and download EGA...
    – Zhen Lin
    Jun 4, 2012 at 13:35
  • 1
    Ask your advisor.
    – Potato
    Jun 4, 2012 at 14:27
  • 1
    @Zhen Lin: Maybe EGA is not the best choice for a topologist... Anyway the website of IHÉS contains a lot of other interesting papers/monographs. Jun 4, 2012 at 14:44
  • Basically, if you just learn to translate with a dictionary, it will not be very helpful later, because it will be too painful to actually use it throughout your career.
    – user781
    Jun 10, 2012 at 12:48

2 Answers 2

  1. Yes. I took both German and French. I was already fairly comfortable with French from having studied it in high school and in university, so I'll write about German instead. I prepared for my German language exam by going on to the GDZ server (a program to retroactively digitise documents, including a lot of classical mathematics papers) and downloading some papers in mathematical analysis (my field, sort of) from around the 1950s to 60s (when still a lot of the papers were published in German). I then went to the library and borrowed the copy of Hyman's German-English Mathematics Dictionary (for the nouns) and a copy of a normal German-English dictionary (for the verbs) and sat myself down and started translating word by word. After about a month I translated about 30 pages or so of math papers (and at the same time learned about uniform spaces), and proceeded to take my exam.

  2. Not much. I have absolutely no conversational German, and only very rarely do I have to look something up in German that I cannot find a write-up in another language. (So far twice only, one a slightly obscure result in algebraic lattices that the only reference I could find is to a paper in the 30s.) (French on the other hand has been useful; mostly because the abovementioned "write-up in another language" often turned out to be in French.) Given how much I've found French to be useful for me, I'm sure that if I had actually properly learned the language, German would be just about as useful. Conversely, if you don't really learn the foreign language, you will probably be able to get by through other means in this day and age.

  3. The canonical place to find French papers to read is NUMDAM. Of course, for practice, you can also check out Bourbaki from the library; it may be slightly less intimidating to start working first on the language by reading something whose mathematics at least you are familiar with.


I think the main purpose of learning another language in the PhD program is to be able to explain your research in that language if necessary. At my former institution, the language exam covered reading and writing about mathematics in french or russian. It would be best to investigate your institution's policy about the exam and their requirements for the PhD degree. That will determine what and how you need to study for that exam.

While I've never had to write a technical paper in another language, knowing these other languages has certainly increased my ability to network with experts from around the world. That's the true value of learning languages: being able to connect with others on another level.

  • 10
    I agree with you in principle about learning languages, but what you describe sounds more ambitious than most grad school language exams. My impression is that (in mathematics) they focus almost exclusively on reading, which is a much lower bar than writing. Jun 11, 2012 at 5:22
  • 1
    I think this is generally true of language exams in scientific fields. I've never heard of a science or engineering program require speaking knowledge of a foreign language.
    – aeismail
    Jun 29, 2012 at 4:42

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .