17

When I was a PhD student, I wrote a collection of papers that comprised a large portion of my dissertation. These papers were written using grant money from one of my advisors. (I had two advisors. The "second" advisor funded me for two semesters).

After I had graduated, this professor expressed a desire to collaborate with me on publishing the papers I had written as part of my dissertation. However, he wants me to entirely re-write each paper in "new" language in order to make the publication truly double blind. He fears that if someone were to Google phrases from my paper, that my dissertation would appear and that the authorship would be known (thus destroying the double-blind process). Because this professor is only topically familiar with my research (enough to be on my committee, not enough to actually write the papers), I am extremely hesitant to even suggest that he do any re-writes himself. (In fact, he attempted to re-write a new draft of the paper and it contained numerous errors in terminology and theory).

I am currently employed in a private research setting (i.e. non-academia) where publication is still important, but I am not free to devote significant amounts of time to my own research. (I have specific, employer directed, research I need to perform). As such, I am not able to allocate numerous hours to re-writing a paper that I already wrote and that I feel is currently in "submission-ready" form.

The ultimate reality is that Google is a powerful tool that could locate my research no matter how hard I tried to obscure it. I cannot take down my LinkedIn, arXiv, and ResearchGate accounts (plus a website) just in case someone wants to devote hours tracing research back to me. Besides, the whole point is to connect myself to my research and make it easier for people to find my research.

Several of the potential journals I am looking at submitting to require double-blindness.

I have two main questions:

Is it usual practice for journals/referees to Google sentences from papers and attempt to determine authorship?

Should I feel inclined to re-write each paper in "new" language in order to avoid anyone from being able to trace my work back to my dissertation?

Added: There have been some questions about authorship on this paper. I am the first author on this paper. The professor in question here contributed what I would consider the bare minimum to be granted recognition as an author. I have already gone through a protracted dispute with my former university over the matter of authorship and they decided that since it was my professor's word against mine as to exactly how much he contributed, they would rule in favor of the professor. (It's much easier to tell a student no than a professor). I was instructed to include this professor as an author on the paper by university administration. I am not directly concerned here with who should be named author on the paper. At this point, dropping the professor from the paper is not an option.

  • 7
    There are very few journals in math that even use double-blind review; single-blind is much more common (the referee knows who the author is, but not vice versa). Are you sure that your journals of interest actually use this process? – Nate Eldredge Aug 28 '18 at 22:59
  • 2
    @NateEldredge A few journals I am familiar with in statistics use double-blinding, namely Annals of Statistics and Journal of the American Statistical Association. – Vladhagen Aug 28 '18 at 23:07
  • 5
    @Vladhagen That doesn't answer Nate's question: do the journals you're actually interested in publishing in use double-blind review? Do they use it exclusively, or do they allow authors to choose single- vs double-blind for their manuscripts? And, if you're allowed to choose, do you (and your co-author) really assess that you will benefit from the double-blinding? It can indeed be beneficial, but plenty of fields get along just fine with single-blinded reviews. – E.P. Aug 29 '18 at 13:10
  • 3
    @Vladhagen This doesn't change the conclusions ("Several of the potential journals I am looking at submitting to require double-blindness") but I'm having trouble locating the Annals of Statistics policy that requires double-blind manuscripts (this points to a single-blind process). If there is such a policy, it will likely have further details. – E.P. Aug 29 '18 at 18:44
  • 2
    @Vladhagen Frankly, if there is a journal of suitable calibre that allows single-blind submissions, then that is the obvious choice to completely sidestep this problem. If your dissertation is already available, I don't think a proper double-blinding (which is actually honest, and doesn't open the door for an accusation of plagiarism from reviewers) is even possible. – E.P. Aug 29 '18 at 19:22
30

Check out double-blind peer review policies from the venue you'll be publishing in. Here's an example from Elsevier. In a nutshell, you want to prevent any reader from immediately connecting the paper to your identity by e.g. not listing your affiliation, anonymizing self-references, removing funding sources, and so on; but you're not aiming to "beat Google".

  • 2
    This sounds sensible to me. If the journal has double blind reviewing then remove any obvious identifiers. But I think re-writing the whole paper is unnecessary, and too much work - I don't imagine many reviewers would bother googling phrases from the manuscript to try and identify you. – rw2 Aug 29 '18 at 8:36
  • 4
    I would actually treat it as foolish to rewrite a paper just to obscure authorship for review. Hopefully you used the best and clearest phrasing when you originally wrote it. Re writing for obscurity can only make it worse, I think. It is the responsibility of the reviewers to keep the process valid without extreme measures from authors. – Buffy Aug 29 '18 at 19:37
10

I would think that this would be an unusual practice, at least in the field of mathematics. Almost all journals would allow for publication of results previously presented by an author in his/her dissertation. There may need to be appropriate attributions made, and you likely could not place the published version of the paper in your dissertation, but if your dissertation is already completed it would seem this would not be an issue.

I published two papers based on my dissertation in abstract algebra and character theory. The topic was such that someone could Google the titles of the papers and usually find my dissertation as well. The two texts are not identical, but they are certainly similar enough that an alert human could quickly determine the authorship of the papers given my dissertation, even without definitively knowing I wrote both papers.

Some questions I have:

If the co-collaborator does not know enough to re-write the paper in "new language," why is he even being given authorship?

Is the topic specific enough that the paper could not be somewhat feasibly re-written in places in order to make it slightly different?

Based on the fact that this paper is in your dissertation, I assume you are the principal/first author?

  • This collaborator is willing to pay the publication fee. And I also am just trying to avoid getting drawn into a dispute about authorship. Otherwise I would just tell the former advisor I do not wish to piggy-back him along on this publication. – Vladhagen Aug 28 '18 at 23:30
  • 2
    "Paying the publication fee" is not a valid reason for co-authorship, just like "paying for your holiday" wouldn't be. Most respected journals will actually make the authors confirm, upon submission, that they have intellectually contributed to the paper; and even if this is not explicit in the agreement, it is certainly an implicit expectation. Both you and your "collaborator" will be in breach of community norms, and possibly of journal policies, by submitting such "co-authored" papers. – Alex B. Aug 29 '18 at 8:34
  • @AlexB. His contribution is more than just paying the publication fee. (Although......I do still believe there would be an argument that he does not warrant authorship). Because I was his graduate student, I will have a hard time showing definitively to my former university that he does not deserve authorship. He came to many meetings on the paper. He just gave no feedback and doesn't know the details of the paper. It becomes my word against his in an authorship dispute. – Vladhagen Aug 29 '18 at 14:40
  • It is not uncommon for collaborators to not be able to defend an entire paper, especially if they became collaborators to fill in each other's weaknesses. For instance, programmers and statisticians often consult and provide meaningful contributions to papers in fields they are not experts in. – TimothyAWiseman Aug 29 '18 at 15:45
  • 1
    @TimothyAWiseman Such is not the case here. He is being given authorship because he has available money and sat through some meetings on the paper. That being said, the issues of who should be named an author are not my direct concern here. – Vladhagen Aug 29 '18 at 18:05
10

A low-quality answer, but: in mathematics, my field, spending time to obscure the authorship of a paper is misguided and un-necessary: established people have a viewpoint and style that would be unmistake-able, for example.

More generally, the goal of work/research is not anonymity, but progress in our collective understanding. Notably, this involves people, not anonymous entities. Referees can barely be anonymous, and it's even harder for authors to be anonymous. This cannot possibly be a high priority...

  • 2
    A short answer, but not a low quality one. – Buffy Aug 29 '18 at 19:35
1

It sounds like you're having a relationship with your co-author that's not very far from the border of turning into adversarial, and I don't think you'll gain much from shying away from that aspect. You have enough high cards on your hand that I think you can get things to a situation where your co-author has little choice but do what you say. And, ultimately, it's your dissertation, and he's frankly pushing things with the amount of undue work that he's demanding of you.

One of the high cards that you have is that at least one of the journals you're considering has an explicit set of criteria of what it considers necessary in a double-blinded manuscript (specifically, this policy for JASA). As such, I would propose as a course of action:

  • double-blinding to the explicit standards set by the journal
  • emailing your coauthor to tell them that the manuscript's current form satisfies the explicit journal standards, and that it is therefore ready to be submitted
  • opening the door to their suggesting edits and re-wordings, but in doing so
    • making it clear that you would have to give full approval to the text
    • making it clear that you would consider suggestions that included technical inaccuracies along the lines of their previous draft to be a waste of everyone's time, and
    • making it clear that you have a pressing need for this paper to be submitted

and then making that last point explicit by directly setting a short deadline (two weeks? three weeks? read: short enough to be infeasible from him) for that re-write. This course of action puts the ball on your co-author's court, and it puts it on them to rise up to the bar that they're setting, so that if they are unable to do that, then it's because they set the bar too high. And if the coauthor decides to raise a stink with your university administration, then you can simply say that he's being intentionally difficult, making requests that fall outside of what the journal requires, and failing to live up to the expectations that he sets.

And also: the 'negative outcome' from your coauthor's perspective is that a paper is submitted in a form that is less perfect than it could be. This is simply something that happens, all the time, when life intervenes into research. It is not a reason to stop a publication from being submitted.


As an alternative to the above, you should also consider taking JASA up on their word:

For answers to specific problems not addressed in these guidelines, please contact the ASA Journals Department, journals@amstat.org.

Explain the paper's situation to them (specifically, where and how your dissertation has been published, and what the relationship is between your paper and your dissertation), and ask them to make a determination of how much blinding is required. (You could even send them a preliminary manuscript, or a sample of it, and ask whether they feel the text is too close.) If they say that your text is fine, then you have extremely strong textual evidence to force your co-author to do what you say, or to convince your university that he's acting unreasonably.

If they say that the text requires re-writing, then it may be that that journal is not for you (i.e. that their policies are such that they don't in fact support dissertation-based papers, or that you should have set a publication embargo on your thesis when you submitted it). In that case, either own up to the re-writing, or move to another journal that considers your non-re-written blinding to be sufficient, or which allows for single-blinded reviews.

And in that process, of course, you have the high card that your co-author's say on where and how that paper gets submitted is frankly very limited. You seem to be forced by circumstance to include them, but if you say "we're submitting to X journal in Y weeks at the maximum", what's the worst thing that your coauthor can do?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.