By "Guido Veloce"
Illustration by Michael Morgenstern
 
As I write, one academic season gives way to the next. We're leaving the season of writing letters of recommendation and entering the one of reading them. The former lasts most of the year. The latter is brief but intense. Both activities are time-consuming, necessary, and may not make a whole lot of difference. At least I hope they don't because the process is flawed, a chain with no strong links.

Difficulties begin at the beginning. When it comes to creating forms and formats for letters of recommendation, colleges, universities, and funding agencies have done the ultimate outsourcing. Somewhere in the bowels of Hell, the Devil's minions work non-stop to complicate the process. Want it electronic? We'll do passwords guaranteed to protect the form from the recommender. We'll stack the odds in favor of accidentally sending it prematurely, with no hope of retrieval. You prefer an anti-electronic format? We'll do forms that can only be filled in by hand. We'll ask for six copies in separate envelopes. We'll throw in irrelevant questions and insist on a response. Sample: "Does the applicant get along well with peers?" (Answer: "From his behavior in my 200-student lecture course, I can attest that he is a sociable young man, always chatting amiably with those around him when he is not reading or sending text messages.")

The next problem point is the applicant. Perhaps out of a concern for confidentiality, many are reluctant to reveal information, like to whom the recommendation should be addressed and when it is due. Or was due. Others give detailed instructions, including how to express the applicant's enormous strengths and dearth of weaknesses. The root assumptions here are that the faculty member has never done a letter of recommendation before, is not sufficiently impressed with the applicant, and can't be trusted to get it right. The first assumption is wrong; the second is arguable; and the third has merit. That brings me to the recommender.

Some letter writers appear not to know the applicant. My department received a recommendation in which the author got the would-be graduate student's gender wrong (unless there was a story we weren't told). Another described in great detail a senior thesis the applicant hadn't written. Still other recommenders pursue counterproductive strategies. One prominent scholar was fond of describing each candidate as "among the top two or three PhDs I've taught." One year five of the top two or three applied for the same job. An equally distinguished scholar had his own version. He made a mid-career move from Bigger State University to Smaller State University, which has a less visible graduate program. His stock phrase for applicants from the latter was "as good or better than the best students I had at" Bigger State. This comment was less compelling to his Bigger State students sitting on search committees than it was to their colleagues.

The final weak link is the letter reader, harried and increasingly impatient trying to decipher cryptic phrases. What was the "bit of a scrape" in one applicant's past? (He came to Johns Hopkins, was a good student, and later explained the episode to me. A more apt phrase would be "a bit of a felony.") What to make of a letter praising a job candidate, then calling him "much like Sammy Glick"? The writer was probably banking on the search committee not getting the reference and on the applicant's attorney never seeing it. Faced with the enormous task of making sense of these letters, readers soon grow waspish, as did one colleague who scrawled an angry comment complaining that the writer didn't know how to spell Milwaukee. The applicant was from Oregon. Milwaukie, Oregon.

The picture is not all dismal, and a significant fraction get it right at every stage. Some letters are even literary gems and, as an added bonus, helpful. But here is the brightest spot of all: Seldom does a bad letter, by itself, sink an applicant, and seldom does a glowing one, by itself, elevate one. That's a consoling thought for applicants, letter writers, and letter readers alike.

"Guido Veloce" is a Johns Hopkins University professor.

Return to February 2008 Table of Contents