Hostage authorship and the problem of dirty hands

Abstract

This article discusses gift authorship, the practice where co-authorship is awarded to a person who has not contributed significantly to the study. From an ethical point of view, gift authorship raises concerns about desert, fairness, honesty and transparency, and its prevalence in research is rightly considered a serious ethical concern. We argue that even though misuse of authorship is always bad, there are instances where accepting requests of gift authorship may nevertheless be the right thing to do. More specifically, we propose that researchers may find themselves in a situation much similar to the problem of dirty hands, which has been frequently discussed in political philosophy and applied ethics. The problem of dirty hands is relevant to what we call hostage authorship, where the researchers include undeserving authors unwillingly, and only because they find it unavoidable in order to accomplish a morally important research goal.

Introduction

Authorship and the question of who ought to be included as authors on research articles occupy a central place in contemporary discussions on research ethics (Shamoo and Resnik, 2009). From the perspective of the researchers, the number of publications listed in their CVs has implications for promotion, tenure, salary and external funding (Jabbehdari and Walsh, 2017; Kornhaber et al., 2015; Smith and Williams-Jones, 2012; Zaki, 2011). From an ethical point of view, and according to widely recognized general authorship guidelines, only those who have contributed substantially to the article should be included as authors (ICMJE, 2016). Where this is not the case, it raises concerns about desert, fairness, honesty and transparency. When researchers are included as authors even though they have not made any substantial contribution to the article in question, it is both unjust – in the sense that they do not deserve the credit associated with the publication – and unfair to researchers who only include those who have earned it. It could also mislead the reader and may have negative consequences in terms of who gets appointed to academic positions and who is awarded research funding, with the risk of awarding the greatest fraudsters instead of the greatest researchers or most promising research projects. The prevalence of so-called gift authorship, where co-authorship is awarded to a person who has not contributed significantly to the study, is therefore rightly considered a serious concern (Eriksson et al., 2018; ¹ Wager, 2009; Zaki, 2011). Despite this, gift authorship is a widespread practice in research (Flanagin et al., 1998; Manton and English, 2006; Marušić et al., 2011).

A common position is that, regardless of the reason for granting a gift authorship, such practice should be strongly discouraged and measures taken to limit its prevalence (see, for example, Eriksson et al., 2018; ¹ Wager, 2009; Zaki, 2011). This certainly is the case when we consider some of the examples commonly acknowledged in discussions on authorship, such as when a non-contributing supervisor is listed as co-author on a PhD student’s work, or when a notable professor is listed as co-author merely for the sake of increasing the visibility or impact of the article.

However, as we argue in this article, it may nevertheless be too hasty a conclusion that it is always wrong to include as co-author a person who has not contributed substantially to the study. More specifically, we propose that researchers may find themselves in a situation much similar to the problem of dirty hands, which has been frequently discussed in political philosophy and applied ethics (Coady, 2014; Gaus, 2003). As typically argued, and as discussed in this article, the problem of dirty hands involves the idea that correct political action sometimes conflicts with deeply held moral norms, but is yet required in order to achieve a greater moral good or avoid serious harm (Archard, 2013; Walzer, 1973; Williams, 1978). As Bernard Williams argues, ‘it is a predictable and probable hazard of public life that there will be these situations in which something morally disagreeable is clearly required. To refuse on moral grounds ever to do anything of that sort is more likely to mean that one cannot seriously pursue even the moral ends of politics’ (Williams, 1978: 62). In other words, the nature of politics sometimes requires politicians to get their hands dirty. At the same time, none of this implies that the ‘moral disagreeableness’ of these acts is cancelled. As some philosophers have argued, such an act is wrongful and regrettable, and remains so even though it is required (Stocker, 1992; Walzer, 1973; Williams, 1978). Thus, the problem of dirty hands involves a kind of contradiction, as it suggests that it is sometimes right to do what is wrongful. ² In the following we argue that there are similar cases in relation to academic authorship, notably in relation to what we call hostage authorship.

Varieties of incorrect authorship inclusion

There are several terms used to denote various morally dubious authorship practices, and little agreement regarding what denotes what. For instance, the term gift authorship is most often used for any authorship practice where an author is included without having made a sufficient contribution to the study (see, for example, Committee on Publication Ethics, 2017), but is also used to mean something more narrow, such as listing an undeserving author ‘as a personal or professional favor’ (Shamoo and Resnik, 2009: 99). Also, honorary authorship seems to be used in a general sense (as above) and a more narrow sense, the latter referring to giving away authorship credit ‘as a sign of respect or gratitude’ (Shamoo and Resnik, 2009: 99). Apart from these two, there are several other labels in use, such as guest authorship, gratuitous authorship, and prestige authorship.

Leaving the terminological cacophony aside, there are a number of specific ideas regarding what motivates or explains the inclusion of undeserving authors on articles (for brevity, we call such an undeserving author X):

• 1. X is included as author in order to increase the likelihood that the article gets accepted for publication, or gets accepted by a higher-impact journal. To accomplish this effect, X needs to be well renowned in his/her field – editors will therefore find the article more attractive to their journals, as well-renowned researchers on published articles tend to attract readers.

• 2. X is included as author in order to increase the visibility and impact of the article, potentially leading to greater attention and more citations. This strategy targets the research community, and its interest in the renowned author, rather than the editors. Editors, of course, share the authors’ interest if the article gets accepted.

• 3. X is included as author as part of a collaborative scheme intended to increase the collaborators’ number of publications, of the type ‘if you include me on your articles, I’ll include you on mine’ (regardless of contribution). If X is included for this reason, then that most likely means that X already has a deal with one of the more influential authors of the article, and will reciprocate in the future.

• 4. X is included as author for strategical reasons relating to the research group. In order to obtain external funding, it might be beneficial for the group to have at least one researcher with an impressive CV to put up as principal investigator in the applications. For this reason, it might benefit everyone in the group if someone is added to every article of the group, even if the person does not contribute to many of the articles.

• 5. X is included as author as a favor, for instance because X needs another publication to get a promotion (such as advancing to associate professor). This may be done out of generosity, or with the expectation to get the favor returned in the future.

• 6. X is included as author out of gratitude or in order to honor X, without any other underlying intentions.

Although it is wrong to include persons as authors to articles to which they have not contributed substantially, regardless of what motivates the decision, we can see that the listed categories nevertheless differ to some extent from an ethical perspective. Some reasons simply seem worse than others. The first three points identify clearly selfish motives. The fourth point can perhaps be characterized as ‘group selfish’ – it benefits the individual doing it as well as the rest of the group, where the motivating reasons might concern one of these aspects or both. The fifth point may or may not be self-interestedly tactical; regardless, recognizing the gift authorship as a favor seems to imply that the researcher doing it is aware of the tactical benefit for the receiver, and the unfairness in relation to his or her competitors. The sixth point identifies a reason that is not selfish, but rather an expression of appreciation of someone else. It may, or may not, take place with recognition of the effects on fair distribution of authorship. Taken together, this provides an argument for the view that it does matter ethically, at least to some extent, why authorship is given away.

However, two more ideas that should be added to the list are the following:

• 7. X is included as an author as a result of having either put pressure on the legitimate author(s) of the article or in other ways coerced the legitimate author(s) to include X. We call this coerced authorship. Here the initiative comes from the (to be) illegitimate author rather than the legitimate ones. Other labels found in the literature for this are ‘coercion authorship’ and ‘pressured authorship’ (Feeser and Simon, 2008). The standard case imagined is that a senior researcher, e.g. the head of the research group or some larger research unit, insists on being included on articles where he or she has not contributed.

Coerced authorship can be of different kinds. In what follows we will focus on the following subcategory:

• 8. X is included as author because there is a hostage-like situation, where the researchers of the article cannot proceed and finish the article unless the conditions raised by X are fulfilled – conditions that include providing an authorship position on the article. We call this hostage authorship. Here the researchers of the article include X unwillingly, and only because they find it unavoidable in order to accomplish the overall research goal of the article.

As hostage authorship will be the focus of our argument in the subsequent sections, let us clarify what we have in mind. We have a case of a hostage authorship if the author has made his way to the article by raising that as a demand in order to allow researchers access to something that he has monopolistic control over. Hostage-like situations may occur, for instance, when someone controls access to rare technical equipment or a unique data base, or owns a patent or a unique reagent necessary for certain kinds of experiments. Hostage-like situations may also occur in the relation between preclinical researchers and clinicians able to provide blood or tissue samples of interest for research. That a hostage-like situation may occur does, of course, not mean that it always does so in the kinds of contexts mentioned.

There may be situations that are not as clear cut. In particular, researchers may foresee the risk of a conflict unless they offer an authorship position that the other party will not have deserved, and they may make such an offer without first having faced an explicit demand. Then the hostage situation has not clearly arisen, but is merely foreseen and avoided before it arises. In subsequent sections, we argue that this way of proceeding is less acceptable, compared to granting authorship to X in cases where the demands are already raised. In situations where monopolistic collaboration is unavoidable, or difficult to circumvent, it is easy to get into the habit of never aiming for the best outcome – that research is carried out and only those deserving it get listed as authors. But if those willing to do the right thing always back down whenever they need to collaborate with less scrupulous researchers/research holders, then the battle against fraud authorship will be lost.

For the record, we should make clear before proceeding that using, for instance, a patent or data base as a leverage to get included as an author on articles without making any substantial contribution is by itself an instance of unethical behavior in research. So, from the perspective of the scientific community, such practices ought to be counteracted and punished if exposed. This does not, however, solve the problem of how researchers should act when facing the problem.

Hostage authorship and dirty hands

What we will argue in what follows is that, in the case of hostage-like situations, there may be times when granting authorship to X is the ethically required course of action, i.e. what we ought to do, even if it requires us to do something that is intrinsically bad. Consider, for example, the following scenario:

Dr Jekyll is a medical researcher with his own research group, working on finding out more about a particular kind of cancer, with the long-term hope of finding a cure. There has been considerable progress in the lab work recently, but Dr Jekyll and his group need to run a new set of experiments to be able to move further. The only place for these experiments is in Lab X, run by Professor Hyde. Hyde is happy to provide the opportunity for Dr Jekyll to do his research, with only one little condition: Hyde should be included as co-author on all articles stemming from the work in his lab, although he has no intention to participate in the work. Dr Jekyll opposes Professor Hyde’s request, but Hyde insists. Dr Jekyll strongly believes that if he does not include Professor Hyde, his research group will not be able to continue their promising work.

Given the importance of Dr Jekyll’s work, and given that he has strong reasons to believe that he will not be able to carry out the research without access to Lab X, it seems that he has a weighty moral reason to accept Professor Hyde’s request, corresponding with the duty of beneficence in biomedical ethics (Beauchamp and Childress, 2013). At the same time, Dr Jekyll has the same reasons as any other researcher to adhere to the rules of conduct concerning authorship. Not doing so would be unfair to other researchers and would also violate core values in research, such as honesty and transparency (Shamoo and Resnik, 2009). What, then, should Dr Jekyll do in this case? To us it seems that Dr Jekyll faces a moral conflict, where he will either violate the norms of justice and fairness or he will fail to do good. Thus, his situation resembles that of the problem of dirty hands as Dr Jekyll is confronted with a situation where he can, and arguably should, violate widely held ethical norms for the sake of a greater moral good.

Should Dr Jekyll get his hands dirty? This depends, but if this is both practically necessary and proportionate to the end, then he should. There should be no other realistic option open to Dr Jekyll besides granting authorship to Professor Hyde, and the research in question must be sufficiently important in order to override the moral constraints of justice and fairness. However, as in dirty hands cases, none of this means that Dr Jekyll should pride himself on what he is doing. Rather, he should feel regret for having to accept this means to the good end and having to betray the ethical norms of the scientific community to which he belongs. The fact that Dr Jekyll should feel reluctant to include Professor Hyde also explains why it would be wrong to include an undeserved author when merely having foreseen the risk of a conflict unless an authorship position is offered. Unlike an explicit hostage-like scenario it has not been ruled out that granting authorship might be practically unnecessary, and it is therefore not clear why one should willingly compromise with the ethical norms of research when this is not required.

One should be aware that this argument, as any dirty hands case, asserts that there sometimes are inevitable moral conflicts where an act may be justified, even obligatory, yet also wrongful and regrettable (Stocker, 1992). Some may object to this, arguing that Dr Jekyll should not feel any regret because of Professor Hyde’s immoral behavior. In response, one should keep in mind that even though Professor Hyde’s behavior is an act of coercion, forcing Dr Jekyll to take part in an immoral endeavor, Dr Jekyll still finds himself in a situation where he needs to do something regrettable.

Implications

In the following, we wish to point to some of the implications of our argument. Firstly, that researchers sometimes face hostage-like situations like that between Jekyll and Hyde is by no means unrealistic. There are many different kinds of contexts, not least in medical research, where a hostage-like situation may occur, for instance, relating to access to blood and tissue from healthcare, to biobanks and data bases, to reagents of different kinds with only one provider, and to expensive technical equipment in the possession or control of a few. This does not mean that one should accept requests of hostage authorship lightheartedly. On the contrary, such authorship requests should be denied either if what is desired for the research can be obtained by other means, or if the research lacks connection to any moral good important enough to justify the wrongdoing – in the latter case it is better not to carry out the research than to give in to the hostage-taker’s demand. So, even though there are instances where researchers should willingly get their hands dirty, not all instances of hostage-like situation render gift authorship defensible. To give another example, junior researchers may very well find themselves in hostage-like scenarios in relation to their supervisors, where it is their career that is on the line. Even though this is an awful situation, it is doubtful whether it is justifiable to save one’s career in situations like these by accepting the hostage-taker’s conditions – rather the junior researcher needs to take some other form of action, like contacting the head of department for support and advice.

Secondly, what needs to be considered, and which is perhaps mostly overlooked in research ethical discussions, is how scenarios of this sort may put researchers in situations where they should not adhere to the standard guidelines concerning authorship. Guidelines such as the Vancouver rules (ICMJE, 2016) provide helpful instructions to the extent that ideal circumstances prevail, but mainly remain silent when it comes to how to proceed in a variety of situations where some circumstances already are not the way they should be. Perhaps one should not require of guidelines to be complete in this regard, but if they are not they leave some important ethical authorship issues untreated.

Thirdly, although researchers may face situations where they should be prepared to get their hands dirty, this nevertheless brings additional responsibilities. Being researchers, they already have a shared responsibility to counteract and discourage all forms of dishonest behavior in science, but because of hands-on experience of corruption in research they have duties to contribute to this problem becoming known and dealt with.

Conclusions

Misuse of authorship is always bad, as it involves dishonesty and is unjust, unfair to competitors, and may have negative consequences in terms of who gets academic positions and research funding. However, in hostage-like situations it may nevertheless be the right thing to do to accept a hostage authorship in order to be able to carry out important research, even though it will leave you with dirty hands. This is the case if doing so is practically necessary and proportionate to the end. This is not to say that bad practices should be accepted without resistance. The research community must make the effort to strike back against fraudulent authorship practices.