The letter of recommendation has outlived its usefulness and is now more trouble than it’s worth. It’s a failed genre. Let’s get rid of it.
Like many of my readers, I have served on both graduate-admissions committees (a few times) and hiring committees (many of them). In this capacity, I have over the years had the opportunity to read thousands of letters of recommendation. And with this experience I have gained a bit of wisdom.
Letters of recommendation are mostly a waste — a predictable exercise that leaves us feeling just a little more self-loathing than we did before. We in the academy continue to insist on them for no other reasons than professional insecurity, bureaucratic anxiety, and, at times, existential dread.
The letter-of-recommendation game is like Kabuki theater but without the artistry. Committees know they need to ask for them, candidates know they need to have them written, and recommenders know they need to write them. The avalanche of formally constrained expectation, the yawning abyss of accountability, and the raging pyre of bad faith that is everywhere in our profession ensure that this hyperstylized performance is mostly show. It’s lunacy to devote so much energy and time to this practice.
On the demand side, we confront a beggar’s army of candidates who face the pro forma, standardized, but somehow still mortifying task of asking at least three professors or mentors or supervisors to write letters on their behalf — letters that in many cases, let’s be honest, won’t be read. On the supply side, mentors are turned into machines for the production of exaggerated claims. Who knew there were so many students who were “one of the top three students I’ve ever taught” or who wrote “easily one of the best papers I’ve ever read — and I’ve been teaching for 25 years”?
And then there’s the consumption side, perilously parsing all this hyperbole. What did Recommender No. 1 mean by “extraordinary”? Are we supposed to read Letter No. 2’s reference to the candidate’s participation in the campus literary club as a veiled expression of concern about the candidate’s scholarly aptitude? In the absence of any commitment to or stabilization of a structure of evaluation, the genre makes us doubt ourselves. Was I right to see the repetition of the word “good” in the third paragraph as a red flag? Is a “brilliant” student better than one who’s “exceptionally gifted”? What about the one who’s “simply put, fantastic”? Should the candidate whose account of Causabon’s poetics of unimportance rates as “a truly original and transformative contribution” to the study of Middlemarch be evaluated more highly than the candidate whose anatomy of affective self-importance “stands to substantially recast the field” of transatlantic 19th-century studies? In a sea of superlatives, all boats sink.
The genre’s manifold failures circulate around its inherent conflicts of interest. Candidates who request letters can easily find themselves in situations where it might be reasonable to worry — for reasons that don’t necessarily have much to do with their preparation or appropriateness for the position for which they’re applying. Letter writers have a vested interest in securing (to the extent that they are able) positive judgments of those for whom they write letters — except of course in circumstances in which they might find it preferable for any number of reasons (not all of which might be considered, shall we say, objective) to elevate one student over another. And committees have to read these letters with a grain of salt big enough to give us all high blood pressure.
Compounding all this, of course, are structural problems of bias, intended or unconscious: Letter-writing swarms with opportunities for the reproduction and aggravation of racism, sexism, classism, and other forms of prejudice. In a profession such as ours, in which the stakes — of everything — are often low and the narcissism of small differences intense, letters end up doing all sorts of things we can’t control for, trust, or desire.
Ideally, a genre represents the stabilization of a set of conventions or functional arrangements between writer and reader that facilitate the transmission of meaning. Or at least that’s what I used to think. But what’s been formalized in this farkakteh genre? Too little of what we read in letters of recommendation means what it says. What’s the point of a genre that can so rarely be read at face value? “Good” in fact means “mediocre”; “bright” can just as easily mean “uninspiring”; “exceptional” means “pretty good”; a reference to Derrida could be meant to signal that the work is old hat and boring, or maybe not; what about Butler? Oooh, look, a letter from Spivak; wait, it’s only 58 words long and doesn’t refer to the candidate by name. Wow, a four-page, single-spaced letter! Whose interests — or ego — is that supposed to serve? Reading letters of recommendation has become a game of Kremlinology.
Too little of what we read in letters of recommendation means what it says.
The fact is that committees should usually be in a position to judge the quality of the candidate’s work without the aid of letters of recommendation. Personal statements, transcripts, and writing samples in the case of graduate admissions, and a job letter and some portfolio of publications or written work and a demonstration of teaching effectiveness in the case of job applicants, provide committees with a pretty good picture of the academic skills of the applicant. What a committee really stands to gain from a reference is a sense of what the candidate — as a person, a human being — is, in the parlance of our times, “like.” Are they a responsible citizen, or are they a cancer on the body politic? Are they a jerk? Might one, I don’t know, want to work in the same department with them?
I’ll admit I don’t really know what might replace the letter of recommendation. Asking for names of recommenders rather than letters — which some departments, of course, already do — might cut down on some waste in that references would be contacted only for candidates who have an actual shot of making it. But this leaves substantially in place nearly all the structural problems of the genre. Having a single member of every department evaluate in a standard format the teaching abilities, say, of all that department’s job candidates might cut down on some problems, but leave in place lots of others, including, importantly, the dangers of bias. In any case, as a profession, we could certainly stand to think more deliberately and self-critically about how admissions and hiring committees can get the supplementary information — mostly about departmental citizenship — that might be useful to them.
I propose we end the practice of requesting, writing, and reading letters of recommendation. Just because it’s the only thing we can envision now doesn’t mean the letter of reference is a good, or the only, option. The genre is just too flawed. A little destruction might help with the job of imagining a path forward. All hail the interregnum; it has to be better than sticking with the letter of recommendation as it currently exercises hegemony. Let’s ditch the letter.
