73

Suppose Researcher A did some joint work with Researcher B. While the work was still in progress, or the manuscript was in preparation, researcher B became persona non grata in Academia, due to wrongdoing. I don't think the exact nature of wrongdoing is relevant here, but let's say it is serious, as in for example:

  • faking research results (the research with A was not faked)
  • severely breaching research ethics (the research with A was not affected)
  • sexual misconduct towards subordinates (that Researcher A was not aware of during joint work)

What's Researcher A to do? I see four options, all of which have their ethical and practical drawbacks.

  • publish the result jointly with B as planned. The ethical drawback is that it can be perceived as siding with B in the scandal/breaching a boycott, and the practical one is that Researcher A's career might suffer from association with the scandal.
  • not publish at all. The practical drawback is that Researcher's A CV suffers, and the ethical drawbacks are obvious: not publishing a worthy research funded by taxpayers is a waste of their money; it may hinder further progress of the field, and, for example, in Maths, if results have been announced or communicated to the community, then codes of conduct explicitly require that the details are published soon.
  • withdraw Researcher A's name. The practical drawback is as above, and the ethical drawback is that generally nobody should receive full credit for the work that was in fact joint, and even less so as a "reward" for the misconduct.
  • ask/pressure Researcher B to withdraw their name. The practical concern is that Researcher B may not agree, and the ethical one is that, again, Researcher A should not get more credit than their contribution to the work, and wrongdoing should not disqualify B from getting their share of credit where it is due.

So, what is the right course of action for A, ethically and practically?

19
  • 20
    The question is very interesting, but you already give a self-contained answer. There is no perfect solution, all solutions will have substantial drawbacks, and one can only weigh-off the tradeoffs to pick the solution that aligns best with one's personal values. – lighthouse keeper Nov 23 '20 at 9:52
  • 57
    The three examples of misconduct are quite different, actually, and probably relevant. – Buffy Nov 23 '20 at 10:46
  • 6
    Also has researcher A published with researcher B before? If so, as researcher A is already associated with B i might suggest being inclined to still publish (as long as A knows for sure that B didn't fake the data or breach ethics on this paper). – Rob Nov 23 '20 at 15:55
  • 18
    faking research results (the research with A was not faked) at least in some fields it would be very hard to say this confidently. And you would need to be extremely confident that you are correct in your assumption -- if you went ahead and published with someone you knew to be deceitful in some of their work , even a minor problem with your paper is going to make you look very bad. – eps Nov 23 '20 at 19:23
  • 9
    Unfortunately the type of misdemeanour is relevant although your reticence is understandable. For example fake results will result in all of B's work being suspect. It won't help much to say that in your case things were different. People will simply shy away from citing B's work. Thus publishing may result in your work being ignored. It's a really tough situation. – chasly - supports Monica Nov 23 '20 at 22:32
45

I expand on a comment by Buffy where this was hinted at:

It is a very different issue whether the misconduct was on the scientific level, which would cast doubt on the results, or on the social/moral level.

In the first case, it is difficult to trust the results if B is on the author list. Here, we have a serious dilemma. Probably OP might want to consider to cut their (and everybody's losses). Perhaps there is a way to publish it that makes it explicitly clear that this work has not been tainted by misconduct (editors note etc.).

To the second point:

Reiser's filesystem, Bieberbach's or Teichmüller's achievements are not devalued by their personal moral failings.

In this case, the research is done. It should be published. I do not know how important or impactful it is, but in principle, it's always a service to science and humanity, not just the respective taxpayer. Ultimately it's the OP's decision whether they are prepared to accept possible flak for their step. The research is separate from the authors behaviour. It's their detractors that would be trying to mix science and mores.

The only exception I can imagine for research to be suppressed for moral failures would be if the data were obtained in a clearly unethical way. The purpose of this is to discourage incentives for future breaking of ethical rules for obtaining it.

20
  • 5
    A filesystem is a code that can be tested by someone much less skilled than its author. A theorem is a sequence of arguments that can be understood and verified by someone with close skills and expertise as the author. Dishonest authors can quickly produce a stream of faked results, that will take years to verify and disprove. That's why academic reputation is so important. – Dmitry Savostyanov Nov 23 '20 at 13:02
  • 23
    @DmitrySavostyanov Exactly. Scientific and extrascientific moral failures are of a different kind and should be treated separately. People can be highly flawed human beings, but their works may still be valid if their flaws are not on the professional level. – Captain Emacs Nov 23 '20 at 13:05
  • 6
    @DmitrySavostyanov Yes, indeed. That's precisely why I make a specific remark taking into account whether the OP is ready to take flak for their position. The fundamental science-ethical question does not or should not - in my opinion - depend on the mob's reaction to it. What one chooses to do, will have to depend on personal courage, position, stage in career, freedom of operation and other factors. I wouldn't blame anyone for taking the safe route, but respect whoever takes the uncomfortable one. – Captain Emacs Nov 23 '20 at 17:35
  • 4
    @user111388 Reiser was found guilty for murdering his wife. Googling "Reiser file system" gives his story on the wikipedia page as second result. I used Bieberbach and Teichmüller to offer concrete examples for research that should not be banned because of the people behind it even under today's criteria. This statement was completely independent of whether it would have been punished in their time. I thus stand by the example. (As aside: Teichmüller effectively "committed suicide" by volunteering for the Eastern Front). – Captain Emacs Nov 24 '20 at 16:03
  • 4
    @ljrk I guess if you apply that rule strictly to past researchers, you will have very few left standing. That will be one great culling. Elimination of achievements is one of the hallmarks of totalitarian ideologies. I believe to remember - but cannot reconstruct the case - that one of the totalitarian states of the 20th century did (try to) erase the achievement of an "undesirable" and instead attributed it to a ideologically amenable one. I.I.Rabi's advice to Feynman comes to mind, paraphrased: "Do not use something good that somebody does against that person, even if they have many vices." – Captain Emacs Nov 24 '20 at 19:10
11

Recommend publishing in both authors true names and modify the paper to include information to make it easier to verify the results. Make the data sets available, publish all the source code, etc. Don't hold anything back.

2
  • 12
    Arguably, this is what any paper is expected to include these days. – Dmitry Savostyanov Nov 23 '20 at 23:33
  • 4
    @DmitrySavostyanov: Yes, but the answer makes a good point: whatever the field’s expected level of transparency is, it makes sense in this situation to go beyond these expectations and aim for an especially high standard. – PLL Nov 24 '20 at 16:04
7

In your question you considered the pros and contras of all approaches. For example, when you considered the option not publish at all, the main counter-arguments are that Researcher A does not benefit, taxpayer's money are wasted, and community does not benefit from the results. Let me discuss them in a little more details.

  • Taxpayers are not benefitting from the results of the research directly. They benefit from the impact of this research: the practical outcomes, embodied in more effective tools, processes and products.
  • Research community and the surrounding community of engineers, enterpreneurs, buisiness people, etc, pick up promising results and take them further towards development and impact. For this to happen, the results need to be clearly communicated, promising, and trustworthy. People are not likely to invest time and resources in results they can't trust.
  • Researcher B commited an academic misconduct and compromised their reputation. The compromised reputation taints the trustworthiness of all their results, past and future. As a co-author, Researcher A might know that the results of their joint paper are not affected and trustworthy. But for the community and general public establishing the trustworthiness is not simple, and they have to rely on reputation instead. Unless Researcher A has a lot of reputational credit (i.e. they are Professor Famous), their word alone is not sufficient to wipe the negativity brought by the Researcher B being part of the team.

Perhaps, the best course of action would be not to publish the research done together with the Researcher B. Instead, Researcher A should use the skills they learnt while working on this project to obtain new results and benefit their career. Finding right people for collaboration is one of critically important skills for success in academia.

7
  • 3
    That's a good point, though not all types of misconduct compromise B's credibility as a researcher. – Kostya_I Nov 23 '20 at 11:50
  • @Kostya_I Even if B's misconduct does not compromise them in A's eyes, the views of the general public might differ. – Dmitry Savostyanov Nov 23 '20 at 12:56
  • Actually, after some thought I don't really understand your argument. The ethical pressure to publish stems from A's belief that they have something valuable to give to the community/humanity. How does the fact that the community/humanity might be unwilling to take it affect this at all? If the research fails to make impact it merits, at least A tried. What is gained by not trying? – Kostya_I Nov 23 '20 at 21:22
  • So, A believes they have a valuable gift for the people. But people don't really trust that the gift is safe, and don't want it. But A believes the gift is valuable and safe, and people are just not very smart to see it. So A forces the gift down people's throats. At least A tried. What is gained by not trying? – Dmitry Savostyanov Nov 23 '20 at 22:41
  • 3
    publishing a scientific paper is not forcing anyone anything, even to read it, much less take down their throat. And it is not granted that people "don't really trust the gift is safe and don't want it" before the paper is published. While people might use authors' credibility as a proxy, eventually scientific papers are evaluated on their merits. Important or promising findings will be independently verified/replicated, and it's by definition not true that people don't want it. – Kostya_I Nov 24 '20 at 8:26
7

Publish your contributions as separate sole author papers

  • note this will only be possible in some scenarios, but it is important to present this option, because it is often the best one when it is possible

Remove the content contributed by the offending co-author and publish the partial work without them (most likely in a lower impact journal). This may not be possible if your independent contributions (your ideas, analyses, and writing) do not make up a significant contribution to the literature without the content from the other author. But often the partial work is worth publishing, even if not as complete as you would like it to be. Also you might be able to take the work in a slightly different direction with a minor revision.

This solution is actually quite generous to the co-author because they are then free to publish their contribution as a separate publication, alone, as well. They can build off of your work, and additionally you get a citation. Science suffers the least in this solution. The only cost to readers is the mental cost of having to read and find the work in 2 papers rather than one. While your paper may be less impactful than the original piece, a sole author publication can also be beneficial on the CV. This solution means you get to avoid the negative consequences of the association, but you ethically aren't doing anything wrong either. If you want to be really transparent you can even add in the acknowledgements that you are grateful for conversations with X on dates Y to Z. Note if Y - Z is before the scandal is revealed, it also provides evidence that these conversations occurred before when you might be made aware of the scandal. In addition, almost no one on hiring committees looks at the acknowledgements on your papers. So your reputation would be quite safe here.

You should of course discuss this with your co-author, but they might be more likely to cooperate with this approach than some of the other options. But ethically, this is the only option (besides not publishing) where you can proceed even without your co-author's permission, as long as you don't publish their contributions. Be sure to save all emails from the collaboration, so you can prove that these were your ideas.

4
  • 1
    The OP doesn't give details about the field or the particular project, but in nearly all truly collaborative projects, it wouldn't be feasible to carve out papers separately from it, based solely upon author contributions. Neither paper would likely be comprehensive or coherent, and hence independently publishable. So this doesn't seem like a generalised answer to the question. – Michael MacAskill Nov 24 '20 at 1:36
  • @MichaelMacAskill In nearly all purely theoretical or computational fields, which is a quite expansive set of disciplines, this can be an option. As you mention the OP doesn't state their field, so it's important to provide answers that would work in various different fields, not just fields with lab or field based experimental work. Additionally, the answer specifically states "This may not be possible if your independent contributions do not make up a significant contribution to the literature without the content from the other author" – WetlabStudent Nov 24 '20 at 1:44
  • 1
    “In nearly all purely theoretical or computational fields […] this can be an option.” What fields do you have in mind? At least in my experience of pure maths and theoretical CS, it is pretty rare that a collaboration can be disentangled to this extent. – PLL Nov 24 '20 at 16:06
  • @PLL I'm thinking mathematical/compuational biology, engineering, physics and other math modelling disciplines. Often a purer paper describing the model and proving some the model's properties could easily be published in an applied math journal, where as an applied paper that uses the model to answer some novel applied questions might go in a science journal. The combined paper might be a very high impact general science paper, but the two papers are easily disentangled, especially if the authors cooperate. – WetlabStudent Nov 25 '20 at 6:51
3

Publish under pseudonym(s)

One or both of you could use pseudonyms. As with all other options, this is a compromise, but in this case you're emphasising getting the publicly funded and/or important research out there, while avoiding further association with a person you don't want to be associated with, and not allowing them further credit, while also not tarnishing the research by association with their name. Of course it would be improper to force them to publish under a pseudonym, so it's still their choice, and if they didn't agree then it might be obvious who you are if you're continuing previous work.

12
  • 9
    This may well be considered unethical in its own right, because the effect is to directly dodge the negative reputation of B. Think about what happens when it's revealed that "C" is actually B, to the reputation of A. – obscurans Nov 23 '20 at 22:35
  • 3
    @obscurans to be clear the model I have in my head is B discovered to have committed serious sexual misconduct, or murder or something else beyond academia. A & B have pending work close to publication which is of some significance, and A & B both agree to publish pseudonymously then sever ties. I might be wrong, but I think that if both A & B were exposed, "I thought this work was important but that B's reputation should not benefit from publishing it" would be acceptable as a defence. – llama Nov 23 '20 at 22:48
  • 4
    @obscurans pseudonyms have been used in science for as long as science has existed. Knuth once published pseudonymously precisely to ensure that his paper was reviewed properly, and not just by the strength of his name. Sure, they're not ideal, but they're not inherently unethical or unworkable and it's ultimately up to the authors in such a situation to decide whether it's an appropriate solution or not. – llama Nov 23 '20 at 23:33
  • 3
    Yes they have been used. In a situation where you're deliberately dodging negative reputation... I think you'll have a hard time defending it. – obscurans Nov 23 '20 at 23:38
  • 1
    @obscurans "..if B's crime was nonacademic, it doesn't diminish their credit": The problem seems to be that I (and OP, and this answerer) believe this is not true as you get evaluated by people (who are driven by emotions), not by machines. Of course, if there is a time gap of 10 years between the research and the misconduct, those feelings have probably diminished. But if you sleep with my wife today, I will not be able to review your papee next month properly, no matter how much I try (and no matter how much at fault my wife is and you are not). I belive that if, say, Jana aus Kassel – user111388 Nov 24 '20 at 17:01
-2

Publish the result jointly with B as planned. There is no hesitation here. If you want ethics, here's a teaching from the Bible Matthew 22:15-21:

Give to Caesar what is Caesar’s, and to God what is God’s.

Did Researcher B contribute his value to the study? Yes, he did, according to what you said.

Did Researcher B commit any crime within your joint study? Faked results, etc.? No, he didn't, according to what you said.

So, the reward of having the result jointly published with you (under his real name, for god's sake) is his, isn't it?

It's not your place to judge him or to take this from him - you are not the judge, right? Not even judge should have that right. Should all your contributions and credits be deleted when you commit a crime? Does it sound like justice?

Now, to your concerns:

The ethical drawback is that it can be perceived as siding with B in the scandal/breaching a boycott, and the practical one is that Researcher A's career might suffer from association with the scandal.

OMG... this is based on a fear of being judged by a very sick society. Only very sick minded society would make these judgements.

You have to make a choice, if your decision shall be driven by real ethics, or fear of being judged by sick minded people.

8
  • Yet, people in this society are "very sick". People are usually driven by emotions. I sincerly believe that a physics paper from someone who publicitly says Corona doesn't exist and people should meet and kiss a lot and cough onto each other would not be properly credited, by (mostly unconsicous) biases people have. – user111388 Nov 25 '20 at 13:30
  • @user111388 that wouldn't be science then, that would be (rather dirty) politics. Scientist should have enough ethics to pursue the truth and make his decisions free of politic bullshit. – Tomas Nov 25 '20 at 16:13
  • @user111388 and yes, some people in this society are very sick, but do you want to take part in spreading that sickness with unethical decisions? – Tomas Nov 25 '20 at 16:17
  • 4
    The bible also says things like "A man or a woman who is a medium or a spiritist must be put to death. They are to be stoned; their blood is on their own hands" if anything quoting it in reference to ethics is just muddying the waters here. If you have a valid point from an academic point of view you're welcome to bring that forward instead. – Lio Elbammalf Nov 25 '20 at 16:30
  • 2
    You seem to think that a decision not to have anything to do with a person because of their moral failures can only be taken out of a fear of a mythical "very sick society", but that is merely a convenient way to denounce such a decision without a need to address its ethical substance. – Kostya_I Nov 25 '20 at 17:05

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.