Letters of Recommendation

Here’s an ethical issue (actually, a pair of them) which I’m sure all of us have faced, or will. Sometimes students come to instructors seeking letters of recommendation – for graduate school, for jobs, for postdocs, or what have you. And sometimes these students are such that we could not, without dishonesty, write a fully positive and utterly enthusiastic letter. But there seems to be an expectation on the part of a lot of people – not only students who request these letters, but also those decision-makers who will read them – that every such letter should be completely positive, so that a letter that contains any negative comment at all will simply doom the person whom it ‘recommends.’

In light of this, there are three obvious strategies. (I’m sure there are more, but these seem to be the most obvious, and probably the most common.)

(1) Write utterly positive letters for everyone who asks, except for those who are so utterly inadequate that one feels unable to recommend them at all.

(2) Write letters only for those whom one sincerely believes to be utterly outstanding, and to have no flaws whatsoever – the top one percent, perhaps. For them, write utterly positive letters.

(3) Write honest letters, reflecting a candidate’s positive and negative qualities, for everyone who asks, except for those who are so utterly inadequate that one feels unable to recommend them at all. Begin the letter with something like the following (I borrow this from Steven M. Cahn’s book, Saints and Scamps: Ethics in Academia, p. 50): “This is a candid recommendation. As such, it necessarily contains criticism as well as praise. Please read it in the spirit in which it is written.”

My policy has generally been to follow (3). (1) is deplorably dishonest; as Cahn writes, “Just as counterfeit bills destroy monetary standards and cause disorder in the economic community, so phony recommendations wreck academic standards and lead to babel in the community of scholars.” As for (2), while it would mostly relieve one of the burden of having to write such letters, it would, if generally followed, leave many students unable to obtain recommendations at all; on the other hand, if followed only by a minority (as is likely to be the case), it would simply leave the field open to the majority who tend more toward (1).

However, I was recently criticized by a colleague who prefers (2). He claims that, against the current background in which thoroughly positive raves are de rigueur, what I take to be an honest evaluation is in fact not honest at all; since (he claims) to express any criticism in a letter of recommendation is to destroy the student’s chances, doing so is tantamount to stating that the student is not worthy of recommendation (whereas the students in this class are meant to be ones whom one regards as worthy of recommendation, but not perfect). Moreover, since many students will simply withdraw their request for the letter when informed of the policy, my colleague claims, (3) in practice tends to reduce to (2): one tends to end up writing letters only for those students who can be reassured that the letters will be completely positive.

I’d like to know how others handle this. Before inviting comment, let me very briefly raise an additional issue. Suppose one has a student who performs well in one’s philosophy classes, but who is morally reprehensible in some way – say, for instance, she holds reprehensible views with respect to racial issues. Is an instructor entitled to refuse to recommend her on the basis of her views in this area? Or is she allowed, or perhaps obligated, to mention these views in the letter? Refusing to write a letter in this case strikes me as uncomfortably close to refusing to write a letter of recommendation for someone whom one simply dislikes; I’m not claiming they are strictly analogous, but they are similar enough to worry me. At any rate, this is not a problem for my colleague, who claims that the mere fact that I dislike a student (whether for moral or more idiosyncratic reasons) entitles me to refuse to write her a letter of recommendation, even if I cannot provide any reason beyond my personal distaste.

12 Replies to “Letters of Recommendation

  1. Troy-
    (3) only collapses into (2) if students know what you’re likely to write (or fear that you won’t write what they’d hoped you’d write). So why inform your students of your policy of following (3)? I don’t see how a student has the right to know what sort of approach you take in writing these letters. These letters are meant to be confidential, after all. I don’t mean that you ought to be less than candid if a student asks how you approach recommendation letters, but I certainly don’t see how we are under an obligation to volunteer that information to students.
    But I also think that (2) collapses into (1), at least as regards its practical effects. Isn’t the point of these letters to enable those with decision making authority to make differentiations, however minute and fine-grained, on the basis of which jobs and other opportunities will be allocated? The content of the letters written under (2) will be identical to the content of these written under (1), thus undermining the possibility that the necessary distinctions among candidates can be made. Admittedly, since you’ve restricted (2) to the top one percent, letters that prevent making these distinctions do not generate a significant injustice, since the top one percent of any group is likely to do fairly well in the end. But still, someone’s got to decide who, for instance, is admitted at the top grad programs and who is admitted with generous financial assistance to boot.
    In any case, it seems to me that the whole purpose of recommendations is collapsing in academia thanks to the routine overselling, puffery, and exaggeration that is now expected, even (as you note!) by the students themselves.

  2. As a newly admitted grad student, I must admit that I have to prefer option 4.
    That is, provide honest evaluations. But also be honest with the student when they come to ask for a recommendation. You should provide teh student with some idea if you can’t write a utterly positive review.
    Why? Because with rec inflation, a moderate rec review can be quite damaging to an admission portfolio.
    I see no problem with writing a mixed rec if you make it clear to the student that that is what you would have to do.

  3. I wanted to explain my position a little better.
    One of the things that struck me about philosophy admissions was the uncertainty I had about how strong a candidate I was. A significant part of this was that I had very little or no clue what my recommenders were saying about me. And this can be problematic for people applying to grad school, since we often have limited financial resources and time.
    I once had a conversation with a professor at the University of Chicago which I assume is representative. He told me flat out, “We are looking for raves.”
    Now, I would not be offended if one of my recommenders told me that he/she could not write a rave. However, wouldn’t it be nice to know if I was going to waste my time by applying?
    I think the way to go is to say straight up: “I am going to write an honest recommendation which will not be an absolute rave even if I will, in general, be quite positive. Can you live with that kind of recommendation, given the programs to which you plan to apply?”
    I am genuinely curious why this isn’t ordinary practice.

  4. Troy,
    I think a common 4th option is (essentially) to damn through faint praise – “Joey regularly attended class sessions, and tended to be quite punctual”. Perhaps this is a variation of option (1); there are no explicit negative comments, just a dearth of positive comments on matters of real importance. Or would you consider these a form of option (3)?
    I think this 4th option captures those who will ‘recommend’ some students, ‘highly recommend’ others, and so on.

  5. As someone who has never had to write such recommendations or had occasion to read them, I have to ask this: What possible value can recommendation letters have if they are all unrelenting praise? If you are trying to decide what grad students to admit, for example, how can letters of recommendation help at all? If they are all (or most) nothing but praise, then the only variable left is the status of the name of the person doing the recommending, unless I am missing something here.
    When one is in a dysfunctional system, it is hard to wage a battle against it alone. So it seems that writing a good letter or declining to write one is the only real choice you have – the dishonest letter is just not right, regardless of the practice of others. But how about this as an idea: write TWO letters. Identify one as the letter you are writing based on the current expected standards of what a letter is supposed to be like – all good without reservations mentioned. Identify the second one as the letter you wish you could write and wished others would write. Maybe a bit of subversive activity like this will lead others who don’t like recommendation inflation to do the same, in the end resulting in more realistic recommendations to become more common. Or they just might think you’re an idiot and ignore any recommendations you make. Your call. (:

  6. As a graduate student, I have already written a few letters of recommendation for my students — none that were meant for anything like a permanent job or graduate school, of course, but letters for summer internships, scholarships, and study abroad programs. Until I read Troy’s post, I thought I had handled these letters well. Now I am suddenly worried that I doomed my students to failure.
    I assumed from the very start that I would have to write very positive letters if they were to be of any use to the students. But of course, it is also the case that most students who ask for letters — especially most who would prefer a letter from their TA rather than someone of more official standing — were not the best student in the class, and I was not prepared to endorse them as such. What did I do, then? I emphasized their most prominent virtues, and omitted the rest.
    As I saw it, these students weren’t applying to graduate school — they were applying to a summer internship or a study abroad program. Every such program must be flooded with less than stellar paper writers. And surely the people who would read my recommendation letter would want to know much more about them than the quality of the philosophy papers they wrote. But fortunately for my students, there are many more virtues that a teacher is uniquely positioned to observe (many students show a great deal of curiosity and motivation, an ability to take in different points of view, a willingness to listen to criticism and adapt, and so on). I recounted specific stories – noting that P often visited me during office hours, for example, or that Q often gave me drafts well in advance of the deadline. If they were also among the best four or five students in the class, I said so. But otherwise I found I had plenty of positive things to say.
    My question for you is this: by writing these sorts of letters, did I inadvertently condemn them to failure? Does the inflated standard of puffed-up recommendation letters demand that I supply praise that they do not deserve? I had assumed, perhaps naively, that it was enough to omit the areas in which they had not excelled, but I fear now that I might actually be expected to ascribe talents to them that they do not actually have.

  7. While I agree that there’s a problem with recommendation inflation, I’m not sure the problem is as bad as some of you suggest. It’s at least sometimes possible to tell the difference between a genuine rave and a pro forma rave. Most people write letters for multiple students, and have some interest in preserving some credibility as a letter writer, so that when a truly excellent student comes along, they will be able to help them out. (Not everyone, of course.)

  8. I do something of the sort that Patrick recommends above. I tell students if I think I could not write them a letter that would do them any good. In fact I recommend to students that they ask for letters in a way that makes it easy for professors to say no. I also try to guide them to apply to programs that are apt to admit someone of their general abilities.
    For what it is worth, most of the students who do ask me are fit for what they are asking me recommendations for. (Most of my students do not go on in philosophy) Those who want to go on in philosophy invariably belong in at least a good program if not a top program. I try to give them letters that are as positive as I can while also such that if they were accepted on my say-so the department would not think that I had misrepresnted their abilities.
    Within this general approach there is still quite a bit of room for stronger and less strong letters. Put any two of my letters next to one another and you would likely be able to tell who I thought was better, or at least who was better at this and who was better at that. I believe that I provide enough information that departments will be able to figure out from my letter in conjunction with the other evidence what sort of a student they would be getting. Remember that a credible letter includes detail, so that there can be a variety of different good letters and some can still be stronger than others. It also helps to be in a program which has successfully placed people into good programs and into the profession so that one can make comparisons with people they know. So I think that Ben is right in his comments immediately above.

  9. Let me say something from the other side of the map. My policy on writing letters has been greatly informed by the letters I’ve *read*. Having read quite a few letters, both of potential grad students and of potential faculty members (about which nothing’s really been said yet on this thread), I think the straight-up honest letter actually does *more* for the good students than do the puff-piece raves. When I come across a flat-out rave, with nothing critical at all about the candidate, I tend to become extremely suspicious and take the positive claims with a grain of salt (unless I know the author well enough to make a solid judgment about its authenticity). When I come across a letter where the writer was unafraid to point out some weaknesses in the candidate, I tend to take it far more seriously — the positive comments then really *mean* something, in other words, and I feel far more confident that we won’t get a nasty “surprise” from the candidate if admitted or hired.
    So I insist on the straight-up honest approach in *writing* letters, and I’ll usually say something right up front commenting on the general atmosphere of praise-inflation in these sorts of letters and how I want to provide a robust description of the student whom I am nevertheless *recommending* to the interested parties.
    On the other hand, if I feel there’s truly nothing to recommend about someone, it’s invariably because I just don’t know that student well enough (surely there’s *something* positive to say about someone who wants to do further study in philosophy!), and so I will decline to write a letter for such folks and offer that (“I just don’t feel I know you well enough”) to the student as the explanation.

  10. Let me add a further comment to supplement David Shoemaker’s post. After you have been teaching for any significant length of time, you will find that you are recommending students to the same programs year in and year out. If many of your letters are raves, you don’t do an able applicant any favors, because your letters will tend to be dismissed. (Years ago we had a position and received letters from a well-known philosopher who raved that each of the three students he recommended were the best he had ever taught! His letters were worthless.)
    One thing that you can do is write, in detail, about the abilities and perceived potential of an applicant along many different dimensions. This helps those reading the letters to have a more rounded picture of each applicant: all may be deserving of a place in a graduate program, but it doesn’t mean that they are all alike! Think, for example, of four or five distinguished philosophers you know well and imagine writing letters of recommendation for them. It would be highly unlikely that you’d say the same positive or negative things about each.
    I have also found it useful to “compare and contrast” students who have attended the same graduate program. Admittedly, you can’t do that much when just beginning to teach, but it happens fairly rapidly. Further, you can — as many graduate programs ask — say where an applicant stands in relation to other students generally, other philosophy students, either for this year or the past decade or among recent graduates.
    Remember, too, that our responsibility isn’t just to our students but to our colleagues and to our vocaton. Here it is helpful to bear in mind what you’d say to a friend and colleague at a graduate institution if she phoned you and asked, “Tell me, is Bloggs really as good as you say?” If you can’t say “Yes” without blushing, perhaps you shouldn’t say it in a letter to a graduate program where you don’t happen to know anyone.
    Finally, you often do a student a favor by saying, “Look: while I think you’d get something valuable out of studying more philosophy, perhaps at a Master’s level, the competition is extremely keen at the Ph.D level, so while I think you’d succeed if you got in, I can’t write as strong a letter as I think you’ll need.” Students are amazingly resilient, and if you are straight with them, they’ll listen to you. It doesn’t mean that they will agree with you. They might well think “I’ll show him just how wrong he is!” And if they do, that’s great. But they might choose a different career and become highly accomplished in it. Philosophy is wonderful — but so are many other endeavors.

Comments are closed.