Does copyright restrict how I can adapt papers when compiling them into a thesis?

Question

I am starting to put together my PhD thesis. The bulk of the thesis will be reprints of papers I have published during my PhD. I am wondering about presentation style.

My personal preference would be to reformat all of my papers into a single LaTeX document so as to have consistent style throughout. It's a bit of extra work, but I would prefer this approach. In several cases my papers have supplementary info which would not fit in the constraints of the main manuscript in the published version, but which could (and probably should) be coherently woven into the main text.

Are these acceptable options? Assuming I get reprint permission from the publishers, is it acceptable to change or partially reorganize the text for the sake of a coherent narrative within the context of the larger thesis? Or do copyright laws require that my reprints be exact reproductions of both content and style as compared to published versions?

Years ago I received a copy of a Danish thesis. There was a cover sheet with the thesis title, author, university, etc., and stapled behind it was a half dozen reprints of journal articles. Sadly, I did not manage to emulate that, since my university did not approve of it. Sigh. — Jon Custer, Mar 21 '18 at 14:30
I mean, on the one hand nobody is ever going to read it again so who really cares how it looks, but on the other that seems like a cheap way to do something you put 4+ years of your life into. — KBriggs, Mar 21 '18 at 14:35
What you describe is good practice. Making the formatting, sectioning, etc. uniform, and writing an intro that ties everything together, is a useful exercise, even if no one reads the whole thing. — , Mar 21 '18 at 14:43
Sure, but will copyright law allow me to alter the published version for the sake of style while still reprinting large sections from it verbatim? — KBriggs, Mar 21 '18 at 14:45
@KBriggs - the 4+ years was to do the work and get the PhD, not to have a beautifully formatted thesis that sits on your (and your parent's) shelf. In physics/materials science people read the papers, not the thesis. Other fields may be different... — Jon Custer, Mar 21 '18 at 14:51
@KBriggs I'm not a lawyer, but I've known many people do this sort of thing in their thesis and not get in trouble. Just cite everything, and be clear about what has been reworked and to what degree. — , Mar 21 '18 at 14:52
@Jon Custer you're certainly right, thanks. I don't think it's going be all that much extra work, though. Maybe an extra week or two to translate my papers into LaTeX and get the references right. — KBriggs, Mar 21 '18 at 14:55
@user37208 I assume that's the case, since I don't see why anyone would go after someone for the content in their thesis. Just wanted to be sure! — KBriggs, Mar 21 '18 at 14:55
Re. the copyright question, this may be of interest: https://academia.stackexchange.com/questions/71135/chapter-in-printed-thesis-vs-journal-article/104040#104040
In most countries, use of copyright materials in a thesis is permitted via statutory exceptions to copyright. However, this could limit your ability to publish your thesis as a book (common in some countries, I think) or to distribute it online. Making sure you meet your institution's rules on self-plagiarism is probably more important than copyright, in your case. — Wandering Chemist, Mar 21 '18 at 15:05
Maybe an extra week or two to translate my papers into LaTeX and get the references right. – I am so grateful to be in a field where papers are written in LaTeX from the beginning. For me, the respective process took about an hour. — Wrzlprmft, Mar 21 '18 at 15:09
@KBriggs if you don't have much formulas, you could export the (Word? Libreoffice?) documents as HTML and then use Pandoc to convert them into LaTeX automatically. — Andrea Lazzarotto, Mar 22 '18 at 00:32
@AndreaLazzarotto: Pandoc does a great job of converting a Word document to a starting point for a LaTeX document, but your job doesn't end there. Pandoc's not perfect, and you'll probably have a couple of hours of work ahead of you (for, say, a ten-page paper—I speak from experience). — wchargin, Mar 22 '18 at 00:54
@JonCuster: this is a way to do your thesis in France and Belgium (thèse sur publications) — WoJ, Mar 22 '18 at 07:42
@wchargin, and that's exactly my point. It's WAY quicker than retyping everything manually. — Andrea Lazzarotto, Mar 22 '18 at 11:23
I did the first one this morning manually. Took about 2 hours. Really not that bad. I'd rather have a LaTeX document I made myself and know all the packages and such involved over something automatically generated which might have interactions I am unaware of. — KBriggs, Mar 22 '18 at 16:28
@KBriggs, unless you explicitly ask for a standalone document, Pandoc doesn't even import any package. It only gives you standard LaTeX code for the document body, leaving the preamble as a responsibility of yours. I'm not criticizing your choice which is absolutely fine, just pointing out how the conversion works. :) — Andrea Lazzarotto, Mar 23 '18 at 00:01

score 19 · Accepted Answer · answered Mar 21 '18 at 15:22

19

I disagree with part of the other answer, in that the copyright holder of a work is not always the creators/authors. For some journals when you publish with them and sign the licensing agreement, you transfer the copyright of your work to the publisher. Then it is up to the publisher to decide how you can reuse your paper. I would advice reviewing the copyright/licensing agreement you signed with your publisher for limitations for reuse, and if still in doubt, contact the publisher.

answered Mar 21 '18 at 15:22

sl1129

430
1
3
10

Good point. I will check carefully with the journals I have published in, thanks. – KBriggs Mar 21 '18 at 15:26
6

@KBriggs Yes, this answer is better guidance than the other. Note that many journals have explicit allowances for their material to be included in a thesis, since this is a very common issue. – Bryan Krause Mar 21 '18 at 16:54
@Bryan Krause Yes, it seems you're right. There is some variation between journals, but some of the ones I have submitted to explicitly take over copyright. – KBriggs Mar 21 '18 at 17:34
1

It's the norm for journals to take copyright, there are a lot of good legal reasons for that to happen (and perhaps counterintuitively, good reasons for you as well as for them). They usually release back certain rights to you in the process, though, which often covers just about anything you'd want anyways: inclusion in a thesis, distribution of your own copies of the work, etc. – Bryan Krause Mar 21 '18 at 17:39
2

@KBriggs This answer sounds correct for the U.S. Further, librarians at your institution may help you navigate this process for your country AND university--this was part of a dissertation workshop that I attended. – cactus_pardner Mar 21 '18 at 20:18
1

@cactus_pardner good suggestion, I will look into that. FYI, I am in Canada. – KBriggs Mar 21 '18 at 20:21

score 8 · Answer 2 · answered Mar 21 '18 at 18:38

As stated in the accepted answer, you should check with the publisher. However, this issue is often clarified specifically online. For example, AIP Publishing states on their website:

Q: May I include my AIP Publishing article in my thesis or dissertation?

AIP Publishing permits authors to include their published articles in a thesis or dissertation. It is understood that the thesis or dissertation may be published in print and/or electronic form and offered for sale on demand, as well as included in a university’s repository. Formal permission from AIP Publishing is not needed. If the university requires written permission, however, we are happy to supply it.

The same is also true for Springer: "Author retains the right to use his/her Contribution for his/her further scientific career by including the final published paper in his/her dissertation or doctoral thesis provided acknowledgment is given to the original source of publication." Source — mrp, Mar 21 '18 at 21:23

score 2 · Answer 3 · answered Mar 21 '18 at 19:17

Checking with publisher was mentioned, but not how. Factually all large publishers have this "Request permissions" link directly on the page with the paper. Basically, you click on the DOI of your paper and search for this link. Then you are taken to another website, typically copyright.com. There you have to make some choices, such as "yes, I am the author", "no, it's for the thesis", "yes, all the paper", "no, I won't translate". Finally, a price tag pops up. It is typically 0 Euro/US$/whatever, but I've also seen larger values. You agree (or pay, if you are unlucky), then you receive an email that you officially may use this paper for the purposes you stated. Rinse, repeat.

Small publishers typically state the availability of the papers somewhere close to their front page.

If you have not signed the copyright transfer form yet (i.e., the paper is not finally accepted), you are good. If the paper is Open Access (i.e., licensed under CC-BY or similar), you are good. If you have not transferred the copyright, but licensed the content (say, to ACM), you should be good, but better ask via the above copyright.com process.

Wrzlprmft · Answer 4 · 2018-03-21T17:37:51.013

1

Or do copyright laws require that my reprints be exact reproductions of both content and style as compared to published versions?

Why should copyright laws require such a thing? The entire point of copyright laws is to protect creators¹ and restrict what others can do with your work. In the sense of copyright law, you and your co-authors are the creators of your paper and you can do it with it whatever you want (presuming that you all agree on this). Of course, you transfer parts of what is included in your copyright to the journal, but for this you have to consult the respective copyright agreement with the journal and not copyright law.

Sidenotes:

Consider using the accepted versions of the respective manuscripts instead of the published ones. This is usually more compatible with copyright agreements. Also, this is closer to your work, as whatever the journal did to your papers is missing.
Your university, supervisor, and committee may have special ideas as to what you can or cannot do in such a thesis, which are usually more restrictive than copyright.
As already noted, if there is any restriction imposed by copyright it is that your co-authors should agree on this – which should usually not be a problem though.

^{¹ or in some legislation copyright holders, which in turn are the authors by default}

edited Mar 21 '18 at 17:37

answered Mar 21 '18 at 15:03

Wrzlprmft

61,194
18
189
288

2

Thank you. You're right that the university probably has restrictions; unfortunately, being a university, they have not placed these anywhere that I've been able to find them.
Good point about using accepted versions. That's what I had to work with anyway so it makes sense.
– KBriggs Mar 21 '18 at 15:08
10

When you transfer the copyright to a journal (unless it is open access), you cease to be the copyright holder. Copyright law is here to protect the commercial interests of copyright holders, not authors. Remember all the well known conflicts between authors of music and publishers/labels. Such music authors certainly cannot do whatever they want with their music and certainly cannot publish it independently anywhere else. Not without the label's approval. – Vladimir F Героям слава Mar 21 '18 at 16:12
9

This answer is clearly wrong. The laws protect the copyright holder not the author. If you have transferred copyright to the journal it is theirs; not yours, and you do not enjoy special protection as the original author. – Jack Aidley Mar 21 '18 at 16:30
2

@JackAidley The question doesn't specify where this takes place; In many places copyright law does indeed protect the author of a work and transferring the copyright post factum isn't actually possible (this does not preclude the granting of exclusive licenses, in that case violating the terms would be a breach of contract rather than a copyright violation). In those places the journal would still hold the copyright on any modification they made though. – Cubic Mar 21 '18 at 16:59
3

@Cubic: That's interesting, I'm no international lawyer so could you give me an example of jurisdictions where this is the case? – Jack Aidley Mar 21 '18 at 17:13
1

@VladimirF: Right, but if you look at the question, the asker is obviously aware of having transferred (portions of) the copyright to the journal and assumes that whatever they do is in accordance with this. I mentioned this exception explicitly (and now made it even more explicit), but that’s not what I understood the question to be about (and nothing anybody can answer anyway). Also note that you rarely transfer the entire copyright (at least, I have never seen such an agreement). – Wrzlprmft Mar 21 '18 at 17:34
3

@JackAidley: Germany, for example. Also see the above comment. – Wrzlprmft Mar 21 '18 at 17:38

score -4 · Answer 5 · answered Mar 21 '18 at 18:05

-4

Just prepare your thesis the way you see fit and don't think too much about publishers or copyrights.

answered Mar 21 '18 at 18:05

Alexey B.

1,205
1
8
8

3

This seems like a recipe for a lot of frustration – KBriggs Mar 21 '18 at 18:08
@KBriggs This answer is actually fine - so long as your thesis is only ever seen by yourself, your supervisor, and your examiners, and you do not make it public in any way. However, since you presumably do want to make your thesis public, the "just ignore the law" advice in this answer is truly terrible. – E.P. Mar 21 '18 at 21:50
@E.P. Not it is not. Often universities have a specialized office taking care of the thesis formalities. I had to provide for each paper in mine a formal authorization from the publishers. – Zenon Mar 22 '18 at 08:20
@Zenon That's because you are making the thesis public. – E.P. Mar 22 '18 at 09:02
@E.P. I guess by university rules it is mandatory yes. I’m surprised it is not everywhere! Interesting thanks. – Zenon Mar 23 '18 at 01:42

Does copyright restrict how I can adapt papers when compiling them into a thesis?

5 Answers5