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…

Tenure Letters: Some Questions

Brian Leiter

MOVING TO FRONT FROM YESTERDAY, in hopes of generating some more comments!

====================================

A philosopher writes:

If your blog has not done so recently, I wonder if you would be willing to start up a discussion about the ethics of writing tenure letters. I would very much like to hear other people’s opinions about this. I have heard that it can be held against a candidate if people decline to consider their tenure case. Is this true? Do people think it to be encouraged? I don’t see why it should. How much pressure do people feel to not decline to consider a tenure case? Also, has anyone found a good solution to the question of how to be able to say that one has done this hard work without outing oneself a letter writer for a particular person?

Good questions, probably of interest to many.  Here are my own impressions:  (1)  many schools want a reason if you decline to do a tenure review, usually because their Administration wants one–this is, I assume, because declining to do a tenure review is often (mis)interpreted as having a negative view of the candidate; (2)  my own practice, which may well be idiosyncratic, is to only do tenure reviews in cases where I am interested to study the candidate’s work more carefully–sometimes this has led to favorable reviews, sometimes not; (3)  I simply list on the CV the names of schools that have asked me to do formal reviews for promotion; I don’t list dates, though I suppose one could draw inferences about the candidates given the likely AOS for which I get asked to review candidates.  But since in most cases there are multiple reviewers, I’m not sure this is especially revealing information.

No anonymous postings, and please post only once–comments may take awhile to appear.

Leave a Reply

Your email address will not be published.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

10 responses to “Tenure Letters: Some Questions”

  1. kid bitzer

    "declining to do a tenure review is often (mis)interpreted as having a negative view of the candidate"

    the inference from silence to condemnation is surely fallacious in any particular case, but surely more solid in the aggregate (which is the level at which deans and provosts have to work).

    people asked to review tenure cases have very limited time. in addition, they normally take no pleasure in criticizing very weak or unqualified cases. speaking ill of a professional colleague is something one does when forced to, but one is not forced to write tenure reviews.

    accordingly, people are more likely to agree to requests if they have good things to say, and by standard bayesian calculi, they are more likely to have bad things to say if they decline.

    knowing that administrators will read my refusal in this way, i decline only when i think the declining will do less damage than what i would be required to say if i agreed.

    sorry–that's not quite right. i will also sometimes decline when i am simply overwhelmed, even if the candidate is tenurable. but then i make sure to include in my letter a full and circumstancial explanation of my inability to write, along with an extra large dose of effusive praise about what nice things i would have said if i had had the time.

    Reply
  2. Keith DeRose

    Doing a tenure review is a big task. I estimate it's roughly equal to refereeing 10 papers. (I usually end up reading more than 10 papers worth of material, but the written evaluation isn't near to 10x as long as in refereeing a paper.) And it's higher stakes: Whether a candidate gets tenure is much more important to her than whether she gets a paper into a particular journal. For me, that makes the task more unpleasant.

    I don't put these on my CV. But I generally include fewer kinds of items on my CV than others seem to. (For example, I also don't list the journals and publishers I do refereeing work for.) I don't know how well your strategy would work for others, Brian. Most philosophy departments don't do all that many tenure reviews, maybe averaging only one every few years. Using areas of specialization and competence (yours & the various tenure candidates a place has had in recent years), it might be pretty easy for someone to figure out whom you were evaluating from your listing that you did a tenure for "X College" — especially if you have your CV posted on-line, and someone can notice when that line is added to your CV. (& I do strongly encourage everyone who teaches in a dept. that has a graduate program to post their CVs on-line, for the benefit of potential graduate students.)

    Reply
  3. Anonymous

    In my own experience, evaluators do interpret refusals to write tenure letters negatively. The assumption (fair or not) is that an expert in the candidate's field should already be familiar with the candidate's work, and have a high opinion of him or her, in which case writing a tenure letter should not be an undue burden. If a potential evaluator gives a specific reason for declining (ideally with a disclaimer that her or his refusal does not reflect any negative opinion of the candidate), then of course that will not be held against the candidate. But if a significant number of potential evaluators decline without giving a specific reason, that will often be seen as a mark against the candidate. One clear moral to be drawn is that if you are asked to write a tenure letter and cannot do so, be sure to provide a specific excuse, and if you have a generally positive opinion of the candidate, say so.

    I do not list tenure reviews that I have done anywhere on my CV, even just by school. It would be relatively easy to infer who I wrote letters for. I doubt that such service would count for much of anything if I did include it. (Likewise, although I include information on my CV about journals, academic presses, and funding agencies for which I have served as a referee, I doubt that anyone looked at this when I was up for my most recent promotion. Sadly, professional service doesn't count for very much.)

    Apologies for the anonymous posting. My school has a very strict policy on leaking any information about tenure cases.

    Reply
  4. Sandy Goldberg

    I think that all people familiar with contemporary academic culture — faculty and administrators alike — should, and typically do, recognize the various reasons why one who is approached to write a tenure letter might decline the invitation to do so. Often the reasons are quite innocuous (too busy, not enough background familiarity with the subject-matter of the tenure-candidate's work, existing commitment to writing x other tenure-letters this year, etc.).

    On the few times I have served on an in-house administrative committee evaluating a tenure file, no one on the committee drew any negative inference from the fact that several people declined to write letters. In most cases those who declined offered an innocuous reason for doing so (and this was made part of the dossier itself). But my sense is that, even without the innocuous excuses, no negative inference would have been drawn unless a disproportionately high percentage of those who were approached declined to write. (In each of the cases in which I participated, the % who declined was a bit less than 50.) I don't know whether my experience is representative; but there you have it.

    Regarding getting credit for writing tenure-letters, my practice has been to include it in the annual review material sent to the Dean of my home institution, but to leave it off of the CV (which, since it is available on the web, is a more public document).

    Reply
  5. Ralph Wedgwood

    The ethics of writing tenure letters strikes me as a distinctly thorny issue. Part of the problem is that as soon as one is asked to write a tenure letter, then however one responds, one's response will be reported to some Dean or Provost or President, who will come up with some interpretation of one's response, and probably give this interpretation non-negligible weight in deciding the fate of some young philosopher.

    One distinguished senior philosopher once told me that he always declines to write tenure letters, on the grounds that he disapproves of the whole system. In his view, the decision should ideally be based solely on the opinions of the members of particular department involved. After all, the members of the particular department are philosophers, and so (unlike all the Deans and Provosts and Presidents) they have the necessary expertise to judge the quality of the candidate's work; and they also have a compelling incentive to make sure that the right decision gets taken (since if they make the wrong choice, the mistake will affect them much more seriously than anyone else). In his view, this would be a more rational approach than what happens at many universities, where enormous power lies in the hands of Deans and Provosts and the like, even though these administrators have little basis for their decisions other than their parsing of the text of these letters — which are in turn written by outsiders who have a significantly less compelling incentive to ensure that the right decision is taken.

    Although I had considerable sympathy for this distinguished philosopher's view, I have in fact acceded to every request to write a tenure letter that I have ever received, on the grounds that in each case, the candidate was a philosopher whose work I already knew and admired — and so was in effect someone whom I already believed to be deserving of getting tenure. But I haven't yet succeeded in devising a policy that I'm happy with for how to respond to requests to write a tenure letter about someone whose work I either don't know or don't admire.

    I've occasionally wondered whether I should adopt this distinguished philosopher's approach after all, and just decline all requests to write such letters that I receive from now on. But given that I can't change the system, it seems preferable to do what good I can within the system rather than just being a refusenik. I'd be very interested to know what readers of this blog think about this issue.

    Reply
  6. Rob Rupert

    "accordingly, people are more likely to agree to requests if they have good things to say, and by standard bayesian calculi, they are more likely to have bad things to say if they decline."

    This sounds like a base-rate mistake to me. Say there are a hundred potential reviewers, 80 positive and 20 negative. All of the negatives decline. Twenty of the positives decline. In this case, it is surely true that people are more likely to agree if they have good things to say (75% acceptance rate versus 0%); but they are _not_ any more likely to have bad things to say if they decline (among the decliners the split is 50-50).

    Although my numbers are artificially tidy, it's pretty plausible that some cases break down along roughly these lines.

    Reply
  7. Rob Rupert

    I should have emphasized a disturbing bit of irony. Cases in which the base-rate of negative evaluators is low are cases in which, other things being equal, the tenure decision should turn out in favor of the candidate. But, these are just the cases in which people (including deans) are likely to commit the base-rate fallacy and misinterpret someone's declining.

    Reply
  8. kid bitzer

    rr:

    the question is not whether there are "some cases" e.g. fred's case or jane's case, in which the numbers go the other way. doubtless there are. but no individual's population of potential reviewers can give the base rate of interest, because no dean or provost can know those facts about that individual's population of reviewers.

    the question that deans-and-provosts have to ask is whether, knowing only what the patterns are for the profession as a whole–indeed, for academia as a whole–the conditional probability of (has a negative opinion, given that they declined to express one) is above 0.5.

    if p(declines, given negative) is close to 1, and p(declines) is close to 0.5 (as e.g. sandy goldberg suggests above), then p(negative) can dip close to 0.25 without driving p(negative, given declines) below more probable than not.

    this is consistent with your story, right? you managed to balance out the evidentiary weight of declining to "not more probably negative than not" by driving a single case's negative rate down to 0.2. if it drifts above 0.2, then the numbers support the prejudicial inference.

    so the question that administrators have to ask themselves is: out of every 100 potential tenure reviewers, not for this candidate but in academia as a whole, what's the percentage of those who have a negative opinion in any given case, whether expressed or not? that's the base rate that matters.

    ponder that with your d-or-p hat on, and pretty soon you're asking "well, what percentage of tenure cases, at the national level, really *ought* to be successful?"

    and now notice that your answer to this, as a d-or-p, will probably have a lot to do with your own institution's standards. if you think that most tenure cases, by the time they come up, should probably be successful (in general or at schools like yours), then you will tend to low-ball p(n), and so give lesser weight to silence. if you think that most tenure cases (in general or at schools like yours) should probably be unsuccessful, then you'll probably work with a higher p(n), and accordingly give more weight to silence. and in between for in betweens.

    (and remember that tenure reviews sometimes say, though not in so many words, "i would not advise tenuring jones at grand university x, but i suppose at your modest institution you ought to be content to have jones' services".)

    so the base rate of interest is not only not fred's or jane's, it is also not exactly the empirical numbers, either. rather, it is the d-or-p's own beliefs about the base rate for reviewers of the sort that their institution uses, making judgements that are relativized to tenuring at the d-or-p's institution.

    it's a mess, i grant you that. but i don't think my original comments involve any base rate mistake on my part. perhaps d-or-p's are making base rate mistakes left and right, but even so the illocutionary force of declining will remain the same, and for the same reasons.

    Reply
  9. Brian

    I am intrigued by Ralph's report of the distinguished philosopher's approach, which I am sure no serious university would accede to. Administrators know all too well the risks of inbreeding, whether by pedigree or methodological approach or substnative doctrine, and so, at least at the better universities, they want evidence that the candidate in question is not simply the local toady and water-carrier but actually someone whose work commands respect elsewhere. To be sure, this process has serious drawbacks: it encourages intellectual timidity and conformity, even in philosophy, which is rather hard-edged as the humanities go. But all-in-all, I think universities are right to want evidence that the candidate for tenure is not just the toady of the local powers-that-be.

    Reply
  10. Michael Rosen

    *Do people decline to write because they have reservations about the candidate's work?*

    I suspect that this is actually rare at the tenure review level. If a junior person is someone about whom the senior scholars who are asked to do tenure reviews have a strong opinion one way or another before carrying out the review then they already have a high enough profile for there to be a presumption of tenure at most institutions.

    *Should people feel obligated to write tenure reviews whatever their feelings about the candidate?*

    It depends on your views about the ethics of tenure. If someone asks me: should I employ X (over Y or Z)?, that is one thing. If you ask me: should I continue to employ X?, that is, I think, another. If you think that the two questions have different answers but that the tenure review presumes that the answer to the second question is the answer to the first (and you're not prepared to massage your views) then it seems to me that non-participation is ethically justified.

    Reply

Designed with WordPress