Three Reasons Why We Should Not Request Letters of Recommendation for Job Applications

A standard element for a job application for a philosophy tenure-track position is three (or more) letters of recommendation, requested up front. Just glancing through this year’s PhilJobs, I see tenure-track positions that require “three recommendation letters,” “three letters of reference signed and dated within the last six months. At least one letter should speak to your teaching skills,” and “Candidates should arrange for three confidential letters of recommendation to be sent separately.” For each of these positions, hundreds of applicants will apply. Most of them will have at least three glowing letters of recommendation, most of which will barely get a glance from the search committee. It is a waste of everyone’s time.

As our department’s placement director, I review all elements of the job application process with our graduate students, stressing the evidential value of each document. For example, the research statement aims to show that candidates have a viable research project that will lead to enough publications to earn them tenure. The teaching dossier demonstrates that they are competent instructors with innovative practices, and includes pedagogical justifications for those practices. The CV indicates whether candidates meet the requirements for the job, and shows in an unadorned way what they’ve accomplished. The diversity statement enables the search committee to identify whether candidates have skills, experience, or at the very least willingness to enhance campus diversity and equity efforts. But I don’t see any such clear purpose for the letter of recommendation.

The time has come to stop requesting letters—at the very least at the stage of application—and perhaps to dispense with them altogether.

Search committees in many other fields already do this: They want references only when a long list or shortlist is compiled. For example, the American Anthropological Association recommends that “Solicitation of letters of recommendation should occur only after an initial screening of candidates to minimize inconvenience to applicants and referees.” Even in philosophy, letters of recommendation for senior positions (at the associate level or above) are often only required at the shortlisting stage. For example, Washington University in St. Louis is hiring this year and requires “at least three letters of reference” for assistant professors but only “a list of references” for associate or full professors. In the UK job market, it is customary to ask for references only at the shortlisting stage.

Perhaps we feel writing recommendations is just one of those things we have to put up with. But we have the power to shape disciplinary practices in all sorts of ways. Consider the first-round interview at the Eastern APA meeting. Presently, only a tiny number of schools (maybe 1-5 percent) still do APA interviews, while everyone else uses Skype or brings candidates directly to campus. This change was not externally enforced but happened gradually through institutional culture. While there are downsides to this practice (e.g., Skype technology can fail), it saves an enormous amount of money and effort. Similarly, the costs of letters of recommendation are large, both for letter writers and for applicants.

(1) Letters of recommendation are time-consuming for everyone involved.

If you belong to a large department with a big cohort of grad students, you will write letters for several candidates. Given the huge number of applicants per job, this demands a great many hours of work. Should you make comparisons? How can you be specific without harming candidates by (unwittingly) raising doubts? What if there is a problem with a candidate (e.g., in terms of their ability to finish up the dissertation), and how candid should you be about this?

The following tweet by Kevin Zollman, a philosophy professor at a research-intensive institution, captures this worry. (Note: Zollman thinks that letters of recommendation are still useful at the initial stage, and that it’s good that “faculty use different methods for evaluating candidates,” as can be seen on this thread.)

Job candidates are also burdened by the letter requirement. Very often, they cannot feasibly apply for jobs when the deadline is very short, even though they are a good fit for the job, because it is not possible to line up letter writers in time. Candidates who are several years out of graduate school and unable to land a tenure-track position typically find themselves with a diminished network and have to scramble to find letter writers. In the job market today, it is not uncommon for people to be three or more years on the market post-PhD. The letters of the supervisor and other committee members will be less and less up to date, and candidates in precarious positions (e.g., adjuncts) do not have the resources to update their network (e.g., by attending conferences). I’ve also heard, in my capacity as mentor for several projects (such as The Job Candididate Mentoring Program for Women in Philosophy), that candidates in precarious positions do not apply for jobs that fit them because they don’t want to keep on bothering the same letter writers.

(2) Letters of recommendation have little or no evidential value.

A common reason for why we should have letters at the stage of application is that they helpfully contextualize the candidate. American letters tend to be longer and more (uniformly) positive than, for example, letters from UK-based academics. I once attended a workshop on the US job market at Oxford intended to teach people working there how to write a (what was perceived to be) an over-the-top positive letter. British writers are more honest, giving both strengths and shortcomings of candidates, but that culture is changing too.

I’ve also seen and heard (especially in Europe, but  I’m not sure how widespread this is elsewhere) that letter writers request the candidate to write their own letter. This demand puts candidates in a terrible bind—it is very difficult to sing one’s own praises with conviction. There are even blogs detailing how to ghostwrite your own letter. The evidential value of a ghostwritten letter is zero.

Maybe letters can help search committees better contextualize a candidate’s research, in a way the candidate is not yet able to do. I’m not denying that such information is useful, but in an already large package of information that search committees during a first selection already have difficulties reading in detail, the added value is marginal. Candidates need to be able to pitch themselves and contextualize themselves anyway, in e.g., their research statement, cover letter, and other materials.

Letters might be useful in raising red flags (e.g., doubts about a candidate’s ability to complete their dissertation on time) but I am not convinced letter writers will, or can, be sufficiently candid to do this. Letter writers have described the process as a Byzantine art. One even said that they used code words such “solid” for what was actually a weak candidate, remaining confident that the recipient would know this shared code and be able to act on it. I’m not so sure the code is really that transparent or widely shared.

(3) Letters of recommendation perpetuate gender bias and prestige bias.

If letters of recommendation are used to spot red flags, it is worrisome that letter writers raise more doubts about women than men. A recent study in the Journal of Business and Psychology found that letters for women (regardless of the gender of the letter writer) contain significantly more negativity.  “She is unlikely to become a superstar, but she is very solid,” states one letter quoted in the article. “I think it would be fair to say that her record on paper would not place her among the top echelon of candidates for first rate programs,” says another. Unsurprisingly, such damning phrases lower the candidate’s chances with search committee members. Earlier studies have also found that letters for women, compared to their male counterparts, are shorter and less keen, and that letters for men contain more “standout” words that emphasize the brilliance of the candidate.

Moreover, letters for women (compared to men) contain more “grindstone” words such as “hardworking” and “diligent,” emphasizing that women, if they are excellent candidates, got there through hard work. Philosophers, more so than practitioners in most other disciplines, regard philosophical talent as innate and something one cannot acquire through hard work or teaching. If that bias is widespread, grindstone words could easily be taken as code for candidates who need to make up for their lack of innate brilliance with hard work.

Consider also prestige. People at more highly regarded institutions will be able to solicit letters from more famous writers, who are more likely to be part of their committee, or who are more likely to be in contact thanks to conferences and other networking events. Such letters carry weight, but it is not clear to what extent one should weigh them over and above the prestige of the school the candidate graduated from. (I’ve argued earlier that prestige should not count as a consideration in evaluating job applications, but even if it does, should it be counted twice?)

Let’s face it: The main reason we are still using letters of reference is because it is an ingrained practice. While prudence might require us to think twice before altering this tradition, letters of recommendation have high costs for both applicants and writers (let alone the search committee members who will need to read them). We can change this on the hiring side by asking for letters only at the shortlisting or longlisting stage, or perhaps by not asking for them at all. If more and more hiring departments follow suit, the norm will slowly shift. Or so I hope.

Helen De Cruz

Website

Helen De Cruz holds the Danforth Chair in the Humanities at Saint Louis University. Her areas of specialization are philosophy of cognitive science and philosophy of religion. Recent publications include De Cruz, De Smedt & Schwitzgebel (Eds.) Philosophy through science fiction stories (Bloomsbury, 2021) and De Cruz (Ed. and illustrator). Philosophy illustrated. 42 thought experiments to broaden your mind (Oxford University Press, forthcoming).