We have had a paper accepted with Mathematische Annalen (Springer). As my university has an open access agreement with Springer, the paper will be published as (gold) open access, with a Creative Commons license. However, when asked to sign the license agreement, it contains the following paragraph:

Immediately after acceptance the Author may deposit the Accepted Manuscript to any location, and under any terms (including, but not limited to, under a CC BY licence), provided it is not made publicly available until after publication. The Author will include an acknowledgement in the Accepted Manuscript, together with a link to the Version of Record on the publisher’s website: “This version of the article has been accepted for publication, after peer review (when applicable) but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/[insert DOI]”.

If the paper is being published open access (which is paid for through the licensing agreement), why are we asked to agree not to upload the accepted manuscript? Why should we be restricted in the wording used in how to acknowledge the "version of record"? Why should we not be allowed to publish it with post-acceptance improvements, if we retain the copyright?

[There is also a paragraph about the "submitted manuscript", stating that this may be made available but that the author must include a specific acknowledgement that this manuscript has not undergone peer review or any post-submission improvements.]

I don't recall such a requirement in previous Open Access agreements. It seems that it would potentially leave us with fewer rights to use the article in which we retain copyright than others who are free to use it under the CC BY license!

Has anyone else seen this type of wording, and experience of how to deal with it? My first reaction is to strike out this part of the license agreement and sign (by hand) without it.

  • 2
    @Roland Well, it seems that by signing the agreement, we would agree to not exercise these rights that the CC-BY license would otherwise give us, when others are not restricted in this manner. I mean, either that or the corresponding paragraphs are meaningless anyway. Commented Nov 14, 2023 at 10:03
  • 1
    @FedericoPoloni The proposed acknowledgement isn't an attribution. Attribution would need to specify a creator and the paragraph doesn't do that.
    – user9482
    Commented Nov 14, 2023 at 12:23
  • 1
    Sounds like the person-in-charge at Springer sent you the wrong license agreement. I'd just write to them and ask them to clarify.
    – Allure
    Commented Nov 14, 2023 at 14:23
  • 1
    The agreement would, if you release the accepted version under an open access license, restrict the ways in which you can distribute the accepted version, restrictions that others are not subject to. However, the agreement does not restrict your use of the version of record.
    – Anyon
    Commented Nov 14, 2023 at 14:52
  • 1
    @Allure This actually is verbiage Springer puts in some OA license agreements. See e.g. faimconference.org/images/FAIM_2022_LTP_OA_SN_Switzerland.docx This is
    – Anyon
    Commented Nov 14, 2023 at 14:53

5 Answers 5


While not a seasoned academic (or copyright lawyer), my reading of that statement is that these rules apply to the article before it is published (i.e., while it is still an 'accepted manuscript'). Once the article is published (after typesetting or in a full issue) you would be relatively free to use it how you like (assuming, of course, it is referenced correctly, like any other article would be).

I think this paragraph of the agreement is mainly to cover them in case of changes made to the manuscript during the publication process. Once it's published, everything is fixed in place; before it's published, minor changes can occur in figures or formatting or data, etc. Most early access papers do the same thing by putting huge watermarks saying:

This version of the article has been accepted for publication, after peer review (when applicable) but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/[insert DOI]

(or something like that). Take a look at any 'accepted manuscripts' in any journal and that is what you will see. For example, one article I just looked up in Chemical Science has this in the margins of each page:

accepted manuscript watermark

(I've only posted the margin because I don't know about whether I'm allowed to post a full page, even if it is open access...)

Again, I'm not a seasoned academic (still a few months from getting my PhD), but that is my interpretation.

  • Also not a seasoned academic or copyright lawyer, but this was my interpretation while reading the question.
    – TripeHound
    Commented Nov 15, 2023 at 20:38

To answer the "why": Springer (and other publishers) want people to use the published version, which has their branding, and preferably to get it directly from them, increasing their download statistics. Letting people use the accepted version, even though it brings in exactly the same direct revenue (i.e. none), would make it difficult to justify that the published version has added value, and jeopardise the continued willingness of institutions to pay open access fees.

What they'd like to do is stop people posting the accepted version, but they know that's not going to happen. So they instead are trying to impose a more "reasonable" condition which will still mean that most people end up looking at "their" version of the paper. They probably aren't concerned about whether it is actually enforceable since they don't want to antagonise authors by enforcing it. They just hope most authors will respect the conditions, and accept that a few people will either ignore them or find a way around them.


I don't think that those restrictions are illegal, because they are a private agreement between you (the copyright holder) and Springer. However, this sort of clauses are moot because they are very easy to circumvent.

Step 1: deposit the accepted manuscript on a USB stick, and give it to a friend. Tell them that you license the work to them under CC-BY. This is your right, because giving the manuscript to a friend does not make it publicly available.

Step 2: ask the friend to upload the manuscript to your repository of choice and make it public. They can do it by following CC-BY, and they are not bound by any additional restrictions, because CC-BY explicitly forbids such restrictions.

  • Academics should not do things in bad faith. Commented Nov 14, 2023 at 12:23
  • 6
    @ScottSeidman This is not bad faith; this is applying CC-BY to its letter and to its spirit. If anything, the poor attempt to add additional restrictions to CC-BY is bad faith from the publisher (and bad lawyers). Commented Nov 14, 2023 at 13:05
  • 1
    But it does seem like intentionally subverting the restrictions and, perhaps, poisoning the relationship with the publisher. Using an intermediary to commit murder isn't a defense, I'll note.
    – Buffy
    Commented Nov 14, 2023 at 13:39
  • 3
    @ScottSeidman I am honoring the agreement to the letter. Commented Nov 14, 2023 at 17:23
  • 2
    And let me add that if there is someone intentionally subverting restrictions here it's Springer. Open access should mean open access, and CC-BY explicitly forbids additional restrictions. Commented Nov 14, 2023 at 18:24

I have seen this type of wording. My university has (probably) the same agreement with Springer. I have normally put the accepted manuscript (without post-acceptance changes) in a repository. I thought that's good enough for me. I have occasionally forgotten to add the requested link/information, and in other occasions I haven't looked up the precise wording and just wrote something to this effect myself. Nothing bad has ever happened because of this.


Thank you for all your comments. I have reflected some more on this, and it seems that the requirements in the agreement are essentially toothless, because it does not and cannot restrict you from uploading some version of the article other than the accepted version. After all, the copyright remains with the author.

The only issue could arise if there is some other entity (such as a funding agency) that explicitly requires you to make the "Author Accepted Manuscript" available after acceptance without an embargo period. At least in the UK, this is not currently the case - for "Gold" open access papers, there is no requirement for the AAM to be made available also.

So I think in the end such an agreement does not genuinely restrict us as authors on how to use our articles, but it still leaves a bad taste in my opinion.

  • As I said in my answer, I don't think the aim of the comment is to restrict your use of the article, and that you will be free to do whatever you want with it once it is typeset and published. Before that, they just want to make sure you make it clear that a journal has already accepted it, and it does not represent the final form of the article. That is literally all they are saying—use it as you wish, so long as you add the disclaimer that it is not the final version of the article, and the journal in question has already claimed it. You're looking for restrictions where there aren't any. Commented Dec 18, 2023 at 2:57
  • @isolatedmatrix I respectfully disagree. As long as I have published, there have been terms in copyright and (more recently) open access agreements about acknowledging the publication of the paper in a journal. This text is new. Just because the restrictions are ineffective, does not mean that they are none. Commented Feb 5 at 10:29

You must log in to answer this question.

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