Leiter Reports: A Philosophy Blog

News and views about philosophy, the academic profession, academic freedom, intellectual culture, and other topics. The world’s most popular philosophy blog, since 2003.

  1. Fool's avatar
  2. Santa Monica's avatar
  3. Charles Bakker's avatar
  4. Matty Silverstein's avatar
  5. Jason's avatar
  6. Nathan Meyvis's avatar
  7. Stefan Sciaraffa's avatar

    The McMaster Department of Philosophy has now put together the following notice commemorating Barry: Barry Allen: A Philosophical Life Barry…

Huemer’s polemic against letters of recommendation

Brian Leiter

David Boonin (Colorado) kindly calls to my attention his colleague Michael Huemer's case against soliciting letters of recommendation.  Prof. Huemer is always pleasingly contrarian, though it seems to me his case is stronger in this instance when it comes to letters of recommendation for senior candidates rather than rookies.  With regard to rookies (1) no appointments committee has time to review in detail the submitted work of all applicants; (2) pedigree (the applicant was trained at a program with which one has had good experiences) is both under- and over-inclusive for narrowing the applicant pool down; (3) the same goes for publications; and (4) letters can be quite helpful, as long as one knows how to read them.  If one does appointments enough, one gets better at deciphering the codes in which they are written, and you learn who is reliable and who isn't (in the late 1980s and 1990s, Hilary Putnam wrote one too many time in letters of recommendation that "now I know what it was like to have Wittgenstein in class" that it became a joke in the profession–none of those young philosophers, to the best of my recollection, turned out to be major figures in philosophy). 

Leave a Reply

Your email address will not be published. Required fields are marked *

12 responses to “Huemer’s polemic against letters of recommendation”

  1. Robert Gressis

    OK, let's abolish letters of recommendation (Huemer's case seems pretty solid to me). What, then, should we use to determine whether or not a person should get a job? It seems at least part of the best answer will be: look at the writing sample. But we don't have time to read 200 twenty-page writing samples. So, perhaps writing samples should be shorter? Or perhaps candidates should send their writing sample, but also a five-page summary of their writing sample?

    Reply
  2. Jason Brennan

    Last hiring round: 520 applicants, average 3.5 letters per applicant, average letter length 2.5 pages. (Forgive me if these are mere approximations; I did not in fact do the math.) So, about 4500 pages of letters to read, if we read them all.

    But we don't, of course. We use other methods to cut the 520 down to a manageable amount, say, 70 who look promising. These get further scrutiny. At this point, you might read a bit of each person's writing sample, and if it seems good, finish the entire thing. If 3 pages in the person still hasn't gave a coherent thesis or account of why the paper matters, move on. Etc.

    It's not clear to me what the letter add, other than that apparently everyone two years ago had the best grad students they've ever had.

    Reply
  3. Brian Leiter

    I realize Jason is making a joke about "everyone two years ago had the best grad students they've ever had," and there are some repeat offenders who use lines like that, but I want to assure job seekers (and letter writers) that most do not. I'm curious how Jason & colleagues cut down from 520 to 70 without the letters. One can read portions of letters and still get enough information for trimming purposes. And of course I read more carefully for those I know from experience to be valuable recommenders.

    Reply
  4. Michael Huemer

    By the way, social scientists have done studies of the reliability and validity of letters of recommendation. The results are not encouraging: https://orgtheory.wordpress.com/?s=letters+of+recommendation

    Reply
  5. Brian

    Mike, a lot of the evidence comes from the professional school context, where I am not inclined to defend the relevance of LOR. And a lot depends on how they are written. The basic fact that every student should know is that if Distinguished Expert X says the student is "the real deal," and Distinguished Expert X doesn't have a Putnam reputation, then that letter matters. Everyone in philosophy knows this!

    Reply
  6. John Gardner

    All this seems to reflect badly, not on letters as such, but on the way they are written and read in the United States. Except when writing to the United States, one is writing a letter of reference, not a letter of recommendation. One is writing for the benefit of the appointing department, not for the benefit of the applicant. A letter that does not explain what is good and bad about the applicant, and what makes her/him suitable or unsuitable for the particular appointment, is immediately suspect. I would not write such a thing – except to the US, where I know that any expression of reservations, far from showing the attentiveness of the letter-writer, is treated as fatal to the applicant's case. So when I write to the US I have to change my approach. I have to convey my balanced opinion using exquisitely subtle modulations of my superlatives rather than open exploration of the pluses and minuses of the candidate. This is a ridiculous situation. Better no letters than this. But better sensible critical letters than no letters. (By the way: if my critical letter will lean towards an overall negative verdict, I advise but do not require the student concerned to look for a different referee. We owe praise to nobody but the praiseworthy.)

    Reply
  7. Steven Hales

    I like the biannual (or so) ritual of chastising ourselves over hiring heuristics. There's a satisfyingly familiar pattern to it, like the Liturgy of Self-Doubt.
    Reverend: We should stop
    Congregation: using letters of recommendation
    Reverend: They are unreliable, and full of inflated adjectives. We should stop
    Congregation: looking at graduate school prestige
    Reverend: It is unreliable, and disadvantages the underprivileged. We should stop
    Congregation: APA interviews
    Reverend: They disenfranchise the poor who cannot attend. We should stop
    Congregation: expecting graduate students to have publications
    Reverend: It clogs the journal system and abandons late bloomers. We should stop
    Congregation: Hiring on buzz and promise
    Reverend: It is biased to the well-connected. We should stop
    Congregation: Deciding primarily on writing samples
    Reverend: No one can read the hundreds submitted or has infallible judgment of quality. And all these methods offend the holy trinity
    Congregation: Race, Class, Gender.
    All: Amen.

    Reply
  8. alexander stingl

    I have given one or two talks about the problem of "academic social capital" and reference letters appear in this context, in my view. I have provided a half essay half draft for a section from an upcoming book here: https://alexstingl.wordpress.com/2016/02/26/matthew-showering-the-burning-of-academic-social-capital-and-economies-of-relevance-of-funding-institutions/
    I am not per se against the use of reference letters, I am against the practice for the process before short-listing, because I think that one can plausibly argue that the practice may reinforce social inequalities. I think that upon short-listing, it can make sense to have input from people who know the candidates. I have noticed, at least for sociology, that this year some universities have begun to try that practice – I actually think that Drexel was one of the first about two years ago, if I recall either for a soc or an sts search. A second thing has to do with the "secret language" of these letters – I recall that Brian has (re)posted some helpful advice here. The problem is precisely, too, that – as a candidate – one not only needs letter writers, but also letter writers who know how to write these in the "proper vernacular" (I know that the practice is quite similar on the German job market, not only in academia, but widely. There are guides on how to write the so-called "Arbeitszeugnis", a kind of report your former employer writes for you, when you leave a German company – either voluntarily or not -, and which has a similar "vernacular", which makes me wonder if anyone has written an international history on these "jargons" and their uses). Now, this vernacular is something that is often learned informally and implicitly, and one can already suspect that this, too, may replicate inequalities. So, besides the removing these letters from the initial stage, I think it would be good to consider letters that are written in a more open style. I cannot speak for philosophy here with the same strength as I can for sociology, but I have had conversations with philosophers on this as well, who confirmed, thus at least anecdotally, that they don't take seriously letters, resp. the referred to candidates, if they are not using the "correct jargon".

    Reply
  9. Michael Huemer

    I suspect Brian Leiter is more familiar than most of us are with the "code" of letters of reference, and the characteristics of individual letter-writers.

    I have been in the profession for 19 years, but I do not know what the code is, nor do I have any idea who is or isn't a trustworthy writer. Nor have I ever tried to write in code. I suspect that most people are more like me than like Brian Leiter. If you're a writer and you think that you're communicating something with code, or you're a reader and you think that you're receiving a message in code, don't be so sure.

    Reply
  10. historygrrrrl

    Prof. Huemer's assessment is great. I've been on search committees and have tried desperately to interpret the "code". I'm also kind of old and kind of junior and kind of on the job market.

    The indicated problems explain, in part, my own conundrum in seeking references/recommendations. Do I ask my friends? That seems somehow dishonest. Do I ask famous people? That seems dishonest in a different way. Do I ask the philosophers who are best qualified to evaluate my work? That seems unlikely to produce a letter at all, since I do not know some of these people. If it were to result in a letter, it would likely be consistent with one of my primary values (honesty), but may not lead to the ultimate end (obtaining a job or fellowship).

    The current system, IMO, seems to privilege those who are shameless and extroverted, two qualities that are not necessary conditions for philosophical aptitude.

    Reply
  11. Aaron Smuts

    I've only been on three searches, but each received over 200 applicants. In my experience letters have been largely worthless. It's not just the hyperbole that's the problem. The bigger issue is that they are typically without helpful content. Most letters contain mere summaries of the applicant's dissertation. I see this as mere padding. And some writers use the same stock letter for multiple applicants on the same search. I saw 3 from the same writer last year that were 95% identical. It was embarrassing. Stop doing that. Please. The main value for which letters have proved useful so far in my experience is to help explain something in the application that may have raised worries.

    I'm not at all surprised that a pool of 520 could be reduced to 70 in a matter of hours without the aid of letters. It's amazing just how many applicants don't even come close to the appropriate AOS on some searches. . . . The real difficulties come when you get down to the large pool of serious contenders that are are shockingly well qualified. All I can say is that it's a rough market. Good luck.

    Reply
  12. Michael Huemer

    I would add that I am not saying that no letters are ever useful. No doubt some are valuable. But the valuable ones have to be weighed against the misleading ones, plus all the negative consequences of the practice.

    —–
    KEYWORDS:
    Primary Blog

    Reply

Designed with WordPress