My goal of writing “Against Desk Rejects!” was to stimulate a discussion about an editorial practice that has stealthily crept into our peer-review process almost without any explicit debate. Without recounts and lawsuits, I am going to declare victory on that objective.Footnote 1
I face two alternatives in this reply to the symposium participants. By one choice, I would seek to rebut each author’s criticism. A second choice is to move the debate forward, letting my peers judge for themselves whether the arguments and analyses of the symposium contributors should be accepted, accepted with revisions, rejected, or even desk rejected. With two exceptions, I have decided to pursue the second choice.
The first point I want to rebut concerns the often-implicit assumption that it is easy to identify both “weak” or “half-baked” papers and the journal level that best “fits” the paper.Footnote 2 When I finish a paper, I often do not have a clue as to how good it is, and I typically must seek advice from friends and colleagues about their views on where I might submit it (their views almost always conflict). Furthermore, I am interested to know of any empirical evidence that to “aim high” in submitting a paper for publication is widely regarded by our peers as improper behavior and a “misuse of the journal-submission process.” Indeed, I believe totally in the peer-review process to sort out what is half-baked and what is fully baked. Precisely because there often is ambiguity about the quality and contribution of a paper,Footnote 3 it is essential that papers be subjected to peer review by subject-matter experts.
Second, I was warned in advance that my proposal for fees would be dismissed by some with the vacuous smear of “elitist”—and this did, in fact, turn out to be true.Footnote 4 Therefore, perhaps it is useful to remind readers that:
-
• Assumptions about research accounts at some institutions and not at others have not been empirically verified (to my knowledge) and ignore a wide range of individual circumstances. Neither are all research accounts exogenous.
-
• Last time I checked, the cost of a room at the Palmer House was around $250 per night and the registration fees for our conferences are typically in approximately the same range. It is easy to imagine that attending a single political science conference might cost as much as $1,000. Is this an elitist practice?
-
• The American Political Science Association (APSA) charges membership and conference registration fees, even for graduate students, that far exceed any submission fees that journals are likely to implement.
-
• According to the American Association of University Professors, the average salary for university professors in the United States in 2017–2018, combining all ranks, was $95,176.Footnote 5
-
• Recall that in my article I discuss possible mechanisms for providing submission fees for those for whom such fees would be a hardship. This is the context in which I raise the issue of submission and publication fees.
Before going further, it is perhaps useful to briefly review my argumentFootnote 6:
-
• My premise is that the frequency of desk-reject editorial decisions (i.e., rejections without peer review) has increased significantly in political science journals. At least one journal editor in this symposium confirmed this assumption.
-
• I did not argue that every manuscript that comes over the transom should receive a full review. There clearly are some papers that are not academic in nature and some academic papers that are so obviously defective (unbaked?) that they warrant no serious peer review. I do not know exactly how to state the criteria but, like the US Supreme Court and pornography, I think virtually all of us “know it when we see it.”
-
• There currently are two problems with our peer-review system: an oversupply of manuscripts and an undersupply of reviewers.
-
• The purpose of my article was to ask whether there are alternative solutions to these two problems—solutions that do not require editors to reject without peer review substantial proportions of the papers submitted to their journal.
-
• To put it mildly, not everyone loves all of my proposed solutions.
Therefore, I assume that current practices are not likely to change radically. Under this assumption, I use the remainder of this article to suggest “best practices” for using the editorial decision to reject a manuscript without peer review.
One issue that has become clear in this symposium is that editorial practices differ greatly across journals. Of the various complaints I raise, it undoubtedly is the case that not all journals are guilty of every “crime,” and it was not my original intent to imply that that was the case. A frequent reply to my critique of desk rejects is: “Well, that particular practice is not true of my journal.” I accept that. Having what I suspect is higher-than-average experience with a wide range of journals, I can attest to (and can provide evidence of) the veracity of most of my critiques for at least one journal (e.g., a desk reject after a manuscript was under review for more than two months). However, further addressing the frailties of the desk-reject system is perhaps not the best use of anyone’s time. Instead, I suggest the following best practices that all editors may want to consider:
The Desk-Reject Decision: If available, an editor who has clearly recognized substantive expertise in the subfield in which the paper is located should make the desk-reject decision. This, of course, is impossible at journals that do not employ field-specific subeditors. To the extent possible, more than a single person should be responsible for the decision to reject a manuscript without peer review.Footnote 7
Double-Blind: To the extent possible, the desk-reject deliberations should be double-blind.
Uncertainty: To the extent that the decision to desk reject a manuscript is not substantially obvious, the manuscript should proceed to full review.
Speed: Desk-reject decisions should be made as quickly as possible—and in no circumstances more than two weeks after the submission of a manuscript.
Justification: The decision to desk reject a paper should be accompanied by a clear statement of the reason why. This justification should be more than gratuitous, although boilerplate material could be developed—for example, “this paper does not fit with the substantive objectives of this journal, as clearly stated in [document X]” or “it is the policy of this journal not to publish papers that are heavily dependent on non-representative samples such as M-Turk.”
Transparency: Journals should state clearly that their review process involves two important decisions. First, the decision will be made on whether to grant peer review to a paper, based on a clear statement of the criteria for the decision. Second, for manuscripts that proceed to peer review, the process will proceed as follows. Regarding the latter, most journals already publish explanations of the process. For the former, it would be useful to authors to know the rates of desk-reject decisions, just as some journals report their acceptance rates.
Hidden Criteria: Although part of transparency, hidden criteria should be unhidden. If a journal has a bias against certain types of papers (e.g., formal theory, normative theory, audit studies, or the papers of authors who have not reviewed for the journal), the criteria should be explicitly stated.Footnote 8
Term Limits: Although I recognize that many journals do not have the luxury of choosing editors from a long list (N>1)Footnote 9 of highly qualified applicants, they should consider whether to impose term limits on an editor’s length of service. This policy might be more appropriate now than in the past, especially in light of the vast amount of discretion and authority vested in editors by the power of desk rejecting. It also might be particularly appropriate when single individuals, rather than teams, serve as the editor for a journal.
Like those who supervise the counting of election ballots, being a journal editor often is a thankless task, one that offers the “opportunity” of alienating 90% of one’s colleagues while pleasing only 10%. I have no doubt that virtually every editor tries to do what he or she believes is best for our science.
Like those who supervise the counting of election ballots, being a journal editor often is a thankless task, one that offers the “opportunity” of alienating 90% of one’s colleagues while pleasing only 10%.
In some ways, I regret focusing this symposium on desk rejects. In hindsight, I should have strictly framed the debate around two problems: too many submissions and too few reviewers. Rejections without peer review are seemingly a consequence of these two problems. I hope that future debate about our editorial practices focuses on ways to solve these two problems rather than specifically on desk rejects. As I acknowledge, my proposed solutions to these two problems might not be the best solutions. I have no doubt that fellow authors will be able to identify solutions that have eluded me. That is my hope.
