Plagiarism, self-plagiarism and text recycling: where the lines are

Few topics cause as much anxiety among researchers as plagiarism. Part of the reason is that the rules feel unclear. Is it plagiarism to reuse a paragraph from your own thesis? To describe a standard method in the same words as your last paper? To translate an English article into Hindi and publish it? And what does a 15 per cent similarity score actually mean?

This guide sets out the main categories, the grey areas, and practical habits that keep your work clean. It is general guidance, not a substitute for your institution’s or journal’s own policy, which should always be your final reference.

Plagiarism: using others’ work as your own

Plagiarism is presenting someone else’s words, ideas, data, images or structure as your own without proper acknowledgement. It is widely treated as a serious form of research misconduct. Common forms include:

• Verbatim copying of text without quotation marks and citation.
• Close paraphrasing (“patchwriting”) that changes a few words but keeps the original sentence structure, even when cited.
• Idea plagiarism: taking an argument, hypothesis or framework without credit, even in entirely new words.
• Copying figures, tables or data without permission and attribution.
• Translation plagiarism: translating someone else’s work into another language and presenting it as original.
• Citing sources you have not read, by lifting references from another paper, which can misrepresent both the source and your own scholarship.

Intent matters for how institutions respond, but it does not change whether plagiarism has occurred. “I forgot to add quotation marks” is an explanation, not a defence.

Self-plagiarism and duplicate publication

“Self-plagiarism” is a slightly awkward term, because you cannot steal from yourself. What it describes is reusing your own previously published work without disclosure, in a way that misleads editors and readers into thinking it is new. The most serious form is duplicate or redundant publication: publishing the same study, data or findings in more than one place as if each were original. This distorts the literature, can double-count evidence in reviews, and often breaches copyright agreements with the first publisher.

Related practices that journals look out for include:

• Salami slicing: splitting one study into several thin papers to increase publication count.
• Simultaneous submission: sending the same manuscript to more than one journal at once.
• Undisclosed republication of a conference paper or book chapter as a journal article.

Text recycling: the grey zone

Text recycling means reusing portions of your own earlier writing in a new document. It is not automatically wrong. Views differ across disciplines and publishers, but some patterns are widely accepted:

• Usually acceptable: reusing short, standard descriptions of methods, especially where there are only a few ways to describe a technique precisely, provided you cite the earlier paper. Publishing a journal article based on your own thesis is also generally accepted, though many journals expect you to disclose it.
• Needs care and disclosure: reusing parts of a literature review or theoretical framing across papers, or developing a conference paper into a full article. Tell the editor, cite the earlier version, and make sure the new work adds substantially.
• Usually not acceptable: reusing results, discussion or conclusions, or large portions of text, without disclosure; or reusing text where the copyright belongs to another publisher and the licence does not permit it.

When in doubt, disclose. A line in your cover letter explaining the relationship to earlier work costs nothing and protects you.

What similarity reports do and do not tell you

Similarity-checking software compares your text with a database and reports overlapping passages. A percentage score is not a verdict. A high score may come from properly quoted material, a reference list, standard phrases or your own preprint. A low score can hide serious problems such as idea plagiarism, translated text or heavily paraphrased copying.

In India, the UGC’s 2018 regulations on promoting academic integrity and preventing plagiarism in higher educational institutions set out graded levels of similarity with corresponding consequences, and they treat similarity of up to 10 per cent as “Level 0”, which carries no penalty, for both theses and research papers. Higher levels (above 10, 40 and 60 per cent) bring increasingly serious consequences. Before the percentage is worked out, certain material is excluded, such as properly quoted and attributed text, references and bibliography, the table of contents and acknowledgements, generic terms, laws and standard equations, and common phrases. Two cautions: the rule applies to Indian higher educational institutions and is not a journal standard (journals set their own thresholds), and a score under 10 per cent does not prove that a work is free of plagiarism. If you are a student or faculty member in India, read the current regulations and your university’s policy, since institutions apply them with their own procedures.

Whatever the score, editors and committees look at what the overlapping text is. Read the report yourself before submission and check every highlighted passage.

Practical habits that keep you safe

1. Separate notes from your own words. When taking notes, mark direct quotations clearly with the page number. Many accidental plagiarism cases start in messy notes.
2. Paraphrase from understanding, not from the page. Close the source, write the idea in your own words, then check you have not mirrored the original.
3. Cite ideas as well as words. If a concept or argument came from someone else, credit it.
4. Use a reference manager so citations do not go missing between drafts.
5. Keep a record of your own prior outputs and cite them when you build on them.
6. Disclose related publications to editors, including theses, preprints, conference papers and chapters.
7. Check co-authors’ sections too. All authors share responsibility for the whole manuscript.
8. Get permission for reused figures and follow licence terms.

If something goes wrong

If you discover a problem in your own published work, contact the journal promptly. Correcting the record through a correction or, where necessary, a retraction is far better than having someone else find it. If you suspect plagiarism in work you are reviewing, raise it confidentially with the editor rather than with the authors directly. Guidance from the Committee on Publication Ethics (COPE) is widely used by journals to handle these cases.

How CollabCite fits in

Research integrity is part of how CollabCite works: members carry a verified ID, and our Integrity Policy sets out what we expect of everyone on the platform. We also only share calls from partner journals that meet our journal standards. If you want collaborators who take these questions seriously, you can join as a founding member.