The Journal calls upon referees to help evaluate manuscripts, to ensure that the broadest range of expertise in the profession is brought to bear on submissions. Ultimately, we are asking referees to help the editorial team with two central questions about a submission:
Answering the first question might involve considering whether the principal line of argument is valid or cogent, and whether the author's chosen methodology is appropriate for the questions they are attempting to address. Answering the second question might involve consideration of the novelty, significance, and overall plausibility of the submission.
We'd expect referees who answer ‘no’ to one or both of these questions to recommend rejection of the submission. If the answer to the question(s) is ‘yes, with reservations’, then we'd expect a recommendation of revise and resubmit; and if the answer is ‘yes’ to both questions, we'd expect a recommendation of acceptance.
Referees should acknowledge they are in a position of considerable influence over the fate of a submission, and should always act professionally, ethically, and demonstrate high standards of probity. Manuscripts must always be evaluated on their merits. We expect referees to disclose any conflicts of interest and to decline any requests to referee manuscripts that are not in their areas of expertise.
We ask that, unless special arrangements have been explicitly made, referees' reports be supplied within one month of accepting the task. Please get in touch if you need more time; we generally prefer a late report to no report at all, but it's always best to know what's going on, and we can often offer an extended deadline which will stop our system sending you increasingly shrill automatic reminders.
ScholarOne Manuscripts allows direct typing of reports into the site, but, as with all typing directly into text boxes, there is a risk of loss of work. When you have completed your report and are ready to submit it, click on 'Submit'. (If you don't do this, we won't know you've supplied it.) If you want to write the report piecemeal, it is essential that you click the 'Save' button before exiting, or all the information that you have entered will be lost.
Referees may prefer to paste their text, or lodge their reports as (or supplement them with) word-processed documents. If they do so, they should take notice of the limitations on file formats described on the Submissions page. We prefer not to receive reports in PDF format, though for especially lengthy reports, or those that make use of special characters or formatting, can be uploaded as a file. This file will go to the Editor and Associate Editor, who will pass on to the author its contents at their discretion.
On the referee page, there is a space for 'Comments to Editor' and a space for 'Comments to the Author'. Please be sure to paste or write your comments in the appropriate space.
If you wish to view the manuscript and the referee form simultaneously, click on the HTML or PDF icon. The manuscript will open in a new window. Leave the new window open, switch back to the main window, and open the score sheet by clicking on the Score Sheet tab. Follow the instructions for referees provided in the ScholarOne Manuscripts site.
A referee is asked to indicate by checking the appropriate box on the ScholarOne form whether acceptance, conditional acceptance, minor revision, major revision, or rejection is recommended. We prefer that referees not indicate their recommendation in the text of their comments to the author.
If either kind of revision is the recommendation, indication should be provided of willingness (or lack of it) to read a revised version of the paper; in the former case, a copy of the original manuscript may be retained for convenience. (A verdict of 'major revision' is strongly discouraged when the paper has already been resubmitted; resubmitted papers are identifiable by an 'R' suffix to their login number.)
In the case of a paper longer than 8,000 words of main text including bibliography (or a Discussion Note longer than 2,000), referees should bear in mind the editorial policy that the acceptance bar rises with increasing length; roughly speaking, a 16,000-word article would have to be good enough to out-compete not just one but two good papers of 8,000 words each, and suggestions about how a paper could profitably be shortened are always very useful. It greatly assists the Editor in coming to a decision if referees provide sufficient commentary so that the basis of their judgment, rather than just the verdict itself, is clear. This also is vital information for authors, who will have a clearer idea of how their work must be improved in order to be of a standard publishable in a leading journal.
Referees should also bear in mind that Journal policy is to make available as much of their reports as possible to the author(s). The AJP is committed to the view that the process of refereeing not only ensures the high quality of the papers appearing in it but, additionally, serves an educative function, especially for less experienced scholars. Referees should accordingly adopt a judicious tone in their assessment, while not forgetting that, if a paper is of very poor quality the report must indicate this. A clear and forthright report is of more use to author and Editor than a tactfully evasive one, but the language should be measured. If in doubt as to whether a given remark should be directed to both author and editor, direct it to the editor alone, who may make use of it in their own comments to the author. Occasionally referees make comments directed to the author that it would not be prudent for the journal to forward to the author; please note that the editor may minimally edit a report in such cases. The Journal endeavours to the extent possible to abide by the COPE guidelines on editing of peer reviews.
In those rare cases where a referee has no comments to offer an author, but only wishes to make comments to the editors, please put 'N/A' or similar in the field for comments to the author to allow you to submit your recommendation.
A paper accepted for publication in the AJP should of course display in a high degree the usual academic virtues—argument, organisation, originality, scholarship, significance and so on—and referees are expected to comment on these matters. But it would be greatly appreciated if referees also asked themselves these sorts of questions about a submission: Is it enjoyable, even exciting, to read? Is it written in such a way that it might interest someone who does not already have detailed knowledge of its subject matter?
Referees may consider whether a submission is well-written. But care must be taken here. Referees must be mindful that the AJP is an international journal, and we receive and eventually publish many submissions by authors whose first language is not English. While the style and fluency of a submission are components of an evaluation, referees should not give undue weight to minor infelicities or unidiomatic language, and issues of prose style should never be determinative of a recommendation. (Of course, if the paper is poorly structured or written in such a way that it cannot be understood, those are problems that go beyond issues of prose style.) See our policy on language and style for more information on this point.
The Journal assesses each submission on its merits, but in calibrating their recommendations referees might wish to consider that (i) the AJP is one of the top-rated journals of its kind, and (ii) we have space for only a small percentage of the very large number of submissions we receive.
As assessing revisions adds significantly to the already high workload, before recommending revision rather than rejection referees should be fairly confident that this standard is likely to be achieved within one round of revision. Comment on whether the paper is built around a really strong and distinctive key idea would be especially welcome. If the paper is currently a little underdeveloped in some of its details, but a sufficiently original idea is being introduced by it, then a recommendation of revise-and-resubmit would be appropriate, all else being equal. This is so, even if it is not entirely clear that just a single revision will be enough to do full justice to the strong idea at the heart of the paper.
Referees should be wary of recommending revisions to an already-revised submission; writing a report is not managing the author from afar to produce the article you would have written. Referees should feel empowered to recommend rejection of a revised manuscript that has made changes in line with an earlier round of comments – often the changes that authors make serve to underline or bring out a weakness in the paper, and referees are not obligated to recommend acceptance just because the author has done something to respond to the referees prior concerns. Sometimes a second look at a submission allows referees to notice a problem they didn't previously notice, and it is okay to recommend rejection on that basis (even if in an ideal world it would have been noticed on the first look).
It hardly needs saying that we would like referees to assess the paper, as far as possible, with regard to the quality of its argumentation, rather than in terms of the compatibility of its conclusions with their own positions and philosophical commitments. Writing a report is different from writing a reply—referees should, for example, resist the temptation to recommend rejection merely because you can think of some ingenious objection to the paper's arguments. (However, if the referee thinks there is a straightforward objection to the argument, one which would occur to most reasonable philosophers familiar with the topic but which the author has not considered, that is a reason to conclude the paper cannot be published in its current form.)
It is worth emphasising that referees are not making a decision about the submission, but providing advice and making a recommendation to the editorial team. The editorial team is probably better placed than individual referees to reflect on the suitability of a submission for the Journal and our audience, and referees are asked to focus primarily on questions about the quality and significance of the paper with respect to broader disciplinary norms.
All intellectual property in submitted manuscripts remains with the author. Referees must always respect the moral and intellectual property rights of the author. Information gained by referees and others involved in the review of manuscripts is confidential to the reviewing process, and cannot be used by referees except for the purposes of conducting their review.
The AJP strives to implement double-anonymous review. Referees should not attempt to identify the authors of manuscripts, either during or after submitting their review (for reviewers may be asked to review a revised version). Likewise, reports themselves must be prepared anonymously. Referees should resist the urge to recommend significant numbers of their own publications to an author in their report; often such recommendations are de facto identifying.
If a paper is accepted for publication in AJP, the editor aims to notify referees of the verdict and of the identity of the author. We do not disclose the identity of referees to authors, even at that stage. We do not notify referees of final verdicts of 'reject'.
Sometimes, particularly for submissions on relatively niche topics, referees do have knowledge or at least justified belief about the identity of the authors. This is increasingly common given the connected nature of the profession today and our requirement that referees be active researchers. This does not automatically disqualify the referee from providing a review. The question referees should ask themselves is: is there a conflict of interest in my reviewing this submission? If the referee knows the author from having seen a conference presentation or a preprint, for example, but has no other professional or personal connection with the author, then it is probably acceptable to provide a review. If a referee has seen a previous version of the submission at another venue, that is also not a disqualifying fact. The editor is asking an expert peer for their opinion of the merits of a submission. The journal's interests are best served by having the best advice, even if that is advice from someone who has seen a previous version of the submission. Put bluntly, the Journal exists for its readers, not its authors, and otherwise qualified referees do not need to recuse themselves on the grounds that a paper may have a better chance of acceptance with another referee.
If in doubt, referees should contact the member of the editorial team who extended the invitation to review for further advice.