“A new form of plagiarism:” When researchers fake co-authors’ names

There’s a new publishing trend in town, says Mario Biagioli: Faking co-authors’ names. Biagioli, distinguished professor of law and science and technology studies and director of the Center for Innovation Studies at the University of California, Davis, writes in an article in Trends in Chemistry that it’s “the emergence of a new form of plagiarism that reflects the new metrics-based economy of scholarly publishing.” We asked him a few questions about what he’s found, and why authors might do this.

Retraction Watch (RW): You write that “A new trend in scientific misconduct involves listing fake coauthors on one’s publication.” Why would authors do that?

Mario Biagioli (MB): Without more fine-grained evidence about these cases (and you know better than I do how little information is generally provided by retraction notices), it is difficult to explain practices like these that seem to defy economic rationality. Why should I give away authorial credit to co-authors that I am making up?  

One hypothesis is that fraudsters might risk to attract too much attention to themselves by publishing at an unusually fast clip, which they can do given that it takes much less time to put out fraudulent publications than properly researched ones.  Adding fake co-authors helps to make anomalous productivity look more normal.  

Another hypothesis is that if you publish fraudulent claims of a kind that, if properly produced, would require the collaboration of scientists with different skills, it makes sense to make up co-authors with those skills and paste them on the byline. That may be especially important if people know that you do not possess all the different skills needed to produce the article you just published.

A last possibility, and the one that I mostly focus on in the article, is that people make up coauthors as fake links to real institutions. Because I am not a well-known scientist from a well-known university, chances of having my submission taken seriously or sent out for review by a high-impact journal may be slim, but I can improve my odds by listing fake co-authors who happen to work at Caltech, Cambridge, etc.

There are at least two variations on this strategy.  One involves listing real people with credible institutional affiliations as co-authors, but without telling them. Whether one lists fake people or real but unaware ones, the main goal is to draw credibility from their institutional affiliations.  What changes are the risks involved. Using real people instead of made-up ones increases the chance to be caught by one’s fictional co-authors when they see their names on unknown publications. The same scenario, however, would be less likely to raise flags in the mind of readers who would not become suspicious by seeing names of colleagues they know listed as co-authors with other scientists whose name they do not recognize.  Conversely, using fake co-authors could not be detected by co-authors who do not exist, but may raise suspicion among readers familiar with that field but not with those names.

Another possibility involves pseudonymous publications. If my goal is to put out a controversial claim without the risk of being identified but also without the loss of credibility that goes with anonymous publications, I could assume a fake name while giving myself a real and prestigious institutional affiliation to confer credibility to my claims while also facilitating their publication. I would relinquish authorial credit, but that does not matter given that I don’t want to be identified anyway.  

RW: You write about the case of Jesús Ángel Lemus, who is up to 13 retractions. Tell us about this case, and what you took from it.

MB: Lemus is a Spanish veterinary doctor who started to pursue a research career in his mid-thirties, obtaining his PhD in 2010, and publishing at a fast clip since 2006, possibly to make up for his late start. He has specialized on the impact of environmental pollution on wildlife — a hot topic in Europe. Acquaintances of his interviewed by El Pais (which has published several good articles on the case) represent him as a wildlife biologist who loved to spend time in the field. But his budding career ended as soon as it was starting.  As you say, he has gained 13 retractions, plus a correction and an expression of concern. Most of his early publications were with his dissertation advisor — Guillermo Blanco. Six of them of them also listed a certain Javier Grande.  

It turns out that Grande is an imaginary co-author that was credited (in two now-retracted PLOS ONE articles that required descriptions of the authors’ contributions) for having “analyzed the data,” “performed pathogens determinations,” and “contributed reagents/materials/analysis/tools.” Unsurprisingly, Javier Grande’s institutional affiliations (as listed in his publications) tracked those of Lemus like a shadow. The overlap also extended to their authorial contributions.

It is surprising that while Lemus was better known as a field biologist and practicing veterinarian than a laboratory scientist, a colleague told El Pais that: “Lemus was very effective. If you gave him samples of birds to analyze pathogens or antbiotics, he was always on time and the best thing is that there was always a publishable result.” These are precisely the skills that were also attributed to Grande. A co-author of Lemus and Grande described the latter as, “a fellow at the IREC [Instituto de Investigación en Recursos Cinegéticos] that analyzed the samples. I had email exchanges with him.” Grande, in sum, was the doppelganger of who Lemus would have liked to be.

The fiction collapsed when Lemus pushed his luck too far, claiming that “almost half of the parrots in Barcelona were infected with the psittacosis bacterium – a disease transmissible to humans,” That got people’s attention, which quickly brought scrutiny to Lemus’ oeuvre. His co-authors failed to replicate his tests (in some cases the samples had decayed) while others could not locate the external testing laboratories he might have relied on.  Eventually his co-authors had to admit that they have never seen or met Javier Grande in person, and that there were no institutional traces of his existence.  (This is rather striking given that at least one article was authored by scientists who — Grande included — were all affiliated with the same department, in the same institution).  

Lemus fits well two categories of the typology I sketched out above.  He made up Grande as a scientist who had the skills that Lemus claimed to have but probably was still developing. He also attributed Grande authoritative institutional affiliations in that field, which certainly did not hurt his publication chances. But there are a few interesting additional twists.  

El Pais reported that in March 2012, as the investigation of Lemus was in full gear, his curriculum vitae was suddenly removed from his institutional website at CSIS. According to some witnesses, the curriculum had listed 6 articles (apparently single-authored), except that those publications did not exist. More precisely, the journals did exist, but they did not contain the publications that Lemus claimed to have published in those journals.  

Then, in addition to the fact that all these journals were highly specialistic and difficult to find, Lemus listed incongruent volume numbers and publication years, thus confusing potential readers who might conclude that, instead of a non-existing publication, they were simply dealing with a typo in his bibliography.  Interestingly, Lemus does not appear to have listed Grande as co-author on these imaginary publications. He did not need to. Being figments of his imagination, the inexistent publications he listed on his disappearing vitae did not need to be accepted for publication, and did not need to convince co-authors and colleagues that he had somebody on the team who could do pathogens’ tests. It would have been beyond the point to increase the credibility of inexistent publications by adding fake co-authors.  

Lemus’ two strategies seem surprisingly symmetric. In one case he made up a fake co-author with real skills and real institutional affiliations.  In the other, he made up fake articles in real journals. What we find in both cases are not carefully falsified claims but rather the pairing of obviously real and obviously fake claims. If the El Pais report is correct, a thorough search of his online curriculum would have easily detected the unambiguously fraudulent nature of his bibliography.  

Similarly, if one decided to track down Javier Grande, s/he would have quickly found that he was a figment of Lemus’ imagination. It is not a complex matter of interpretation: either a person by that name and affiliations exists in Madrid or he doesn’t.  But in both cases delays and deflections were put in place to buffer Lemus from easy detection.  His fake curriculum listed intentional typos and obscure journals, and Javier Grande happened to have a flesh-and-blood “decoy” in Madrid: Javier Grande Ortiz.  

Ortiz is a legitimate veterinarian in private practice, and a former fellow student of Lemus’ in veterinary school.  Being in the same field (but not in academia) Javier Grande Ortiz might have helped to make “Javier Grande” a little less strange a name in that local community. None of that thin linguistic camouflage would have provided safety for Lemus, but may have been enough to provide “plausible deniability” to those (inside and outside of Lemus’ team) who did not feel like probing.

In my article I emphasize how fake co-authors are often introduced to maximize the credibility of a submission or publication in the eyes of editors or readers. Lemus’ case suggests that he might have targeted a more local, internal audience, introducing Grande to make his other flesh-and-blood co-authors feel that there was a team member with the right skills for the job — a fictional collaborator that may have helped Lemus to cover up for the skills that he may not have had. (Interestingly, the six articles that Lemus “co-authored” with Grande all appeared from 2006 to 2009, when Lemus was a graduate student. As soon as Lemus graduated early in 2010, Grande vanished). 

But while things did not pan out for Lemus, we should at least give him credit for a new articulation of the game of musical chairs that typically plays when fraud accusations start flying. Ghost authorship often enables senior authors to claim that their ghostliness was benign. They had nothing to do with the mess; their name was lent out of generosity rather than forced in the byline by the desire for undue authorial credit.

Instead, when asked, “Who is Javier Grande?”, Lemus responded: “You ask me? Ask Guillermo Blanco.”  Blanco is Lemus’ PhD advisor and the only other person who, together with Lemus, co-authored all of Grande’s retracted articles.  Not so subtly (and probably a little more than self-servingly) Lemus suggested that the ghost of Javier Grande may have been made up by another ghost – a senior one.  

RW: Are you familiar with the case of Stronzo Bestiale? Does it remind you of the cases you describe?

MB: Yes, in some sense Stronzo Bestiale would seem to mark the beginning of the trend I am describing in the article. If I did not mention it is because I have not quite figured out what that story is really about.  Based on the information you read on the web — I have not found a scholarly treatment of this episode – this sounds like a case of made up co-authorship, but the details are unclear.

According to one source, in the late 1980s, a US physicist working at Livermore — William G. Hoover — struggled to get some of his research published, apparently because of its particularly innovative nature.  When interviewed decades later, Hoover reminisced that: “while I was traveling on a flight to Paris, next to me were two Italian women who spoke among themselves, saying continually: ‘Che stronzo (what an asshole)!’, ‘Stronzo bestiale (total asshole).’ […] I thought that Stronzo Bestiale would have been the perfect co-author for a refused publication. So I decided to submit my papers again, simply by changing the title and adding the name of that author. And the researches were published.”  The Scopus Index now lists a “Bestiale, Stronzo” affiliated with the University of Vienna Institute for Experimental Physics, who has authored two 1987 articles with William G. Hoover.  

Given the scattered information, it is difficult to tell whether Hoover strategically affiliated Bestiale with the University of Vienna to facilitate the publication of his previously rejected work, or whether he simply picked up an institutional affiliation because, as a physicist, Dr. Bestiale needed one.  But people who plan to gain something by using fake co-authors do their best to fly below the radar and do not use insulting words on the very first page of their publications. That Hoover resubmitted the articles to some of the same journals that had previously rejected them may suggest that he simply wanted to stick it to the editors by inserting a strong insult in the resubmission – an insult that they did not understand and thus reprinted with the article.  Because they did not get the joke, they reproduced it, turning it into a bigger joke on them.

If this is a plausible reading, then while Stronzo Bestiale may look like the kind of fake co-author that we see today, it is in fact something quite different. And I do not think we can say it was a hoax. Hoaxes become hoaxes only if they are made public shortly after they have been performed, but Stronzo Bestiale surfaced decades after he was deployed. Also, physicists have not treated this as misconduct either. The Retraction Watch database does not show any retraction for either Stronzo Bestiale or William G. Hoover.  Of course we can say that it was just a joke (good or bad depending on one’s taste), though that’s not exactly an analytical description. In sum, I do not know how to categorize Dr. Bestiale.

RW: What can be done to combat this trend?

MB: A simple answer, for a change.  Editors could email all co-authors (corresponding or not) at their institutional email address to check if they agreed to author the submitted paper. That simple step would probably solve all the problems involved in the cases I have discussed. ORCID identifiers would not hurt either.  

Like Retraction Watch? You can make a tax-deductible contribution to support our growth, follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up for an email every time there’s a new post (look for the “follow” button at the lower right part of your screen), or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].