The Editor's Department

THE EDITOR'S DEPARTMENT In my December 1990 Editor's Department column I described the refereeing process for submissions to Language and gave advice to referees about how to write a good referee report. This column also concerns the evaluation of manuscripts, but this time I will focus on giving authors information that might help them understand the process and its results; and I will offer suggestions about useful (and not so useful) ways of reacting to those results. The ideal referee has a number of characteristics. First, s/he is an excellent scholar: it takes one to know one. Second, s/he knows a great deal about the topic of the manuscript, both its substance and the scholarly literature that has been devoted to it; in this context, 'topic' includes both the linguistic phenomena that are being analyzed and the language(s) from which examples are drawn. Third, s/he is probing and tough-minded: weak argumentation, implausible ideas that are not well supported by evidence, good ideas that are not well supported by evidence, failure to cite crucially relevant published discussions of the topic, mistakes in data, and seriously flawed exposition will be identified by the ideal referee, and will not be viewed favorably. The ideal referee will suppress any tendency to let sympathy for an author who tries hard (but doesn't succeed) outweigh the primary goal, which is to select papers on the basis of merit of content and presentation. And, of course, this referee will never recommend acceptance of a non-first-rate paper, even if its author is famous. Fourth, s/he is extremely fair, able to view all approaches to a particular question as potentially valuable and to avoid letting personal theoretical prejudices get in the way of assessment of an author's analyses. Fifth, s/he has excellent judgment about what is and is not important: minor flaws in a paper never lead to a recommendation to reject, and major flaws always do. Sixth, s/he gives the author courteous and helpful feedback, regardless of whether the recommendation to the editor is positive or negative. And seventh, s/he is reliable, prompt, and willing to make room in a busy schedule for this selfless contribution to the health of the field. Editors spend a lot of time searching for and then cultivating referees who approximate the ideal. Not surprisingly, the vast majority of referees are good, conscientious scholars who provide useful recommendations and feedback but who fall short of the ideal in one respect or another. Inevitably, any editor will also fail in the effort to achieve complete knowledge, perfect judgment, and other attributes of the ideal editor. Against this background of well-meaning but fallible editor and referees, consider the author's role in the evaluation process. Of course the best way to increase the chances of acceptance is to send the editor an excellent paper that deserves to be published in a highly selective journal. (The acceptance rate for Language has ranged from 11% to 15% in the past three years.) But the author can do more. To begin with, the author can suggest referees to the editor. Often I can't take an author's suggestions, for various reasons; but often I can, and even when I can't, the author's suggestions let me know what kinds of referees I 428 THE EDITOR'S DEPARTMENT429 should be looking for. I am most likely to adopt an author's suggestions if it looks as if the author is proposing truly objective scholars—not his/her colleagues or (former) teachers, and not only people who will be predictably enthusiastic about the approach. Authors who write first-rate papers are likely to suggest referees whose critical comments they themselves would most like to see, and who will be objective. An author can also request that the paper not be sent to a particular scholar. Like most editors, I will normally honor such a request. Authors should think about what the editor needs to know when submitting a paper. A brief cover letter should always be included with a submission, but the author should resist the urge to send along reprints of...