Skip to main content

Free guide

Who Owns Your Paper: what you sign away, what you keep, and the clause that contradicts your funder

9 min read · verified August 21, 2026

Every clause quoted here was read off the publisher or funder page that sets it, on the date above, and every one is linked at the end.

Most researchers sign the publishing agreement without reading it, assume they handed over the copyright, and are wrong about which parts they gave up. The parts you keep are broader than you think. The parts you lose are narrower and stranger. And if you are funded by a cOAlition S member, two organisations have told you to do opposite things with the same file, which is the part nobody warns you about at all.

Yours to keep, free

The whole guide is on this page. The PDF is the same thing, formatted to print or keep on your phone.

Download the PDF (PDF file)

01You may not have signed away copyright at all

The agreement has a name, and the name tells you what happened. Three of the largest publishers do three different things, so "I signed the copyright over" is a guess until you look.

  • Nature Portfolio does not require authors of original primary research to assign copyright. You grant an exclusive licence to publish, and in exchange you can reuse your paper in your own future printed work without asking permission first.
  • Wiley does one of two things depending on the journal: a Copyright Transfer Agreement, which is an actual transfer, or an Exclusive License Agreement, which is not. Same publisher, different documents, and the acronym on the form is the whole answer.
  • Elsevier splits it by route. Publish subscription and you typically transfer copyright. Publish open access and you keep it, granting a publishing licence instead.
  • PLOS is the simplest: you keep copyright, full stop. In their words, if you submit your paper for publication by PLOS, you retain copyright to your work.
  • So the first action is not to memorise a rule. It is to find the PDF of what you signed, read the title of the document, and know whether you are an owner or a licensor before you argue with anyone about it.

02The inversion nobody expects: your review article is the one you do not own

This is the sharpest thing in the guide and it is one sentence on a public policy page. At Nature Portfolio the copyright position flips depending on what kind of article you wrote.

  • Primary research: no copyright assignment required, as above. You keep it.
  • Non-primary articles, which Nature Portfolio lists as Reviews, Perspectives, Comments and News and Views, are different: copyright is retained by Springer Nature. That explicitly includes commissioned content and everything published in the Nature Reviews journals.
  • Read those two together and the result is backwards from the intuition. The invited review, the piece that arrives as a career compliment, is the one where you hand over ownership. The original research you fought through peer review is the one you keep.
  • It compounds with the money: non-primary articles are also not eligible for gold open access at Nature and can only go the subscription route. So the invited review is both the one you own least and the one you cannot open up by paying.
  • None of this is a reason to decline an invitation. It is a reason to know, before you agree, that reusing that text later is a permissions request rather than a right.

03What you keep no matter what you signed

Author rights survive a copyright transfer. These are carve-outs written into the agreements themselves, not favours you have to request, and most researchers underuse every one of them.

  • The preprint stays yours. Elsevier states that authors may publicly share the preprint anywhere at any time, and Nature Portfolio actively encourages posting preprints and confirms it will not jeopardise consideration. Posting a preprint is not a risk you are taking, it is a right you already have.
  • Reuse in your own next paper. Elsevier grants authors reuse of their article in their own new works without permission from Elsevier and without payment. Nature Portfolio grants reuse in future printed work without requiring permission first.
  • Your thesis. Including the article in your own thesis or dissertation is standard across the agreements we read.
  • Teaching. Classroom use is preserved, though at Elsevier the broader teaching and non-commercial MOOC rights for the published version sit behind the embargo.
  • The practical upshot: you almost never need to ask permission to build on your own work. People routinely email publishers for permission they were already granted in the contract they signed.

04What you actually lose is a file, not your ideas

The restriction is on a specific version of the document. Understanding which version you are allowed to post is the difference between a compliant deposit and a takedown notice.

  • There are three versions and only the middle one is contested. The preprint is yours. The Version of Record, the typeset PDF with the journal branding, is the publisher’s. The Author Accepted Manuscript, which is your text after peer review and before copyediting, is the one every policy argues about.
  • At Nature Portfolio the accepted manuscript can be made openly available six months after online publication, and the deposited version has to carry the reference and DOI on the first page with a link back to the published article.
  • The published PDF is the thing you may not post. This is what people get wrong when they upload to a personal site or a scholarly network, and it is also the version they most want to share, because it looks finished.
  • The requirement to link back is not bureaucratic. A deposited manuscript with no pointer to the published version is how two slightly different copies of the same study end up circulating, which is a citation and correction problem before it is a copyright one.

05The contradiction: your funder and your publisher have told you opposite things

If a cOAlition S funder supports your work, you have been given two instructions about the same file, and they cannot both be followed. This is live, it is documented on both sides, and almost nobody flags it at submission.

  • What the funder tells you: apply a CC BY licence to your Author Accepted Manuscript, and declare it in the submission itself so the licence predates anything you sign later. That priority in time is the entire mechanism.
  • What Nature Portfolio tells you, on its own self-archiving page: please note that the accepted manuscript may not be released under a Creative Commons license.
  • Read plainly, one says put a CC BY licence on the accepted manuscript and the other says you may not. There is no wording that satisfies both, which means an author following their funder in good faith may be in breach of the publishing agreement they signed, and an author following the publisher may be out of compliance with their grant.
  • What to do about it: put the declaration in at submission, in writing, where it is on the record. That is what makes it a prior licence rather than an afterthought.
  • Then ask the journal directly, before acceptance, whether it will publish on those terms. Some will, some will not, and finding out at acceptance is finding out too late to choose differently.
  • If the journal says no, that is real information about the journal, not a defeat. Your funder chose that requirement deliberately and the compliant venues are the ones that honour it.

The rights retention declaration, and what to check before you sign

  • Put the funder statement in the manuscript and the cover letter, at submission. A licence asserted after acceptance is not a prior licence, it is a request.
  • Name the grant. The declaration is a statement about a specific funded piece of work and it is weaker without the number.
  • Before you sign anything, find the document name: Copyright Transfer Agreement, Exclusive License Agreement, or a licence to publish. That word decides whether you are transferring or lending.
  • Check the embargo and which version it applies to, then diary the date. A right you forget to exercise is worth the same as one you never had.
  • Check whether the licence you are choosing can be changed later. Usually it cannot.

This is the cOAlition S rights retention wording, reproduced from their own page. It works by priority in time: the licence exists before you sign the publishing agreement. Applied after acceptance it achieves nothing, which is why the timing matters more than the phrasing.

06The licence you pick decides what the world may do with your work, and you pick once

Open access is not one thing. The letters after CC are the actual terms, they are enforced, and the choice is usually irreversible.

  • CC BY lets anyone share and adapt the work for any purpose, including commercially, as long as you are credited. It is the most permissive and it is what PLOS applies by default.
  • CC BY-NC-ND is much narrower. NonCommercial rules out use primarily intended for commercial advantage or monetary compensation. NoDerivatives means someone may remix or build on your work but may not distribute the modified version. Nature Portfolio offers both this and CC BY on its open access and hybrid journals.
  • NonCommercial is vaguer than it sounds and that vagueness is a real cost. Whether a fee-charging course, a company reading group, or a textbook counts as commercial is a question your future readers have to answer without you, and the safe answer for them is to not use it.
  • The AI clause is now explicit and it is worth reading twice. Wiley states that for works under a Copyright Transfer Agreement, an Exclusive License Agreement, CC BY-NC or CC BY-ND, authors and third parties must seek permission from Wiley for uses such as AI training or retrieval augmented generation. Authors, not only third parties. Under those terms you ask the publisher for permission to use your own article.
  • Assume the choice is permanent. PNAS states plainly that changes to licence selections, or to open access status, after publication may not be possible, and that is the general pattern rather than a quirk.
  • Choose by asking who you want to be able to use this and how, not by which line is cheapest. The price difference is one-off. The licence is forever.

07Read your institution’s IP policy, because it may outrank all of this

Every agreement above assumes you had the rights to grant. Your employer may have a prior claim, and that claim was made when you signed your contract rather than when you wrote the paper.

  • Universities and institutes commonly assert rights over work produced in the course of employment. Where that applies, the publishing agreement you sign is downstream of a claim that already exists.
  • We are not going to summarise these policies, because they genuinely differ and a confident general rule would be wrong for most readers. This is the one part of the guide where the answer is local.
  • Find yours before your next submission and read the sections on scholarly works and on copyright. Many carve out academic publications explicitly, which is worth knowing in advance rather than discovering in a dispute.
  • Ask your research office or librarian rather than a colleague. This is exactly their job, the answer is institution-specific, and it takes one email.
  • The general point stands even when the details do not: ownership was decided in several documents, signed at different times, and the publishing agreement is only the last of them.

Where to check

The databases this guide sends you to, all free and open without an account.

Where these numbers come from

  1. 01No copyright assignment for primary research; copyright retained by the publisher for Reviews, Perspectives, Comments and News and Views; six month embargo on the accepted manuscript; and the accepted manuscript may not be released under a Creative Commons licence Nature Portfolio, Self archiving and license to publish
  2. 02The rights retention wording authors are told to include at submission, and the priority-in-time mechanism behind it cOAlition S, Rights Retention Strategy
  3. 03Copyright transferred on the subscription route and retained on open access; preprint sharing anywhere at any time; reuse in the author’s own new works without permission or payment Elsevier, Copyright
  4. 04Copyright Transfer Agreement or Exclusive License Agreement on the subscription route, copyright retained on open access, and permission required from the publisher for AI training or retrieval augmented generation under CTA, ELA, CC BY-NC or CC BY-ND Wiley, Licensing and copyright
  5. 05Authors retain copyright and the default licence is CC BY 4.0 PLOS, Licenses and copyright
  6. 06NonCommercial and NoDerivatives terms, including that modified material may not be distributed Creative Commons, CC BY-NC-ND 4.0 licence deed
  7. 07Changes to licence selections, or to open access status, after publication may not be possible PNAS, Publication Charges

This guide will go out of date

I will tell you when it does

One email a week on what changed in academic publishing and whether it actually matters for your paper. Every number links its source.

One email a week, Mondays. Leave whenever you like.

The next step

None of this matters until the paper is accepted somewhere.

Rights are a good problem to have, because they only arrive with an acceptance. The step before it is the one that decides everything here, and it is decided by the manuscript rather than by the paperwork. The Academy works on that part, on your own draft, one task at a time. Come back to this guide when the agreement lands in your inbox, and read it before you sign rather than after.

Get published, one task at a time.

See what the Academy is