Rejecta Mathematica

Rejecta Mathematica is a new "open-access, online journal that publishes only papers that have been rejected from peer-reviewed journals in the mathematical sciences." "Every paper appearing in Rejecta Mathematica includes an open letter from its authors discussing the paper's original review process, disclosing any known flaws in the paper, and stating the case for the paper's value to the community."

I have to admit that reading these open letters in their inaugural issue, I was surprised by the histories of some of these papers. Some were caught in the gap between subjects (not enough physics to be published in a physics journal, not of broad enough interest to be published in a math journal). Other papers received rejections from reviewers on rather slim or debatable grounds (of course, we only hear from the authors in their own words, so we only get one side of the story). Other authors gave up or moved on, and didn't even try to revise and resubmit their article to the same or a different journal.

This raises the interesting question of whether such a journal would be welcomed in the atmospheric sciences community (Rejecta Meteorologica?). Certainly, I have seen some manuscripts rejected at Monthly Weather Review and other journals that I edit or review for that would fit into one of these categories. What are your thoughts?

Regarding Prof. Jim Steenburgh's posting "about submitting a paper"

I agree wholeheartedly with Professor Steenburgh's comments. As a reviewer, I have often felt that a manuscript that I was being asked to review was not nearly in as "ready" a state as possible. Not uncommon are manuscripts of this type with multiple authors; it seems doubtful that all of the "authors" have participated in the preparation of, or even proofread, the final submitted version.

The submission of poorly edited manuscripts or the premature submission of manuscripts can be viewed as an abuse of the peer review process. Asking reviewers - whose time is valuable - to review these manuscripts is disrespectful and ultimately destructive to the peer review process - reviewers become unenthusiastic and wary about agreeing to do reviews.

It would be desirable for MWR and the AMS to institute policies to discourage the submission of ill-prepared and poorly edited manuscripts.
An issue is that editors and reviewers are currently often unwilling to reject a manuscript, which has some scientific merit, solely on the grounds that it is poorly edited and inadequately prepared.
One possibility would be to ask editors and reviewers to identify such manuscripts. An editor, assigned such a manuscript, could return it to the authors without sending it out for review; and a reviewer, who received such a manuscript, could decline to review it on the grounds that it was not in a suitable state for submission in the first place. Such a decision or recommendation would be different from a rejection - the author(s) would be asked to submit an adequately prepared revised manuscript.

Herschel Mitchell
Editor, MWR

How to respond to reviewers

Anonymous raised some interesting questions about responding to reviewers in a comment on a previous post. My response was getting quite rich in content, so I thought I would submit a full blog entry instead. Bottom line to Anonymous: I think you are approaching your response to reviews correctly.

As a reviewer and generally as an editor, I would be thrilled if every author responded positively to all the reviewer concerns and the manuscript was improved as a result. We could have one round of reviews, and the paper would be published. That situation would be easiest for everyone (reviewers, editor, authors). :-)

Although reviewers generally provide more useful comments than not, reviewers are not always 100% right in their reviews. Therefore, I tell authors that they should usually accept the majority of reviewer comments---especially the major comments---if they want to have smooth sailing through the review process on the way to eventual publication. I call this knowing how to play the game. A positive response to 70-80% of the comments, including intelligent responses to the major issues, usually makes me feel pretty happy about the way the review process goes. No one gets everything they want, but the result is a dramatically improved manuscript that should be acceptable for publication. When authors and reviewers know how to play the game, they submit decent manuscripts that are improved during the peer-review process and are published after one or two rounds of reviews. As an editor, I don't have to write long decision letters for such manuscripts---I trust that the authors will know how to take the reviews and revise the manuscript accordingly.

If an author does not meet these standards with their revision (isn't playing the game), then I as editor am put in an awkward and potentially compromised position. I am trying to balance competing effects: publish good, if not the best, manuscripts and publish them quickly. In most cases, I respect the reviewers (I selected them!) and their opinions, so I want to see that the authors have listened to what the reviewers said and have made positive changes to their manuscript. But I also want to give the authors some license to write the manuscript as they envision it and to publish their work quickly. I am handling 20-30 manuscripts at any given time, and the more peer-reviewed manuscripts that I can send to the publisher, the fewer are in my in-box, so to speak.

Like most conflicts in life, you have to know when to pick your fights. If you are going to battle the reviewer on a major comment or two, then accepting most of the other comments will give me as editor some relief that you are taking the review process seriously. That is why I say 70-80% as an estimate. I don't expect reviewers to always be right or provide exceptional comments all the time, and I feel authors should be able to defend some of their material against possible reviewer misunderstandings. But, if an author blows off half or more of the reviewer comments, then I think the author is being too resistive to change and I get annoyed. I am particularly troubled if the author blows off too many minor comments, which should be relatively straightforward to implement in a revised manuscript.

Good editors know when to bring this back and forth between reviewers and authors to a close and quickly. For manuscripts for which major revisions are required (about 50% of manuscripts submitted to MWR require major revisions), I try not to go more than two rounds. If I have to have a third round of reviews, someone didn't do a good enough job (usually the author).

Quote about submitting a paper

As a reviewer, I see a lot of papers that are sent in with the idea that they will do the final editing after the reviews (or perhaps that the reviewers will provide what they need to edit to final form). My personal view is that when you submit a paper it should be in final form and that you should be comfortable with the paper going directly to press as is. It is a waste of time for all of us to review anything less.
-Prof. Jim Steenburgh, Chair, Department of Atmospheric Sciences, University of Utah

Welcome a new Editor-Herschel Mitchell

To accommodate the growing number of manuscripts on data assimilation, we have added a new editor, Herschel Mitchell of Environment Canada. Herschel will be handling manuscripts on the topics of data assimilation, Kalman filtering, ensemble data assimilation, the ensemble Kalman filter, and operational global and regional NWP.

Welcome aboard Hersch!

Dave

The Unacceptable Practice of Shopping Manuscripts Around

Lately, I have seen situations at MWR and other journals where a manuscript was rejected, only to be resubmitted to the same journal or a different journal with only minor changes, if any, being made. In the cases I am familiar with, the reviews recommending rejection contained reasonable requests, including rerunning model simulations, fixing improper English language, and replotting figures.

Let me be perfectly clear.

Submitting your manuscript to the journal is a privilege, not a right. It is a privilege that can be revoked by the editor or publisher of the journal.

You are imposing upon an editor and several reviewers, all volunteers, to improve your manuscript for publication. Even the most critical reviews offer advice that can make your manuscript better. To ignore their efforts and resubmit the manuscript with only minor changes is a blatant disregard for the time of others. Most editors do not tolerate such behavior, and your manuscript will be rejected.

A similar infraction occurs when authors are found to be “shopping around” rejected manuscripts between journals. Because atmospheric science is a relatively small discipline (compared to physics or chemistry, for example) and your area of specialty may be even smaller still, chances are that some of the same people that knew about your original manuscript at the first journal will see it again at the next one. Not making major revisions to a rejected manuscript, whether or not it was submitted to the same journal, is simply unacceptable.

(Thanks to Andrea and Russ Schumacher for identifying an error in this post, now fixed.)

December 2008 issue of MWR

As the AMS processes the backlog of manuscripts, speeding up the time to publication (now about 200 days from the time of acceptance, down from 240 days last year), you'll notice the issues of MWR getting thicker. The December 2008 issue probably sets the record for length of a single issue among AMS journals: 727 pages.