A senior philosopher elsewhere writes:
Over the years dossiers for applicants for junior positions have gotten larger and larger. Some dossiers we have received for our current search number more than 80 pages, One thing that is particularly striking to me is the number of letters of recommendation that each dossier contains. Back in the dark ages, when I was in graduate school, dossiers included on average three letters. In this year's search not a single dossier that I have seen includes that few. Many include six, seven or even eight letters. I don't fully understand the reason why, though. I am tempted to attribute at least part of the ever increasing number of letters to understandable anxiety on the part of job seekers. There seems to be a view that the more people you can get to say good things about you, the better off you will be. I certainly understand and sympathize with the anxiety. On the other hand, I doubt that there is a single search committee that is eager to read an average of six or seven letters per file. Nor do I think that more letters means more information or a more honest assessment of the candidate. Just for comparison sake, a typical tenure file at my own university will generally include around eight outside letters. Do we really need the same number of letters for a junior appointment as for a tenure case? Personally, I don't think so. But I am sure reasonable people can disagree on this score and would like to know what others think.
To stir the pot just a little, let me add the following. In some number of cases -- though I'm not prepared to quantify whether it be many, most, or just a few -- the additional letters come from people not from the candidate's home institution. Sometimes this makes some sense and is helpful -- if, for example, the outside writer has taught, supervised, collaborated with, or been colleagues with the candidate. Outside PhD examiners are often asked to write, for example. And you would certainly expect people who have moved a bit from place to place to have letters reflecting that fact. Or, in another vein, if a candidate has worked in somebody's lab, say, as sort of the philosopher in residence. All that makes sense and I have no beef with it. On the other hand, many outside letters seem to be the result of proactive networking on the part of candidates. Perhaps the thought is that it is important to have letters from prominent people in other departments with whom one has had some degree of contact. The thought may be that this networking will help increase and demonstrate the reach and impact of one' s work. In a few rare cases, such letters may carry some tiny bit of weight. But that, I think, really is the exception rather than the rule. In my experience, most such letters are relatively bland and pointless. They are neither particularly helpful nor, thank goodness for the candidates, particular hurtful. I suspect that it's an issue of divided loyalties. Seldom have I seen a letter from prominent philosopher P from institution X writing for candidate C from institution Y that directly compares C to P's own students at P's own institution. Most of what such letters add is generic praise. But there's usually enough of that in the "core" letters to suffice.
I'm not sure anything can stop the momentum for more and more letters. And maybe I wrong to think the momentum needs to be stopped. Maybe I'm just tired from reading thousands of pages of files. But I thought this might make a good discussion point for your blog.
I generally recommend candidates have not more than five or six letters, on the assumption that the 5th and/or 6th letters really add something--e.g., a letter from someone outside the department who is expert in the candidate's area and can add something useful about the candidate's work. What do readers think? Signed comments preferred, though grad students and job seekers may post with a pseudonym (pick a distinctive one).