When publishing in a non-open access journal, one (Author) is usually required to sign a copyright transfer agreement with the Publisher such as this.

Now, when the paper includes some work, say a photo, by a third party XYZ, obtainable under a CC license and properly noted ("Photo by XYZ, licensed under CC BY-ND 4.0"), I assume that it is understood that this is not part of the transfer, but Author assures the Publisher that Author and Publisher actually have the permission to use it. (Since the CC license is non-exclusive, distribution of the article by Publisher will then be business between Publisher and XYZ, right?)

If that understanding is correct, now my actual question: what happens if Author and XYZ are identical? Say Author has produced some photo, and previously posted it somewhere under a CC license. They then reuse the photo in their own paper, with the same kind of attribution ("Photo by Author, licensed under CC BY-ND 4.0").

Author in this case actually holds the copyright to the photo, and it looks like they are in a position to transfer it, which they cannot escape from when signing the (unmodified) agreement. At the same time, it seems to me that this transfer does not work -- there are previous rights granted to everyone else! Can Author in this case (implicitely) transfer the copyright of the rest of the work, and independently grant usage rights for the photo throught the pre-exising CC license?

And even worse: suppose the work in question is not just an arbitrary photo, but depicts something obviously specific to the paper (say, photography of experimental setup, or a diagram relating to the research). In such a case, if Author dutyfully adds a copyright notice, they practically reveal their identity to the reviewers! How is this case resolved?

Background: I'm new to academic publishing and was a bit appaled when I saw what kind of copyright transfer agreements I am supposed to sign. I came up with this idea to at least be able to CC license my diagrams, and was wondering how much sense it makes. Related to this question, but not the same.

  • In copyright law, nothing prevents the author from providing different licenses for the same material to two different parties. In this case, however, the author may desire to explicitly not convey copyright of all the material to the journal, and the form usually has ways to denote this.
    – Jon Custer
    Commented Jan 12, 2022 at 13:39
  • "In such a case, if Author dutyfully adds a copyright notice, they practically reveal their identity to the reviewers!" No journal I'm aware of forwards the copyright agreement to the reviewers. In either case, accidental unblinding of authors is not a big deal, particularly this late in a refereeing process. Commented Jan 12, 2022 at 14:14
  • 2
    Also, the question at hand doesn't depend at all on whether the author of the paper is identical to the author of the image. In either case, rights already granted are taken back, which is legally rather questionable. This is an interesting question, with a wide span of possible good answers (my impression is that the de-facto standard behavior differs widely from what is legally justified), so I'm trying to simplify it to make it work as a canonical FAQ. Commented Jan 12, 2022 at 14:17
  • @darijgrinberg Good point. Also, in my experience copyright transfers are signed only after reviewing has finished. Commented Jan 12, 2022 at 16:00
  • @darijgrinberg CC licenses are irrevocable, so it’s not just legally questionable, it’s actually impossible to take back the legal rights in this particular situation.
    – Dan Romik
    Commented Jan 12, 2022 at 16:50

2 Answers 2


The situation of transferring copyright over material which is already licensed under a Creative Commons license is very common in mathematics, where papers almost always get uploaded to arXiv before they are accepted to a journal. The publishers are aware of this, and the copyright transfer agreement will typically list a series of standard exceptions to the publisher’s ability to assert the copyright, to account for the rights the author has already waived away by uploading their content to the arXiv (and to give authors certain freedoms that they generally like to have, like the right to post a preprint version of the paper on their home page). So, there is nothing unusual about such a situation, and it does not make it impossible to transfer copyright, it just means this must be done with appropriate caveats.

In your case you should also make it clear to the publisher that even though are transferring the copyright, you have already waived certain rights in connection with the image. Do so before you sign the copyright transfer form to allow them to suggest how to proceed before you sign away legal rights that you effectively no longer have. But the example above with mathematics papers shows that this is not a difficult situation for publishers to accommodate, so I’m reasonably confident they will find a way to allow the process to move forward, probably by making a slight adjustment to the language of the copyright transfer agreement.


If you give a license you don't hold unrestricted copyright any more since you have yielded some rights. You can only give rights you hold. If you license something under CC as you suggest, then I have some rights to use it that you can't withdraw. You can yield any and all of your remaining rights to the publisher or anyone else.

So, the situation is exactly as before. You can give a publisher copyright, but only the rights you retain after giving a license. The publisher will understand the implications as long as they know of the license. And, you should inform them to avoid problems in the future. Note that the publisher already has license for that photo (or whatever) since you granted it to the commons, of which they are a part.

In short, the boilerplate agreement with the publisher needs amendment before you can sign it. If the publisher isn't willing to do this, then they won't publish your work.

With respect to the no derivatives clause, that still applies to "the commons". But you have the right to make derivative work yourself. You haven't given that up. And if you yield all remaining rights to the publisher then they get that right as well. So there is no conflict. But you then don't have the right to make derivative works thereafter unless the publisher permits it.

Note that CC licenses are irrevocable and hold for the term of the copyright (seemingly infinite these days in the US) and become void (only) if the terms are broken by the licensee.

So, if you want to CC your images before yielding copyright you can do this. That isn't a problem. The problem arises if the publisher won't accept an already restricted copyright in exchange for publishing your work. I suspect that the risk is small but not zero, depending on how much of the "value" of the work is in he images.

Note that the purpose of copyright is actually to retain value to creators for a limited amount of time.

  • 1
    Usually agreements contain a clause where you certify that you are the sole author of the work and you have "full power" to transfer copyright; for instance clause D in the agreement linked by OP. You would be violating that clause, I presume? Commented Jan 12, 2022 at 14:00
  • 2
    @FedericoPoloni, you need to inform the publisher that license has been given. This is the same whether it is you or another that has given the license. You can't give what you don't own. The agreement requires amendment.
    – Buffy
    Commented Jan 12, 2022 at 14:05
  • 2
    Correct, but I suspect the OP would be happier with some sample language for amending the copyright agreement. (I'm not the one who -1ed this answer.) Commented Jan 12, 2022 at 14:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .