Late recommendation letter

Question

I'm about to apply for postdoc positions in the US through mathjobs. I already did it last year, so I have my reference letters from the previous time (which are probably pretty good).

Now here's my situation: During the last year, I was lucky to solve some well-known open problems in my field (well, it's not going to earn me a Wikipedia entry, but the tiny group of researchers in my esoteric field is impressed), which will probably turn my application from "good" to "outstanding". In order to really make my application look better, my reference letters should address the recent results, so I asked my letter writers to update their letters. Most of them did so, except for one (who didn't reply to my emails). The problem is that I should submit some of my applications by November 1, and as I didn't hear from that letter writer for several weeks now, I'm afraid that he might not update his letter by November 1.

Question 1: When job ads on mathjobs say something like "screening of applications will begin on November 1", do they really mean it, or do they mean something like "screening of applications will begin on mid December, but we want you to submit your application early"?

Question 2: What would you do in my situation (assuming that the relevant letter writer is currently abroad and I have no other way to contact him)? Should I contact the relevant search committees and ask them to check that reference letter a bit later, once it gets updated?

Question 3: If a letter of recommendation is updated on mathjobs past the submission of an application, which version of the letters is available to the search committee? What should I do in order to make sure that they read the updated version?

Answer 1

Question 1: When job ads on mathjobs say something like "screening of applications will begin on November 1", do they really mean it, or do they mean something like "screening of applications will begin on mid December, but we want you to submit your application early"?

It depends on the department. It is however well known that many faculty members drag their feet with recommendation letters, and this happens more or less independently of the quality of the applicant. So in my experience it is common to allow a grace period of a couple of weeks to allow the letters to trickle in.

Question 2: What would you do in my situation (assuming that the relevant letter writer is currently abroad and I have no other way to contact him)? Should I contact the relevant search committees and ask them to check that reference letter a bit later, once it gets updated?

Frankly I think you're too worried about the entire situation. You did some exciting new work: great. All but one of the letters will describe this exciting new work. The other letter may not. So when people read the letters they will see that your exciting new work is so new that one of your writers is a little out of the loop. I see no big problem here. If one of the letters is more negative than the others, that drags your application down. If one of the letters is less informative than the others...frankly, that usually happens and is really much of the reason why you get multiple letters.

Question 3: If a letter of recommendation is updated on mathjobs past the submission of an application, which version of the letters is available to the search committee? What should I do in order to make sure that they read the updated version?

Hmm. I believe what happens is that if the file has not been read by anyone, the letter updates without recording the change. If it has been read, then both versions of the letter appear. But I am not completely sure about this.

In summary: you solved some well-known open problems in your field. That's the dream. You say that this will probably turn your application from "good" to "outstanding." Well, I agree. Congratulations, and I hope you enjoy your postdoc.

Answer 2

Question 1: It depends. In my experience on hiring committees it means that no one will look at the applications until after November 1, that some industrious members of the committee will immediately start looking at applications on November 1, and that other members of the committee won't get around to it for some weeks after November 1. A well organized committee might meet in mid November and screen out 90% of the applicants at that point. A badly organized committee might not do anything until January. You have no way of knowing, so the best that you can do is to have a complete application in place by November 1.

Question 2: I'd get a complete application together now, and submit it for those jobs with early deadlines, even though you don't have this new letter of recommendation yet.

Question 3. You can update the letter of recommendation later, but there's no way to be sure that the committee will look at it. In particular, if you've been screened out in the first round, they'll probably never look at any updates to your application. If you're invited to participate in some kind of an interview (at a meeting, by skype, etc.) than they might look at the updated material. When/if you're contacted to arrange an interview, be sure to mention that you've updated your material on mathjobs.org and ask them to look at the updated material.