The relation between errors in the published literature and publication ethics remains in many ways something of a contested concept. Not only is there little agreement as to the scope of the problem, there is also lack of consensus as to what activities, by definition, count as misconduct, particularly in the reporting of research. Nevertheless, it is still possible to acknowledge that specific practices involving corrections to the scientific record make good publishing and ethical sense, and on this basis we firmly recommend their adoption as a matter of policy. It is precisely this purpose that the following document fulfills.
There are four primary motivations for those involved in the publication of scientific research to take seriously as a mandate that the record of scientific activity should be an up-to-date and accurate representation thereof. These roughly correspond to the accepted functions of a published journal, namely: registration, certification, dissemination, and archiving. By briefly discussing each of these functions in turn, it will become clear that no matter how complicated, and even conflicted, discussion concerning the application of ethical principles to the reporting and publication of scientific research might become at times, it is still a practical virtue to hold the publication process and record to a high standard of accountability.
Registration refers to the date-stamping of a submitted manuscript as the assignment of priority for review and possible acceptance by a journal. It is understood that, as a matter of both professional courtesy and commercial viability, a journal’s editorial office will strive for timely, expeditious, and efficient fulfillment of its publication tasks. It is therefore assumed that a journal has already at some point undertaken to make clear and transparent its own ethical policies regarding both submission and review of manuscripts, within the journal’s print or web Instructions for Authors text, and for the explicit purpose of avoiding all doubt regarding expectations and responsibilities.
Certification refers to the guarantee of quality that completion of a journal’s peer review process confers upon an accepted article. This means that a journal’s reputation is staked on the demonstrated validity of the criteria used to evaluate submitted manuscripts. The pressing ethical concern underlying this assumption of proof is therefore the extent to which a journal’s peer review process is both stringent and trustworthy. It is therefore necessary for a journal, as a function of maintaining the standing and reputation it might have as a desirable venue for publication, to put in place procedures which will aid in promptly and appropriately addressing potential inaccuracies occurring within its portion of the published record.
Dissemination refers to the ways in which a journal provides access to the published version of an accepted manuscript. The process of dissemination should itself be carried out in a timely fashion. Like registration, however, dissemination poses, although from a different standpoint, the same ethical questions regarding how a journal might best balance the potentially competing interests of operational efficiency and due diligence. It is certainly a given that every journal’s production office has but limited resources of time and attention to allot for the purpose of detecting errors, including errors resulting from potential misconduct. It is therefore also assumed that a journal has already at some point undertaken to make clear and transparent its own policies regarding both correction and alteration of manuscripts prior to final acceptance for publication, again with the explicit purpose of avoiding all doubt regarding expectations and responsibilities.
Archiving refers to the creation and maintenance of a system for the preservation of a journal’s published record. From the standpoint of the individual journal, then, permanence as a matter of historical reference demands that any and all potential errors not resolved prior to final acceptance should be noted and corrected. It is therefore necessary for a journal, yet again as a function of maintaining the standing and reputation it might have as a desirable venue for publication, to put in place procedures that will aid in promptly and appropriately addressing such potential errors.
Complexity is embedded within the very process of reporting and publishing the results of scientific research. It is a common-sense virtue, in light of this situation, to strive for cultivating an attitude of constant vigilance throughout each and every stage of the publication process. Timeliness and accuracy being the recommended watchwords, it is possible for the individual journal to create policies which support work that strikes a balance between these demands. In doing so, a journal may not find that it is able to directly discourage individuals from choosing an ethically questionable course of action, but might succeed in making a substantial contribution to fostering a climate which rewards responsible decision making.