The Ethical Obligations of Journals Vis-a-Vis Authors

A while ago, there was a healthy debate in the Proceedings of the APA about the amount of time that journals take to get a decision back to authors who have submitted papers. This discussion has thankfully been picked up here, over on Brian Weatherson’s blog, and we at PEA Soup think that it is worth furthering the discussion by endorsing the following proposals about this and other related issues:

(1) Journals should publish statistics on the time between the receipt of a manuscript and the notification of the author of the initial decision.

(2) Journals should publish statistics on the time between the acceptance of a manuscript and its appearance in print.

(3) Journals should make public the details of the journal's editorial procedures, as Analysis, the Philosophical Review, and the Australasian Journal of Philosophy do on their respective web sites. (These sites describe, for instance, the various stages of the review process, how decisions are reached, and whether the process is blind or not.)

(4) Journals should keep the requirements in preparing manuscripts for submission to a bare minimum. Authors should not have to ensure that their manuscripts conform to the journal’s style at the submission stage. Of course, some requirements that will make the review of manuscripts easier on editors and referees -- such as, double-spacing and wide margins -- are in order.

(5) Journals ought to notify authors of receipt of their manuscripts and where possible indicate the approximate time needed for evaluation. After this period of time, inquiries from the author concerning the status of his or her submission should be met with a prompt and courteous reply.

(6) In keeping with the Association of Philosophy Journal Editors’ guidelines on the handling manuscripts: “If articles are held for two months or more, letters of rejection shall normally include one of the following: a) the comments of the referees, b) a brief summary of the referee's comments, or c) the editor's reasons for rejecting the paper.”

(7) Where feasible, the review of an article should be double-blind. Neither the referees nor the editors charged with making a decision should know the identity of the author.

Let us emphasize both that these are merely proposals (to stimulate discussion) and that we take these ought-statements to be defeasible -- thus, we can imagine circumstances, where a journal ought not, all things considered, adhere to one or more of the above seven. Let us also note what should be obvious: these are prescriptive, not descriptive, claims. We readily acknowledge that many, perhaps most journals, are commendable with respect to our concerns. Now with these caveats in mind, let us provide some justifications for our proposals.

Justifications for (1)-(3): For many of us in the profession, whether an article gets published by a certain date affects our chances of securing employment, receiving tenure, or getting a promotion. Given the high stakes involved, potential contributors to a journal need to make informed decisions as to where they should submit their manuscripts. Now many journals, such as Ethics, annually publish some of their vital statistics (e.g., numbers of manuscripts received, manuscripts accepted, manuscripts accepted after revision, etc.). These statistics are crucial to authors in making informed decisions. But in addition to statistics about acceptance rates, it would be helpful if journals provided statistics about the length of time between submission and the initial and final decisions. Specifically, it would be helpful if journals would publicize in each case the mean, the median, the 25th and 75th percentiles, and the extreme long and shorts of the prior year. Also, it’s important for authors to know how their manuscripts will be reviewed and whether they will likely receive any comments back. Will the process be blind? Will it always be sent out to referees? In the event that it is sent out to referees, will the referees’ comments be passed on to the author? Many who are just starting out or who are at the beginning of their careers are still learning how best to write publishable papers. They are, for the most part, reliant on the valuable feedback that referees and editors provide to learn how to do this. Those starting out may also be concerned that their ideas will not be taken as seriously as the ideas of those who are well known to the editor or referees. Thus some authors will want to ensure that their manuscripts are blind reviewed. For all these reasons, we propose that journals respect authors’ rights to make informed decisions by providing them with the information needed to do so.

Justification for (4): The typical journal has an acceptance rate somewhere between five and fifteen percent. Clearly, then, most potential contributors will be submitting to a number of different journals (although, hopefully, not concurrently). It seems, therefore, unnecessarily burdensome to require authors to revise their manuscripts each time so that they conform to each journal’s style guidelines. It also seems unnecessarily burdensome to require authors to submit abstracts or biographical notes unless these items will be used in the review process. Thus we encourage journals to make or keep their requirements for submissions to the bare minimum needed to facilitate the review process.

Justification for (5): Letters, e-mails, and manuscripts can get lost or misplaced, and major delays due to the transfer of a journal’s editorship are known to happen. Authors should, therefore, be told upfront when they should expect to receive an initial decision, and this estimate should take into account any anticipated delays. This will give authors the opportunity to withdraw their submission if the anticipated delay is too considerable and will let authors know when they should inquire as to the status of their submissions. It’s terrible to wait patiently for five months only to learn that your manuscript has been misplaced and hasn’t gone out to referees yet. If only you had known that you should have received a decision two months ago, the error could have been uncovered then.

Justification for (6): If an article is held for more than two months, this normally means that it has been sent out to one or more referees. In this case, it should be possible to provide the author with the referees’ comments or, barring that, a brief statement of the editor’s reasons for rejecting the paper. Editors and referees provide an invaluable and largely uncompensated service, but this service should not be construed narrowly to include only the vetting of submissions. Just as important is the fact that editors and referees provide valuable feedback to authors that help them improve their papers.

Justification for (7): Most of us have some sort of bias or another. For instance, it would be difficult for most of us to evaluate a paper that is bold and suggestive the same way whether we knew it was written by some no-name graduate student or an internationally renowned philosopher. Even more worrisome is the possibility that some will have a bias against women or minorities. It seems best, then, to avoid any potential for bias by making the process double-blind where feasible.

Signed: Dan Boisvert, Josh Glasgow, Doug Portmore, and Dave Shoemaker

Comments

[Lindsay Beyerstein]

In in 1978 the International Committee of Medical Journal Editors drafted uniform submission guidelines for manuscripts submitted to biomedical journals.

I think that the editors of philosophy journals should draft uniform submission guidelines to standardize manuscript preparation and submission.

[Michael Cholbi]

Agreed to all seven proposals. Having recently started an electronic journal (www.historyofethics.org), I strongly encourage all journals to follow them assiduously.

Some additional ideas:
1) In the case of revise and resubmit letters, journals ought to include some statement about the likelihood (almost certain, more likely than not, perhaps, etc.) the manuscript would be accepted were the author to meet the referees' concerns. It can be difficult to read the gradations in enthusiasm from some of these letters.

2) Editors should provide to all referees a detailed document describing the journal's aims and its criteria for publication. In particular, the document ought to underscore the tone the journal wants the reviewer's comments to take., emphasizing that the reviewer has a duty to provide constructive, sympathetic, and critical feedback. Doubtless everyone in the profession has a story about callous, unnecessarily vicious, or unhelpful referees' comments.

Finally, a query, prompted by your remark: "Clearly, then, most potential contributors will be submitting to a number of different journals (although, hopefully, not concurrently)."

What exactly is the rationale for the near-univerasl exclusive submission policy? To my understanding, this is nearly unique to philosophy. IMHO, one of the great frustrations is the great imbalance in power between the journal and the author created by exclusive submission policies. The only justification I can imagine for them is that they ensure journals that any work they consider and accept can be published by them. Meanwhile, authors wait patiently, thankful that the journals deign to read their submissions at all and hoping (usually in vain) that their work will be published there. But why not a more open market for philosophical research? Suppose an author submits a manuscript to First-Tier Journal and to Second-Tier Journal. First-Tier dilly-dallies and by six months has not responded. Second-Tier Journal accepts the paper in three months. The author now confronts a choice: Take sure publication in Second-Tier or hope that First-Tier accepts at some point. That's up to her. But I cannot understand the justice of prohibiting this sort of justice. Sociologically, of course, I recognize that it would have First-Tier and its ilk worried. Yet there's something monopolistic about the present arrangement.

[David]

Re: Michael's question, what exactly are the potential penalties if a journal finds out that you have submitted a paper to more than one journal? Will they simply refuse to publish the paper, even if they would have otherwise? Will you be blacklisted forever? Also, how would journals find out you'd done this unless you told them?

[Dan Boisvert]

Re: Michael's comment. I believe the best justification for the "near universal exclusive submission policy" is to ensure that the limited resources of most journals are not wasted. These resources include the usually limited budgets of many journals and the time and energy of its editors and copy editors. Perhaps the most important resource each journal would like to preserve is the time and energy of its referees, who serve the journal in that capacity for little or no personal or professional gain. Imagine that a journal had no such "exclusivity policy." The number of its submissions would, in all likelihood, increase rather dramatically. Since most submissions to a journal require the time and energy of at least two referees to review, the resources for that journal—especially the time and energy of its referees—would deplete rapidly. Such a depletion of its resources might be worth the cost, provided the articles turn out to be worthy of publication and there are a sufficient number of "spots" in upcoming issues in which the articles could appear; but it would not be worthwhile at all to deplete these resources if many of the articles positively reviewed by the journal's referees are then published elsewhere. And it would surely be the case that if I were to referee for a journal even a small number of articles that I reviewed positively and that, then, appeared elsewhere, I would be very hesitant to referee for that journal again, since doing so would seem to be a significant waste of my time and energy. Of course, this justification is defeasible, but I think it is the best available and, I think, quite strong.

[Christopher Morris]

Two comments.

(1) Reviewing for journals is a pain. Most of the pain is writing the comments. It may take an hour or two to read a piece, but it takes me many hours more to write comments. I can turn around a paper in ten days if I don't have to write comments and rarely get one done in less than two months if comments are needed. My recommendation is for more journals not to provide comments and to guarantee a response (yes/no/wait another four weeks) within a month.

(2) As we are on the subject of pains, it's also a pain to read a double-spaced manuscript with endnotes. Granted that copy-editors need manuscripts formatted thus, why can't readers have nicely printed papers with footnotes? Much easier to read. The required format for submissions is ridiculously outdated. Drop this suggestion.

I'll add a third:

(3) I think blind refereeing is nice. But I'd prefer a simpler system: rapid responses, no attempt to hide biases. I regularly advise grad. students about where to send articles ("no, not there. They tend not to like this kind of stuff. Try this one instead."). There are so many journals now. Why not cut out the pretense of careful, unbiased examination, and just encourage editors to publish what they think is good. In other words, Phil & Public Affairs without dissimulation.

[Jamie]

Very interesting.

I disagree with Chris on both (1) and (3).

It is indeed a pain to write comments on papers you reject as a referee, but it is very, very annoying to get a rejection without comments. It's not merely that I can get good help from ETHICS referees; it drives me nuts to get a commentless 'reject'. And one time, though just one, I did get comments and wrote to the editor pointing out that the reasons for rejection were confused; after a brief exchange (with me), the editor asked the advice of a third party who recommended publication. (Very satisfying!)
Furthermore, I suspect that a referee who doesn't have to write comments is apt to be less scrupulous and careful. I can't remember specific instances, but I am pretty sure that on at least one occasion I planned to reject a paper, then as I wrote out my reasons I realized they weren't very good and I accepted the paper after all.
How about a compromise: write about 250 words for a rejection?

On (3), well, I guess I'll give the short version: I think PPA is not very good, and that its refereeing policy is largely responsible.

[Hilary Bok]

The real problem with only letting people submit papers to one journal at a time, in my view, is its effects on people who are trying to get tenure. When I submitted my first ever paper, it was rejected several times, often on grounds that I found totally bizarre. (One referee objected to the fact that I had failed to justify the claim that people sometimes make choices. Really.) Moreover, the turnaround time was rarely under six months. If you submit a paper in, say, your fourth year and get a few odd rejections in a row, all with long delays, a perfectly good paper that will rightly end up accepted at a good journal will not make it into your dossier as a publication, and that is a real problem.

For this reason, I have a modest proposal: that journals allow all and only untenured people to submit papers to multiple journals.

[Nathan]

This is off topic, but related. Has anyone ever heard any statistics on how often the "average" philosophy article (whatever that is) is read, or cited? I have heard numbers for other fields, or perhaps just academic articles in general. I have no idea if these numbers are at all reliable, or how anyone might determine this (except for citations), but I wonder if anyone has ever heard anything about this for philosophy.

[David Sobel]

I strongly agree with the points made in the initial post. I think that in most cases the case is so obviously correct, and most people would see that this is so, that there is an opportunity to unite around such a proposal and put pressure on journals that do not comply with these reasonable requests.

I think we should seek to get a number of philosophers to sign a statement saying that we will, in one way or another, boycott journals that do not comply with a list of demands within a specified time frame. I believe we could get enough philosophers to sign such a thing that it could make a difference to how the journals conduct themselves.

Here is one specific proposal, but the general idea could take one of many forms. I say we come up with a list of the most basic and generally accepted demands to place on journals (I would urge using the first 5 recommendations above). Then we ask a wide array of philosophers to sign a statement saying that if a journal does not comply with these demands, the people that sign will no longer referee papers for that journal. To be effective, we would have to get a lot of people to sign and if many of them were prominent, that would be even better. I think we should give journals something like 2 years to make the changes. One could, although I fear we would get fewer to sign, also pledge to no longer submit papers to the journals that do not comply.

In sum, comrades, I quite seriously urge that we take to the (metaphorical) streets. I urge that we not merely make a good case for change but find plausible ways to effect the warranted change.

[Sergio Tenenbaum]

I heartily second David's call to arms. I have thought for a while about putting forward a suggestion that would strike a balance between the interest of the journals in not wasting refereeing resources, and the interest of authors to have their papers published in a reasonable amount of time. One could institute the following policy: when an author submits a paper to a journal, the journal has a two or three month period in which the author is committed to publish in that journal if the paper is accepted. After that period, the author is free to send the paper to another journal, while it's still being reviewed in the first journal. The second journal would now have first dibs on the paper for the same period. So the system would work like this: author sends to journal A, and stays put for three months. After the three month period is over, author sends not only traditional "What the *&*$#@ happened to my paper" e-mail, but also submits paper to journal B. If now journal A accepts paper, author tells A to wait for B since they have first dibs now. If author does not hear from journal A or B in the next three months, now author sends not only two copies of the traditional e-mail, but also submits paper to a third journal.
Note that in the system no punctual referee labours in vain (unless of course one has a non-punctual co-referee...)
I'm sure there are very good reasons not to adopt this system, but I haven't thought of any yet.

[Douglas Portmore]

I urge anyone interested in these issues to also see the very interesting discussion thread on Brian Leiter's blog: http://leiterreports.typepad.com/blog/2004/11/philosophy_jour_1.html.