First of all, I'm quite familiar with the phenomenon of "reference inflation." I love the way Dean Dad puts it: "An arms raise of puffery." I recall reading a very interesting discussion of it in Stephen Carter's book Integrity. I don't have my copy handy, so I can't look it up, but I give a wholehearted recommendation to anyone to read it.
Since reading Integrity, I've tried hard to make my own letters of recommenation reasonable and realistic. I've tried to avoid the trap of calling every student "one of the best students that I've ever had," as Carter describes the problem.
But this leads to a new problem. How is the the person reading this letter of recommendation supposed to distinguish between my letter, an honest evaluation of a good student, and a similar letter that some chronic exaggerator has written for some slacker? Or, for that matter, if I really do have a terrific student to write a letter for, how am I to communicate it that I really mean it when I say they are one of the best students I've ever had.
I think it's also true that there can be a disconnect between the candidate's goals and the letter writer's. I think this happened to me. I got my Ph.D. from a fairly prestigious, large research university. My goal, however, was and is to teach at a smaller school that values me for my classroom presence. I never really intended to have a major research program, and as things have worked out, I haven't done a lick of real research in 10 years. I did get letters from people who knew my teaching, and who understood my goals. However, I had to have a letter from my advisor (How could I not?) and he just didn't get it. He is a big-name researcher, and it's really all he understands. I'm quite sure that his letter of reference was "research oriented," and I believe that this turned off some schools that were potentially a good match for me. I do know that the first few interviews I got were from schools that wanted more research than I was prepared to give. I did finally find a school that was an excellent match, but I suspect that was because they ignored my advisor's letter.
Having said all that, I'll reverse field and say that I don't think we should abolish letters of recommendation. When I've sat on hiring committees, the things that have told me the most about the candidates are their own personal teaching statement and their letters of recommendation. I feel like a get a good picture from these taken together, a much clearer picture than I get from a transcript or a C.V. that lists a bunch of fairly ordinary accomplishments. Yes, I do discount the odd letter that is poorly written, or that seems to paint a different picture of the candidate than the others taken together.
So, as Dean Dad and his commenters make clear, it's a complicated issue. It will of course be up to the local hiring committee whether to require letters or not, and it always has been. I can't say that I'd mind it if such a requirement became less standard, and I'd certainly support DD in his effort to eliminate them in his own searches. But I really expect we'll keep them here at my institution, and I think I prefer it that way.
technorati tag: teaching-carnival