Nature’s npj series lays out a strict line on plagiarism and duplicate publication
The npj series is explicit about what happens when originality is compromised: the integrity of the scientific record comes first. Its policy treats plagiarism, duplicate publication, and text recycling as threats not only to individual manuscripts, but to trust in authorship itself. The language is firm because the stakes are firm. If research claims are misrepresented, the damage reaches far beyond one paper and can distort how readers judge the evidence base.
At the center of the policy is a straightforward principle. Manuscripts submitted to the series must be original, must not already be published elsewhere, and must not be under concurrent consideration by another journal. Related material that exists in press or under review elsewhere must be disclosed clearly at submission, with a marked copy provided to editors. When a submission reuses part of an earlier work, the responsibility shifts to the author to make the overlap transparent and properly cited.
Text overlap is treated as a credibility problem, not a cosmetic one
The policy defines plagiarism as unacknowledged copying or misattribution of ideas, text, or results. Large blocks of text cut and pasted without clear attribution are singled out as unacceptable, and manuscripts containing that kind of borrowing would not be considered for publication. The same standard applies when authors paraphrase or summarize the work of others: close attention to attribution is required so readers are not left with the impression that a fresh contribution is being presented when it is not.
A separate warning applies to text recycling, sometimes called self-plagiarism. Reusing language from an author’s own earlier publication can be problematic when the duplication is not disclosed, because it can create a misleading sense of novelty. The policy does not ban every form of overlap, but it insists on transparency, proper citation, and compliance with copyright requirements when reuse is unavoidable. In practice, that means authors need to show editors exactly what has been reused and why.
Duplicate publication is handled just as bluntly. Reissuing the same paper in multiple journals, or adding only a small amount of new data to a substantially identical report, falls squarely into the category of redundant publication. That kind of repetition can inflate the appearance of evidence and complicate interpretation across the literature. For readers trying to assess the weight of a finding, a duplicated paper is not simply extra text; it is a distortion of the record.
One sentence in the policy captures the editorial posture clearly: manuscripts must be original, and any exception must be identified up front.
Editorial action can extend beyond acceptance or rejection
The policy also describes what happens after publication if plagiarism becomes evident. Depending on the degree of overlap, the surrounding context, and the effect on the integrity of the published work, editors may correct, retract, or otherwise amend the article. That flexibility matters because not every case is identical. A minor citation failure is not the same as wholesale unattributed copying, and the policy leaves room for proportionate action rather than a single automatic response.
The journal group uses Similarity Check, a software screening tool that looks for text overlap in submitted manuscripts. The presence of automated screening does not replace editorial judgment; it supplements it. Editors review cases individually and decide whether overlap is accidental, acceptable with disclosure, or serious enough to block publication or trigger post-publication correction. That process reflects a broader attempt to detect misconduct before it enters the record, while preserving discretion when overlap has legitimate explanations.
The policy extends the same standards to material drawn from others’ unpublished work and from published work. Unpublished data cited in a manuscript must be properly credited, and the necessary permissions must be in place. Published work must be described accurately, with both intellectual and technical contributions acknowledged in a way that matches the actual history of the science. The policy also requires authors to be transparent when they reuse figures or other copyrighted material, including documentation that permission has been granted.
A practical consequence follows from all of this: publication is not treated as a clerical endpoint, but as part of a chain of accountability.
Originality, disclosure, and permission define the submission standard
The npj series allows prior publication on recognized preprint servers and permits meeting abstracts before full submission, but those materials must be disclosed. It also allows consideration of work that has appeared as part of a thesis, provided institutional requirements have been met. When the material has already appeared in another language or in another format, editors decide whether the main result, conclusion, or implication remains sufficiently distinct to justify consideration.
For authors, the message is unambiguous. If a manuscript contains any material that has appeared elsewhere, the overlap has to be named openly. If data, text, or figures come from outside sources, the proper credit and permissions must be in place before submission. If a case is uncertain, the editor handling the manuscript is the person to contact.
The result is a policy built around one clinical-adjacent but universal scientific aim: protecting the reliability of the published record. In this framework, plagiarism is not a minor stylistic lapse, duplicate publication is not a bookkeeping error, and undisclosed text recycling is not harmless repetition. Each can alter how evidence is understood, and each is treated as a matter of editorial integrity rather than convenience.